diff options
Diffstat (limited to 'youtube_dl')
242 files changed, 14349 insertions, 5125 deletions
diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 055433362..ba72ec6f3 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -64,6 +64,7 @@ from .utils import ( PostProcessingError, preferredencoding, prepend_extension, + register_socks_protocols, render_table, replace_extension, SameFileError, @@ -195,8 +196,8 @@ class YoutubeDL(object): prefer_insecure: Use HTTP instead of HTTPS to retrieve information. At the moment, this is only supported by YouTube. proxy: URL of the proxy server to use - cn_verification_proxy: URL of the proxy to use for IP address verification - on Chinese sites. (Experimental) + geo_verification_proxy: URL of the proxy to use for IP address verification + on geo-restricted sites. (Experimental) socket_timeout: Time to wait for unresponsive hosts, in seconds bidi_workaround: Work around buggy terminals without bidirectional text support, using fridibi @@ -303,6 +304,11 @@ class YoutubeDL(object): self.params.update(params) self.cache = Cache(self) + if self.params.get('cn_verification_proxy') is not None: + self.report_warning('--cn-verification-proxy is deprecated. Use --geo-verification-proxy instead.') + if self.params.get('geo_verification_proxy') is None: + self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] + if params.get('bidi_workaround', False): try: import pty @@ -325,7 +331,7 @@ class YoutubeDL(object): ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) self._output_channel = os.fdopen(master, 'rb') except OSError as ose: - if ose.errno == 2: + if ose.errno == errno.ENOENT: self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.') else: raise @@ -361,6 +367,8 @@ class YoutubeDL(object): for ph in self.params.get('progress_hooks', []): self.add_progress_hook(ph) + register_socks_protocols() + def warn_if_short_id(self, argv): # short YouTube ID starting with dash? idxs = [ @@ -580,7 +588,7 @@ class YoutubeDL(object): is_id=(k == 'id')) template_dict = dict((k, sanitize(k, v)) for k, v in template_dict.items() - if v is not None) + if v is not None and not isinstance(v, (list, tuple, dict))) template_dict = collections.defaultdict(lambda: 'NA', template_dict) outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) @@ -717,6 +725,7 @@ class YoutubeDL(object): result_type = ie_result.get('_type', 'video') if result_type in ('url', 'url_transparent'): + ie_result['url'] = sanitize_url(ie_result['url']) extract_flat = self.params.get('extract_flat', False) if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or extract_flat is True): @@ -1219,6 +1228,10 @@ class YoutubeDL(object): if 'title' not in info_dict: raise ExtractorError('Missing "title" field in extractor result') + if not isinstance(info_dict['id'], compat_str): + self.report_warning('"id" field is not a string - forcing string conversion') + info_dict['id'] = compat_str(info_dict['id']) + if 'playlist' not in info_dict: # It isn't part of a playlist info_dict['playlist'] = None @@ -1639,7 +1652,7 @@ class YoutubeDL(object): # Just a single file success = dl(filename, info_dict) except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self.report_error('unable to download video data: %s' % str(err)) + self.report_error('unable to download video data: %s' % error_to_compat_str(err)) return except (OSError, IOError) as err: raise UnavailableVideoError(err) @@ -2018,6 +2031,7 @@ class YoutubeDL(object): if opts_cookiefile is None: self.cookiejar = compat_cookiejar.CookieJar() else: + opts_cookiefile = compat_expanduser(opts_cookiefile) self.cookiejar = compat_cookiejar.MozillaCookieJar( opts_cookiefile) if os.access(opts_cookiefile, os.R_OK): diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 737f6545d..2b34bf9c2 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -18,7 +18,6 @@ from .options import ( from .compat import ( compat_expanduser, compat_getpass, - compat_print, compat_shlex_split, workaround_optparse_bug9161, ) @@ -67,16 +66,16 @@ def _real_main(argv=None): # Custom HTTP headers if opts.headers is not None: for h in opts.headers: - if h.find(':', 1) < 0: + if ':' not in h: parser.error('wrong header formatting, it should be key:value, not "%s"' % h) - key, value = h.split(':', 2) + key, value = h.split(':', 1) if opts.verbose: write_string('[debug] Adding header from command line option %s:%s\n' % (key, value)) std_headers[key] = value # Dump user agent if opts.dump_user_agent: - compat_print(std_headers['User-Agent']) + write_string(std_headers['User-Agent'] + '\n', out=sys.stdout) sys.exit(0) # Batch file verification @@ -86,7 +85,9 @@ def _real_main(argv=None): if opts.batchfile == '-': batchfd = sys.stdin else: - batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore') + batchfd = io.open( + compat_expanduser(opts.batchfile), + 'r', encoding='utf-8', errors='ignore') batch_urls = read_batch_urls(batchfd) if opts.verbose: write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n') @@ -99,10 +100,10 @@ def _real_main(argv=None): if opts.list_extractors: for ie in list_extractors(opts.age_limit): - compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '')) + write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '') + '\n', out=sys.stdout) matchedUrls = [url for url in all_urls if ie.suitable(url)] for mu in matchedUrls: - compat_print(' ' + mu) + write_string(' ' + mu + '\n', out=sys.stdout) sys.exit(0) if opts.list_extractor_descriptions: for ie in list_extractors(opts.age_limit): @@ -115,7 +116,7 @@ def _real_main(argv=None): _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow') _COUNTS = ('', '5', '10', 'all') desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES)) - compat_print(desc) + write_string(desc + '\n', out=sys.stdout) sys.exit(0) # Conflicting, missing and erroneous options @@ -381,6 +382,8 @@ def _real_main(argv=None): 'external_downloader_args': external_downloader_args, 'postprocessor_args': postprocessor_args, 'cn_verification_proxy': opts.cn_verification_proxy, + 'geo_verification_proxy': opts.geo_verification_proxy, + } with YoutubeDL(ydl_opts) as ydl: @@ -404,7 +407,7 @@ def _real_main(argv=None): try: if opts.load_info_filename is not None: - retcode = ydl.download_with_info_file(opts.load_info_filename) + retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename)) else: retcode = ydl.download(all_urls) except MaxDownloadsReached: diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index 0b6c5ca7a..b8aaf5a46 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -1,3 +1,4 @@ +# coding: utf-8 from __future__ import unicode_literals import binascii @@ -11,6 +12,7 @@ import re import shlex import shutil import socket +import struct import subprocess import sys import itertools @@ -62,6 +64,2244 @@ try: except ImportError: # Python 2 import htmlentitydefs as compat_html_entities +try: # Python >= 3.3 + compat_html_entities_html5 = compat_html_entities.html5 +except AttributeError: + # Copied from CPython 3.5.1 html/entities.py + compat_html_entities_html5 = { + 'Aacute': '\xc1', + 'aacute': '\xe1', + 'Aacute;': '\xc1', + 'aacute;': '\xe1', + 'Abreve;': '\u0102', + 'abreve;': '\u0103', + 'ac;': '\u223e', + 'acd;': '\u223f', + 'acE;': '\u223e\u0333', + 'Acirc': '\xc2', + 'acirc': '\xe2', + 'Acirc;': '\xc2', + 'acirc;': '\xe2', + 'acute': '\xb4', + 'acute;': '\xb4', + 'Acy;': '\u0410', + 'acy;': '\u0430', + 'AElig': '\xc6', + 'aelig': '\xe6', + 'AElig;': '\xc6', + 'aelig;': '\xe6', + 'af;': '\u2061', + 'Afr;': '\U0001d504', + 'afr;': '\U0001d51e', + 'Agrave': '\xc0', + 'agrave': '\xe0', + 'Agrave;': '\xc0', + 'agrave;': '\xe0', + 'alefsym;': '\u2135', + 'aleph;': '\u2135', + 'Alpha;': '\u0391', + 'alpha;': '\u03b1', + 'Amacr;': '\u0100', + 'amacr;': '\u0101', + 'amalg;': '\u2a3f', + 'AMP': '&', + 'amp': '&', + 'AMP;': '&', + 'amp;': '&', + 'And;': '\u2a53', + 'and;': '\u2227', + 'andand;': '\u2a55', + 'andd;': '\u2a5c', + 'andslope;': '\u2a58', + 'andv;': '\u2a5a', + 'ang;': '\u2220', + 'ange;': '\u29a4', + 'angle;': '\u2220', + 'angmsd;': '\u2221', + 'angmsdaa;': '\u29a8', + 'angmsdab;': '\u29a9', + 'angmsdac;': '\u29aa', + 'angmsdad;': '\u29ab', + 'angmsdae;': '\u29ac', + 'angmsdaf;': '\u29ad', + 'angmsdag;': '\u29ae', + 'angmsdah;': '\u29af', + 'angrt;': '\u221f', + 'angrtvb;': '\u22be', + 'angrtvbd;': '\u299d', + 'angsph;': '\u2222', + 'angst;': '\xc5', + 'angzarr;': '\u237c', + 'Aogon;': '\u0104', + 'aogon;': '\u0105', + 'Aopf;': '\U0001d538', + 'aopf;': '\U0001d552', + 'ap;': '\u2248', + 'apacir;': '\u2a6f', + 'apE;': '\u2a70', + 'ape;': '\u224a', + 'apid;': '\u224b', + 'apos;': "'", + 'ApplyFunction;': '\u2061', + 'approx;': '\u2248', + 'approxeq;': '\u224a', + 'Aring': '\xc5', + 'aring': '\xe5', + 'Aring;': '\xc5', + 'aring;': '\xe5', + 'Ascr;': '\U0001d49c', + 'ascr;': '\U0001d4b6', + 'Assign;': '\u2254', + 'ast;': '*', + 'asymp;': '\u2248', + 'asympeq;': '\u224d', + 'Atilde': '\xc3', + 'atilde': '\xe3', + 'Atilde;': '\xc3', + 'atilde;': '\xe3', + 'Auml': '\xc4', + 'auml': '\xe4', + 'Auml;': '\xc4', + 'auml;': '\xe4', + 'awconint;': '\u2233', + 'awint;': '\u2a11', + 'backcong;': '\u224c', + 'backepsilon;': '\u03f6', + 'backprime;': '\u2035', + 'backsim;': '\u223d', + 'backsimeq;': '\u22cd', + 'Backslash;': '\u2216', + 'Barv;': '\u2ae7', + 'barvee;': '\u22bd', + 'Barwed;': '\u2306', + 'barwed;': '\u2305', + 'barwedge;': '\u2305', + 'bbrk;': '\u23b5', + 'bbrktbrk;': '\u23b6', + 'bcong;': '\u224c', + 'Bcy;': '\u0411', + 'bcy;': '\u0431', + 'bdquo;': '\u201e', + 'becaus;': '\u2235', + 'Because;': '\u2235', + 'because;': '\u2235', + 'bemptyv;': '\u29b0', + 'bepsi;': '\u03f6', + 'bernou;': '\u212c', + 'Bernoullis;': '\u212c', + 'Beta;': '\u0392', + 'beta;': '\u03b2', + 'beth;': '\u2136', + 'between;': '\u226c', + 'Bfr;': '\U0001d505', + 'bfr;': '\U0001d51f', + 'bigcap;': '\u22c2', + 'bigcirc;': '\u25ef', + 'bigcup;': '\u22c3', + 'bigodot;': '\u2a00', + 'bigoplus;': '\u2a01', + 'bigotimes;': '\u2a02', + 'bigsqcup;': '\u2a06', + 'bigstar;': '\u2605', + 'bigtriangledown;': '\u25bd', + 'bigtriangleup;': '\u25b3', + 'biguplus;': '\u2a04', + 'bigvee;': '\u22c1', + 'bigwedge;': '\u22c0', + 'bkarow;': '\u290d', + 'blacklozenge;': '\u29eb', + 'blacksquare;': '\u25aa', + 'blacktriangle;': '\u25b4', + 'blacktriangledown;': '\u25be', + 'blacktriangleleft;': '\u25c2', + 'blacktriangleright;': '\u25b8', + 'blank;': '\u2423', + 'blk12;': '\u2592', + 'blk14;': '\u2591', + 'blk34;': '\u2593', + 'block;': '\u2588', + 'bne;': '=\u20e5', + 'bnequiv;': '\u2261\u20e5', + 'bNot;': '\u2aed', + 'bnot;': '\u2310', + 'Bopf;': '\U0001d539', + 'bopf;': '\U0001d553', + 'bot;': '\u22a5', + 'bottom;': '\u22a5', + 'bowtie;': '\u22c8', + 'boxbox;': '\u29c9', + 'boxDL;': '\u2557', + 'boxDl;': '\u2556', + 'boxdL;': '\u2555', + 'boxdl;': '\u2510', + 'boxDR;': '\u2554', + 'boxDr;': '\u2553', + 'boxdR;': '\u2552', + 'boxdr;': '\u250c', + 'boxH;': '\u2550', + 'boxh;': '\u2500', + 'boxHD;': '\u2566', + 'boxHd;': '\u2564', + 'boxhD;': '\u2565', + 'boxhd;': '\u252c', + 'boxHU;': '\u2569', + 'boxHu;': '\u2567', + 'boxhU;': '\u2568', + 'boxhu;': '\u2534', + 'boxminus;': '\u229f', + 'boxplus;': '\u229e', + 'boxtimes;': '\u22a0', + 'boxUL;': '\u255d', + 'boxUl;': '\u255c', + 'boxuL;': '\u255b', + 'boxul;': '\u2518', + 'boxUR;': '\u255a', + 'boxUr;': '\u2559', + 'boxuR;': '\u2558', + 'boxur;': '\u2514', + 'boxV;': '\u2551', + 'boxv;': '\u2502', + 'boxVH;': '\u256c', + 'boxVh;': '\u256b', + 'boxvH;': '\u256a', + 'boxvh;': '\u253c', + 'boxVL;': '\u2563', + 'boxVl;': '\u2562', + 'boxvL;': '\u2561', + 'boxvl;': '\u2524', + 'boxVR;': '\u2560', + 'boxVr;': '\u255f', + 'boxvR;': '\u255e', + 'boxvr;': '\u251c', + 'bprime;': '\u2035', + 'Breve;': '\u02d8', + 'breve;': '\u02d8', + 'brvbar': '\xa6', + 'brvbar;': '\xa6', + 'Bscr;': '\u212c', + 'bscr;': '\U0001d4b7', + 'bsemi;': '\u204f', + 'bsim;': '\u223d', + 'bsime;': '\u22cd', + 'bsol;': '\\', + 'bsolb;': '\u29c5', + 'bsolhsub;': '\u27c8', + 'bull;': '\u2022', + 'bullet;': '\u2022', + 'bump;': '\u224e', + 'bumpE;': '\u2aae', + 'bumpe;': '\u224f', + 'Bumpeq;': '\u224e', + 'bumpeq;': '\u224f', + 'Cacute;': '\u0106', + 'cacute;': '\u0107', + 'Cap;': '\u22d2', + 'cap;': '\u2229', + 'capand;': '\u2a44', + 'capbrcup;': '\u2a49', + 'capcap;': '\u2a4b', + 'capcup;': '\u2a47', + 'capdot;': '\u2a40', + 'CapitalDifferentialD;': '\u2145', + 'caps;': '\u2229\ufe00', + 'caret;': '\u2041', + 'caron;': '\u02c7', + 'Cayleys;': '\u212d', + 'ccaps;': '\u2a4d', + 'Ccaron;': '\u010c', + 'ccaron;': '\u010d', + 'Ccedil': '\xc7', + 'ccedil': '\xe7', + 'Ccedil;': '\xc7', + 'ccedil;': '\xe7', + 'Ccirc;': '\u0108', + 'ccirc;': '\u0109', + 'Cconint;': '\u2230', + 'ccups;': '\u2a4c', + 'ccupssm;': '\u2a50', + 'Cdot;': '\u010a', + 'cdot;': '\u010b', + 'cedil': '\xb8', + 'cedil;': '\xb8', + 'Cedilla;': '\xb8', + 'cemptyv;': '\u29b2', + 'cent': '\xa2', + 'cent;': '\xa2', + 'CenterDot;': '\xb7', + 'centerdot;': '\xb7', + 'Cfr;': '\u212d', + 'cfr;': '\U0001d520', + 'CHcy;': '\u0427', + 'chcy;': '\u0447', + 'check;': '\u2713', + 'checkmark;': '\u2713', + 'Chi;': '\u03a7', + 'chi;': '\u03c7', + 'cir;': '\u25cb', + 'circ;': '\u02c6', + 'circeq;': '\u2257', + 'circlearrowleft;': '\u21ba', + 'circlearrowright;': '\u21bb', + 'circledast;': '\u229b', + 'circledcirc;': '\u229a', + 'circleddash;': '\u229d', + 'CircleDot;': '\u2299', + 'circledR;': '\xae', + 'circledS;': '\u24c8', + 'CircleMinus;': '\u2296', + 'CirclePlus;': '\u2295', + 'CircleTimes;': '\u2297', + 'cirE;': '\u29c3', + 'cire;': '\u2257', + 'cirfnint;': '\u2a10', + 'cirmid;': '\u2aef', + 'cirscir;': '\u29c2', + 'ClockwiseContourIntegral;': '\u2232', + 'CloseCurlyDoubleQuote;': '\u201d', + 'CloseCurlyQuote;': '\u2019', + 'clubs;': '\u2663', + 'clubsuit;': '\u2663', + 'Colon;': '\u2237', + 'colon;': ':', + 'Colone;': '\u2a74', + 'colone;': '\u2254', + 'coloneq;': '\u2254', + 'comma;': ',', + 'commat;': '@', + 'comp;': '\u2201', + 'compfn;': '\u2218', + 'complement;': '\u2201', + 'complexes;': '\u2102', + 'cong;': '\u2245', + 'congdot;': '\u2a6d', + 'Congruent;': '\u2261', + 'Conint;': '\u222f', + 'conint;': '\u222e', + 'ContourIntegral;': '\u222e', + 'Copf;': '\u2102', + 'copf;': '\U0001d554', + 'coprod;': '\u2210', + 'Coproduct;': '\u2210', + 'COPY': '\xa9', + 'copy': '\xa9', + 'COPY;': '\xa9', + 'copy;': '\xa9', + 'copysr;': '\u2117', + 'CounterClockwiseContourIntegral;': '\u2233', + 'crarr;': '\u21b5', + 'Cross;': '\u2a2f', + 'cross;': '\u2717', + 'Cscr;': '\U0001d49e', + 'cscr;': '\U0001d4b8', + 'csub;': '\u2acf', + 'csube;': '\u2ad1', + 'csup;': '\u2ad0', + 'csupe;': '\u2ad2', + 'ctdot;': '\u22ef', + 'cudarrl;': '\u2938', + 'cudarrr;': '\u2935', + 'cuepr;': '\u22de', + 'cuesc;': '\u22df', + 'cularr;': '\u21b6', + 'cularrp;': '\u293d', + 'Cup;': '\u22d3', + 'cup;': '\u222a', + 'cupbrcap;': '\u2a48', + 'CupCap;': '\u224d', + 'cupcap;': '\u2a46', + 'cupcup;': '\u2a4a', + 'cupdot;': '\u228d', + 'cupor;': '\u2a45', + 'cups;': '\u222a\ufe00', + 'curarr;': '\u21b7', + 'curarrm;': '\u293c', + 'curlyeqprec;': '\u22de', + 'curlyeqsucc;': '\u22df', + 'curlyvee;': '\u22ce', + 'curlywedge;': '\u22cf', + 'curren': '\xa4', + 'curren;': '\xa4', + 'curvearrowleft;': '\u21b6', + 'curvearrowright;': '\u21b7', + 'cuvee;': '\u22ce', + 'cuwed;': '\u22cf', + 'cwconint;': '\u2232', + 'cwint;': '\u2231', + 'cylcty;': '\u232d', + 'Dagger;': '\u2021', + 'dagger;': '\u2020', + 'daleth;': '\u2138', + 'Darr;': '\u21a1', + 'dArr;': '\u21d3', + 'darr;': '\u2193', + 'dash;': '\u2010', + 'Dashv;': '\u2ae4', + 'dashv;': '\u22a3', + 'dbkarow;': '\u290f', + 'dblac;': '\u02dd', + 'Dcaron;': '\u010e', + 'dcaron;': '\u010f', + 'Dcy;': '\u0414', + 'dcy;': '\u0434', + 'DD;': '\u2145', + 'dd;': '\u2146', + 'ddagger;': '\u2021', + 'ddarr;': '\u21ca', + 'DDotrahd;': '\u2911', + 'ddotseq;': '\u2a77', + 'deg': '\xb0', + 'deg;': '\xb0', + 'Del;': '\u2207', + 'Delta;': '\u0394', + 'delta;': '\u03b4', + 'demptyv;': '\u29b1', + 'dfisht;': '\u297f', + 'Dfr;': '\U0001d507', + 'dfr;': '\U0001d521', + 'dHar;': '\u2965', + 'dharl;': '\u21c3', + 'dharr;': '\u21c2', + 'DiacriticalAcute;': '\xb4', + 'DiacriticalDot;': '\u02d9', + 'DiacriticalDoubleAcute;': '\u02dd', + 'DiacriticalGrave;': '`', + 'DiacriticalTilde;': '\u02dc', + 'diam;': '\u22c4', + 'Diamond;': '\u22c4', + 'diamond;': '\u22c4', + 'diamondsuit;': '\u2666', + 'diams;': '\u2666', + 'die;': '\xa8', + 'DifferentialD;': '\u2146', + 'digamma;': '\u03dd', + 'disin;': '\u22f2', + 'div;': '\xf7', + 'divide': '\xf7', + 'divide;': '\xf7', + 'divideontimes;': '\u22c7', + 'divonx;': '\u22c7', + 'DJcy;': '\u0402', + 'djcy;': '\u0452', + 'dlcorn;': '\u231e', + 'dlcrop;': '\u230d', + 'dollar;': '$', + 'Dopf;': '\U0001d53b', + 'dopf;': '\U0001d555', + 'Dot;': '\xa8', + 'dot;': '\u02d9', + 'DotDot;': '\u20dc', + 'doteq;': '\u2250', + 'doteqdot;': '\u2251', + 'DotEqual;': '\u2250', + 'dotminus;': '\u2238', + 'dotplus;': '\u2214', + 'dotsquare;': '\u22a1', + 'doublebarwedge;': '\u2306', + 'DoubleContourIntegral;': '\u222f', + 'DoubleDot;': '\xa8', + 'DoubleDownArrow;': '\u21d3', + 'DoubleLeftArrow;': '\u21d0', + 'DoubleLeftRightArrow;': '\u21d4', + 'DoubleLeftTee;': '\u2ae4', + 'DoubleLongLeftArrow;': '\u27f8', + 'DoubleLongLeftRightArrow;': '\u27fa', + 'DoubleLongRightArrow;': '\u27f9', + 'DoubleRightArrow;': '\u21d2', + 'DoubleRightTee;': '\u22a8', + 'DoubleUpArrow;': '\u21d1', + 'DoubleUpDownArrow;': '\u21d5', + 'DoubleVerticalBar;': '\u2225', + 'DownArrow;': '\u2193', + 'Downarrow;': '\u21d3', + 'downarrow;': '\u2193', + 'DownArrowBar;': '\u2913', + 'DownArrowUpArrow;': '\u21f5', + 'DownBreve;': '\u0311', + 'downdownarrows;': '\u21ca', + 'downharpoonleft;': '\u21c3', + 'downharpoonright;': '\u21c2', + 'DownLeftRightVector;': '\u2950', + 'DownLeftTeeVector;': '\u295e', + 'DownLeftVector;': '\u21bd', + 'DownLeftVectorBar;': '\u2956', + 'DownRightTeeVector;': '\u295f', + 'DownRightVector;': '\u21c1', + 'DownRightVectorBar;': '\u2957', + 'DownTee;': '\u22a4', + 'DownTeeArrow;': '\u21a7', + 'drbkarow;': '\u2910', + 'drcorn;': '\u231f', + 'drcrop;': '\u230c', + 'Dscr;': '\U0001d49f', + 'dscr;': '\U0001d4b9', + 'DScy;': '\u0405', + 'dscy;': '\u0455', + 'dsol;': '\u29f6', + 'Dstrok;': '\u0110', + 'dstrok;': '\u0111', + 'dtdot;': '\u22f1', + 'dtri;': '\u25bf', + 'dtrif;': '\u25be', + 'duarr;': '\u21f5', + 'duhar;': '\u296f', + 'dwangle;': '\u29a6', + 'DZcy;': '\u040f', + 'dzcy;': '\u045f', + 'dzigrarr;': '\u27ff', + 'Eacute': '\xc9', + 'eacute': '\xe9', + 'Eacute;': '\xc9', + 'eacute;': '\xe9', + 'easter;': '\u2a6e', + 'Ecaron;': '\u011a', + 'ecaron;': '\u011b', + 'ecir;': '\u2256', + 'Ecirc': '\xca', + 'ecirc': '\xea', + 'Ecirc;': '\xca', + 'ecirc;': '\xea', + 'ecolon;': '\u2255', + 'Ecy;': '\u042d', + 'ecy;': '\u044d', + 'eDDot;': '\u2a77', + 'Edot;': '\u0116', + 'eDot;': '\u2251', + 'edot;': '\u0117', + 'ee;': '\u2147', + 'efDot;': '\u2252', + 'Efr;': '\U0001d508', + 'efr;': '\U0001d522', + 'eg;': '\u2a9a', + 'Egrave': '\xc8', + 'egrave': '\xe8', + 'Egrave;': '\xc8', + 'egrave;': '\xe8', + 'egs;': '\u2a96', + 'egsdot;': '\u2a98', + 'el;': '\u2a99', + 'Element;': '\u2208', + 'elinters;': '\u23e7', + 'ell;': '\u2113', + 'els;': '\u2a95', + 'elsdot;': '\u2a97', + 'Emacr;': '\u0112', + 'emacr;': '\u0113', + 'empty;': '\u2205', + 'emptyset;': '\u2205', + 'EmptySmallSquare;': '\u25fb', + 'emptyv;': '\u2205', + 'EmptyVerySmallSquare;': '\u25ab', + 'emsp13;': '\u2004', + 'emsp14;': '\u2005', + 'emsp;': '\u2003', + 'ENG;': '\u014a', + 'eng;': '\u014b', + 'ensp;': '\u2002', + 'Eogon;': '\u0118', + 'eogon;': '\u0119', + 'Eopf;': '\U0001d53c', + 'eopf;': '\U0001d556', + 'epar;': '\u22d5', + 'eparsl;': '\u29e3', + 'eplus;': '\u2a71', + 'epsi;': '\u03b5', + 'Epsilon;': '\u0395', + 'epsilon;': '\u03b5', + 'epsiv;': '\u03f5', + 'eqcirc;': '\u2256', + 'eqcolon;': '\u2255', + 'eqsim;': '\u2242', + 'eqslantgtr;': '\u2a96', + 'eqslantless;': '\u2a95', + 'Equal;': '\u2a75', + 'equals;': '=', + 'EqualTilde;': '\u2242', + 'equest;': '\u225f', + 'Equilibrium;': '\u21cc', + 'equiv;': '\u2261', + 'equivDD;': '\u2a78', + 'eqvparsl;': '\u29e5', + 'erarr;': '\u2971', + 'erDot;': '\u2253', + 'Escr;': '\u2130', + 'escr;': '\u212f', + 'esdot;': '\u2250', + 'Esim;': '\u2a73', + 'esim;': '\u2242', + 'Eta;': '\u0397', + 'eta;': '\u03b7', + 'ETH': '\xd0', + 'eth': '\xf0', + 'ETH;': '\xd0', + 'eth;': '\xf0', + 'Euml': '\xcb', + 'euml': '\xeb', + 'Euml;': '\xcb', + 'euml;': '\xeb', + 'euro;': '\u20ac', + 'excl;': '!', + 'exist;': '\u2203', + 'Exists;': '\u2203', + 'expectation;': '\u2130', + 'ExponentialE;': '\u2147', + 'exponentiale;': '\u2147', + 'fallingdotseq;': '\u2252', + 'Fcy;': '\u0424', + 'fcy;': '\u0444', + 'female;': '\u2640', + 'ffilig;': '\ufb03', + 'fflig;': '\ufb00', + 'ffllig;': '\ufb04', + 'Ffr;': '\U0001d509', + 'ffr;': '\U0001d523', + 'filig;': '\ufb01', + 'FilledSmallSquare;': '\u25fc', + 'FilledVerySmallSquare;': '\u25aa', + 'fjlig;': 'fj', + 'flat;': '\u266d', + 'fllig;': '\ufb02', + 'fltns;': '\u25b1', + 'fnof;': '\u0192', + 'Fopf;': '\U0001d53d', + 'fopf;': '\U0001d557', + 'ForAll;': '\u2200', + 'forall;': '\u2200', + 'fork;': '\u22d4', + 'forkv;': '\u2ad9', + 'Fouriertrf;': '\u2131', + 'fpartint;': '\u2a0d', + 'frac12': '\xbd', + 'frac12;': '\xbd', + 'frac13;': '\u2153', + 'frac14': '\xbc', + 'frac14;': '\xbc', + 'frac15;': '\u2155', + 'frac16;': '\u2159', + 'frac18;': '\u215b', + 'frac23;': '\u2154', + 'frac25;': '\u2156', + 'frac34': '\xbe', + 'frac34;': '\xbe', + 'frac35;': '\u2157', + 'frac38;': '\u215c', + 'frac45;': '\u2158', + 'frac56;': '\u215a', + 'frac58;': '\u215d', + 'frac78;': '\u215e', + 'frasl;': '\u2044', + 'frown;': '\u2322', + 'Fscr;': '\u2131', + 'fscr;': '\U0001d4bb', + 'gacute;': '\u01f5', + 'Gamma;': '\u0393', + 'gamma;': '\u03b3', + 'Gammad;': '\u03dc', + 'gammad;': '\u03dd', + 'gap;': '\u2a86', + 'Gbreve;': '\u011e', + 'gbreve;': '\u011f', + 'Gcedil;': '\u0122', + 'Gcirc;': '\u011c', + 'gcirc;': '\u011d', + 'Gcy;': '\u0413', + 'gcy;': '\u0433', + 'Gdot;': '\u0120', + 'gdot;': '\u0121', + 'gE;': '\u2267', + 'ge;': '\u2265', + 'gEl;': '\u2a8c', + 'gel;': '\u22db', + 'geq;': '\u2265', + 'geqq;': '\u2267', + 'geqslant;': '\u2a7e', + 'ges;': '\u2a7e', + 'gescc;': '\u2aa9', + 'gesdot;': '\u2a80', + 'gesdoto;': '\u2a82', + 'gesdotol;': '\u2a84', + 'gesl;': '\u22db\ufe00', + 'gesles;': '\u2a94', + 'Gfr;': '\U0001d50a', + 'gfr;': '\U0001d524', + 'Gg;': '\u22d9', + 'gg;': '\u226b', + 'ggg;': '\u22d9', + 'gimel;': '\u2137', + 'GJcy;': '\u0403', + 'gjcy;': '\u0453', + 'gl;': '\u2277', + 'gla;': '\u2aa5', + 'glE;': '\u2a92', + 'glj;': '\u2aa4', + 'gnap;': '\u2a8a', + 'gnapprox;': '\u2a8a', + 'gnE;': '\u2269', + 'gne;': '\u2a88', + 'gneq;': '\u2a88', + 'gneqq;': '\u2269', + 'gnsim;': '\u22e7', + 'Gopf;': '\U0001d53e', + 'gopf;': '\U0001d558', + 'grave;': '`', + 'GreaterEqual;': '\u2265', + 'GreaterEqualLess;': '\u22db', + 'GreaterFullEqual;': '\u2267', + 'GreaterGreater;': '\u2aa2', + 'GreaterLess;': '\u2277', + 'GreaterSlantEqual;': '\u2a7e', + 'GreaterTilde;': '\u2273', + 'Gscr;': '\U0001d4a2', + 'gscr;': '\u210a', + 'gsim;': '\u2273', + 'gsime;': '\u2a8e', + 'gsiml;': '\u2a90', + 'GT': '>', + 'gt': '>', + 'GT;': '>', + 'Gt;': '\u226b', + 'gt;': '>', + 'gtcc;': '\u2aa7', + 'gtcir;': '\u2a7a', + 'gtdot;': '\u22d7', + 'gtlPar;': '\u2995', + 'gtquest;': '\u2a7c', + 'gtrapprox;': '\u2a86', + 'gtrarr;': '\u2978', + 'gtrdot;': '\u22d7', + 'gtreqless;': '\u22db', + 'gtreqqless;': '\u2a8c', + 'gtrless;': '\u2277', + 'gtrsim;': '\u2273', + 'gvertneqq;': '\u2269\ufe00', + 'gvnE;': '\u2269\ufe00', + 'Hacek;': '\u02c7', + 'hairsp;': '\u200a', + 'half;': '\xbd', + 'hamilt;': '\u210b', + 'HARDcy;': '\u042a', + 'hardcy;': '\u044a', + 'hArr;': '\u21d4', + 'harr;': '\u2194', + 'harrcir;': '\u2948', + 'harrw;': '\u21ad', + 'Hat;': '^', + 'hbar;': '\u210f', + 'Hcirc;': '\u0124', + 'hcirc;': '\u0125', + 'hearts;': '\u2665', + 'heartsuit;': '\u2665', + 'hellip;': '\u2026', + 'hercon;': '\u22b9', + 'Hfr;': '\u210c', + 'hfr;': '\U0001d525', + 'HilbertSpace;': '\u210b', + 'hksearow;': '\u2925', + 'hkswarow;': '\u2926', + 'hoarr;': '\u21ff', + 'homtht;': '\u223b', + 'hookleftarrow;': '\u21a9', + 'hookrightarrow;': '\u21aa', + 'Hopf;': '\u210d', + 'hopf;': '\U0001d559', + 'horbar;': '\u2015', + 'HorizontalLine;': '\u2500', + 'Hscr;': '\u210b', + 'hscr;': '\U0001d4bd', + 'hslash;': '\u210f', + 'Hstrok;': '\u0126', + 'hstrok;': '\u0127', + 'HumpDownHump;': '\u224e', + 'HumpEqual;': '\u224f', + 'hybull;': '\u2043', + 'hyphen;': '\u2010', + 'Iacute': '\xcd', + 'iacute': '\xed', + 'Iacute;': '\xcd', + 'iacute;': '\xed', + 'ic;': '\u2063', + 'Icirc': '\xce', + 'icirc': '\xee', + 'Icirc;': '\xce', + 'icirc;': '\xee', + 'Icy;': '\u0418', + 'icy;': '\u0438', + 'Idot;': '\u0130', + 'IEcy;': '\u0415', + 'iecy;': '\u0435', + 'iexcl': '\xa1', + 'iexcl;': '\xa1', + 'iff;': '\u21d4', + 'Ifr;': '\u2111', + 'ifr;': '\U0001d526', + 'Igrave': '\xcc', + 'igrave': '\xec', + 'Igrave;': '\xcc', + 'igrave;': '\xec', + 'ii;': '\u2148', + 'iiiint;': '\u2a0c', + 'iiint;': '\u222d', + 'iinfin;': '\u29dc', + 'iiota;': '\u2129', + 'IJlig;': '\u0132', + 'ijlig;': '\u0133', + 'Im;': '\u2111', + 'Imacr;': '\u012a', + 'imacr;': '\u012b', + 'image;': '\u2111', + 'ImaginaryI;': '\u2148', + 'imagline;': '\u2110', + 'imagpart;': '\u2111', + 'imath;': '\u0131', + 'imof;': '\u22b7', + 'imped;': '\u01b5', + 'Implies;': '\u21d2', + 'in;': '\u2208', + 'incare;': '\u2105', + 'infin;': '\u221e', + 'infintie;': '\u29dd', + 'inodot;': '\u0131', + 'Int;': '\u222c', + 'int;': '\u222b', + 'intcal;': '\u22ba', + 'integers;': '\u2124', + 'Integral;': '\u222b', + 'intercal;': '\u22ba', + 'Intersection;': '\u22c2', + 'intlarhk;': '\u2a17', + 'intprod;': '\u2a3c', + 'InvisibleComma;': '\u2063', + 'InvisibleTimes;': '\u2062', + 'IOcy;': '\u0401', + 'iocy;': '\u0451', + 'Iogon;': '\u012e', + 'iogon;': '\u012f', + 'Iopf;': '\U0001d540', + 'iopf;': '\U0001d55a', + 'Iota;': '\u0399', + 'iota;': '\u03b9', + 'iprod;': '\u2a3c', + 'iquest': '\xbf', + 'iquest;': '\xbf', + 'Iscr;': '\u2110', + 'iscr;': '\U0001d4be', + 'isin;': '\u2208', + 'isindot;': '\u22f5', + 'isinE;': '\u22f9', + 'isins;': '\u22f4', + 'isinsv;': '\u22f3', + 'isinv;': '\u2208', + 'it;': '\u2062', + 'Itilde;': '\u0128', + 'itilde;': '\u0129', + 'Iukcy;': '\u0406', + 'iukcy;': '\u0456', + 'Iuml': '\xcf', + 'iuml': '\xef', + 'Iuml;': '\xcf', + 'iuml;': '\xef', + 'Jcirc;': '\u0134', + 'jcirc;': '\u0135', + 'Jcy;': '\u0419', + 'jcy;': '\u0439', + 'Jfr;': '\U0001d50d', + 'jfr;': '\U0001d527', + 'jmath;': '\u0237', + 'Jopf;': '\U0001d541', + 'jopf;': '\U0001d55b', + 'Jscr;': '\U0001d4a5', + 'jscr;': '\U0001d4bf', + 'Jsercy;': '\u0408', + 'jsercy;': '\u0458', + 'Jukcy;': '\u0404', + 'jukcy;': '\u0454', + 'Kappa;': '\u039a', + 'kappa;': '\u03ba', + 'kappav;': '\u03f0', + 'Kcedil;': '\u0136', + 'kcedil;': '\u0137', + 'Kcy;': '\u041a', + 'kcy;': '\u043a', + 'Kfr;': '\U0001d50e', + 'kfr;': '\U0001d528', + 'kgreen;': '\u0138', + 'KHcy;': '\u0425', + 'khcy;': '\u0445', + 'KJcy;': '\u040c', + 'kjcy;': '\u045c', + 'Kopf;': '\U0001d542', + 'kopf;': '\U0001d55c', + 'Kscr;': '\U0001d4a6', + 'kscr;': '\U0001d4c0', + 'lAarr;': '\u21da', + 'Lacute;': '\u0139', + 'lacute;': '\u013a', + 'laemptyv;': '\u29b4', + 'lagran;': '\u2112', + 'Lambda;': '\u039b', + 'lambda;': '\u03bb', + 'Lang;': '\u27ea', + 'lang;': '\u27e8', + 'langd;': '\u2991', + 'langle;': '\u27e8', + 'lap;': '\u2a85', + 'Laplacetrf;': '\u2112', + 'laquo': '\xab', + 'laquo;': '\xab', + 'Larr;': '\u219e', + 'lArr;': '\u21d0', + 'larr;': '\u2190', + 'larrb;': '\u21e4', + 'larrbfs;': '\u291f', + 'larrfs;': '\u291d', + 'larrhk;': '\u21a9', + 'larrlp;': '\u21ab', + 'larrpl;': '\u2939', + 'larrsim;': '\u2973', + 'larrtl;': '\u21a2', + 'lat;': '\u2aab', + 'lAtail;': '\u291b', + 'latail;': '\u2919', + 'late;': '\u2aad', + 'lates;': '\u2aad\ufe00', + 'lBarr;': '\u290e', + 'lbarr;': '\u290c', + 'lbbrk;': '\u2772', + 'lbrace;': '{', + 'lbrack;': '[', + 'lbrke;': '\u298b', + 'lbrksld;': '\u298f', + 'lbrkslu;': '\u298d', + 'Lcaron;': '\u013d', + 'lcaron;': '\u013e', + 'Lcedil;': '\u013b', + 'lcedil;': '\u013c', + 'lceil;': '\u2308', + 'lcub;': '{', + 'Lcy;': '\u041b', + 'lcy;': '\u043b', + 'ldca;': '\u2936', + 'ldquo;': '\u201c', + 'ldquor;': '\u201e', + 'ldrdhar;': '\u2967', + 'ldrushar;': '\u294b', + 'ldsh;': '\u21b2', + 'lE;': '\u2266', + 'le;': '\u2264', + 'LeftAngleBracket;': '\u27e8', + 'LeftArrow;': '\u2190', + 'Leftarrow;': '\u21d0', + 'leftarrow;': '\u2190', + 'LeftArrowBar;': '\u21e4', + 'LeftArrowRightArrow;': '\u21c6', + 'leftarrowtail;': '\u21a2', + 'LeftCeiling;': '\u2308', + 'LeftDoubleBracket;': '\u27e6', + 'LeftDownTeeVector;': '\u2961', + 'LeftDownVector;': '\u21c3', + 'LeftDownVectorBar;': '\u2959', + 'LeftFloor;': '\u230a', + 'leftharpoondown;': '\u21bd', + 'leftharpoonup;': '\u21bc', + 'leftleftarrows;': '\u21c7', + 'LeftRightArrow;': '\u2194', + 'Leftrightarrow;': '\u21d4', + 'leftrightarrow;': '\u2194', + 'leftrightarrows;': '\u21c6', + 'leftrightharpoons;': '\u21cb', + 'leftrightsquigarrow;': '\u21ad', + 'LeftRightVector;': '\u294e', + 'LeftTee;': '\u22a3', + 'LeftTeeArrow;': '\u21a4', + 'LeftTeeVector;': '\u295a', + 'leftthreetimes;': '\u22cb', + 'LeftTriangle;': '\u22b2', + 'LeftTriangleBar;': '\u29cf', + 'LeftTriangleEqual;': '\u22b4', + 'LeftUpDownVector;': '\u2951', + 'LeftUpTeeVector;': '\u2960', + 'LeftUpVector;': '\u21bf', + 'LeftUpVectorBar;': '\u2958', + 'LeftVector;': '\u21bc', + 'LeftVectorBar;': '\u2952', + 'lEg;': '\u2a8b', + 'leg;': '\u22da', + 'leq;': '\u2264', + 'leqq;': '\u2266', + 'leqslant;': '\u2a7d', + 'les;': '\u2a7d', + 'lescc;': '\u2aa8', + 'lesdot;': '\u2a7f', + 'lesdoto;': '\u2a81', + 'lesdotor;': '\u2a83', + 'lesg;': '\u22da\ufe00', + 'lesges;': '\u2a93', + 'lessapprox;': '\u2a85', + 'lessdot;': '\u22d6', + 'lesseqgtr;': '\u22da', + 'lesseqqgtr;': '\u2a8b', + 'LessEqualGreater;': '\u22da', + 'LessFullEqual;': '\u2266', + 'LessGreater;': '\u2276', + 'lessgtr;': '\u2276', + 'LessLess;': '\u2aa1', + 'lesssim;': '\u2272', + 'LessSlantEqual;': '\u2a7d', + 'LessTilde;': '\u2272', + 'lfisht;': '\u297c', + 'lfloor;': '\u230a', + 'Lfr;': '\U0001d50f', + 'lfr;': '\U0001d529', + 'lg;': '\u2276', + 'lgE;': '\u2a91', + 'lHar;': '\u2962', + 'lhard;': '\u21bd', + 'lharu;': '\u21bc', + 'lharul;': '\u296a', + 'lhblk;': '\u2584', + 'LJcy;': '\u0409', + 'ljcy;': '\u0459', + 'Ll;': '\u22d8', + 'll;': '\u226a', + 'llarr;': '\u21c7', + 'llcorner;': '\u231e', + 'Lleftarrow;': '\u21da', + 'llhard;': '\u296b', + 'lltri;': '\u25fa', + 'Lmidot;': '\u013f', + 'lmidot;': '\u0140', + 'lmoust;': '\u23b0', + 'lmoustache;': '\u23b0', + 'lnap;': '\u2a89', + 'lnapprox;': '\u2a89', + 'lnE;': '\u2268', + 'lne;': '\u2a87', + 'lneq;': '\u2a87', + 'lneqq;': '\u2268', + 'lnsim;': '\u22e6', + 'loang;': '\u27ec', + 'loarr;': '\u21fd', + 'lobrk;': '\u27e6', + 'LongLeftArrow;': '\u27f5', + 'Longleftarrow;': '\u27f8', + 'longleftarrow;': '\u27f5', + 'LongLeftRightArrow;': '\u27f7', + 'Longleftrightarrow;': '\u27fa', + 'longleftrightarrow;': '\u27f7', + 'longmapsto;': '\u27fc', + 'LongRightArrow;': '\u27f6', + 'Longrightarrow;': '\u27f9', + 'longrightarrow;': '\u27f6', + 'looparrowleft;': '\u21ab', + 'looparrowright;': '\u21ac', + 'lopar;': '\u2985', + 'Lopf;': '\U0001d543', + 'lopf;': '\U0001d55d', + 'loplus;': '\u2a2d', + 'lotimes;': '\u2a34', + 'lowast;': '\u2217', + 'lowbar;': '_', + 'LowerLeftArrow;': '\u2199', + 'LowerRightArrow;': '\u2198', + 'loz;': '\u25ca', + 'lozenge;': '\u25ca', + 'lozf;': '\u29eb', + 'lpar;': '(', + 'lparlt;': '\u2993', + 'lrarr;': '\u21c6', + 'lrcorner;': '\u231f', + 'lrhar;': '\u21cb', + 'lrhard;': '\u296d', + 'lrm;': '\u200e', + 'lrtri;': '\u22bf', + 'lsaquo;': '\u2039', + 'Lscr;': '\u2112', + 'lscr;': '\U0001d4c1', + 'Lsh;': '\u21b0', + 'lsh;': '\u21b0', + 'lsim;': '\u2272', + 'lsime;': '\u2a8d', + 'lsimg;': '\u2a8f', + 'lsqb;': '[', + 'lsquo;': '\u2018', + 'lsquor;': '\u201a', + 'Lstrok;': '\u0141', + 'lstrok;': '\u0142', + 'LT': '<', + 'lt': '<', + 'LT;': '<', + 'Lt;': '\u226a', + 'lt;': '<', + 'ltcc;': '\u2aa6', + 'ltcir;': '\u2a79', + 'ltdot;': '\u22d6', + 'lthree;': '\u22cb', + 'ltimes;': '\u22c9', + 'ltlarr;': '\u2976', + 'ltquest;': '\u2a7b', + 'ltri;': '\u25c3', + 'ltrie;': '\u22b4', + 'ltrif;': '\u25c2', + 'ltrPar;': '\u2996', + 'lurdshar;': '\u294a', + 'luruhar;': '\u2966', + 'lvertneqq;': '\u2268\ufe00', + 'lvnE;': '\u2268\ufe00', + 'macr': '\xaf', + 'macr;': '\xaf', + 'male;': '\u2642', + 'malt;': '\u2720', + 'maltese;': '\u2720', + 'Map;': '\u2905', + 'map;': '\u21a6', + 'mapsto;': '\u21a6', + 'mapstodown;': '\u21a7', + 'mapstoleft;': '\u21a4', + 'mapstoup;': '\u21a5', + 'marker;': '\u25ae', + 'mcomma;': '\u2a29', + 'Mcy;': '\u041c', + 'mcy;': '\u043c', + 'mdash;': '\u2014', + 'mDDot;': '\u223a', + 'measuredangle;': '\u2221', + 'MediumSpace;': '\u205f', + 'Mellintrf;': '\u2133', + 'Mfr;': '\U0001d510', + 'mfr;': '\U0001d52a', + 'mho;': '\u2127', + 'micro': '\xb5', + 'micro;': '\xb5', + 'mid;': '\u2223', + 'midast;': '*', + 'midcir;': '\u2af0', + 'middot': '\xb7', + 'middot;': '\xb7', + 'minus;': '\u2212', + 'minusb;': '\u229f', + 'minusd;': '\u2238', + 'minusdu;': '\u2a2a', + 'MinusPlus;': '\u2213', + 'mlcp;': '\u2adb', + 'mldr;': '\u2026', + 'mnplus;': '\u2213', + 'models;': '\u22a7', + 'Mopf;': '\U0001d544', + 'mopf;': '\U0001d55e', + 'mp;': '\u2213', + 'Mscr;': '\u2133', + 'mscr;': '\U0001d4c2', + 'mstpos;': '\u223e', + 'Mu;': '\u039c', + 'mu;': '\u03bc', + 'multimap;': '\u22b8', + 'mumap;': '\u22b8', + 'nabla;': '\u2207', + 'Nacute;': '\u0143', + 'nacute;': '\u0144', + 'nang;': '\u2220\u20d2', + 'nap;': '\u2249', + 'napE;': '\u2a70\u0338', + 'napid;': '\u224b\u0338', + 'napos;': '\u0149', + 'napprox;': '\u2249', + 'natur;': '\u266e', + 'natural;': '\u266e', + 'naturals;': '\u2115', + 'nbsp': '\xa0', + 'nbsp;': '\xa0', + 'nbump;': '\u224e\u0338', + 'nbumpe;': '\u224f\u0338', + 'ncap;': '\u2a43', + 'Ncaron;': '\u0147', + 'ncaron;': '\u0148', + 'Ncedil;': '\u0145', + 'ncedil;': '\u0146', + 'ncong;': '\u2247', + 'ncongdot;': '\u2a6d\u0338', + 'ncup;': '\u2a42', + 'Ncy;': '\u041d', + 'ncy;': '\u043d', + 'ndash;': '\u2013', + 'ne;': '\u2260', + 'nearhk;': '\u2924', + 'neArr;': '\u21d7', + 'nearr;': '\u2197', + 'nearrow;': '\u2197', + 'nedot;': '\u2250\u0338', + 'NegativeMediumSpace;': '\u200b', + 'NegativeThickSpace;': '\u200b', + 'NegativeThinSpace;': '\u200b', + 'NegativeVeryThinSpace;': '\u200b', + 'nequiv;': '\u2262', + 'nesear;': '\u2928', + 'nesim;': '\u2242\u0338', + 'NestedGreaterGreater;': '\u226b', + 'NestedLessLess;': '\u226a', + 'NewLine;': '\n', + 'nexist;': '\u2204', + 'nexists;': '\u2204', + 'Nfr;': '\U0001d511', + 'nfr;': '\U0001d52b', + 'ngE;': '\u2267\u0338', + 'nge;': '\u2271', + 'ngeq;': '\u2271', + 'ngeqq;': '\u2267\u0338', + 'ngeqslant;': '\u2a7e\u0338', + 'nges;': '\u2a7e\u0338', + 'nGg;': '\u22d9\u0338', + 'ngsim;': '\u2275', + 'nGt;': '\u226b\u20d2', + 'ngt;': '\u226f', + 'ngtr;': '\u226f', + 'nGtv;': '\u226b\u0338', + 'nhArr;': '\u21ce', + 'nharr;': '\u21ae', + 'nhpar;': '\u2af2', + 'ni;': '\u220b', + 'nis;': '\u22fc', + 'nisd;': '\u22fa', + 'niv;': '\u220b', + 'NJcy;': '\u040a', + 'njcy;': '\u045a', + 'nlArr;': '\u21cd', + 'nlarr;': '\u219a', + 'nldr;': '\u2025', + 'nlE;': '\u2266\u0338', + 'nle;': '\u2270', + 'nLeftarrow;': '\u21cd', + 'nleftarrow;': '\u219a', + 'nLeftrightarrow;': '\u21ce', + 'nleftrightarrow;': '\u21ae', + 'nleq;': '\u2270', + 'nleqq;': '\u2266\u0338', + 'nleqslant;': '\u2a7d\u0338', + 'nles;': '\u2a7d\u0338', + 'nless;': '\u226e', + 'nLl;': '\u22d8\u0338', + 'nlsim;': '\u2274', + 'nLt;': '\u226a\u20d2', + 'nlt;': '\u226e', + 'nltri;': '\u22ea', + 'nltrie;': '\u22ec', + 'nLtv;': '\u226a\u0338', + 'nmid;': '\u2224', + 'NoBreak;': '\u2060', + 'NonBreakingSpace;': '\xa0', + 'Nopf;': '\u2115', + 'nopf;': '\U0001d55f', + 'not': '\xac', + 'Not;': '\u2aec', + 'not;': '\xac', + 'NotCongruent;': '\u2262', + 'NotCupCap;': '\u226d', + 'NotDoubleVerticalBar;': '\u2226', + 'NotElement;': '\u2209', + 'NotEqual;': '\u2260', + 'NotEqualTilde;': '\u2242\u0338', + 'NotExists;': '\u2204', + 'NotGreater;': '\u226f', + 'NotGreaterEqual;': '\u2271', + 'NotGreaterFullEqual;': '\u2267\u0338', + 'NotGreaterGreater;': '\u226b\u0338', + 'NotGreaterLess;': '\u2279', + 'NotGreaterSlantEqual;': '\u2a7e\u0338', + 'NotGreaterTilde;': '\u2275', + 'NotHumpDownHump;': '\u224e\u0338', + 'NotHumpEqual;': '\u224f\u0338', + 'notin;': '\u2209', + 'notindot;': '\u22f5\u0338', + 'notinE;': '\u22f9\u0338', + 'notinva;': '\u2209', + 'notinvb;': '\u22f7', + 'notinvc;': '\u22f6', + 'NotLeftTriangle;': '\u22ea', + 'NotLeftTriangleBar;': '\u29cf\u0338', + 'NotLeftTriangleEqual;': '\u22ec', + 'NotLess;': '\u226e', + 'NotLessEqual;': '\u2270', + 'NotLessGreater;': '\u2278', + 'NotLessLess;': '\u226a\u0338', + 'NotLessSlantEqual;': '\u2a7d\u0338', + 'NotLessTilde;': '\u2274', + 'NotNestedGreaterGreater;': '\u2aa2\u0338', + 'NotNestedLessLess;': '\u2aa1\u0338', + 'notni;': '\u220c', + 'notniva;': '\u220c', + 'notnivb;': '\u22fe', + 'notnivc;': '\u22fd', + 'NotPrecedes;': '\u2280', + 'NotPrecedesEqual;': '\u2aaf\u0338', + 'NotPrecedesSlantEqual;': '\u22e0', + 'NotReverseElement;': '\u220c', + 'NotRightTriangle;': '\u22eb', + 'NotRightTriangleBar;': '\u29d0\u0338', + 'NotRightTriangleEqual;': '\u22ed', + 'NotSquareSubset;': '\u228f\u0338', + 'NotSquareSubsetEqual;': '\u22e2', + 'NotSquareSuperset;': '\u2290\u0338', + 'NotSquareSupersetEqual;': '\u22e3', + 'NotSubset;': '\u2282\u20d2', + 'NotSubsetEqual;': '\u2288', + 'NotSucceeds;': '\u2281', + 'NotSucceedsEqual;': '\u2ab0\u0338', + 'NotSucceedsSlantEqual;': '\u22e1', + 'NotSucceedsTilde;': '\u227f\u0338', + 'NotSuperset;': '\u2283\u20d2', + 'NotSupersetEqual;': '\u2289', + 'NotTilde;': '\u2241', + 'NotTildeEqual;': '\u2244', + 'NotTildeFullEqual;': '\u2247', + 'NotTildeTilde;': '\u2249', + 'NotVerticalBar;': '\u2224', + 'npar;': '\u2226', + 'nparallel;': '\u2226', + 'nparsl;': '\u2afd\u20e5', + 'npart;': '\u2202\u0338', + 'npolint;': '\u2a14', + 'npr;': '\u2280', + 'nprcue;': '\u22e0', + 'npre;': '\u2aaf\u0338', + 'nprec;': '\u2280', + 'npreceq;': '\u2aaf\u0338', + 'nrArr;': '\u21cf', + 'nrarr;': '\u219b', + 'nrarrc;': '\u2933\u0338', + 'nrarrw;': '\u219d\u0338', + 'nRightarrow;': '\u21cf', + 'nrightarrow;': '\u219b', + 'nrtri;': '\u22eb', + 'nrtrie;': '\u22ed', + 'nsc;': '\u2281', + 'nsccue;': '\u22e1', + 'nsce;': '\u2ab0\u0338', + 'Nscr;': '\U0001d4a9', + 'nscr;': '\U0001d4c3', + 'nshortmid;': '\u2224', + 'nshortparallel;': '\u2226', + 'nsim;': '\u2241', + 'nsime;': '\u2244', + 'nsimeq;': '\u2244', + 'nsmid;': '\u2224', + 'nspar;': '\u2226', + 'nsqsube;': '\u22e2', + 'nsqsupe;': '\u22e3', + 'nsub;': '\u2284', + 'nsubE;': '\u2ac5\u0338', + 'nsube;': '\u2288', + 'nsubset;': '\u2282\u20d2', + 'nsubseteq;': '\u2288', + 'nsubseteqq;': '\u2ac5\u0338', + 'nsucc;': '\u2281', + 'nsucceq;': '\u2ab0\u0338', + 'nsup;': '\u2285', + 'nsupE;': '\u2ac6\u0338', + 'nsupe;': '\u2289', + 'nsupset;': '\u2283\u20d2', + 'nsupseteq;': '\u2289', + 'nsupseteqq;': '\u2ac6\u0338', + 'ntgl;': '\u2279', + 'Ntilde': '\xd1', + 'ntilde': '\xf1', + 'Ntilde;': '\xd1', + 'ntilde;': '\xf1', + 'ntlg;': '\u2278', + 'ntriangleleft;': '\u22ea', + 'ntrianglelefteq;': '\u22ec', + 'ntriangleright;': '\u22eb', + 'ntrianglerighteq;': '\u22ed', + 'Nu;': '\u039d', + 'nu;': '\u03bd', + 'num;': '#', + 'numero;': '\u2116', + 'numsp;': '\u2007', + 'nvap;': '\u224d\u20d2', + 'nVDash;': '\u22af', + 'nVdash;': '\u22ae', + 'nvDash;': '\u22ad', + 'nvdash;': '\u22ac', + 'nvge;': '\u2265\u20d2', + 'nvgt;': '>\u20d2', + 'nvHarr;': '\u2904', + 'nvinfin;': '\u29de', + 'nvlArr;': '\u2902', + 'nvle;': '\u2264\u20d2', + 'nvlt;': '<\u20d2', + 'nvltrie;': '\u22b4\u20d2', + 'nvrArr;': '\u2903', + 'nvrtrie;': '\u22b5\u20d2', + 'nvsim;': '\u223c\u20d2', + 'nwarhk;': '\u2923', + 'nwArr;': '\u21d6', + 'nwarr;': '\u2196', + 'nwarrow;': '\u2196', + 'nwnear;': '\u2927', + 'Oacute': '\xd3', + 'oacute': '\xf3', + 'Oacute;': '\xd3', + 'oacute;': '\xf3', + 'oast;': '\u229b', + 'ocir;': '\u229a', + 'Ocirc': '\xd4', + 'ocirc': '\xf4', + 'Ocirc;': '\xd4', + 'ocirc;': '\xf4', + 'Ocy;': '\u041e', + 'ocy;': '\u043e', + 'odash;': '\u229d', + 'Odblac;': '\u0150', + 'odblac;': '\u0151', + 'odiv;': '\u2a38', + 'odot;': '\u2299', + 'odsold;': '\u29bc', + 'OElig;': '\u0152', + 'oelig;': '\u0153', + 'ofcir;': '\u29bf', + 'Ofr;': '\U0001d512', + 'ofr;': '\U0001d52c', + 'ogon;': '\u02db', + 'Ograve': '\xd2', + 'ograve': '\xf2', + 'Ograve;': '\xd2', + 'ograve;': '\xf2', + 'ogt;': '\u29c1', + 'ohbar;': '\u29b5', + 'ohm;': '\u03a9', + 'oint;': '\u222e', + 'olarr;': '\u21ba', + 'olcir;': '\u29be', + 'olcross;': '\u29bb', + 'oline;': '\u203e', + 'olt;': '\u29c0', + 'Omacr;': '\u014c', + 'omacr;': '\u014d', + 'Omega;': '\u03a9', + 'omega;': '\u03c9', + 'Omicron;': '\u039f', + 'omicron;': '\u03bf', + 'omid;': '\u29b6', + 'ominus;': '\u2296', + 'Oopf;': '\U0001d546', + 'oopf;': '\U0001d560', + 'opar;': '\u29b7', + 'OpenCurlyDoubleQuote;': '\u201c', + 'OpenCurlyQuote;': '\u2018', + 'operp;': '\u29b9', + 'oplus;': '\u2295', + 'Or;': '\u2a54', + 'or;': '\u2228', + 'orarr;': '\u21bb', + 'ord;': '\u2a5d', + 'order;': '\u2134', + 'orderof;': '\u2134', + 'ordf': '\xaa', + 'ordf;': '\xaa', + 'ordm': '\xba', + 'ordm;': '\xba', + 'origof;': '\u22b6', + 'oror;': '\u2a56', + 'orslope;': '\u2a57', + 'orv;': '\u2a5b', + 'oS;': '\u24c8', + 'Oscr;': '\U0001d4aa', + 'oscr;': '\u2134', + 'Oslash': '\xd8', + 'oslash': '\xf8', + 'Oslash;': '\xd8', + 'oslash;': '\xf8', + 'osol;': '\u2298', + 'Otilde': '\xd5', + 'otilde': '\xf5', + 'Otilde;': '\xd5', + 'otilde;': '\xf5', + 'Otimes;': '\u2a37', + 'otimes;': '\u2297', + 'otimesas;': '\u2a36', + 'Ouml': '\xd6', + 'ouml': '\xf6', + 'Ouml;': '\xd6', + 'ouml;': '\xf6', + 'ovbar;': '\u233d', + 'OverBar;': '\u203e', + 'OverBrace;': '\u23de', + 'OverBracket;': '\u23b4', + 'OverParenthesis;': '\u23dc', + 'par;': '\u2225', + 'para': '\xb6', + 'para;': '\xb6', + 'parallel;': '\u2225', + 'parsim;': '\u2af3', + 'parsl;': '\u2afd', + 'part;': '\u2202', + 'PartialD;': '\u2202', + 'Pcy;': '\u041f', + 'pcy;': '\u043f', + 'percnt;': '%', + 'period;': '.', + 'permil;': '\u2030', + 'perp;': '\u22a5', + 'pertenk;': '\u2031', + 'Pfr;': '\U0001d513', + 'pfr;': '\U0001d52d', + 'Phi;': '\u03a6', + 'phi;': '\u03c6', + 'phiv;': '\u03d5', + 'phmmat;': '\u2133', + 'phone;': '\u260e', + 'Pi;': '\u03a0', + 'pi;': '\u03c0', + 'pitchfork;': '\u22d4', + 'piv;': '\u03d6', + 'planck;': '\u210f', + 'planckh;': '\u210e', + 'plankv;': '\u210f', + 'plus;': '+', + 'plusacir;': '\u2a23', + 'plusb;': '\u229e', + 'pluscir;': '\u2a22', + 'plusdo;': '\u2214', + 'plusdu;': '\u2a25', + 'pluse;': '\u2a72', + 'PlusMinus;': '\xb1', + 'plusmn': '\xb1', + 'plusmn;': '\xb1', + 'plussim;': '\u2a26', + 'plustwo;': '\u2a27', + 'pm;': '\xb1', + 'Poincareplane;': '\u210c', + 'pointint;': '\u2a15', + 'Popf;': '\u2119', + 'popf;': '\U0001d561', + 'pound': '\xa3', + 'pound;': '\xa3', + 'Pr;': '\u2abb', + 'pr;': '\u227a', + 'prap;': '\u2ab7', + 'prcue;': '\u227c', + 'prE;': '\u2ab3', + 'pre;': '\u2aaf', + 'prec;': '\u227a', + 'precapprox;': '\u2ab7', + 'preccurlyeq;': '\u227c', + 'Precedes;': '\u227a', + 'PrecedesEqual;': '\u2aaf', + 'PrecedesSlantEqual;': '\u227c', + 'PrecedesTilde;': '\u227e', + 'preceq;': '\u2aaf', + 'precnapprox;': '\u2ab9', + 'precneqq;': '\u2ab5', + 'precnsim;': '\u22e8', + 'precsim;': '\u227e', + 'Prime;': '\u2033', + 'prime;': '\u2032', + 'primes;': '\u2119', + 'prnap;': '\u2ab9', + 'prnE;': '\u2ab5', + 'prnsim;': '\u22e8', + 'prod;': '\u220f', + 'Product;': '\u220f', + 'profalar;': '\u232e', + 'profline;': '\u2312', + 'profsurf;': '\u2313', + 'prop;': '\u221d', + 'Proportion;': '\u2237', + 'Proportional;': '\u221d', + 'propto;': '\u221d', + 'prsim;': '\u227e', + 'prurel;': '\u22b0', + 'Pscr;': '\U0001d4ab', + 'pscr;': '\U0001d4c5', + 'Psi;': '\u03a8', + 'psi;': '\u03c8', + 'puncsp;': '\u2008', + 'Qfr;': '\U0001d514', + 'qfr;': '\U0001d52e', + 'qint;': '\u2a0c', + 'Qopf;': '\u211a', + 'qopf;': '\U0001d562', + 'qprime;': '\u2057', + 'Qscr;': '\U0001d4ac', + 'qscr;': '\U0001d4c6', + 'quaternions;': '\u210d', + 'quatint;': '\u2a16', + 'quest;': '?', + 'questeq;': '\u225f', + 'QUOT': '"', + 'quot': '"', + 'QUOT;': '"', + 'quot;': '"', + 'rAarr;': '\u21db', + 'race;': '\u223d\u0331', + 'Racute;': '\u0154', + 'racute;': '\u0155', + 'radic;': '\u221a', + 'raemptyv;': '\u29b3', + 'Rang;': '\u27eb', + 'rang;': '\u27e9', + 'rangd;': '\u2992', + 'range;': '\u29a5', + 'rangle;': '\u27e9', + 'raquo': '\xbb', + 'raquo;': '\xbb', + 'Rarr;': '\u21a0', + 'rArr;': '\u21d2', + 'rarr;': '\u2192', + 'rarrap;': '\u2975', + 'rarrb;': '\u21e5', + 'rarrbfs;': '\u2920', + 'rarrc;': '\u2933', + 'rarrfs;': '\u291e', + 'rarrhk;': '\u21aa', + 'rarrlp;': '\u21ac', + 'rarrpl;': '\u2945', + 'rarrsim;': '\u2974', + 'Rarrtl;': '\u2916', + 'rarrtl;': '\u21a3', + 'rarrw;': '\u219d', + 'rAtail;': '\u291c', + 'ratail;': '\u291a', + 'ratio;': '\u2236', + 'rationals;': '\u211a', + 'RBarr;': '\u2910', + 'rBarr;': '\u290f', + 'rbarr;': '\u290d', + 'rbbrk;': '\u2773', + 'rbrace;': '}', + 'rbrack;': ']', + 'rbrke;': '\u298c', + 'rbrksld;': '\u298e', + 'rbrkslu;': '\u2990', + 'Rcaron;': '\u0158', + 'rcaron;': '\u0159', + 'Rcedil;': '\u0156', + 'rcedil;': '\u0157', + 'rceil;': '\u2309', + 'rcub;': '}', + 'Rcy;': '\u0420', + 'rcy;': '\u0440', + 'rdca;': '\u2937', + 'rdldhar;': '\u2969', + 'rdquo;': '\u201d', + 'rdquor;': '\u201d', + 'rdsh;': '\u21b3', + 'Re;': '\u211c', + 'real;': '\u211c', + 'realine;': '\u211b', + 'realpart;': '\u211c', + 'reals;': '\u211d', + 'rect;': '\u25ad', + 'REG': '\xae', + 'reg': '\xae', + 'REG;': '\xae', + 'reg;': '\xae', + 'ReverseElement;': '\u220b', + 'ReverseEquilibrium;': '\u21cb', + 'ReverseUpEquilibrium;': '\u296f', + 'rfisht;': '\u297d', + 'rfloor;': '\u230b', + 'Rfr;': '\u211c', + 'rfr;': '\U0001d52f', + 'rHar;': '\u2964', + 'rhard;': '\u21c1', + 'rharu;': '\u21c0', + 'rharul;': '\u296c', + 'Rho;': '\u03a1', + 'rho;': '\u03c1', + 'rhov;': '\u03f1', + 'RightAngleBracket;': '\u27e9', + 'RightArrow;': '\u2192', + 'Rightarrow;': '\u21d2', + 'rightarrow;': '\u2192', + 'RightArrowBar;': '\u21e5', + 'RightArrowLeftArrow;': '\u21c4', + 'rightarrowtail;': '\u21a3', + 'RightCeiling;': '\u2309', + 'RightDoubleBracket;': '\u27e7', + 'RightDownTeeVector;': '\u295d', + 'RightDownVector;': '\u21c2', + 'RightDownVectorBar;': '\u2955', + 'RightFloor;': '\u230b', + 'rightharpoondown;': '\u21c1', + 'rightharpoonup;': '\u21c0', + 'rightleftarrows;': '\u21c4', + 'rightleftharpoons;': '\u21cc', + 'rightrightarrows;': '\u21c9', + 'rightsquigarrow;': '\u219d', + 'RightTee;': '\u22a2', + 'RightTeeArrow;': '\u21a6', + 'RightTeeVector;': '\u295b', + 'rightthreetimes;': '\u22cc', + 'RightTriangle;': '\u22b3', + 'RightTriangleBar;': '\u29d0', + 'RightTriangleEqual;': '\u22b5', + 'RightUpDownVector;': '\u294f', + 'RightUpTeeVector;': '\u295c', + 'RightUpVector;': '\u21be', + 'RightUpVectorBar;': '\u2954', + 'RightVector;': '\u21c0', + 'RightVectorBar;': '\u2953', + 'ring;': '\u02da', + 'risingdotseq;': '\u2253', + 'rlarr;': '\u21c4', + 'rlhar;': '\u21cc', + 'rlm;': '\u200f', + 'rmoust;': '\u23b1', + 'rmoustache;': '\u23b1', + 'rnmid;': '\u2aee', + 'roang;': '\u27ed', + 'roarr;': '\u21fe', + 'robrk;': '\u27e7', + 'ropar;': '\u2986', + 'Ropf;': '\u211d', + 'ropf;': '\U0001d563', + 'roplus;': '\u2a2e', + 'rotimes;': '\u2a35', + 'RoundImplies;': '\u2970', + 'rpar;': ')', + 'rpargt;': '\u2994', + 'rppolint;': '\u2a12', + 'rrarr;': '\u21c9', + 'Rrightarrow;': '\u21db', + 'rsaquo;': '\u203a', + 'Rscr;': '\u211b', + 'rscr;': '\U0001d4c7', + 'Rsh;': '\u21b1', + 'rsh;': '\u21b1', + 'rsqb;': ']', + 'rsquo;': '\u2019', + 'rsquor;': '\u2019', + 'rthree;': '\u22cc', + 'rtimes;': '\u22ca', + 'rtri;': '\u25b9', + 'rtrie;': '\u22b5', + 'rtrif;': '\u25b8', + 'rtriltri;': '\u29ce', + 'RuleDelayed;': '\u29f4', + 'ruluhar;': '\u2968', + 'rx;': '\u211e', + 'Sacute;': '\u015a', + 'sacute;': '\u015b', + 'sbquo;': '\u201a', + 'Sc;': '\u2abc', + 'sc;': '\u227b', + 'scap;': '\u2ab8', + 'Scaron;': '\u0160', + 'scaron;': '\u0161', + 'sccue;': '\u227d', + 'scE;': '\u2ab4', + 'sce;': '\u2ab0', + 'Scedil;': '\u015e', + 'scedil;': '\u015f', + 'Scirc;': '\u015c', + 'scirc;': '\u015d', + 'scnap;': '\u2aba', + 'scnE;': '\u2ab6', + 'scnsim;': '\u22e9', + 'scpolint;': '\u2a13', + 'scsim;': '\u227f', + 'Scy;': '\u0421', + 'scy;': '\u0441', + 'sdot;': '\u22c5', + 'sdotb;': '\u22a1', + 'sdote;': '\u2a66', + 'searhk;': '\u2925', + 'seArr;': '\u21d8', + 'searr;': '\u2198', + 'searrow;': '\u2198', + 'sect': '\xa7', + 'sect;': '\xa7', + 'semi;': ';', + 'seswar;': '\u2929', + 'setminus;': '\u2216', + 'setmn;': '\u2216', + 'sext;': '\u2736', + 'Sfr;': '\U0001d516', + 'sfr;': '\U0001d530', + 'sfrown;': '\u2322', + 'sharp;': '\u266f', + 'SHCHcy;': '\u0429', + 'shchcy;': '\u0449', + 'SHcy;': '\u0428', + 'shcy;': '\u0448', + 'ShortDownArrow;': '\u2193', + 'ShortLeftArrow;': '\u2190', + 'shortmid;': '\u2223', + 'shortparallel;': '\u2225', + 'ShortRightArrow;': '\u2192', + 'ShortUpArrow;': '\u2191', + 'shy': '\xad', + 'shy;': '\xad', + 'Sigma;': '\u03a3', + 'sigma;': '\u03c3', + 'sigmaf;': '\u03c2', + 'sigmav;': '\u03c2', + 'sim;': '\u223c', + 'simdot;': '\u2a6a', + 'sime;': '\u2243', + 'simeq;': '\u2243', + 'simg;': '\u2a9e', + 'simgE;': '\u2aa0', + 'siml;': '\u2a9d', + 'simlE;': '\u2a9f', + 'simne;': '\u2246', + 'simplus;': '\u2a24', + 'simrarr;': '\u2972', + 'slarr;': '\u2190', + 'SmallCircle;': '\u2218', + 'smallsetminus;': '\u2216', + 'smashp;': '\u2a33', + 'smeparsl;': '\u29e4', + 'smid;': '\u2223', + 'smile;': '\u2323', + 'smt;': '\u2aaa', + 'smte;': '\u2aac', + 'smtes;': '\u2aac\ufe00', + 'SOFTcy;': '\u042c', + 'softcy;': '\u044c', + 'sol;': '/', + 'solb;': '\u29c4', + 'solbar;': '\u233f', + 'Sopf;': '\U0001d54a', + 'sopf;': '\U0001d564', + 'spades;': '\u2660', + 'spadesuit;': '\u2660', + 'spar;': '\u2225', + 'sqcap;': '\u2293', + 'sqcaps;': '\u2293\ufe00', + 'sqcup;': '\u2294', + 'sqcups;': '\u2294\ufe00', + 'Sqrt;': '\u221a', + 'sqsub;': '\u228f', + 'sqsube;': '\u2291', + 'sqsubset;': '\u228f', + 'sqsubseteq;': '\u2291', + 'sqsup;': '\u2290', + 'sqsupe;': '\u2292', + 'sqsupset;': '\u2290', + 'sqsupseteq;': '\u2292', + 'squ;': '\u25a1', + 'Square;': '\u25a1', + 'square;': '\u25a1', + 'SquareIntersection;': '\u2293', + 'SquareSubset;': '\u228f', + 'SquareSubsetEqual;': '\u2291', + 'SquareSuperset;': '\u2290', + 'SquareSupersetEqual;': '\u2292', + 'SquareUnion;': '\u2294', + 'squarf;': '\u25aa', + 'squf;': '\u25aa', + 'srarr;': '\u2192', + 'Sscr;': '\U0001d4ae', + 'sscr;': '\U0001d4c8', + 'ssetmn;': '\u2216', + 'ssmile;': '\u2323', + 'sstarf;': '\u22c6', + 'Star;': '\u22c6', + 'star;': '\u2606', + 'starf;': '\u2605', + 'straightepsilon;': '\u03f5', + 'straightphi;': '\u03d5', + 'strns;': '\xaf', + 'Sub;': '\u22d0', + 'sub;': '\u2282', + 'subdot;': '\u2abd', + 'subE;': '\u2ac5', + 'sube;': '\u2286', + 'subedot;': '\u2ac3', + 'submult;': '\u2ac1', + 'subnE;': '\u2acb', + 'subne;': '\u228a', + 'subplus;': '\u2abf', + 'subrarr;': '\u2979', + 'Subset;': '\u22d0', + 'subset;': '\u2282', + 'subseteq;': '\u2286', + 'subseteqq;': '\u2ac5', + 'SubsetEqual;': '\u2286', + 'subsetneq;': '\u228a', + 'subsetneqq;': '\u2acb', + 'subsim;': '\u2ac7', + 'subsub;': '\u2ad5', + 'subsup;': '\u2ad3', + 'succ;': '\u227b', + 'succapprox;': '\u2ab8', + 'succcurlyeq;': '\u227d', + 'Succeeds;': '\u227b', + 'SucceedsEqual;': '\u2ab0', + 'SucceedsSlantEqual;': '\u227d', + 'SucceedsTilde;': '\u227f', + 'succeq;': '\u2ab0', + 'succnapprox;': '\u2aba', + 'succneqq;': '\u2ab6', + 'succnsim;': '\u22e9', + 'succsim;': '\u227f', + 'SuchThat;': '\u220b', + 'Sum;': '\u2211', + 'sum;': '\u2211', + 'sung;': '\u266a', + 'sup1': '\xb9', + 'sup1;': '\xb9', + 'sup2': '\xb2', + 'sup2;': '\xb2', + 'sup3': '\xb3', + 'sup3;': '\xb3', + 'Sup;': '\u22d1', + 'sup;': '\u2283', + 'supdot;': '\u2abe', + 'supdsub;': '\u2ad8', + 'supE;': '\u2ac6', + 'supe;': '\u2287', + 'supedot;': '\u2ac4', + 'Superset;': '\u2283', + 'SupersetEqual;': '\u2287', + 'suphsol;': '\u27c9', + 'suphsub;': '\u2ad7', + 'suplarr;': '\u297b', + 'supmult;': '\u2ac2', + 'supnE;': '\u2acc', + 'supne;': '\u228b', + 'supplus;': '\u2ac0', + 'Supset;': '\u22d1', + 'supset;': '\u2283', + 'supseteq;': '\u2287', + 'supseteqq;': '\u2ac6', + 'supsetneq;': '\u228b', + 'supsetneqq;': '\u2acc', + 'supsim;': '\u2ac8', + 'supsub;': '\u2ad4', + 'supsup;': '\u2ad6', + 'swarhk;': '\u2926', + 'swArr;': '\u21d9', + 'swarr;': '\u2199', + 'swarrow;': '\u2199', + 'swnwar;': '\u292a', + 'szlig': '\xdf', + 'szlig;': '\xdf', + 'Tab;': '\t', + 'target;': '\u2316', + 'Tau;': '\u03a4', + 'tau;': '\u03c4', + 'tbrk;': '\u23b4', + 'Tcaron;': '\u0164', + 'tcaron;': '\u0165', + 'Tcedil;': '\u0162', + 'tcedil;': '\u0163', + 'Tcy;': '\u0422', + 'tcy;': '\u0442', + 'tdot;': '\u20db', + 'telrec;': '\u2315', + 'Tfr;': '\U0001d517', + 'tfr;': '\U0001d531', + 'there4;': '\u2234', + 'Therefore;': '\u2234', + 'therefore;': '\u2234', + 'Theta;': '\u0398', + 'theta;': '\u03b8', + 'thetasym;': '\u03d1', + 'thetav;': '\u03d1', + 'thickapprox;': '\u2248', + 'thicksim;': '\u223c', + 'ThickSpace;': '\u205f\u200a', + 'thinsp;': '\u2009', + 'ThinSpace;': '\u2009', + 'thkap;': '\u2248', + 'thksim;': '\u223c', + 'THORN': '\xde', + 'thorn': '\xfe', + 'THORN;': '\xde', + 'thorn;': '\xfe', + 'Tilde;': '\u223c', + 'tilde;': '\u02dc', + 'TildeEqual;': '\u2243', + 'TildeFullEqual;': '\u2245', + 'TildeTilde;': '\u2248', + 'times': '\xd7', + 'times;': '\xd7', + 'timesb;': '\u22a0', + 'timesbar;': '\u2a31', + 'timesd;': '\u2a30', + 'tint;': '\u222d', + 'toea;': '\u2928', + 'top;': '\u22a4', + 'topbot;': '\u2336', + 'topcir;': '\u2af1', + 'Topf;': '\U0001d54b', + 'topf;': '\U0001d565', + 'topfork;': '\u2ada', + 'tosa;': '\u2929', + 'tprime;': '\u2034', + 'TRADE;': '\u2122', + 'trade;': '\u2122', + 'triangle;': '\u25b5', + 'triangledown;': '\u25bf', + 'triangleleft;': '\u25c3', + 'trianglelefteq;': '\u22b4', + 'triangleq;': '\u225c', + 'triangleright;': '\u25b9', + 'trianglerighteq;': '\u22b5', + 'tridot;': '\u25ec', + 'trie;': '\u225c', + 'triminus;': '\u2a3a', + 'TripleDot;': '\u20db', + 'triplus;': '\u2a39', + 'trisb;': '\u29cd', + 'tritime;': '\u2a3b', + 'trpezium;': '\u23e2', + 'Tscr;': '\U0001d4af', + 'tscr;': '\U0001d4c9', + 'TScy;': '\u0426', + 'tscy;': '\u0446', + 'TSHcy;': '\u040b', + 'tshcy;': '\u045b', + 'Tstrok;': '\u0166', + 'tstrok;': '\u0167', + 'twixt;': '\u226c', + 'twoheadleftarrow;': '\u219e', + 'twoheadrightarrow;': '\u21a0', + 'Uacute': '\xda', + 'uacute': '\xfa', + 'Uacute;': '\xda', + 'uacute;': '\xfa', + 'Uarr;': '\u219f', + 'uArr;': '\u21d1', + 'uarr;': '\u2191', + 'Uarrocir;': '\u2949', + 'Ubrcy;': '\u040e', + 'ubrcy;': '\u045e', + 'Ubreve;': '\u016c', + 'ubreve;': '\u016d', + 'Ucirc': '\xdb', + 'ucirc': '\xfb', + 'Ucirc;': '\xdb', + 'ucirc;': '\xfb', + 'Ucy;': '\u0423', + 'ucy;': '\u0443', + 'udarr;': '\u21c5', + 'Udblac;': '\u0170', + 'udblac;': '\u0171', + 'udhar;': '\u296e', + 'ufisht;': '\u297e', + 'Ufr;': '\U0001d518', + 'ufr;': '\U0001d532', + 'Ugrave': '\xd9', + 'ugrave': '\xf9', + 'Ugrave;': '\xd9', + 'ugrave;': '\xf9', + 'uHar;': '\u2963', + 'uharl;': '\u21bf', + 'uharr;': '\u21be', + 'uhblk;': '\u2580', + 'ulcorn;': '\u231c', + 'ulcorner;': '\u231c', + 'ulcrop;': '\u230f', + 'ultri;': '\u25f8', + 'Umacr;': '\u016a', + 'umacr;': '\u016b', + 'uml': '\xa8', + 'uml;': '\xa8', + 'UnderBar;': '_', + 'UnderBrace;': '\u23df', + 'UnderBracket;': '\u23b5', + 'UnderParenthesis;': '\u23dd', + 'Union;': '\u22c3', + 'UnionPlus;': '\u228e', + 'Uogon;': '\u0172', + 'uogon;': '\u0173', + 'Uopf;': '\U0001d54c', + 'uopf;': '\U0001d566', + 'UpArrow;': '\u2191', + 'Uparrow;': '\u21d1', + 'uparrow;': '\u2191', + 'UpArrowBar;': '\u2912', + 'UpArrowDownArrow;': '\u21c5', + 'UpDownArrow;': '\u2195', + 'Updownarrow;': '\u21d5', + 'updownarrow;': '\u2195', + 'UpEquilibrium;': '\u296e', + 'upharpoonleft;': '\u21bf', + 'upharpoonright;': '\u21be', + 'uplus;': '\u228e', + 'UpperLeftArrow;': '\u2196', + 'UpperRightArrow;': '\u2197', + 'Upsi;': '\u03d2', + 'upsi;': '\u03c5', + 'upsih;': '\u03d2', + 'Upsilon;': '\u03a5', + 'upsilon;': '\u03c5', + 'UpTee;': '\u22a5', + 'UpTeeArrow;': '\u21a5', + 'upuparrows;': '\u21c8', + 'urcorn;': '\u231d', + 'urcorner;': '\u231d', + 'urcrop;': '\u230e', + 'Uring;': '\u016e', + 'uring;': '\u016f', + 'urtri;': '\u25f9', + 'Uscr;': '\U0001d4b0', + 'uscr;': '\U0001d4ca', + 'utdot;': '\u22f0', + 'Utilde;': '\u0168', + 'utilde;': '\u0169', + 'utri;': '\u25b5', + 'utrif;': '\u25b4', + 'uuarr;': '\u21c8', + 'Uuml': '\xdc', + 'uuml': '\xfc', + 'Uuml;': '\xdc', + 'uuml;': '\xfc', + 'uwangle;': '\u29a7', + 'vangrt;': '\u299c', + 'varepsilon;': '\u03f5', + 'varkappa;': '\u03f0', + 'varnothing;': '\u2205', + 'varphi;': '\u03d5', + 'varpi;': '\u03d6', + 'varpropto;': '\u221d', + 'vArr;': '\u21d5', + 'varr;': '\u2195', + 'varrho;': '\u03f1', + 'varsigma;': '\u03c2', + 'varsubsetneq;': '\u228a\ufe00', + 'varsubsetneqq;': '\u2acb\ufe00', + 'varsupsetneq;': '\u228b\ufe00', + 'varsupsetneqq;': '\u2acc\ufe00', + 'vartheta;': '\u03d1', + 'vartriangleleft;': '\u22b2', + 'vartriangleright;': '\u22b3', + 'Vbar;': '\u2aeb', + 'vBar;': '\u2ae8', + 'vBarv;': '\u2ae9', + 'Vcy;': '\u0412', + 'vcy;': '\u0432', + 'VDash;': '\u22ab', + 'Vdash;': '\u22a9', + 'vDash;': '\u22a8', + 'vdash;': '\u22a2', + 'Vdashl;': '\u2ae6', + 'Vee;': '\u22c1', + 'vee;': '\u2228', + 'veebar;': '\u22bb', + 'veeeq;': '\u225a', + 'vellip;': '\u22ee', + 'Verbar;': '\u2016', + 'verbar;': '|', + 'Vert;': '\u2016', + 'vert;': '|', + 'VerticalBar;': '\u2223', + 'VerticalLine;': '|', + 'VerticalSeparator;': '\u2758', + 'VerticalTilde;': '\u2240', + 'VeryThinSpace;': '\u200a', + 'Vfr;': '\U0001d519', + 'vfr;': '\U0001d533', + 'vltri;': '\u22b2', + 'vnsub;': '\u2282\u20d2', + 'vnsup;': '\u2283\u20d2', + 'Vopf;': '\U0001d54d', + 'vopf;': '\U0001d567', + 'vprop;': '\u221d', + 'vrtri;': '\u22b3', + 'Vscr;': '\U0001d4b1', + 'vscr;': '\U0001d4cb', + 'vsubnE;': '\u2acb\ufe00', + 'vsubne;': '\u228a\ufe00', + 'vsupnE;': '\u2acc\ufe00', + 'vsupne;': '\u228b\ufe00', + 'Vvdash;': '\u22aa', + 'vzigzag;': '\u299a', + 'Wcirc;': '\u0174', + 'wcirc;': '\u0175', + 'wedbar;': '\u2a5f', + 'Wedge;': '\u22c0', + 'wedge;': '\u2227', + 'wedgeq;': '\u2259', + 'weierp;': '\u2118', + 'Wfr;': '\U0001d51a', + 'wfr;': '\U0001d534', + 'Wopf;': '\U0001d54e', + 'wopf;': '\U0001d568', + 'wp;': '\u2118', + 'wr;': '\u2240', + 'wreath;': '\u2240', + 'Wscr;': '\U0001d4b2', + 'wscr;': '\U0001d4cc', + 'xcap;': '\u22c2', + 'xcirc;': '\u25ef', + 'xcup;': '\u22c3', + 'xdtri;': '\u25bd', + 'Xfr;': '\U0001d51b', + 'xfr;': '\U0001d535', + 'xhArr;': '\u27fa', + 'xharr;': '\u27f7', + 'Xi;': '\u039e', + 'xi;': '\u03be', + 'xlArr;': '\u27f8', + 'xlarr;': '\u27f5', + 'xmap;': '\u27fc', + 'xnis;': '\u22fb', + 'xodot;': '\u2a00', + 'Xopf;': '\U0001d54f', + 'xopf;': '\U0001d569', + 'xoplus;': '\u2a01', + 'xotime;': '\u2a02', + 'xrArr;': '\u27f9', + 'xrarr;': '\u27f6', + 'Xscr;': '\U0001d4b3', + 'xscr;': '\U0001d4cd', + 'xsqcup;': '\u2a06', + 'xuplus;': '\u2a04', + 'xutri;': '\u25b3', + 'xvee;': '\u22c1', + 'xwedge;': '\u22c0', + 'Yacute': '\xdd', + 'yacute': '\xfd', + 'Yacute;': '\xdd', + 'yacute;': '\xfd', + 'YAcy;': '\u042f', + 'yacy;': '\u044f', + 'Ycirc;': '\u0176', + 'ycirc;': '\u0177', + 'Ycy;': '\u042b', + 'ycy;': '\u044b', + 'yen': '\xa5', + 'yen;': '\xa5', + 'Yfr;': '\U0001d51c', + 'yfr;': '\U0001d536', + 'YIcy;': '\u0407', + 'yicy;': '\u0457', + 'Yopf;': '\U0001d550', + 'yopf;': '\U0001d56a', + 'Yscr;': '\U0001d4b4', + 'yscr;': '\U0001d4ce', + 'YUcy;': '\u042e', + 'yucy;': '\u044e', + 'yuml': '\xff', + 'Yuml;': '\u0178', + 'yuml;': '\xff', + 'Zacute;': '\u0179', + 'zacute;': '\u017a', + 'Zcaron;': '\u017d', + 'zcaron;': '\u017e', + 'Zcy;': '\u0417', + 'zcy;': '\u0437', + 'Zdot;': '\u017b', + 'zdot;': '\u017c', + 'zeetrf;': '\u2128', + 'ZeroWidthSpace;': '\u200b', + 'Zeta;': '\u0396', + 'zeta;': '\u03b6', + 'Zfr;': '\u2128', + 'zfr;': '\U0001d537', + 'ZHcy;': '\u0416', + 'zhcy;': '\u0436', + 'zigrarr;': '\u21dd', + 'Zopf;': '\u2124', + 'zopf;': '\U0001d56b', + 'Zscr;': '\U0001d4b5', + 'zscr;': '\U0001d4cf', + 'zwj;': '\u200d', + 'zwnj;': '\u200c', + } + try: import http.client as compat_http_client except ImportError: # Python 2 @@ -82,7 +2322,6 @@ try: except ImportError: # Python 2 from HTMLParser import HTMLParser as compat_HTMLParser - try: from subprocess import DEVNULL compat_subprocess_get_DEVNULL = lambda: DEVNULL @@ -244,13 +2483,20 @@ try: except ImportError: # Python 2.6 from xml.parsers.expat import ExpatError as compat_xml_parse_error + +etree = xml.etree.ElementTree + + +class _TreeBuilder(etree.TreeBuilder): + def doctype(self, name, pubid, system): + pass + if sys.version_info[0] >= 3: - compat_etree_fromstring = xml.etree.ElementTree.fromstring + def compat_etree_fromstring(text): + return etree.XML(text, parser=etree.XMLParser(target=_TreeBuilder())) else: # python 2.x tries to encode unicode strings with ascii (see the # XMLParser._fixtext method) - etree = xml.etree.ElementTree - try: _etree_iter = etree.Element.iter except AttributeError: # Python <=2.6 @@ -264,7 +2510,7 @@ else: # 2.7 source def _XML(text, parser=None): if not parser: - parser = etree.XMLParser(target=etree.TreeBuilder()) + parser = etree.XMLParser(target=_TreeBuilder()) parser.feed(text) return parser.close() @@ -276,7 +2522,7 @@ else: return el def compat_etree_fromstring(text): - doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory))) + doc = _XML(text, parser=etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory))) for el in _etree_iter(doc): if el.text is not None and isinstance(el.text, bytes): el.text = el.text.decode('utf-8') @@ -340,24 +2586,28 @@ except ImportError: # Python 2 return parsed_result try: - from shlex import quote as shlex_quote + from shlex import quote as compat_shlex_quote except ImportError: # Python < 3.3 - def shlex_quote(s): + def compat_shlex_quote(s): if re.match(r'^[-_\w./]+$', s): return s else: return "'" + s.replace("'", "'\"'\"'") + "'" -if sys.version_info >= (2, 7, 3): +try: + args = shlex.split('中文') + assert (isinstance(args, list) and + isinstance(args[0], compat_str) and + args[0] == '中文') compat_shlex_split = shlex.split -else: +except (AssertionError, UnicodeEncodeError): # Working around shlex issue with unicode strings on some python 2 # versions (see http://bugs.python.org/issue1548891) def compat_shlex_split(s, comments=False, posix=True): if isinstance(s, compat_str): s = s.encode('utf-8') - return shlex.split(s, comments, posix) + return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix))) def compat_ord(c): @@ -373,6 +2623,9 @@ compat_os_name = os._name if os.name == 'java' else os.name if sys.version_info >= (3, 0): compat_getenv = os.getenv compat_expanduser = os.path.expanduser + + def compat_setenv(key, value, env=os.environ): + env[key] = value else: # Environment variables should be decoded with filesystem encoding. # Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918) @@ -384,6 +2637,12 @@ else: env = env.decode(get_filesystem_encoding()) return env + def compat_setenv(key, value, env=os.environ): + def encode(v): + from .utils import get_filesystem_encoding + return v.encode(get_filesystem_encoding()) if isinstance(v, compat_str) else v + env[encode(key)] = encode(value) + # HACK: The default implementations of os.path.expanduser from cpython do not decode # environment variables with filesystem encoding. We will work around this by # providing adjusted implementations. @@ -456,18 +2715,6 @@ else: print(s) -try: - subprocess_check_output = subprocess.check_output -except AttributeError: - def subprocess_check_output(*args, **kwargs): - assert 'input' not in kwargs - p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs) - output, _ = p.communicate() - ret = p.poll() - if ret: - raise subprocess.CalledProcessError(ret, p.args, output=output) - return output - if sys.version_info < (3, 0) and sys.platform == 'win32': def compat_getpass(prompt, *args, **kwargs): if isinstance(prompt, compat_str): @@ -477,6 +2724,11 @@ if sys.version_info < (3, 0) and sys.platform == 'win32': else: compat_getpass = getpass.getpass +try: + compat_input = raw_input +except NameError: # Python 3 + compat_input = input + # Python < 2.6.5 require kwargs to be bytes try: def _testfunc(x): @@ -583,6 +2835,26 @@ if sys.version_info >= (3, 0): else: from tokenize import generate_tokens as compat_tokenize_tokenize + +try: + struct.pack('!I', 0) +except TypeError: + # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument + # See https://bugs.python.org/issue19099 + def compat_struct_pack(spec, *args): + if isinstance(spec, compat_str): + spec = spec.encode('ascii') + return struct.pack(spec, *args) + + def compat_struct_unpack(spec, *args): + if isinstance(spec, compat_str): + spec = spec.encode('ascii') + return struct.unpack(spec, *args) +else: + compat_struct_pack = struct.pack + compat_struct_unpack = struct.unpack + + __all__ = [ 'compat_HTMLParser', 'compat_HTTPError', @@ -596,17 +2868,23 @@ __all__ = [ 'compat_getenv', 'compat_getpass', 'compat_html_entities', + 'compat_html_entities_html5', 'compat_http_client', 'compat_http_server', + 'compat_input', 'compat_itertools_count', 'compat_kwargs', 'compat_ord', 'compat_os_name', 'compat_parse_qs', 'compat_print', + 'compat_setenv', + 'compat_shlex_quote', 'compat_shlex_split', 'compat_socket_create_connection', 'compat_str', + 'compat_struct_pack', + 'compat_struct_unpack', 'compat_subprocess_get_DEVNULL', 'compat_tokenize_tokenize', 'compat_urllib_error', @@ -623,7 +2901,5 @@ __all__ = [ 'compat_urlretrieve', 'compat_xml_parse_error', 'compat_xpath', - 'shlex_quote', - 'subprocess_check_output', 'workaround_optparse_bug9161', ] diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index 8d642fc3e..fae245024 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -6,6 +6,7 @@ import sys import re from .common import FileDownloader +from ..compat import compat_setenv from ..postprocessor.ffmpeg import FFmpegPostProcessor, EXT_TO_OUT_FORMATS from ..utils import ( cli_option, @@ -84,7 +85,7 @@ class ExternalFD(FileDownloader): cmd, stderr=subprocess.PIPE) _, stderr = p.communicate() if p.returncode != 0: - self.to_stderr(stderr) + self.to_stderr(stderr.decode('utf-8', 'replace')) return p.returncode @@ -198,6 +199,19 @@ class FFmpegFD(ExternalFD): '-headers', ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())] + env = None + proxy = self.params.get('proxy') + if proxy: + if not re.match(r'^[\da-zA-Z]+://', proxy): + proxy = 'http://%s' % proxy + # Since December 2015 ffmpeg supports -http_proxy option (see + # http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd) + # We could switch to the following code if we are able to detect version properly + # args += ['-http_proxy', proxy] + env = os.environ.copy() + compat_setenv('HTTP_PROXY', proxy, env=env) + compat_setenv('http_proxy', proxy, env=env) + protocol = info_dict.get('protocol') if protocol == 'rtmp': @@ -224,7 +238,7 @@ class FFmpegFD(ExternalFD): args += ['-rtmp_live', 'live'] args += ['-i', url, '-c', 'copy'] - if protocol == 'm3u8': + if protocol in ('m3u8', 'm3u8_native'): if self.params.get('hls_use_mpegts', False) or tmpfilename == '-': args += ['-f', 'mpegts'] else: @@ -239,7 +253,7 @@ class FFmpegFD(ExternalFD): self._debug_cmd(args) - proc = subprocess.Popen(args, stdin=subprocess.PIPE) + proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env) try: retval = proc.wait() except KeyboardInterrupt: diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index 664d87543..80c21d40b 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -12,37 +12,49 @@ from ..compat import ( compat_urlparse, compat_urllib_error, compat_urllib_parse_urlparse, + compat_struct_pack, + compat_struct_unpack, ) from ..utils import ( encodeFilename, fix_xml_ampersands, sanitize_open, - struct_pack, - struct_unpack, xpath_text, ) +class DataTruncatedError(Exception): + pass + + class FlvReader(io.BytesIO): """ Reader for Flv files The file format is documented in https://www.adobe.com/devnet/f4v.html """ + def read_bytes(self, n): + data = self.read(n) + if len(data) < n: + raise DataTruncatedError( + 'FlvReader error: need %d bytes while only %d bytes got' % ( + n, len(data))) + return data + # Utility functions for reading numbers and strings def read_unsigned_long_long(self): - return struct_unpack('!Q', self.read(8))[0] + return compat_struct_unpack('!Q', self.read_bytes(8))[0] def read_unsigned_int(self): - return struct_unpack('!I', self.read(4))[0] + return compat_struct_unpack('!I', self.read_bytes(4))[0] def read_unsigned_char(self): - return struct_unpack('!B', self.read(1))[0] + return compat_struct_unpack('!B', self.read_bytes(1))[0] def read_string(self): res = b'' while True: - char = self.read(1) + char = self.read_bytes(1) if char == b'\x00': break res += char @@ -53,18 +65,18 @@ class FlvReader(io.BytesIO): Read a box and return the info as a tuple: (box_size, box_type, box_data) """ real_size = size = self.read_unsigned_int() - box_type = self.read(4) + box_type = self.read_bytes(4) header_end = 8 if size == 1: real_size = self.read_unsigned_long_long() header_end = 16 - return real_size, box_type, self.read(real_size - header_end) + return real_size, box_type, self.read_bytes(real_size - header_end) def read_asrt(self): # version self.read_unsigned_char() # flags - self.read(3) + self.read_bytes(3) quality_entry_count = self.read_unsigned_char() # QualityEntryCount for i in range(quality_entry_count): @@ -85,7 +97,7 @@ class FlvReader(io.BytesIO): # version self.read_unsigned_char() # flags - self.read(3) + self.read_bytes(3) # time scale self.read_unsigned_int() @@ -119,7 +131,7 @@ class FlvReader(io.BytesIO): # version self.read_unsigned_char() # flags - self.read(3) + self.read_bytes(3) self.read_unsigned_int() # BootstrapinfoVersion # Profile,Live,Update,Reserved @@ -184,6 +196,11 @@ def build_fragments_list(boot_info): first_frag_number = fragment_run_entry_table[0]['first'] fragments_counter = itertools.count(first_frag_number) for segment, fragments_count in segment_run_table['segment_run']: + # In some live HDS streams (for example Rai), `fragments_count` is + # abnormal and causing out-of-memory errors. It's OK to change the + # number of fragments for live streams as they are updated periodically + if fragments_count == 4294967295 and boot_info['live']: + fragments_count = 2 for _ in range(fragments_count): res.append((segment, next(fragments_counter))) @@ -194,11 +211,11 @@ def build_fragments_list(boot_info): def write_unsigned_int(stream, val): - stream.write(struct_pack('!I', val)) + stream.write(compat_struct_pack('!I', val)) def write_unsigned_int_24(stream, val): - stream.write(struct_pack('!I', val)[1:]) + stream.write(compat_struct_pack('!I', val)[1:]) def write_flv_header(stream): @@ -307,7 +324,7 @@ class F4mFD(FragmentFD): doc = compat_etree_fromstring(manifest) formats = [(int(f.attrib.get('bitrate', -1)), f) for f in self._get_unencrypted_media(doc)] - if requested_bitrate is None: + if requested_bitrate is None or len(formats) == 1: # get the best format formats = sorted(formats, key=lambda f: f[0]) rate, media = formats[-1] @@ -317,7 +334,11 @@ class F4mFD(FragmentFD): base_url = compat_urlparse.urljoin(man_url, media.attrib['url']) bootstrap_node = doc.find(_add_ns('bootstrapInfo')) - boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url) + # From Adobe F4M 3.0 spec: + # The <baseURL> element SHALL be the base URL for all relative + # (HTTP-based) URLs in the manifest. If <baseURL> is not present, said + # URLs should be relative to the location of the containing document. + boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, man_url) live = boot_info['live'] metadata_node = media.find(_add_ns('metadata')) if metadata_node is not None: @@ -374,7 +395,17 @@ class F4mFD(FragmentFD): down.close() reader = FlvReader(down_data) while True: - _, box_type, box_data = reader.read_box_info() + try: + _, box_type, box_data = reader.read_box_info() + except DataTruncatedError: + if test: + # In tests, segments may be truncated, and thus + # FlvReader may not be able to parse the whole + # chunk. If so, write the segment as is + # See https://github.com/rg3/youtube-dl/issues/9214 + dest_stream.write(down_data) + break + raise if box_type == b'mdat': dest_stream.write(box_data) break diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index a01dac031..3b7bb3508 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -2,13 +2,24 @@ from __future__ import unicode_literals import os.path import re +import binascii +try: + from Crypto.Cipher import AES + can_decrypt_frag = True +except ImportError: + can_decrypt_frag = False from .fragment import FragmentFD +from .external import FFmpegFD -from ..compat import compat_urlparse +from ..compat import ( + compat_urlparse, + compat_struct_pack, +) from ..utils import ( encodeFilename, sanitize_open, + parse_m3u8_attributes, ) @@ -17,42 +28,101 @@ class HlsFD(FragmentFD): FD_NAME = 'hlsnative' + @staticmethod + def can_download(manifest): + UNSUPPORTED_FEATURES = ( + r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)', # encrypted streams [1] + r'#EXT-X-BYTERANGE', # playlists composed of byte ranges of media files [2] + + # Live streams heuristic does not always work (e.g. geo restricted to Germany + # http://hls-geo.daserste.de/i/videoportal/Film/c_620000/622873/format,716451,716457,716450,716458,716459,.mp4.csmil/index_4_av.m3u8?null=0) + # r'#EXT-X-MEDIA-SEQUENCE:(?!0$)', # live streams [3] + + # This heuristic also is not correct since segments may not be appended as well. + # Twitch vods of finished streams have EXT-X-PLAYLIST-TYPE:EVENT despite + # no segments will definitely be appended to the end of the playlist. + # r'#EXT-X-PLAYLIST-TYPE:EVENT', # media segments may be appended to the end of + # # event media playlists [4] + + # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.4 + # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2 + # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2 + # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5 + ) + check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES] + check_results.append(can_decrypt_frag or '#EXT-X-KEY:METHOD=AES-128' not in manifest) + return all(check_results) + def real_download(self, filename, info_dict): man_url = info_dict['url'] self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME) manifest = self.ydl.urlopen(man_url).read() s = manifest.decode('utf-8', 'ignore') - fragment_urls = [] + + if not self.can_download(s): + self.report_warning( + 'hlsnative has detected features it does not support, ' + 'extraction will be delegated to ffmpeg') + fd = FFmpegFD(self.ydl, self.params) + for ph in self._progress_hooks: + fd.add_progress_hook(ph) + return fd.real_download(filename, info_dict) + + total_frags = 0 for line in s.splitlines(): line = line.strip() if line and not line.startswith('#'): - segment_url = ( - line - if re.match(r'^https?://', line) - else compat_urlparse.urljoin(man_url, line)) - fragment_urls.append(segment_url) - # We only download the first fragment during the test - if self.params.get('test', False): - break + total_frags += 1 ctx = { 'filename': filename, - 'total_frags': len(fragment_urls), + 'total_frags': total_frags, } self._prepare_and_start_frag_download(ctx) + i = 0 + media_sequence = 0 + decrypt_info = {'METHOD': 'NONE'} frags_filenames = [] - for i, frag_url in enumerate(fragment_urls): - frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i) - success = ctx['dl'].download(frag_filename, {'url': frag_url}) - if not success: - return False - down, frag_sanitized = sanitize_open(frag_filename, 'rb') - ctx['dest_stream'].write(down.read()) - down.close() - frags_filenames.append(frag_sanitized) + for line in s.splitlines(): + line = line.strip() + if line: + if not line.startswith('#'): + frag_url = ( + line + if re.match(r'^https?://', line) + else compat_urlparse.urljoin(man_url, line)) + frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i) + success = ctx['dl'].download(frag_filename, {'url': frag_url}) + if not success: + return False + down, frag_sanitized = sanitize_open(frag_filename, 'rb') + frag_content = down.read() + down.close() + if decrypt_info['METHOD'] == 'AES-128': + iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence) + frag_content = AES.new( + decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content) + ctx['dest_stream'].write(frag_content) + frags_filenames.append(frag_sanitized) + # We only download the first fragment during the test + if self.params.get('test', False): + break + i += 1 + media_sequence += 1 + elif line.startswith('#EXT-X-KEY'): + decrypt_info = parse_m3u8_attributes(line[11:]) + if decrypt_info['METHOD'] == 'AES-128': + if 'IV' in decrypt_info: + decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:]) + if not re.match(r'^https?://', decrypt_info['URI']): + decrypt_info['URI'] = compat_urlparse.urljoin( + man_url, decrypt_info['URI']) + decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read() + elif line.startswith('#EXT-X-MEDIA-SEQUENCE'): + media_sequence = int(line[22:]) self._finish_frag_download(ctx) diff --git a/youtube_dl/extractor/abcnews.py b/youtube_dl/extractor/abcnews.py new file mode 100644 index 000000000..b61a6327c --- /dev/null +++ b/youtube_dl/extractor/abcnews.py @@ -0,0 +1,135 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import calendar +import re +import time + +from .amp import AMPIE +from .common import InfoExtractor +from ..compat import compat_urlparse + + +class AbcNewsVideoIE(AMPIE): + IE_NAME = 'abcnews:video' + _VALID_URL = 'http://abcnews.go.com/[^/]+/video/(?P<display_id>[0-9a-z-]+)-(?P<id>\d+)' + + _TESTS = [{ + 'url': 'http://abcnews.go.com/ThisWeek/video/week-exclusive-irans-foreign-minister-zarif-20411932', + 'info_dict': { + 'id': '20411932', + 'ext': 'mp4', + 'display_id': 'week-exclusive-irans-foreign-minister-zarif', + 'title': '\'This Week\' Exclusive: Iran\'s Foreign Minister Zarif', + 'description': 'George Stephanopoulos goes one-on-one with Iranian Foreign Minister Dr. Javad Zarif.', + 'duration': 180, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('display_id') + video_id = mobj.group('id') + info_dict = self._extract_feed_info( + 'http://abcnews.go.com/video/itemfeed?id=%s' % video_id) + info_dict.update({ + 'id': video_id, + 'display_id': display_id, + }) + return info_dict + + +class AbcNewsIE(InfoExtractor): + IE_NAME = 'abcnews' + _VALID_URL = 'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)' + + _TESTS = [{ + 'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY', + 'info_dict': { + 'id': '10498713', + 'ext': 'flv', + 'display_id': 'dramatic-video-rare-death-job-america', + 'title': 'Occupational Hazards', + 'description': 'Nightline investigates the dangers that lurk at various jobs.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'upload_date': '20100428', + 'timestamp': 1272412800, + }, + 'add_ie': ['AbcNewsVideo'], + }, { + 'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818', + 'info_dict': { + 'id': '39125818', + 'ext': 'mp4', + 'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016', + 'title': 'Justin Timberlake Drops Hints For Secret Single', + 'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.', + 'upload_date': '20160515', + 'timestamp': 1463329500, + }, + 'params': { + # m3u8 download + 'skip_download': True, + # The embedded YouTube video is blocked due to copyright issues + 'playlist_items': '1', + }, + 'add_ie': ['AbcNewsVideo'], + }, { + 'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('display_id') + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + video_url = self._search_regex( + r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL') + full_video_url = compat_urlparse.urljoin(url, video_url) + + youtube_url = self._html_search_regex( + r'<iframe[^>]+src="(https://www\.youtube\.com/embed/[^"]+)"', + webpage, 'YouTube URL', default=None) + + timestamp = None + date_str = self._html_search_regex( + r'<span[^>]+class="timestamp">([^<]+)</span>', + webpage, 'timestamp', fatal=False) + if date_str: + tz_offset = 0 + if date_str.endswith(' ET'): # Eastern Time + tz_offset = -5 + date_str = date_str[:-3] + date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p'] + for date_format in date_formats: + try: + timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format)) + except ValueError: + continue + if timestamp is not None: + timestamp -= tz_offset * 3600 + + entry = { + '_type': 'url_transparent', + 'ie_key': AbcNewsVideoIE.ie_key(), + 'url': full_video_url, + 'id': video_id, + 'display_id': display_id, + 'timestamp': timestamp, + } + + if youtube_url: + entries = [entry, self.url_result(youtube_url, 'Youtube')] + return self.playlist_result(entries) + + return entry diff --git a/youtube_dl/extractor/adobetv.py b/youtube_dl/extractor/adobetv.py index 8753ee2cf..5ae16fa16 100644 --- a/youtube_dl/extractor/adobetv.py +++ b/youtube_dl/extractor/adobetv.py @@ -156,7 +156,10 @@ class AdobeTVVideoIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - video_data = self._download_json(url + '?format=json', video_id) + webpage = self._download_webpage(url, video_id) + + video_data = self._parse_json(self._search_regex( + r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id) formats = [{ 'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')), diff --git a/youtube_dl/extractor/aenetworks.py b/youtube_dl/extractor/aenetworks.py index 1bbfe2641..8f53050c9 100644 --- a/youtube_dl/extractor/aenetworks.py +++ b/youtube_dl/extractor/aenetworks.py @@ -2,41 +2,33 @@ from __future__ import unicode_literals import re -from .common import InfoExtractor +from .theplatform import ThePlatformIE from ..utils import ( smuggle_url, update_url_query, unescapeHTML, + extract_attributes, + get_element_by_attribute, ) +from ..compat import ( + compat_urlparse, +) + +class AENetworksBaseIE(ThePlatformIE): + _THEPLATFORM_KEY = 'crazyjava' + _THEPLATFORM_SECRET = 's3cr3t' -class AENetworksIE(InfoExtractor): + +class AENetworksIE(AENetworksBaseIE): IE_NAME = 'aenetworks' IE_DESC = 'A+E Networks: A&E, Lifetime, History.com, FYI Network' - _VALID_URL = r'https?://(?:www\.)?(?:(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?P<type>[^/]+)/(?:[^/]+/)+(?P<id>[^/]+?)(?:$|[?#])' - + _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?:shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|movies/(?P<movie_display_id>[^/]+)/full-movie)' _TESTS = [{ - 'url': 'http://www.history.com/topics/valentines-day/history-of-valentines-day/videos/bet-you-didnt-know-valentines-day?m=528e394da93ae&s=undefined&f=1&free=false', - 'info_dict': { - 'id': 'g12m5Gyt3fdR', - 'ext': 'mp4', - 'title': "Bet You Didn't Know: Valentine's Day", - 'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7', - 'timestamp': 1375819729, - 'upload_date': '20130806', - 'uploader': 'AENE-NEW', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - 'add_ie': ['ThePlatform'], - 'expected_warnings': ['JSON-LD'], - }, { 'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1', 'md5': '8ff93eb073449f151d6b90c0ae1ef0c7', 'info_dict': { - 'id': 'eg47EERs_JsZ', + 'id': '22253814', 'ext': 'mp4', 'title': 'Winter Is Coming', 'description': 'md5:641f424b7a19d8e24f26dea22cf59d74', @@ -46,42 +38,168 @@ class AENetworksIE(InfoExtractor): }, 'add_ie': ['ThePlatform'], }, { - 'url': 'http://www.aetv.com/shows/duck-dynasty/video/inlawful-entry', + 'url': 'http://www.history.com/shows/ancient-aliens/season-1', + 'info_dict': { + 'id': '71889446852', + }, + 'playlist_mincount': 5, + }, { + 'url': 'http://www.mylifetime.com/shows/atlanta-plastic', + 'info_dict': { + 'id': 'SERIES4317', + 'title': 'Atlanta Plastic', + }, + 'playlist_mincount': 2, + }, { + 'url': 'http://www.aetv.com/shows/duck-dynasty/season-9/episode-1', 'only_matching': True }, { - 'url': 'http://www.fyi.tv/shows/tiny-house-nation/videos/207-sq-ft-minnesota-prairie-cottage', + 'url': 'http://www.fyi.tv/shows/tiny-house-nation/season-1/episode-8', 'only_matching': True }, { - 'url': 'http://www.mylifetime.com/shows/project-runway-junior/video/season-1/episode-6/superstar-clients', + 'url': 'http://www.mylifetime.com/shows/project-runway-junior/season-1/episode-6', + 'only_matching': True + }, { + 'url': 'http://www.mylifetime.com/movies/center-stage-on-pointe/full-movie', 'only_matching': True }] + _DOMAIN_TO_REQUESTOR_ID = { + 'history.com': 'HISTORY', + 'aetv.com': 'AETV', + 'mylifetime.com': 'LIFETIME', + 'fyi.tv': 'FYI', + } def _real_extract(self, url): - page_type, video_id = re.match(self._VALID_URL, url).groups() + domain, show_path, movie_display_id = re.match(self._VALID_URL, url).groups() + display_id = show_path or movie_display_id + webpage = self._download_webpage(url, display_id) + if show_path: + url_parts = show_path.split('/') + url_parts_len = len(url_parts) + if url_parts_len == 1: + entries = [] + for season_url_path in re.findall(r'(?s)<li[^>]+data-href="(/shows/%s/season-\d+)"' % url_parts[0], webpage): + entries.append(self.url_result( + compat_urlparse.urljoin(url, season_url_path), 'AENetworks')) + return self.playlist_result( + entries, self._html_search_meta('aetn:SeriesId', webpage), + self._html_search_meta('aetn:SeriesTitle', webpage)) + elif url_parts_len == 2: + entries = [] + for episode_item in re.findall(r'(?s)<div[^>]+class="[^"]*episode-item[^"]*"[^>]*>', webpage): + episode_attributes = extract_attributes(episode_item) + episode_url = compat_urlparse.urljoin( + url, episode_attributes['data-canonical']) + entries.append(self.url_result( + episode_url, 'AENetworks', + episode_attributes['data-videoid'])) + return self.playlist_result( + entries, self._html_search_meta('aetn:SeasonId', webpage)) + + query = { + 'mbr': 'true', + 'assetTypes': 'medium_video_s3' + } + video_id = self._html_search_meta('aetn:VideoID', webpage) + media_url = self._search_regex( + r"media_url\s*=\s*'([^']+)'", webpage, 'video url') + theplatform_metadata = self._download_theplatform_metadata(self._search_regex( + r'https?://link.theplatform.com/s/([^?]+)', media_url, 'theplatform_path'), video_id) + info = self._parse_theplatform_metadata(theplatform_metadata) + if theplatform_metadata.get('AETN$isBehindWall'): + requestor_id = self._DOMAIN_TO_REQUESTOR_ID[domain] + resource = '<rss version="2.0" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>%s</title><item><title>%s</title><guid>%s</guid><media:rating scheme="urn:v-chip">%s</media:rating></item></channel></rss>' % (requestor_id, theplatform_metadata['title'], theplatform_metadata['AETN$PPL_pplProgramId'], theplatform_metadata['ratings'][0]['rating']) + query['auth'] = self._extract_mvpd_auth( + url, video_id, requestor_id, resource) + info.update(self._search_json_ld(webpage, video_id, fatal=False)) + media_url = update_url_query(media_url, query) + media_url = self._sign_url(media_url, self._THEPLATFORM_KEY, self._THEPLATFORM_SECRET) + formats, subtitles = self._extract_theplatform_smil(media_url, video_id) + self._sort_formats(formats) + info.update({ + 'id': video_id, + 'formats': formats, + 'subtitles': subtitles, + }) + return info - webpage = self._download_webpage(url, video_id) - video_url_re = [ - r'data-href="[^"]*/%s"[^>]+data-release-url="([^"]+)"' % video_id, - r"media_url\s*=\s*'([^']+)'" - ] - video_url = unescapeHTML(self._search_regex(video_url_re, webpage, 'video url')) - query = {'mbr': 'true'} - if page_type == 'shows': - query['assetTypes'] = 'medium_video_s3' - if 'switch=hds' in video_url: - query['switch'] = 'hls' +class HistoryTopicIE(AENetworksBaseIE): + IE_NAME = 'history:topic' + IE_DESC = 'History.com Topic' + _VALID_URL = r'https?://(?:www\.)?history\.com/topics/(?:[^/]+/)?(?P<topic_id>[^/]+)(?:/[^/]+(?:/(?P<video_display_id>[^/?#]+))?)?' + _TESTS = [{ + 'url': 'http://www.history.com/topics/valentines-day/history-of-valentines-day/videos/bet-you-didnt-know-valentines-day?m=528e394da93ae&s=undefined&f=1&free=false', + 'info_dict': { + 'id': '40700995724', + 'ext': 'mp4', + 'title': "Bet You Didn't Know: Valentine's Day", + 'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7', + 'timestamp': 1375819729, + 'upload_date': '20130806', + 'uploader': 'AENE-NEW', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + 'add_ie': ['ThePlatform'], + }, { + 'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/videos', + 'info_dict': + { + 'id': 'world-war-i-history', + 'title': 'World War I History', + }, + 'playlist_mincount': 24, + }, { + 'url': 'http://www.history.com/topics/world-war-i-history/videos', + 'only_matching': True, + }, { + 'url': 'http://www.history.com/topics/world-war-i/world-war-i-history', + 'only_matching': True, + }, { + 'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/speeches', + 'only_matching': True, + }] - info = self._search_json_ld(webpage, video_id, fatal=False) - info.update({ + def theplatform_url_result(self, theplatform_url, video_id, query): + return { '_type': 'url_transparent', + 'id': video_id, 'url': smuggle_url( - update_url_query(video_url, query), + update_url_query(theplatform_url, query), { 'sig': { - 'key': 'crazyjava', - 'secret': 's3cr3t'}, + 'key': self._THEPLATFORM_KEY, + 'secret': self._THEPLATFORM_SECRET, + }, 'force_smil_url': True }), - }) - return info + 'ie_key': 'ThePlatform', + } + + def _real_extract(self, url): + topic_id, video_display_id = re.match(self._VALID_URL, url).groups() + if video_display_id: + webpage = self._download_webpage(url, video_display_id) + release_url, video_id = re.search(r"_videoPlayer.play\('([^']+)'\s*,\s*'[^']+'\s*,\s*'(\d+)'\)", webpage).groups() + release_url = unescapeHTML(release_url) + + return self.theplatform_url_result( + release_url, video_id, { + 'mbr': 'true', + 'switch': 'hls' + }) + else: + webpage = self._download_webpage(url, topic_id) + entries = [] + for episode_item in re.findall(r'<a.+?data-release-url="[^"]+"[^>]*>', webpage): + video_attributes = extract_attributes(episode_item) + entries.append(self.theplatform_url_result( + video_attributes['data-release-url'], video_attributes['data-id'], { + 'mbr': 'true', + 'switch': 'hls' + })) + return self.playlist_result(entries, topic_id, get_element_by_attribute('class', 'show-title', webpage)) diff --git a/youtube_dl/extractor/afreecatv.py b/youtube_dl/extractor/afreecatv.py new file mode 100644 index 000000000..518c61f67 --- /dev/null +++ b/youtube_dl/extractor/afreecatv.py @@ -0,0 +1,133 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse_urlparse, + compat_urlparse, +) +from ..utils import ( + ExtractorError, + int_or_none, + xpath_element, + xpath_text, +) + + +class AfreecaTVIE(InfoExtractor): + IE_DESC = 'afreecatv.com' + _VALID_URL = r'''(?x)^ + https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)? + (?: + /app/(?:index|read_ucc_bbs)\.cgi| + /player/[Pp]layer\.(?:swf|html)) + \?.*?\bnTitleNo=(?P<id>\d+)''' + _TESTS = [{ + 'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=', + 'md5': 'f72c89fe7ecc14c1b5ce506c4996046e', + 'info_dict': { + 'id': '36164052', + 'ext': 'mp4', + 'title': '데일리 에이프릴 요정들의 시상식!', + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': 'dailyapril', + 'uploader_id': 'dailyapril', + 'upload_date': '20160503', + } + }, { + 'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867', + 'info_dict': { + 'id': '36153164', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'", + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': 'dailyapril', + 'uploader_id': 'dailyapril', + }, + 'playlist_count': 2, + 'playlist': [{ + 'md5': 'd8b7c174568da61d774ef0203159bf97', + 'info_dict': { + 'id': '36153164_1', + 'ext': 'mp4', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'", + 'upload_date': '20160502', + }, + }, { + 'md5': '58f2ce7f6044e34439ab2d50612ab02b', + 'info_dict': { + 'id': '36153164_2', + 'ext': 'mp4', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'", + 'upload_date': '20160502', + }, + }], + }, { + 'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652', + 'only_matching': True, + }] + + @staticmethod + def parse_video_key(key): + video_key = {} + m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key) + if m: + video_key['upload_date'] = m.group('upload_date') + video_key['part'] = m.group('part') + return video_key + + def _real_extract(self, url): + video_id = self._match_id(url) + parsed_url = compat_urllib_parse_urlparse(url) + info_url = compat_urlparse.urlunparse(parsed_url._replace( + netloc='afbbs.afreecatv.com:8080', + path='/api/video/get_video_info.php')) + video_xml = self._download_xml(info_url, video_id) + + if xpath_element(video_xml, './track/video/file') is None: + raise ExtractorError('Specified AfreecaTV video does not exist', + expected=True) + + title = xpath_text(video_xml, './track/title', 'title') + uploader = xpath_text(video_xml, './track/nickname', 'uploader') + uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id') + duration = int_or_none(xpath_text(video_xml, './track/duration', + 'duration')) + thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail') + + entries = [] + for i, video_file in enumerate(video_xml.findall('./track/video/file')): + video_key = self.parse_video_key(video_file.get('key', '')) + if not video_key: + continue + entries.append({ + 'id': '%s_%s' % (video_id, video_key.get('part', i + 1)), + 'title': title, + 'upload_date': video_key.get('upload_date'), + 'duration': int_or_none(video_file.get('duration')), + 'url': video_file.text, + }) + + info = { + 'id': video_id, + 'title': title, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'duration': duration, + 'thumbnail': thumbnail, + } + + if len(entries) > 1: + info['_type'] = 'multi_video' + info['entries'] = entries + elif len(entries) == 1: + info['url'] = entries[0]['url'] + info['upload_date'] = entries[0].get('upload_date') + else: + raise ExtractorError( + 'No files found for the specified AfreecaTV video, either' + ' the URL is incorrect or the video has been made private.', + expected=True) + + return info diff --git a/youtube_dl/extractor/aftonbladet.py b/youtube_dl/extractor/aftonbladet.py index d548592fe..5766b4fe8 100644 --- a/youtube_dl/extractor/aftonbladet.py +++ b/youtube_dl/extractor/aftonbladet.py @@ -24,10 +24,10 @@ class AftonbladetIE(InfoExtractor): webpage = self._download_webpage(url, video_id) # find internal video meta data - meta_url = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json' + meta_url = 'http://aftonbladet-play-metadata.cdn.drvideo.aptoma.no/video/%s.json' player_config = self._parse_json(self._html_search_regex( r'data-player-config="([^"]+)"', webpage, 'player config'), video_id) - internal_meta_id = player_config['videoId'] + internal_meta_id = player_config['aptomaVideoId'] internal_meta_url = meta_url % internal_meta_id internal_meta_json = self._download_json( internal_meta_url, video_id, 'Downloading video meta data') diff --git a/youtube_dl/extractor/amp.py b/youtube_dl/extractor/amp.py index 138fa0808..e8e40126b 100644 --- a/youtube_dl/extractor/amp.py +++ b/youtube_dl/extractor/amp.py @@ -5,6 +5,8 @@ from .common import InfoExtractor from ..utils import ( int_or_none, parse_iso8601, + mimetype2ext, + determine_ext, ) @@ -50,21 +52,25 @@ class AMPIE(InfoExtractor): if isinstance(media_content, dict): media_content = [media_content] for media_data in media_content: - media = media_data['@attributes'] - media_type = media['type'] - if media_type == 'video/f4m': + media = media_data.get('@attributes', {}) + media_url = media.get('url') + if not media_url: + continue + ext = mimetype2ext(media.get('type')) or determine_ext(media_url) + if ext == 'f4m': formats.extend(self._extract_f4m_formats( - media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124', + media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124', video_id, f4m_id='hds', fatal=False)) - elif media_type == 'application/x-mpegURL': + elif ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False)) + media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) else: formats.append({ - 'format_id': media_data['media-category']['@attributes']['label'], + 'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'), 'url': media['url'], 'tbr': int_or_none(media.get('bitrate')), 'filesize': int_or_none(media.get('fileSize')), + 'ext': ext, }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/animeondemand.py b/youtube_dl/extractor/animeondemand.py index 9b01e38f5..159c6ef5a 100644 --- a/youtube_dl/extractor/animeondemand.py +++ b/youtube_dl/extractor/animeondemand.py @@ -22,6 +22,7 @@ class AnimeOnDemandIE(InfoExtractor): _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply' _NETRC_MACHINE = 'animeondemand' _TESTS = [{ + # jap, OmU 'url': 'https://www.anime-on-demand.de/anime/161', 'info_dict': { 'id': '161', @@ -30,17 +31,21 @@ class AnimeOnDemandIE(InfoExtractor): }, 'playlist_mincount': 4, }, { - # Film wording is used instead of Episode + # Film wording is used instead of Episode, ger/jap, Dub/OmU 'url': 'https://www.anime-on-demand.de/anime/39', 'only_matching': True, }, { - # Episodes without titles + # Episodes without titles, jap, OmU 'url': 'https://www.anime-on-demand.de/anime/162', 'only_matching': True, }, { # ger/jap, Dub/OmU, account required 'url': 'https://www.anime-on-demand.de/anime/169', 'only_matching': True, + }, { + # Full length film, non-series, ger/jap, Dub/OmU, account required + 'url': 'https://www.anime-on-demand.de/anime/185', + 'only_matching': True, }] def _login(self): @@ -110,35 +115,12 @@ class AnimeOnDemandIE(InfoExtractor): entries = [] - for num, episode_html in enumerate(re.findall( - r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 1): - episodebox_title = self._search_regex( - (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1', - r'class="episodebox-title"[^>]+>(?P<title>.+?)<'), - episode_html, 'episodebox title', default=None, group='title') - if not episodebox_title: - continue - - episode_number = int(self._search_regex( - r'(?:Episode|Film)\s*(\d+)', - episodebox_title, 'episode number', default=num)) - episode_title = self._search_regex( - r'(?:Episode|Film)\s*\d+\s*-\s*(.+)', - episodebox_title, 'episode title', default=None) - - video_id = 'episode-%d' % episode_number - - common_info = { - 'id': video_id, - 'series': anime_title, - 'episode': episode_title, - 'episode_number': episode_number, - } - + def extract_info(html, video_id, num=None): + title, description = [None] * 2 formats = [] for input_ in re.findall( - r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', episode_html): + r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', html): attributes = extract_attributes(input_) playlist_urls = [] for playlist_key in ('data-playlist', 'data-otherplaylist'): @@ -161,7 +143,7 @@ class AnimeOnDemandIE(InfoExtractor): format_id_list.append(lang) if kind: format_id_list.append(kind) - if not format_id_list: + if not format_id_list and num is not None: format_id_list.append(compat_str(num)) format_id = '-'.join(format_id_list) format_note = ', '.join(filter(None, (kind, lang_note))) @@ -215,28 +197,74 @@ class AnimeOnDemandIE(InfoExtractor): }) formats.extend(file_formats) - if formats: - self._sort_formats(formats) + return { + 'title': title, + 'description': description, + 'formats': formats, + } + + def extract_entries(html, video_id, common_info, num): + info = extract_info(html, video_id, num) + + if info['formats']: + self._sort_formats(info['formats']) f = common_info.copy() - f.update({ - 'title': title, - 'description': description, - 'formats': formats, - }) + f.update(info) entries.append(f) - # Extract teaser only when full episode is not available - if not formats: + # Extract teaser/trailer only when full episode is not available + if not info['formats']: m = re.search( - r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<', - episode_html) + r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<', + html) if m: f = common_info.copy() f.update({ - 'id': '%s-teaser' % f['id'], + 'id': '%s-%s' % (f['id'], m.group('kind').lower()), 'title': m.group('title'), 'url': compat_urlparse.urljoin(url, m.group('href')), }) entries.append(f) + def extract_episodes(html): + for num, episode_html in enumerate(re.findall( + r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1): + episodebox_title = self._search_regex( + (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1', + r'class="episodebox-title"[^>]+>(?P<title>.+?)<'), + episode_html, 'episodebox title', default=None, group='title') + if not episodebox_title: + continue + + episode_number = int(self._search_regex( + r'(?:Episode|Film)\s*(\d+)', + episodebox_title, 'episode number', default=num)) + episode_title = self._search_regex( + r'(?:Episode|Film)\s*\d+\s*-\s*(.+)', + episodebox_title, 'episode title', default=None) + + video_id = 'episode-%d' % episode_number + + common_info = { + 'id': video_id, + 'series': anime_title, + 'episode': episode_title, + 'episode_number': episode_number, + } + + extract_entries(episode_html, video_id, common_info) + + def extract_film(html, video_id): + common_info = { + 'id': anime_id, + 'title': anime_title, + 'description': anime_description, + } + extract_entries(html, video_id, common_info) + + extract_episodes(webpage) + + if not entries: + extract_film(webpage, anime_id) + return self.playlist_result(entries, anime_id, anime_title, anime_description) diff --git a/youtube_dl/extractor/anvato.py b/youtube_dl/extractor/anvato.py new file mode 100644 index 000000000..cb29cf111 --- /dev/null +++ b/youtube_dl/extractor/anvato.py @@ -0,0 +1,224 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import base64 +import hashlib +import json +import random +import time + +from .common import InfoExtractor +from ..aes import aes_encrypt +from ..compat import compat_str +from ..utils import ( + bytes_to_intlist, + determine_ext, + intlist_to_bytes, + int_or_none, + strip_jsonp, +) + + +def md5_text(s): + if not isinstance(s, compat_str): + s = compat_str(s) + return hashlib.md5(s.encode('utf-8')).hexdigest() + + +class AnvatoIE(InfoExtractor): + # Copied from anvplayer.min.js + _ANVACK_TABLE = { + 'nbcu_nbcd_desktop_web_prod_93d8ead38ce2024f8f544b78306fbd15895ae5e6': 'NNemUkySjxLyPTKvZRiGntBIjEyK8uqicjMakIaQ', + 'nbcu_nbcd_desktop_web_qa_1a6f01bdd0dc45a439043b694c8a031d': 'eSxJUbA2UUKBTXryyQ2d6NuM8oEqaPySvaPzfKNA', + 'nbcu_nbcd_desktop_web_acc_eb2ff240a5d4ae9a63d4c297c32716b6c523a129': '89JR3RtUGbvKuuJIiKOMK0SoarLb5MUx8v89RcbP', + 'nbcu_nbcd_watchvod_web_prod_e61107507180976724ec8e8319fe24ba5b4b60e1': 'Uc7dFt7MJ9GsBWB5T7iPvLaMSOt8BBxv4hAXk5vv', + 'nbcu_nbcd_watchvod_web_qa_42afedba88a36203db5a4c09a5ba29d045302232': 'T12oDYVFP2IaFvxkmYMy5dKxswpLHtGZa4ZAXEi7', + 'nbcu_nbcd_watchvod_web_acc_9193214448e2e636b0ffb78abacfd9c4f937c6ca': 'MmobcxUxMedUpohNWwXaOnMjlbiyTOBLL6d46ZpR', + 'nbcu_local_monitor_web_acc_f998ad54eaf26acd8ee033eb36f39a7b791c6335': 'QvfIoPYrwsjUCcASiw3AIkVtQob2LtJHfidp9iWg', + 'nbcu_cable_monitor_web_acc_a413759603e8bedfcd3c61b14767796e17834077': 'uwVPJLShvJWSs6sWEIuVem7MTF8A4IknMMzIlFto', + 'nbcu_nbcd_mcpstage_web_qa_4c43a8f6e95a88dbb40276c0630ba9f693a63a4e': 'PxVYZVwjhgd5TeoPRxL3whssb5OUPnM3zyAzq8GY', + 'nbcu_comcast_comcast_web_prod_074080762ad4ce956b26b43fb22abf153443a8c4': 'afnaRZfDyg1Z3WZHdupKfy6xrbAG2MHqe3VfuSwh', + 'nbcu_comcast_comcast_web_qa_706103bb93ead3ef70b1de12a0e95e3c4481ade0': 'DcjsVbX9b3uoPlhdriIiovgFQZVxpISZwz0cx1ZK', + 'nbcu_comcast_comcastcable_web_prod_669f04817536743563d7331c9293e59fbdbe3d07': '0RwMN2cWy10qhAhOscq3eK7aEe0wqnKt3vJ0WS4D', + 'nbcu_comcast_comcastcable_web_qa_3d9d2d66219094127f0f6b09cc3c7bb076e3e1ca': '2r8G9DEya7PCqBceKZgrn2XkXgASjwLMuaFE1Aad', + 'hearst_hearst_demo_web_stage_960726dfef3337059a01a78816e43b29ec04dfc7': 'cuZBPXTR6kSdoTCVXwk5KGA8rk3NrgGn4H6e9Dsp', + 'anvato_mcpqa_demo_web_stage_18b55e00db5a13faa8d03ae6e41f6f5bcb15b922': 'IOaaLQ8ymqVyem14QuAvE5SndQynTcH5CrLkU2Ih', + 'anvato_nextmedia_demo_web_stage_9787d56a02ff6b9f43e9a2b0920d8ca88beb5818': 'Pqu9zVzI1ApiIzbVA3VkGBEQHvdKSUuKpD6s2uaR', + 'anvato_scripps_app_web_prod_0837996dbe373629133857ae9eb72e740424d80a': 'du1ccmn7RxzgizwbWU7hyUaGodNlJn7HtXI0WgXW', + 'anvato_scripps_app_web_stage_360797e00fe2826be142155c4618cc52fce6c26c': '2PMrQ0BRoqCWl7nzphj0GouIMEh2mZYivAT0S1Su', + 'fs2go_fs2go_go_all_prod_21934911ccfafc03a075894ead2260d11e2ddd24': 'RcuHlKikW2IJw6HvVoEkqq2UsuEJlbEl11pWXs4Q', + 'fs2go_fs2go_go_web_prod_ead4b0eec7460c1a07783808db21b49cf1f2f9a7': '4K0HTT2u1zkQA2MaGaZmkLa1BthGSBdr7jllrhk5', + 'fs2go_fs2go_go_web_stage_407585454a4400355d4391691c67f361': 'ftnc37VKRJBmHfoGGi3kT05bHyeJzilEzhKJCyl3', + 'fs2go_fs2go_go_android_stage_44b714db6f8477f29afcba15a41e1d30': 'CtxpPvVpo6AbZGomYUhkKs7juHZwNml9b9J0J2gI', + 'anvato_cbslocal_app_web_prod_547f3e49241ef0e5d30c79b2efbca5d92c698f67': 'Pw0XX5KBDsyRnPS0R2JrSrXftsy8Jnz5pAjaYC8s', + 'anvato_cbslocal_app_web_stage_547a5f096594cd3e00620c6f825cad1096d28c80': '37OBUhX2uwNyKhhrNzSSNHSRPZpApC3trdqDBpuz', + 'fs2go_att_att_web_prod_1042dddd089a05438b6a08f972941176f699ffd8': 'JLcF20JwYvpv6uAGcLWIaV12jKwaL1R8us4b6Zkg', + 'fs2go_att_att_web_stage_807c5001955fc114a3331fe027ddc76e': 'gbu1oO1y0JiOFh4SUipt86P288JHpyjSqolrrT1x', + 'fs2go_fs2go_tudor_web_prod_a7dd8e5a7cdc830cae55eae6f3e9fee5ee49eb9b': 'ipcp87VCEZXPPe868j3orLqzc03oTy7DXsGkAXXH', + 'anvato_mhz_app_web_prod_b808218b30de7fdf60340cbd9831512bc1bf6d37': 'Stlm5Gs6BEhJLRTZHcNquyzxGqr23EuFmE5DCgjX', + 'fs2go_charter_charter_web_stage_c2c6e5a68375a1bf00fff213d3ff8f61a835a54c': 'Lz4hbJp1fwL6jlcz4M2PMzghM4jp4aAmybtT5dPc', + 'fs2go_charter_charter_web_prod_ebfe3b10f1af215a7321cd3d629e0b81dfa6fa8c': 'vUJsK345A1bVmyYDRhZX0lqFIgVXuqhmuyp1EtPK', + 'anvato_epfox_app_web_prod_b3373168e12f423f41504f207000188daf88251b': 'GDKq1ixvX3MoBNdU5IOYmYa2DTUXYOozPjrCJnW7', + 'anvato_epfox_app_web_stage_a3c2ce60f8f83ef374a88b68ee73a950f8ab87ce': '2jz2NH4BsXMaDsoJ5qkHMbcczAfIReo2eFYuVC1C', + 'fs2go_verizon_verizon_web_stage_08e6df0354a4803f1b1f2428b5a9a382e8dbcd62': 'rKTVapNaAcmnUbGL4ZcuOoY4SE7VmZSQsblPFr7e', + 'fs2go_verizon_verizon_web_prod_f909564cb606eff1f731b5e22e0928676732c445': 'qLSUuHerM3u9eNPzaHyUK52obai5MvE4XDJfqYe1', + 'fs2go_foxcom_synd_web_stage_f7b9091f00ea25a4fdaaae77fca5b54cdc7e7043': '96VKF2vLd24fFiDfwPFpzM5llFN4TiIGAlodE0Re', + 'fs2go_foxcom_synd_web_prod_0f2cdd64d87e4ab6a1d54aada0ff7a7c8387a064': 'agiPjbXEyEZUkbuhcnmVPhe9NNVbDjCFq2xkcx51', + 'anvato_own_app_web_stage_1214ade5d28422c4dae9d03c1243aba0563c4dba': 'mzhamNac3swG4WsJAiUTacnGIODi6SWeVWk5D7ho', + 'anvato_own_app_web_prod_944e162ed927ec3e9ed13eb68ed2f1008ee7565e': '9TSxh6G2TXOLBoYm9ro3LdNjjvnXpKb8UR8KoIP9', + 'anvato_scripps_app_ftv_prod_a10a10468edd5afb16fb48171c03b956176afad1': 'COJ2i2UIPK7xZqIWswxe7FaVBOVgRkP1F6O6qGoH', + 'anvato_scripps_app_ftv_stage_77d3ad2bdb021ec37ca2e35eb09acd396a974c9a': 'Q7nnopNLe2PPfGLOTYBqxSaRpl209IhqaEuDZi1F', + 'anvato_univision_app_web_stage_551236ef07a0e17718c3995c35586b5ed8cb5031': 'D92PoLS6UitwxDRA191HUGT9OYcOjV6mPMa5wNyo', + 'anvato_univision_app_web_prod_039a5c0a6009e637ae8ac906718a79911e0e65e1': '5mVS5u4SQjtw6NGw2uhMbKEIONIiLqRKck5RwQLR', + 'nbcu_cnbc_springfield_ios_prod_670207fae43d6e9a94c351688851a2ce': 'M7fqCCIP9lW53oJbHs19OlJlpDrVyc2OL8gNeuTa', + 'nbcu_cnbc_springfieldvod_ios_prod_7a5f04b1ceceb0e9c9e2264a44aa236e08e034c2': 'Yia6QbJahW0S7K1I0drksimhZb4UFq92xLBmmMvk', + 'anvato_cox_app_web_prod_ce45cda237969f93e7130f50ee8bb6280c1484ab': 'cc0miZexpFtdoqZGvdhfXsLy7FXjRAOgb9V0f5fZ', + 'anvato_cox_app_web_stage_c23dbe016a8e9d8c7101d10172b92434f6088bf9': 'yivU3MYHd2eDZcOfmLbINVtqxyecKTOp8OjOuoGJ', + 'anvato_chnzero_app_web_stage_b1164d1352b579e792e542fddf13ee34c0eeb46b': 'A76QkXMmVH8lTCfU15xva1mZnSVcqeY4Xb22Kp7m', + 'anvato_chnzero_app_web_prod_253d358928dc08ec161eda2389d53707288a730c': 'OA5QI3ZWZZkdtUEDqh28AH8GedsF6FqzJI32596b', + 'anvato_discovery_vodpoc_web_stage_9fa7077b5e8af1f8355f65d4fb8d2e0e9d54e2b7': 'q3oT191tTQ5g3JCP67PkjLASI9s16DuWZ6fYmry3', + 'anvato_discovery_vodpoc_web_prod_688614983167a1af6cdf6d76343fda10a65223c1': 'qRvRQCTVHd0VVOHsMvvfidyWmlYVrTbjby7WqIuK', + 'nbcu_cnbc_springfieldvod_ftv_stage_826040aad1925a46ac5dfb4b3c5143e648c6a30d': 'JQaSb5a8Tz0PT4ti329DNmzDO30TnngTHmvX8Vua', + 'nbcu_cnbc_springfield_ftv_stage_826040aad1925a46ac5dfb4b3c5143e648c6a30d': 'JQaSb5a8Tz0PT4ti329DNmzDO30TnngTHmvX8Vua', + 'nbcu_nbcd_capture_web_stage_4dd9d585bfb984ebf856dee35db027b2465cc4ae': '0j1Ov4Vopyi2HpBZJYdL2m8ERJVGYh3nNpzPiO8F', + 'nbcu_nbcd_watch3_android_prod_7712ca5fcf1c22f19ec1870a9650f9c37db22dcf': '3LN2UB3rPUAMu7ZriWkHky9vpLMXYha8JbSnxBlx', + 'nbcu_nbcd_watchvod3_android_prod_0910a3a4692d57c0b5ff4316075bc5d096be45b9': 'mJagcQ2II30vUOAauOXne7ERwbf5S9nlB3IP17lQ', + 'anvato_scripps_app_atv_prod_790deda22e16e71e83df58f880cd389908a45d52': 'CB6trI1mpoDIM5o54DNTsji90NDBQPZ4z4RqBNSH', + 'nbcu_nbcd_watchv4_android_prod_ff67cef9cb409158c6f8c3533edddadd0b750507': 'j8CHQCUWjlYERj4NFRmUYOND85QNbHViH09UwuKm', + 'nbcu_nbcd_watchvodv4_android_prod_a814d781609989dea6a629d50ae4c7ad8cc8e907': 'rkVnUXxdA9rawVLUlDQtMue9Y4Q7lFEaIotcUhjt', + 'rvVKpA50qlOPLFxMjrCGf5pdkdQDm7qn': '1J7ZkY5Qz5lMLi93QOH9IveE7EYB3rLl', + 'nbcu_dtv_local_web_prod_b266cf49defe255fd4426a97e27c09e513e9f82f': 'HuLnJDqzLa4saCzYMJ79zDRSQpEduw1TzjMNQu2b', + 'nbcu_att_local_web_prod_4cef038b2d969a6b7d700a56a599040b6a619f67': 'Q0Em5VDc2KpydUrVwzWRXAwoNBulWUxCq2faK0AV', + 'nbcu_dish_local_web_prod_c56dcaf2da2e9157a4266c82a78195f1dd570f6b': 'bC1LWmRz9ayj2AlzizeJ1HuhTfIaJGsDBnZNgoRg', + 'nbcu_verizon_local_web_prod_88bebd2ce006d4ed980de8133496f9a74cb9b3e1': 'wzhDKJZpgvUSS1EQvpCQP8Q59qVzcPixqDGJefSk', + 'nbcu_charter_local_web_prod_9ad90f7fc4023643bb718f0fe0fd5beea2382a50': 'PyNbxNhEWLzy1ZvWEQelRuIQY88Eub7xbSVRMdfT', + 'nbcu_suddenlink_local_web_prod_20fb711725cac224baa1c1cb0b1c324d25e97178': '0Rph41lPXZbb3fqeXtHjjbxfSrNbtZp1Ygq7Jypa', + 'nbcu_wow_local_web_prod_652d9ce4f552d9c2e7b5b1ed37b8cb48155174ad': 'qayIBZ70w1dItm2zS42AptXnxW15mkjRrwnBjMPv', + 'nbcu_centurylink_local_web_prod_2034402b029bf3e837ad46814d9e4b1d1345ccd5': 'StePcPMkjsX51PcizLdLRMzxMEl5k2FlsMLUNV4k', + 'nbcu_atlanticbrd_local_web_prod_8d5f5ecbf7f7b2f5e6d908dd75d90ae3565f682e': 'NtYLb4TFUS0pRs3XTkyO5sbVGYjVf17bVbjaGscI', + 'nbcu_nbcd_watchvod_web_dev_08bc05699be47c4f31d5080263a8cfadc16d0f7c': 'hwxi2dgDoSWgfmVVXOYZm14uuvku4QfopstXckhr', + 'anvato_nextmedia_app_web_prod_a4fa8c7204aa65e71044b57aaf63711980cfe5a0': 'tQN1oGPYY1nM85rJYePWGcIb92TG0gSqoVpQTWOw', + 'anvato_mcp_lin_web_prod_4c36fbfd4d8d8ecae6488656e21ac6d1ac972749': 'GUXNf5ZDX2jFUpu4WT2Go4DJ5nhUCzpnwDRRUx1K', + 'anvato_mcp_univision_web_prod_37fe34850c99a3b5cdb71dab10a417dd5cdecafa': 'bLDYF8JqfG42b7bwKEgQiU9E2LTIAtnKzSgYpFUH', + 'anvato_mcp_fs2go_web_prod_c7b90a93e171469cdca00a931211a2f556370d0a': 'icgGoYGipQMMSEvhplZX1pwbN69srwKYWksz3xWK', + 'anvato_mcp_sps_web_prod_54bdc90dd6ba21710e9f7074338365bba28da336': 'fA2iQdI7RDpynqzQYIpXALVS83NTPr8LLFK4LFsu', + 'anvato_mcp_anv_web_prod_791407490f4c1ef2a4bcb21103e0cb1bcb3352b3': 'rMOUZqe9lwcGq2mNgG3EDusm6lKgsUnczoOX3mbg', + 'anvato_mcp_gray_web_prod_4c10f067c393ed8fc453d3930f8ab2b159973900': 'rMOUZqe9lwcGq2mNgG3EDusm6lKgsUnczoOX3mbg', + 'anvato_mcp_hearst_web_prod_5356c3de0fc7c90a3727b4863ca7fec3a4524a99': 'P3uXJ0fXXditBPCGkfvlnVScpPEfKmc64Zv7ZgbK', + 'anvato_mcp_cbs_web_prod_02f26581ff80e5bda7aad28226a8d369037f2cbe': 'mGPvo5ZA5SgjOFAPEPXv7AnOpFUICX8hvFQVz69n', + 'anvato_mcp_telemundo_web_prod_c5278d51ad46fda4b6ca3d0ea44a7846a054f582': 'qyT6PXXLjVNCrHaRVj0ugAhalNRS7Ee9BP7LUokD', + 'nbcu_nbcd_watchvodv4_web_stage_4108362fba2d4ede21f262fea3c4162cbafd66c7': 'DhaU5lj0W2gEdcSSsnxURq8t7KIWtJfD966crVDk', + 'anvato_scripps_app_ios_prod_409c41960c60b308db43c3cc1da79cab9f1c3d93': 'WPxj5GraLTkYCyj3M7RozLqIycjrXOEcDGFMIJPn', + 'EZqvRyKBJLrgpClDPDF8I7Xpdp40Vx73': '4OxGd2dEakylntVKjKF0UK9PDPYB6A9W', + 'M2v78QkpleXm9hPp9jUXI63x5vA6BogR': 'ka6K32k7ZALmpINkjJUGUo0OE42Md1BQ', + 'nbcu_nbcd_desktop_web_prod_93d8ead38ce2024f8f544b78306fbd15895ae5e6_secure': 'NNemUkySjxLyPTKvZRiGntBIjEyK8uqicjMakIaQ' + } + + _AUTH_KEY = b'\x31\xc2\x42\x84\x9e\x73\xa0\xce' + + def __init__(self, *args, **kwargs): + super(AnvatoIE, self).__init__(*args, **kwargs) + self.__server_time = None + + def _server_time(self, access_key, video_id): + if self.__server_time is not None: + return self.__server_time + + self.__server_time = int(self._download_json( + self._api_prefix(access_key) + 'server_time?anvack=' + access_key, video_id, + note='Fetching server time')['server_time']) + + return self.__server_time + + def _api_prefix(self, access_key): + return 'https://tkx2-%s.anvato.net/rest/v2/' % ('prod' if 'prod' in access_key else 'stage') + + def _get_video_json(self, access_key, video_id): + # See et() in anvplayer.min.js, which is an alias of getVideoJSON() + video_data_url = self._api_prefix(access_key) + 'mcp/video/%s?anvack=%s' % (video_id, access_key) + server_time = self._server_time(access_key, video_id) + input_data = '%d~%s~%s' % (server_time, md5_text(video_data_url), md5_text(server_time)) + + auth_secret = intlist_to_bytes(aes_encrypt( + bytes_to_intlist(input_data[:64]), bytes_to_intlist(self._AUTH_KEY))) + + video_data_url += '&X-Anvato-Adst-Auth=' + base64.b64encode(auth_secret).decode('ascii') + anvrid = md5_text(time.time() * 1000 * random.random())[:30] + payload = { + 'api': { + 'anvrid': anvrid, + 'anvstk': md5_text('%s|%s|%d|%s' % ( + access_key, anvrid, server_time, self._ANVACK_TABLE[access_key])), + 'anvts': server_time, + }, + } + + return self._download_json( + video_data_url, video_id, transform_source=strip_jsonp, + data=json.dumps(payload).encode('utf-8')) + + def _extract_anvato_videos(self, webpage, video_id): + anvplayer_data = self._parse_json(self._html_search_regex( + r'<script[^>]+data-anvp=\'([^\']+)\'', webpage, + 'Anvato player data'), video_id) + + video_id = anvplayer_data['video'] + access_key = anvplayer_data['accessKey'] + + video_data = self._get_video_json(access_key, video_id) + + formats = [] + for published_url in video_data['published_urls']: + video_url = published_url['embed_url'] + ext = determine_ext(video_url) + + if ext == 'smil': + formats.extend(self._extract_smil_formats(video_url, video_id)) + continue + + tbr = int_or_none(published_url.get('kbps')) + a_format = { + 'url': video_url, + 'format_id': ('-'.join(filter(None, ['http', published_url.get('cdn_name')]))).lower(), + 'tbr': tbr if tbr != 0 else None, + } + + if ext == 'm3u8': + # Not using _extract_m3u8_formats here as individual media + # playlists are also included in published_urls. + if tbr is None: + formats.append(self._m3u8_meta_format(video_url, ext='mp4', m3u8_id='hls')) + continue + else: + a_format.update({ + 'format_id': '-'.join(filter(None, ['hls', compat_str(tbr)])), + 'ext': 'mp4', + }) + elif ext == 'mp3': + a_format['vcodec'] = 'none' + else: + a_format.update({ + 'width': int_or_none(published_url.get('width')), + 'height': int_or_none(published_url.get('height')), + }) + formats.append(a_format) + + self._sort_formats(formats) + + subtitles = {} + for caption in video_data.get('captions', []): + a_caption = { + 'url': caption['url'], + 'ext': 'tt' if caption.get('format') == 'SMPTE-TT' else None + } + subtitles.setdefault(caption['language'], []).append(a_caption) + + return { + 'id': video_id, + 'formats': formats, + 'title': video_data.get('def_title'), + 'description': video_data.get('def_description'), + 'categories': video_data.get('categories'), + 'thumbnail': video_data.get('thumbnail'), + 'subtitles': subtitles, + } diff --git a/youtube_dl/extractor/aol.py b/youtube_dl/extractor/aol.py index 24df8fe93..42c21bf41 100644 --- a/youtube_dl/extractor/aol.py +++ b/youtube_dl/extractor/aol.py @@ -12,7 +12,7 @@ from ..utils import ( class AolIE(InfoExtractor): IE_NAME = 'on.aol.com' - _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/.*-)(?P<id>[^/?-]+)' + _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/(?:[^/]+/)*(?:[^/?#&]+-)?)(?P<id>[^/?#&]+)' _TESTS = [{ # video with 5min ID @@ -53,6 +53,12 @@ class AolIE(InfoExtractor): }, { 'url': 'http://on.aol.com/shows/park-bench-shw518173474-559a1b9be4b0c3bfad3357a7?context=SH:SHW518173474:PL4327:1460619712763', 'only_matching': True, + }, { + 'url': 'http://on.aol.com/video/519442220', + 'only_matching': True, + }, { + 'url': 'aol-video:5707d6b8e4b090497b04f706', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py index be40f85b4..a6801f3d4 100644 --- a/youtube_dl/extractor/appletrailers.py +++ b/youtube_dl/extractor/appletrailers.py @@ -7,6 +7,8 @@ from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( int_or_none, + parse_duration, + unified_strdate, ) @@ -16,7 +18,8 @@ class AppleTrailersIE(InfoExtractor): _TESTS = [{ 'url': 'http://trailers.apple.com/trailers/wb/manofsteel/', 'info_dict': { - 'id': 'manofsteel', + 'id': '5111', + 'title': 'Man of Steel', }, 'playlist': [ { @@ -70,6 +73,15 @@ class AppleTrailersIE(InfoExtractor): 'id': 'blackthorn', }, 'playlist_mincount': 2, + 'expected_warnings': ['Unable to download JSON metadata'], + }, { + # json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json + 'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/', + 'info_dict': { + 'id': '15881', + 'title': 'Kung Fu Panda 3', + }, + 'playlist_mincount': 4, }, { 'url': 'http://trailers.apple.com/ca/metropole/autrui/', 'only_matching': True, @@ -85,6 +97,45 @@ class AppleTrailersIE(InfoExtractor): movie = mobj.group('movie') uploader_id = mobj.group('company') + webpage = self._download_webpage(url, movie) + film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id') + film_data = self._download_json( + 'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id, + film_id, fatal=False) + + if film_data: + entries = [] + for clip in film_data.get('clips', []): + clip_title = clip['title'] + + formats = [] + for version, version_data in clip.get('versions', {}).items(): + for size, size_data in version_data.get('sizes', {}).items(): + src = size_data.get('src') + if not src: + continue + formats.append({ + 'format_id': '%s-%s' % (version, size), + 'url': re.sub(r'_(\d+p.mov)', r'_h\1', src), + 'width': int_or_none(size_data.get('width')), + 'height': int_or_none(size_data.get('height')), + 'language': version[:2], + }) + self._sort_formats(formats) + + entries.append({ + 'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(), + 'formats': formats, + 'title': clip_title, + 'thumbnail': clip.get('screen') or clip.get('thumb'), + 'duration': parse_duration(clip.get('runtime') or clip.get('faded')), + 'upload_date': unified_strdate(clip.get('posted')), + 'uploader_id': uploader_id, + }) + + page_data = film_data.get('page', {}) + return self.playlist_result(entries, film_id, page_data.get('movie_title')) + playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc') def fix_html(s): diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 26446c2fe..13a06396d 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -8,12 +8,12 @@ from .generic import GenericIE from ..utils import ( determine_ext, ExtractorError, - get_element_by_attribute, qualities, int_or_none, parse_duration, unified_strdate, xpath_text, + update_url_query, ) from ..compat import compat_etree_fromstring @@ -35,6 +35,7 @@ class ARDMediathekIE(InfoExtractor): # m3u8 download 'skip_download': True, }, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916', 'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e', @@ -45,6 +46,7 @@ class ARDMediathekIE(InfoExtractor): 'description': 'md5:196392e79876d0ac94c94e8cdb2875f1', 'duration': 5252, }, + 'skip': 'HTTP Error 404: Not Found', }, { # audio 'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086', @@ -56,6 +58,7 @@ class ARDMediathekIE(InfoExtractor): 'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef', 'duration': 3240, }, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', 'only_matching': True, @@ -114,11 +117,14 @@ class ARDMediathekIE(InfoExtractor): continue if ext == 'f4m': formats.extend(self._extract_f4m_formats( - stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', - video_id, preference=-1, f4m_id='hds', fatal=False)) + update_url_query(stream_url, { + 'hdcore': '3.1.1', + 'plugin': 'aasp-3.1.1.69.124' + }), + video_id, f4m_id='hds', fatal=False)) elif ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False)) + stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) else: if server and server.startswith('rtmp'): f = { @@ -232,7 +238,8 @@ class ARDIE(InfoExtractor): 'title': 'Die Story im Ersten: Mission unter falscher Flagge', 'upload_date': '20140804', 'thumbnail': 're:^https?://.*\.jpg$', - } + }, + 'skip': 'HTTP Error 404: Not Found', } def _real_extract(self, url): @@ -274,41 +281,3 @@ class ARDIE(InfoExtractor): 'upload_date': upload_date, 'thumbnail': thumbnail, } - - -class SportschauIE(ARDMediathekIE): - IE_NAME = 'Sportschau' - _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html' - _TESTS = [{ - 'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html', - 'info_dict': { - 'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100', - 'ext': 'mp4', - 'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"', - 'thumbnail': 're:^https?://.*\.jpg$', - 'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - base_url = mobj.group('baseurl') - - webpage = self._download_webpage(url, video_id) - title = get_element_by_attribute('class', 'headline', webpage) - description = self._html_search_meta('description', webpage, 'description') - - info = self._extract_media_info( - base_url + '-mc_defaultQuality-h.json', webpage, video_id) - - info.update({ - 'title': title, - 'description': description, - }) - - return info diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index a9e3266dc..e0c5c1804 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -61,10 +61,7 @@ class ArteTvIE(InfoExtractor): } -class ArteTVPlus7IE(InfoExtractor): - IE_NAME = 'arte.tv:+7' - _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])' - +class ArteTVBaseIE(InfoExtractor): @classmethod def _extract_url_info(cls, url): mobj = re.match(cls._VALID_URL, url) @@ -78,60 +75,6 @@ class ArteTVPlus7IE(InfoExtractor): video_id = mobj.group('id') return video_id, lang - def _real_extract(self, url): - video_id, lang = self._extract_url_info(url) - webpage = self._download_webpage(url, video_id) - return self._extract_from_webpage(webpage, video_id, lang) - - def _extract_from_webpage(self, webpage, video_id, lang): - patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']') - ids = (video_id, '') - # some pages contain multiple videos (like - # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D), - # so we first try to look for json URLs that contain the video id from - # the 'vid' parameter. - patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates] - json_url = self._html_search_regex( - patterns, webpage, 'json vp url', default=None) - if not json_url: - def find_iframe_url(webpage, default=NO_DEFAULT): - return self._html_search_regex( - r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1', - webpage, 'iframe url', group='url', default=default) - - iframe_url = find_iframe_url(webpage, None) - if not iframe_url: - embed_url = self._html_search_regex( - r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None) - if embed_url: - player = self._download_json( - embed_url, video_id, 'Downloading player page') - iframe_url = find_iframe_url(player['html']) - # en and es URLs produce react-based pages with different layout (e.g. - # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world) - if not iframe_url: - program = self._search_regex( - r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n', - webpage, 'program', default=None) - if program: - embed_html = self._parse_json(program, video_id) - if embed_html: - iframe_url = find_iframe_url(embed_html['embed_html']) - if iframe_url: - json_url = compat_parse_qs( - compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0] - if json_url: - title = self._search_regex( - r'<h3[^>]+title=(["\'])(?P<title>.+?)\1', - webpage, 'title', default=None, group='title') - return self._extract_from_json_url(json_url, video_id, lang, title=title) - # Different kind of embed URL (e.g. - # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium) - embed_url = self._search_regex( - r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', - webpage, 'embed url', group='url') - return self.url_result(embed_url) - def _extract_from_json_url(self, json_url, video_id, lang, title=None): info = self._download_json(json_url, video_id) player_info = info['videoJsonPlayer'] @@ -161,24 +104,53 @@ class ArteTVPlus7IE(InfoExtractor): 'es': 'E[ESP]', } + langcode = LANGS.get(lang, lang) + formats = [] for format_id, format_dict in player_info['VSR'].items(): f = dict(format_dict) versionCode = f.get('versionCode') - langcode = LANGS.get(lang, lang) - lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)] - lang_pref = None - if versionCode: - matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)] - lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs) - source_pref = 0 - if versionCode is not None: - # The original version with subtitles has lower relevance - if re.match(r'VO-ST(F|A|E)', versionCode): - source_pref -= 10 - # The version with sourds/mal subtitles has also lower relevance - elif re.match(r'VO?(F|A|E)-STM\1', versionCode): - source_pref -= 9 + l = re.escape(langcode) + + # Language preference from most to least priority + # Reference: section 5.6.3 of + # http://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-05.pdf + PREFERENCES = ( + # original version in requested language, without subtitles + r'VO{0}$'.format(l), + # original version in requested language, with partial subtitles in requested language + r'VO{0}-ST{0}$'.format(l), + # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language + r'VO{0}-STM{0}$'.format(l), + # non-original (dubbed) version in requested language, without subtitles + r'V{0}$'.format(l), + # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language + r'V{0}-ST{0}$'.format(l), + # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language + r'V{0}-STM{0}$'.format(l), + # original version in requested language, with partial subtitles in different language + r'VO{0}-ST(?!{0}).+?$'.format(l), + # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language + r'VO{0}-STM(?!{0}).+?$'.format(l), + # original version in different language, with partial subtitles in requested language + r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l), + # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language + r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l), + # original version in different language, without subtitles + r'VO(?:(?!{0}))?$'.format(l), + # original version in different language, with partial subtitles in different language + r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l), + # original version in different language, with subtitles for the deaf and hard-of-hearing in different language + r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l), + ) + + for pref, p in enumerate(PREFERENCES): + if re.match(p, versionCode): + lang_pref = len(PREFERENCES) - pref + break + else: + lang_pref = -1 + format = { 'format_id': format_id, 'preference': -10 if f.get('videoFormat') == 'M3U8' else None, @@ -188,7 +160,6 @@ class ArteTVPlus7IE(InfoExtractor): 'height': int_or_none(f.get('height')), 'tbr': int_or_none(f.get('bitrate')), 'quality': qfunc(f.get('quality')), - 'source_preference': source_pref, } if f.get('mediaType') == 'rtmp': @@ -207,28 +178,94 @@ class ArteTVPlus7IE(InfoExtractor): return info_dict +class ArteTVPlus7IE(ArteTVBaseIE): + IE_NAME = 'arte.tv:+7' + _VALID_URL = r'https?://(?:(?:www|sites)\.)?arte\.tv/[^/]+/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D', + 'only_matching': True, + }, { + 'url': 'http://sites.arte.tv/karambolage/de/video/karambolage-22', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if ArteTVPlaylistIE.suitable(url) else super(ArteTVPlus7IE, cls).suitable(url) + + def _real_extract(self, url): + video_id, lang = self._extract_url_info(url) + webpage = self._download_webpage(url, video_id) + return self._extract_from_webpage(webpage, video_id, lang) + + def _extract_from_webpage(self, webpage, video_id, lang): + patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']') + ids = (video_id, '') + # some pages contain multiple videos (like + # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D), + # so we first try to look for json URLs that contain the video id from + # the 'vid' parameter. + patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates] + json_url = self._html_search_regex( + patterns, webpage, 'json vp url', default=None) + if not json_url: + def find_iframe_url(webpage, default=NO_DEFAULT): + return self._html_search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1', + webpage, 'iframe url', group='url', default=default) + + iframe_url = find_iframe_url(webpage, None) + if not iframe_url: + embed_url = self._html_search_regex( + r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None) + if embed_url: + player = self._download_json( + embed_url, video_id, 'Downloading player page') + iframe_url = find_iframe_url(player['html']) + # en and es URLs produce react-based pages with different layout (e.g. + # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world) + if not iframe_url: + program = self._search_regex( + r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n', + webpage, 'program', default=None) + if program: + embed_html = self._parse_json(program, video_id) + if embed_html: + iframe_url = find_iframe_url(embed_html['embed_html']) + if iframe_url: + json_url = compat_parse_qs( + compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0] + if json_url: + title = self._search_regex( + r'<h3[^>]+title=(["\'])(?P<title>.+?)\1', + webpage, 'title', default=None, group='title') + return self._extract_from_json_url(json_url, video_id, lang, title=title) + # Different kind of embed URL (e.g. + # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium) + entries = [ + self.url_result(url) + for _, url in re.findall(r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', webpage)] + return self.playlist_result(entries) + + # It also uses the arte_vp_url url from the webpage to extract the information class ArteTVCreativeIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:creative' _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' _TESTS = [{ - 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design', + 'url': 'http://creative.arte.tv/fr/episode/osmosis-episode-1', 'info_dict': { - 'id': '72176', + 'id': '057405-001-A', 'ext': 'mp4', - 'title': 'Folge 2 - Corporate Design', - 'upload_date': '20131004', + 'title': 'OSMOSIS - N\'AYEZ PLUS PEUR D\'AIMER (1)', + 'upload_date': '20150716', }, }, { 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion', - 'info_dict': { - 'id': '160676', - 'ext': 'mp4', - 'title': 'Monty Python live (mostly)', - 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n', - 'upload_date': '20140805', - } + 'playlist_count': 11, + 'add_ie': ['Youtube'], }, { 'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde', 'only_matching': True, @@ -239,7 +276,7 @@ class ArteTVInfoIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:info' _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere', 'info_dict': { 'id': '067528-000-A', @@ -247,7 +284,7 @@ class ArteTVInfoIE(ArteTVPlus7IE): 'title': 'Service civique, un cache misère ?', 'upload_date': '20160403', }, - } + }] class ArteTVFutureIE(ArteTVPlus7IE): @@ -272,6 +309,8 @@ class ArteTVDDCIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:ddc' _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)' + _TESTS = [] + def _real_extract(self, url): video_id, lang = self._extract_url_info(url) if lang == 'folge': @@ -290,7 +329,7 @@ class ArteTVConcertIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:concert' _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde', 'md5': '9ea035b7bd69696b67aa2ccaaa218161', 'info_dict': { @@ -300,24 +339,23 @@ class ArteTVConcertIE(ArteTVPlus7IE): 'upload_date': '20140128', 'description': 'md5:486eb08f991552ade77439fe6d82c305', }, - } + }] class ArteTVCinemaIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:cinema' _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)' - _TEST = { - 'url': 'http://cinema.arte.tv/de/node/38291', - 'md5': '6b275511a5107c60bacbeeda368c3aa1', + _TESTS = [{ + 'url': 'http://cinema.arte.tv/fr/article/les-ailes-du-desir-de-julia-reck', + 'md5': 'a5b9dd5575a11d93daf0e3f404f45438', 'info_dict': { - 'id': '055876-000_PWA12025-D', + 'id': '062494-000-A', 'ext': 'mp4', - 'title': 'Tod auf dem Nil', - 'upload_date': '20160122', - 'description': 'md5:7f749bbb77d800ef2be11d54529b96bc', + 'title': 'Film lauréat du concours web - "Les ailes du désir" de Julia Reck', + 'upload_date': '20150807', }, - } + }] class ArteTVMagazineIE(ArteTVPlus7IE): @@ -362,9 +400,42 @@ class ArteTVEmbedIE(ArteTVPlus7IE): ) ''' + _TESTS = [] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') lang = mobj.group('lang') json_url = mobj.group('json_url') return self._extract_from_json_url(json_url, video_id, lang) + + +class ArteTVPlaylistIE(ArteTVBaseIE): + IE_NAME = 'arte.tv:playlist' + _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/[^#]*#collection/(?P<id>PL-\d+)' + + _TESTS = [{ + 'url': 'http://www.arte.tv/guide/de/plus7/?country=DE#collection/PL-013263/ARTETV', + 'info_dict': { + 'id': 'PL-013263', + 'title': 'Areva & Uramin', + 'description': 'md5:a1dc0312ce357c262259139cfd48c9bf', + }, + 'playlist_mincount': 6, + }, { + 'url': 'http://www.arte.tv/guide/de/playlists?country=DE#collection/PL-013190/ARTETV', + 'only_matching': True, + }] + + def _real_extract(self, url): + playlist_id, lang = self._extract_url_info(url) + collection = self._download_json( + 'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos' + % (lang, playlist_id), playlist_id) + title = collection.get('title') + description = collection.get('shortDescription') or collection.get('teaserText') + entries = [ + self._extract_from_json_url( + video['jsonUrl'], video.get('programId') or playlist_id, lang) + for video in collection['videos'] if video.get('jsonUrl')] + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/audiomack.py b/youtube_dl/extractor/audiomack.py index a52d26cec..f3bd4d444 100644 --- a/youtube_dl/extractor/audiomack.py +++ b/youtube_dl/extractor/audiomack.py @@ -6,6 +6,7 @@ import time from .common import InfoExtractor from .soundcloud import SoundcloudIE +from ..compat import compat_str from ..utils import ( ExtractorError, url_basename, @@ -136,7 +137,7 @@ class AudiomackAlbumIE(InfoExtractor): result[resultkey] = api_response[apikey] song_id = url_basename(api_response['url']).rpartition('.')[0] result['entries'].append({ - 'id': api_response.get('id', song_id), + 'id': compat_str(api_response.get('id', song_id)), 'uploader': api_response.get('artist'), 'title': api_response.get('title', song_id), 'url': api_response['url'], diff --git a/youtube_dl/extractor/azubu.py b/youtube_dl/extractor/azubu.py index efa624de1..a813eb429 100644 --- a/youtube_dl/extractor/azubu.py +++ b/youtube_dl/extractor/azubu.py @@ -46,6 +46,7 @@ class AzubuIE(InfoExtractor): 'uploader_id': 272749, 'view_count': int, }, + 'skip': 'Channel offline', }, ] @@ -56,22 +57,26 @@ class AzubuIE(InfoExtractor): 'http://www.azubu.tv/api/video/%s' % video_id, video_id)['data'] title = data['title'].strip() - description = data['description'] - thumbnail = data['thumbnail'] - view_count = data['view_count'] - uploader = data['user']['username'] - uploader_id = data['user']['id'] + description = data.get('description') + thumbnail = data.get('thumbnail') + view_count = data.get('view_count') + user = data.get('user', {}) + uploader = user.get('username') + uploader_id = user.get('id') stream_params = json.loads(data['stream_params']) - timestamp = float_or_none(stream_params['creationDate'], 1000) - duration = float_or_none(stream_params['length'], 1000) + timestamp = float_or_none(stream_params.get('creationDate'), 1000) + duration = float_or_none(stream_params.get('length'), 1000) renditions = stream_params.get('renditions') or [] video = stream_params.get('FLVFullLength') or stream_params.get('videoFullLength') if video: renditions.append(video) + if not renditions and not user.get('channel', {}).get('is_live', True): + raise ExtractorError('%s said: channel is offline.' % self.IE_NAME, expected=True) + formats = [{ 'url': fmt['url'], 'width': fmt['frameWidth'], diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index c1ef8051d..991ab0676 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -29,7 +29,7 @@ class BandcampIE(InfoExtractor): '_skip': 'There is a limit of 200 free downloads / month for the test song' }, { 'url': 'http://benprunty.bandcamp.com/track/lanius-battle', - 'md5': '2b68e5851514c20efdff2afc5603b8b4', + 'md5': '73d0b3171568232574e45652f8720b5c', 'info_dict': { 'id': '2650410135', 'ext': 'mp3', @@ -48,6 +48,10 @@ class BandcampIE(InfoExtractor): if m_trackinfo: json_code = m_trackinfo.group(1) data = json.loads(json_code)[0] + track_id = compat_str(data['id']) + + if not data.get('file'): + raise ExtractorError('Not streamable', video_id=track_id, expected=True) formats = [] for format_id, format_url in data['file'].items(): @@ -64,7 +68,7 @@ class BandcampIE(InfoExtractor): self._sort_formats(formats) return { - 'id': compat_str(data['id']), + 'id': track_id, 'title': data['title'], 'formats': formats, 'duration': float_or_none(data.get('duration')), diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py index 74c4510f9..4b3cd8c65 100644 --- a/youtube_dl/extractor/bbc.py +++ b/youtube_dl/extractor/bbc.py @@ -31,7 +31,7 @@ class BBCCoUkIE(InfoExtractor): music/clips[/#]| radio/player/ ) - (?P<id>%s) + (?P<id>%s)(?!/(?:episodes|broadcasts|clips)) ''' % _ID_REGEX _MEDIASELECTOR_URLS = [ @@ -192,6 +192,7 @@ class BBCCoUkIE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'Now it\'s really geo-restricted', }, { # compact player (https://github.com/rg3/youtube-dl/issues/8147) 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player', @@ -698,7 +699,9 @@ class BBCIE(BBCCoUkIE): @classmethod def suitable(cls, url): - return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url) + EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE) + return (False if any(ie.suitable(url) for ie in EXCLUDE_IE) + else super(BBCIE, cls).suitable(url)) def _extract_from_media_meta(self, media_meta, video_id): # Direct links to media in media metadata (e.g. @@ -975,3 +978,72 @@ class BBCCoUkArticleIE(InfoExtractor): r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)] return self.playlist_result(entries, playlist_id, title, description) + + +class BBCCoUkPlaylistBaseIE(InfoExtractor): + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result(self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key()) + for video_id in re.findall( + self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage)] + + title, description = self._extract_title_and_description(webpage) + + return self.playlist_result(entries, playlist_id, title, description) + + +class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE): + IE_NAME = 'bbc.co.uk:iplayer:playlist' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/episodes/(?P<id>%s)' % BBCCoUkIE._ID_REGEX + _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s' + _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)' + _TEST = { + 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v', + 'info_dict': { + 'id': 'b05rcz9v', + 'title': 'The Disappearance', + 'description': 'French thriller serial about a missing teenager.', + }, + 'playlist_mincount': 6, + } + + def _extract_title_and_description(self, webpage): + title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False) + description = self._search_regex( + r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>', + webpage, 'description', fatal=False, group='value') + return title, description + + +class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE): + IE_NAME = 'bbc.co.uk:playlist' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX + _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s' + _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)' + _TESTS = [{ + 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips', + 'info_dict': { + 'id': 'b05rcz9v', + 'title': 'The Disappearance - Clips - BBC Four', + 'description': 'French thriller serial about a missing teenager.', + }, + 'playlist_mincount': 7, + }, { + 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player', + 'only_matching': True, + }] + + def _extract_title_and_description(self, webpage): + title = self._og_search_title(webpage, fatal=False) + description = self._og_search_description(webpage) + return title, description diff --git a/youtube_dl/extractor/bet.py b/youtube_dl/extractor/bet.py index 986245bf0..bd3ee2e2e 100644 --- a/youtube_dl/extractor/bet.py +++ b/youtube_dl/extractor/bet.py @@ -1,31 +1,27 @@ from __future__ import unicode_literals -from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote -from ..utils import ( - xpath_text, - xpath_with_ns, - int_or_none, - parse_iso8601, -) +from .mtv import MTVServicesInfoExtractor +from ..utils import unified_strdate +from ..compat import compat_urllib_parse_urlencode -class BetIE(InfoExtractor): +class BetIE(MTVServicesInfoExtractor): _VALID_URL = r'https?://(?:www\.)?bet\.com/(?:[^/]+/)+(?P<id>.+?)\.html' _TESTS = [ { 'url': 'http://www.bet.com/news/politics/2014/12/08/in-bet-exclusive-obama-talks-race-and-racism.html', 'info_dict': { - 'id': 'news/national/2014/a-conversation-with-president-obama', + 'id': '07e96bd3-8850-3051-b856-271b457f0ab8', 'display_id': 'in-bet-exclusive-obama-talks-race-and-racism', 'ext': 'flv', 'title': 'A Conversation With President Obama', - 'description': 'md5:699d0652a350cf3e491cd15cc745b5da', + 'description': 'President Obama urges persistence in confronting racism and bias.', 'duration': 1534, - 'timestamp': 1418075340, 'upload_date': '20141208', - 'uploader': 'admin', 'thumbnail': 're:(?i)^https?://.*\.jpg$', + 'subtitles': { + 'en': 'mincount:2', + } }, 'params': { # rtmp download @@ -35,16 +31,17 @@ class BetIE(InfoExtractor): { 'url': 'http://www.bet.com/video/news/national/2014/justice-for-ferguson-a-community-reacts.html', 'info_dict': { - 'id': 'news/national/2014/justice-for-ferguson-a-community-reacts', + 'id': '9f516bf1-7543-39c4-8076-dd441b459ba9', 'display_id': 'justice-for-ferguson-a-community-reacts', 'ext': 'flv', 'title': 'Justice for Ferguson: A Community Reacts', 'description': 'A BET News special.', 'duration': 1696, - 'timestamp': 1416942360, 'upload_date': '20141125', - 'uploader': 'admin', 'thumbnail': 're:(?i)^https?://.*\.jpg$', + 'subtitles': { + 'en': 'mincount:2', + } }, 'params': { # rtmp download @@ -53,57 +50,32 @@ class BetIE(InfoExtractor): } ] - def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - - media_url = compat_urllib_parse_unquote(self._search_regex( - [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"], - webpage, 'media URL')) + _FEED_URL = "http://feeds.mtvnservices.com/od/feed/bet-mrss-player" - video_id = self._search_regex( - r'/video/(.*)/_jcr_content/', media_url, 'video id') + def _get_feed_query(self, uri): + return compat_urllib_parse_urlencode({ + 'uuid': uri, + }) - mrss = self._download_xml(media_url, display_id) - - item = mrss.find('./channel/item') - - NS_MAP = { - 'dc': 'http://purl.org/dc/elements/1.1/', - 'media': 'http://search.yahoo.com/mrss/', - 'ka': 'http://kickapps.com/karss', - } + def _extract_mgid(self, webpage): + return self._search_regex(r'data-uri="([^"]+)', webpage, 'mgid') - title = xpath_text(item, './title', 'title') - description = xpath_text( - item, './description', 'description', fatal=False) - - timestamp = parse_iso8601(xpath_text( - item, xpath_with_ns('./dc:date', NS_MAP), - 'upload date', fatal=False)) - uploader = xpath_text( - item, xpath_with_ns('./dc:creator', NS_MAP), - 'uploader', fatal=False) + def _real_extract(self, url): + display_id = self._match_id(url) - media_content = item.find( - xpath_with_ns('./media:content', NS_MAP)) - duration = int_or_none(media_content.get('duration')) - smil_url = media_content.get('url') + webpage = self._download_webpage(url, display_id) + mgid = self._extract_mgid(webpage) + videos_info = self._get_videos_info(mgid) - thumbnail = media_content.find( - xpath_with_ns('./media:thumbnail', NS_MAP)).get('url') + info_dict = videos_info['entries'][0] - formats = self._extract_smil_formats(smil_url, display_id) - self._sort_formats(formats) + upload_date = unified_strdate(self._html_search_meta('date', webpage)) + description = self._html_search_meta('description', webpage) - return { - 'id': video_id, + info_dict.update({ 'display_id': display_id, - 'title': title, 'description': description, - 'thumbnail': thumbnail, - 'timestamp': timestamp, - 'uploader': uploader, - 'duration': duration, - 'formats': formats, - } + 'upload_date': upload_date, + }) + + return info_dict diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index 8baff2041..b17047b39 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -1,34 +1,42 @@ # coding: utf-8 from __future__ import unicode_literals +import calendar +import datetime import re from .common import InfoExtractor -from ..compat import compat_str +from ..compat import ( + compat_etree_fromstring, + compat_str, + compat_parse_qs, + compat_xml_parse_error, +) from ..utils import ( - int_or_none, - unescapeHTML, ExtractorError, + int_or_none, + float_or_none, xpath_text, ) class BiliBiliIE(InfoExtractor): - _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)(?:/index_(?P<page_num>\d+).html)?' + _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)' _TESTS = [{ 'url': 'http://www.bilibili.tv/video/av1074402/', - 'md5': '2c301e4dab317596e837c3e7633e7d86', + 'md5': '5f7d29e1a2872f3df0cf76b1f87d3788', 'info_dict': { 'id': '1554319', 'ext': 'flv', 'title': '【金坷垃】金泡沫', - 'duration': 308313, + 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923', + 'duration': 308.067, + 'timestamp': 1398012660, 'upload_date': '20140420', 'thumbnail': 're:^https?://.+\.jpg', - 'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923', - 'timestamp': 1397983878, 'uploader': '菊子桑', + 'uploader_id': '156160', }, }, { 'url': 'http://www.bilibili.com/video/av1041170/', @@ -36,75 +44,186 @@ class BiliBiliIE(InfoExtractor): 'id': '1041170', 'title': '【BD1080P】刀语【诸神&异域】', 'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦！~', - 'uploader': '枫叶逝去', - 'timestamp': 1396501299, }, 'playlist_count': 9, + }, { + 'url': 'http://www.bilibili.com/video/av4808130/', + 'info_dict': { + 'id': '4808130', + 'title': '【长篇】哆啦A梦443【钉铛】', + 'description': '(2016.05.27)来组合客人的脸吧&amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;illust_id=56912929', + }, + 'playlist': [{ + 'md5': '55cdadedf3254caaa0d5d27cf20a8f9c', + 'info_dict': { + 'id': '4808130_part1', + 'ext': 'flv', + 'title': '【长篇】哆啦A梦443【钉铛】', + 'description': '(2016.05.27)来组合客人的脸吧&amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;illust_id=56912929', + 'timestamp': 1464564180, + 'upload_date': '20160529', + 'uploader': '喜欢拉面', + 'uploader_id': '151066', + }, + }, { + 'md5': '926f9f67d0c482091872fbd8eca7ea3d', + 'info_dict': { + 'id': '4808130_part2', + 'ext': 'flv', + 'title': '【长篇】哆啦A梦443【钉铛】', + 'description': '(2016.05.27)来组合客人的脸吧&amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;illust_id=56912929', + 'timestamp': 1464564180, + 'upload_date': '20160529', + 'uploader': '喜欢拉面', + 'uploader_id': '151066', + }, + }, { + 'md5': '4b7b225b968402d7c32348c646f1fd83', + 'info_dict': { + 'id': '4808130_part3', + 'ext': 'flv', + 'title': '【长篇】哆啦A梦443【钉铛】', + 'description': '(2016.05.27)来组合客人的脸吧&amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;illust_id=56912929', + 'timestamp': 1464564180, + 'upload_date': '20160529', + 'uploader': '喜欢拉面', + 'uploader_id': '151066', + }, + }, { + 'md5': '7b795e214166501e9141139eea236e91', + 'info_dict': { + 'id': '4808130_part4', + 'ext': 'flv', + 'title': '【长篇】哆啦A梦443【钉铛】', + 'description': '(2016.05.27)来组合客人的脸吧&amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;illust_id=56912929', + 'timestamp': 1464564180, + 'upload_date': '20160529', + 'uploader': '喜欢拉面', + 'uploader_id': '151066', + }, + }], + }, { + # Missing upload time + 'url': 'http://www.bilibili.com/video/av1867637/', + 'info_dict': { + 'id': '2880301', + 'ext': 'flv', + 'title': '【HDTV】【喜剧】岳父岳母真难当 （2014）【法国票房冠军】', + 'description': '一个信奉天主教的法国旧式传统资产阶级家庭中有四个女儿。三个女儿却分别找了阿拉伯、犹太、中国丈夫，老夫老妻唯独期盼剩下未嫁的小女儿能找一个信奉天主教的法国白人，结果没想到小女儿找了一位非裔黑人……【这次应该不会跳帧了】', + 'uploader': '黑夜为猫', + 'uploader_id': '610729', + }, + 'params': { + # Just to test metadata extraction + 'skip_download': True, + }, + 'expected_warnings': ['upload time'], }] + # BiliBili blocks keys from time to time. The current key is extracted from + # the Android client + # TODO: find the sign algorithm used in the flash player + _APP_KEY = '86385cdc024c0f6c' + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - page_num = mobj.group('page_num') or '1' - view_data = self._download_json( - 'http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s&page=%s' % (video_id, page_num), - video_id) - if 'error' in view_data: - raise ExtractorError('%s said: %s' % (self.IE_NAME, view_data['error']), expected=True) + webpage = self._download_webpage(url, video_id) - cid = view_data['cid'] - title = unescapeHTML(view_data['title']) + params = compat_parse_qs(self._search_regex( + [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)', + r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'], + webpage, 'player parameters')) + cid = params['cid'][0] - doc = self._download_xml( - 'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid, - cid, - 'Downloading page %s/%s' % (page_num, view_data['pages']) - ) + info_xml_str = self._download_webpage( + 'http://interface.bilibili.com/v_cdn_play', + cid, query={'appkey': self._APP_KEY, 'cid': cid}, + note='Downloading video info page') + + err_msg = None + durls = None + info_xml = None + try: + info_xml = compat_etree_fromstring(info_xml_str.encode('utf-8')) + except compat_xml_parse_error: + info_json = self._parse_json(info_xml_str, video_id, fatal=False) + err_msg = (info_json or {}).get('error_text') + else: + err_msg = xpath_text(info_xml, './message') - if xpath_text(doc, './result') == 'error': - raise ExtractorError('%s said: %s' % (self.IE_NAME, xpath_text(doc, './message')), expected=True) + if info_xml is not None: + durls = info_xml.findall('./durl') + if not durls: + if err_msg: + raise ExtractorError('%s said: %s' % (self.IE_NAME, err_msg), expected=True) + else: + raise ExtractorError('No videos found!') entries = [] - for durl in doc.findall('./durl'): + for durl in durls: size = xpath_text(durl, ['./filesize', './size']) formats = [{ 'url': durl.find('./url').text, 'filesize': int_or_none(size), - 'ext': 'flv', }] - backup_urls = durl.find('./backup_url') - if backup_urls is not None: - for backup_url in backup_urls.findall('./url'): - formats.append({'url': backup_url.text}) - formats.reverse() + for backup_url in durl.findall('./backup_url/url'): + formats.append({ + 'url': backup_url.text, + # backup URLs have lower priorities + 'preference': -2 if 'hd.mp4' in backup_url.text else -3, + }) + + self._sort_formats(formats) entries.append({ 'id': '%s_part%s' % (cid, xpath_text(durl, './order')), - 'title': title, 'duration': int_or_none(xpath_text(durl, './length'), 1000), 'formats': formats, }) + title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title') + description = self._html_search_meta('description', webpage) + datetime_str = self._html_search_regex( + r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False) + timestamp = None + if datetime_str: + timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple()) + + # TODO 'view_count' requires deobfuscating Javascript info = { 'id': compat_str(cid), 'title': title, - 'description': view_data.get('description'), - 'thumbnail': view_data.get('pic'), - 'uploader': view_data.get('author'), - 'timestamp': int_or_none(view_data.get('created')), - 'view_count': int_or_none(view_data.get('play')), - 'duration': int_or_none(xpath_text(doc, './timelength')), + 'description': description, + 'timestamp': timestamp, + 'thumbnail': self._html_search_meta('thumbnailUrl', webpage), + 'duration': float_or_none(xpath_text(info_xml, './timelength'), scale=1000), } + uploader_mobj = re.search( + r'<a[^>]+href="https?://space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"', + webpage) + if uploader_mobj: + info.update({ + 'uploader': uploader_mobj.group('name'), + 'uploader_id': uploader_mobj.group('id'), + }) + + for entry in entries: + entry.update(info) + if len(entries) == 1: - entries[0].update(info) return entries[0] else: - info.update({ + for idx, entry in enumerate(entries): + entry['id'] = '%s_part%d' % (video_id, (idx + 1)) + + return { '_type': 'multi_video', 'id': video_id, + 'title': title, + 'description': description, 'entries': entries, - }) - return info + } diff --git a/youtube_dl/extractor/biqle.py b/youtube_dl/extractor/biqle.py new file mode 100644 index 000000000..ae4579b33 --- /dev/null +++ b/youtube_dl/extractor/biqle.py @@ -0,0 +1,39 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class BIQLEIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?biqle\.(?:com|org|ru)/watch/(?P<id>-?\d+_\d+)' + _TESTS = [{ + 'url': 'http://www.biqle.ru/watch/847655_160197695', + 'md5': 'ad5f746a874ccded7b8f211aeea96637', + 'info_dict': { + 'id': '160197695', + 'ext': 'mp4', + 'title': 'Foo Fighters - The Pretender (Live at Wembley Stadium)', + 'uploader': 'Andrey Rogozin', + 'upload_date': '20110605', + } + }, { + 'url': 'https://biqle.org/watch/-44781847_168547604', + 'md5': '7f24e72af1db0edf7c1aaba513174f97', + 'info_dict': { + 'id': '168547604', + 'ext': 'mp4', + 'title': 'Ребенок в шоке от автоматической мойки', + 'uploader': 'Dmitry Kotov', + } + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + embed_url = self._proto_relative_url(self._search_regex( + r'<iframe.+?src="((?:http:)?//daxab\.com/[^"]+)".*?></iframe>', webpage, 'embed url')) + + return { + '_type': 'url_transparent', + 'url': embed_url, + } diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 13343bc25..bd538be50 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -17,6 +17,9 @@ class BloombergIE(InfoExtractor): 'title': 'Shah\'s Presentation on Foreign-Exchange Strategies', 'description': 'md5:a8ba0302912d03d246979735c17d2761', }, + 'params': { + 'format': 'best[format_id^=hds]', + }, }, { 'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets', 'only_matching': True, diff --git a/youtube_dl/extractor/br.py b/youtube_dl/extractor/br.py index 11cf49851..ff0aa11b1 100644 --- a/youtube_dl/extractor/br.py +++ b/youtube_dl/extractor/br.py @@ -29,7 +29,8 @@ class BRIE(InfoExtractor): 'duration': 180, 'uploader': 'Reinhard Weber', 'upload_date': '20150422', - } + }, + 'skip': '404 not found', }, { 'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html', @@ -40,7 +41,8 @@ class BRIE(InfoExtractor): 'title': 'Manfred Schreiber ist tot', 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97', 'duration': 26, - } + }, + 'skip': '404 not found', }, { 'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html', @@ -51,7 +53,8 @@ class BRIE(InfoExtractor): 'title': 'Kurzweilig und sehr bewegend', 'description': 'md5:0351996e3283d64adeb38ede91fac54e', 'duration': 296, - } + }, + 'skip': '404 not found', }, { 'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html', diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index f0781fc27..57ce0c174 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -90,6 +90,7 @@ class BrightcoveLegacyIE(InfoExtractor): 'description': 'md5:363109c02998fee92ec02211bd8000df', 'uploader': 'National Ballet of Canada', }, + 'skip': 'Video gone', }, { # test flv videos served by akamaihd.net @@ -108,7 +109,7 @@ class BrightcoveLegacyIE(InfoExtractor): }, }, { - # playlist test + # playlist with 'videoList' # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL', 'info_dict': { @@ -117,6 +118,15 @@ class BrightcoveLegacyIE(InfoExtractor): }, 'playlist_mincount': 7, }, + { + # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965) + 'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg', + 'info_dict': { + 'id': '1522758701001', + 'title': 'Lesson 08', + }, + 'playlist_mincount': 10, + }, ] FLV_VCODECS = { 1: 'SORENSON', @@ -298,18 +308,25 @@ class BrightcoveLegacyIE(InfoExtractor): info_url, player_key, 'Downloading playlist information') json_data = json.loads(playlist_info) - if 'videoList' not in json_data: + if 'videoList' in json_data: + playlist_info = json_data['videoList'] + playlist_dto = playlist_info['mediaCollectionDTO'] + elif 'playlistTabs' in json_data: + playlist_info = json_data['playlistTabs'] + playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0] + else: raise ExtractorError('Empty playlist') - playlist_info = json_data['videoList'] - videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']] + + videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']] return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'], - playlist_title=playlist_info['mediaCollectionDTO']['displayName']) + playlist_title=playlist_dto['displayName']) def _extract_video_info(self, video_info): + video_id = compat_str(video_info['id']) publisher_id = video_info.get('publisherId') info = { - 'id': compat_str(video_info['id']), + 'id': video_id, 'title': video_info['displayName'].strip(), 'description': video_info.get('shortDescription'), 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'), @@ -331,7 +348,8 @@ class BrightcoveLegacyIE(InfoExtractor): url_comp = compat_urllib_parse_urlparse(url) if url_comp.path.endswith('.m3u8'): formats.extend( - self._extract_m3u8_formats(url, info['id'], 'mp4')) + self._extract_m3u8_formats( + url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) continue elif 'akamaihd.net' in url_comp.netloc: # This type of renditions are served through @@ -365,7 +383,7 @@ class BrightcoveLegacyIE(InfoExtractor): a_format.update({ 'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''), 'ext': 'mp4', - 'protocol': 'm3u8', + 'protocol': 'm3u8_native', }) formats.append(a_format) @@ -395,7 +413,7 @@ class BrightcoveLegacyIE(InfoExtractor): return ad_info if 'url' not in info and not info.get('formats'): - raise ExtractorError('Unable to extract video url for %s' % info['id']) + raise ExtractorError('Unable to extract video url for %s' % video_id) return info @@ -442,6 +460,10 @@ class BrightcoveNewIE(InfoExtractor): # non numeric ref: prefixed video id 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356', 'only_matching': True, + }, { + # unavailable video without message but with error_code + 'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001', + 'only_matching': True, }] @staticmethod @@ -512,8 +534,9 @@ class BrightcoveNewIE(InfoExtractor): }) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: - json_data = self._parse_json(e.cause.read().decode(), video_id) - raise ExtractorError(json_data[0]['message'], expected=True) + json_data = self._parse_json(e.cause.read().decode(), video_id)[0] + raise ExtractorError( + json_data.get('message') or json_data['error_code'], expected=True) raise title = json_data['name'].strip() @@ -527,7 +550,7 @@ class BrightcoveNewIE(InfoExtractor): if not src: continue formats.extend(self._extract_m3u8_formats( - src, video_id, 'mp4', m3u8_id='hls', fatal=False)) + src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) elif source_type == 'application/dash+xml': if not src: continue @@ -578,6 +601,13 @@ class BrightcoveNewIE(InfoExtractor): 'format_id': build_format_id('rtmp'), }) formats.append(f) + + errors = json_data.get('errors') + if not formats and errors: + error = errors[0] + raise ExtractorError( + error.get('message') or error.get('error_subcode') or error['error_code'], expected=True) + self._sort_formats(formats) subtitles = {} diff --git a/youtube_dl/extractor/buzzfeed.py b/youtube_dl/extractor/buzzfeed.py index df503ecc0..75fa92d7c 100644 --- a/youtube_dl/extractor/buzzfeed.py +++ b/youtube_dl/extractor/buzzfeed.py @@ -5,6 +5,7 @@ import json import re from .common import InfoExtractor +from .facebook import FacebookIE class BuzzFeedIE(InfoExtractor): @@ -20,11 +21,11 @@ class BuzzFeedIE(InfoExtractor): 'info_dict': { 'id': 'aVCR29aE_OQ', 'ext': 'mp4', + 'title': 'Angry Ram destroys a punching bag..', + 'description': 'md5:c59533190ef23fd4458a5e8c8c872345', 'upload_date': '20141024', 'uploader_id': 'Buddhanz1', - 'description': 'He likes to stay in shape with his heavy bag, he wont stop until its on the ground\n\nFollow Angry Ram on Facebook for regular updates -\nhttps://www.facebook.com/pages/Angry-Ram/1436897249899558?ref=hl', - 'uploader': 'Buddhanz', - 'title': 'Angry Ram destroys a punching bag', + 'uploader': 'Angry Ram', } }] }, { @@ -41,13 +42,30 @@ class BuzzFeedIE(InfoExtractor): 'info_dict': { 'id': 'mVmBL8B-In0', 'ext': 'mp4', + 'title': 're:Munchkin the Teddy Bear gets her exercise', + 'description': 'md5:28faab95cda6e361bcff06ec12fc21d8', 'upload_date': '20141124', 'uploader_id': 'CindysMunchkin', - 'description': 're:© 2014 Munchkin the', 'uploader': 're:^Munchkin the', - 'title': 're:Munchkin the Teddy Bear gets her exercise', }, }] + }, { + 'url': 'http://www.buzzfeed.com/craigsilverman/the-most-adorable-crash-landing-ever#.eq7pX0BAmK', + 'info_dict': { + 'id': 'the-most-adorable-crash-landing-ever', + 'title': 'Watch This Baby Goose Make The Most Adorable Crash Landing', + 'description': 'This gosling knows how to stick a landing.', + }, + 'playlist': [{ + 'md5': '763ca415512f91ca62e4621086900a23', + 'info_dict': { + 'id': '971793786185728', + 'ext': 'mp4', + 'title': 'We set up crash pads so that the goslings on our roof would have a safe landi...', + 'uploader': 'Calgary Outdoor Centre-University of Calgary', + }, + }], + 'add_ie': ['Facebook'], }] def _real_extract(self, url): @@ -66,6 +84,10 @@ class BuzzFeedIE(InfoExtractor): continue entries.append(self.url_result(video['url'])) + facebook_url = FacebookIE._extract_url(webpage) + if facebook_url: + entries.append(self.url_result(facebook_url)) + return { '_type': 'playlist', 'id': playlist_id, diff --git a/youtube_dl/extractor/byutv.py b/youtube_dl/extractor/byutv.py index dda98059e..3aec601f8 100644 --- a/youtube_dl/extractor/byutv.py +++ b/youtube_dl/extractor/byutv.py @@ -11,6 +11,7 @@ class BYUtvIE(InfoExtractor): _VALID_URL = r'^https?://(?:www\.)?byutv.org/watch/[0-9a-f-]+/(?P<video_id>[^/?#]+)' _TEST = { 'url': 'http://www.byutv.org/watch/6587b9a3-89d2-42a6-a7f7-fd2f81840a7d/studio-c-season-5-episode-5', + 'md5': '05850eb8c749e2ee05ad5a1c34668493', 'info_dict': { 'id': 'studio-c-season-5-episode-5', 'ext': 'mp4', @@ -21,7 +22,8 @@ class BYUtvIE(InfoExtractor): }, 'params': { 'skip_download': True, - } + }, + 'add_ie': ['Ooyala'], } def _real_extract(self, url): diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 25b2d4efe..61463f249 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -4,11 +4,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_urllib_parse_urlparse from ..utils import ( ExtractorError, HEADRequest, unified_strdate, - url_basename, qualities, int_or_none, ) @@ -16,24 +16,38 @@ from ..utils import ( class CanalplusIE(InfoExtractor): IE_DESC = 'canalplus.fr, piwiplus.fr and d8.tv' - _VALID_URL = r'https?://(?:www\.(?P<site>canalplus\.fr|piwiplus\.fr|d8\.tv|itele\.fr)/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>[0-9]+))' + _VALID_URL = r'''(?x) + https?:// + (?: + (?: + (?:(?:www|m)\.)?canalplus\.fr| + (?:www\.)?piwiplus\.fr| + (?:www\.)?d8\.tv| + (?:www\.)?d17\.tv| + (?:www\.)?itele\.fr + )/(?:(?:[^/]+/)*(?P<display_id>[^/?#&]+))?(?:\?.*\bvid=(?P<vid>\d+))?| + player\.canalplus\.fr/#/(?P<id>\d+) + ) + + ''' _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s?format=json' _SITE_ID_MAP = { - 'canalplus.fr': 'cplus', - 'piwiplus.fr': 'teletoon', - 'd8.tv': 'd8', - 'itele.fr': 'itele', + 'canalplus': 'cplus', + 'piwiplus': 'teletoon', + 'd8': 'd8', + 'd17': 'd17', + 'itele': 'itele', } _TESTS = [{ - 'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1263092', - 'md5': '12164a6f14ff6df8bd628e8ba9b10b78', + 'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814', + 'md5': '41f438a4904f7664b91b4ed0dec969dc', 'info_dict': { - 'id': '1263092', + 'id': '1192814', 'ext': 'mp4', - 'title': 'Le Zapping - 13/05/15', - 'description': 'md5:09738c0d06be4b5d06a0940edb0da73f', - 'upload_date': '20150513', + 'title': "L'Année du Zapping 2014 - L'Année du Zapping 2014", + 'description': "Toute l'année 2014 dans un Zapping exceptionnel !", + 'upload_date': '20150105', }, }, { 'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190', @@ -46,35 +60,45 @@ class CanalplusIE(InfoExtractor): }, 'skip': 'Only works from France', }, { - 'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html', + 'url': 'http://www.d8.tv/d8-docs-mags/pid5198-d8-en-quete-d-actualite.html?vid=1390231', 'info_dict': { - 'id': '966289', - 'ext': 'flv', - 'title': 'Campagne intime - Documentaire exceptionnel', - 'description': 'md5:d2643b799fb190846ae09c61e59a859f', - 'upload_date': '20131108', + 'id': '1390231', + 'ext': 'mp4', + 'title': "Vacances pas chères : prix discount ou grosses dépenses ? - En quête d'actualité", + 'description': 'md5:edb6cf1cb4a1e807b5dd089e1ac8bfc6', + 'upload_date': '20160512', + }, + 'params': { + 'skip_download': True, }, - 'skip': 'videos get deleted after a while', }, { - 'url': 'http://www.itele.fr/france/video/aubervilliers-un-lycee-en-colere-111559', - 'md5': '38b8f7934def74f0d6f3ba6c036a5f82', + 'url': 'http://www.itele.fr/chroniques/invite-bruce-toussaint/thierry-solere-nicolas-sarkozy-officialisera-sa-candidature-a-la-primaire-quand-il-le-voudra-167224', 'info_dict': { - 'id': '1213714', + 'id': '1398334', 'ext': 'mp4', - 'title': 'Aubervilliers : un lycée en colère - Le 11/02/2015 à 06h45', - 'description': 'md5:8216206ec53426ea6321321f3b3c16db', - 'upload_date': '20150211', + 'title': "L'invité de Bruce Toussaint du 07/06/2016 - ", + 'description': 'md5:40ac7c9ad0feaeb6f605bad986f61324', + 'upload_date': '20160607', }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://m.canalplus.fr/?vid=1398231', + 'only_matching': True, + }, { + 'url': 'http://www.d17.tv/emissions/pid8303-lolywood.html?vid=1397061', + 'only_matching': True, }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.groupdict().get('id') + video_id = mobj.groupdict().get('id') or mobj.groupdict().get('vid') - site_id = self._SITE_ID_MAP[mobj.group('site') or 'canal'] + site_id = self._SITE_ID_MAP[compat_urllib_parse_urlparse(url).netloc.rsplit('.', 2)[-2]] # Beware, some subclasses do not define an id group - display_id = url_basename(mobj.group('path')) + display_id = mobj.group('display_id') or video_id if video_id is None: webpage = self._download_webpage(url, display_id) diff --git a/youtube_dl/extractor/carambatv.py b/youtube_dl/extractor/carambatv.py new file mode 100644 index 000000000..5797fb951 --- /dev/null +++ b/youtube_dl/extractor/carambatv.py @@ -0,0 +1,88 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + float_or_none, + int_or_none, + try_get, +) + + +class CarambaTVIE(InfoExtractor): + _VALID_URL = r'(?:carambatv:|https?://video1\.carambatv\.ru/v/)(?P<id>\d+)' + _TESTS = [{ + 'url': 'http://video1.carambatv.ru/v/191910501', + 'md5': '2f4a81b7cfd5ab866ee2d7270cb34a2a', + 'info_dict': { + 'id': '191910501', + 'ext': 'mp4', + 'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 2678.31, + }, + }, { + 'url': 'carambatv:191910501', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'http://video1.carambatv.ru/v/%s/videoinfo.js' % video_id, + video_id) + + title = video['title'] + + base_url = video.get('video') or 'http://video1.carambatv.ru/v/%s/' % video_id + + formats = [{ + 'url': base_url + f['fn'], + 'height': int_or_none(f.get('height')), + 'format_id': '%sp' % f['height'] if f.get('height') else None, + } for f in video['qualities'] if f.get('fn')] + self._sort_formats(formats) + + thumbnail = video.get('splash') + duration = float_or_none(try_get( + video, lambda x: x['annotations'][0]['end_time'], compat_str)) + + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + } + + +class CarambaTVPageIE(InfoExtractor): + _VALID_URL = r'https?://carambatv\.ru/(?:[^/]+/)+(?P<id>[^/?#&]+)' + _TEST = { + 'url': 'http://carambatv.ru/movie/bad-comedian/razborka-v-manile/', + 'md5': '', + 'info_dict': { + 'id': '191910501', + 'ext': 'mp4', + 'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 2678.31, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + video_url = self._og_search_property('video:iframe', webpage, default=None) + + if not video_url: + video_id = self._search_regex( + r'(?:video_id|crmb_vuid)\s*[:=]\s*["\']?(\d+)', + webpage, 'video id') + video_url = 'carambatv:%s' % video_id + + return self.url_result(video_url, CarambaTVIE.ie_key()) diff --git a/youtube_dl/extractor/cbc.py b/youtube_dl/extractor/cbc.py index d8aa31038..ff663d079 100644 --- a/youtube_dl/extractor/cbc.py +++ b/youtube_dl/extractor/cbc.py @@ -4,64 +4,66 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import js_to_json +from ..utils import ( + js_to_json, + smuggle_url, +) class CBCIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:[^/]+/)+(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)' _TESTS = [{ # with mediaId 'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs', + 'md5': '97e24d09672fc4cf56256d6faa6c25bc', 'info_dict': { 'id': '2682904050', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Don Cherry – All-Stars', 'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.', - 'timestamp': 1454475540, + 'timestamp': 1454463000, 'upload_date': '20160203', - }, - 'params': { - # rtmp download - 'skip_download': True, + 'uploader': 'CBCC-NEW', }, }, { # with clipId 'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live', + 'md5': '0274a90b51a9b4971fe005c63f592f12', 'info_dict': { 'id': '2487345465', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Robin Williams freestyles on 90 Minutes Live', 'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.', - 'upload_date': '19700101', - }, - 'params': { - # rtmp download - 'skip_download': True, + 'upload_date': '19780210', + 'uploader': 'CBCC-NEW', + 'timestamp': 255977160, }, }, { # multiple iframes 'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot', 'playlist': [{ + 'md5': '377572d0b49c4ce0c9ad77470e0b96b4', 'info_dict': { 'id': '2680832926', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'An Eagle\'s-Eye View Off Burrard Bridge', 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.', - 'upload_date': '19700101', + 'upload_date': '20160201', + 'timestamp': 1454342820, + 'uploader': 'CBCC-NEW', }, }, { + 'md5': '415a0e3f586113894174dfb31aa5bb1a', 'info_dict': { 'id': '2658915080', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Fly like an eagle!', 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower', - 'upload_date': '19700101', + 'upload_date': '20150315', + 'timestamp': 1426443984, + 'uploader': 'CBCC-NEW', }, }], - 'params': { - # rtmp download - 'skip_download': True, - }, }] @classmethod @@ -90,24 +92,54 @@ class CBCIE(InfoExtractor): class CBCPlayerIE(InfoExtractor): _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.cbc.ca/player/play/2683190193', + 'md5': '64d25f841ddf4ddb28a235338af32e2c', 'info_dict': { 'id': '2683190193', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Gerry Runs a Sweat Shop', 'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0', - 'timestamp': 1455067800, + 'timestamp': 1455071400, 'upload_date': '20160210', + 'uploader': 'CBCC-NEW', }, - 'params': { - # rtmp download - 'skip_download': True, + }, { + # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/ + 'url': 'http://www.cbc.ca/player/play/2657631896', + 'md5': 'e5e708c34ae6fca156aafe17c43e8b75', + 'info_dict': { + 'id': '2657631896', + 'ext': 'mp3', + 'title': 'CBC Montreal is organizing its first ever community hackathon!', + 'description': 'The modern technology we tend to depend on so heavily, is never without it\'s share of hiccups and headaches. Next weekend - CBC Montreal will be getting members of the public for its first Hackathon.', + 'timestamp': 1425704400, + 'upload_date': '20150307', + 'uploader': 'CBCC-NEW', }, - } + }, { + # available only when we add `formats=MPEG4,FLV,MP3` to theplatform url + 'url': 'http://www.cbc.ca/player/play/2164402062', + 'md5': '17a61eb813539abea40618d6323a7f82', + 'info_dict': { + 'id': '2164402062', + 'ext': 'flv', + 'title': 'Cancer survivor four times over', + 'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.', + 'timestamp': 1320410746, + 'upload_date': '20111104', + 'uploader': 'CBCC-NEW', + }, + }] def _real_extract(self, url): video_id = self._match_id(url) - return self.url_result( - 'http://feed.theplatform.com/f/ExhSPC/vms_5akSXx4Ng_Zn?byGuid=%s' % video_id, - 'ThePlatformFeed', video_id) + return { + '_type': 'url_transparent', + 'ie_key': 'ThePlatform', + 'url': smuggle_url( + 'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, { + 'force_smil_url': True + }), + 'id': video_id, + } diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index 051d783a2..a23173d6f 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -1,15 +1,13 @@ from __future__ import unicode_literals -from .theplatform import ThePlatformIE +from .theplatform import ThePlatformFeedIE from ..utils import ( - xpath_text, - xpath_element, int_or_none, find_xpath_attr, ) -class CBSBaseIE(ThePlatformIE): +class CBSBaseIE(ThePlatformFeedIE): def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL') return { @@ -19,9 +17,22 @@ class CBSBaseIE(ThePlatformIE): }] } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else [] + def _extract_video_info(self, filter_query, video_id): + return self._extract_feed_info( + 'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: { + 'series': entry.get('cbs$SeriesTitle'), + 'season_number': int_or_none(entry.get('cbs$SeasonNumber')), + 'episode': entry.get('cbs$EpisodeTitle'), + 'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')), + }, { + 'StreamPack': { + 'manifest': 'm3u', + } + }) + class CBSIE(CBSBaseIE): - _VALID_URL = r'https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<id>[^/]+)' + _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)' _TESTS = [{ 'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', @@ -36,25 +47,7 @@ class CBSIE(CBSBaseIE): 'upload_date': '20131127', 'uploader': 'CBSI-NEW', }, - 'params': { - # rtmp download - 'skip_download': True, - }, - '_skip': 'Blocked outside the US', - }, { - 'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/', - 'info_dict': { - 'id': 'WWF_5KqY3PK1', - 'display_id': 'st-vincent', - 'ext': 'flv', - 'title': 'Live on Letterman - St. Vincent', - 'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.', - 'duration': 3221, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, + 'expected_warnings': ['Failed to download m3u8 information'], '_skip': 'Blocked outside the US', }, { 'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/', @@ -66,43 +59,5 @@ class CBSIE(CBSBaseIE): TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true' def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - content_id = self._search_regex( - [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"], - webpage, 'content id') - items_data = self._download_xml( - 'http://can.cbs.com/thunder/player/videoPlayerService.php', - content_id, query={'partner': 'cbs', 'contentId': content_id}) - video_data = xpath_element(items_data, './/item') - title = xpath_text(video_data, 'videoTitle', 'title', True) - - subtitles = {} - formats = [] - for item in items_data.findall('.//item'): - pid = xpath_text(item, 'pid') - if not pid: - continue - tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid - if '.m3u8' in xpath_text(item, 'contentUrl', default=''): - tp_release_url += '&manifest=m3u' - tp_formats, tp_subtitles = self._extract_theplatform_smil( - tp_release_url, content_id, 'Downloading %s SMIL data' % pid) - formats.extend(tp_formats) - subtitles = self._merge_subtitles(subtitles, tp_subtitles) - self._sort_formats(formats) - - info = self.get_metadata('dJ5BDC/media/guid/2198311517/%s' % content_id, content_id) - info.update({ - 'id': content_id, - 'display_id': display_id, - 'title': title, - 'series': xpath_text(video_data, 'seriesTitle'), - 'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')), - 'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')), - 'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000), - 'thumbnail': xpath_text(video_data, 'previewImageURL'), - 'formats': formats, - 'subtitles': subtitles, - }) - return info + content_id = self._match_id(url) + return self._extract_video_info('byGuid=%s' % content_id, content_id) diff --git a/youtube_dl/extractor/cbsinteractive.py b/youtube_dl/extractor/cbsinteractive.py index 0011c3029..821db20b2 100644 --- a/youtube_dl/extractor/cbsinteractive.py +++ b/youtube_dl/extractor/cbsinteractive.py @@ -80,9 +80,6 @@ class CBSInteractiveIE(ThePlatformIE): media_guid_path = 'media/guid/%d/%s' % (self.MPX_ACCOUNTS[site], vdata['mpxRefId']) formats, subtitles = [], {} - if site == 'cnet': - formats, subtitles = self._extract_theplatform_smil( - self.TP_RELEASE_URL_TEMPLATE % media_guid_path, video_id) for (fkey, vid) in vdata['files'].items(): if fkey == 'hls_phone' and 'hls_tablet' in vdata['files']: continue @@ -94,7 +91,7 @@ class CBSInteractiveIE(ThePlatformIE): subtitles = self._merge_subtitles(subtitles, tp_subtitles) self._sort_formats(formats) - info = self.get_metadata('kYEXFC/%s' % media_guid_path, video_id) + info = self._extract_theplatform_metadata('kYEXFC/%s' % media_guid_path, video_id) info.update({ 'id': video_id, 'display_id': display_id, diff --git a/youtube_dl/extractor/cbslocal.py b/youtube_dl/extractor/cbslocal.py new file mode 100644 index 000000000..74adb38a6 --- /dev/null +++ b/youtube_dl/extractor/cbslocal.py @@ -0,0 +1,84 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import calendar +import datetime + +from .anvato import AnvatoIE +from .sendtonews import SendtoNewsIE +from ..compat import compat_urlparse + + +class CBSLocalIE(AnvatoIE): + _VALID_URL = r'https?://[a-z]+\.cbslocal\.com/\d+/\d+/\d+/(?P<id>[0-9a-z-]+)' + + _TESTS = [{ + # Anvato backend + 'url': 'http://losangeles.cbslocal.com/2016/05/16/safety-advocates-say-fatal-car-seat-failures-are-public-health-crisis', + 'md5': 'f0ee3081e3843f575fccef901199b212', + 'info_dict': { + 'id': '3401037', + 'ext': 'mp4', + 'title': 'Safety Advocates Say Fatal Car Seat Failures Are \'Public Health Crisis\'', + 'description': 'Collapsing seats have been the focus of scrutiny for decades, though experts say remarkably little has been done to address the issue. Randy Paige reports.', + 'thumbnail': 're:^https?://.*', + 'timestamp': 1463440500, + 'upload_date': '20160516', + 'subtitles': { + 'en': 'mincount:5', + }, + 'categories': [ + 'Stations\\Spoken Word\\KCBSTV', + 'Syndication\\MSN', + 'Syndication\\NDN', + 'Syndication\\AOL', + 'Syndication\\Yahoo', + 'Syndication\\Tribune', + 'Syndication\\Curb.tv', + 'Content\\News' + ], + }, + }, { + # SendtoNews embed + 'url': 'http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/', + 'info_dict': { + 'id': 'GxfCe0Zo7D-175909-5588', + 'ext': 'mp4', + 'title': 'Recap: CLE 15, CIN 6', + 'description': '5/16/16: Indians\' bats explode for 15 runs in a win', + 'upload_date': '20160516', + 'timestamp': 1463433840, + 'duration': 49, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + + sendtonews_url = SendtoNewsIE._extract_url(webpage) + if sendtonews_url: + info_dict = { + '_type': 'url_transparent', + 'url': compat_urlparse.urljoin(url, sendtonews_url), + } + else: + info_dict = self._extract_anvato_videos(webpage, display_id) + + time_str = self._html_search_regex( + r'class="entry-date">([^<]+)<', webpage, 'released date', fatal=False) + timestamp = None + if time_str: + timestamp = calendar.timegm(datetime.datetime.strptime( + time_str, '%b %d, %Y %I:%M %p').timetuple()) + + info_dict.update({ + 'display_id': display_id, + 'timestamp': timestamp, + }) + + return info_dict diff --git a/youtube_dl/extractor/cbsnews.py b/youtube_dl/extractor/cbsnews.py index 79ddc20a0..387537e76 100644 --- a/youtube_dl/extractor/cbsnews.py +++ b/youtube_dl/extractor/cbsnews.py @@ -30,9 +30,12 @@ class CBSNewsIE(CBSBaseIE): { 'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/', 'info_dict': { - 'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack', + 'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y', 'ext': 'mp4', 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack', + 'description': 'md5:4a6983e480542d8b333a947bfc64ddc7', + 'upload_date': '19700101', + 'uploader': 'CBSI-NEW', 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 205, 'subtitles': { @@ -58,30 +61,8 @@ class CBSNewsIE(CBSBaseIE): webpage, 'video JSON info'), video_id) item = video_info['item'] if 'item' in video_info else video_info - title = item.get('articleTitle') or item.get('hed') - duration = item.get('duration') - thumbnail = item.get('mediaImage') or item.get('thumbnail') - - subtitles = {} - formats = [] - for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']: - pid = item.get('media' + format_id) - if not pid: - continue - release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true' % pid - tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid) - formats.extend(tp_formats) - subtitles = self._merge_subtitles(subtitles, tp_subtitles) - self._sort_formats(formats) - - return { - 'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - } + guid = item['mpxRefId'] + return self._extract_video_info('byGuid=%s' % guid, guid) class CBSNewsLiveVideoIE(InfoExtractor): diff --git a/youtube_dl/extractor/cbssports.py b/youtube_dl/extractor/cbssports.py index 549ae32f3..78ca44b02 100644 --- a/youtube_dl/extractor/cbssports.py +++ b/youtube_dl/extractor/cbssports.py @@ -1,30 +1,28 @@ from __future__ import unicode_literals -import re +from .cbs import CBSBaseIE -from .common import InfoExtractor +class CBSSportsIE(CBSBaseIE): + _VALID_URL = r'https?://www\.cbssports\.com/video/player/[^/]+/(?P<id>\d+)' -class CBSSportsIE(InfoExtractor): - _VALID_URL = r'https?://www\.cbssports\.com/video/player/(?P<section>[^/]+)/(?P<id>[^/]+)' - - _TEST = { - 'url': 'http://www.cbssports.com/video/player/tennis/318462531970/0/us-open-flashbacks-1990s', + _TESTS = [{ + 'url': 'http://www.cbssports.com/video/player/videos/708337219968/0/ben-simmons-the-next-lebron?-not-so-fast', 'info_dict': { - 'id': '_d5_GbO8p1sT', - 'ext': 'flv', - 'title': 'US Open flashbacks: 1990s', - 'description': 'Bill Macatee relives the best moments in US Open history from the 1990s.', + 'id': '708337219968', + 'ext': 'mp4', + 'title': 'Ben Simmons the next LeBron? Not so fast', + 'description': 'md5:854294f627921baba1f4b9a990d87197', + 'timestamp': 1466293740, + 'upload_date': '20160618', + 'uploader': 'CBSI-NEW', }, - } + 'params': { + # m3u8 download + 'skip_download': True, + } + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - section = mobj.group('section') - video_id = mobj.group('id') - all_videos = self._download_json( - 'http://www.cbssports.com/data/video/player/getVideos/%s?as=json' % section, - video_id) - # The json file contains the info of all the videos in the section - video_info = next(v for v in all_videos if v['pcid'] == video_id) - return self.url_result('theplatform:%s' % video_info['pid'], 'ThePlatform') + video_id = self._match_id(url) + return self._extract_video_info('byId=%s' % video_id, video_id) diff --git a/youtube_dl/extractor/ccc.py b/youtube_dl/extractor/ccc.py index dda2c0959..8f7f09e22 100644 --- a/youtube_dl/extractor/ccc.py +++ b/youtube_dl/extractor/ccc.py @@ -1,13 +1,9 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( int_or_none, - parse_duration, - qualities, - unified_strdate, + parse_iso8601, ) @@ -19,14 +15,14 @@ class CCCIE(InfoExtractor): 'url': 'https://media.ccc.de/v/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor#video', 'md5': '3a1eda8f3a29515d27f5adb967d7e740', 'info_dict': { - 'id': '30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor', + 'id': '1839', 'ext': 'mp4', 'title': 'Introduction to Processor Design', - 'description': 'md5:80be298773966f66d56cb11260b879af', + 'description': 'md5:df55f6d073d4ceae55aae6f2fd98a0ac', 'thumbnail': 're:^https?://.*\.jpg$', - 'view_count': int, 'upload_date': '20131228', - 'duration': 3660, + 'timestamp': 1388188800, + 'duration': 3710, } }, { 'url': 'https://media.ccc.de/v/32c3-7368-shopshifting#download', @@ -34,79 +30,48 @@ class CCCIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - if self._downloader.params.get('prefer_free_formats'): - preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd']) - else: - preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd']) - - title = self._html_search_regex( - r'(?s)<h1>(.*?)</h1>', webpage, 'title') - description = self._html_search_regex( - r'(?s)<h3>About</h3>(.+?)<h3>', - webpage, 'description', fatal=False) - upload_date = unified_strdate(self._html_search_regex( - r"(?s)<span[^>]+class='[^']*fa-calendar-o'[^>]*>(.+?)</span>", - webpage, 'upload date', fatal=False)) - view_count = int_or_none(self._html_search_regex( - r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>", - webpage, 'view count', fatal=False)) - duration = parse_duration(self._html_search_regex( - r'(?s)<span[^>]+class=(["\']).*?fa-clock-o.*?\1[^>]*></span>(?P<duration>.+?)</li', - webpage, 'duration', fatal=False, group='duration')) + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + event_id = self._search_regex("data-id='(\d+)'", webpage, 'event id') + event_data = self._download_json('https://media.ccc.de/public/events/%s' % event_id, event_id) - matches = re.finditer(r'''(?xs) - <(?:span|div)\s+class='label\s+filetype'>(?P<format>[^<]*)</(?:span|div)>\s* - <(?:span|div)\s+class='label\s+filetype'>(?P<lang>[^<]*)</(?:span|div)>\s* - <a\s+download\s+href='(?P<http_url>[^']+)'>\s* - (?: - .*? - <a\s+(?:download\s+)?href='(?P<torrent_url>[^']+\.torrent)' - )?''', webpage) formats = [] - for m in matches: - format = m.group('format') - format_id = self._search_regex( - r'.*/([a-z0-9_-]+)/[^/]*$', - m.group('http_url'), 'format id', default=None) - if format_id: - format_id = m.group('lang') + '-' + format_id - vcodec = 'h264' if 'h264' in format_id else ( - 'none' if format_id in ('mp3', 'opus') else None + for recording in event_data.get('recordings', []): + recording_url = recording.get('recording_url') + if not recording_url: + continue + language = recording.get('language') + folder = recording.get('folder') + format_id = None + if language: + format_id = language + if folder: + if language: + format_id += '-' + folder + else: + format_id = folder + vcodec = 'h264' if 'h264' in folder else ( + 'none' if folder in ('mp3', 'opus') else None ) formats.append({ 'format_id': format_id, - 'format': format, - 'language': m.group('lang'), - 'url': m.group('http_url'), + 'url': recording_url, + 'width': int_or_none(recording.get('width')), + 'height': int_or_none(recording.get('height')), + 'filesize': int_or_none(recording.get('size'), invscale=1024 * 1024), + 'language': language, 'vcodec': vcodec, - 'preference': preference(format_id), }) - - if m.group('torrent_url'): - formats.append({ - 'format_id': 'torrent-%s' % (format if format_id is None else format_id), - 'format': '%s (torrent)' % format, - 'proto': 'torrent', - 'format_note': '(unsupported; will just download the .torrent file)', - 'vcodec': vcodec, - 'preference': -100 + preference(format_id), - 'url': m.group('torrent_url'), - }) self._sort_formats(formats) - thumbnail = self._html_search_regex( - r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False) - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'view_count': view_count, - 'upload_date': upload_date, - 'duration': duration, + 'id': event_id, + 'display_id': display_id, + 'title': event_data['title'], + 'description': event_data.get('description'), + 'thumbnail': event_data.get('thumb_url'), + 'timestamp': parse_iso8601(event_data.get('date')), + 'duration': int_or_none(event_data.get('length')), + 'tags': event_data.get('tags'), 'formats': formats, } diff --git a/youtube_dl/extractor/cda.py b/youtube_dl/extractor/cda.py index 498d2c0d8..8af318703 100755 --- a/youtube_dl/extractor/cda.py +++ b/youtube_dl/extractor/cda.py @@ -58,7 +58,8 @@ class CDAIE(InfoExtractor): def extract_format(page, version): unpacked = decode_packed_codes(page) format_url = self._search_regex( - r"url:\\'(.+?)\\'", unpacked, '%s url' % version, fatal=False) + r"(?:file|url)\s*:\s*(\\?[\"'])(?P<url>http.+?)\1", unpacked, + '%s url' % version, fatal=False, group='url') if not format_url: return f = { @@ -75,7 +76,8 @@ class CDAIE(InfoExtractor): info_dict['formats'].append(f) if not info_dict['duration']: info_dict['duration'] = parse_duration(self._search_regex( - r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False)) + r"duration\s*:\s*(\\?[\"'])(?P<duration>.+?)\1", + unpacked, 'duration', fatal=False, group='duration')) extract_format(webpage, 'default') diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index 6652c8e42..5a58d1777 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -33,20 +33,34 @@ class CeskaTelevizeIE(InfoExtractor): 'skip_download': True, }, }, { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + 'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en', 'info_dict': { - 'id': '61924494876844374', + 'id': '61924494877028507', 'ext': 'mp4', - 'title': 'První republika: Zpěvačka z Dupárny Bobina', - 'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.', + 'title': 'Hyde Park Civilizace: Bonus 01 - En', + 'description': 'English Subtittles', 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 88.4, + 'duration': 81.3, }, 'params': { # m3u8 download 'skip_download': True, }, }, { + # live stream + 'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/', + 'info_dict': { + 'id': 402, + 'ext': 'mp4', + 'title': 're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + 'skip': 'Georestricted to Czech Republic', + }, { # video with 18+ caution trailer 'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/', 'info_dict': { @@ -118,19 +132,21 @@ class CeskaTelevizeIE(InfoExtractor): req = sanitized_Request(compat_urllib_parse_unquote(playlist_url)) req.add_header('Referer', url) - playlist_title = self._og_search_title(webpage) - playlist_description = self._og_search_description(webpage) + playlist_title = self._og_search_title(webpage, default=None) + playlist_description = self._og_search_description(webpage, default=None) playlist = self._download_json(req, playlist_id)['playlist'] playlist_len = len(playlist) entries = [] for item in playlist: + is_live = item.get('type') == 'LIVE' formats = [] for format_id, stream_url in item['streamUrls'].items(): formats.extend(self._extract_m3u8_formats( stream_url, playlist_id, 'mp4', - entry_protocol='m3u8_native', fatal=False)) + entry_protocol='m3u8' if is_live else 'm3u8_native', + fatal=False)) self._sort_formats(formats) item_id = item.get('id') or item['assetId'] @@ -145,14 +161,22 @@ class CeskaTelevizeIE(InfoExtractor): if subs: subtitles = self.extract_subtitles(episode_id, subs) + if playlist_len == 1: + final_title = playlist_title or title + if is_live: + final_title = self._live_title(final_title) + else: + final_title = '%s (%s)' % (playlist_title, title) + entries.append({ 'id': item_id, - 'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title), + 'title': final_title, 'description': playlist_description if playlist_len == 1 else None, 'thumbnail': thumbnail, 'duration': duration, 'formats': formats, 'subtitles': subtitles, + 'is_live': is_live, }) return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py index c74553dcf..34d4e6156 100644 --- a/youtube_dl/extractor/channel9.py +++ b/youtube_dl/extractor/channel9.py @@ -20,54 +20,64 @@ class Channel9IE(InfoExtractor): ''' IE_DESC = 'Channel 9' IE_NAME = 'channel9' - _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?' - - _TESTS = [ - { - 'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002', - 'md5': 'bbd75296ba47916b754e73c3a4bbdf10', - 'info_dict': { - 'id': 'Events/TechEd/Australia/2013/KOS002', - 'ext': 'mp4', - 'title': 'Developer Kick-Off Session: Stuff We Love', - 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f', - 'duration': 4576, - 'thumbnail': 're:http://.*\.jpg', - 'session_code': 'KOS002', - 'session_day': 'Day 1', - 'session_room': 'Arena 1A', - 'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen'], - }, + _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)' + + _TESTS = [{ + 'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002', + 'md5': 'bbd75296ba47916b754e73c3a4bbdf10', + 'info_dict': { + 'id': 'Events/TechEd/Australia/2013/KOS002', + 'ext': 'mp4', + 'title': 'Developer Kick-Off Session: Stuff We Love', + 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f', + 'duration': 4576, + 'thumbnail': 're:http://.*\.jpg', + 'session_code': 'KOS002', + 'session_day': 'Day 1', + 'session_room': 'Arena 1A', + 'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', + 'Mads Kristensen'], }, - { - 'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing', - 'md5': 'b43ee4529d111bc37ba7ee4f34813e68', - 'info_dict': { - 'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing', - 'ext': 'mp4', - 'title': 'Self-service BI with Power BI - nuclear testing', - 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', - 'duration': 1540, - 'thumbnail': 're:http://.*\.jpg', - 'authors': ['Mike Wilmot'], - }, + }, { + 'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing', + 'md5': 'b43ee4529d111bc37ba7ee4f34813e68', + 'info_dict': { + 'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing', + 'ext': 'mp4', + 'title': 'Self-service BI with Power BI - nuclear testing', + 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', + 'duration': 1540, + 'thumbnail': 're:http://.*\.jpg', + 'authors': ['Mike Wilmot'], }, - { - # low quality mp4 is best - 'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', - 'info_dict': { - 'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', - 'ext': 'mp4', - 'title': 'Ranges for the Standard Library', - 'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d', - 'duration': 5646, - 'thumbnail': 're:http://.*\.jpg', - }, - 'params': { - 'skip_download': True, - }, - } - ] + }, { + # low quality mp4 is best + 'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', + 'info_dict': { + 'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library', + 'ext': 'mp4', + 'title': 'Ranges for the Standard Library', + 'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d', + 'duration': 5646, + 'thumbnail': 're:http://.*\.jpg', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'https://channel9.msdn.com/Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b/RSS', + 'info_dict': { + 'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b', + 'title': 'Channel 9', + }, + 'playlist_count': 2, + }, { + 'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS', + 'only_matching': True, + }, { + 'url': 'https://channel9.msdn.com/Events/Speakers/scott-hanselman/RSS?UrlSafeName=scott-hanselman', + 'only_matching': True, + }] _RSS_URL = 'http://channel9.msdn.com/%s/RSS' @@ -254,22 +264,30 @@ class Channel9IE(InfoExtractor): return self.playlist_result(contents) - def _extract_list(self, content_path): - rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS') + def _extract_list(self, video_id, rss_url=None): + if not rss_url: + rss_url = self._RSS_URL % video_id + rss = self._download_xml(rss_url, video_id, 'Downloading RSS') entries = [self.url_result(session_url.text, 'Channel9') for session_url in rss.findall('./channel/item/link')] title_text = rss.find('./channel/title').text - return self.playlist_result(entries, content_path, title_text) + return self.playlist_result(entries, video_id, title_text) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) content_path = mobj.group('contentpath') + rss = mobj.group('rss') + + if rss: + return self._extract_list(content_path, url) - webpage = self._download_webpage(url, content_path, 'Downloading web page') + webpage = self._download_webpage( + url, content_path, 'Downloading web page') - page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage) - if page_type_m is not None: - page_type = page_type_m.group('pagetype') + page_type = self._search_regex( + r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2', + webpage, 'page type', default=None, group='pagetype') + if page_type: if page_type == 'Entry': # Any 'item'-like page, may contain downloadable content return self._extract_entry_item(webpage, content_path) elif page_type == 'Session': # Event session page, may contain downloadable content @@ -278,6 +296,5 @@ class Channel9IE(InfoExtractor): return self._extract_list(content_path) else: raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True) - else: # Assuming list return self._extract_list(content_path) diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py deleted file mode 100644 index 042c4f2f1..000000000 --- a/youtube_dl/extractor/cinemassacre.py +++ /dev/null @@ -1,119 +0,0 @@ -# encoding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ExtractorError -from .screenwavemedia import ScreenwaveMediaIE - - -class CinemassacreIE(InfoExtractor): - _VALID_URL = 'https?://(?:www\.)?cinemassacre\.com/(?P<date_y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/(?P<display_id>[^?#/]+)' - _TESTS = [ - { - 'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/', - 'md5': 'fde81fbafaee331785f58cd6c0d46190', - 'info_dict': { - 'id': 'Cinemassacre-19911', - 'ext': 'mp4', - 'upload_date': '20121110', - 'title': '“Angry Video Game Nerd: The Movie” – Trailer', - 'description': 'md5:fb87405fcb42a331742a0dce2708560b', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }, - { - 'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940', - 'md5': 'd72f10cd39eac4215048f62ab477a511', - 'info_dict': { - 'id': 'Cinemassacre-521be8ef82b16', - 'ext': 'mp4', - 'upload_date': '20131002', - 'title': 'The Mummy’s Hand (1940)', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }, - { - # Youtube embedded video - 'url': 'http://cinemassacre.com/2006/12/07/chronologically-confused-about-bad-movie-and-video-game-sequel-titles/', - 'md5': 'ec9838a5520ef5409b3e4e42fcb0a3b9', - 'info_dict': { - 'id': 'OEVzPCY2T-g', - 'ext': 'webm', - 'title': 'AVGN: Chronologically Confused about Bad Movie and Video Game Sequel Titles', - 'upload_date': '20061207', - 'uploader': 'Cinemassacre', - 'uploader_id': 'JamesNintendoNerd', - 'description': 'md5:784734696c2b8b7f4b8625cc799e07f6', - } - }, - { - # Youtube embedded video - 'url': 'http://cinemassacre.com/2006/09/01/mckids/', - 'md5': '7393c4e0f54602ad110c793eb7a6513a', - 'info_dict': { - 'id': 'FnxsNhuikpo', - 'ext': 'webm', - 'upload_date': '20060901', - 'uploader': 'Cinemassacre Extra', - 'description': 'md5:de9b751efa9e45fbaafd9c8a1123ed53', - 'uploader_id': 'Cinemassacre', - 'title': 'AVGN: McKids', - } - }, - { - 'url': 'http://cinemassacre.com/2015/05/25/mario-kart-64-nintendo-64-james-mike-mondays/', - 'md5': '1376908e49572389e7b06251a53cdd08', - 'info_dict': { - 'id': 'Cinemassacre-555779690c440', - 'ext': 'mp4', - 'description': 'Let’s Play Mario Kart 64 !! Mario Kart 64 is a classic go-kart racing game released for the Nintendo 64 (N64). Today James & Mike do 4 player Battle Mode with Kyle and Bootsy!', - 'title': 'Mario Kart 64 (Nintendo 64) James & Mike Mondays', - 'upload_date': '20150525', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - } - ] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') - video_date = mobj.group('date_y') + mobj.group('date_m') + mobj.group('date_d') - - webpage = self._download_webpage(url, display_id) - - playerdata_url = self._search_regex( - [ - ScreenwaveMediaIE.EMBED_PATTERN, - r'<iframe[^>]+src="(?P<url>(?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"', - ], - webpage, 'player data URL', default=None, group='url') - if not playerdata_url: - raise ExtractorError('Unable to find player data') - - video_title = self._html_search_regex( - r'<title>(?P<title>.+?)\|', webpage, 'title') - video_description = self._html_search_regex( - r'<div class="entry-content">(?P<description>.+?)</div>', - webpage, 'description', flags=re.DOTALL, fatal=False) - video_thumbnail = self._og_search_thumbnail(webpage) - - return { - '_type': 'url_transparent', - 'display_id': display_id, - 'title': video_title, - 'description': video_description, - 'upload_date': video_date, - 'thumbnail': video_thumbnail, - 'url': playerdata_url, - } diff --git a/youtube_dl/extractor/cliprs.py b/youtube_dl/extractor/cliprs.py index 4f9320ea5..d55b26d59 100644 --- a/youtube_dl/extractor/cliprs.py +++ b/youtube_dl/extractor/cliprs.py @@ -1,16 +1,10 @@ # coding: utf-8 from __future__ import unicode_literals -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - float_or_none, - int_or_none, - parse_iso8601, -) +from .onet import OnetBaseIE -class ClipRsIE(InfoExtractor): +class ClipRsIE(OnetBaseIE): _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+' _TEST = { 'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732', @@ -27,64 +21,13 @@ class ClipRsIE(InfoExtractor): } def _real_extract(self, url): - video_id = self._match_id(url) + display_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, display_id) - video_id = self._search_regex( - r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id') + mvp_id = self._search_mvp_id(webpage) - response = self._download_json( - 'http://qi.ckm.onetapi.pl/', video_id, - query={ - 'body[id]': video_id, - 'body[jsonrpc]': '2.0', - 'body[method]': 'get_asset_detail', - 'body[params][ID_Publikacji]': video_id, - 'body[params][Service]': 'www.onet.pl', - 'content-type': 'application/jsonp', - 'x-onet-app': 'player.front.onetapi.pl', - }) + info_dict = self._extract_from_id(mvp_id, webpage) + info_dict['display_id'] = display_id - error = response.get('error') - if error: - raise ExtractorError( - '%s said: %s' % (self.IE_NAME, error['message']), expected=True) - - video = response['result'].get('0') - - formats = [] - for _, formats_dict in video['formats'].items(): - if not isinstance(formats_dict, dict): - continue - for format_id, format_list in formats_dict.items(): - if not isinstance(format_list, list): - continue - for f in format_list: - if not f.get('url'): - continue - formats.append({ - 'url': f['url'], - 'format_id': format_id, - 'height': int_or_none(f.get('vertical_resolution')), - 'width': int_or_none(f.get('horizontal_resolution')), - 'abr': float_or_none(f.get('audio_bitrate')), - 'vbr': float_or_none(f.get('video_bitrate')), - }) - self._sort_formats(formats) - - meta = video.get('meta', {}) - - title = self._og_search_title(webpage, default=None) or meta['title'] - description = self._og_search_description(webpage, default=None) or meta.get('description') - duration = meta.get('length') or meta.get('lenght') - timestamp = parse_iso8601(meta.get('addDate'), ' ') - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'timestamp': timestamp, - 'formats': formats, - } + return info_dict diff --git a/youtube_dl/extractor/closertotruth.py b/youtube_dl/extractor/closertotruth.py new file mode 100644 index 000000000..26243d52d --- /dev/null +++ b/youtube_dl/extractor/closertotruth.py @@ -0,0 +1,92 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class CloserToTruthIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?closertotruth\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)' + _TESTS = [{ + 'url': 'http://closertotruth.com/series/solutions-the-mind-body-problem#video-3688', + 'info_dict': { + 'id': '0_zof1ktre', + 'display_id': 'solutions-the-mind-body-problem', + 'ext': 'mov', + 'title': 'Solutions to the Mind-Body Problem?', + 'upload_date': '20140221', + 'timestamp': 1392956007, + 'uploader_id': 'CTTXML' + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://closertotruth.com/episodes/how-do-brains-work', + 'info_dict': { + 'id': '0_iuxai6g6', + 'display_id': 'how-do-brains-work', + 'ext': 'mov', + 'title': 'How do Brains Work?', + 'upload_date': '20140221', + 'timestamp': 1392956024, + 'uploader_id': 'CTTXML' + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://closertotruth.com/interviews/1725', + 'info_dict': { + 'id': '1725', + 'title': 'AyaFr-002', + }, + 'playlist_mincount': 2, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + partner_id = self._search_regex( + r'<script[^>]+src=["\'].*?\b(?:partner_id|p)/(\d+)', + webpage, 'kaltura partner_id') + + title = self._search_regex( + r'<title>(.+?)\s*\|\s*.+?</title>', webpage, 'video title') + + select = self._search_regex( + r'(?s)<select[^>]+id="select-version"[^>]*>(.+?)</select>', + webpage, 'select version', default=None) + if select: + entry_ids = set() + entries = [] + for mobj in re.finditer( + r'<option[^>]+value=(["\'])(?P<id>[0-9a-z_]+)(?:#.+?)?\1[^>]*>(?P<title>[^<]+)', + webpage): + entry_id = mobj.group('id') + if entry_id in entry_ids: + continue + entry_ids.add(entry_id) + entries.append({ + '_type': 'url_transparent', + 'url': 'kaltura:%s:%s' % (partner_id, entry_id), + 'ie_key': 'Kaltura', + 'title': mobj.group('title'), + }) + if entries: + return self.playlist_result(entries, display_id, title) + + entry_id = self._search_regex( + r'<a[^>]+id=(["\'])embed-kaltura\1[^>]+data-kaltura=(["\'])(?P<id>[0-9a-z_]+)\2', + webpage, 'kaltura entry_id', group='id') + + return { + '_type': 'url_transparent', + 'display_id': display_id, + 'url': 'kaltura:%s:%s' % (partner_id, entry_id), + 'ie_key': 'Kaltura', + 'title': title + } diff --git a/youtube_dl/extractor/cloudy.py b/youtube_dl/extractor/cloudy.py index 9e267e6c0..9a28ef354 100644 --- a/youtube_dl/extractor/cloudy.py +++ b/youtube_dl/extractor/cloudy.py @@ -19,7 +19,7 @@ from ..utils import ( class CloudyIE(InfoExtractor): _IE_DESC = 'cloudy.ec and videoraj.ch' _VALID_URL = r'''(?x) - https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.ch)/ + https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.(?:ch|to))/ (?:v/|embed\.php\?id=) (?P<id>[A-Za-z0-9]+) ''' @@ -37,7 +37,7 @@ class CloudyIE(InfoExtractor): } }, { - 'url': 'http://www.videoraj.ch/v/47f399fd8bb60', + 'url': 'http://www.videoraj.to/v/47f399fd8bb60', 'md5': '7d0f8799d91efd4eda26587421c3c3b0', 'info_dict': { 'id': '47f399fd8bb60', diff --git a/youtube_dl/extractor/collegehumor.py b/youtube_dl/extractor/collegehumor.py deleted file mode 100644 index 002b24037..000000000 --- a/youtube_dl/extractor/collegehumor.py +++ /dev/null @@ -1,101 +0,0 @@ -from __future__ import unicode_literals - -import json -import re - -from .common import InfoExtractor -from ..utils import int_or_none - - -class CollegeHumorIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$' - - _TESTS = [ - { - 'url': 'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe', - 'md5': 'dcc0f5c1c8be98dc33889a191f4c26bd', - 'info_dict': { - 'id': '6902724', - 'ext': 'mp4', - 'title': 'Comic-Con Cosplay Catastrophe', - 'description': "Fans get creative this year at San Diego. Too creative. And yes, that's really Joss Whedon.", - 'age_limit': 13, - 'duration': 187, - }, - }, { - 'url': 'http://www.collegehumor.com/video/3505939/font-conference', - 'md5': '72fa701d8ef38664a4dbb9e2ab721816', - 'info_dict': { - 'id': '3505939', - 'ext': 'mp4', - 'title': 'Font Conference', - 'description': "This video wasn't long enough, so we made it double-spaced.", - 'age_limit': 10, - 'duration': 179, - }, - }, { - # embedded youtube video - 'url': 'http://www.collegehumor.com/embed/6950306', - 'info_dict': { - 'id': 'Z-bao9fg6Yc', - 'ext': 'mp4', - 'title': 'Young Americans Think President John F. Kennedy Died THIS MORNING IN A CAR ACCIDENT!!!', - 'uploader': 'Mark Dice', - 'uploader_id': 'MarkDice', - 'description': 'md5:62c3dab9351fac7bb44b53b69511d87f', - 'upload_date': '20140127', - }, - 'params': { - 'skip_download': True, - }, - 'add_ie': ['Youtube'], - }, - ] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - - jsonUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id + '.json' - data = json.loads(self._download_webpage( - jsonUrl, video_id, 'Downloading info JSON')) - vdata = data['video'] - if vdata.get('youtubeId') is not None: - return { - '_type': 'url', - 'url': vdata['youtubeId'], - 'ie_key': 'Youtube', - } - - AGE_LIMITS = {'nc17': 18, 'r': 18, 'pg13': 13, 'pg': 10, 'g': 0} - rating = vdata.get('rating') - if rating: - age_limit = AGE_LIMITS.get(rating.lower()) - else: - age_limit = None # None = No idea - - PREFS = {'high_quality': 2, 'low_quality': 0} - formats = [] - for format_key in ('mp4', 'webm'): - for qname, qurl in vdata.get(format_key, {}).items(): - formats.append({ - 'format_id': format_key + '_' + qname, - 'url': qurl, - 'format': format_key, - 'preference': PREFS.get(qname), - }) - self._sort_formats(formats) - - duration = int_or_none(vdata.get('duration'), 1000) - like_count = int_or_none(vdata.get('likes')) - - return { - 'id': video_id, - 'title': vdata['title'], - 'description': vdata.get('description'), - 'thumbnail': vdata.get('thumbnail'), - 'formats': formats, - 'age_limit': age_limit, - 'duration': duration, - 'like_count': like_count, - } diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 0c59102e0..2b6aaa3aa 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -44,10 +44,10 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor): # or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524 _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow) |https?://(:www\.)? - (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/ + (?P<showname>thedailyshow|thecolbertreport|tosh)\.(?:cc\.)?com/ ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)| (?P<clip> - (?:(?:guests/[^/]+|videos|video-playlists|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+)) + (?:(?:guests/[^/]+|videos|video-(?:clips|playlists)|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+)) |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?)) |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)) )| @@ -129,6 +129,9 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor): }, { 'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel', 'only_matching': True, + }, { + 'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans', + 'only_matching': True, }] _available_formats = ['3500', '2200', '1700', '1200', '750', '400'] diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 02cd2c003..816baa424 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -44,7 +44,9 @@ from ..utils import ( sanitized_Request, unescapeHTML, unified_strdate, + unified_timestamp, url_basename, + xpath_element, xpath_text, xpath_with_ns, determine_protocol, @@ -52,6 +54,7 @@ from ..utils import ( mimetype2ext, update_Request, update_url_query, + parse_m3u8_attributes, ) @@ -159,11 +162,12 @@ class InfoExtractor(object): * "height" (optional, int) * "resolution" (optional, string "{width}x{height"}, deprecated) + * "filesize" (optional, int) thumbnail: Full URL to a video thumbnail image. description: Full video description. uploader: Full name of the video uploader. license: License name the video is licensed under. - creator: The main artist who created the video. + creator: The creator of the video. release_date: The date (YYYYMMDD) when the video was released. timestamp: UNIX timestamp of the moment the video became available. upload_date: Video upload date (YYYYMMDD). @@ -747,10 +751,12 @@ class InfoExtractor(object): return self._og_search_property('url', html, **kargs) def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): + if not isinstance(name, (list, tuple)): + name = [name] if display_name is None: - display_name = name + display_name = name[0] return self._html_search_regex( - self._meta_regex(name), + [self._meta_regex(n) for n in name], html, display_name, fatal=fatal, group='content', **kwargs) def _dc_search_uploader(self, html): @@ -799,15 +805,17 @@ class InfoExtractor(object): return self._html_search_meta('twitter:player', html, 'twitter card player') - def _search_json_ld(self, html, video_id, **kwargs): + def _search_json_ld(self, html, video_id, expected_type=None, **kwargs): json_ld = self._search_regex( r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>', html, 'JSON-LD', group='json_ld', **kwargs) if not json_ld: return {} - return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True)) + return self._json_ld( + json_ld, video_id, fatal=kwargs.get('fatal', True), + expected_type=expected_type) - def _json_ld(self, json_ld, video_id, fatal=True): + def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None): if isinstance(json_ld, compat_str): json_ld = self._parse_json(json_ld, video_id, fatal=fatal) if not json_ld: @@ -815,6 +823,8 @@ class InfoExtractor(object): info = {} if json_ld.get('@context') == 'http://schema.org': item_type = json_ld.get('@type') + if expected_type is not None and expected_type != item_type: + return info if item_type == 'TVEpisode': info.update({ 'episode': unescapeHTML(json_ld.get('name')), @@ -833,6 +843,19 @@ class InfoExtractor(object): 'title': unescapeHTML(json_ld.get('headline')), 'description': unescapeHTML(json_ld.get('articleBody')), }) + elif item_type == 'VideoObject': + info.update({ + 'url': json_ld.get('contentUrl'), + 'title': unescapeHTML(json_ld.get('name')), + 'description': unescapeHTML(json_ld.get('description')), + 'thumbnail': json_ld.get('thumbnailUrl'), + 'duration': parse_duration(json_ld.get('duration')), + 'timestamp': unified_timestamp(json_ld.get('uploadDate')), + 'filesize': float_or_none(json_ld.get('contentSize')), + 'tbr': int_or_none(json_ld.get('bitrate')), + 'width': int_or_none(json_ld.get('width')), + 'height': int_or_none(json_ld.get('height')), + }) return dict((k, v) for k, v in info.items() if v is not None) @staticmethod @@ -874,7 +897,11 @@ class InfoExtractor(object): f['ext'] = determine_ext(f['url']) if isinstance(field_preference, (list, tuple)): - return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference) + return tuple( + f.get(field) + if f.get(field) is not None + else ('' if field == 'format_id' else -1) + for field in field_preference) preference = f.get('preference') if preference is None: @@ -987,7 +1014,7 @@ class InfoExtractor(object): def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, transform_source=lambda s: fix_xml_ampersands(s).strip(), - fatal=True): + fatal=True, m3u8_id=None): manifest = self._download_xml( manifest_url, video_id, 'Downloading f4m manifest', 'Unable to download f4m manifest', @@ -1001,11 +1028,18 @@ class InfoExtractor(object): return self._parse_f4m_formats( manifest, manifest_url, video_id, preference=preference, f4m_id=f4m_id, - transform_source=transform_source, fatal=fatal) + transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id) def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None, transform_source=lambda s: fix_xml_ampersands(s).strip(), - fatal=True): + fatal=True, m3u8_id=None): + # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy + akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0') + if akamai_pv is not None and ';' in akamai_pv.text: + playerVerificationChallenge = akamai_pv.text.split(';')[0] + if playerVerificationChallenge.strip() != '': + return [] + formats = [] manifest_version = '1.0' media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media') @@ -1022,9 +1056,26 @@ class InfoExtractor(object): 'base URL', default=None) if base_url: base_url = base_url.strip() + + bootstrap_info = xpath_element( + manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'], + 'bootstrap info', default=None) + for i, media_el in enumerate(media_nodes): - if manifest_version == '2.0': - media_url = media_el.attrib.get('href') or media_el.attrib.get('url') + tbr = int_or_none(media_el.attrib.get('bitrate')) + width = int_or_none(media_el.attrib.get('width')) + height = int_or_none(media_el.attrib.get('height')) + format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])) + # If <bootstrapInfo> is present, the specified f4m is a + # stream-level manifest, and only set-level manifests may refer to + # external resources. See section 11.4 and section 4 of F4M spec + if bootstrap_info is None: + media_url = None + # @href is introduced in 2.0, see section 11.6 of F4M spec + if manifest_version == '2.0': + media_url = media_el.attrib.get('href') + if media_url is None: + media_url = media_el.attrib.get('url') if not media_url: continue manifest_url = ( @@ -1034,29 +1085,43 @@ class InfoExtractor(object): # since bitrates in parent manifest (this one) and media_url manifest # may differ leading to inability to resolve the format by requested # bitrate in f4m downloader - if determine_ext(manifest_url) == 'f4m': - formats.extend(self._extract_f4m_formats( + ext = determine_ext(manifest_url) + if ext == 'f4m': + f4m_formats = self._extract_f4m_formats( manifest_url, video_id, preference=preference, f4m_id=f4m_id, - transform_source=transform_source, fatal=fatal)) + transform_source=transform_source, fatal=fatal) + # Sometimes stream-level manifest contains single media entry that + # does not contain any quality metadata (e.g. http://matchtv.ru/#live-player). + # At the same time parent's media entry in set-level manifest may + # contain it. We will copy it from parent in such cases. + if len(f4m_formats) == 1: + f = f4m_formats[0] + f.update({ + 'tbr': f.get('tbr') or tbr, + 'width': f.get('width') or width, + 'height': f.get('height') or height, + 'format_id': f.get('format_id') if not tbr else format_id, + }) + formats.extend(f4m_formats) + continue + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + manifest_url, video_id, 'mp4', preference=preference, + m3u8_id=m3u8_id, fatal=fatal)) continue - tbr = int_or_none(media_el.attrib.get('bitrate')) formats.append({ - 'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])), + 'format_id': format_id, 'url': manifest_url, - 'ext': 'flv', + 'ext': 'flv' if bootstrap_info is not None else None, 'tbr': tbr, - 'width': int_or_none(media_el.attrib.get('width')), - 'height': int_or_none(media_el.attrib.get('height')), + 'width': width, + 'height': height, 'preference': preference, }) return formats - def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, - entry_protocol='m3u8', preference=None, - m3u8_id=None, note=None, errnote=None, - fatal=True): - - formats = [{ + def _m3u8_meta_format(self, m3u8_url, ext=None, preference=None, m3u8_id=None): + return { 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])), 'url': m3u8_url, 'ext': ext, @@ -1064,7 +1129,14 @@ class InfoExtractor(object): 'preference': preference - 1 if preference else -1, 'resolution': 'multiple', 'format_note': 'Quality selection URL', - }] + } + + def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None, + entry_protocol='m3u8', preference=None, + m3u8_id=None, note=None, errnote=None, + fatal=True, live=False): + + formats = [self._m3u8_meta_format(m3u8_url, ext, preference, m3u8_id)] format_url = lambda u: ( u @@ -1104,23 +1176,11 @@ class InfoExtractor(object): }] last_info = None last_media = None - kv_rex = re.compile( - r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)') for line in m3u8_doc.splitlines(): if line.startswith('#EXT-X-STREAM-INF:'): - last_info = {} - for m in kv_rex.finditer(line): - v = m.group('val') - if v.startswith('"'): - v = v[1:-1] - last_info[m.group('key')] = v + last_info = parse_m3u8_attributes(line) elif line.startswith('#EXT-X-MEDIA:'): - last_media = {} - for m in kv_rex.finditer(line): - v = m.group('val') - if v.startswith('"'): - v = v[1:-1] - last_media[m.group('key')] = v + last_media = parse_m3u8_attributes(line) elif line.startswith('#') or not line.strip(): continue else: @@ -1131,8 +1191,15 @@ class InfoExtractor(object): format_id = [] if m3u8_id: format_id.append(m3u8_id) - last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None - format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats))) + last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') not in ('SUBTITLES', 'CLOSED-CAPTIONS') else None + # Despite specification does not mention NAME attribute for + # EXT-X-STREAM-INF it still sometimes may be present + stream_name = last_info.get('NAME') or last_media_name + # Bandwidth of live streams may differ over time thus making + # format_id unpredictable. So it's better to keep provided + # format_id intact. + if not live: + format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats))) f = { 'format_id': '-'.join(format_id), 'url': format_url(line.strip()), @@ -1264,21 +1331,21 @@ class InfoExtractor(object): m3u8_count = 0 srcs = [] - videos = smil.findall(self._xpath_ns('.//video', namespace)) - for video in videos: - src = video.get('src') + media = smil.findall(self._xpath_ns('.//video', namespace)) + smil.findall(self._xpath_ns('.//audio', namespace)) + for medium in media: + src = medium.get('src') if not src or src in srcs: continue srcs.append(src) - bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) - filesize = int_or_none(video.get('size') or video.get('fileSize')) - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - proto = video.get('proto') - ext = video.get('ext') + bitrate = float_or_none(medium.get('system-bitrate') or medium.get('systemBitrate'), 1000) + filesize = int_or_none(medium.get('size') or medium.get('fileSize')) + width = int_or_none(medium.get('width')) + height = int_or_none(medium.get('height')) + proto = medium.get('proto') + ext = medium.get('ext') src_ext = determine_ext(src) - streamer = video.get('streamer') or base + streamer = medium.get('streamer') or base if proto == 'rtmp' or streamer.startswith('rtmp'): rtmp_count += 1 @@ -1681,6 +1748,13 @@ class InfoExtractor(object): def _mark_watched(self, *args, **kwargs): raise NotImplementedError('This method must be implemented by subclasses') + def geo_verification_headers(self): + headers = {} + geo_verification_proxy = self._downloader.params.get('geo_verification_proxy') + if geo_verification_proxy: + headers['Ytdl-request-proxy'] = geo_verification_proxy + return headers + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/coub.py b/youtube_dl/extractor/coub.py new file mode 100644 index 000000000..a901b8d22 --- /dev/null +++ b/youtube_dl/extractor/coub.py @@ -0,0 +1,143 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + float_or_none, + int_or_none, + parse_iso8601, + qualities, +) + + +class CoubIE(InfoExtractor): + _VALID_URL = r'(?:coub:|https?://(?:coub\.com/(?:view|embed|coubs)/|c-cdn\.coub\.com/fb-player\.swf\?.*\bcoub(?:ID|id)=))(?P<id>[\da-z]+)' + + _TESTS = [{ + 'url': 'http://coub.com/view/5u5n1', + 'info_dict': { + 'id': '5u5n1', + 'ext': 'mp4', + 'title': 'The Matrix Moonwalk', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 4.6, + 'timestamp': 1428527772, + 'upload_date': '20150408', + 'uploader': 'Артём Лоскутников', + 'uploader_id': 'artyom.loskutnikov', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + 'age_limit': 0, + }, + }, { + 'url': 'http://c-cdn.coub.com/fb-player.swf?bot_type=vk&coubID=7w5a4', + 'only_matching': True, + }, { + 'url': 'coub:5u5n1', + 'only_matching': True, + }, { + # longer video id + 'url': 'http://coub.com/view/237d5l5h', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + coub = self._download_json( + 'http://coub.com/api/v2/coubs/%s.json' % video_id, video_id) + + if coub.get('error'): + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, coub['error']), expected=True) + + title = coub['title'] + + file_versions = coub['file_versions'] + + QUALITIES = ('low', 'med', 'high') + + MOBILE = 'mobile' + IPHONE = 'iphone' + HTML5 = 'html5' + + SOURCE_PREFERENCE = (MOBILE, IPHONE, HTML5) + + quality_key = qualities(QUALITIES) + preference_key = qualities(SOURCE_PREFERENCE) + + formats = [] + + for kind, items in file_versions.get(HTML5, {}).items(): + if kind not in ('video', 'audio'): + continue + if not isinstance(items, dict): + continue + for quality, item in items.items(): + if not isinstance(item, dict): + continue + item_url = item.get('url') + if not item_url: + continue + formats.append({ + 'url': item_url, + 'format_id': '%s-%s-%s' % (HTML5, kind, quality), + 'filesize': int_or_none(item.get('size')), + 'vcodec': 'none' if kind == 'audio' else None, + 'quality': quality_key(quality), + 'preference': preference_key(HTML5), + }) + + iphone_url = file_versions.get(IPHONE, {}).get('url') + if iphone_url: + formats.append({ + 'url': iphone_url, + 'format_id': IPHONE, + 'preference': preference_key(IPHONE), + }) + + mobile_url = file_versions.get(MOBILE, {}).get('audio_url') + if mobile_url: + formats.append({ + 'url': mobile_url, + 'format_id': '%s-audio' % MOBILE, + 'preference': preference_key(MOBILE), + }) + + self._sort_formats(formats) + + thumbnail = coub.get('picture') + duration = float_or_none(coub.get('duration')) + timestamp = parse_iso8601(coub.get('published_at') or coub.get('created_at')) + uploader = coub.get('channel', {}).get('title') + uploader_id = coub.get('channel', {}).get('permalink') + + view_count = int_or_none(coub.get('views_count') or coub.get('views_increase_count')) + like_count = int_or_none(coub.get('likes_count')) + repost_count = int_or_none(coub.get('recoubs_count')) + comment_count = int_or_none(coub.get('comments_count')) + + age_restricted = coub.get('age_restricted', coub.get('age_restricted_by_admin')) + if age_restricted is not None: + age_limit = 18 if age_restricted is True else 0 + else: + age_limit = None + + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'view_count': view_count, + 'like_count': like_count, + 'repost_count': repost_count, + 'comment_count': comment_count, + 'age_limit': age_limit, + 'formats': formats, + } diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index 8ae3f2890..90a64303d 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -11,7 +11,6 @@ from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( compat_etree_fromstring, - compat_urllib_parse_unquote, compat_urllib_parse_urlencode, compat_urllib_request, compat_urlparse, @@ -27,6 +26,7 @@ from ..utils import ( unified_strdate, urlencode_postdata, xpath_text, + extract_attributes, ) from ..aes import ( aes_cbc_decrypt, @@ -306,28 +306,36 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage, 'video_uploader', fatal=False) - playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url')) - playerdata_req = sanitized_Request(playerdata_url) - playerdata_req.data = urlencode_postdata({'current_page': webpage_url}) - playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') - playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info') - - stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id') - video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False) - + available_fmts = [] + for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage): + attrs = extract_attributes(a) + href = attrs.get('href') + if href and '/freetrial' in href: + continue + available_fmts.append(fmt) + if not available_fmts: + for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'): + available_fmts = re.findall(p, webpage) + if available_fmts: + break + video_encode_ids = [] formats = [] - for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage): + for fmt in available_fmts: stream_quality, stream_format = self._FORMAT_IDS[fmt] video_format = fmt + 'p' streamdata_req = sanitized_Request( 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s' - % (stream_id, stream_format, stream_quality), + % (video_id, stream_format, stream_quality), compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8')) streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') streamdata = self._download_xml( streamdata_req, video_id, note='Downloading media info for %s' % video_format) stream_info = streamdata.find('./{default}preload/stream_info') + video_encode_id = xpath_text(stream_info, './video_encode_id') + if video_encode_id in video_encode_ids: + continue + video_encode_ids.append(video_encode_id) video_url = xpath_text(stream_info, './host') video_play_path = xpath_text(stream_info, './file') if not video_url or not video_play_path: @@ -359,6 +367,14 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'ext': 'flv', }) formats.append(format_info) + self._sort_formats(formats) + + metadata = self._download_xml( + 'http://www.crunchyroll.com/xml', video_id, + note='Downloading media info', query={ + 'req': 'RpcApiVideoPlayer_GetMediaMetadata', + 'media_id': video_id, + }) subtitles = self.extract_subtitles(video_id, webpage) @@ -366,9 +382,12 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'id': video_id, 'title': video_title, 'description': video_description, - 'thumbnail': video_thumbnail, + 'thumbnail': xpath_text(metadata, 'episode_image_url'), 'uploader': video_uploader, 'upload_date': video_upload_date, + 'series': xpath_text(metadata, 'series_title'), + 'episode': xpath_text(metadata, 'episode_title'), + 'episode_number': int_or_none(xpath_text(metadata, 'episode_number')), 'subtitles': subtitles, 'formats': formats, } diff --git a/youtube_dl/extractor/ctv.py b/youtube_dl/extractor/ctv.py new file mode 100644 index 000000000..5807fbac9 --- /dev/null +++ b/youtube_dl/extractor/ctv.py @@ -0,0 +1,30 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class CTVIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?ctv\.ca/video/player\?vid=(?P<id>[0-9.]+)' + _TESTS = [{ + 'url': 'http://www.ctv.ca/video/player?vid=706966', + 'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0', + 'info_dict': { + 'id': '706966', + 'ext': 'mp4', + 'title': 'Larry Day and Richard Jutras on the TIFF red carpet of \'Stonewall\'', + 'description': 'etalk catches up with Larry Day and Richard Jutras on the TIFF red carpet of "Stonewall”.', + 'upload_date': '20150919', + 'timestamp': 1442624700, + }, + 'expected_warnings': ['HTTP Error 404'], + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + return { + '_type': 'url_transparent', + 'id': video_id, + 'url': '9c9media:ctv_web:%s' % video_id, + 'ie_key': 'NineCNineMedia', + } diff --git a/youtube_dl/extractor/ctvnews.py b/youtube_dl/extractor/ctvnews.py new file mode 100644 index 000000000..1023b6130 --- /dev/null +++ b/youtube_dl/extractor/ctvnews.py @@ -0,0 +1,65 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import orderedSet + + +class CTVNewsIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)' + _TESTS = [{ + 'url': 'http://www.ctvnews.ca/video?clipId=901995', + 'md5': '10deb320dc0ccb8d01d34d12fc2ea672', + 'info_dict': { + 'id': '901995', + 'ext': 'mp4', + 'title': 'Extended: \'That person cannot be me\' Johnson says', + 'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285', + 'timestamp': 1467286284, + 'upload_date': '20160630', + } + }, { + 'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224', + 'info_dict': + { + 'id': '1.2966224', + }, + 'playlist_mincount': 19, + }, { + 'url': 'http://www.ctvnews.ca/video?binId=1.2876780', + 'info_dict': + { + 'id': '1.2876780', + }, + 'playlist_mincount': 100, + }, { + 'url': 'http://www.ctvnews.ca/1.810401', + 'only_matching': True, + }, { + 'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231', + 'only_matching': True, + }] + + def _real_extract(self, url): + page_id = self._match_id(url) + + def ninecninemedia_url_result(clip_id): + return { + '_type': 'url_transparent', + 'id': clip_id, + 'url': '9c9media:ctvnews_web:%s' % clip_id, + 'ie_key': 'NineCNineMedia', + } + + if page_id.isdigit(): + return ninecninemedia_url_result(page_id) + else: + webpage = self._download_webpage('http://www.ctvnews.ca/%s' % page_id, page_id, query={ + 'ot': 'example.AjaxPageLayout.ot', + 'maxItemsPerPage': 1000000, + }) + entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet( + re.findall(r'clip\.id\s*=\s*(\d+);', webpage))] + return self.playlist_result(entries, page_id) diff --git a/youtube_dl/extractor/cwtv.py b/youtube_dl/extractor/cwtv.py index f5cefd966..ebd14cb16 100644 --- a/youtube_dl/extractor/cwtv.py +++ b/youtube_dl/extractor/cwtv.py @@ -9,7 +9,7 @@ from ..utils import ( class CWTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/shows/(?:[^/]+/){2}\?play=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})' + _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/(?:shows/)?(?:[^/]+/){2}\?.*\bplay=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})' _TESTS = [{ 'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?play=6b15e985-9345-4f60-baf8-56e96be57c63', 'info_dict': { @@ -48,6 +48,9 @@ class CWTVIE(InfoExtractor): # m3u8 download 'skip_download': True, } + }, { + 'url': 'http://cwtv.com/thecw/chroniclesofcisco/?play=8adebe35-f447-465f-ab52-e863506ff6d6', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/dailymail.py b/youtube_dl/extractor/dailymail.py new file mode 100644 index 000000000..b60a1d813 --- /dev/null +++ b/youtube_dl/extractor/dailymail.py @@ -0,0 +1,61 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + determine_protocol, +) + + +class DailyMailIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.dailymail.co.uk/video/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html', + 'md5': '2f639d446394f53f3a33658b518b6615', + 'info_dict': { + 'id': '1288527', + 'ext': 'mp4', + 'title': 'Turn any video into an impressionist masterpiece', + 'description': 'md5:88ddbcb504367987b2708bb38677c9d2', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + video_data = self._parse_json(self._search_regex( + r"data-opts='({.+?})'", webpage, 'video data'), video_id) + title = video_data['title'] + video_sources = self._download_json(video_data.get( + 'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id) + + formats = [] + for rendition in video_sources['renditions']: + rendition_url = rendition.get('url') + if not rendition_url: + continue + tbr = int_or_none(rendition.get('encodingRate'), 1000) + container = rendition.get('videoContainer') + is_hls = container == 'M2TS' + protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url}) + formats.append({ + 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''), + 'url': rendition_url, + 'width': int_or_none(rendition.get('frameWidth')), + 'height': int_or_none(rendition.get('frameHeight')), + 'tbr': tbr, + 'vcodec': rendition.get('videoCodec'), + 'container': container, + 'protocol': protocol, + 'ext': 'mp4' if is_hls else None, + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': video_data.get('descr'), + 'thumbnail': video_data.get('poster') or video_data.get('thumbnail'), + 'formats': formats, + } diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 2e6226ea0..1f92823b7 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -16,6 +16,7 @@ from ..utils import ( sanitized_Request, str_to_int, unescapeHTML, + mimetype2ext, ) @@ -111,6 +112,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor): } ] + @staticmethod + def _extract_urls(webpage): + # Look for embedded Dailymotion player + matches = re.findall( + r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage) + return list(map(lambda m: unescapeHTML(m[1]), matches)) + def _real_extract(self, url): video_id = self._match_id(url) @@ -153,18 +161,19 @@ class DailymotionIE(DailymotionBaseInfoExtractor): type_ = media.get('type') if type_ == 'application/vnd.lumberjack.manifest': continue - ext = determine_ext(media_url) - if type_ == 'application/x-mpegURL' or ext == 'm3u8': + ext = mimetype2ext(type_) or determine_ext(media_url) + if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( media_url, video_id, 'mp4', preference=-1, m3u8_id='hls', fatal=False)) - elif type_ == 'application/f4m' or ext == 'f4m': + elif ext == 'f4m': formats.extend(self._extract_f4m_formats( media_url, video_id, preference=-1, f4m_id='hds', fatal=False)) else: f = { 'url': media_url, 'format_id': 'http-%s' % quality, + 'ext': ext, } m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url) if m: diff --git a/youtube_dl/extractor/daum.py b/youtube_dl/extractor/daum.py index 86024a745..b5c310ccb 100644 --- a/youtube_dl/extractor/daum.py +++ b/youtube_dl/extractor/daum.py @@ -66,22 +66,32 @@ class DaumIE(InfoExtractor): 'view_count': int, 'comment_count': int, }, + }, { + # Requires dte_type=WEB (#9972) + 'url': 'http://tvpot.daum.net/v/s3794Uf1NZeZ1qMpGpeqeRU', + 'md5': 'a8917742069a4dd442516b86e7d66529', + 'info_dict': { + 'id': 's3794Uf1NZeZ1qMpGpeqeRU', + 'ext': 'mp4', + 'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny) [쇼! 음악중심] 508회 20160611', + 'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\n\n[쇼! 음악중심] 20160611, 507회', + 'upload_date': '20160611', + }, }] def _real_extract(self, url): video_id = compat_urllib_parse_unquote(self._match_id(url)) - query = compat_urllib_parse_urlencode({'vid': video_id}) movie_data = self._download_json( - 'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query, - video_id, 'Downloading video formats info') + 'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json', + video_id, 'Downloading video formats info', query={'vid': video_id, 'dte_type': 'WEB'}) # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id): return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id) info = self._download_xml( - 'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id, - 'Downloading video info') + 'http://tvpot.daum.net/clip/ClipInfoXml.do', video_id, + 'Downloading video info', query={'vid': video_id}) formats = [] for format_el in movie_data['output_list']['output_list']: diff --git a/youtube_dl/extractor/dcn.py b/youtube_dl/extractor/dcn.py index 5deff5f30..efb8585e8 100644 --- a/youtube_dl/extractor/dcn.py +++ b/youtube_dl/extractor/dcn.py @@ -20,7 +20,7 @@ from ..utils import ( class DCNIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?' + _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?' def _real_extract(self, url): show_id, video_id, season_id = re.match(self._VALID_URL, url).groups() @@ -55,30 +55,32 @@ class DCNBaseIE(InfoExtractor): 'is_live': is_live, } - def _extract_video_formats(self, webpage, video_id, entry_protocol): + def _extract_video_formats(self, webpage, video_id, m3u8_entry_protocol): formats = [] - m3u8_url = self._html_search_regex( - r'file\s*:\s*"([^"]+)', webpage, 'm3u8 url', fatal=False) - if m3u8_url: - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=None)) - - rtsp_url = self._search_regex( - r'<a[^>]+href="(rtsp://[^"]+)"', webpage, 'rtsp url', fatal=False) - if rtsp_url: - formats.append({ - 'url': rtsp_url, - 'format_id': 'rtsp', - }) - + format_url_base = 'http' + self._html_search_regex( + [ + r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8', + r'<a[^>]+href="rtsp(://[^"]+)"' + ], webpage, 'format url') + # TODO: Current DASH formats are broken - $Time$ pattern in + # <SegmentTemplate> not implemented yet + # formats.extend(self._extract_mpd_formats( + # format_url_base + '/manifest.mpd', + # video_id, mpd_id='dash', fatal=False)) + formats.extend(self._extract_m3u8_formats( + format_url_base + '/playlist.m3u8', video_id, 'mp4', + m3u8_entry_protocol, m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + format_url_base + '/manifest.f4m', + video_id, f4m_id='hds', fatal=False)) self._sort_formats(formats) return formats class DCNVideoIE(DCNBaseIE): IE_NAME = 'dcn:video' - _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/[^/]+|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?(?:video(?:/[^/]+)?|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)' + _TESTS = [{ 'url': 'http://www.dcndigital.ae/#/video/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375', 'info_dict': { @@ -94,7 +96,10 @@ class DCNVideoIE(DCNBaseIE): # m3u8 download 'skip_download': True, }, - } + }, { + 'url': 'http://awaan.ae/video/26723981/%D8%AF%D8%A7%D8%B1-%D8%A7%D9%84%D8%B3%D9%84%D8%A7%D9%85:-%D8%AE%D9%8A%D8%B1-%D8%AF%D9%88%D8%B1-%D8%A7%D9%84%D8%A3%D9%86%D8%B5%D8%A7%D8%B1', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -120,7 +125,7 @@ class DCNVideoIE(DCNBaseIE): class DCNLiveIE(DCNBaseIE): IE_NAME = 'dcn:live' - _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?live/(?P<id>\d+)' + _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?live/(?P<id>\d+)' def _real_extract(self, url): channel_id = self._match_id(url) @@ -147,7 +152,7 @@ class DCNLiveIE(DCNBaseIE): class DCNSeasonIE(InfoExtractor): IE_NAME = 'dcn:season' - _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))' + _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))' _TEST = { 'url': 'http://dcndigital.ae/#/program/205024/%D9%85%D8%AD%D8%A7%D8%B6%D8%B1%D8%A7%D8%AA-%D8%A7%D9%84%D8%B4%D9%8A%D8%AE-%D8%A7%D9%84%D8%B4%D8%B9%D8%B1%D8%A7%D9%88%D9%8A', 'info_dict': diff --git a/youtube_dl/extractor/dfb.py b/youtube_dl/extractor/dfb.py index cdfeccacb..a4d0448c2 100644 --- a/youtube_dl/extractor/dfb.py +++ b/youtube_dl/extractor/dfb.py @@ -12,39 +12,46 @@ class DFBIE(InfoExtractor): _TEST = { 'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/', - # The md5 is different each time + 'md5': 'ac0f98a52a330f700b4b3034ad240649', 'info_dict': { 'id': '11633', 'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'U 19-EM: Stimmen zum Spiel gegen Russland', 'upload_date': '20150714', }, } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') + display_id, video_id = re.match(self._VALID_URL, url).groups() - webpage = self._download_webpage(url, display_id) player_info = self._download_xml( 'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id, display_id) video_info = player_info.find('video') - - f4m_info = self._download_xml( - self._proto_relative_url(video_info.find('url').text.strip()), display_id) - token_el = f4m_info.find('token') - manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0' - formats = self._extract_f4m_formats(manifest_url, display_id) + stream_access_url = self._proto_relative_url(video_info.find('url').text.strip()) + + formats = [] + # see http://tv.dfb.de/player/js/ajax.js for the method to extract m3u8 formats + for sa_url in (stream_access_url, stream_access_url + '&area=&format=iphone'): + stream_access_info = self._download_xml(sa_url, display_id) + token_el = stream_access_info.find('token') + manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + if '.f4m' in manifest_url: + formats.extend(self._extract_f4m_formats( + manifest_url + '&hdcore=3.2.0', + display_id, f4m_id='hds', fatal=False)) + else: + formats.extend(self._extract_m3u8_formats( + manifest_url, display_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) self._sort_formats(formats) return { 'id': video_id, 'display_id': display_id, 'title': video_info.find('title').text, - 'thumbnail': self._og_search_thumbnail(webpage), + 'thumbnail': 'http://tv.dfb.de/images/%s_640x360.jpg' % video_id, 'upload_date': unified_strdate(video_info.find('time_date').text), 'formats': formats, } diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py index 5f1275b39..55853f76f 100644 --- a/youtube_dl/extractor/discovery.py +++ b/youtube_dl/extractor/discovery.py @@ -33,6 +33,7 @@ class DiscoveryIE(InfoExtractor): 'duration': 156, 'timestamp': 1302032462, 'upload_date': '20110405', + 'uploader_id': '103207', }, 'params': { 'skip_download': True, # requires ffmpeg @@ -54,7 +55,11 @@ class DiscoveryIE(InfoExtractor): 'upload_date': '20140725', 'timestamp': 1406246400, 'duration': 116, + 'uploader_id': '103207', }, + 'params': { + 'skip_download': True, # requires ffmpeg + } }] def _real_extract(self, url): @@ -66,13 +71,19 @@ class DiscoveryIE(InfoExtractor): entries = [] for idx, video_info in enumerate(info['playlist']): - formats = self._extract_m3u8_formats( - video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls', - note='Download m3u8 information for video %d' % (idx + 1)) - self._sort_formats(formats) + subtitles = {} + caption_url = video_info.get('captionsUrl') + if caption_url: + subtitles = { + 'en': [{ + 'url': caption_url, + }] + } + entries.append({ + '_type': 'url_transparent', + 'url': 'http://players.brightcove.net/103207/default_default/index.html?videoId=ref:%s' % video_info['referenceId'], 'id': compat_str(video_info['id']), - 'formats': formats, 'title': video_info['title'], 'description': video_info.get('description'), 'duration': parse_duration(video_info.get('video_length')), @@ -80,6 +91,7 @@ class DiscoveryIE(InfoExtractor): 'thumbnail': video_info.get('thumbnailURL'), 'alt_title': video_info.get('secondary_title'), 'timestamp': parse_iso8601(video_info.get('publishedDate')), + 'subtitles': subtitles, }) return self.playlist_result(entries, display_id, video_title) diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py index 3915cb182..ce6962755 100644 --- a/youtube_dl/extractor/douyutv.py +++ b/youtube_dl/extractor/douyutv.py @@ -18,7 +18,7 @@ class DouyuTVIE(InfoExtractor): 'display_id': 'iseven', 'ext': 'flv', 'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', - 'description': 'md5:f34981259a03e980a3c6404190a3ed61', + 'description': 're:.*m7show@163\.com.*', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': '7师傅', 'uploader_id': '431925', @@ -43,7 +43,7 @@ class DouyuTVIE(InfoExtractor): 'params': { 'skip_download': True, }, - 'skip': 'Romm not found', + 'skip': 'Room not found', }, { 'url': 'http://www.douyutv.com/17732', 'info_dict': { @@ -51,7 +51,7 @@ class DouyuTVIE(InfoExtractor): 'display_id': '17732', 'ext': 'flv', 'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', - 'description': 'md5:f34981259a03e980a3c6404190a3ed61', + 'description': 're:.*m7show@163\.com.*', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': '7师傅', 'uploader_id': '431925', @@ -75,13 +75,28 @@ class DouyuTVIE(InfoExtractor): room_id = self._html_search_regex( r'"room_id"\s*:\s*(\d+),', page, 'room id') - prefix = 'room/%s?aid=android&client_sys=android&time=%d' % ( - room_id, int(time.time())) - - auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest() - config = self._download_json( - 'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth), - video_id) + config = None + # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache" + # Retry with different parameters - same parameters cause same errors + for i in range(5): + prefix = 'room/%s?aid=android&client_sys=android&time=%d' % ( + room_id, int(time.time())) + auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest() + + config_page = self._download_webpage( + 'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth), + video_id) + try: + config = self._parse_json(config_page, video_id, fatal=False) + except ExtractorError: + # Wait some time before retrying to get a different time() value + self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. ' + 'Waiting for %(timeout)s seconds before retrying') + continue + else: + break + if config is None: + raise ExtractorError('Unable to fetch API result') data = config['data'] diff --git a/youtube_dl/extractor/dw.py b/youtube_dl/extractor/dw.py index ae7c571bd..d740652f1 100644 --- a/youtube_dl/extractor/dw.py +++ b/youtube_dl/extractor/dw.py @@ -2,13 +2,16 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import int_or_none +from ..utils import ( + int_or_none, + unified_strdate, +) from ..compat import compat_urlparse class DWIE(InfoExtractor): IE_NAME = 'dw' - _VALID_URL = r'https?://(?:www\.)?dw\.com/(?:[^/]+/)+av-(?P<id>\d+)' + _VALID_URL = r'https?://(?:www\.)?dw\.com/(?:[^/]+/)+(?:av|e)-(?P<id>\d+)' _TESTS = [{ # video 'url': 'http://www.dw.com/en/intelligent-light/av-19112290', @@ -31,6 +34,18 @@ class DWIE(InfoExtractor): 'description': 'md5:bc9ca6e4e063361e21c920c53af12405', 'upload_date': '20160311', } + }, { + # DW documentaries, only last for one or two weeks + 'url': 'http://www.dw.com/en/documentaries-welcome-to-the-90s-2016-05-21/e-19220158-9798', + 'md5': '56b6214ef463bfb9a3b71aeb886f3cf1', + 'info_dict': { + 'id': '19274438', + 'ext': 'mp4', + 'title': 'Welcome to the 90s – Hip Hop', + 'description': 'Welcome to the 90s - The Golden Decade of Hip Hop', + 'upload_date': '20160521', + }, + 'skip': 'Video removed', }] def _real_extract(self, url): @@ -38,6 +53,7 @@ class DWIE(InfoExtractor): webpage = self._download_webpage(url, media_id) hidden_inputs = self._hidden_inputs(webpage) title = hidden_inputs['media_title'] + media_id = hidden_inputs.get('media_id') or media_id if hidden_inputs.get('player_type') == 'video' and hidden_inputs.get('stream_file') == '1': formats = self._extract_smil_formats( @@ -49,13 +65,20 @@ class DWIE(InfoExtractor): else: formats = [{'url': hidden_inputs['file_name']}] + upload_date = hidden_inputs.get('display_date') + if not upload_date: + upload_date = self._html_search_regex( + r'<span[^>]+class="date">([0-9.]+)\s*\|', webpage, + 'upload date', default=None) + upload_date = unified_strdate(upload_date) + return { 'id': media_id, 'title': title, 'description': self._og_search_description(webpage), 'thumbnail': hidden_inputs.get('preview_image'), 'duration': int_or_none(hidden_inputs.get('file_duration')), - 'upload_date': hidden_inputs.get('display_date'), + 'upload_date': upload_date, 'formats': formats, } diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index 0f8c73fd7..12d28d3b9 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -23,7 +23,7 @@ class EaglePlatformIE(InfoExtractor): _TESTS = [{ # http://lenta.ru/news/2015/03/06/navalny/ 'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201', - 'md5': '881ee8460e1b7735a8be938e2ffb362b', + # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used 'info_dict': { 'id': '227304', 'ext': 'mp4', @@ -51,6 +51,14 @@ class EaglePlatformIE(InfoExtractor): }] @staticmethod + def _extract_url(webpage): + mobj = re.search( + r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1', + webpage) + if mobj is not None: + return mobj.group('url') + + @staticmethod def _handle_error(response): status = int_or_none(response.get('status', 200)) if status != 200: @@ -109,8 +117,11 @@ class EaglePlatformIE(InfoExtractor): mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url']) if mobj: http_format = m3u8_format.copy() + video_url = mp4_url.replace(mp4_url_basename, mobj.group(1)) + if not self._is_valid_url(video_url, video_id): + continue http_format.update({ - 'url': mp4_url.replace(mp4_url_basename, mobj.group(1)), + 'url': video_url, 'format_id': m3u8_format['format_id'].replace('hls', 'http'), 'protocol': 'http', }) diff --git a/youtube_dl/extractor/eporner.py b/youtube_dl/extractor/eporner.py index e006921ec..ac5d0fe24 100644 --- a/youtube_dl/extractor/eporner.py +++ b/youtube_dl/extractor/eporner.py @@ -11,8 +11,8 @@ from ..utils import ( class EpornerIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\d+)/(?P<display_id>[\w-]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)/(?P<display_id>[\w-]+)' + _TESTS = [{ 'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/', 'md5': '39d486f046212d8e1b911c52ab4691f8', 'info_dict': { @@ -23,8 +23,12 @@ class EpornerIE(InfoExtractor): 'duration': 1838, 'view_count': int, 'age_limit': 18, - } - } + }, + }, { + # New (May 2016) URL layout + 'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) diff --git a/youtube_dl/extractor/espn.py b/youtube_dl/extractor/espn.py index db4b263bc..66c08bec4 100644 --- a/youtube_dl/extractor/espn.py +++ b/youtube_dl/extractor/espn.py @@ -8,6 +8,7 @@ class ESPNIE(InfoExtractor): _VALID_URL = r'https?://espn\.go\.com/(?:[^/]+/)*(?P<id>[^/]+)' _TESTS = [{ 'url': 'http://espn.go.com/video/clip?id=10365079', + 'md5': '60e5d097a523e767d06479335d1bdc58', 'info_dict': { 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG', 'ext': 'mp4', @@ -15,21 +16,22 @@ class ESPNIE(InfoExtractor): 'description': None, }, 'params': { - # m3u8 download 'skip_download': True, }, + 'add_ie': ['OoyalaExternal'], }, { # intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season 'url': 'http://espn.go.com/video/clip?id=2743663', + 'md5': 'f4ac89b59afc7e2d7dbb049523df6768', 'info_dict': { 'id': '50NDFkeTqRHB0nXBOK-RGdSG5YQPuxHg', 'ext': 'mp4', 'title': 'Must-See Moments: Best of the MLS season', }, 'params': { - # m3u8 download 'skip_download': True, }, + 'add_ie': ['OoyalaExternal'], }, { 'url': 'https://espn.go.com/video/iframe/twitter/?cms=espn&id=10365079', 'only_matching': True, diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 6de3438fc..9f70ce752 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -3,6 +3,10 @@ from __future__ import unicode_literals from .abc import ABCIE from .abc7news import Abc7NewsIE +from .abcnews import ( + AbcNewsIE, + AbcNewsVideoIE, +) from .academicearth import AcademicEarthCourseIE from .acast import ( ACastIE, @@ -16,7 +20,11 @@ from .adobetv import ( AdobeTVVideoIE, ) from .adultswim import AdultSwimIE -from .aenetworks import AENetworksIE +from .aenetworks import ( + AENetworksIE, + HistoryTopicIE, +) +from .afreecatv import AfreecaTVIE from .aftonbladet import AftonbladetIE from .airmozilla import AirMozillaIE from .aljazeera import AlJazeeraIE @@ -39,7 +47,6 @@ from .archiveorg import ArchiveOrgIE from .ard import ( ARDIE, ARDMediathekIE, - SportschauIE, ) from .arte import ( ArteTvIE, @@ -52,6 +59,7 @@ from .arte import ( ArteTVDDCIE, ArteTVMagazineIE, ArteTVEmbedIE, + ArteTVPlaylistIE, ) from .atresplayer import AtresPlayerIE from .atttechchannel import ATTTechChannelIE @@ -65,6 +73,8 @@ from .bandcamp import BandcampIE, BandcampAlbumIE from .bbc import ( BBCCoUkIE, BBCCoUkArticleIE, + BBCCoUkIPlayerPlaylistIE, + BBCCoUkPlaylistIE, BBCIE, ) from .beeg import BeegIE @@ -75,6 +85,7 @@ from .bigflix import BigflixIE from .bild import BildIE from .bilibili import BiliBiliIE from .biobiochiletv import BioBioChileTVIE +from .biqle import BIQLEIE from .bleacherreport import ( BleacherReportIE, BleacherReportCMSIE, @@ -101,11 +112,16 @@ from .camwithher import CamWithHerIE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE from .canvas import CanvasIE +from .carambatv import ( + CarambaTVIE, + CarambaTVPageIE, +) from .cbc import ( CBCIE, CBCPlayerIE, ) from .cbs import CBSIE +from .cbslocal import CBSLocalIE from .cbsinteractive import CBSInteractiveIE from .cbsnews import ( CBSNewsIE, @@ -123,11 +139,11 @@ from .chirbit import ( ChirbitProfileIE, ) from .cinchcast import CinchcastIE -from .cinemassacre import CinemassacreIE -from .cliprs import ClipRsIE from .clipfish import ClipfishIE from .cliphunter import CliphunterIE +from .cliprs import ClipRsIE from .clipsyndicate import ClipsyndicateIE +from .closertotruth import CloserToTruthIE from .cloudy import CloudyIE from .clubic import ClubicIE from .clyp import ClypIE @@ -138,7 +154,7 @@ from .cnn import ( CNNBlogsIE, CNNArticleIE, ) -from .collegehumor import CollegeHumorIE +from .coub import CoubIE from .collegerama import CollegeRamaIE from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE from .comcarcoff import ComCarCoffIE @@ -155,8 +171,11 @@ from .crunchyroll import ( ) from .cspan import CSpanIE from .ctsnews import CtsNewsIE +from .ctv import CTVIE +from .ctvnews import CTVNewsIE from .cultureunplugged import CultureUnpluggedIE from .cwtv import CWTVIE +from .dailymail import DailyMailIE from .dailymotion import ( DailymotionIE, DailymotionPlaylistIE, @@ -226,6 +245,7 @@ from .everyonesmixtape import EveryonesMixtapeIE from .exfm import ExfmIE from .expotv import ExpoTVIE from .extremetube import ExtremeTubeIE +from .eyedotv import EyedoTVIE from .facebook import FacebookIE from .faz import FazIE from .fc2 import FC2IE @@ -236,8 +256,10 @@ from .fivemin import FiveMinIE from .fivetv import FiveTVIE from .fktv import FKTVIE from .flickr import FlickrIE +from .flipagram import FlipagramIE from .folketinget import FolketingetIE from .footyroom import FootyRoomIE +from .formula1 import Formula1IE from .fourtube import FourTubeIE from .fox import FOXIE from .foxgay import FoxgayIE @@ -260,6 +282,7 @@ from .freespeech import FreespeechIE from .freevideo import FreeVideoIE from .funimation import FunimationIE from .funnyordie import FunnyOrDieIE +from .fusion import FusionIE from .gameinformer import GameInformerIE from .gamekings import GamekingsIE from .gameone import ( @@ -269,7 +292,6 @@ from .gameone import ( from .gamersyde import GamersydeIE from .gamespot import GameSpotIE from .gamestar import GameStarIE -from .gametrailers import GametrailersIE from .gazeta import GazetaIE from .gdcvault import GDCVaultIE from .generic import GenericIE @@ -282,6 +304,7 @@ from .globo import ( GloboArticleIE, ) from .godtube import GodTubeIE +from .godtv import GodTVIE from .goldenmoustache import GoldenMoustacheIE from .golem import GolemIE from .googledrive import GoogleDriveIE @@ -304,6 +327,10 @@ from .hotnewhiphop import HotNewHipHopIE from .hotstar import HotStarIE from .howcast import HowcastIE from .howstuffworks import HowStuffWorksIE +from .hrti import ( + HRTiIE, + HRTiPlaylistIE, +) from .huffpost import HuffPostIE from .hypem import HypemIE from .iconosquare import IconosquareIE @@ -342,6 +369,7 @@ from .jove import JoveIE from .jwplatform import JWPlatformIE from .jpopsukitv import JpopsukiIE from .kaltura import KalturaIE +from .kamcord import KamcordIE from .kanalplay import KanalPlayIE from .kankan import KankanIE from .karaoketv import KaraoketvIE @@ -365,6 +393,7 @@ from .kuwo import ( ) from .la7 import LA7IE from .laola1tv import Laola1TvIE +from .learnr import LearnrIE from .lecture2go import Lecture2GoIE from .lemonde import LemondeIE from .leeco import ( @@ -372,6 +401,7 @@ from .leeco import ( LePlaylistIE, LetvCloudIE, ) +from .libraryofcongress import LibraryOfCongressIE from .libsyn import LibsynIE from .lifenews import ( LifeNewsIE, @@ -382,6 +412,7 @@ from .limelight import ( LimelightChannelIE, LimelightChannelListIE, ) +from .litv import LiTVIE from .liveleak import LiveLeakIE from .livestream import ( LivestreamIE, @@ -389,6 +420,7 @@ from .livestream import ( LivestreamShortenerIE, ) from .lnkgo import LnkGoIE +from .localnews8 import LocalNews8IE from .lovehomeporn import LoveHomePornIE from .lrt import LRTIE from .lynda import ( @@ -400,13 +432,17 @@ from .macgamestore import MacGameStoreIE from .mailru import MailRuIE from .makerschannel import MakersChannelIE from .makertv import MakerTVIE -from .malemotion import MalemotionIE from .matchtv import MatchTVIE from .mdr import MDRIE +from .meta import METAIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mgoon import MgoonIE from .mgtv import MGTVIE +from .microsoftvirtualacademy import ( + MicrosoftVirtualAcademyIE, + MicrosoftVirtualAcademyCourseIE, +) from .minhateca import MinhatecaIE from .ministrygrid import MinistryGridIE from .minoto import MinotoIE @@ -431,6 +467,7 @@ from .motherless import MotherlessIE from .motorsport import MotorsportIE from .movieclips import MovieClipsIE from .moviezine import MoviezineIE +from .msn import MSNIE from .mtv import ( MTVIE, MTVServicesEmbeddedIE, @@ -439,8 +476,7 @@ from .mtv import ( ) from .muenchentv import MuenchenTVIE from .musicplayon import MusicPlayOnIE -from .muzu import MuzuTVIE -from .mwave import MwaveIE +from .mwave import MwaveIE, MwaveMeetGreetIE from .myspace import MySpaceIE, MySpaceAlbumIE from .myspass import MySpassIE from .myvi import MyviIE @@ -458,7 +494,6 @@ from .nbc import ( NBCNewsIE, NBCSportsIE, NBCSportsVPlayerIE, - MSNBCIE, ) from .ndr import ( NDRIE, @@ -495,8 +530,12 @@ from .nhl import ( NHLVideocenterCategoryIE, NHLIE, ) -from .nick import NickIE +from .nick import ( + NickIE, + NickDeIE, +) from .niconico import NiconicoIE, NiconicoPlaylistIE +from .ninecninemedia import NineCNineMediaIE from .ninegag import NineGagIE from .noco import NocoIE from .normalboots import NormalbootsIE @@ -544,6 +583,10 @@ from .nytimes import ( from .nuvid import NuvidIE from .odnoklassniki import OdnoklassnikiIE from .oktoberfesttv import OktoberfestTVIE +from .onet import ( + OnetIE, + OnetChannelIE, +) from .onionstudios import OnionStudiosIE from .ooyala import ( OoyalaIE, @@ -562,7 +605,10 @@ from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE from .pbs import PBSIE from .people import PeopleIE -from .periscope import PeriscopeIE +from .periscope import ( + PeriscopeIE, + PeriscopeUserIE, +) from .philharmoniedeparis import PhilharmonieDeParisIE from .phoenix import PhoenixIE from .photobucket import PhotobucketIE @@ -579,6 +625,7 @@ from .pluralsight import ( PluralsightCourseIE, ) from .podomatic import PodomaticIE +from .polskieradio import PolskieRadioIE from .porn91 import Porn91IE from .pornhd import PornHdIE from .pornhub import ( @@ -602,7 +649,14 @@ from .qqmusic import ( QQMusicToplistIE, QQMusicPlaylistIE, ) -from .r7 import R7IE +from .r7 import ( + R7IE, + R7ArticleIE, +) +from .radiocanada import ( + RadioCanadaIE, + RadioCanadaAudioVideoIE, +) from .radiode import RadioDeIE from .radiojavan import RadioJavanIE from .radiobremen import RadioBremenIE @@ -616,11 +670,16 @@ from .rds import RDSIE from .redtube import RedTubeIE from .regiotv import RegioTVIE from .restudy import RestudyIE +from .reuters import ReutersIE from .reverbnation import ReverbNationIE -from .revision3 import Revision3IE +from .revision3 import ( + Revision3EmbedIE, + Revision3IE, +) from .rice import RICEIE from .ringtv import RingTVIE from .ro220 import Ro220IE +from .rockstargames import RockstarGamesIE from .rottentomatoes import RottenTomatoesIE from .roxwel import RoxwelIE from .rtbf import RTBFIE @@ -656,18 +715,21 @@ from .screencast import ScreencastIE from .screencastomatic import ScreencastOMaticIE from .screenjunkies import ScreenJunkiesIE from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE +from .seeker import SeekerIE from .senateisvp import SenateISVPIE +from .sendtonews import SendtoNewsIE from .servingsys import ServingSysIE from .sexu import SexuIE -from .sexykarma import SexyKarmaIE from .shahid import ShahidIE from .shared import SharedIE from .sharesix import ShareSixIE from .sina import SinaIE +from .sixplay import SixPlayIE from .skynewsarabia import ( SkyNewsArabiaIE, SkyNewsArabiaArticleIE, ) +from .skysports import SkySportsIE from .slideshare import SlideshareIE from .slutload import SlutloadIE from .smotri import ( @@ -676,10 +738,6 @@ from .smotri import ( SmotriUserIE, SmotriBroadcastIE, ) -from .snagfilms import ( - SnagFilmsIE, - SnagFilmsEmbedIE, -) from .snotr import SnotrIE from .sohu import SohuIE from .soundcloud import ( @@ -712,6 +770,7 @@ from .sportbox import ( SportBoxEmbedIE, ) from .sportdeutschland import SportDeutschlandIE +from .sportschau import SportschauIE from .srgssr import ( SRGSSRIE, SRGSSRPlayIE, @@ -731,7 +790,10 @@ from .svt import ( from .swrmediathek import SWRMediathekIE from .syfy import SyfyIE from .sztvhu import SztvHuIE -from .tagesschau import TagesschauIE +from .tagesschau import ( + TagesschauPlayerIE, + TagesschauIE, +) from .tapely import TapelyIE from .tass import TassIE from .tdslifeway import TDSLifewayIE @@ -749,6 +811,7 @@ from .telecinco import TelecincoIE from .telegraaf import TelegraafIE from .telemb import TeleMBIE from .teletask import TeleTaskIE +from .telewebion import TelewebionIE from .testurl import TestURLIE from .tf1 import TF1IE from .theintercept import TheInterceptIE @@ -761,6 +824,7 @@ from .thesixtyone import TheSixtyOneIE from .thestar import TheStarIE from .thisamericanlife import ThisAmericanLifeIE from .thisav import ThisAVIE +from .threeqsdn import ThreeQSDNIE from .tinypic import TinyPicIE from .tlc import TlcDeIE from .tmz import ( @@ -813,7 +877,10 @@ from .tvc import ( ) from .tvigle import TvigleIE from .tvland import TVLandIE -from .tvp import TvpIE, TvpSeriesIE +from .tvp import ( + TVPIE, + TVPSeriesIE, +) from .tvplay import TVPlayIE from .tweakers import TweakersIE from .twentyfourvideo import TwentyFourVideoIE @@ -828,8 +895,8 @@ from .twitch import ( TwitchVodIE, TwitchProfileIE, TwitchPastBroadcastsIE, - TwitchBookmarksIE, TwitchStreamIE, + TwitchClipsIE, ) from .twitter import ( TwitterCardIE, @@ -844,16 +911,23 @@ from .udn import UDNEmbedIE from .digiteka import DigitekaIE from .unistra import UnistraIE from .urort import UrortIE +from .urplay import URPlayIE from .usatoday import USATodayIE from .ustream import UstreamIE, UstreamChannelIE -from .ustudio import UstudioIE +from .ustudio import ( + UstudioIE, + UstudioEmbedIE, +) from .varzesh3 import Varzesh3IE from .vbox7 import Vbox7IE from .veehd import VeeHDIE from .veoh import VeohIE from .vessel import VesselIE from .vesti import VestiIE -from .vevo import VevoIE +from .vevo import ( + VevoIE, + VevoPlaylistIE, +) from .vgtv import ( BTArticleIE, BTVestlendingenIE, @@ -864,6 +938,7 @@ from .vice import ( ViceIE, ViceShowIE, ) +from .vidbit import VidbitIE from .viddler import ViddlerIE from .videodetective import VideoDetectiveIE from .videofyme import VideofyMeIE @@ -875,6 +950,7 @@ from .videomore import ( ) from .videopremium import VideoPremiumIE from .videott import VideoTtIE +from .vidio import VidioIE from .vidme import ( VidmeIE, VidmeUserIE, @@ -882,6 +958,10 @@ from .vidme import ( ) from .vidzi import VidziIE from .vier import VierIE, VierVideosIE +from .viewlift import ( + ViewLiftIE, + ViewLiftEmbedIE, +) from .viewster import ViewsterIE from .viidea import ViideaIE from .vimeo import ( @@ -916,25 +996,29 @@ from .vporn import VpornIE from .vrt import VRTIE from .vube import VubeIE from .vuclip import VuClipIE -from .vulture import VultureIE from .walla import WallaIE -from .washingtonpost import WashingtonPostIE +from .washingtonpost import ( + WashingtonPostIE, + WashingtonPostArticleIE, +) from .wat import WatIE +from .watchindianporn import WatchIndianPornIE from .wdr import ( WDRIE, WDRMobileIE, - WDRMausIE, ) from .webofstories import ( WebOfStoriesIE, WebOfStoriesPlaylistIE, ) -from .weibo import WeiboIE from .weiqitv import WeiqiTVIE from .wimp import WimpIE from .wistia import WistiaIE from .worldstarhiphop import WorldStarHipHopIE -from .wrzuta import WrzutaIE +from .wrzuta import ( + WrzutaIE, + WrzutaPlaylistIE, +) from .wsj import WSJIE from .xbef import XBefIE from .xboxclips import XboxClipsIE @@ -943,6 +1027,12 @@ from .xhamster import ( XHamsterIE, XHamsterEmbedIE, ) +from .xiami import ( + XiamiSongIE, + XiamiAlbumIE, + XiamiArtistIE, + XiamiCollectionIE +) from .xminus import XMinusIE from .xnxx import XNXXIE from .xstream import XstreamIE @@ -964,7 +1054,10 @@ from .yesjapan import YesJapanIE from .yinyuetai import YinYueTaiIE from .ynet import YnetIE from .youjizz import YouJizzIE -from .youku import YoukuIE +from .youku import ( + YoukuIE, + YoukuShowIE, +) from .youporn import YouPornIE from .yourupload import YourUploadIE from .youtube import ( @@ -979,6 +1072,7 @@ from .youtube import ( YoutubeSearchDateIE, YoutubeSearchIE, YoutubeSearchURLIE, + YoutubeSharedVideoIE, YoutubeShowIE, YoutubeSubscriptionsIE, YoutubeTruncatedIDIE, diff --git a/youtube_dl/extractor/eyedotv.py b/youtube_dl/extractor/eyedotv.py new file mode 100644 index 000000000..2f3035147 --- /dev/null +++ b/youtube_dl/extractor/eyedotv.py @@ -0,0 +1,64 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + xpath_text, + parse_duration, + ExtractorError, +) + + +class EyedoTVIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?eyedo\.tv/[^/]+/(?:#!/)?Live/Detail/(?P<id>[0-9]+)' + _TEST = { + 'url': 'https://www.eyedo.tv/en-US/#!/Live/Detail/16301', + 'md5': 'ba14f17995cdfc20c36ba40e21bf73f7', + 'info_dict': { + 'id': '16301', + 'ext': 'mp4', + 'title': 'Journée du conseil scientifique de l\'Afnic 2015', + 'description': 'md5:4abe07293b2f73efc6e1c37028d58c98', + 'uploader': 'Afnic Live', + 'uploader_id': '8023', + } + } + _ROOT_URL = 'http://live.eyedo.net:1935/' + + def _real_extract(self, url): + video_id = self._match_id(url) + video_data = self._download_xml('http://eyedo.tv/api/live/GetLive/%s' % video_id, video_id) + + def _add_ns(path): + return self._xpath_ns(path, 'http://schemas.datacontract.org/2004/07/EyeDo.Core.Implementation.Web.ViewModels.Api') + + title = xpath_text(video_data, _add_ns('Titre'), 'title', True) + state_live_code = xpath_text(video_data, _add_ns('StateLiveCode'), 'title', True) + if state_live_code == 'avenir': + raise ExtractorError( + '%s said: We\'re sorry, but this video is not yet available.' % self.IE_NAME, + expected=True) + + is_live = state_live_code == 'live' + m3u8_url = None + # http://eyedo.tv/Content/Html5/Scripts/html5view.js + if is_live: + if xpath_text(video_data, 'Cdn') == 'true': + m3u8_url = 'http://rrr.sz.xlcdn.com/?account=eyedo&file=A%s&type=live&service=wowza&protocol=http&output=playlist.m3u8' % video_id + else: + m3u8_url = self._ROOT_URL + 'w/%s/eyedo_720p/playlist.m3u8' % video_id + else: + m3u8_url = self._ROOT_URL + 'replay-w/%s/mp4:%s.mp4/playlist.m3u8' % (video_id, video_id) + + return { + 'id': video_id, + 'title': title, + 'formats': self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8' if is_live else 'm3u8_native'), + 'description': xpath_text(video_data, _add_ns('Description')), + 'duration': parse_duration(xpath_text(video_data, _add_ns('Duration'))), + 'uploader': xpath_text(video_data, _add_ns('Createur')), + 'uploader_id': xpath_text(video_data, _add_ns('CreateurId')), + 'chapter': xpath_text(video_data, _add_ns('ChapitreTitre')), + 'chapter_id': xpath_text(video_data, _add_ns('ChapitreId')), + } diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index f5bbd39d2..0d43acc4a 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -129,6 +129,21 @@ class FacebookIE(InfoExtractor): 'only_matching': True, }] + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage) + if mobj is not None: + return mobj.group('url') + + # Facebook API embed + # see https://developers.facebook.com/docs/plugins/embedded-video-player + mobj = re.search(r'''(?x)<div[^>]+ + class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+ + data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage) + if mobj is not None: + return mobj.group('url') + def _login(self): (useremail, password) = self._get_login_info() if useremail is None: @@ -204,12 +219,23 @@ class FacebookIE(InfoExtractor): BEFORE = '{swf.addParam(param[0], param[1]);});' AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});' - m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage) - if m: - swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"') + PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER) + + for m in re.findall(PATTERN, webpage): + swf_params = m.replace('\\\\', '\\').replace('\\"', '"') data = dict(json.loads(swf_params)) params_raw = compat_urllib_parse_unquote(data['params']) - video_data = json.loads(params_raw)['video_data'] + video_data_candidate = json.loads(params_raw)['video_data'] + for _, f in video_data_candidate.items(): + if not f: + continue + if isinstance(f, dict): + f = [f] + if isinstance(f, list): + continue + if f[0].get('video_id') == video_id: + video_data = video_data_candidate + break def video_data_list2dict(video_data): ret = {} @@ -239,6 +265,8 @@ class FacebookIE(InfoExtractor): formats = [] for format_id, f in video_data.items(): + if f and isinstance(f, dict): + f = [f] if not f or not isinstance(f, list): continue for quality in ('sd', 'hd'): diff --git a/youtube_dl/extractor/fczenit.py b/youtube_dl/extractor/fczenit.py index f1f150ef2..8d1010b88 100644 --- a/youtube_dl/extractor/fczenit.py +++ b/youtube_dl/extractor/fczenit.py @@ -1,20 +1,19 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..compat import compat_urlparse class FczenitIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/gl(?P<id>[0-9]+)' + _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/(?P<id>[0-9]+)' _TEST = { - 'url': 'http://fc-zenit.ru/video/gl6785/', - 'md5': '458bacc24549173fe5a5aa29174a5606', + 'url': 'http://fc-zenit.ru/video/41044/', + 'md5': '0e3fab421b455e970fa1aa3891e57df0', 'info_dict': { - 'id': '6785', + 'id': '41044', 'ext': 'mp4', - 'title': '«Зенит-ТВ»: как Олег Шатов играл против «Урала»', + 'title': 'Так пишется история: казанский разгром ЦСКА на «Зенит-ТВ»', }, } @@ -22,15 +21,23 @@ class FczenitIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_title = self._html_search_regex(r'<div class=\"photoalbum__title\">([^<]+)', webpage, 'title') + video_title = self._html_search_regex( + r'<[^>]+class=\"photoalbum__title\">([^<]+)', webpage, 'title') + + video_items = self._parse_json(self._search_regex( + r'arrPath\s*=\s*JSON\.parse\(\'(.+)\'\)', webpage, 'video items'), + video_id) - bitrates_raw = self._html_search_regex(r'bitrates:.*\n(.*)\]', webpage, 'video URL') - bitrates = re.findall(r'url:.?\'(.+?)\'.*?bitrate:.?([0-9]{3}?)', bitrates_raw) + def merge_dicts(*dicts): + ret = {} + for a_dict in dicts: + ret.update(a_dict) + return ret formats = [{ - 'url': furl, - 'tbr': tbr, - } for furl, tbr in bitrates] + 'url': compat_urlparse.urljoin(url, video_url), + 'tbr': int(tbr), + } for tbr, video_url in merge_dicts(*video_items).items()] self._sort_formats(formats) diff --git a/youtube_dl/extractor/flickr.py b/youtube_dl/extractor/flickr.py index 0a3de1498..a8e1bf42a 100644 --- a/youtube_dl/extractor/flickr.py +++ b/youtube_dl/extractor/flickr.py @@ -24,13 +24,28 @@ class FlickrIE(InfoExtractor): 'upload_date': '20110423', 'uploader_id': '10922353@N03', 'uploader': 'Forest Wander', + 'uploader_url': 'https://www.flickr.com/photos/forestwander-nature-pictures/', 'comment_count': int, 'view_count': int, 'tags': list, + 'license': 'Attribution-ShareAlike', } } - _API_BASE_URL = 'https://api.flickr.com/services/rest?' + # https://help.yahoo.com/kb/flickr/SLN25525.html + _LICENSES = { + '0': 'All Rights Reserved', + '1': 'Attribution-NonCommercial-ShareAlike', + '2': 'Attribution-NonCommercial', + '3': 'Attribution-NonCommercial-NoDerivs', + '4': 'Attribution', + '5': 'Attribution-ShareAlike', + '6': 'Attribution-NoDerivs', + '7': 'No known copyright restrictions', + '8': 'United States government work', + '9': 'Public Domain Dedication (CC0)', + '10': 'Public Domain Work', + } def _call_api(self, method, video_id, api_key, note, secret=None): query = { @@ -75,6 +90,9 @@ class FlickrIE(InfoExtractor): self._sort_formats(formats) owner = video_info.get('owner', {}) + uploader_id = owner.get('nsid') + uploader_path = owner.get('path_alias') or uploader_id + uploader_url = 'https://www.flickr.com/photos/%s/' % uploader_path if uploader_path else None return { 'id': video_id, @@ -83,11 +101,13 @@ class FlickrIE(InfoExtractor): 'formats': formats, 'timestamp': int_or_none(video_info.get('dateuploaded')), 'duration': int_or_none(video_info.get('video', {}).get('duration')), - 'uploader_id': owner.get('nsid'), + 'uploader_id': uploader_id, 'uploader': owner.get('realname'), + 'uploader_url': uploader_url, 'comment_count': int_or_none(video_info.get('comments', {}).get('_content')), 'view_count': int_or_none(video_info.get('views')), - 'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])] + 'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])], + 'license': self._LICENSES.get(video_info.get('license')), } else: raise ExtractorError('not a video', expected=True) diff --git a/youtube_dl/extractor/flipagram.py b/youtube_dl/extractor/flipagram.py new file mode 100644 index 000000000..acb6133ff --- /dev/null +++ b/youtube_dl/extractor/flipagram.py @@ -0,0 +1,115 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + int_or_none, + float_or_none, + try_get, + unified_timestamp, +) + + +class FlipagramIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?#&]+)' + _TEST = { + 'url': 'https://flipagram.com/f/nyvTSJMKId', + 'md5': '888dcf08b7ea671381f00fab74692755', + 'info_dict': { + 'id': 'nyvTSJMKId', + 'ext': 'mp4', + 'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction', + 'description': 'md5:d55e32edc55261cae96a41fa85ff630e', + 'duration': 35.571, + 'timestamp': 1461244995, + 'upload_date': '20160421', + 'uploader': 'kitty juria', + 'uploader_id': 'sjuria101', + 'creator': 'kitty juria', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + 'comments': list, + 'formats': 'mincount:2', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + video_data = self._parse_json( + self._search_regex( + r'window\.reactH2O\s*=\s*({.+});', webpage, 'video data'), + video_id) + + flipagram = video_data['flipagram'] + video = flipagram['video'] + + json_ld = self._search_json_ld(webpage, video_id, default=False) + title = json_ld.get('title') or flipagram['captionText'] + description = json_ld.get('description') or flipagram.get('captionText') + + formats = [{ + 'url': video['url'], + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), + 'filesize': int_or_none(video_data.get('size')), + }] + + preview_url = try_get( + flipagram, lambda x: x['music']['track']['previewUrl'], compat_str) + if preview_url: + formats.append({ + 'url': preview_url, + 'ext': 'm4a', + 'vcodec': 'none', + }) + + self._sort_formats(formats) + + counts = flipagram.get('counts', {}) + user = flipagram.get('user', {}) + video_data = flipagram.get('video', {}) + + thumbnails = [{ + 'url': self._proto_relative_url(cover['url']), + 'width': int_or_none(cover.get('width')), + 'height': int_or_none(cover.get('height')), + 'filesize': int_or_none(cover.get('size')), + } for cover in flipagram.get('covers', []) if cover.get('url')] + + # Note that this only retrieves comments that are initally loaded. + # For videos with large amounts of comments, most won't be retrieved. + comments = [] + for comment in video_data.get('comments', {}).get(video_id, {}).get('items', []): + text = comment.get('comment') + if not text or not isinstance(text, list): + continue + comments.append({ + 'author': comment.get('user', {}).get('name'), + 'author_id': comment.get('user', {}).get('username'), + 'id': comment.get('id'), + 'text': text[0], + 'timestamp': unified_timestamp(comment.get('created')), + }) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'duration': float_or_none(flipagram.get('duration'), 1000), + 'thumbnails': thumbnails, + 'timestamp': unified_timestamp(flipagram.get('iso8601Created')), + 'uploader': user.get('name'), + 'uploader_id': user.get('username'), + 'creator': user.get('name'), + 'view_count': int_or_none(counts.get('plays')), + 'like_count': int_or_none(counts.get('likes')), + 'repost_count': int_or_none(counts.get('reflips')), + 'comment_count': int_or_none(counts.get('comments')), + 'comments': comments, + 'formats': formats, + } diff --git a/youtube_dl/extractor/formula1.py b/youtube_dl/extractor/formula1.py new file mode 100644 index 000000000..322c41e5a --- /dev/null +++ b/youtube_dl/extractor/formula1.py @@ -0,0 +1,26 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class Formula1IE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?formula1\.com/content/fom-website/en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html' + _TEST = { + 'url': 'http://www.formula1.com/content/fom-website/en/video/2016/5/Race_highlights_-_Spain_2016.html', + 'md5': '8c79e54be72078b26b89e0e111c0502b', + 'info_dict': { + 'id': 'JvYXJpMzE6pArfHWm5ARp5AiUmD-gibV', + 'ext': 'flv', + 'title': 'Race highlights - Spain 2016', + }, + 'add_ie': ['Ooyala'], + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + ooyala_embed_code = self._search_regex( + r'data-videoid="([^"]+)"', webpage, 'ooyala embed code') + return self.url_result( + 'ooyala:%s' % ooyala_embed_code, 'Ooyala', ooyala_embed_code) diff --git a/youtube_dl/extractor/foxsports.py b/youtube_dl/extractor/foxsports.py index df7665176..a3bb98377 100644 --- a/youtube_dl/extractor/foxsports.py +++ b/youtube_dl/extractor/foxsports.py @@ -1,7 +1,10 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import smuggle_url +from ..utils import ( + smuggle_url, + update_url_query, +) class FoxSportsIE(InfoExtractor): @@ -9,11 +12,15 @@ class FoxSportsIE(InfoExtractor): _TEST = { 'url': 'http://www.foxsports.com/video?vid=432609859715', + 'md5': 'b49050e955bebe32c301972e4012ac17', 'info_dict': { - 'id': 'gA0bHB3Ladz3', - 'ext': 'flv', + 'id': 'i0qKWsk3qJaM', + 'ext': 'mp4', 'title': 'Courtney Lee on going up 2-0 in series vs. Blazers', 'description': 'Courtney Lee talks about Memphis being focused.', + 'upload_date': '20150423', + 'timestamp': 1429761109, + 'uploader': 'NEWA-FNG-FOXSPORTS', }, 'add_ie': ['ThePlatform'], } @@ -28,5 +35,8 @@ class FoxSportsIE(InfoExtractor): r"data-player-config='([^']+)'", webpage, 'data player config'), video_id) - return self.url_result(smuggle_url( - config['releaseURL'] + '&manifest=f4m', {'force_smil_url': True})) + return self.url_result(smuggle_url(update_url_query( + config['releaseURL'], { + 'mbr': 'true', + 'switch': 'http', + }), {'force_smil_url': True})) diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index ad94e31f3..7653975e3 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -14,7 +14,10 @@ from ..utils import ( parse_duration, determine_ext, ) -from .dailymotion import DailymotionCloudIE +from .dailymotion import ( + DailymotionIE, + DailymotionCloudIE, +) class FranceTVBaseInfoExtractor(InfoExtractor): @@ -188,6 +191,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'params': { 'skip_download': True, }, + }, { + # Dailymotion embed + 'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html', + 'md5': 'ee7f1828f25a648addc90cb2687b1f12', + 'info_dict': { + 'id': 'x4iiko0', + 'ext': 'mp4', + 'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen', + 'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016', + 'timestamp': 1467011958, + 'upload_date': '20160627', + 'uploader': 'France Inter', + 'uploader_id': 'x2q2ez', + }, + 'add_ie': ['Dailymotion'], }] def _real_extract(self, url): @@ -197,7 +215,13 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage) if dmcloud_url: - return self.url_result(dmcloud_url, 'DailymotionCloud') + return self.url_result(dmcloud_url, DailymotionCloudIE.ie_key()) + + dailymotion_urls = DailymotionIE._extract_urls(webpage) + if dailymotion_urls: + return self.playlist_result([ + self.url_result(dailymotion_url, DailymotionIE.ie_key()) + for dailymotion_url in dailymotion_urls]) video_id, catalogue = self._search_regex( (r'id-video=([^@]+@[^"]+)', diff --git a/youtube_dl/extractor/funimation.py b/youtube_dl/extractor/funimation.py index 1eb528f31..0ad0d9b6a 100644 --- a/youtube_dl/extractor/funimation.py +++ b/youtube_dl/extractor/funimation.py @@ -2,6 +2,10 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..compat import ( + compat_HTTPError, + compat_urllib_parse_unquote_plus, +) from ..utils import ( clean_html, determine_ext, @@ -27,6 +31,7 @@ class FunimationIE(InfoExtractor): 'description': 'md5:1769f43cd5fc130ace8fd87232207892', 'thumbnail': 're:https?://.*\.jpg', }, + 'skip': 'Access without user interaction is forbidden by CloudFlare, and video removed', }, { 'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play', 'info_dict': { @@ -37,6 +42,7 @@ class FunimationIE(InfoExtractor): 'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd', 'thumbnail': 're:https?://.*\.jpg', }, + 'skip': 'Access without user interaction is forbidden by CloudFlare', }, { 'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview', 'info_dict': { @@ -47,8 +53,36 @@ class FunimationIE(InfoExtractor): 'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803', 'thumbnail': 're:https?://.*\.(?:jpg|png)', }, + 'skip': 'Access without user interaction is forbidden by CloudFlare', }] + _LOGIN_URL = 'http://www.funimation.com/login' + + def _download_webpage(self, *args, **kwargs): + try: + return super(FunimationIE, self)._download_webpage(*args, **kwargs) + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: + response = ee.cause.read() + if b'>Please complete the security check to access<' in response: + raise ExtractorError( + 'Access to funimation.com is blocked by CloudFlare. ' + 'Please browse to http://www.funimation.com/, solve ' + 'the reCAPTCHA, export browser cookies to a text file,' + ' and then try again with --cookies YOUR_COOKIE_FILE.', + expected=True) + raise + + def _extract_cloudflare_session_ua(self, url): + ci_session_cookie = self._get_cookies(url).get('ci_session') + if ci_session_cookie: + ci_session = compat_urllib_parse_unquote_plus(ci_session_cookie.value) + # ci_session is a string serialized by PHP function serialize() + # This case is simple enough to use regular expressions only + return self._search_regex( + r'"user_agent";s:\d+:"([^"]+)"', ci_session, 'user agent', + default=None) + def _login(self): (username, password) = self._get_login_info() if username is None: @@ -57,8 +91,11 @@ class FunimationIE(InfoExtractor): 'email_field': username, 'password_field': password, }) - login_request = sanitized_Request('http://www.funimation.com/login', data, headers={ - 'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0', + user_agent = self._extract_cloudflare_session_ua(self._LOGIN_URL) + if not user_agent: + user_agent = 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0' + login_request = sanitized_Request(self._LOGIN_URL, data, headers={ + 'User-Agent': user_agent, 'Content-Type': 'application/x-www-form-urlencoded' }) login_page = self._download_webpage( @@ -103,11 +140,16 @@ class FunimationIE(InfoExtractor): ('mobile', 'Mozilla/5.0 (Linux; Android 4.4.2; Nexus 4 Build/KOT49H) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.114 Mobile Safari/537.36'), ) + user_agent = self._extract_cloudflare_session_ua(url) + if user_agent: + USER_AGENTS = ((None, user_agent),) + for kind, user_agent in USER_AGENTS: request = sanitized_Request(url) request.add_header('User-Agent', user_agent) webpage = self._download_webpage( - request, display_id, 'Downloading %s webpage' % kind) + request, display_id, + 'Downloading %s webpage' % kind if kind else 'Downloading webpage') playlist = self._parse_json( self._search_regex( diff --git a/youtube_dl/extractor/fusion.py b/youtube_dl/extractor/fusion.py new file mode 100644 index 000000000..b4ab4cbb7 --- /dev/null +++ b/youtube_dl/extractor/fusion.py @@ -0,0 +1,35 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from .ooyala import OoyalaIE + + +class FusionIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?fusion\.net/video/(?P<id>\d+)' + _TESTS = [{ + 'url': 'http://fusion.net/video/201781/u-s-and-panamanian-forces-work-together-to-stop-a-vessel-smuggling-drugs/', + 'info_dict': { + 'id': 'ZpcWNoMTE6x6uVIIWYpHh0qQDjxBuq5P', + 'ext': 'mp4', + 'title': 'U.S. and Panamanian forces work together to stop a vessel smuggling drugs', + 'description': 'md5:0cc84a9943c064c0f46b128b41b1b0d7', + 'duration': 140.0, + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': ['Ooyala'], + }, { + 'url': 'http://fusion.net/video/201781', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + + ooyala_code = self._search_regex( + r'data-video-id=(["\'])(?P<code>.+?)\1', + webpage, 'ooyala code', group='code') + + return OoyalaIE._build_url_result(ooyala_code) diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 4ffdd7515..621257c9f 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -1,19 +1,19 @@ from __future__ import unicode_literals import re -import json -from .common import InfoExtractor +from .once import OnceIE from ..compat import ( compat_urllib_parse_unquote, - compat_urlparse, ) from ..utils import ( unescapeHTML, + url_basename, + dict_get, ) -class GameSpotIE(InfoExtractor): +class GameSpotIE(OnceIE): _VALID_URL = r'https?://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?' _TESTS = [{ 'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/', @@ -39,29 +39,73 @@ class GameSpotIE(InfoExtractor): webpage = self._download_webpage(url, page_id) data_video_json = self._search_regex( r'data-video=["\'](.*?)["\']', webpage, 'data video') - data_video = json.loads(unescapeHTML(data_video_json)) + data_video = self._parse_json(unescapeHTML(data_video_json), page_id) streams = data_video['videoStreams'] + manifest_url = None formats = [] f4m_url = streams.get('f4m_stream') - if f4m_url is not None: - # Transform the manifest url to a link to the mp4 files - # they are used in mobile devices. - f4m_path = compat_urlparse.urlparse(f4m_url).path - QUALITIES_RE = r'((,\d+)+,?)' - qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',') - http_path = f4m_path[1:].split('/', 1)[1] - http_template = re.sub(QUALITIES_RE, r'%s', http_path) - http_template = http_template.replace('.csmil/manifest.f4m', '') - http_template = compat_urlparse.urljoin( - 'http://video.gamespotcdn.com/', http_template) - for q in qualities: - formats.append({ - 'url': http_template % q, - 'ext': 'mp4', - 'format_id': q, - }) - else: + if f4m_url: + manifest_url = f4m_url + formats.extend(self._extract_f4m_formats( + f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False)) + m3u8_url = streams.get('m3u8_stream') + if m3u8_url: + manifest_url = m3u8_url + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, page_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False) + formats.extend(m3u8_formats) + progressive_url = dict_get( + streams, ('progressive_hd', 'progressive_high', 'progressive_low')) + if progressive_url and manifest_url: + qualities_basename = self._search_regex( + '/([^/]+)\.csmil/', + manifest_url, 'qualities basename', default=None) + if qualities_basename: + QUALITIES_RE = r'((,\d+)+,?)' + qualities = self._search_regex( + QUALITIES_RE, qualities_basename, + 'qualities', default=None) + if qualities: + qualities = list(map(lambda q: int(q), qualities.strip(',').split(','))) + qualities.sort() + http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename) + http_url_basename = url_basename(progressive_url) + if m3u8_formats: + self._sort_formats(m3u8_formats) + m3u8_formats = list(filter( + lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + m3u8_formats)) + if len(qualities) == len(m3u8_formats): + for q, m3u8_format in zip(qualities, m3u8_formats): + f = m3u8_format.copy() + f.update({ + 'url': progressive_url.replace( + http_url_basename, http_template % q), + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) + else: + for q in qualities: + formats.append({ + 'url': progressive_url.replace( + http_url_basename, http_template % q), + 'ext': 'mp4', + 'format_id': 'http-%d' % q, + 'tbr': q, + }) + + onceux_json = self._search_regex( + r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None) + if onceux_json: + onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri') + if onceux_url: + formats.extend(self._extract_once_formats(re.sub( + r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url).replace('ads/vmap/', ''))) + + if not formats: for quality in ['sd', 'hd']: # It's actually a link to a flv file flv_url = streams.get('f4m_{0}'.format(quality)) @@ -71,6 +115,7 @@ class GameSpotIE(InfoExtractor): 'ext': 'flv', 'format_id': quality, }) + self._sort_formats(formats) return { 'id': data_video['guid'], diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py deleted file mode 100644 index 1e7948ab8..000000000 --- a/youtube_dl/extractor/gametrailers.py +++ /dev/null @@ -1,62 +0,0 @@ -from __future__ import unicode_literals - -from .common import InfoExtractor -from ..utils import ( - int_or_none, - parse_age_limit, - url_basename, -) - - -class GametrailersIE(InfoExtractor): - _VALID_URL = r'https?://www\.gametrailers\.com/videos/view/[^/]+/(?P<id>.+)' - - _TEST = { - 'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review', - 'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a', - 'info_dict': { - 'id': '2983958', - 'ext': 'mp4', - 'display_id': '116437-Just-Cause-3-Review', - 'title': 'Just Cause 3 - Review', - 'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?', - }, - } - - def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - title = self._html_search_regex( - r'<title>(.+?)\|', webpage, 'title').strip() - embed_url = self._proto_relative_url( - self._search_regex( - r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage, - 'embed url'), - scheme='http:') - video_id = url_basename(embed_url) - embed_page = self._download_webpage(embed_url, video_id) - embed_vars_json = self._search_regex( - r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page, - 'embed vars') - info = self._parse_json(embed_vars_json, video_id) - - formats = [] - for media in info['media']: - if media['mediaPurpose'] == 'play': - formats.append({ - 'url': media['uri'], - 'height': media['height'], - 'width:': media['width'], - }) - self._sort_formats(formats) - - return { - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'formats': formats, - 'thumbnail': info.get('thumbUri'), - 'description': self._og_search_description(webpage), - 'duration': int_or_none(info.get('videoLengthInSeconds')), - 'age_limit': parse_age_limit(info.get('audienceRating')), - } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 95d233259..4efdf146e 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -49,9 +49,12 @@ from .pornhub import PornHubIE from .xhamster import XHamsterEmbedIE from .tnaflix import TNAFlixNetworkEmbedIE from .vimeo import VimeoIE -from .dailymotion import DailymotionCloudIE +from .dailymotion import ( + DailymotionIE, + DailymotionCloudIE, +) from .onionstudios import OnionStudiosIE -from .snagfilms import SnagFilmsEmbedIE +from .viewlift import ViewLiftEmbedIE from .screenwavemedia import ScreenwaveMediaIE from .mtv import MTVServicesEmbeddedIE from .pladform import PladformIE @@ -61,6 +64,12 @@ from .jwplatform import JWPlatformIE from .digiteka import DigitekaIE from .instagram import InstagramIE from .liveleak import LiveLeakIE +from .threeqsdn import ThreeQSDNIE +from .theplatform import ThePlatformIE +from .vessel import VesselIE +from .kaltura import KalturaIE +from .eagleplatform import EaglePlatformIE +from .facebook import FacebookIE class GenericIE(InfoExtractor): @@ -237,6 +246,7 @@ class GenericIE(InfoExtractor): 'ext': 'mp4', 'title': 'car-20120827-manifest', 'formats': 'mincount:9', + 'upload_date': '20130904', }, 'params': { 'format': 'bestvideo', @@ -596,7 +606,11 @@ class GenericIE(InfoExtractor): 'id': 'k2mm4bCdJ6CQ2i7c8o2', 'ext': 'mp4', 'title': 'Le Zap de Spi0n n°216 - Zapping du Web', + 'description': 'md5:faf028e48a461b8b7fad38f1e104b119', 'uploader': 'Spi0n', + 'uploader_id': 'xgditw', + 'upload_date': '20140425', + 'timestamp': 1398441542, }, 'add_ie': ['Dailymotion'], }, @@ -619,13 +633,13 @@ class GenericIE(InfoExtractor): }, # MTVSercices embed { - 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too', - 'md5': '35727f82f58c76d996fc188f9755b0d5', + 'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html', + 'md5': 'ca1aef97695ef2c1d6973256a57e5252', 'info_dict': { - 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9', + 'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1', 'ext': 'mp4', - 'title': 'Review', - 'description': 'Mario\'s life in the fast lane has never looked so good.', + 'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored', + 'description': 'Two valets share their love for movie star Liam Neesons.', }, }, # YouTube embed via <data-embed-url=""> @@ -711,15 +725,18 @@ class GenericIE(InfoExtractor): }, # Wistia embed { - 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson', - 'md5': '8788b683c777a5cf25621eaf286d0c23', + 'url': 'http://study.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson', + 'md5': '1953f3a698ab51cfc948ed3992a0b7ff', 'info_dict': { - 'id': '1cfaf6b7ea', + 'id': '6e2wtrbdaf', 'ext': 'mov', - 'title': 'md5:51364a8d3d009997ba99656004b5e20d', - 'duration': 643.0, - 'filesize': 182808282, - 'uploader': 'education-portal.com', + 'title': 'paywall_north-american-exploration-failed-colonies-of-spain-france-england', + 'description': 'a Paywall Videos video from Remilon', + 'duration': 644.072, + 'uploader': 'study.com', + 'timestamp': 1459678540, + 'upload_date': '20160403', + 'filesize': 24687186, }, }, { @@ -728,10 +745,29 @@ class GenericIE(InfoExtractor): 'info_dict': { 'id': 'uxjb0lwrcz', 'ext': 'mp4', - 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks', + 'title': 'Conversation about Hexagonal Rails Part 1', + 'description': 'a Martin Fowler video from ThoughtWorks', 'duration': 1715.0, 'uploader': 'thoughtworks.wistia.com', + 'timestamp': 1401832161, + 'upload_date': '20140603', + }, + }, + # Wistia standard embed (async) + { + 'url': 'https://www.getdrip.com/university/brennan-dunn-drip-workshop/', + 'info_dict': { + 'id': '807fafadvk', + 'ext': 'mp4', + 'title': 'Drip Brennan Dunn Workshop', + 'description': 'a JV Webinars video from getdrip-1', + 'duration': 4986.95, + 'timestamp': 1463607249, + 'upload_date': '20160518', }, + 'params': { + 'skip_download': True, + } }, # Soundcloud embed { @@ -755,6 +791,19 @@ class GenericIE(InfoExtractor): 'title': 'Rosetta #CometLanding webcast HL 10', } }, + # Another Livestream embed, without 'new.' in URL + { + 'url': 'https://www.freespeech.org/', + 'info_dict': { + 'id': '123537347', + 'ext': 'mp4', + 'title': 're:^FSTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + }, + 'params': { + # Live stream + 'skip_download': True, + }, + }, # LazyYT { 'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986', @@ -839,18 +888,6 @@ class GenericIE(InfoExtractor): 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !', } }, - # Kaltura embed - { - 'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15', - 'info_dict': { - 'id': '1_eergr3h1', - 'ext': 'mp4', - 'upload_date': '20150226', - 'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com', - 'timestamp': int, - 'title': 'John Carlson Postgame 2/25/15', - }, - }, # Kaltura embed (different embed code) { 'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014', @@ -876,9 +913,41 @@ class GenericIE(InfoExtractor): 'uploader_id': 'echojecka', }, }, + # Kaltura embed with single quotes + { + 'url': 'http://fod.infobase.com/p_ViewPlaylist.aspx?AssignmentID=NUN8ZY', + 'info_dict': { + 'id': '0_izeg5utt', + 'ext': 'mp4', + 'title': '35871', + 'timestamp': 1355743100, + 'upload_date': '20121217', + 'uploader_id': 'batchUser', + }, + 'add_ie': ['Kaltura'], + }, + { + # Kaltura embedded via quoted entry_id + 'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures', + 'info_dict': { + 'id': '0_utuok90b', + 'ext': 'mp4', + 'title': '06_matthew_brender_raj_dutt', + 'timestamp': 1466638791, + 'upload_date': '20160622', + }, + 'add_ie': ['Kaltura'], + 'expected_warnings': [ + 'Could not send HEAD request' + ], + 'params': { + 'skip_download': True, + } + }, # Eagle.Platform embed (generic URL) { 'url': 'http://lenta.ru/news/2015/03/06/navalny/', + # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used 'info_dict': { 'id': '227304', 'ext': 'mp4', @@ -893,6 +962,7 @@ class GenericIE(InfoExtractor): # ClipYou (Eagle.Platform) embed (custom URL) { 'url': 'http://muz-tv.ru/play/7129/', + # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used 'info_dict': { 'id': '12820', 'ext': 'mp4', @@ -981,18 +1051,36 @@ class GenericIE(InfoExtractor): 'ext': 'flv', 'title': "PFT Live: New leader in the 'new-look' defense", 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e', + 'uploader': 'NBCU-SPORTS', + 'upload_date': '20140107', + 'timestamp': 1389118457, + }, + }, + # NBC News embed + { + 'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html', + 'md5': '1aa589c675898ae6d37a17913cf68d66', + 'info_dict': { + 'id': '701714499682', + 'ext': 'mp4', + 'title': 'PREVIEW: On Assignment: David Letterman', + 'description': 'A preview of Tom Brokaw\'s interview with David Letterman as part of the On Assignment series powered by Dateline. Airs Sunday June 12 at 7/6c.', }, }, # UDN embed { - 'url': 'http://www.udn.com/news/story/7314/822787', + 'url': 'https://video.udn.com/news/300346', 'md5': 'fd2060e988c326991037b9aff9df21a6', 'info_dict': { 'id': '300346', 'ext': 'mp4', 'title': '中一中男師變性 全校師生力挺', 'thumbnail': 're:^https?://.*\.jpg$', - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, # Ooyala embed { @@ -1009,20 +1097,6 @@ class GenericIE(InfoExtractor): 'skip_download': True, } }, - # Contains a SMIL manifest - { - 'url': 'http://www.telewebion.com/fa/1263668/%D9%82%D8%B1%D8%B9%D9%87%E2%80%8C%DA%A9%D8%B4%DB%8C-%D9%84%DB%8C%DA%AF-%D9%82%D9%87%D8%B1%D9%85%D8%A7%D9%86%D8%A7%D9%86-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7/%2B-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84.html', - 'info_dict': { - 'id': 'file', - 'ext': 'flv', - 'title': '+ Football: Lottery Champions League Europe', - 'uploader': 'www.telewebion.com', - }, - 'params': { - # rtmpe downloads - 'skip_download': True, - } - }, # Brightcove URL in single quotes { 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/', @@ -1033,17 +1107,25 @@ class GenericIE(InfoExtractor): 'title': 'SN Presents: Russell Martin, World Citizen', 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.', 'uploader': 'Rogers Sportsnet', + 'uploader_id': '1704050871', + 'upload_date': '20150525', + 'timestamp': 1432570283, }, }, # Dailymotion Cloud video { 'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910', - 'md5': '49444254273501a64675a7e68c502681', + 'md5': 'dcaf23ad0c67a256f4278bce6e0bae38', 'info_dict': { - 'id': '5585de919473990de4bee11b', + 'id': 'x2uy8t3', 'ext': 'mp4', - 'title': 'Le débat', + 'title': 'Sauvons les abeilles ! - Le débat', + 'description': 'md5:d9082128b1c5277987825d684939ca26', 'thumbnail': 're:^https?://.*\.jpe?g$', + 'timestamp': 1434970506, + 'upload_date': '20150622', + 'uploader': 'Public Sénat', + 'uploader_id': 'xa9gza', } }, # OnionStudios embed @@ -1124,6 +1206,9 @@ class GenericIE(InfoExtractor): 'title': 'The Cardinal Pell Interview', 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ', 'uploader': 'GlobeCast Australia - GlobeStream', + 'uploader_id': '2733773828001', + 'upload_date': '20160304', + 'timestamp': 1457083087, }, 'params': { # m3u8 downloads @@ -1154,6 +1239,97 @@ class GenericIE(InfoExtractor): 'uploader': 'Lake8737', } }, + # Duplicated embedded video URLs + { + 'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443', + 'info_dict': { + 'id': '149298443_480_16c25b74_2', + 'ext': 'mp4', + 'title': 'vs. Blue Orange Spring Game', + 'uploader': 'www.hudl.com', + }, + }, + # twitter:player embed + { + 'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/', + 'md5': 'a3e0df96369831de324f0778e126653c', + 'info_dict': { + 'id': '4909620399001', + 'ext': 'mp4', + 'title': 'What Do Black Holes Sound Like?', + 'description': 'what do black holes sound like', + 'upload_date': '20160524', + 'uploader_id': '29913724001', + 'timestamp': 1464107587, + 'uploader': 'TheAtlantic', + }, + 'add_ie': ['BrightcoveLegacy'], + }, + # Facebook <iframe> embed + { + 'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html', + 'md5': 'fbcde74f534176ecb015849146dd3aee', + 'info_dict': { + 'id': '599637780109885', + 'ext': 'mp4', + 'title': 'Facebook video #599637780109885', + }, + }, + # Facebook API embed + { + 'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/', + 'md5': 'a47372ee61b39a7b90287094d447d94e', + 'info_dict': { + 'id': '10153467542406923', + 'ext': 'mp4', + 'title': 'Facebook video #10153467542406923', + }, + }, + # Wordpress "YouTube Video Importer" plugin + { + 'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/', + 'md5': 'd16797741b560b485194eddda8121b48', + 'info_dict': { + 'id': 'HNTXWDXV9Is', + 'ext': 'mp4', + 'title': 'Blue Devils Drumline Stanford lot 2016', + 'upload_date': '20160627', + 'uploader_id': 'GENOCIDE8GENERAL10', + 'uploader': 'cylus cyrus', + }, + }, + { + # video stored on custom kaltura server + 'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv', + 'md5': '537617d06e64dfed891fa1593c4b30cc', + 'info_dict': { + 'id': '0_1iotm5bh', + 'ext': 'mp4', + 'title': 'Elecciones británicas: 5 lecciones para Rajoy', + 'description': 'md5:435a89d68b9760b92ce67ed227055f16', + 'uploader_id': 'videos.expansion@el-mundo.net', + 'upload_date': '20150429', + 'timestamp': 1430303472, + }, + 'add_ie': ['Kaltura'], + }, + # { + # # TODO: find another test + # # http://schema.org/VideoObject + # 'url': 'https://flipagram.com/f/nyvTSJMKId', + # 'md5': '888dcf08b7ea671381f00fab74692755', + # 'info_dict': { + # 'id': 'nyvTSJMKId', + # 'ext': 'mp4', + # 'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction', + # 'description': '#love for cats.', + # 'timestamp': 1461244995, + # 'upload_date': '20160421', + # }, + # 'params': { + # 'force_generic_extractor': True, + # }, + # } ] def report_following_redirect(self, new_url): @@ -1408,7 +1584,8 @@ class GenericIE(InfoExtractor): # Site Name | Video Title # Video Title - Tagline | Site Name # and so on and so forth; it's just not practical - video_title = self._html_search_regex( + video_title = self._og_search_title( + webpage, default=None) or self._html_search_regex( r'(?s)<title>(.*?)</title>', webpage, 'video title', default='video') @@ -1426,6 +1603,9 @@ class GenericIE(InfoExtractor): video_uploader = self._search_regex( r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') + video_description = self._og_search_description(webpage, default=None) + video_thumbnail = self._og_search_thumbnail(webpage, default=None) + # Helper method def _playlist_from_matches(matches, getter=None, ie=None): urlrs = orderedSet( @@ -1456,6 +1636,16 @@ class GenericIE(InfoExtractor): if bc_urls: return _playlist_from_matches(bc_urls, ie='BrightcoveNew') + # Look for ThePlatform embeds + tp_urls = ThePlatformIE._extract_urls(webpage) + if tp_urls: + return _playlist_from_matches(tp_urls, ie='ThePlatform') + + # Look for Vessel embeds + vessel_urls = VesselIE._extract_urls(webpage) + if vessel_urls: + return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key()) + # Look for embedded rtl.nl player matches = re.findall( r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"', @@ -1496,12 +1686,16 @@ class GenericIE(InfoExtractor): if matches: return _playlist_from_matches(matches, lambda m: unescapeHTML(m)) - # Look for embedded Dailymotion player - matches = re.findall( - r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage) + # Look for Wordpress "YouTube Video Importer" plugin + matches = re.findall(r'''(?x)<div[^>]+ + class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+ + data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage) if matches: - return _playlist_from_matches( - matches, lambda m: unescapeHTML(m[1])) + return _playlist_from_matches(matches, lambda m: m[-1]) + + matches = DailymotionIE._extract_urls(webpage) + if matches: + return _playlist_from_matches(matches) # Look for embedded Dailymotion playlist player (#3822) m = re.search( @@ -1524,21 +1718,26 @@ class GenericIE(InfoExtractor): 'url': embed_url, 'ie_key': 'Wistia', 'uploader': video_uploader, - 'title': video_title, - 'id': video_id, } match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage) if match: return { '_type': 'url_transparent', - 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')), + 'url': 'wistia:%s' % match.group('id'), 'ie_key': 'Wistia', 'uploader': video_uploader, - 'title': video_title, - 'id': match.group('id') } + match = re.search( + r'''(?sx) + <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*? + <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2 + ''', webpage) + if match: + return self.url_result(self._proto_relative_url( + 'wistia:%s' % match.group('id')), 'Wistia') + # Look for SVT player svt_url = SVTIE._extract_url(webpage) if svt_url: @@ -1633,10 +1832,9 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) # Look for embedded Facebook player - mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage) - if mobj is not None: - return self.url_result(mobj.group('url'), 'Facebook') + facebook_url = FacebookIE._extract_url(webpage) + if facebook_url is not None: + return self.url_result(facebook_url, 'Facebook') # Look for embedded VK player mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage) @@ -1758,14 +1956,6 @@ class GenericIE(InfoExtractor): url = unescapeHTML(mobj.group('url')) return self.url_result(url) - # Look for embedded vulture.com player - mobj = re.search( - r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"', - webpage) - if mobj is not None: - url = unescapeHTML(mobj.group('url')) - return self.url_result(url, ie='Vulture') - # Look for embedded mtvservices player mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage) if mtvservices_url: @@ -1814,7 +2004,7 @@ class GenericIE(InfoExtractor): return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast') mobj = re.search( - r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"', + r'<iframe[^>]+src="(?P<url>https?://(?:new\.)?livestream\.com/[^"]+/player[^"]+)"', webpage) if mobj is not None: return self.url_result(mobj.group('url'), 'Livestream') @@ -1826,18 +2016,14 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url'), 'Zapiks') # Look for Kaltura embeds - mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or - re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage)) - if mobj is not None: - return self.url_result(smuggle_url( - 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), - {'source_url': url}), 'Kaltura') + kaltura_url = KalturaIE._extract_url(webpage) + if kaltura_url: + return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key()) # Look for Eagle.Platform embeds - mobj = re.search( - r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage) - if mobj is not None: - return self.url_result(mobj.group('url'), 'EaglePlatform') + eagleplatform_url = EaglePlatformIE._extract_url(webpage) + if eagleplatform_url: + return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key()) # Look for ClipYou (uses Eagle.Platform) embeds mobj = re.search( @@ -1878,6 +2064,12 @@ class GenericIE(InfoExtractor): if nbc_sports_url: return self.url_result(nbc_sports_url, 'NBCSportsVPlayer') + # Look for NBC News embeds + nbc_news_embed_url = re.search( + r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage) + if nbc_news_embed_url: + return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews') + # Look for Google Drive embeds google_drive_url = GoogleDriveIE._extract_url(webpage) if google_drive_url: @@ -1905,10 +2097,10 @@ class GenericIE(InfoExtractor): if onionstudios_url: return self.url_result(onionstudios_url) - # Look for SnagFilms embeds - snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage) - if snagfilms_url: - return self.url_result(snagfilms_url) + # Look for ViewLift embeds + viewlift_url = ViewLiftEmbedIE._extract_url(webpage) + if viewlift_url: + return self.url_result(viewlift_url) # Look for JWPlatform embeds jwplatform_url = JWPlatformIE._extract_url(webpage) @@ -1964,6 +2156,37 @@ class GenericIE(InfoExtractor): if liveleak_url: return self.url_result(liveleak_url, 'LiveLeak') + # Look for 3Q SDN embeds + threeqsdn_url = ThreeQSDNIE._extract_url(webpage) + if threeqsdn_url: + return { + '_type': 'url_transparent', + 'ie_key': ThreeQSDNIE.ie_key(), + 'url': self._proto_relative_url(threeqsdn_url), + 'title': video_title, + 'description': video_description, + 'thumbnail': video_thumbnail, + 'uploader': video_uploader, + } + + # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser + embed_url = self._html_search_meta('twitter:player', webpage, default=None) + if embed_url: + return self.url_result(embed_url) + + # Looking for http://schema.org/VideoObject + json_ld = self._search_json_ld( + webpage, video_id, default=None, expected_type='VideoObject') + if json_ld and json_ld.get('url'): + info_dict.update({ + 'title': video_title or info_dict['title'], + 'description': video_description, + 'thumbnail': video_thumbnail, + 'age_limit': age_limit + }) + info_dict.update(json_ld) + return info_dict + def check_video(vurl): if YoutubeIE.suitable(vurl): return True @@ -2044,7 +2267,8 @@ class GenericIE(InfoExtractor): raise UnsupportedError(url) entries = [] - for video_url in found: + for video_url in orderedSet(found): + video_url = unescapeHTML(video_url) video_url = video_url.replace('\\/', '/') video_url = compat_urlparse.urljoin(url, video_url) video_id = compat_urllib_parse_unquote(os.path.basename(video_url)) diff --git a/youtube_dl/extractor/godtv.py b/youtube_dl/extractor/godtv.py new file mode 100644 index 000000000..c5d3b4e6a --- /dev/null +++ b/youtube_dl/extractor/godtv.py @@ -0,0 +1,66 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from .ooyala import OoyalaIE +from ..utils import js_to_json + + +class GodTVIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?god\.tv(?:/[^/]+)*/(?P<id>[^/?#&]+)' + _TESTS = [{ + 'url': 'http://god.tv/jesus-image/video/jesus-conference-2016/randy-needham', + 'info_dict': { + 'id': 'lpd3g2MzE6D1g8zFAKz8AGpxWcpu6o_3', + 'ext': 'mp4', + 'title': 'Randy Needham', + 'duration': 3615.08, + }, + 'params': { + 'skip_download': True, + } + }, { + 'url': 'http://god.tv/playlist/bible-study', + 'info_dict': { + 'id': 'bible-study', + }, + 'playlist_mincount': 37, + }, { + 'url': 'http://god.tv/node/15097', + 'only_matching': True, + }, { + 'url': 'http://god.tv/live/africa', + 'only_matching': True, + }, { + 'url': 'http://god.tv/liveevents', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + settings = self._parse_json( + self._search_regex( + r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);', + webpage, 'settings', default='{}'), + display_id, transform_source=js_to_json, fatal=False) + + ooyala_id = None + + if settings: + playlist = settings.get('playlist') + if playlist and isinstance(playlist, list): + entries = [ + OoyalaIE._build_url_result(video['content_id']) + for video in playlist if video.get('content_id')] + if entries: + return self.playlist_result(entries, display_id) + ooyala_id = settings.get('ooyala', {}).get('content_id') + + if not ooyala_id: + ooyala_id = self._search_regex( + r'["\']content_id["\']\s*:\s*(["\'])(?P<id>[\w-]+)\1', + webpage, 'ooyala id', group='id') + + return OoyalaIE._build_url_result(ooyala_id) diff --git a/youtube_dl/extractor/groupon.py b/youtube_dl/extractor/groupon.py index f6b69662b..a6da90931 100644 --- a/youtube_dl/extractor/groupon.py +++ b/youtube_dl/extractor/groupon.py @@ -4,7 +4,7 @@ from .common import InfoExtractor class GrouponIE(InfoExtractor): - _VALID_URL = r'https?://www\.groupon\.com/deals/(?P<id>[^?#]+)' + _VALID_URL = r'https?://(?:www\.)?groupon\.com/deals/(?P<id>[^/?#&]+)' _TEST = { 'url': 'https://www.groupon.com/deals/bikram-yoga-huntington-beach-2#ooid=tubGNycTo_9Uxg82uESj4i61EYX8nyuf', @@ -14,17 +14,27 @@ class GrouponIE(InfoExtractor): 'description': 'Studio kept at 105 degrees and 40% humidity with anti-microbial and anti-slip Flotex flooring; certified instructors', }, 'playlist': [{ + 'md5': '42428ce8a00585f9bc36e49226eae7a1', 'info_dict': { - 'id': 'tubGNycTo_9Uxg82uESj4i61EYX8nyuf', - 'ext': 'flv', - 'title': 'Bikram Yoga Huntington Beach | Orange County', + 'id': 'fk6OhWpXgIQ', + 'ext': 'mp4', + 'title': 'Bikram Yoga Huntington Beach | Orange County !tubGNycTo@9Uxg82uESj4i61EYX8nyuf', 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e', - 'duration': 44.961, + 'duration': 45, + 'upload_date': '20160405', + 'uploader_id': 'groupon', + 'uploader': 'Groupon', }, + 'add_ie': ['Youtube'], }], 'params': { - 'skip_download': 'HDS', - } + 'skip_download': True, + }, + } + + _PROVIDERS = { + 'ooyala': ('ooyala:%s', 'Ooyala'), + 'youtube': ('%s', 'Youtube'), } def _real_extract(self, url): @@ -36,12 +46,17 @@ class GrouponIE(InfoExtractor): videos = payload['carousel'].get('dealVideos', []) entries = [] for v in videos: - if v.get('provider') != 'OOYALA': + provider = v.get('provider') + video_id = v.get('media') or v.get('id') or v.get('baseURL') + if not provider or not video_id: + continue + url_pattern, ie_key = self._PROVIDERS.get(provider.lower()) + if not url_pattern: self.report_warning( '%s: Unsupported video provider %s, skipping video' % - (playlist_id, v.get('provider'))) + (playlist_id, provider)) continue - entries.append(self.url_result('ooyala:%s' % v['media'])) + entries.append(self.url_result(url_pattern % video_id, ie_key)) return { '_type': 'playlist', diff --git a/youtube_dl/extractor/hearthisat.py b/youtube_dl/extractor/hearthisat.py index 7d8698655..256453882 100644 --- a/youtube_dl/extractor/hearthisat.py +++ b/youtube_dl/extractor/hearthisat.py @@ -7,6 +7,7 @@ from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( HEADRequest, + KNOWN_EXTENSIONS, sanitized_Request, str_to_int, urlencode_postdata, @@ -17,7 +18,7 @@ from ..utils import ( class HearThisAtIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?hearthis\.at/(?P<artist>[^/]+)/(?P<title>[A-Za-z0-9\-]+)/?$' _PLAYLIST_URL = 'https://hearthis.at/playlist.php' - _TEST = { + _TESTS = [{ 'url': 'https://hearthis.at/moofi/dr-kreep', 'md5': 'ab6ec33c8fed6556029337c7885eb4e0', 'info_dict': { @@ -26,7 +27,7 @@ class HearThisAtIE(InfoExtractor): 'title': 'Moofi - Dr. Kreep', 'thumbnail': 're:^https?://.*\.jpg$', 'timestamp': 1421564134, - 'description': 'Creepy Patch. Mutable Instruments Braids Vowel + Formant Mode.', + 'description': 'Listen to Dr. Kreep by Moofi on hearthis.at - Modular, Eurorack, Mutable Intruments Braids, Valhalla-DSP', 'upload_date': '20150118', 'comment_count': int, 'view_count': int, @@ -34,7 +35,25 @@ class HearThisAtIE(InfoExtractor): 'duration': 71, 'categories': ['Experimental'], } - } + }, { + # 'download' link redirects to the original webpage + 'url': 'https://hearthis.at/twitchsf/dj-jim-hopkins-totally-bitchin-80s-dance-mix/', + 'md5': '5980ceb7c461605d30f1f039df160c6e', + 'info_dict': { + 'id': '811296', + 'ext': 'mp3', + 'title': 'TwitchSF - DJ Jim Hopkins - Totally Bitchin\' 80\'s Dance Mix!', + 'description': 'Listen to DJ Jim Hopkins - Totally Bitchin\' 80\'s Dance Mix! by TwitchSF on hearthis.at - Dance', + 'upload_date': '20160328', + 'timestamp': 1459186146, + 'thumbnail': 're:^https?://.*\.jpg$', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'duration': 4360, + 'categories': ['Dance'], + }, + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) @@ -90,13 +109,14 @@ class HearThisAtIE(InfoExtractor): ext_handle = self._request_webpage( ext_req, display_id, note='Determining extension') ext = urlhandle_detect_ext(ext_handle) - formats.append({ - 'format_id': 'download', - 'vcodec': 'none', - 'ext': ext, - 'url': download_url, - 'preference': 2, # Usually better quality - }) + if ext in KNOWN_EXTENSIONS: + formats.append({ + 'format_id': 'download', + 'vcodec': 'none', + 'ext': ext, + 'url': download_url, + 'preference': 2, # Usually better quality + }) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/howcast.py b/youtube_dl/extractor/howcast.py index e8f51e545..7e36b85ad 100644 --- a/youtube_dl/extractor/howcast.py +++ b/youtube_dl/extractor/howcast.py @@ -8,7 +8,7 @@ class HowcastIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?howcast\.com/videos/(?P<id>\d+)' _TEST = { 'url': 'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly', - 'md5': '8b743df908c42f60cf6496586c7f12c3', + 'md5': '7d45932269a288149483144f01b99789', 'info_dict': { 'id': '390161', 'ext': 'mp4', @@ -19,9 +19,9 @@ class HowcastIE(InfoExtractor): 'duration': 56.823, }, 'params': { - # m3u8 download 'skip_download': True, }, + 'add_ie': ['Ooyala'], } def _real_extract(self, url): diff --git a/youtube_dl/extractor/hrti.py b/youtube_dl/extractor/hrti.py new file mode 100644 index 000000000..656ce6d05 --- /dev/null +++ b/youtube_dl/extractor/hrti.py @@ -0,0 +1,202 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import json +import re + +from .common import InfoExtractor +from ..compat import compat_HTTPError +from ..utils import ( + clean_html, + ExtractorError, + int_or_none, + parse_age_limit, + sanitized_Request, + try_get, +) + + +class HRTiBaseIE(InfoExtractor): + """ + Base Information Extractor for Croatian Radiotelevision + video on demand site https://hrti.hrt.hr + Reverse engineered from the JavaScript app in app.min.js + """ + _NETRC_MACHINE = 'hrti' + + _APP_LANGUAGE = 'hr' + _APP_VERSION = '1.1' + _APP_PUBLICATION_ID = 'all_in_one' + _API_URL = 'http://clientapi.hrt.hr/client_api.php/config/identify/format/json' + + def _initialize_api(self): + init_data = { + 'application_publication_id': self._APP_PUBLICATION_ID + } + + uuid = self._download_json( + self._API_URL, None, note='Downloading uuid', + errnote='Unable to download uuid', + data=json.dumps(init_data).encode('utf-8'))['uuid'] + + app_data = { + 'uuid': uuid, + 'application_publication_id': self._APP_PUBLICATION_ID, + 'application_version': self._APP_VERSION + } + + req = sanitized_Request(self._API_URL, data=json.dumps(app_data).encode('utf-8')) + req.get_method = lambda: 'PUT' + + resources = self._download_json( + req, None, note='Downloading session information', + errnote='Unable to download session information') + + self._session_id = resources['session_id'] + + modules = resources['modules'] + + self._search_url = modules['vod_catalog']['resources']['search']['uri'].format( + language=self._APP_LANGUAGE, + application_id=self._APP_PUBLICATION_ID) + + self._login_url = (modules['user']['resources']['login']['uri'] + + '/format/json').format(session_id=self._session_id) + + self._logout_url = modules['user']['resources']['logout']['uri'] + + def _login(self): + (username, password) = self._get_login_info() + # TODO: figure out authentication with cookies + if username is None or password is None: + self.raise_login_required() + + auth_data = { + 'username': username, + 'password': password, + } + + try: + auth_info = self._download_json( + self._login_url, None, note='Logging in', errnote='Unable to log in', + data=json.dumps(auth_data).encode('utf-8')) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 406: + auth_info = self._parse_json(e.cause.read().encode('utf-8'), None) + else: + raise + + error_message = auth_info.get('error', {}).get('message') + if error_message: + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, error_message), + expected=True) + + self._token = auth_info['secure_streaming_token'] + + def _real_initialize(self): + self._initialize_api() + self._login() + + +class HRTiIE(HRTiBaseIE): + _VALID_URL = r'''(?x) + (?: + hrti:(?P<short_id>[0-9]+)| + https?:// + hrti\.hrt\.hr/\#/video/show/(?P<id>[0-9]+)/(?P<display_id>[^/]+)? + ) + ''' + _TESTS = [{ + 'url': 'https://hrti.hrt.hr/#/video/show/2181385/republika-dokumentarna-serija-16-hd', + 'info_dict': { + 'id': '2181385', + 'display_id': 'republika-dokumentarna-serija-16-hd', + 'ext': 'mp4', + 'title': 'REPUBLIKA, dokumentarna serija (1/6) (HD)', + 'description': 'md5:48af85f620e8e0e1df4096270568544f', + 'duration': 2922, + 'view_count': int, + 'average_rating': int, + 'episode_number': int, + 'season_number': int, + 'age_limit': 12, + }, + 'skip': 'Requires account credentials', + }, { + 'url': 'https://hrti.hrt.hr/#/video/show/2181385/', + 'only_matching': True, + }, { + 'url': 'hrti:2181385', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('short_id') or mobj.group('id') + display_id = mobj.group('display_id') or video_id + + video = self._download_json( + '%s/video_id/%s/format/json' % (self._search_url, video_id), + display_id, 'Downloading video metadata JSON')['video'][0] + + title_info = video['title'] + title = title_info['title_long'] + + movie = video['video_assets']['movie'][0] + m3u8_url = movie['url'].format(TOKEN=self._token) + formats = self._extract_m3u8_formats( + m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls') + self._sort_formats(formats) + + description = clean_html(title_info.get('summary_long')) + age_limit = parse_age_limit(video.get('parental_control', {}).get('rating')) + view_count = int_or_none(video.get('views')) + average_rating = int_or_none(video.get('user_rating')) + duration = int_or_none(movie.get('duration')) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'duration': duration, + 'view_count': view_count, + 'average_rating': average_rating, + 'age_limit': age_limit, + 'formats': formats, + } + + +class HRTiPlaylistIE(HRTiBaseIE): + _VALID_URL = r'https?://hrti.hrt.hr/#/video/list/category/(?P<id>[0-9]+)/(?P<display_id>[^/]+)?' + _TESTS = [{ + 'url': 'https://hrti.hrt.hr/#/video/list/category/212/ekumena', + 'info_dict': { + 'id': '212', + 'title': 'ekumena', + }, + 'playlist_mincount': 8, + 'skip': 'Requires account credentials', + }, { + 'url': 'https://hrti.hrt.hr/#/video/list/category/212/', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + category_id = mobj.group('id') + display_id = mobj.group('display_id') or category_id + + response = self._download_json( + '%s/category_id/%s/format/json' % (self._search_url, category_id), + display_id, 'Downloading video metadata JSON') + + video_ids = try_get( + response, lambda x: x['video_listings'][0]['alternatives'][0]['list'], + list) or [video['id'] for video in response.get('videos', []) if video.get('id')] + + entries = [self.url_result('hrti:%s' % video_id) for video_id in video_ids] + + return self.playlist_result(entries, category_id, display_id) diff --git a/youtube_dl/extractor/imdb.py b/youtube_dl/extractor/imdb.py index 8bed8ccd0..0acce9f4c 100644 --- a/youtube_dl/extractor/imdb.py +++ b/youtube_dl/extractor/imdb.py @@ -1,10 +1,10 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..utils import ( + mimetype2ext, qualities, ) @@ -12,9 +12,9 @@ from ..utils import ( class ImdbIE(InfoExtractor): IE_NAME = 'imdb' IE_DESC = 'Internet Movie Database trailers' - _VALID_URL = r'https?://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)' + _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video/[^/]+/|title/tt\d+.*?#lb-)vi(?P<id>\d+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.imdb.com/video/imdb/vi2524815897', 'info_dict': { 'id': '2524815897', @@ -22,7 +22,16 @@ class ImdbIE(InfoExtractor): 'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb', 'description': 'md5:9061c2219254e5d14e03c25c98e96a81', } - } + }, { + 'url': 'http://www.imdb.com/video/_/vi2524815897', + 'only_matching': True, + }, { + 'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897', + 'only_matching': True, + }, { + 'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -48,13 +57,27 @@ class ImdbIE(InfoExtractor): json_data = self._search_regex( r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>', format_page, 'json data', flags=re.DOTALL) - info = json.loads(json_data) - format_info = info['videoPlayerObject']['video'] - f_id = format_info['ffname'] + info = self._parse_json(json_data, video_id, fatal=False) + if not info: + continue + format_info = info.get('videoPlayerObject', {}).get('video', {}) + if not format_info: + continue + video_info_list = format_info.get('videoInfoList') + if not video_info_list or not isinstance(video_info_list, list): + continue + video_info = video_info_list[0] + if not video_info or not isinstance(video_info, dict): + continue + video_url = video_info.get('videoUrl') + if not video_url: + continue + format_id = format_info.get('ffname') formats.append({ - 'format_id': f_id, - 'url': format_info['videoInfoList'][0]['videoUrl'], - 'quality': quality(f_id), + 'format_id': format_id, + 'url': video_url, + 'ext': mimetype2ext(video_info.get('videoMimeType')), + 'quality': quality(format_id), }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/indavideo.py b/youtube_dl/extractor/indavideo.py index 9622f198a..c6f080484 100644 --- a/youtube_dl/extractor/indavideo.py +++ b/youtube_dl/extractor/indavideo.py @@ -60,7 +60,8 @@ class IndavideoEmbedIE(InfoExtractor): formats = [{ 'url': video_url, - 'height': self._search_regex(r'\.(\d{3,4})\.mp4$', video_url, 'height', default=None), + 'height': int_or_none(self._search_regex( + r'\.(\d{3,4})\.mp4(?:\?|$)', video_url, 'height', default=None)), } for video_url in video_urls] self._sort_formats(formats) diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index 3cbe77ad8..fc0197ae1 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -8,6 +8,7 @@ from ..utils import ( int_or_none, limit_length, lowercase_escape, + try_get, ) @@ -19,10 +20,16 @@ class InstagramIE(InfoExtractor): 'info_dict': { 'id': 'aye83DjauH', 'ext': 'mp4', - 'uploader_id': 'naomipq', 'title': 'Video by naomipq', 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', - } + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1371748545, + 'upload_date': '20130620', + 'uploader_id': 'naomipq', + 'uploader': 'Naomi Leonor Phan-Quang', + 'like_count': int, + 'comment_count': int, + }, }, { # missing description 'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears', @@ -31,6 +38,13 @@ class InstagramIE(InfoExtractor): 'ext': 'mp4', 'uploader_id': 'britneyspears', 'title': 'Video by britneyspears', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1453760977, + 'upload_date': '20160125', + 'uploader_id': 'britneyspears', + 'uploader': 'Britney Spears', + 'like_count': int, + 'comment_count': int, }, 'params': { 'skip_download': True, @@ -67,21 +81,57 @@ class InstagramIE(InfoExtractor): url = mobj.group('url') webpage = self._download_webpage(url, video_id) - uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"', - webpage, 'uploader id', fatal=False) - desc = self._search_regex( - r'"caption":"(.+?)"', webpage, 'description', default=None) - if desc is not None: - desc = lowercase_escape(desc) + + (video_url, description, thumbnail, timestamp, uploader, + uploader_id, like_count, comment_count) = [None] * 8 + + shared_data = self._parse_json( + self._search_regex( + r'window\._sharedData\s*=\s*({.+?});', + webpage, 'shared data', default='{}'), + video_id, fatal=False) + if shared_data: + media = try_get( + shared_data, lambda x: x['entry_data']['PostPage'][0]['media'], dict) + if media: + video_url = media.get('video_url') + description = media.get('caption') + thumbnail = media.get('display_src') + timestamp = int_or_none(media.get('date')) + uploader = media.get('owner', {}).get('full_name') + uploader_id = media.get('owner', {}).get('username') + like_count = int_or_none(media.get('likes', {}).get('count')) + comment_count = int_or_none(media.get('comments', {}).get('count')) + + if not video_url: + video_url = self._og_search_video_url(webpage, secure=False) + + if not uploader_id: + uploader_id = self._search_regex( + r'"owner"\s*:\s*{\s*"username"\s*:\s*"(.+?)"', + webpage, 'uploader id', fatal=False) + + if not description: + description = self._search_regex( + r'"caption"\s*:\s*"(.+?)"', webpage, 'description', default=None) + if description is not None: + description = lowercase_escape(description) + + if not thumbnail: + thumbnail = self._og_search_thumbnail(webpage) return { 'id': video_id, - 'url': self._og_search_video_url(webpage, secure=False), + 'url': video_url, 'ext': 'mp4', 'title': 'Video by %s' % uploader_id, - 'thumbnail': self._og_search_thumbnail(webpage), + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, 'uploader_id': uploader_id, - 'description': desc, + 'uploader': uploader, + 'like_count': like_count, + 'comment_count': comment_count, } diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index ffb8008ce..01c7b3042 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -3,28 +3,22 @@ from __future__ import unicode_literals import hashlib import itertools -import math -import os -import random import re import time -import uuid from .common import InfoExtractor from ..compat import ( - compat_parse_qs, compat_str, compat_urllib_parse_urlencode, - compat_urllib_parse_urlparse, ) from ..utils import ( + clean_html, decode_packed_codes, + get_element_by_id, + get_element_by_attribute, ExtractorError, ohdave_rsa_encrypt, remove_start, - sanitized_Request, - urlencode_postdata, - url_basename, ) @@ -171,70 +165,21 @@ class IqiyiIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.iqiyi.com/v_19rrojlavg.html', - 'md5': '2cb594dc2781e6c941a110d8f358118b', + # MD5 checksum differs on my machine and Travis CI 'info_dict': { 'id': '9c1fb1b99d192b21c559e5a1a2cb3c73', + 'ext': 'mp4', 'title': '美国德州空中惊现奇异云团 酷似UFO', - 'ext': 'f4v', } }, { 'url': 'http://www.iqiyi.com/v_19rrhnnclk.html', + 'md5': '667171934041350c5de3f5015f7f1152', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb', - 'title': '名侦探柯南第752集', - }, - 'playlist': [{ - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }, { - 'info_dict': { - 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8', - 'ext': 'f4v', - 'title': '名侦探柯南第752集', - }, - }], - 'params': { - 'skip_download': True, + 'ext': 'mp4', + 'title': '名侦探柯南 国语版：第752集 迫近灰原秘密的黑影 下篇', }, + 'skip': 'Geo-restricted to China', }, { 'url': 'http://www.iqiyi.com/w_19rt6o8t9p.html', 'only_matching': True, @@ -250,22 +195,10 @@ class IqiyiIE(InfoExtractor): 'url': 'http://www.iqiyi.com/v_19rrny4w8w.html', 'info_dict': { 'id': 'f3cf468b39dddb30d676f89a91200dc1', + 'ext': 'mp4', 'title': '泰坦尼克号', }, - 'playlist': [{ - 'info_dict': { - 'id': 'f3cf468b39dddb30d676f89a91200dc1_part1', - 'ext': 'f4v', - 'title': '泰坦尼克号', - }, - }, { - 'info_dict': { - 'id': 'f3cf468b39dddb30d676f89a91200dc1_part2', - 'ext': 'f4v', - 'title': '泰坦尼克号', - }, - }], - 'expected_warnings': ['Needs a VIP account for full video'], + 'skip': 'Geo-restricted to China', }, { 'url': 'http://www.iqiyi.com/a_19rrhb8ce1.html', 'info_dict': { @@ -278,20 +211,15 @@ class IqiyiIE(InfoExtractor): 'only_matching': True, }] - _FORMATS_MAP = [ - ('1', 'h6'), - ('2', 'h5'), - ('3', 'h4'), - ('4', 'h3'), - ('5', 'h2'), - ('10', 'h1'), - ] - - AUTH_API_ERRORS = { - # No preview available (不允许试看鉴权失败) - 'Q00505': 'This video requires a VIP account', - # End of preview time (试看结束鉴权失败) - 'Q00506': 'Needs a VIP account for full video', + _FORMATS_MAP = { + '96': 1, # 216p, 240p + '1': 2, # 336p, 360p + '2': 3, # 480p, 504p + '21': 4, # 504p + '4': 5, # 720p + '17': 5, # 720p + '5': 6, # 1072p, 1080p + '18': 7, # 1080p } def _real_initialize(self): @@ -352,174 +280,23 @@ class IqiyiIE(InfoExtractor): return True - def _authenticate_vip_video(self, api_video_url, video_id, tvid, _uuid, do_report_warning): - auth_params = { - # version and platform hard-coded in com/qiyi/player/core/model/remote/AuthenticationRemote.as - 'version': '2.0', - 'platform': 'b6c13e26323c537d', - 'aid': tvid, - 'tvid': tvid, - 'uid': '', - 'deviceId': _uuid, - 'playType': 'main', # XXX: always main? - 'filename': os.path.splitext(url_basename(api_video_url))[0], - } + def get_raw_data(self, tvid, video_id): + tm = int(time.time() * 1000) - qd_items = compat_parse_qs(compat_urllib_parse_urlparse(api_video_url).query) - for key, val in qd_items.items(): - auth_params[key] = val[0] - - auth_req = sanitized_Request( - 'http://api.vip.iqiyi.com/services/ckn.action', - urlencode_postdata(auth_params)) - # iQiyi server throws HTTP 405 error without the following header - auth_req.add_header('Content-Type', 'application/x-www-form-urlencoded') - auth_result = self._download_json( - auth_req, video_id, - note='Downloading video authentication JSON', - errnote='Unable to download video authentication JSON') - - code = auth_result.get('code') - msg = self.AUTH_API_ERRORS.get(code) or auth_result.get('msg') or code - if code == 'Q00506': - if do_report_warning: - self.report_warning(msg) - return False - if 'data' not in auth_result: - if msg is not None: - raise ExtractorError('%s said: %s' % (self.IE_NAME, msg), expected=True) - raise ExtractorError('Unexpected error from Iqiyi auth API') - - return auth_result['data'] - - def construct_video_urls(self, data, video_id, _uuid, tvid): - def do_xor(x, y): - a = y % 3 - if a == 1: - return x ^ 121 - if a == 2: - return x ^ 72 - return x ^ 103 - - def get_encode_code(l): - a = 0 - b = l.split('-') - c = len(b) - s = '' - for i in range(c - 1, -1, -1): - a = do_xor(int(b[c - i - 1], 16), i) - s += chr(a) - return s[::-1] - - def get_path_key(x, format_id, segment_index): - mg = ')(*&^flash@#$%a' - tm = self._download_json( - 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id, - note='Download path key of segment %d for format %s' % (segment_index + 1, format_id) - )['t'] - t = str(int(math.floor(int(tm) / (600.0)))) - return md5_text(t + mg + x) - - video_urls_dict = {} - need_vip_warning_report = True - for format_item in data['vp']['tkl'][0]['vs']: - if 0 < int(format_item['bid']) <= 10: - format_id = self.get_format(format_item['bid']) - else: - continue - - video_urls = [] - - video_urls_info = format_item['fs'] - if not format_item['fs'][0]['l'].startswith('/'): - t = get_encode_code(format_item['fs'][0]['l']) - if t.endswith('mp4'): - video_urls_info = format_item['flvs'] - - for segment_index, segment in enumerate(video_urls_info): - vl = segment['l'] - if not vl.startswith('/'): - vl = get_encode_code(vl) - is_vip_video = '/vip/' in vl - filesize = segment['b'] - base_url = data['vp']['du'].split('/') - if not is_vip_video: - key = get_path_key( - vl.split('/')[-1].split('.')[0], format_id, segment_index) - base_url.insert(-1, key) - base_url = '/'.join(base_url) - param = { - 'su': _uuid, - 'qyid': uuid.uuid4().hex, - 'client': '', - 'z': '', - 'bt': '', - 'ct': '', - 'tn': str(int(time.time())) - } - api_video_url = base_url + vl - if is_vip_video: - api_video_url = api_video_url.replace('.f4v', '.hml') - auth_result = self._authenticate_vip_video( - api_video_url, video_id, tvid, _uuid, need_vip_warning_report) - if auth_result is False: - need_vip_warning_report = False - break - param.update({ - 't': auth_result['t'], - # cid is hard-coded in com/qiyi/player/core/player/RuntimeData.as - 'cid': 'afbe8fd3d73448c9', - 'vid': video_id, - 'QY00001': auth_result['u'], - }) - api_video_url += '?' if '?' not in api_video_url else '&' - api_video_url += compat_urllib_parse_urlencode(param) - js = self._download_json( - api_video_url, video_id, - note='Download video info of segment %d for format %s' % (segment_index + 1, format_id)) - video_url = js['l'] - video_urls.append( - (video_url, filesize)) - - video_urls_dict[format_id] = video_urls - return video_urls_dict - - def get_format(self, bid): - matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)] - return matched_format_ids[0] if len(matched_format_ids) else None - - def get_bid(self, format_id): - matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id] - return matched_bids[0] if len(matched_bids) else None - - def get_raw_data(self, tvid, video_id, enc_key, _uuid): - tm = str(int(time.time())) - tail = tm + tvid - param = { - 'key': 'fvip', - 'src': md5_text('youtube-dl'), - 'tvId': tvid, + key = 'd5fb4bd9d50c4be6948c97edd7254b0e' + sc = md5_text(compat_str(tm) + key + tvid) + params = { + 'tvid': tvid, 'vid': video_id, - 'vinfo': 1, - 'tm': tm, - 'enc': md5_text(enc_key + tail), - 'qyid': _uuid, - 'tn': random.random(), - 'um': 0, - 'authkey': md5_text(md5_text('') + tail), - 'k_tag': 1, + 'src': '76f90cbd92f94a2e925d83e8ccd22cb7', + 'sc': sc, + 't': tm, } - api_url = 'http://cache.video.qiyi.com/vms' + '?' + \ - compat_urllib_parse_urlencode(param) - raw_data = self._download_json(api_url, video_id) - return raw_data - - def get_enc_key(self, video_id): - # TODO: automatic key extraction - # last update at 2016-01-22 for Zombie::bite - enc_key = '4a1caba4b4465345366f28da7c117d20' - return enc_key + return self._download_json( + 'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id), + video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='), + query=params, headers=self.geo_verification_headers()) def _extract_playlist(self, webpage): PAGE_SIZE = 50 @@ -568,58 +345,41 @@ class IqiyiIE(InfoExtractor): r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid') video_id = self._search_regex( r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id') - _uuid = uuid.uuid4().hex - - enc_key = self.get_enc_key(video_id) - - raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid) - - if raw_data['code'] != 'A000000': - raise ExtractorError('Unable to load data. Error code: ' + raw_data['code']) - - data = raw_data['data'] - - title = data['vi']['vn'] - - # generate video_urls_dict - video_urls_dict = self.construct_video_urls( - data, video_id, _uuid, tvid) - - # construct info - entries = [] - for format_id in video_urls_dict: - video_urls = video_urls_dict[format_id] - for i, video_url_info in enumerate(video_urls): - if len(entries) < i + 1: - entries.append({'formats': []}) - entries[i]['formats'].append( - { - 'url': video_url_info[0], - 'filesize': video_url_info[-1], - 'format_id': format_id, - 'preference': int(self.get_bid(format_id)) - } - ) - - for i in range(len(entries)): - self._sort_formats(entries[i]['formats']) - entries[i].update( - { - 'id': '%s_part%d' % (video_id, i + 1), - 'title': title, - } - ) - - if len(entries) > 1: - info = { - '_type': 'multi_video', - 'id': video_id, - 'title': title, - 'entries': entries, - } - else: - info = entries[0] - info['id'] = video_id - info['title'] = title - - return info + + formats = [] + for _ in range(5): + raw_data = self.get_raw_data(tvid, video_id) + + if raw_data['code'] != 'A00000': + if raw_data['code'] == 'A00111': + self.raise_geo_restricted() + raise ExtractorError('Unable to load data. Error code: ' + raw_data['code']) + + data = raw_data['data'] + + for stream in data['vidl']: + if 'm3utx' not in stream: + continue + vd = compat_str(stream['vd']) + formats.append({ + 'url': stream['m3utx'], + 'format_id': vd, + 'ext': 'mp4', + 'preference': self._FORMATS_MAP.get(vd, -1), + 'protocol': 'm3u8_native', + }) + + if formats: + break + + self._sleep(5, video_id) + + self._sort_formats(formats) + title = (get_element_by_id('widget-videotitle', webpage) or + clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage))) + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + } diff --git a/youtube_dl/extractor/jwplatform.py b/youtube_dl/extractor/jwplatform.py index 8a5e562db..e44e31104 100644 --- a/youtube_dl/extractor/jwplatform.py +++ b/youtube_dl/extractor/jwplatform.py @@ -5,33 +5,76 @@ import re from .common import InfoExtractor from ..utils import ( + determine_ext, float_or_none, int_or_none, ) class JWPlatformBaseIE(InfoExtractor): - def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True): + @staticmethod + def _find_jwplayer_data(webpage): + # TODO: Merge this with JWPlayer-related codes in generic.py + + mobj = re.search( + 'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\((?P<options>[^)]+)\)', + webpage) + if mobj: + return mobj.group('options') + + def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs): + jwplayer_data = self._parse_json( + self._find_jwplayer_data(webpage), video_id) + return self._parse_jwplayer_data( + jwplayer_data, video_id, *args, **kwargs) + + def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True, m3u8_id=None, rtmp_params=None): + # JWPlayer backward compatibility: flattened playlists + # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96 + if 'playlist' not in jwplayer_data: + jwplayer_data = {'playlist': [jwplayer_data]} + video_data = jwplayer_data['playlist'][0] + # JWPlayer backward compatibility: flattened sources + # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35 + if 'sources' not in video_data: + video_data['sources'] = [video_data] + formats = [] for source in video_data['sources']: source_url = self._proto_relative_url(source['file']) source_type = source.get('type') or '' - if source_type in ('application/vnd.apple.mpegurl', 'hls'): + if source_type in ('application/vnd.apple.mpegurl', 'hls') or determine_ext(source_url) == 'm3u8': formats.extend(self._extract_m3u8_formats( - source_url, video_id, 'mp4', 'm3u8_native', fatal=False)) + source_url, video_id, 'mp4', 'm3u8_native', m3u8_id=m3u8_id, fatal=False)) elif source_type.startswith('audio'): formats.append({ 'url': source_url, 'vcodec': 'none', }) else: - formats.append({ + a_format = { 'url': source_url, 'width': int_or_none(source.get('width')), 'height': int_or_none(source.get('height')), - }) + } + if source_url.startswith('rtmp'): + a_format['ext'] = 'flv', + + # See com/longtailvideo/jwplayer/media/RTMPMediaProvider.as + # of jwplayer.flash.swf + rtmp_url_parts = re.split( + r'((?:mp4|mp3|flv):)', source_url, 1) + if len(rtmp_url_parts) == 3: + rtmp_url, prefix, play_path = rtmp_url_parts + a_format.update({ + 'url': rtmp_url, + 'play_path': prefix + play_path, + }) + if rtmp_params: + a_format.update(rtmp_params) + formats.append(a_format) self._sort_formats(formats) subtitles = {} diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index a65697ff5..1729f5bfb 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -6,7 +6,6 @@ import base64 from .common import InfoExtractor from ..compat import ( - compat_urllib_parse_urlencode, compat_urlparse, compat_parse_qs, ) @@ -15,6 +14,7 @@ from ..utils import ( ExtractorError, int_or_none, unsmuggle_url, + smuggle_url, ) @@ -34,7 +34,8 @@ class KalturaIE(InfoExtractor): )(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))? ) ''' - _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?' + _SERVICE_URL = 'http://cdnapi.kaltura.com' + _SERVICE_BASE = '/api_v3/index.php' _TESTS = [ { 'url': 'kaltura:269692:1_1jc2y3e4', @@ -64,16 +65,50 @@ class KalturaIE(InfoExtractor): } ] - def _kaltura_api_call(self, video_id, actions, *args, **kwargs): + @staticmethod + def _extract_url(webpage): + mobj = ( + re.search( + r"""(?xs) + kWidget\.(?:thumb)?[Ee]mbed\( + \{.*? + (?P<q1>['\"])wid(?P=q1)\s*:\s* + (?P<q2>['\"])_?(?P<partner_id>[^'\"]+)(?P=q2),.*? + (?P<q3>['\"])entry_?[Ii]d(?P=q3)\s*:\s* + (?P<q4>['\"])(?P<id>[^'\"]+)(?P=q4), + """, webpage) or + re.search( + r'''(?xs) + (?P<q1>["\']) + (?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*? + (?P=q1).*? + (?: + entry_?[Ii]d| + (?P<q2>["\'])entry_?[Ii]d(?P=q2) + )\s*:\s* + (?P<q3>["\'])(?P<id>.+?)(?P=q3) + ''', webpage)) + if mobj: + embed_info = mobj.groupdict() + url = 'kaltura:%(partner_id)s:%(id)s' % embed_info + escaped_pid = re.escape(embed_info['partner_id']) + service_url = re.search( + r'<script[^>]+src=["\']((?:https?:)?//.+?)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid), + webpage) + if service_url: + url = smuggle_url(url, {'service_url': service_url.group(1)}) + return url + + def _kaltura_api_call(self, video_id, actions, service_url=None, *args, **kwargs): params = actions[0] if len(actions) > 1: for i, a in enumerate(actions[1:], start=1): for k, v in a.items(): params['%d:%s' % (i, k)] = v - query = compat_urllib_parse_urlencode(params) - url = self._API_BASE + query - data = self._download_json(url, video_id, *args, **kwargs) + data = self._download_json( + (service_url or self._SERVICE_URL) + self._SERVICE_BASE, + video_id, query=params, *args, **kwargs) status = data if len(actions) == 1 else data[0] if status.get('objectType') == 'KalturaAPIException': @@ -82,7 +117,7 @@ class KalturaIE(InfoExtractor): return data - def _get_kaltura_signature(self, video_id, partner_id): + def _get_kaltura_signature(self, video_id, partner_id, service_url=None): actions = [{ 'apiVersion': '3.1', 'expiry': 86400, @@ -92,10 +127,10 @@ class KalturaIE(InfoExtractor): 'widgetId': '_%s' % partner_id, }] return self._kaltura_api_call( - video_id, actions, note='Downloading Kaltura signature')['ks'] + video_id, actions, service_url, note='Downloading Kaltura signature')['ks'] - def _get_video_info(self, video_id, partner_id): - signature = self._get_kaltura_signature(video_id, partner_id) + def _get_video_info(self, video_id, partner_id, service_url=None): + signature = self._get_kaltura_signature(video_id, partner_id, service_url) actions = [ { 'action': 'null', @@ -118,7 +153,7 @@ class KalturaIE(InfoExtractor): }, ] return self._kaltura_api_call( - video_id, actions, note='Downloading video info JSON') + video_id, actions, service_url, note='Downloading video info JSON') def _real_extract(self, url): url, smuggled_data = unsmuggle_url(url, {}) @@ -127,7 +162,7 @@ class KalturaIE(InfoExtractor): partner_id, entry_id = mobj.group('partner_id', 'id') ks = None if partner_id and entry_id: - info, flavor_assets = self._get_video_info(entry_id, partner_id) + info, flavor_assets = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url')) else: path, query = mobj.group('path', 'query') if not path and not query: @@ -175,12 +210,17 @@ class KalturaIE(InfoExtractor): unsigned_url += '?referrer=%s' % referrer return unsigned_url + data_url = info['dataUrl'] + if '/flvclipper/' in data_url: + data_url = re.sub(r'/flvclipper/.*', '/serveFlavor', data_url) + formats = [] for f in flavor_assets: # Continue if asset is not ready if f['status'] != 2: continue - video_url = sign_url('%s/flavorId/%s' % (info['dataUrl'], f['id'])) + video_url = sign_url( + '%s/flavorId/%s' % (data_url, f['id'])) formats.append({ 'format_id': '%(fileExt)s-%(bitrate)s' % f, 'ext': f.get('fileExt'), @@ -193,9 +233,12 @@ class KalturaIE(InfoExtractor): 'width': int_or_none(f.get('width')), 'url': video_url, }) - m3u8_url = sign_url(info['dataUrl'].replace('format/url', 'format/applehttp')) - formats.extend(self._extract_m3u8_formats( - m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + if '/playManifest/' in data_url: + m3u8_url = sign_url(data_url.replace( + 'format/url', 'format/applehttp')) + formats.extend(self._extract_m3u8_formats( + m3u8_url, entry_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) self._check_formats(formats, entry_id) self._sort_formats(formats) diff --git a/youtube_dl/extractor/kamcord.py b/youtube_dl/extractor/kamcord.py new file mode 100644 index 000000000..b50120d98 --- /dev/null +++ b/youtube_dl/extractor/kamcord.py @@ -0,0 +1,71 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + int_or_none, + qualities, +) + + +class KamcordIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?kamcord\.com/v/(?P<id>[^/?#&]+)' + _TEST = { + 'url': 'https://www.kamcord.com/v/hNYRduDgWb4', + 'md5': 'c3180e8a9cfac2e86e1b88cb8751b54c', + 'info_dict': { + 'id': 'hNYRduDgWb4', + 'ext': 'mp4', + 'title': 'Drinking Madness', + 'uploader': 'jacksfilms', + 'uploader_id': '3044562', + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + video = self._parse_json( + self._search_regex( + r'window\.__props\s*=\s*({.+?});?(?:\n|\s*</script)', + webpage, 'video'), + video_id)['video'] + + title = video['title'] + + formats = self._extract_m3u8_formats( + video['play']['hls'], video_id, 'mp4', entry_protocol='m3u8_native') + self._sort_formats(formats) + + uploader = video.get('user', {}).get('username') + uploader_id = video.get('user', {}).get('id') + + view_count = int_or_none(video.get('viewCount')) + like_count = int_or_none(video.get('heartCount')) + comment_count = int_or_none(video.get('messageCount')) + + preference_key = qualities(('small', 'medium', 'large')) + + thumbnails = [{ + 'url': thumbnail_url, + 'id': thumbnail_id, + 'preference': preference_key(thumbnail_id), + } for thumbnail_id, thumbnail_url in (video.get('thumbnail') or {}).items() + if isinstance(thumbnail_id, compat_str) and isinstance(thumbnail_url, compat_str)] + + return { + 'id': video_id, + 'title': title, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'view_count': view_count, + 'like_count': like_count, + 'comment_count': comment_count, + 'thumbnails': thumbnails, + 'formats': formats, + } diff --git a/youtube_dl/extractor/kuwo.py b/youtube_dl/extractor/kuwo.py index c0ece5113..b1d460599 100644 --- a/youtube_dl/extractor/kuwo.py +++ b/youtube_dl/extractor/kuwo.py @@ -26,11 +26,6 @@ class KuwoBaseIE(InfoExtractor): def _get_formats(self, song_id, tolerate_ip_deny=False): formats = [] for file_format in self._FORMATS: - headers = {} - cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') - if cn_verification_proxy: - headers['Ytdl-request-proxy'] = cn_verification_proxy - query = { 'format': file_format['ext'], 'br': file_format.get('br', ''), @@ -42,7 +37,7 @@ class KuwoBaseIE(InfoExtractor): song_url = self._download_webpage( 'http://antiserver.kuwo.cn/anti.s', song_id, note='Download %s url info' % file_format['format'], - query=query, headers=headers, + query=query, headers=self.geo_verification_headers(), ) if song_url == 'IPDeny' and not tolerate_ip_deny: @@ -81,7 +76,7 @@ class KuwoIE(KuwoBaseIE): 'id': '6446136', 'ext': 'mp3', 'title': '心', - 'description': 'md5:b2ab6295d014005bfc607525bfc1e38a', + 'description': 'md5:5d0e947b242c35dc0eb1d2fce9fbf02c', 'creator': 'IU', 'upload_date': '20150518', }, @@ -102,10 +97,10 @@ class KuwoIE(KuwoBaseIE): raise ExtractorError('this song has been offline because of copyright issues', expected=True) song_name = self._html_search_regex( - r'(?s)class="(?:[^"\s]+\s+)*title(?:\s+[^"\s]+)*".*?<h1[^>]+title="([^"]+)"', webpage, 'song name') - singer_name = self._html_search_regex( - r'<div[^>]+class="s_img">\s*<a[^>]+title="([^>]+)"', - webpage, 'singer name', fatal=False) + r'<p[^>]+id="lrcName">([^<]+)</p>', webpage, 'song name') + singer_name = remove_start(self._html_search_regex( + r'<a[^>]+href="http://www\.kuwo\.cn/artist/content\?name=([^"]+)">', + webpage, 'singer name', fatal=False), '歌手') lrc_content = clean_html(get_element_by_id('lrcContent', webpage)) if lrc_content == '暂无': # indicates no lyrics lrc_content = None @@ -114,7 +109,7 @@ class KuwoIE(KuwoBaseIE): self._sort_formats(formats) album_id = self._html_search_regex( - r'<p[^>]+class="album"[^<]+<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"', + r'<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"', webpage, 'album id', fatal=False) publish_time = None @@ -148,8 +143,8 @@ class KuwoAlbumIE(InfoExtractor): 'url': 'http://www.kuwo.cn/album/502294/', 'info_dict': { 'id': '502294', - 'title': 'M', - 'description': 'md5:6a7235a84cc6400ec3b38a7bdaf1d60c', + 'title': 'Made\xa0Series\xa0《M》', + 'description': 'md5:d463f0d8a0ff3c3ea3d6ed7452a9483f', }, 'playlist_count': 2, } @@ -209,7 +204,7 @@ class KuwoSingerIE(InfoExtractor): 'url': 'http://www.kuwo.cn/mingxing/bruno+mars/', 'info_dict': { 'id': 'bruno+mars', - 'title': 'Bruno Mars', + 'title': 'Bruno\xa0Mars', }, 'playlist_mincount': 329, }, { @@ -268,7 +263,7 @@ class KuwoCategoryIE(InfoExtractor): 'title': '八十年代精选', 'description': '这些都是属于八十年代的回忆！', }, - 'playlist_count': 24, + 'playlist_mincount': 24, } def _real_extract(self, url): @@ -283,6 +278,8 @@ class KuwoCategoryIE(InfoExtractor): category_desc = remove_start( get_element_by_id('intro', webpage).strip(), '%s简介：' % category_name) + if category_desc == '暂无': + category_desc = None jsonm = self._parse_json(self._html_search_regex( r'var\s+jsonm\s*=\s*([^;]+);', webpage, 'category songs'), category_id) @@ -304,7 +301,7 @@ class KuwoMvIE(KuwoBaseIE): 'id': '6480076', 'ext': 'mp4', 'title': 'My HouseMV', - 'creator': '2PM', + 'creator': 'PM02:00', }, # In this video, music URLs (anti.s) are blocked outside China and # USA, while the MV URL (mvurl) is available globally, so force the MV diff --git a/youtube_dl/extractor/la7.py b/youtube_dl/extractor/la7.py index b08f6e3c9..da5a5de4a 100644 --- a/youtube_dl/extractor/la7.py +++ b/youtube_dl/extractor/la7.py @@ -1,60 +1,65 @@ +# coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - parse_duration, + js_to_json, + smuggle_url, ) class LA7IE(InfoExtractor): - IE_NAME = 'la7.tv' - _VALID_URL = r'''(?x) - https?://(?:www\.)?la7\.tv/ - (?: - richplayer/\?assetid=| - \?contentId= - ) - (?P<id>[0-9]+)''' - - _TEST = { - 'url': 'http://www.la7.tv/richplayer/?assetid=50355319', - 'md5': 'ec7d1f0224d20ba293ab56cf2259651f', + IE_NAME = 'la7.it' + _VALID_URL = r'''(?x)(https?://)?(?: + (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video)/| + tg\.la7\.it/repliche-tgla7\?id= + )(?P<id>.+)''' + + _TESTS = [{ + # 'src' is a plain URL + 'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722', + 'md5': '8b613ffc0c4bf9b9e377169fc19c214c', 'info_dict': { - 'id': '50355319', + 'id': 'inccool8-02-10-2015-163722', 'ext': 'mp4', - 'title': 'IL DIVO', - 'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti e Flavio Bucci', - 'duration': 6254, + 'title': 'Inc.Cool8', + 'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto atletico', + 'thumbnail': 're:^https?://.*', + 'uploader_id': 'kdla7pillole@iltrovatore.it', + 'timestamp': 1443814869, + 'upload_date': '20151002', }, - 'skip': 'Blocked in the US', - } + }, { + # 'src' is a dictionary + 'url': 'http://tg.la7.it/repliche-tgla7?id=189080', + 'md5': '6b0d8888d286e39870208dfeceaf456b', + 'info_dict': { + 'id': '189080', + 'ext': 'mp4', + 'title': 'TG LA7', + }, + }, { + 'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - xml_url = 'http://www.la7.tv/repliche/content/index.php?contentId=%s' % video_id - doc = self._download_xml(xml_url, video_id) - - video_title = doc.find('title').text - description = doc.find('description').text - duration = parse_duration(doc.find('duration').text) - thumbnail = doc.find('img').text - view_count = int(doc.find('views').text) - prefix = doc.find('.//fqdn').text.strip().replace('auto:', 'http:') + webpage = self._download_webpage(url, video_id) - formats = [{ - 'format': vnode.find('quality').text, - 'tbr': int(vnode.find('quality').text), - 'url': vnode.find('fms').text.strip().replace('mp4:', prefix), - } for vnode in doc.findall('.//videos/video')] - self._sort_formats(formats) + player_data = self._parse_json( + self._search_regex(r'videoLa7\(({[^;]+})\);', webpage, 'player data'), + video_id, transform_source=js_to_json) return { + '_type': 'url_transparent', + 'url': smuggle_url('kaltura:103:%s' % player_data['vid'], { + 'service_url': 'http://kdam.iltrovatore.it', + }), 'id': video_id, - 'title': video_title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - 'view_count': view_count, + 'title': player_data['title'], + 'description': self._og_search_description(webpage, default=None), + 'thumbnail': player_data.get('poster'), + 'ie_key': 'Kaltura', } diff --git a/youtube_dl/extractor/learnr.py b/youtube_dl/extractor/learnr.py new file mode 100644 index 000000000..1435e090e --- /dev/null +++ b/youtube_dl/extractor/learnr.py @@ -0,0 +1,33 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class LearnrIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?learnr\.pro/view/video/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.learnr.pro/view/video/51624-web-development-tutorial-for-beginners-1-how-to-build-webpages-with-html-css-javascript', + 'md5': '3719fdf0a68397f49899e82c308a89de', + 'info_dict': { + 'id': '51624', + 'ext': 'mp4', + 'title': 'Web Development Tutorial for Beginners (#1) - How to build webpages with HTML, CSS, Javascript', + 'description': 'md5:b36dbfa92350176cdf12b4d388485503', + 'uploader': 'LearnCode.academy', + 'uploader_id': 'learncodeacademy', + 'upload_date': '20131021', + }, + 'add_ie': ['Youtube'], + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + return { + '_type': 'url_transparent', + 'url': self._search_regex( + r"videoId\s*:\s*'([^']+)'", webpage, 'youtube id'), + 'id': video_id, + } diff --git a/youtube_dl/extractor/leeco.py b/youtube_dl/extractor/leeco.py index 375fdaed1..e9cc9aa59 100644 --- a/youtube_dl/extractor/leeco.py +++ b/youtube_dl/extractor/leeco.py @@ -20,15 +20,16 @@ from ..utils import ( int_or_none, orderedSet, parse_iso8601, - sanitized_Request, str_or_none, url_basename, + urshift, + update_url_query, ) class LeIE(InfoExtractor): IE_DESC = '乐视网' - _VALID_URL = r'https?://www\.le\.com/ptv/vplay/(?P<id>\d+)\.html' + _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|sports\.le\.com/video)/(?P<id>\d+)\.html' _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html' @@ -69,17 +70,16 @@ class LeIE(InfoExtractor): 'hls_prefer_native': True, }, 'skip': 'Only available in China', + }, { + 'url': 'http://sports.le.com/video/25737697.html', + 'only_matching': True, }] - @staticmethod - def urshift(val, n): - return val >> n if val >= 0 else (val + 0x100000000) >> n - # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf def ror(self, param1, param2): _loc3_ = 0 while _loc3_ < param2: - param1 = self.urshift(param1, 1) + ((param1 & 1) << 31) + param1 = urshift(param1, 1) + ((param1 & 1) << 31) _loc3_ += 1 return param1 @@ -90,6 +90,10 @@ class LeIE(InfoExtractor): _loc3_ = self.ror(_loc3_, _loc2_ % 17) return _loc3_ + # reversed from http://jstatic.letvcdn.com/sdk/player.js + def get_mms_key(self, time): + return self.ror(time, 8) ^ 185025305 + # see M3U8Encryption class in KLetvPlayer.swf @staticmethod def decrypt_m3u8(encrypted_data): @@ -110,28 +114,7 @@ class LeIE(InfoExtractor): return bytes(_loc7_) - def _real_extract(self, url): - media_id = self._match_id(url) - page = self._download_webpage(url, media_id) - params = { - 'id': media_id, - 'platid': 1, - 'splatid': 101, - 'format': 1, - 'tkey': self.calc_time_key(int(time.time())), - 'domain': 'www.le.com' - } - play_json_req = sanitized_Request( - 'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse_urlencode(params) - ) - cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') - if cn_verification_proxy: - play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy) - - play_json = self._download_json( - play_json_req, - media_id, 'Downloading playJson data') - + def _check_errors(self, play_json): # Check for errors playstatus = play_json['playstatus'] if playstatus['status'] == 0: @@ -142,43 +125,99 @@ class LeIE(InfoExtractor): msg = 'Generic error. flag = %d' % flag raise ExtractorError(msg, expected=True) - playurl = play_json['playurl'] - - formats = ['350', '1000', '1300', '720p', '1080p'] - dispatch = playurl['dispatch'] + def _real_extract(self, url): + media_id = self._match_id(url) + page = self._download_webpage(url, media_id) - urls = [] - for format_id in formats: - if format_id in dispatch: - media_url = playurl['domain'][0] + dispatch[format_id][0] - media_url += '&' + compat_urllib_parse_urlencode({ - 'm3v': 1, + play_json_h5 = self._download_json( + 'http://api.le.com/mms/out/video/playJsonH5', + media_id, 'Downloading html5 playJson data', query={ + 'id': media_id, + 'platid': 3, + 'splatid': 304, + 'format': 1, + 'tkey': self.get_mms_key(int(time.time())), + 'domain': 'www.le.com', + 'tss': 'no', + }, + headers=self.geo_verification_headers()) + self._check_errors(play_json_h5) + + play_json_flash = self._download_json( + 'http://api.le.com/mms/out/video/playJson', + media_id, 'Downloading flash playJson data', query={ + 'id': media_id, + 'platid': 1, + 'splatid': 101, + 'format': 1, + 'tkey': self.calc_time_key(int(time.time())), + 'domain': 'www.le.com', + }, + headers=self.geo_verification_headers()) + self._check_errors(play_json_flash) + + def get_h5_urls(media_url, format_id): + location = self._download_json( + media_url, media_id, + 'Download JSON metadata for format %s' % format_id, query={ 'format': 1, 'expect': 3, - 'rateid': format_id, - }) + 'tss': 'no', + })['location'] - nodes_data = self._download_json( - media_url, media_id, - 'Download JSON metadata for format %s' % format_id) + return { + 'http': update_url_query(location, {'tss': 'no'}), + 'hls': update_url_query(location, {'tss': 'ios'}), + } - req = self._request_webpage( - nodes_data['nodelist'][0]['location'], media_id, - note='Downloading m3u8 information for format %s' % format_id) + def get_flash_urls(media_url, format_id): + media_url += '&' + compat_urllib_parse_urlencode({ + 'm3v': 1, + 'format': 1, + 'expect': 3, + 'rateid': format_id, + }) - m3u8_data = self.decrypt_m3u8(req.read()) + nodes_data = self._download_json( + media_url, media_id, + 'Download JSON metadata for format %s' % format_id) - url_info_dict = { - 'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'), - 'ext': determine_ext(dispatch[format_id][1]), - 'format_id': format_id, - 'protocol': 'm3u8', - } + req = self._request_webpage( + nodes_data['nodelist'][0]['location'], media_id, + note='Downloading m3u8 information for format %s' % format_id) - if format_id[-1:] == 'p': - url_info_dict['height'] = int_or_none(format_id[:-1]) + m3u8_data = self.decrypt_m3u8(req.read()) - urls.append(url_info_dict) + return { + 'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'), + } + + extracted_formats = [] + formats = [] + for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)): + playurl = play_json['playurl'] + play_domain = playurl['domain'][0] + + for format_id, format_data in playurl.get('dispatch', []).items(): + if format_id in extracted_formats: + continue + extracted_formats.append(format_id) + + media_url = play_domain + format_data[0] + for protocol, format_url in get_urls(media_url, format_id).items(): + f = { + 'url': format_url, + 'ext': determine_ext(format_data[1]), + 'format_id': '%s-%s' % (protocol, format_id), + 'protocol': 'm3u8_native' if protocol == 'hls' else 'http', + 'quality': int_or_none(format_id), + } + + if format_id[-1:] == 'p': + f['height'] = int_or_none(format_id[:-1]) + + formats.append(f) + self._sort_formats(formats, ('height', 'quality', 'format_id')) publish_time = parse_iso8601(self._html_search_regex( r'发布时间 ([^<>]+) ', page, 'publish time', default=None), @@ -187,7 +226,7 @@ class LeIE(InfoExtractor): return { 'id': media_id, - 'formats': urls, + 'formats': formats, 'title': playurl['title'], 'thumbnail': playurl['pic'], 'description': description, @@ -196,7 +235,7 @@ class LeIE(InfoExtractor): class LePlaylistIE(InfoExtractor): - _VALID_URL = r'https?://[a-z]+\.le\.com/[a-z]+/(?P<id>[a-z0-9_]+)' + _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)' _TESTS = [{ 'url': 'http://www.le.com/tv/46177.html', diff --git a/youtube_dl/extractor/libraryofcongress.py b/youtube_dl/extractor/libraryofcongress.py new file mode 100644 index 000000000..0a94366fd --- /dev/null +++ b/youtube_dl/extractor/libraryofcongress.py @@ -0,0 +1,143 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + +from ..utils import ( + determine_ext, + float_or_none, + int_or_none, + parse_filesize, +) + + +class LibraryOfCongressIE(InfoExtractor): + IE_NAME = 'loc' + IE_DESC = 'Library of Congress' + _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9]+)' + _TESTS = [{ + # embedded via <div class="media-player" + 'url': 'http://loc.gov/item/90716351/', + 'md5': '353917ff7f0255aa6d4b80a034833de8', + 'info_dict': { + 'id': '90716351', + 'ext': 'mp4', + 'title': "Pa's trip to Mars", + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 0, + 'view_count': int, + }, + }, { + # webcast embedded via mediaObjectId + 'url': 'https://www.loc.gov/today/cyberlc/feature_wdesc.php?rec=5578', + 'info_dict': { + 'id': '5578', + 'ext': 'mp4', + 'title': 'Help! Preservation Training Needs Here, There & Everywhere', + 'duration': 3765, + 'view_count': int, + 'subtitles': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, { + # with direct download links + 'url': 'https://www.loc.gov/item/78710669/', + 'info_dict': { + 'id': '78710669', + 'ext': 'mp4', + 'title': 'La vie et la passion de Jesus-Christ', + 'duration': 0, + 'view_count': int, + 'formats': 'mincount:4', + }, + 'params': { + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + media_id = self._search_regex( + (r'id=(["\'])media-player-(?P<id>.+?)\1', + r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1', + r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1', + r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1'), + webpage, 'media id', group='id') + + data = self._download_json( + 'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id, + video_id)['mediaObject'] + + derivative = data['derivatives'][0] + media_url = derivative['derivativeUrl'] + + title = derivative.get('shortName') or data.get('shortName') or self._og_search_title( + webpage) + + # Following algorithm was extracted from setAVSource js function + # found in webpage + media_url = media_url.replace('rtmp', 'https') + + is_video = data.get('mediaType', 'v').lower() == 'v' + ext = determine_ext(media_url) + if ext not in ('mp4', 'mp3'): + media_url += '.mp4' if is_video else '.mp3' + + if 'vod/mp4:' in media_url: + formats = [{ + 'url': media_url.replace('vod/mp4:', 'hls-vod/media/') + '.m3u8', + 'format_id': 'hls', + 'ext': 'mp4', + 'protocol': 'm3u8_native', + 'quality': 1, + }] + elif 'vod/mp3:' in media_url: + formats = [{ + 'url': media_url.replace('vod/mp3:', ''), + 'vcodec': 'none', + }] + + download_urls = set() + for m in re.finditer( + r'<option[^>]+value=(["\'])(?P<url>.+?)\1[^>]+data-file-download=[^>]+>\s*(?P<id>.+?)(?:(?: |\s+)\((?P<size>.+?)\))?\s*<', webpage): + format_id = m.group('id').lower() + if format_id == 'gif': + continue + download_url = m.group('url') + if download_url in download_urls: + continue + download_urls.add(download_url) + formats.append({ + 'url': download_url, + 'format_id': format_id, + 'filesize_approx': parse_filesize(m.group('size')), + }) + + self._sort_formats(formats) + + duration = float_or_none(data.get('duration')) + view_count = int_or_none(data.get('viewCount')) + + subtitles = {} + cc_url = data.get('ccUrl') + if cc_url: + subtitles.setdefault('en', []).append({ + 'url': cc_url, + 'ext': 'ttml', + }) + + return { + 'id': video_id, + 'title': title, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'view_count': view_count, + 'formats': formats, + 'subtitles': subtitles, + } diff --git a/youtube_dl/extractor/lifenews.py b/youtube_dl/extractor/lifenews.py index ba2f80a75..c2b4490c4 100644 --- a/youtube_dl/extractor/lifenews.py +++ b/youtube_dl/extractor/lifenews.py @@ -7,48 +7,53 @@ from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( determine_ext, + ExtractorError, int_or_none, + parse_iso8601, remove_end, - unified_strdate, - ExtractorError, ) class LifeNewsIE(InfoExtractor): - IE_NAME = 'lifenews' - IE_DESC = 'LIFE | NEWS' - _VALID_URL = r'https?://lifenews\.ru/(?:mobile/)?(?P<section>news|video)/(?P<id>\d+)' + IE_NAME = 'life' + IE_DESC = 'Life.ru' + _VALID_URL = r'https?://life\.ru/t/[^/]+/(?P<id>\d+)' _TESTS = [{ # single video embedded via video/source - 'url': 'http://lifenews.ru/news/98736', + 'url': 'https://life.ru/t/новости/98736', 'md5': '77c95eaefaca216e32a76a343ad89d23', 'info_dict': { 'id': '98736', 'ext': 'mp4', 'title': 'Мужчина нашел дома архив оборонного завода', 'description': 'md5:3b06b1b39b5e2bea548e403d99b8bf26', + 'timestamp': 1344154740, 'upload_date': '20120805', + 'view_count': int, } }, { # single video embedded via iframe - 'url': 'http://lifenews.ru/news/152125', + 'url': 'https://life.ru/t/новости/152125', 'md5': '77d19a6f0886cd76bdbf44b4d971a273', 'info_dict': { 'id': '152125', 'ext': 'mp4', 'title': 'В Сети появилось видео захвата «Правым сектором» колхозных полей ', 'description': 'Жители двух поселков Днепропетровской области не простили радикалам угрозу лишения плодородных земель и пошли в лобовую. ', + 'timestamp': 1427961840, 'upload_date': '20150402', + 'view_count': int, } }, { # two videos embedded via iframe - 'url': 'http://lifenews.ru/news/153461', + 'url': 'https://life.ru/t/новости/153461', 'info_dict': { 'id': '153461', 'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве', 'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.', - 'upload_date': '20150505', + 'timestamp': 1430825520, + 'view_count': int, }, 'playlist': [{ 'md5': '9b6ef8bc0ffa25aebc8bdb40d89ab795', @@ -57,6 +62,7 @@ class LifeNewsIE(InfoExtractor): 'ext': 'mp4', 'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве (Видео 1)', 'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.', + 'timestamp': 1430825520, 'upload_date': '20150505', }, }, { @@ -66,22 +72,25 @@ class LifeNewsIE(InfoExtractor): 'ext': 'mp4', 'title': 'В Москве спасли потерявшегося медвежонка, который спрятался на дереве (Видео 2)', 'description': 'Маленький хищник не смог найти дорогу домой и обрел временное убежище на тополе недалеко от жилого массива, пока его не нашла соседская собака.', + 'timestamp': 1430825520, 'upload_date': '20150505', }, }], }, { - 'url': 'http://lifenews.ru/video/13035', + 'url': 'https://life.ru/t/новости/213035', + 'only_matching': True, + }, { + 'url': 'https://life.ru/t/%D0%BD%D0%BE%D0%B2%D0%BE%D1%81%D1%82%D0%B8/153461', + 'only_matching': True, + }, { + 'url': 'https://life.ru/t/новости/411489/manuel_vals_nazval_frantsiiu_tsieliu_nomier_odin_dlia_ighil', 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - section = mobj.group('section') + video_id = self._match_id(url) - webpage = self._download_webpage( - 'http://lifenews.ru/%s/%s' % (section, video_id), - video_id, 'Downloading page') + webpage = self._download_webpage(url, video_id) video_urls = re.findall( r'<video[^>]+><source[^>]+src=["\'](.+?)["\']', webpage) @@ -95,26 +104,22 @@ class LifeNewsIE(InfoExtractor): title = remove_end( self._og_search_title(webpage), - ' - Первый по срочным новостям — LIFE | NEWS') + ' - Life.ru') description = self._og_search_description(webpage) view_count = self._html_search_regex( - r'<div class=\'views\'>\s*(\d+)\s*</div>', webpage, 'view count', fatal=False) - comment_count = self._html_search_regex( - r'=\'commentCount\'[^>]*>\s*(\d+)\s*<', - webpage, 'comment count', fatal=False) + r'<div[^>]+class=(["\']).*?\bhits-count\b.*?\1[^>]*>\s*(?P<value>\d+)\s*</div>', + webpage, 'view count', fatal=False, group='value') - upload_date = self._html_search_regex( - r'<time[^>]*datetime=\'([^\']+)\'', webpage, 'upload date', fatal=False) - if upload_date is not None: - upload_date = unified_strdate(upload_date) + timestamp = parse_iso8601(self._search_regex( + r'<time[^>]+datetime=(["\'])(?P<value>.+?)\1', + webpage, 'upload date', fatal=False, group='value')) common_info = { 'description': description, 'view_count': int_or_none(view_count), - 'comment_count': int_or_none(comment_count), - 'upload_date': upload_date, + 'timestamp': timestamp, } def make_entry(video_id, video_url, index=None): @@ -183,7 +188,8 @@ class LifeEmbedIE(InfoExtractor): ext = determine_ext(video_url) if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', m3u8_id='m3u8')) + video_url, video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='m3u8')) else: formats.append({ 'url': video_url, diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py index 2599d45c3..5d2c3e256 100644 --- a/youtube_dl/extractor/limelight.py +++ b/youtube_dl/extractor/limelight.py @@ -98,13 +98,19 @@ class LimelightBaseIE(InfoExtractor): } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')] subtitles = {} - for caption in properties.get('captions', {}): + for caption in properties.get('captions', []): lang = caption.get('language_code') subtitles_url = caption.get('url') if lang and subtitles_url: - subtitles[lang] = [{ + subtitles.setdefault(lang, []).append({ 'url': subtitles_url, - }] + }) + closed_captions_url = properties.get('closed_captions_url') + if closed_captions_url: + subtitles.setdefault('en', []).append({ + 'url': closed_captions_url, + 'ext': 'ttml', + }) return { 'id': video_id, @@ -123,7 +129,18 @@ class LimelightBaseIE(InfoExtractor): class LimelightMediaIE(LimelightBaseIE): IE_NAME = 'limelight' - _VALID_URL = r'(?:limelight:media:|https?://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})' + _VALID_URL = r'''(?x) + (?: + limelight:media:| + https?:// + (?: + link\.videoplatform\.limelight\.com/media/| + assets\.delvenetworks\.com/player/loader\.swf + ) + \?.*?\bmediaId= + ) + (?P<id>[a-z0-9]{32}) + ''' _TESTS = [{ 'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86', 'info_dict': { @@ -158,6 +175,9 @@ class LimelightMediaIE(LimelightBaseIE): # rtmp download 'skip_download': True, }, + }, { + 'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452', + 'only_matching': True, }] _PLAYLIST_SERVICE_PATH = 'media' _API_PATH = 'media' @@ -176,15 +196,29 @@ class LimelightMediaIE(LimelightBaseIE): class LimelightChannelIE(LimelightBaseIE): IE_NAME = 'limelight:channel' - _VALID_URL = r'(?:limelight:channel:|https?://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})' - _TEST = { + _VALID_URL = r'''(?x) + (?: + limelight:channel:| + https?:// + (?: + link\.videoplatform\.limelight\.com/media/| + assets\.delvenetworks\.com/player/loader\.swf + ) + \?.*?\bchannelId= + ) + (?P<id>[a-z0-9]{32}) + ''' + _TESTS = [{ 'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082', 'info_dict': { 'id': 'ab6a524c379342f9b23642917020c082', 'title': 'Javascript Sample Code', }, 'playlist_mincount': 3, - } + }, { + 'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082', + 'only_matching': True, + }] _PLAYLIST_SERVICE_PATH = 'channel' _API_PATH = 'channels' @@ -207,15 +241,29 @@ class LimelightChannelIE(LimelightBaseIE): class LimelightChannelListIE(LimelightBaseIE): IE_NAME = 'limelight:channel_list' - _VALID_URL = r'(?:limelight:channel_list:|https?://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})' - _TEST = { + _VALID_URL = r'''(?x) + (?: + limelight:channel_list:| + https?:// + (?: + link\.videoplatform\.limelight\.com/media/| + assets\.delvenetworks\.com/player/loader\.swf + ) + \?.*?\bchannelListId= + ) + (?P<id>[a-z0-9]{32}) + ''' + _TESTS = [{ 'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b', 'info_dict': { 'id': '301b117890c4465c8179ede21fd92e2b', 'title': 'Website - Hero Player', }, 'playlist_mincount': 2, - } + }, { + 'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b', + 'only_matching': True, + }] _PLAYLIST_SERVICE_PATH = 'channel_list' def _real_extract(self, url): diff --git a/youtube_dl/extractor/litv.py b/youtube_dl/extractor/litv.py new file mode 100644 index 000000000..3356d015d --- /dev/null +++ b/youtube_dl/extractor/litv.py @@ -0,0 +1,137 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import json +import re + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + int_or_none, + smuggle_url, + unsmuggle_url, +) + + +class LiTVIE(InfoExtractor): + _VALID_URL = r'https?://www\.litv\.tv/vod/[^/]+/content\.do\?.*?\bid=(?P<id>[^&]+)' + + _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?id=%s' + + _TESTS = [{ + 'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1', + 'info_dict': { + 'id': 'VOD00041606', + 'title': '花千骨', + }, + 'playlist_count': 50, + }, { + 'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1', + 'info_dict': { + 'id': 'VOD00041610', + 'ext': 'mp4', + 'title': '花千骨第1集', + 'thumbnail': 're:https?://.*\.jpg$', + 'description': 'md5:c7017aa144c87467c4fb2909c4b05d6f', + 'episode_number': 1, + }, + 'params': { + 'noplaylist': True, + 'skip_download': True, # m3u8 download + }, + 'skip': 'Georestricted to Taiwan', + }] + + def _extract_playlist(self, season_list, video_id, vod_data, view_data, prompt=True): + episode_title = view_data['title'] + content_id = season_list['contentId'] + + if prompt: + self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (content_id, video_id)) + + all_episodes = [ + self.url_result(smuggle_url( + self._URL_TEMPLATE % (view_data['contentType'], episode['contentId']), + {'force_noplaylist': True})) # To prevent infinite recursion + for episode in season_list['episode']] + + return self.playlist_result(all_episodes, content_id, episode_title) + + def _real_extract(self, url): + url, data = unsmuggle_url(url, {}) + + video_id = self._match_id(url) + + noplaylist = self._downloader.params.get('noplaylist') + noplaylist_prompt = True + if 'force_noplaylist' in data: + noplaylist = data['force_noplaylist'] + noplaylist_prompt = False + + webpage = self._download_webpage(url, video_id) + + view_data = dict(map(lambda t: (t[0], t[2]), re.findall( + r'viewData\.([a-zA-Z]+)\s*=\s*(["\'])([^"\']+)\2', + webpage))) + + vod_data = self._parse_json(self._search_regex( + 'var\s+vod\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'), + video_id) + + season_list = list(vod_data.get('seasonList', {}).values()) + if season_list: + if not noplaylist: + return self._extract_playlist( + season_list[0], video_id, vod_data, view_data, + prompt=noplaylist_prompt) + + if noplaylist_prompt: + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + + # In browsers `getMainUrl` request is always issued. Usually this + # endpoint gives the same result as the data embedded in the webpage. + # If georestricted, there are no embedded data, so an extra request is + # necessary to get the error code + video_data = self._parse_json(self._search_regex( + r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);', + webpage, 'video data', default='{}'), video_id) + if not video_data: + payload = { + 'assetId': view_data['assetId'], + 'watchDevices': vod_data['watchDevices'], + 'contentType': view_data['contentType'], + } + video_data = self._download_json( + 'https://www.litv.tv/vod/getMainUrl', video_id, + data=json.dumps(payload).encode('utf-8'), + headers={'Content-Type': 'application/json'}) + + if not video_data.get('fullpath'): + error_msg = video_data.get('errorMessage') + if error_msg == 'vod.error.outsideregionerror': + self.raise_geo_restricted('This video is available in Taiwan only') + if error_msg: + raise ExtractorError('%s said: %s' % (self.IE_NAME, error_msg), expected=True) + raise ExtractorError('Unexpected result from %s' % self.IE_NAME) + + formats = self._extract_m3u8_formats( + video_data['fullpath'], video_id, ext='mp4', m3u8_id='hls') + for a_format in formats: + # LiTV HLS segments doesn't like compressions + a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = True + + title = view_data['title'] + view_data.get('secondaryMark', '') + description = view_data.get('description') + thumbnail = view_data.get('imageFile') + categories = [item['name'] for item in vod_data.get('category', [])] + episode = int_or_none(view_data.get('episode')) + + return { + 'id': video_id, + 'formats': formats, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'categories': categories, + 'episode_number': episode, + } diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py index 29fba5f30..ea0565ac0 100644 --- a/youtube_dl/extractor/liveleak.py +++ b/youtube_dl/extractor/liveleak.py @@ -17,7 +17,8 @@ class LiveLeakIE(InfoExtractor): 'ext': 'flv', 'description': 'extremely bad day for this guy..!', 'uploader': 'ljfriel2', - 'title': 'Most unlucky car accident' + 'title': 'Most unlucky car accident', + 'thumbnail': 're:^https?://.*\.jpg$' } }, { 'url': 'http://www.liveleak.com/view?i=f93_1390833151', @@ -28,6 +29,7 @@ class LiveLeakIE(InfoExtractor): 'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.', 'uploader': 'ARD_Stinkt', 'title': 'German Television does first Edward Snowden Interview (ENGLISH)', + 'thumbnail': 're:^https?://.*\.jpg$' } }, { 'url': 'http://www.liveleak.com/view?i=4f7_1392687779', @@ -49,7 +51,8 @@ class LiveLeakIE(InfoExtractor): 'ext': 'mp4', 'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.', 'uploader': 'bony333', - 'title': 'Crazy Hungarian tourist films close call waterspout in Croatia' + 'title': 'Crazy Hungarian tourist films close call waterspout in Croatia', + 'thumbnail': 're:^https?://.*\.jpg$' } }] @@ -72,6 +75,7 @@ class LiveLeakIE(InfoExtractor): age_limit = int_or_none(self._search_regex( r'you confirm that you are ([0-9]+) years and over.', webpage, 'age limit', default=None)) + video_thumbnail = self._og_search_thumbnail(webpage) sources_raw = self._search_regex( r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None) @@ -124,4 +128,5 @@ class LiveLeakIE(InfoExtractor): 'uploader': video_uploader, 'formats': formats, 'age_limit': age_limit, + 'thumbnail': video_thumbnail, } diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index eada7c299..bc7894bf1 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -150,7 +150,7 @@ class LivestreamIE(InfoExtractor): } def _extract_stream_info(self, stream_info): - broadcast_id = stream_info['broadcast_id'] + broadcast_id = compat_str(stream_info['broadcast_id']) is_live = stream_info.get('is_live') formats = [] @@ -203,9 +203,10 @@ class LivestreamIE(InfoExtractor): if not videos_info: break for v in videos_info: + v_id = compat_str(v['id']) entries.append(self.url_result( - 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v['id']), - 'Livestream', v['id'], v['caption'])) + 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v_id), + 'Livestream', v_id, v.get('caption'))) last_video = videos_info[-1]['id'] return self.playlist_result(entries, event_id, event_data['full_name']) diff --git a/youtube_dl/extractor/localnews8.py b/youtube_dl/extractor/localnews8.py new file mode 100644 index 000000000..aad396135 --- /dev/null +++ b/youtube_dl/extractor/localnews8.py @@ -0,0 +1,47 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class LocalNews8IE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?localnews8\.com/(?:[^/]+/)*(?P<display_id>[^/]+)/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.localnews8.com/news/rexburg-business-turns-carbon-fiber-scraps-into-wedding-rings/35183304', + 'md5': 'be4d48aea61aa2bde7be2ee47691ad20', + 'info_dict': { + 'id': '35183304', + 'display_id': 'rexburg-business-turns-carbon-fiber-scraps-into-wedding-rings', + 'ext': 'mp4', + 'title': 'Rexburg business turns carbon fiber scraps into wedding ring', + 'description': 'The process was first invented by Lamborghini and less than a dozen companies around the world use it.', + 'duration': 153, + 'timestamp': 1441844822, + 'upload_date': '20150910', + 'uploader_id': 'api', + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + display_id = mobj.group('display_id') + + webpage = self._download_webpage(url, display_id) + + partner_id = self._search_regex( + r'partnerId\s*[:=]\s*(["\'])(?P<id>\d+)\1', + webpage, 'partner id', group='id') + kaltura_id = self._search_regex( + r'videoIdString\s*[:=]\s*(["\'])kaltura:(?P<id>[0-9a-z_]+)\1', + webpage, 'videl id', group='id') + + return { + '_type': 'url_transparent', + 'url': 'kaltura:%s:%s' % (partner_id, kaltura_id), + 'ie_key': 'Kaltura', + 'id': video_id, + 'display_id': display_id, + } diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 86d47266f..1237e1573 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -1,106 +1,106 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor -from ..compat import compat_str +from ..compat import ( + compat_HTTPError, + compat_str, + compat_urlparse, +) from ..utils import ( ExtractorError, - clean_html, int_or_none, - sanitized_Request, urlencode_postdata, ) class LyndaBaseIE(InfoExtractor): - _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' + _SIGNIN_URL = 'https://www.lynda.com/signin' + _PASSWORD_URL = 'https://www.lynda.com/signin/password' + _USER_URL = 'https://www.lynda.com/signin/user' _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' _NETRC_MACHINE = 'lynda' def _real_initialize(self): self._login() + @staticmethod + def _check_error(json_string, key_or_keys): + keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys + for key in keys: + error = json_string.get(key) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) + + def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url): + action_url = self._search_regex( + r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html, + 'post url', default=fallback_action_url, group='url') + + if not action_url.startswith('http'): + action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url) + + form_data = self._hidden_inputs(form_html) + form_data.update(extra_form_data) + + try: + response = self._download_json( + action_url, None, note, + data=urlencode_postdata(form_data), + headers={ + 'Referer': referrer_url, + 'X-Requested-With': 'XMLHttpRequest', + }) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500: + response = self._parse_json(e.cause.read().decode('utf-8'), None) + self._check_error(response, ('email', 'password')) + raise + + self._check_error(response, 'ErrorMessage') + + return response, action_url + def _login(self): username, password = self._get_login_info() if username is None: return - login_form = { - 'username': username, - 'password': password, - 'remember': 'false', - 'stayPut': 'false' - } - request = sanitized_Request( - self._LOGIN_URL, urlencode_postdata(login_form)) - login_page = self._download_webpage( - request, None, 'Logging in as %s' % username) - - # Not (yet) logged in - m = re.search(r'loginResultJson\s*=\s*\'(?P<json>[^\']+)\';', login_page) - if m is not None: - response = m.group('json') - response_json = json.loads(response) - state = response_json['state'] - - if state == 'notlogged': - raise ExtractorError( - 'Unable to login, incorrect username and/or password', - expected=True) - - # This is when we get popup: - # > You're already logged in to lynda.com on two devices. - # > If you log in here, we'll log you out of another device. - # So, we need to confirm this. - if state == 'conflicted': - confirm_form = { - 'username': '', - 'password': '', - 'resolve': 'true', - 'remember': 'false', - 'stayPut': 'false', - } - request = sanitized_Request( - self._LOGIN_URL, urlencode_postdata(confirm_form)) - login_page = self._download_webpage( - request, None, - 'Confirming log in and log out from another device') - - if all(not re.search(p, login_page) for p in ('isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')): - if 'login error' in login_page: - mobj = re.search( - r'(?s)<h1[^>]+class="topmost">(?P<title>[^<]+)</h1>\s*<div>(?P<description>.+?)</div>', - login_page) - if mobj: - raise ExtractorError( - 'lynda returned error: %s - %s' - % (mobj.group('title'), clean_html(mobj.group('description'))), - expected=True) - raise ExtractorError('Unable to log in') - - def _logout(self): - username, _ = self._get_login_info() - if username is None: + # Step 1: download signin page + signin_page = self._download_webpage( + self._SIGNIN_URL, None, 'Downloading signin page') + + # Already logged in + if any(re.search(p, signin_page) for p in ( + 'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')): return - self._download_webpage( - 'http://www.lynda.com/ajax/logout.aspx', None, - 'Logging out', 'Unable to log out', fatal=False) + # Step 2: submit email + signin_form = self._search_regex( + r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)', + signin_page, 'signin form') + signin_page, signin_url = self._login_step( + signin_form, self._PASSWORD_URL, {'email': username}, + 'Submitting email', self._SIGNIN_URL) + + # Step 3: submit password + password_form = signin_page['body'] + self._login_step( + password_form, self._USER_URL, {'email': username, 'password': password}, + 'Submitting password', signin_url) class LyndaIE(LyndaBaseIE): IE_NAME = 'lynda' IE_DESC = 'lynda.com videos' _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(?P<id>\d+)' - _NETRC_MACHINE = 'lynda' _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]' _TESTS = [{ 'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', - 'md5': 'ecfc6862da89489161fb9cd5f5a6fac1', + 'md5': '679734f6786145da3546585de9a356be', 'info_dict': { 'id': '114408', 'ext': 'mp4', @@ -212,8 +212,6 @@ class LyndaCourseIE(LyndaBaseIE): 'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, course_id, 'Downloading course JSON') - self._logout() - if course.get('Status') == 'NotFound': raise ExtractorError( 'Course %s does not exist' % course_id, expected=True) @@ -246,5 +244,6 @@ class LyndaCourseIE(LyndaBaseIE): % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT) course_title = course.get('Title') + course_description = course.get('Description') - return self.playlist_result(entries, course_id, course_title) + return self.playlist_result(entries, course_id, course_title, course_description) diff --git a/youtube_dl/extractor/m6.py b/youtube_dl/extractor/m6.py index d5945ad66..39d2742c8 100644 --- a/youtube_dl/extractor/m6.py +++ b/youtube_dl/extractor/m6.py @@ -1,8 +1,6 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor @@ -23,34 +21,5 @@ class M6IE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - rss = self._download_xml('http://ws.m6.fr/v1/video/info/m6/bonus/%s' % video_id, video_id, - 'Downloading video RSS') - - title = rss.find('./channel/item/title').text - description = rss.find('./channel/item/description').text - thumbnail = rss.find('./channel/item/visuel_clip_big').text - duration = int(rss.find('./channel/item/duration').text) - view_count = int(rss.find('./channel/item/nombre_vues').text) - - formats = [] - for format_id in ['lq', 'sd', 'hq', 'hd']: - video_url = rss.find('./channel/item/url_video_%s' % format_id) - if video_url is None: - continue - formats.append({ - 'url': video_url.text, - 'format_id': format_id, - }) - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'view_count': view_count, - 'formats': formats, - } + video_id = self._match_id(url) + return self.url_result('6play:%s' % video_id, 'SixPlay', video_id) diff --git a/youtube_dl/extractor/malemotion.py b/youtube_dl/extractor/malemotion.py deleted file mode 100644 index 92511a671..000000000 --- a/youtube_dl/extractor/malemotion.py +++ /dev/null @@ -1,46 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote - - -class MalemotionIE(InfoExtractor): - _VALID_URL = r'https?://malemotion\.com/video/(.+?)\.(?P<id>.+?)(#|$)' - _TEST = { - 'url': 'http://malemotion.com/video/bete-de-concours.ltc', - 'md5': '3013e53a0afbde2878bc39998c33e8a5', - 'info_dict': { - 'id': 'ltc', - 'ext': 'mp4', - 'title': 'Bête de Concours', - 'age_limit': 18, - }, - } - - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - video_url = compat_urllib_parse_unquote(self._search_regex( - r'<source type="video/mp4" src="(.+?)"', webpage, 'video URL')) - video_title = self._html_search_regex( - r'<title>(.*?)</title', webpage, 'title') - video_thumbnail = self._search_regex( - r'<video .+?poster="(.+?)"', webpage, 'thumbnail', fatal=False) - - formats = [{ - 'url': video_url, - 'ext': 'mp4', - 'format_id': 'mp4', - 'preference': 1, - }] - self._sort_formats(formats) - - return { - 'id': video_id, - 'formats': formats, - 'title': video_title, - 'thumbnail': video_thumbnail, - 'age_limit': 18, - } diff --git a/youtube_dl/extractor/matchtv.py b/youtube_dl/extractor/matchtv.py index 80a0d7013..33b0b539f 100644 --- a/youtube_dl/extractor/matchtv.py +++ b/youtube_dl/extractor/matchtv.py @@ -4,16 +4,12 @@ from __future__ import unicode_literals import random from .common import InfoExtractor -from ..compat import compat_urllib_parse_urlencode -from ..utils import ( - sanitized_Request, - xpath_text, -) +from ..utils import xpath_text class MatchTVIE(InfoExtractor): - _VALID_URL = r'https?://matchtv\.ru/?#live-player' - _TEST = { + _VALID_URL = r'https?://matchtv\.ru(?:/on-air|/?#live-player)' + _TESTS = [{ 'url': 'http://matchtv.ru/#live-player', 'info_dict': { 'id': 'matchtv-live', @@ -24,12 +20,16 @@ class MatchTVIE(InfoExtractor): 'params': { 'skip_download': True, }, - } + }, { + 'url': 'http://matchtv.ru/on-air/', + 'only_matching': True, + }] def _real_extract(self, url): video_id = 'matchtv-live' - request = sanitized_Request( - 'http://player.matchtv.ntvplus.tv/player/smil?%s' % compat_urllib_parse_urlencode({ + video_url = self._download_json( + 'http://player.matchtv.ntvplus.tv/player/smil', video_id, + query={ 'ts': '', 'quality': 'SD', 'contentId': '561d2c0df7159b37178b4567', @@ -40,11 +40,10 @@ class MatchTVIE(InfoExtractor): 'contentType': 'channel', 'timeShift': '0', 'platform': 'portal', - }), + }, headers={ 'Referer': 'http://player.matchtv.ntvplus.tv/embed-player/NTVEmbedPlayer.swf', - }) - video_url = self._download_json(request, video_id)['data']['videoUrl'] + })['data']['videoUrl'] f4m_url = xpath_text(self._download_xml(video_url, video_id), './to') formats = self._extract_f4m_formats(f4m_url, video_id) self._sort_formats(formats) diff --git a/youtube_dl/extractor/meta.py b/youtube_dl/extractor/meta.py new file mode 100644 index 000000000..cdb46e163 --- /dev/null +++ b/youtube_dl/extractor/meta.py @@ -0,0 +1,73 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from .pladform import PladformIE +from ..utils import ( + unescapeHTML, + int_or_none, + ExtractorError, +) + + +class METAIE(InfoExtractor): + _VALID_URL = r'https?://video\.meta\.ua/(?:iframe/)?(?P<id>[0-9]+)' + _TESTS = [{ + 'url': 'http://video.meta.ua/5502115.video', + 'md5': '71b6f3ee274bef16f1ab410f7f56b476', + 'info_dict': { + 'id': '5502115', + 'ext': 'mp4', + 'title': 'Sony Xperia Z camera test [HQ]', + 'description': 'Xperia Z shoots video in FullHD HDR.', + 'uploader_id': 'nomobile', + 'uploader': 'CHЁZA.TV', + 'upload_date': '20130211', + }, + 'add_ie': ['Youtube'], + }, { + 'url': 'http://video.meta.ua/iframe/5502115', + 'only_matching': True, + }, { + # pladform embed + 'url': 'http://video.meta.ua/7121015.video', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + st_html5 = self._search_regex( + r"st_html5\s*=\s*'#([^']+)'", webpage, 'uppod html5 st', default=None) + + if st_html5: + # uppod st decryption algorithm is reverse engineered from function un(s) at uppod.js + json_str = '' + for i in range(0, len(st_html5), 3): + json_str += '�%s;' % st_html5[i:i + 3] + uppod_data = self._parse_json(unescapeHTML(json_str), video_id) + error = uppod_data.get('customnotfound') + if error: + raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True) + + video_url = uppod_data['file'] + info = { + 'id': video_id, + 'url': video_url, + 'title': uppod_data.get('comment') or self._og_search_title(webpage), + 'description': self._og_search_description(webpage, default=None), + 'thumbnail': uppod_data.get('poster') or self._og_search_thumbnail(webpage), + 'duration': int_or_none(self._og_search_property( + 'video:duration', webpage, default=None)), + } + if 'youtube.com/' in video_url: + info.update({ + '_type': 'url_transparent', + 'ie_key': 'Youtube', + }) + return info + + pladform_url = PladformIE._extract_url(webpage) + if pladform_url: + return self.url_result(pladform_url) diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py index 61dadb7a7..e6e7659a1 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -11,13 +11,14 @@ from ..utils import ( determine_ext, ExtractorError, int_or_none, - sanitized_Request, urlencode_postdata, + get_element_by_attribute, + mimetype2ext, ) class MetacafeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*' + _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/(?P<video_id>[^/]+)/(?P<display_id>[^/?#]+)' _DISCLAIMER = 'http://www.metacafe.com/family_filter/' _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user' IE_NAME = 'metacafe' @@ -47,6 +48,7 @@ class MetacafeIE(InfoExtractor): 'uploader': 'ign', 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.', }, + 'skip': 'Page is temporarily unavailable.', }, # AnyClip video { @@ -55,8 +57,8 @@ class MetacafeIE(InfoExtractor): 'id': 'an-dVVXnuY7Jh77J', 'ext': 'mp4', 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3', - 'uploader': 'anyclip', - 'description': 'md5:38c711dd98f5bb87acf973d573442e67', + 'uploader': 'AnyClip', + 'description': 'md5:cbef0460d31e3807f6feb4e7a5952e5b', }, }, # age-restricted video @@ -81,6 +83,9 @@ class MetacafeIE(InfoExtractor): 'title': 'Open: This is Face the Nation, February 9', 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476', 'duration': 96, + 'uploader': 'CBSI-NEW', + 'upload_date': '20140209', + 'timestamp': 1391959800, }, 'params': { # rtmp download @@ -107,28 +112,25 @@ class MetacafeIE(InfoExtractor): def report_disclaimer(self): self.to_screen('Retrieving disclaimer') - def _real_initialize(self): + def _confirm_age(self): # Retrieve disclaimer self.report_disclaimer() self._download_webpage(self._DISCLAIMER, None, False, 'Unable to retrieve disclaimer') # Confirm age - disclaimer_form = { - 'filters': '0', - 'submit': "Continue - I'm over 18", - } - request = sanitized_Request(self._FILTER_POST, urlencode_postdata(disclaimer_form)) - request.add_header('Content-Type', 'application/x-www-form-urlencoded') self.report_age_confirmation() - self._download_webpage(request, None, False, 'Unable to confirm age') + self._download_webpage( + self._FILTER_POST, None, False, 'Unable to confirm age', + data=urlencode_postdata({ + 'filters': '0', + 'submit': "Continue - I'm over 18", + }), headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + }) def _real_extract(self, url): # Extract id and simplified title from URL - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError('Invalid URL: %s' % url) - - video_id = mobj.group(1) + video_id, display_id = re.match(self._VALID_URL, url).groups() # the video may come from an external site m_external = re.match('^(\w{2})-(.*)$', video_id) @@ -141,15 +143,24 @@ class MetacafeIE(InfoExtractor): if prefix == 'cb': return self.url_result('theplatform:%s' % ext_id, 'ThePlatform') - # Retrieve video webpage to extract further information - req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id) + # self._confirm_age() # AnyClip videos require the flashversion cookie so that we get the link # to the mp4 file - mobj_an = re.match(r'^an-(.*?)$', video_id) - if mobj_an: - req.headers['Cookie'] = 'flashVersion=0;' - webpage = self._download_webpage(req, video_id) + headers = {} + if video_id.startswith('an-'): + headers['Cookie'] = 'flashVersion=0;' + + # Retrieve video webpage to extract further information + webpage = self._download_webpage(url, video_id, headers=headers) + + error = get_element_by_attribute( + 'class', 'notfound-page-title', webpage) + if error: + raise ExtractorError(error, expected=True) + + video_title = self._html_search_meta( + ['og:title', 'twitter:title'], webpage, 'title', default=None) or self._search_regex(r'<h1>(.*?)</h1>', webpage, 'title') # Extract URL, uploader and title from webpage self.report_extraction(video_id) @@ -213,20 +224,40 @@ class MetacafeIE(InfoExtractor): 'player_url': player_url, 'ext': play_path.partition(':')[0], }) + if video_url is None: + flashvars = self._parse_json(self._search_regex( + r'flashvars\s*=\s*({.*});', webpage, 'flashvars', + default=None), video_id, fatal=False) + if flashvars: + video_url = [] + for source in flashvars.get('sources'): + source_url = source.get('src') + if not source_url: + continue + ext = mimetype2ext(source.get('type')) or determine_ext(source_url) + if ext == 'm3u8': + video_url.extend(self._extract_m3u8_formats( + source_url, video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + else: + video_url.append({ + 'url': source_url, + 'ext': ext, + }) if video_url is None: raise ExtractorError('Unsupported video type') - video_title = self._html_search_regex( - r'(?im)<title>(.*) - Video</title>', webpage, 'title') - description = self._og_search_description(webpage) - thumbnail = self._og_search_thumbnail(webpage) + description = self._html_search_meta( + ['og:description', 'twitter:description', 'description'], + webpage, 'title', fatal=False) + thumbnail = self._html_search_meta( + ['og:image', 'twitter:image'], webpage, 'title', fatal=False) video_uploader = self._html_search_regex( r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);', webpage, 'uploader nickname', fatal=False) duration = int_or_none( - self._html_search_meta('video:duration', webpage)) - + self._html_search_meta('video:duration', webpage, default=None)) age_limit = ( 18 if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage) @@ -239,10 +270,11 @@ class MetacafeIE(InfoExtractor): 'url': video_url, 'ext': video_ext, }] - self._sort_formats(formats) + return { 'id': video_id, + 'display_id': display_id, 'description': description, 'uploader': video_uploader, 'title': video_title, diff --git a/youtube_dl/extractor/mgtv.py b/youtube_dl/extractor/mgtv.py index a14d176a5..d970e94ec 100644 --- a/youtube_dl/extractor/mgtv.py +++ b/youtube_dl/extractor/mgtv.py @@ -11,7 +11,7 @@ class MGTVIE(InfoExtractor): _TEST = { 'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html', - 'md5': '', + 'md5': '1bdadcf760a0b90946ca68ee9a2db41a', 'info_dict': { 'id': '3116640', 'ext': 'mp4', @@ -20,37 +20,39 @@ class MGTVIE(InfoExtractor): 'duration': 7461, 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - 'skip_download': True, # m3u8 download - }, - } - - _FORMAT_MAP = { - '标清': ('Standard', 0), - '高清': ('High', 1), - '超清': ('SuperHigh', 2), } def _real_extract(self, url): video_id = self._match_id(url) api_data = self._download_json( 'http://v.api.mgtv.com/player/video', video_id, - query={'video_id': video_id})['data'] + query={'video_id': video_id}, + headers=self.geo_verification_headers())['data'] info = api_data['info'] formats = [] for idx, stream in enumerate(api_data['stream']): - format_name = stream.get('name') - format_id, preference = self._FORMAT_MAP.get(format_name, (None, None)) - format_info = self._download_json( - stream['url'], video_id, - note='Download video info for format %s' % format_id or '#%d' % idx) - formats.append({ - 'format_id': format_id, - 'url': format_info['info'], - 'ext': 'mp4', # These are m3u8 playlists - 'preference': preference, - }) + stream_url = stream.get('url') + if not stream_url: + continue + tbr = int_or_none(self._search_regex( + r'(\d+)\.mp4', stream_url, 'tbr', default=None)) + + def extract_format(stream_url, format_id, idx, query={}): + format_info = self._download_json( + stream_url, video_id, + note='Download video info for format %s' % format_id or '#%d' % idx, query=query) + return { + 'format_id': format_id, + 'url': format_info['info'], + 'ext': 'mp4', + 'tbr': tbr, + } + + formats.append(extract_format( + stream_url, 'hls-%d' % tbr if tbr else None, idx * 2)) + formats.append(extract_format(stream_url.replace( + '/playlist.m3u8', ''), 'http-%d' % tbr if tbr else None, idx * 2 + 1, {'pno': 1031})) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/microsoftvirtualacademy.py b/youtube_dl/extractor/microsoftvirtualacademy.py new file mode 100644 index 000000000..afd3e98ec --- /dev/null +++ b/youtube_dl/extractor/microsoftvirtualacademy.py @@ -0,0 +1,192 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_xpath, +) +from ..utils import ( + int_or_none, + parse_duration, + smuggle_url, + unsmuggle_url, + xpath_text, +) + + +class MicrosoftVirtualAcademyBaseIE(InfoExtractor): + def _extract_base_url(self, course_id, display_id): + return self._download_json( + 'https://api-mlxprod.microsoft.com/services/products/anonymous/%s' % course_id, + display_id, 'Downloading course base URL') + + def _extract_chapter_and_title(self, title): + if not title: + return None, None + m = re.search(r'(?P<chapter>\d+)\s*\|\s*(?P<title>.+)', title) + return (int(m.group('chapter')), m.group('title')) if m else (None, title) + + +class MicrosoftVirtualAcademyIE(MicrosoftVirtualAcademyBaseIE): + IE_NAME = 'mva' + IE_DESC = 'Microsoft Virtual Academy videos' + _VALID_URL = r'(?:%s:|https?://(?:mva\.microsoft|(?:www\.)?microsoftvirtualacademy)\.com/[^/]+/training-courses/[^/?#&]+-)(?P<course_id>\d+)(?::|\?l=)(?P<id>[\da-zA-Z]+_\d+)' % IE_NAME + + _TESTS = [{ + 'url': 'https://mva.microsoft.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788?l=gfVXISmEB_6804984382', + 'md5': '7826c44fc31678b12ad8db11f6b5abb9', + 'info_dict': { + 'id': 'gfVXISmEB_6804984382', + 'ext': 'mp4', + 'title': 'Course Introduction', + 'formats': 'mincount:3', + 'subtitles': { + 'en': [{ + 'ext': 'ttml', + }], + }, + } + }, { + 'url': 'mva:11788:gfVXISmEB_6804984382', + 'only_matching': True, + }] + + def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + + mobj = re.match(self._VALID_URL, url) + course_id = mobj.group('course_id') + video_id = mobj.group('id') + + base_url = smuggled_data.get('base_url') or self._extract_base_url(course_id, video_id) + + settings = self._download_xml( + '%s/content/content_%s/videosettings.xml?v=1' % (base_url, video_id), + video_id, 'Downloading video settings XML') + + _, title = self._extract_chapter_and_title(xpath_text( + settings, './/Title', 'title', fatal=True)) + + formats = [] + + for sources in settings.findall(compat_xpath('.//MediaSources')): + if sources.get('videoType') == 'smoothstreaming': + continue + for source in sources.findall(compat_xpath('./MediaSource')): + video_url = source.text + if not video_url or not video_url.startswith('http'): + continue + video_mode = source.get('videoMode') + height = int_or_none(self._search_regex( + r'^(\d+)[pP]$', video_mode or '', 'height', default=None)) + codec = source.get('codec') + acodec, vcodec = [None] * 2 + if codec: + codecs = codec.split(',') + if len(codecs) == 2: + acodec, vcodec = codecs + elif len(codecs) == 1: + vcodec = codecs[0] + formats.append({ + 'url': video_url, + 'format_id': video_mode, + 'height': height, + 'acodec': acodec, + 'vcodec': vcodec, + }) + self._sort_formats(formats) + + subtitles = {} + for source in settings.findall(compat_xpath('.//MarkerResourceSource')): + subtitle_url = source.text + if not subtitle_url: + continue + subtitles.setdefault('en', []).append({ + 'url': '%s/%s' % (base_url, subtitle_url), + 'ext': source.get('type'), + }) + + return { + 'id': video_id, + 'title': title, + 'subtitles': subtitles, + 'formats': formats + } + + +class MicrosoftVirtualAcademyCourseIE(MicrosoftVirtualAcademyBaseIE): + IE_NAME = 'mva:course' + IE_DESC = 'Microsoft Virtual Academy courses' + _VALID_URL = r'(?:%s:|https?://(?:mva\.microsoft|(?:www\.)?microsoftvirtualacademy)\.com/[^/]+/training-courses/(?P<display_id>[^/?#&]+)-)(?P<id>\d+)' % IE_NAME + + _TESTS = [{ + 'url': 'https://mva.microsoft.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788', + 'info_dict': { + 'id': '11788', + 'title': 'Microsoft Azure Fundamentals: Virtual Machines', + }, + 'playlist_count': 36, + }, { + # with emphasized chapters + 'url': 'https://mva.microsoft.com/en-US/training-courses/developing-windows-10-games-with-construct-2-16335', + 'info_dict': { + 'id': '16335', + 'title': 'Developing Windows 10 Games with Construct 2', + }, + 'playlist_count': 10, + }, { + 'url': 'https://www.microsoftvirtualacademy.com/en-US/training-courses/microsoft-azure-fundamentals-virtual-machines-11788', + 'only_matching': True, + }, { + 'url': 'mva:course:11788', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if MicrosoftVirtualAcademyIE.suitable(url) else super( + MicrosoftVirtualAcademyCourseIE, cls).suitable(url) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + course_id = mobj.group('id') + display_id = mobj.group('display_id') + + base_url = self._extract_base_url(course_id, display_id) + + manifest = self._download_json( + '%s/imsmanifestlite.json' % base_url, + display_id, 'Downloading course manifest JSON')['manifest'] + + organization = manifest['organizations']['organization'][0] + + entries = [] + for chapter in organization['item']: + chapter_number, chapter_title = self._extract_chapter_and_title(chapter.get('title')) + chapter_id = chapter.get('@identifier') + for item in chapter.get('item', []): + item_id = item.get('@identifier') + if not item_id: + continue + metadata = item.get('resource', {}).get('metadata') or {} + if metadata.get('learningresourcetype') != 'Video': + continue + _, title = self._extract_chapter_and_title(item.get('title')) + duration = parse_duration(metadata.get('duration')) + description = metadata.get('description') + entries.append({ + '_type': 'url_transparent', + 'url': smuggle_url( + 'mva:%s:%s' % (course_id, item_id), {'base_url': base_url}), + 'title': title, + 'description': description, + 'duration': duration, + 'chapter': chapter_title, + 'chapter_number': chapter_number, + 'chapter_id': chapter_id, + }) + + title = organization.get('title') or manifest.get('metadata', {}).get('title') + + return self.playlist_result(entries, course_id, title) diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py index 7b4581dc5..cd169f361 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -1,5 +1,8 @@ +# coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..compat import ( compat_urllib_parse_urlencode, @@ -8,25 +11,104 @@ from ..compat import ( from ..utils import ( get_element_by_attribute, int_or_none, + remove_start, + extract_attributes, + determine_ext, ) -class MiTeleIE(InfoExtractor): +class MiTeleBaseIE(InfoExtractor): + def _get_player_info(self, url, webpage): + player_data = extract_attributes(self._search_regex( + r'(?s)(<ms-video-player.+?</ms-video-player>)', + webpage, 'ms video player')) + video_id = player_data['data-media-id'] + config_url = compat_urlparse.urljoin(url, player_data['data-config']) + config = self._download_json( + config_url, video_id, 'Downloading config JSON') + mmc_url = config['services']['mmc'] + + duration = None + formats = [] + for m_url in (mmc_url, mmc_url.replace('/flash.json', '/html5.json')): + mmc = self._download_json( + m_url, video_id, 'Downloading mmc JSON') + if not duration: + duration = int_or_none(mmc.get('duration')) + for location in mmc['locations']: + gat = self._proto_relative_url(location.get('gat'), 'http:') + bas = location.get('bas') + loc = location.get('loc') + ogn = location.get('ogn') + if None in (gat, bas, loc, ogn): + continue + token_data = { + 'bas': bas, + 'icd': loc, + 'ogn': ogn, + 'sta': '0', + } + media = self._download_json( + '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)), + video_id, 'Downloading %s JSON' % location['loc']) + file_ = media.get('file') + if not file_: + continue + ext = determine_ext(file_) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats( + file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18', + video_id, f4m_id='hds', fatal=False)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + self._sort_formats(formats) + + return { + 'id': video_id, + 'formats': formats, + 'thumbnail': player_data.get('data-poster') or config.get('poster', {}).get('imageUrl'), + 'duration': duration, + } + + +class MiTeleIE(MiTeleBaseIE): IE_DESC = 'mitele.es' - _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/' + _VALID_URL = r'https?://www\.mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/' _TESTS = [{ 'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', - 'md5': '0ff1a13aebb35d9bc14081ff633dd324', + # MD5 is unstable 'info_dict': { 'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'display_id': 'programa-144', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Tor, la web invisible', 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', + 'series': 'Diario de', + 'season': 'La redacción', + 'episode': 'Programa 144', 'thumbnail': 're:(?i)^https?://.*\.jpg$', 'duration': 2913, }, + }, { + # no explicit title + 'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/temporada-6/programa-226/', + 'info_dict': { + 'id': 'eLZSwoEd1S3pVyUm8lc6F', + 'display_id': 'programa-226', + 'ext': 'mp4', + 'title': 'Cuarto Milenio - Temporada 6 - Programa 226', + 'description': 'md5:50daf9fadefa4e62d9fc866d0c015701', + 'series': 'Cuarto Milenio', + 'season': 'Temporada 6', + 'episode': 'Programa 226', + 'thumbnail': 're:(?i)^https?://.*\.jpg$', + 'duration': 7312, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -34,56 +116,32 @@ class MiTeleIE(InfoExtractor): webpage = self._download_webpage(url, display_id) - config_url = self._search_regex( - r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url') - config_url = compat_urlparse.urljoin(url, config_url) - - config = self._download_json( - config_url, display_id, 'Downloading config JSON') - - mmc = self._download_json( - config['services']['mmc'], display_id, 'Downloading mmc JSON') - - formats = [] - for location in mmc['locations']: - gat = self._proto_relative_url(location.get('gat'), 'http:') - bas = location.get('bas') - loc = location.get('loc') - ogn = location.get('ogn') - if None in (gat, bas, loc, ogn): - continue - token_data = { - 'bas': bas, - 'icd': loc, - 'ogn': ogn, - 'sta': '0', - } - media = self._download_json( - '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)), - display_id, 'Downloading %s JSON' % location['loc']) - file_ = media.get('file') - if not file_: - continue - formats.extend(self._extract_f4m_formats( - file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18', - display_id, f4m_id=loc)) - self._sort_formats(formats) + info = self._get_player_info(url, webpage) title = self._search_regex( - r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title') + r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', + webpage, 'title', default=None) - video_id = self._search_regex( - r'data-media-id\s*=\s*"([^"]+)"', webpage, - 'data media id', default=None) or display_id - thumbnail = config.get('poster', {}).get('imageUrl') - duration = int_or_none(mmc.get('duration')) + mobj = re.search(r'''(?sx) + class="Destacado-text"[^>]*>.*?<h1>\s* + <span>(?P<series>[^<]+)</span>\s* + <span>(?P<season>[^<]+)</span>\s* + <span>(?P<episode>[^<]+)</span>''', webpage) + series, season, episode = mobj.groups() if mobj else [None] * 3 - return { - 'id': video_id, + if not title: + if mobj: + title = '%s - %s - %s' % (series, season, episode) + else: + title = remove_start(self._search_regex( + r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ') + + info.update({ 'display_id': display_id, 'title': title, 'description': get_element_by_attribute('class', 'text', webpage), - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - } + 'series': series, + 'season': season, + 'episode': episode, + }) + return info diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 483f6925f..560fe188b 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -102,11 +102,11 @@ class MixcloudIE(InfoExtractor): description = self._og_search_description(webpage) like_count = parse_count(self._search_regex( r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)', - webpage, 'like count', fatal=False)) + webpage, 'like count', default=None)) view_count = str_to_int(self._search_regex( [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"', r'/listeners/?">([0-9,.]+)</a>'], - webpage, 'play count', fatal=False)) + webpage, 'play count', default=None)) return { 'id': track_id, diff --git a/youtube_dl/extractor/msn.py b/youtube_dl/extractor/msn.py new file mode 100644 index 000000000..1ec8e0f50 --- /dev/null +++ b/youtube_dl/extractor/msn.py @@ -0,0 +1,122 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + determine_ext, + ExtractorError, + int_or_none, + unescapeHTML, +) + + +class MSNIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)' + _TESTS = [{ + 'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/criminal-minds-shemar-moore-shares-a-touching-goodbye-message/vp-BBqQYNE', + 'md5': '8442f66c116cbab1ff7098f986983458', + 'info_dict': { + 'id': 'BBqQYNE', + 'display_id': 'criminal-minds-shemar-moore-shares-a-touching-goodbye-message', + 'ext': 'mp4', + 'title': 'Criminal Minds - Shemar Moore Shares A Touching Goodbye Message', + 'description': 'md5:e8e89b897b222eb33a6b5067a8f1bc25', + 'duration': 104, + 'uploader': 'CBS Entertainment', + 'uploader_id': 'IT0X5aoJ6bJgYerJXSDCgFmYPB1__54v', + }, + }, { + 'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf', + 'only_matching': True, + }, { + 'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH', + 'only_matching': True, + }, { + # geo restricted + 'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU', + 'only_matching': True, + }, { + 'url': 'http://www.msn.com/en-ae/entertainment/bollywood/watch-how-salman-khan-reacted-when-asked-if-he-would-apologize-for-his-‘raped-woman’-comment/vi-AAhvzW6', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id, display_id = mobj.group('id', 'display_id') + + webpage = self._download_webpage(url, display_id) + + video = self._parse_json( + self._search_regex( + r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1', + webpage, 'video data', default='{}', group='data'), + display_id, transform_source=unescapeHTML) + + if not video: + error = unescapeHTML(self._search_regex( + r'data-error=(["\'])(?P<error>.+?)\1', + webpage, 'error', group='error')) + raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True) + + title = video['title'] + + formats = [] + for file_ in video.get('videoFiles', []): + format_url = file_.get('url') + if not format_url: + continue + ext = determine_ext(format_url) + # .ism is not yet supported (see + # https://github.com/rg3/youtube-dl/issues/8118) + if ext == 'ism': + continue + if 'm3u8' in format_url: + # m3u8_native should not be used here until + # https://github.com/rg3/youtube-dl/issues/9913 is fixed + m3u8_formats = self._extract_m3u8_formats( + format_url, display_id, 'mp4', + m3u8_id='hls', fatal=False) + # Despite metadata in m3u8 all video+audio formats are + # actually video-only (no audio) + for f in m3u8_formats: + if f.get('acodec') != 'none' and f.get('vcodec') != 'none': + f['acodec'] = 'none' + formats.extend(m3u8_formats) + else: + formats.append({ + 'url': format_url, + 'ext': 'mp4', + 'format_id': 'http', + 'width': int_or_none(file_.get('width')), + 'height': int_or_none(file_.get('height')), + }) + self._sort_formats(formats) + + subtitles = {} + for file_ in video.get('files', []): + format_url = file_.get('url') + format_code = file_.get('formatCode') + if not format_url or not format_code: + continue + if compat_str(format_code) == '3100': + subtitles.setdefault(file_.get('culture', 'en'), []).append({ + 'ext': determine_ext(format_url, 'ttml'), + 'url': format_url, + }) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'description': video.get('description'), + 'thumbnail': video.get('headlineImage', {}).get('url'), + 'duration': int_or_none(video.get('durationSecs')), + 'uploader': video.get('sourceFriendly'), + 'uploader_id': video.get('providerId'), + 'creator': video.get('creator'), + 'subtitles': subtitles, + 'formats': formats, + } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 640ee3d93..dd0639589 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -6,6 +6,7 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse_urlencode, compat_str, + compat_xpath, ) from ..utils import ( ExtractorError, @@ -84,9 +85,10 @@ class MTVServicesInfoExtractor(InfoExtractor): rtmp_video_url = rendition.find('./src').text if rtmp_video_url.endswith('siteunavail.png'): continue + new_url = self._transform_rtmp_url(rtmp_video_url) formats.append({ - 'ext': ext, - 'url': self._transform_rtmp_url(rtmp_video_url), + 'ext': 'flv' if new_url.startswith('rtmp') else ext, + 'url': new_url, 'format_id': rendition.get('bitrate'), 'width': int(rendition.get('width')), 'height': int(rendition.get('height')), @@ -139,9 +141,9 @@ class MTVServicesInfoExtractor(InfoExtractor): itemdoc, './/{http://search.yahoo.com/mrss/}category', 'scheme', 'urn:mtvn:video_title') if title_el is None: - title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title') + title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title')) if title_el is None: - title_el = itemdoc.find('.//title') or itemdoc.find('./title') + title_el = itemdoc.find(compat_xpath('.//title')) if title_el.text is None: title_el = None diff --git a/youtube_dl/extractor/muzu.py b/youtube_dl/extractor/muzu.py deleted file mode 100644 index cbc800481..000000000 --- a/youtube_dl/extractor/muzu.py +++ /dev/null @@ -1,63 +0,0 @@ -from __future__ import unicode_literals - -from .common import InfoExtractor -from ..compat import compat_urllib_parse_urlencode - - -class MuzuTVIE(InfoExtractor): - _VALID_URL = r'https?://www\.muzu\.tv/(.+?)/(.+?)/(?P<id>\d+)' - IE_NAME = 'muzu.tv' - - _TEST = { - 'url': 'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/', - 'md5': '98f8b2c7bc50578d6a0364fff2bfb000', - 'info_dict': { - 'id': '1981454', - 'ext': 'mp4', - 'title': 'Cat Walk (Original Mix)', - 'description': 'md5:90e868994de201b2570e4e5854e19420', - 'uploader': 'MarcAshken featuring SOS', - }, - } - - def _real_extract(self, url): - video_id = self._match_id(url) - - info_data = compat_urllib_parse_urlencode({ - 'format': 'json', - 'url': url, - }) - info = self._download_json( - 'http://www.muzu.tv/api/oembed/?%s' % info_data, - video_id, 'Downloading video info') - - player_info = self._download_json( - 'http://player.muzu.tv/player/playerInit?ai=%s' % video_id, - video_id, 'Downloading player info') - video_info = player_info['videos'][0] - for quality in ['1080', '720', '480', '360']: - if video_info.get('v%s' % quality): - break - - data = compat_urllib_parse_urlencode({ - 'ai': video_id, - # Even if each time you watch a video the hash changes, - # it seems to work for different videos, and it will work - # even if you use any non empty string as a hash - 'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k', - 'device': 'web', - 'qv': quality, - }) - video_url_info = self._download_json( - 'http://player.muzu.tv/player/requestVideo?%s' % data, - video_id, 'Downloading video url') - video_url = video_url_info['url'] - - return { - 'id': video_id, - 'title': info['title'], - 'url': video_url, - 'thumbnail': info['thumbnail_url'], - 'description': info['description'], - 'uploader': info['author_name'], - } diff --git a/youtube_dl/extractor/mwave.py b/youtube_dl/extractor/mwave.py index 66b523197..a103e0323 100644 --- a/youtube_dl/extractor/mwave.py +++ b/youtube_dl/extractor/mwave.py @@ -10,9 +10,10 @@ from ..utils import ( class MwaveIE(InfoExtractor): _VALID_URL = r'https?://mwave\.interest\.me/mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)' + _URL_TEMPLATE = 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=%s' _TEST = { 'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859', - 'md5': 'c930e27b7720aaa3c9d0018dfc8ff6cc', + # md5 is unstable 'info_dict': { 'id': '168859', 'ext': 'flv', @@ -56,3 +57,28 @@ class MwaveIE(InfoExtractor): 'view_count': int_or_none(vod_info.get('hit')), 'formats': formats, } + + +class MwaveMeetGreetIE(InfoExtractor): + _VALID_URL = r'https?://mwave\.interest\.me/meetgreet/view/(?P<id>\d+)' + _TEST = { + 'url': 'http://mwave.interest.me/meetgreet/view/256', + 'info_dict': { + 'id': '173294', + 'ext': 'flv', + 'title': '[MEET&GREET] Park BoRam', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Mwave', + 'duration': 3634, + 'view_count': int, + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + clip_id = self._html_search_regex( + r'<iframe[^>]+src="/mnettv/ifr_clip\.m\?searchVideoDetailVO\.clip_id=(\d+)', + webpage, 'clip ID') + clip_url = MwaveIE._URL_TEMPLATE % clip_id + return self.url_result(clip_url, 'Mwave', clip_id) diff --git a/youtube_dl/extractor/nationalgeographic.py b/youtube_dl/extractor/nationalgeographic.py index 722518663..e717abb9f 100644 --- a/youtube_dl/extractor/nationalgeographic.py +++ b/youtube_dl/extractor/nationalgeographic.py @@ -1,6 +1,7 @@ from __future__ import unicode_literals from .common import InfoExtractor +from .theplatform import ThePlatformIE from ..utils import ( smuggle_url, url_basename, @@ -61,7 +62,7 @@ class NationalGeographicIE(InfoExtractor): } -class NationalGeographicChannelIE(InfoExtractor): +class NationalGeographicChannelIE(ThePlatformIE): IE_NAME = 'natgeo:channel' _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/videos/(?P<id>[^/?]+)' @@ -102,12 +103,22 @@ class NationalGeographicChannelIE(InfoExtractor): release_url = self._search_regex( r'video_auth_playlist_url\s*=\s*"([^"]+)"', webpage, 'release url') + query = { + 'mbr': 'true', + 'switch': 'http', + } + is_auth = self._search_regex(r'video_is_auth\s*=\s*"([^"]+)"', webpage, 'is auth', fatal=False) + if is_auth == 'auth': + auth_resource_id = self._search_regex( + r"video_auth_resourceId\s*=\s*'([^']+)'", + webpage, 'auth resource id') + query['auth'] = self._extract_mvpd_auth(url, display_id, 'natgeo', auth_resource_id) or '' return { '_type': 'url_transparent', 'ie_key': 'ThePlatform', 'url': smuggle_url( - update_url_query(release_url, {'mbr': 'true', 'switch': 'http'}), + update_url_query(release_url, query), {'force_smil_url': True}), 'display_id': display_id, } diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index f9d42d07a..f694e210b 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -9,10 +9,6 @@ from ..utils import ( lowercase_escape, smuggle_url, unescapeHTML, - update_url_query, - int_or_none, - HEADRequest, - parse_iso8601, ) @@ -67,6 +63,23 @@ class NBCIE(InfoExtractor): # This video has expired but with an escaped embedURL 'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515', 'only_matching': True, + }, + { + # HLS streams requires the 'hdnea3' cookie + 'url': 'http://www.nbc.com/Kings/video/goliath/n1806', + 'info_dict': { + 'id': 'n1806', + 'ext': 'mp4', + 'title': 'Goliath', + 'description': 'When an unknown soldier saves the life of the King\'s son in battle, he\'s thrust into the limelight and politics of the kingdom.', + 'timestamp': 1237100400, + 'upload_date': '20090315', + 'uploader': 'NBCU-COM', + }, + 'params': { + 'skip_download': True, + }, + 'skip': 'Only works from US', } ] @@ -134,6 +147,9 @@ class NBCSportsIE(InfoExtractor): 'ext': 'flv', 'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke', 'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113', + 'uploader': 'NBCU-SPORTS', + 'upload_date': '20150330', + 'timestamp': 1427726529, } } @@ -172,9 +188,9 @@ class CSNNEIE(InfoExtractor): class NBCNewsIE(ThePlatformIE): - _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/ + _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/ (?:video/.+?/(?P<id>\d+)| - ([^/]+/)*(?P<display_id>[^/?]+)) + ([^/]+/)*(?:.*-)?(?P<mpx_id>[^/?]+)) ''' _TESTS = [ @@ -196,13 +212,16 @@ class NBCNewsIE(ThePlatformIE): 'ext': 'mp4', 'title': 'How Twitter Reacted To The Snowden Interview', 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64', + 'uploader': 'NBCU-NEWS', + 'timestamp': 1401363060, + 'upload_date': '20140529', }, }, { 'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156', 'md5': 'fdbf39ab73a72df5896b6234ff98518a', 'info_dict': { - 'id': 'Wjf9EDR3A_60', + 'id': '529953347624', 'ext': 'mp4', 'title': 'FULL EPISODE: Family Business', 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04', @@ -217,6 +236,9 @@ class NBCNewsIE(ThePlatformIE): 'ext': 'mp4', 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)', 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5', + 'timestamp': 1423104900, + 'uploader': 'NBCU-NEWS', + 'upload_date': '20150205', }, }, { @@ -225,10 +247,12 @@ class NBCNewsIE(ThePlatformIE): 'info_dict': { 'id': '529953347624', 'ext': 'mp4', - 'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'', - 'description': 'md5:d22d1281a24f22ea0880741bb4dd6301', + 'title': 'Volkswagen U.S. Chief:\xa0 We Have Totally Screwed Up', + 'description': 'md5:c8be487b2d80ff0594c005add88d8351', + 'upload_date': '20150922', + 'timestamp': 1442917800, + 'uploader': 'NBCU-NEWS', }, - 'expected_warnings': ['http-6000 is not available'] }, { 'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788', @@ -240,12 +264,33 @@ class NBCNewsIE(ThePlatformIE): 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1', 'upload_date': '20160420', 'timestamp': 1461152093, + 'uploader': 'NBCU-NEWS', + }, + }, + { + 'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924', + 'md5': '6d236bf4f3dddc226633ce6e2c3f814d', + 'info_dict': { + 'id': '314487875924', + 'ext': 'mp4', + 'title': 'The chaotic GOP immigration vote', + 'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1406937606, + 'upload_date': '20140802', + 'uploader': 'NBCU-NEWS', + 'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'], }, }, { 'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952', 'only_matching': True, }, + { + # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html + 'url': 'http://www.nbcnews.com/widget/video-embed/701714499682', + 'only_matching': True, + }, ] def _real_extract(self, url): @@ -265,106 +310,28 @@ class NBCNewsIE(ThePlatformIE): } else: # "feature" and "nightly-news" pages use theplatform.com - display_id = mobj.group('display_id') - webpage = self._download_webpage(url, display_id) - info = None - bootstrap_json = self._search_regex( - r'(?m)var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$', - webpage, 'bootstrap json', default=None) - if bootstrap_json: - bootstrap = self._parse_json(bootstrap_json, display_id) - info = bootstrap['results'][0]['video'] - else: - player_instance_json = self._search_regex( - r'videoObj\s*:\s*({.+})', webpage, 'player instance', default=None) - if not player_instance_json: - player_instance_json = self._html_search_regex( - r'data-video="([^"]+)"', webpage, 'video json') - info = self._parse_json(player_instance_json, display_id) - video_id = info['mpxId'] - title = info['title'] - - subtitles = {} - caption_links = info.get('captionLinks') - if caption_links: - for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')): - sub_url = caption_links.get(sub_key) - if sub_url: - subtitles.setdefault('en', []).append({ - 'url': sub_url, - 'ext': sub_ext, - }) - - formats = [] - for video_asset in info['videoAssets']: - video_url = video_asset.get('publicUrl') - if not video_url: - continue - container = video_asset.get('format') - asset_type = video_asset.get('assetType') or '' - if container == 'ISM' or asset_type == 'FireTV-Once': - continue - elif asset_type == 'OnceURL': - tp_formats, tp_subtitles = self._extract_theplatform_smil( - video_url, video_id) - formats.extend(tp_formats) - subtitles = self._merge_subtitles(subtitles, tp_subtitles) + video_id = mobj.group('mpx_id') + if not video_id.isdigit(): + webpage = self._download_webpage(url, video_id) + info = None + bootstrap_json = self._search_regex( + [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$', + r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'], + webpage, 'bootstrap json', default=None) + bootstrap = self._parse_json( + bootstrap_json, video_id, transform_source=unescapeHTML) + if 'results' in bootstrap: + info = bootstrap['results'][0]['video'] + elif 'video' in bootstrap: + info = bootstrap['video'] else: - tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000) - format_id = 'http%s' % ('-%d' % tbr if tbr else '') - video_url = update_url_query( - video_url, {'format': 'redirect'}) - # resolve the url so that we can check availability and detect the correct extension - head = self._request_webpage( - HEADRequest(video_url), video_id, - 'Checking %s url' % format_id, - '%s is not available' % format_id, - fatal=False) - if head: - video_url = head.geturl() - formats.append({ - 'format_id': format_id, - 'url': video_url, - 'width': int_or_none(video_asset.get('width')), - 'height': int_or_none(video_asset.get('height')), - 'tbr': tbr, - 'container': video_asset.get('format'), - }) - self._sort_formats(formats) + info = bootstrap + video_id = info['mpxId'] return { + '_type': 'url_transparent', 'id': video_id, - 'title': title, - 'description': info.get('description'), - 'thumbnail': info.get('thumbnail'), - 'duration': int_or_none(info.get('duration')), - 'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')), - 'formats': formats, - 'subtitles': subtitles, + # http://feed.theplatform.com/f/2E2eJC/nbcnews also works + 'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byId=%s' % video_id, + 'ie_key': 'ThePlatformFeed', } - - -class MSNBCIE(InfoExtractor): - # https URLs redirect to corresponding http ones - _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)' - _TEST = { - 'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924', - 'md5': '6d236bf4f3dddc226633ce6e2c3f814d', - 'info_dict': { - 'id': 'n_hayes_Aimm_140801_272214', - 'ext': 'mp4', - 'title': 'The chaotic GOP immigration vote', - 'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.', - 'thumbnail': 're:^https?://.*\.jpg$', - 'timestamp': 1406937606, - 'upload_date': '20140802', - 'uploader': 'NBCU-NEWS', - 'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'], - }, - } - - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - embed_url = self._html_search_meta('embedURL', webpage) - return self.url_result(embed_url) diff --git a/youtube_dl/extractor/ndtv.py b/youtube_dl/extractor/ndtv.py index 2a1ca80df..96528f649 100644 --- a/youtube_dl/extractor/ndtv.py +++ b/youtube_dl/extractor/ndtv.py @@ -1,19 +1,18 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( - month_by_name, int_or_none, + remove_end, + unified_strdate, ) class NDTVIE(InfoExtractor): - _VALID_URL = r'^https?://(?:www\.)?ndtv\.com/video/player/[^/]*/[^/]*/(?P<id>[a-z0-9]+)' + _VALID_URL = r'https?://(?:www\.)?ndtv\.com/video/(?:[^/]+/)+[^/?^&]+-(?P<id>\d+)' _TEST = { - 'url': 'http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710', + 'url': 'http://www.ndtv.com/video/news/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal-300710', 'md5': '39f992dbe5fb531c395d8bbedb1e5e88', 'info_dict': { 'id': '300710', @@ -22,7 +21,7 @@ class NDTVIE(InfoExtractor): 'description': 'md5:ab2d4b4a6056c5cb4caa6d729deabf02', 'upload_date': '20131208', 'duration': 1327, - 'thumbnail': 'http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg', + 'thumbnail': 're:https?://.*\.jpg', }, } @@ -30,36 +29,19 @@ class NDTVIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) + title = remove_end(self._og_search_title(webpage), ' - NDTV') + filename = self._search_regex( r"__filename='([^']+)'", webpage, 'video filename') - video_url = ('http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' % - filename) + video_url = 'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' % filename duration = int_or_none(self._search_regex( r"__duration='([^']+)'", webpage, 'duration', fatal=False)) - date_m = re.search(r'''(?x) - <p\s+class="vod_dateline">\s* - Published\s+On:\s* - (?P<monthname>[A-Za-z]+)\s+(?P<day>[0-9]+),\s*(?P<year>[0-9]+) - ''', webpage) - upload_date = None - - if date_m is not None: - month = month_by_name(date_m.group('monthname')) - if month is not None: - upload_date = '%s%02d%02d' % ( - date_m.group('year'), month, int(date_m.group('day'))) - - description = self._og_search_description(webpage) - READ_MORE = ' (Read more)' - if description.endswith(READ_MORE): - description = description[:-len(READ_MORE)] + upload_date = unified_strdate(self._html_search_meta( + 'publish-date', webpage, 'upload date', fatal=False)) - title = self._og_search_title(webpage) - TITLE_SUFFIX = ' - NDTV' - if title.endswith(TITLE_SUFFIX): - title = title[:-len(TITLE_SUFFIX)] + description = remove_end(self._og_search_description(webpage), ' (Read more)') return { 'id': video_id, diff --git a/youtube_dl/extractor/nfb.py b/youtube_dl/extractor/nfb.py index 51e4a34f7..adcc636bc 100644 --- a/youtube_dl/extractor/nfb.py +++ b/youtube_dl/extractor/nfb.py @@ -2,8 +2,12 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - sanitized_Request, + clean_html, + determine_ext, + int_or_none, + qualities, urlencode_postdata, + xpath_text, ) @@ -16,12 +20,12 @@ class NFBIE(InfoExtractor): 'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny', 'info_dict': { 'id': 'qallunaat_why_white_people_are_funny', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Qallunaat! Why White People Are Funny ', - 'description': 'md5:836d8aff55e087d04d9f6df554d4e038', + 'description': 'md5:6b8e32dde3abf91e58857b174916620c', 'duration': 3128, + 'creator': 'Mark Sandiford', 'uploader': 'Mark Sandiford', - 'uploader_id': 'mark-sandiford', }, 'params': { # rtmp download @@ -31,65 +35,78 @@ class NFBIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - page = self._download_webpage( - 'https://www.nfb.ca/film/%s' % video_id, video_id, - 'Downloading film page') - uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"', - page, 'director id', fatal=False) - uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>', - page, 'director name', fatal=False) - - request = sanitized_Request( + config = self._download_xml( 'https://www.nfb.ca/film/%s/player_config' % video_id, - urlencode_postdata({'getConfig': 'true'})) - request.add_header('Content-Type', 'application/x-www-form-urlencoded') - request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf') - - config = self._download_xml(request, video_id, 'Downloading player config XML') + video_id, 'Downloading player config XML', + data=urlencode_postdata({'getConfig': 'true'}), + headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + 'X-NFB-Referer': 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf' + }) - title = None - description = None - thumbnail = None - duration = None - formats = [] - - def extract_thumbnail(media): - thumbnails = {} - for asset in media.findall('assets/asset'): - thumbnails[asset.get('quality')] = asset.find('default/url').text - if not thumbnails: - return None - if 'high' in thumbnails: - return thumbnails['high'] - return list(thumbnails.values())[0] + title, description, thumbnail, duration, uploader, author = [None] * 6 + thumbnails, formats = [[]] * 2 + subtitles = {} for media in config.findall('./player/stream/media'): if media.get('type') == 'posterImage': - thumbnail = extract_thumbnail(media) + quality_key = qualities(('low', 'high')) + thumbnails = [] + for asset in media.findall('assets/asset'): + asset_url = xpath_text(asset, 'default/url', default=None) + if not asset_url: + continue + quality = asset.get('quality') + thumbnails.append({ + 'url': asset_url, + 'id': quality, + 'preference': quality_key(quality), + }) elif media.get('type') == 'video': - duration = int(media.get('duration')) - title = media.find('title').text - description = media.find('description').text - # It seems assets always go from lower to better quality, so no need to sort + title = xpath_text(media, 'title', fatal=True) for asset in media.findall('assets/asset'): - for x in asset: + quality = asset.get('quality') + height = int_or_none(self._search_regex( + r'^(\d+)[pP]$', quality or '', 'height', default=None)) + for node in asset: + streamer = xpath_text(node, 'streamerURI', default=None) + if not streamer: + continue + play_path = xpath_text(node, 'url', default=None) + if not play_path: + continue formats.append({ - 'url': x.find('streamerURI').text, - 'app': x.find('streamerURI').text.split('/', 3)[3], - 'play_path': x.find('url').text, + 'url': streamer, + 'app': streamer.split('/', 3)[3], + 'play_path': play_path, 'rtmp_live': False, - 'ext': 'mp4', - 'format_id': '%s-%s' % (x.tag, asset.get('quality')), + 'ext': 'flv', + 'format_id': '%s-%s' % (node.tag, quality) if quality else node.tag, + 'height': height, }) + self._sort_formats(formats) + description = clean_html(xpath_text(media, 'description')) + uploader = xpath_text(media, 'author') + duration = int_or_none(media.get('duration')) + for subtitle in media.findall('./subtitles/subtitle'): + subtitle_url = xpath_text(subtitle, 'url', default=None) + if not subtitle_url: + continue + lang = xpath_text(subtitle, 'lang', default='en') + subtitles.setdefault(lang, []).append({ + 'url': subtitle_url, + 'ext': (subtitle.get('format') or determine_ext(subtitle_url)).lower(), + }) return { 'id': video_id, 'title': title, 'description': description, - 'thumbnail': thumbnail, + 'thumbnails': thumbnails, 'duration': duration, + 'creator': uploader, 'uploader': uploader, - 'uploader_id': uploader_id, 'formats': formats, + 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/nick.py b/youtube_dl/extractor/nick.py index ce065f2b0..4935002d0 100644 --- a/youtube_dl/extractor/nick.py +++ b/youtube_dl/extractor/nick.py @@ -3,11 +3,12 @@ from __future__ import unicode_literals from .mtv import MTVServicesInfoExtractor from ..compat import compat_urllib_parse_urlencode +from ..utils import update_url_query class NickIE(MTVServicesInfoExtractor): IE_NAME = 'nick.com' - _VALID_URL = r'https?://(?:www\.)?nick\.com/videos/clip/(?P<id>[^/?#.]+)' + _VALID_URL = r'https?://(?:www\.)?nick(?:jr)?\.com/(?:videos/clip|[^/]+/videos)/(?P<id>[^/?#.]+)' _FEED_URL = 'http://udat.mtvnservices.com/service1/dispatch.htm' _TESTS = [{ 'url': 'http://www.nick.com/videos/clip/alvinnn-and-the-chipmunks-112-full-episode.html', @@ -51,6 +52,9 @@ class NickIE(MTVServicesInfoExtractor): } }, ], + }, { + 'url': 'http://www.nickjr.com/paw-patrol/videos/pups-save-a-goldrush-s3-ep302-full-episode/', + 'only_matching': True, }] def _get_feed_query(self, uri): @@ -61,3 +65,26 @@ class NickIE(MTVServicesInfoExtractor): def _extract_mgid(self, webpage): return self._search_regex(r'data-contenturi="([^"]+)', webpage, 'mgid') + + +class NickDeIE(MTVServicesInfoExtractor): + IE_NAME = 'nick.de' + _VALID_URL = r'https?://(?:www\.)?nick\.de/(?:playlist|shows)/(?:[^/]+/)*(?P<id>[^/?#&]+)' + _TESTS = [{ + 'url': 'http://www.nick.de/playlist/3773-top-videos/videos/episode/17306-zu-wasser-und-zu-land-rauchende-erdnusse', + 'only_matching': True, + }, { + 'url': 'http://www.nick.de/shows/342-icarly', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + mrss_url = update_url_query(self._search_regex( + r'data-mrss=(["\'])(?P<url>http.+?)\1', webpage, 'mrss url', group='url'), + {'siteKey': 'nick.de'}) + + return self._get_videos_info_from_url(mrss_url, video_id) diff --git a/youtube_dl/extractor/ninecninemedia.py b/youtube_dl/extractor/ninecninemedia.py new file mode 100644 index 000000000..d889245ad --- /dev/null +++ b/youtube_dl/extractor/ninecninemedia.py @@ -0,0 +1,55 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + parse_iso8601, + parse_duration, + ExtractorError +) + + +class NineCNineMediaIE(InfoExtractor): + _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)' + + def _real_extract(self, url): + destination_code, video_id = re.match(self._VALID_URL, url).groups() + api_base_url = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' % (destination_code, video_id) + content = self._download_json(api_base_url, video_id, query={ + '$include': '[contentpackages]', + }) + title = content['Name'] + if len(content['ContentPackages']) > 1: + raise ExtractorError('multiple content packages') + content_package = content['ContentPackages'][0] + stacks_base_url = api_base_url + 'contentpackages/%s/stacks/' % content_package['Id'] + stacks = self._download_json(stacks_base_url, video_id)['Items'] + if len(stacks) > 1: + raise ExtractorError('multiple stacks') + stack = stacks[0] + stack_base_url = '%s%s/manifest.' % (stacks_base_url, stack['Id']) + formats = [] + formats.extend(self._extract_m3u8_formats( + stack_base_url + 'm3u8', video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + stack_base_url + 'f4m', video_id, + f4m_id='hds', fatal=False)) + mp4_url = self._download_webpage(stack_base_url + 'pd', video_id, fatal=False) + if mp4_url: + formats.append({ + 'url': mp4_url, + 'format_id': 'mp4', + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': content.get('Desc') or content.get('ShortDesc'), + 'timestamp': parse_iso8601(content.get('BroadcastDateTime')), + 'duration': parse_duration(content.get('BroadcastTime')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/normalboots.py b/youtube_dl/extractor/normalboots.py index 77e091072..af44c3bb5 100644 --- a/youtube_dl/extractor/normalboots.py +++ b/youtube_dl/extractor/normalboots.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor +from .screenwavemedia import ScreenwaveMediaIE from ..utils import ( unified_strdate, @@ -12,7 +13,6 @@ class NormalbootsIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?normalboots\.com/video/(?P<id>[0-9a-z-]*)/?$' _TEST = { 'url': 'http://normalboots.com/video/home-alone-games-jontron/', - 'md5': '8bf6de238915dd501105b44ef5f1e0f6', 'info_dict': { 'id': 'home-alone-games-jontron', 'ext': 'mp4', @@ -22,9 +22,10 @@ class NormalbootsIE(InfoExtractor): 'upload_date': '20140125', }, 'params': { - # rtmp download + # m3u8 download 'skip_download': True, }, + 'add_ie': ['ScreenwaveMedia'], } def _real_extract(self, url): @@ -38,16 +39,15 @@ class NormalbootsIE(InfoExtractor): r'<span style="text-transform:uppercase; font-size:inherit;">[A-Za-z]+, (?P<date>.*)</span>', webpage, 'date', fatal=False)) - player_url = self._html_search_regex( - r'<iframe\swidth="[0-9]+"\sheight="[0-9]+"\ssrc="(?P<url>[\S]+)"', - webpage, 'player url') - player_page = self._download_webpage(player_url, video_id) - video_url = self._html_search_regex( - r"file:\s'(?P<file>[^']+\.mp4)'", player_page, 'file') + screenwavemedia_url = self._html_search_regex( + ScreenwaveMediaIE.EMBED_PATTERN, webpage, 'screenwave URL', + group='url') return { + '_type': 'url_transparent', 'id': video_id, - 'url': video_url, + 'url': screenwavemedia_url, + 'ie_key': ScreenwaveMediaIE.ie_key(), 'title': self._og_search_title(webpage), 'description': self._og_search_description(webpage), 'thumbnail': self._og_search_thumbnail(webpage), diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 9df200822..6ded5bd45 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -4,90 +4,218 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urlparse, - compat_urllib_parse_unquote, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( - determine_ext, ExtractorError, - float_or_none, + int_or_none, + parse_age_limit, parse_duration, - unified_strdate, ) -class NRKIE(InfoExtractor): - _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)' - - _TESTS = [ - { - 'url': 'http://www.nrk.no/video/PS*150533', - 'md5': 'bccd850baebefe23b56d708a113229c2', - 'info_dict': { - 'id': '150533', - 'ext': 'flv', - 'title': 'Dompap og andre fugler i Piip-Show', - 'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f', - 'duration': 263, - } - }, - { - 'url': 'http://www.nrk.no/video/PS*154915', - 'md5': '0b1493ba1aae7d9579a5ad5531bc395a', - 'info_dict': { - 'id': '154915', - 'ext': 'flv', - 'title': 'Slik høres internett ut når du er blind', - 'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568', - 'duration': 20, - } - }, - ] +class NRKBaseIE(InfoExtractor): + def _extract_formats(self, manifest_url, video_id, fatal=True): + formats = [] + formats.extend(self._extract_f4m_formats( + manifest_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', + video_id, f4m_id='hds', fatal=fatal)) + formats.extend(self._extract_m3u8_formats(manifest_url.replace( + 'akamaihd.net/z/', 'akamaihd.net/i/').replace('/manifest.f4m', '/master.m3u8'), + video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=fatal)) + return formats def _real_extract(self, url): video_id = self._match_id(url) data = self._download_json( - 'http://v8.psapi.nrk.no/mediaelement/%s' % video_id, - video_id, 'Downloading media JSON') + 'http://%s/mediaelement/%s' % (self._API_HOST, video_id), + video_id, 'Downloading mediaelement JSON') + + title = data.get('fullTitle') or data.get('mainTitle') or data['title'] + video_id = data.get('id') or video_id + + entries = [] + + media_assets = data.get('mediaAssets') + if media_assets and isinstance(media_assets, list): + def video_id_and_title(idx): + return ((video_id, title) if len(media_assets) == 1 + else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx))) + for num, asset in enumerate(media_assets, 1): + asset_url = asset.get('url') + if not asset_url: + continue + formats = self._extract_formats(asset_url, video_id, fatal=False) + if not formats: + continue + self._sort_formats(formats) + entry_id, entry_title = video_id_and_title(num) + duration = parse_duration(asset.get('duration')) + subtitles = {} + for subtitle in ('webVtt', 'timedText'): + subtitle_url = asset.get('%sSubtitlesUrl' % subtitle) + if subtitle_url: + subtitles.setdefault('no', []).append({ + 'url': compat_urllib_parse_unquote(subtitle_url) + }) + entries.append({ + 'id': asset.get('carrierId') or entry_id, + 'title': entry_title, + 'duration': duration, + 'subtitles': subtitles, + 'formats': formats, + }) - media_url = data.get('mediaUrl') + if not entries: + media_url = data.get('mediaUrl') + if media_url: + formats = self._extract_formats(media_url, video_id) + self._sort_formats(formats) + duration = parse_duration(data.get('duration')) + entries = [{ + 'id': video_id, + 'title': title, + 'duration': duration, + 'formats': formats, + }] - if not media_url: - if data['usageRights']['isGeoBlocked']: + if not entries: + if data.get('usageRights', {}).get('isGeoBlocked'): raise ExtractorError( 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', expected=True) - if determine_ext(media_url) == 'f4m': - formats = self._extract_f4m_formats( - media_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id, f4m_id='hds') - self._sort_formats(formats) - else: - formats = [{ - 'url': media_url, - 'ext': 'flv', - }] - - duration = parse_duration(data.get('duration')) + conviva = data.get('convivaStatistics') or {} + series = conviva.get('seriesName') or data.get('seriesTitle') + episode = conviva.get('episodeName') or data.get('episodeNumberOrDate') + thumbnails = None images = data.get('images') - if images: - thumbnails = images['webImages'] - thumbnails.sort(key=lambda image: image['pixelWidth']) - thumbnail = thumbnails[-1]['imageUrl'] - else: - thumbnail = None - - return { - 'id': video_id, - 'title': data['title'], - 'description': data['description'], - 'duration': duration, - 'thumbnail': thumbnail, - 'formats': formats, + if images and isinstance(images, dict): + web_images = images.get('webImages') + if isinstance(web_images, list): + thumbnails = [{ + 'url': image['imageUrl'], + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in web_images if image.get('imageUrl')] + + description = data.get('description') + + common_info = { + 'description': description, + 'series': series, + 'episode': episode, + 'age_limit': parse_age_limit(data.get('legalAge')), + 'thumbnails': thumbnails, + } + + vcodec = 'none' if data.get('mediaType') == 'Audio' else None + + # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged + + for entry in entries: + entry.update(common_info) + for f in entry['formats']: + f['vcodec'] = vcodec + + return self.playlist_result(entries, video_id, title, description) + + +class NRKIE(NRKBaseIE): + _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)' + _API_HOST = 'v8.psapi.nrk.no' + _TESTS = [{ + # video + 'url': 'http://www.nrk.no/video/PS*150533', + 'md5': '2f7f6eeb2aacdd99885f355428715cfa', + 'info_dict': { + 'id': '150533', + 'ext': 'mp4', + 'title': 'Dompap og andre fugler i Piip-Show', + 'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f', + 'duration': 263, + } + }, { + # audio + 'url': 'http://www.nrk.no/video/PS*154915', + # MD5 is unstable + 'info_dict': { + 'id': '154915', + 'ext': 'flv', + 'title': 'Slik høres internett ut når du er blind', + 'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568', + 'duration': 20, } + }] + + +class NRKTVIE(NRKBaseIE): + IE_DESC = 'NRK TV and NRK Radio' + _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?' + _API_HOST = 'psapi-we.nrk.no' + + _TESTS = [{ + 'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014', + 'md5': '4e9ca6629f09e588ed240fb11619922a', + 'info_dict': { + 'id': 'MUHH48000314AA', + 'ext': 'mp4', + 'title': '20 spørsmål 23.05.2014', + 'description': 'md5:bdea103bc35494c143c6a9acdd84887a', + 'duration': 1741, + }, + }, { + 'url': 'https://tv.nrk.no/program/mdfp15000514', + 'md5': '43d0be26663d380603a9cf0c24366531', + 'info_dict': { + 'id': 'MDFP15000514CA', + 'ext': 'mp4', + 'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014', + 'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db', + 'duration': 4605, + }, + }, { + # single playlist video + 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2', + 'md5': 'adbd1dbd813edaf532b0a253780719c2', + 'info_dict': { + 'id': 'MSPO40010515-part2', + 'ext': 'flv', + 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', + 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + }, + 'skip': 'Only works from Norway', + }, { + 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015', + 'playlist': [{ + 'md5': '9480285eff92d64f06e02a5367970a7a', + 'info_dict': { + 'id': 'MSPO40010515-part1', + 'ext': 'flv', + 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)', + 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + }, + }, { + 'md5': 'adbd1dbd813edaf532b0a253780719c2', + 'info_dict': { + 'id': 'MSPO40010515-part2', + 'ext': 'flv', + 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', + 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + }, + }], + 'info_dict': { + 'id': 'MSPO40010515', + 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn', + 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', + 'duration': 6947.52, + }, + 'skip': 'Only works from Norway', + }, { + 'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#', + 'only_matching': True, + }] class NRKPlaylistIE(InfoExtractor): @@ -132,206 +260,34 @@ class NRKPlaylistIE(InfoExtractor): class NRKSkoleIE(InfoExtractor): IE_DESC = 'NRK Skole' - _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/klippdetalj?.*\btopic=(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)' _TESTS = [{ - 'url': 'http://nrk.no/skole/klippdetalj?topic=nrk:klipp/616532', - 'md5': '04cd85877cc1913bce73c5d28a47e00f', + 'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099', + 'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6', 'info_dict': { 'id': '6021', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Genetikk og eneggede tvillinger', 'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d', 'duration': 399, }, }, { - 'url': 'http://www.nrk.no/skole/klippdetalj?topic=nrk%3Aklipp%2F616532#embed', - 'only_matching': True, - }, { - 'url': 'http://www.nrk.no/skole/klippdetalj?topic=urn:x-mediadb:21379', + 'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355', 'only_matching': True, }] def _real_extract(self, url): - video_id = compat_urllib_parse_unquote(self._match_id(url)) - - webpage = self._download_webpage(url, video_id) - - nrk_id = self._search_regex(r'data-nrk-id=["\'](\d+)', webpage, 'nrk id') - return self.url_result('nrk:%s' % nrk_id) - - -class NRKTVIE(InfoExtractor): - IE_DESC = 'NRK TV and NRK Radio' - _VALID_URL = r'(?P<baseurl>https?://(?:tv|radio)\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?' - - _TESTS = [ - { - 'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014', - 'info_dict': { - 'id': 'MUHH48000314', - 'ext': 'mp4', - 'title': '20 spørsmål', - 'description': 'md5:bdea103bc35494c143c6a9acdd84887a', - 'upload_date': '20140523', - 'duration': 1741.52, - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }, - { - 'url': 'https://tv.nrk.no/program/mdfp15000514', - 'info_dict': { - 'id': 'mdfp15000514', - 'ext': 'mp4', - 'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting', - 'description': 'md5:654c12511f035aed1e42bdf5db3b206a', - 'upload_date': '20140524', - 'duration': 4605.08, - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }, - { - # single playlist video - 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2', - 'md5': 'adbd1dbd813edaf532b0a253780719c2', - 'info_dict': { - 'id': 'MSPO40010515-part2', - 'ext': 'flv', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', - 'upload_date': '20150106', - }, - 'skip': 'Only works from Norway', - }, - { - 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015', - 'playlist': [ - { - 'md5': '9480285eff92d64f06e02a5367970a7a', - 'info_dict': { - 'id': 'MSPO40010515-part1', - 'ext': 'flv', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', - 'upload_date': '20150106', - }, - }, - { - 'md5': 'adbd1dbd813edaf532b0a253780719c2', - 'info_dict': { - 'id': 'MSPO40010515-part2', - 'ext': 'flv', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', - 'upload_date': '20150106', - }, - }, - ], - 'info_dict': { - 'id': 'MSPO40010515', - 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn', - 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', - 'upload_date': '20150106', - 'duration': 6947.5199999999995, - }, - 'skip': 'Only works from Norway', - }, - { - 'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#', - 'only_matching': True, - } - ] + video_id = self._match_id(url) - def _extract_f4m(self, manifest_url, video_id): - return self._extract_f4m_formats( - manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds') + webpage = self._download_webpage( + 'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id, + video_id) - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - part_id = mobj.group('part_id') - base_url = mobj.group('baseurl') - - webpage = self._download_webpage(url, video_id) - - title = self._html_search_meta( - 'title', webpage, 'title') - description = self._html_search_meta( - 'description', webpage, 'description') - - thumbnail = self._html_search_regex( - r'data-posterimage="([^"]+)"', - webpage, 'thumbnail', fatal=False) - upload_date = unified_strdate(self._html_search_meta( - 'rightsfrom', webpage, 'upload date', fatal=False)) - duration = float_or_none(self._html_search_regex( - r'data-duration="([^"]+)"', - webpage, 'duration', fatal=False)) - - # playlist - parts = re.findall( - r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage) - if parts: - entries = [] - for current_part_id, stream_url, part_title in parts: - if part_id and current_part_id != part_id: - continue - video_part_id = '%s-part%s' % (video_id, current_part_id) - formats = self._extract_f4m(stream_url, video_part_id) - entries.append({ - 'id': video_part_id, - 'title': part_title, - 'description': description, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'formats': formats, - }) - if part_id: - if entries: - return entries[0] - else: - playlist = self.playlist_result(entries, video_id, title, description) - playlist.update({ - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'duration': duration, - }) - return playlist - - formats = [] + nrk_id = self._parse_json( + self._search_regex( + r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>', + webpage, 'application json'), + video_id)['activeMedia']['psId'] - f4m_url = re.search(r'data-media="([^"]+)"', webpage) - if f4m_url: - formats.extend(self._extract_f4m(f4m_url.group(1), video_id)) - - m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage) - if m3u8_url: - formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4', m3u8_id='hls')) - self._sort_formats(formats) - - subtitles_url = self._html_search_regex( - r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1', - webpage, 'subtitle URL', default=None, group='url') - subtitles = {} - if subtitles_url: - subtitles['no'] = [{ - 'ext': 'ttml', - 'url': compat_urlparse.urljoin(base_url, subtitles_url), - }] - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - } + return self.url_result('nrk:%s' % nrk_id) diff --git a/youtube_dl/extractor/nuvid.py b/youtube_dl/extractor/nuvid.py index 9fa7cefad..ab6bfcd7f 100644 --- a/youtube_dl/extractor/nuvid.py +++ b/youtube_dl/extractor/nuvid.py @@ -5,8 +5,6 @@ import re from .common import InfoExtractor from ..utils import ( parse_duration, - sanitized_Request, - unified_strdate, ) @@ -20,7 +18,6 @@ class NuvidIE(InfoExtractor): 'ext': 'mp4', 'title': 'Horny babes show their awesome bodeis and', 'duration': 129, - 'upload_date': '20140508', 'age_limit': 18, } } @@ -28,28 +25,31 @@ class NuvidIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - formats = [] + page_url = 'http://m.nuvid.com/video/%s' % video_id + webpage = self._download_webpage( + page_url, video_id, 'Downloading video page') + # When dwnld_speed exists and has a value larger than the MP4 file's + # bitrate, Nuvid returns the MP4 URL + # It's unit is 100bytes/millisecond, see mobile-nuvid-min.js for the algorithm + self._set_cookie('nuvid.com', 'dwnld_speed', '10.0') + mp4_webpage = self._download_webpage( + page_url, video_id, 'Downloading video page for MP4 format') - for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]: - request = sanitized_Request( - 'http://m.nuvid.com/play/%s' % video_id) - request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed) - webpage = self._download_webpage( - request, video_id, 'Downloading %s page' % format_id) - video_url = self._html_search_regex( - r'<a\s+href="([^"]+)"\s+class="b_link">', webpage, '%s video URL' % format_id, fatal=False) - if not video_url: - continue + html5_video_re = r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', + video_url = self._html_search_regex(html5_video_re, webpage, video_id) + mp4_video_url = self._html_search_regex(html5_video_re, mp4_webpage, video_id) + formats = [{ + 'url': video_url, + }] + if mp4_video_url != video_url: formats.append({ - 'url': video_url, - 'format_id': format_id, + 'url': mp4_video_url, }) - webpage = self._download_webpage( - 'http://m.nuvid.com/video/%s' % video_id, video_id, 'Downloading video page') title = self._html_search_regex( [r'<span title="([^"]+)">', - r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>'], webpage, 'title').strip() + r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>', + r'<span[^>]+class="title_thumb">([^<]+)</span>'], webpage, 'title').strip() thumbnails = [ { 'url': thumb_url, @@ -57,9 +57,8 @@ class NuvidIE(InfoExtractor): ] thumbnail = thumbnails[0]['url'] if thumbnails else None duration = parse_duration(self._html_search_regex( - r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})', webpage, 'duration', fatal=False)) - upload_date = unified_strdate(self._html_search_regex( - r'<i class="fa fa-user"></i>\s*(\d{4}-\d{2}-\d{2})', webpage, 'upload date', fatal=False)) + [r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})', + r'<span[^>]+class="view_time">([^<]+)</span>'], webpage, 'duration', fatal=False)) return { 'id': video_id, @@ -67,7 +66,6 @@ class NuvidIE(InfoExtractor): 'thumbnails': thumbnails, 'thumbnail': thumbnail, 'duration': duration, - 'upload_date': upload_date, 'age_limit': 18, 'formats': formats, } diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index f9e064a60..986708e75 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -2,7 +2,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_unquote, + compat_urllib_parse_urlparse, +) from ..utils import ( ExtractorError, unified_strdate, @@ -32,7 +36,7 @@ class OdnoklassnikiIE(InfoExtractor): 'skip': 'Video has been blocked', }, { # metadataUrl - 'url': 'http://ok.ru/video/63567059965189-0', + 'url': 'http://ok.ru/video/63567059965189-0?fromTime=5', 'md5': '9676cf86eff5391d35dea675d224e131', 'info_dict': { 'id': '63567059965189-0', @@ -44,6 +48,7 @@ class OdnoklassnikiIE(InfoExtractor): 'uploader': '☭ Андрей Мещанинов ☭', 'like_count': int, 'age_limit': 0, + 'start_time': 5, }, }, { # YouTube embed (metadataUrl, provider == USER_YOUTUBE) @@ -61,6 +66,22 @@ class OdnoklassnikiIE(InfoExtractor): 'age_limit': 0, }, }, { + # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field) + 'url': 'http://ok.ru/video/62036049272859-0', + 'info_dict': { + 'id': '62036049272859-0', + 'ext': 'mp4', + 'title': 'МУЗЫКА ДОЖДЯ .', + 'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0', + 'upload_date': '20120106', + 'uploader_id': '473534735899', + 'uploader': 'МARINA D', + 'age_limit': 0, + }, + 'params': { + 'skip_download': True, + }, + }, { 'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452', 'only_matching': True, }, { @@ -78,6 +99,9 @@ class OdnoklassnikiIE(InfoExtractor): }] def _real_extract(self, url): + start_time = int_or_none(compat_parse_qs( + compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0]) + video_id = self._match_id(url) webpage = self._download_webpage( @@ -106,7 +130,14 @@ class OdnoklassnikiIE(InfoExtractor): video_id, 'Downloading metadata JSON') movie = metadata['movie'] - title = movie['title'] + + # Some embedded videos may not contain title in movie dict (e.g. + # http://ok.ru/video/62036049272859-0) thus we allow missing title + # here and it's going to be extracted later by an extractor that + # will process the actual embed. + provider = metadata.get('provider') + title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title') + thumbnail = movie.get('poster') duration = int_or_none(movie.get('duration')) @@ -135,9 +166,10 @@ class OdnoklassnikiIE(InfoExtractor): 'uploader_id': uploader_id, 'like_count': like_count, 'age_limit': age_limit, + 'start_time': start_time, } - if metadata.get('provider') == 'USER_YOUTUBE': + if provider == 'USER_YOUTUBE': info.update({ '_type': 'url_transparent', 'url': movie['contentId'], diff --git a/youtube_dl/extractor/onet.py b/youtube_dl/extractor/onet.py new file mode 100644 index 000000000..402d3a9f7 --- /dev/null +++ b/youtube_dl/extractor/onet.py @@ -0,0 +1,172 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + ExtractorError, + float_or_none, + get_element_by_class, + int_or_none, + js_to_json, + parse_iso8601, + remove_start, + strip_or_none, + url_basename, +) + + +class OnetBaseIE(InfoExtractor): + def _search_mvp_id(self, webpage): + return self._search_regex( + r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id') + + def _extract_from_id(self, video_id, webpage): + response = self._download_json( + 'http://qi.ckm.onetapi.pl/', video_id, + query={ + 'body[id]': video_id, + 'body[jsonrpc]': '2.0', + 'body[method]': 'get_asset_detail', + 'body[params][ID_Publikacji]': video_id, + 'body[params][Service]': 'www.onet.pl', + 'content-type': 'application/jsonp', + 'x-onet-app': 'player.front.onetapi.pl', + }) + + error = response.get('error') + if error: + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, error['message']), expected=True) + + video = response['result'].get('0') + + formats = [] + for _, formats_dict in video['formats'].items(): + if not isinstance(formats_dict, dict): + continue + for format_id, format_list in formats_dict.items(): + if not isinstance(format_list, list): + continue + for f in format_list: + video_url = f.get('url') + if not video_url: + continue + ext = determine_ext(video_url) + if format_id == 'ism': + # TODO: Support Microsoft Smooth Streaming + continue + elif ext == 'mpd': + # TODO: Current DASH formats are broken - $Time$ pattern in + # <SegmentTemplate> not implemented yet + # formats.extend(self._extract_mpd_formats( + # video_url, video_id, mpd_id='dash', fatal=False)) + continue + else: + formats.append({ + 'url': video_url, + 'format_id': format_id, + 'height': int_or_none(f.get('vertical_resolution')), + 'width': int_or_none(f.get('horizontal_resolution')), + 'abr': float_or_none(f.get('audio_bitrate')), + 'vbr': float_or_none(f.get('video_bitrate')), + }) + self._sort_formats(formats) + + meta = video.get('meta', {}) + + title = self._og_search_title(webpage, default=None) or meta['title'] + description = self._og_search_description(webpage, default=None) or meta.get('description') + duration = meta.get('length') or meta.get('lenght') + timestamp = parse_iso8601(meta.get('addDate'), ' ') + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } + + +class OnetIE(OnetBaseIE): + _VALID_URL = 'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)' + IE_NAME = 'onet.tv' + + _TEST = { + 'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc', + 'md5': 'e3ffbf47590032ac3f27249204173d50', + 'info_dict': { + 'id': 'qbpyqc', + 'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd', + 'ext': 'mp4', + 'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd', + 'description': 'Trzy samochody, których nigdy nie użyto, prywatne spa, hotel dekorowany czarnym suknem czy nielegalne używki. Organizatorzy koncertów i festiwali muszą stawać przed nie lada wyzwaniem zapraszając gwia...', + 'upload_date': '20160705', + 'timestamp': 1467721580, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id, video_id = mobj.group('display_id', 'id') + + webpage = self._download_webpage(url, display_id) + + mvp_id = self._search_mvp_id(webpage) + + info_dict = self._extract_from_id(mvp_id, webpage) + info_dict.update({ + 'id': video_id, + 'display_id': display_id, + }) + + return info_dict + + +class OnetChannelIE(OnetBaseIE): + _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)' + IE_NAME = 'onet.tv:channel' + + _TEST = { + 'url': 'http://onet.tv/k/openerfestival', + 'info_dict': { + 'id': 'openerfestival', + 'title': 'Open\'er Festival Live', + 'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.', + }, + 'playlist_mincount': 46, + } + + def _real_extract(self, url): + channel_id = self._match_id(url) + + webpage = self._download_webpage(url, channel_id) + + current_clip_info = self._parse_json(self._search_regex( + r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id, + transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s))) + video_id = remove_start(current_clip_info['ckmId'], 'mvp:') + video_name = url_basename(current_clip_info['url']) + + if self._downloader.params.get('noplaylist'): + self.to_screen( + 'Downloading just video %s because of --no-playlist' % video_name) + return self._extract_from_id(video_id, webpage) + + self.to_screen( + 'Downloading channel %s - add --no-playlist to just download video %s' % ( + channel_id, video_name)) + matches = re.findall( + r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)', + webpage) + entries = [ + self.url_result(video_link, OnetIE.ie_key()) + for video_link in matches] + + channel_title = strip_or_none(get_element_by_class('o_channelName', webpage)) + channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage)) + return self.playlist_result(entries, channel_id, channel_title, channel_description) diff --git a/youtube_dl/extractor/onionstudios.py b/youtube_dl/extractor/onionstudios.py index 6e843c327..6fb1a3fcc 100644 --- a/youtube_dl/extractor/onionstudios.py +++ b/youtube_dl/extractor/onionstudios.py @@ -7,6 +7,8 @@ from .common import InfoExtractor from ..utils import ( determine_ext, int_or_none, + float_or_none, + mimetype2ext, ) @@ -15,15 +17,14 @@ class OnionStudiosIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937', - 'md5': 'd4851405d31adfadf71cd7a487b765bb', + 'md5': 'e49f947c105b8a78a675a0ee1bddedfe', 'info_dict': { 'id': '2937', 'ext': 'mp4', 'title': 'Hannibal charges forward, stops for a cocktail', - 'description': 'md5:e786add7f280b7f0fe237b64cc73df76', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'The A.V. Club', - 'uploader_id': 'TheAVClub', + 'uploader_id': 'the-av-club', }, }, { 'url': 'http://www.onionstudios.com/embed?id=2855&autoplay=true', @@ -40,50 +41,38 @@ class OnionStudiosIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage( - 'http://www.onionstudios.com/embed?id=%s' % video_id, video_id) + video_data = self._download_json( + 'http://www.onionstudios.com/video/%s.json' % video_id, video_id) + + title = video_data['title'] formats = [] - for src in re.findall(r'<source[^>]+src="([^"]+)"', webpage): - ext = determine_ext(src) + for source in video_data.get('sources', []): + source_url = source.get('url') + if not source_url: + continue + ext = mimetype2ext(source.get('content_type')) or determine_ext(source_url) if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + source_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) else: - height = int_or_none(self._search_regex( - r'/(\d+)\.%s' % ext, src, 'height', default=None)) + tbr = int_or_none(source.get('bitrate')) formats.append({ - 'format_id': ext + ('-%sp' % height if height else ''), - 'url': src, - 'height': height, + 'format_id': ext + ('-%d' % tbr if tbr else ''), + 'url': source_url, + 'width': int_or_none(source.get('width')), + 'tbr': tbr, 'ext': ext, - 'preference': 1, }) self._sort_formats(formats) - title = self._search_regex( - r'share_title\s*=\s*(["\'])(?P<title>[^\1]+?)\1', - webpage, 'title', group='title') - description = self._search_regex( - r'share_description\s*=\s*(["\'])(?P<description>[^\1]+?)\1', - webpage, 'description', default=None, group='description') - thumbnail = self._search_regex( - r'poster\s*=\s*(["\'])(?P<thumbnail>[^\1]+?)\1', - webpage, 'thumbnail', default=False, group='thumbnail') - - uploader_id = self._search_regex( - r'twitter_handle\s*=\s*(["\'])(?P<uploader_id>[^\1]+?)\1', - webpage, 'uploader id', fatal=False, group='uploader_id') - uploader = self._search_regex( - r'window\.channelName\s*=\s*(["\'])Embedded:(?P<uploader>[^\1]+?)\1', - webpage, 'uploader', default=False, group='uploader') - return { 'id': video_id, 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'uploader': uploader, - 'uploader_id': uploader_id, + 'thumbnail': video_data.get('poster_url'), + 'uploader': video_data.get('channel_name'), + 'uploader_id': video_data.get('channel_slug'), + 'duration': float_or_none(video_data.get('duration', 1000)), + 'tags': video_data.get('tags'), 'formats': formats, } diff --git a/youtube_dl/extractor/ooyala.py b/youtube_dl/extractor/ooyala.py index 16f040191..2038a6ba5 100644 --- a/youtube_dl/extractor/ooyala.py +++ b/youtube_dl/extractor/ooyala.py @@ -8,6 +8,7 @@ from ..utils import ( float_or_none, ExtractorError, unsmuggle_url, + determine_ext, ) from ..compat import compat_urllib_parse_urlencode @@ -15,71 +16,80 @@ from ..compat import compat_urllib_parse_urlencode class OoyalaBaseIE(InfoExtractor): _PLAYER_BASE = 'http://player.ooyala.com/' _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/' - _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v1/authorization/embed_code/%s/%s?' + _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s?' def _extract(self, content_tree_url, video_id, domain='example.org'): content_tree = self._download_json(content_tree_url, video_id)['content_tree'] metadata = content_tree[list(content_tree)[0]] embed_code = metadata['embed_code'] pcode = metadata.get('asset_pcode') or embed_code - video_info = { - 'id': embed_code, - 'title': metadata['title'], - 'description': metadata.get('description'), - 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'), - 'duration': float_or_none(metadata.get('duration'), 1000), - } + title = metadata['title'] + + auth_data = self._download_json( + self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) + + compat_urllib_parse_urlencode({ + 'domain': domain, + 'supportedFormats': 'mp4,rtmp,m3u8,hds', + }), video_id) + + cur_auth_data = auth_data['authorization_data'][embed_code] urls = [] formats = [] - for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'): - auth_data = self._download_json( - self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) + - compat_urllib_parse_urlencode({ - 'domain': domain, - 'supportedFormats': supported_format - }), - video_id, 'Downloading %s JSON' % supported_format) - - cur_auth_data = auth_data['authorization_data'][embed_code] - - if cur_auth_data['authorized']: - for stream in cur_auth_data['streams']: - url = base64.b64decode( - stream['url']['data'].encode('ascii')).decode('utf-8') - if url in urls: - continue - urls.append(url) - delivery_type = stream['delivery_type'] - if delivery_type == 'hls' or '.m3u8' in url: - formats.extend(self._extract_m3u8_formats( - url, embed_code, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) - elif delivery_type == 'hds' or '.f4m' in url: - formats.extend(self._extract_f4m_formats( - url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False)) - elif '.smil' in url: - formats.extend(self._extract_smil_formats( - url, embed_code, fatal=False)) - else: - formats.append({ - 'url': url, - 'ext': stream.get('delivery_type'), - 'vcodec': stream.get('video_codec'), - 'format_id': delivery_type, - 'width': int_or_none(stream.get('width')), - 'height': int_or_none(stream.get('height')), - 'abr': int_or_none(stream.get('audio_bitrate')), - 'vbr': int_or_none(stream.get('video_bitrate')), - 'fps': float_or_none(stream.get('framerate')), - }) - else: - raise ExtractorError('%s said: %s' % ( - self.IE_NAME, cur_auth_data['message']), expected=True) + if cur_auth_data['authorized']: + for stream in cur_auth_data['streams']: + s_url = base64.b64decode( + stream['url']['data'].encode('ascii')).decode('utf-8') + if s_url in urls: + continue + urls.append(s_url) + ext = determine_ext(s_url, None) + delivery_type = stream['delivery_type'] + if delivery_type == 'hls' or ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + s_url, embed_code, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif delivery_type == 'hds' or ext == 'f4m': + formats.extend(self._extract_f4m_formats( + s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False)) + elif ext == 'smil': + formats.extend(self._extract_smil_formats( + s_url, embed_code, fatal=False)) + else: + formats.append({ + 'url': s_url, + 'ext': ext or stream.get('delivery_type'), + 'vcodec': stream.get('video_codec'), + 'format_id': delivery_type, + 'width': int_or_none(stream.get('width')), + 'height': int_or_none(stream.get('height')), + 'abr': int_or_none(stream.get('audio_bitrate')), + 'vbr': int_or_none(stream.get('video_bitrate')), + 'fps': float_or_none(stream.get('framerate')), + }) + else: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, cur_auth_data['message']), expected=True) self._sort_formats(formats) - video_info['formats'] = formats - return video_info + subtitles = {} + for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items(): + sub_url = sub.get('url') + if not sub_url: + continue + subtitles[lang] = [{ + 'url': sub_url, + }] + + return { + 'id': embed_code, + 'title': title, + 'description': metadata.get('description'), + 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'), + 'duration': float_or_none(metadata.get('duration'), 1000), + 'subtitles': subtitles, + 'formats': formats, + } class OoyalaIE(OoyalaBaseIE): @@ -96,6 +106,8 @@ class OoyalaIE(OoyalaBaseIE): 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.', 'duration': 853.386, }, + # The video in the original webpage now uses PlayWire + 'skip': 'Ooyala said: movie expired', }, { # Only available for ipad 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0', diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 4468f31fc..6415b8fdc 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -6,13 +6,15 @@ import re from .common import InfoExtractor from ..compat import compat_chr from ..utils import ( + determine_ext, encode_base_n, ExtractorError, + mimetype2ext, ) class OpenloadIE(InfoExtractor): - _VALID_URL = r'https://openload.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-]+)' + _VALID_URL = r'https://openload.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)' _TESTS = [{ 'url': 'https://openload.co/f/kUEfGclsU9o', @@ -29,6 +31,14 @@ class OpenloadIE(InfoExtractor): }, { 'url': 'https://openload.io/f/ZAn6oz-VZGE/', 'only_matching': True, + }, { + 'url': 'https://openload.co/f/_-ztPaZtMhM/', + 'only_matching': True, + }, { + # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout + # for title and ext + 'url': 'https://openload.co/embed/Sxz5sADo82g/', + 'only_matching': True, }] @staticmethod @@ -93,15 +103,28 @@ class OpenloadIE(InfoExtractor): raise ExtractorError('File not found', expected=True) code = self._search_regex( - r'<video[^>]+>\s*<script[^>]+>([^<]+)</script>', + r'</video>\s*</div>\s*<script[^>]+>[^>]+</script>\s*<script[^>]+>([^<]+)</script>', webpage, 'JS code') + decoded = self.openload_decode(code) + video_url = self._search_regex( - r'return\s+"(https?://[^"]+)"', self.openload_decode(code), 'video URL') + r'return\s+"(https?://[^"]+)"', decoded, 'video URL') + + title = self._og_search_title(webpage, default=None) or self._search_regex( + r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage, + 'title', default=None) or self._html_search_meta( + 'description', webpage, 'title', fatal=True) + + ext = mimetype2ext(self._search_regex( + r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded, + 'mimetype', default=None, group='mimetype')) or determine_ext( + video_url, 'mp4') return { 'id': video_id, - 'title': self._og_search_title(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), + 'title': title, + 'ext': ext, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), 'url': video_url, } diff --git a/youtube_dl/extractor/ora.py b/youtube_dl/extractor/ora.py index 8545fb1b8..1d42be39b 100644 --- a/youtube_dl/extractor/ora.py +++ b/youtube_dl/extractor/ora.py @@ -12,8 +12,8 @@ from ..utils import ( class OraTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?ora\.tv/([^/]+/)*(?P<id>[^/\?#]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:ora\.tv|unsafespeech\.com)/([^/]+/)*(?P<id>[^/\?#]+)' + _TESTS = [{ 'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq', 'md5': 'fa33717591c631ec93b04b0e330df786', 'info_dict': { @@ -22,7 +22,10 @@ class OraTVIE(InfoExtractor): 'title': 'Vine & YouTube Stars Zach King & King Bach On Their Viral Videos!', 'description': 'md5:ebbc5b1424dd5dba7be7538148287ac1', } - } + }, { + 'url': 'http://www.unsafespeech.com/video/2016/5/10/student-self-censorship-and-the-thought-police-on-university-campuses-0_6622bnkppw4d', + 'only_matching': True, + }] def _real_extract(self, url): display_id = self._match_id(url) diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py index 66c75f8b3..4e3864f0d 100644 --- a/youtube_dl/extractor/orf.py +++ b/youtube_dl/extractor/orf.py @@ -185,6 +185,7 @@ class ORFFM4IE(InfoExtractor): 'timestamp': 1452456073, 'upload_date': '20160110', }, + 'skip': 'Live streams on FM4 got deleted soon', } def _real_extract(self, url): diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index f43e3a146..f6f423597 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -196,7 +196,7 @@ class PBSIE(InfoExtractor): _TESTS = [ { 'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/', - 'md5': 'ce1888486f0908d555a8093cac9a7362', + 'md5': '173dc391afd361fa72eab5d3d918968d', 'info_dict': { 'id': '2365006249', 'ext': 'mp4', @@ -204,13 +204,10 @@ class PBSIE(InfoExtractor): 'description': 'md5:36f341ae62e251b8f5bd2b754b95a071', 'duration': 3190, }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { 'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/', - 'md5': '143c98aa54a346738a3d78f54c925321', + 'md5': '6f722cb3c3982186d34b0f13374499c7', 'info_dict': { 'id': '2365297690', 'ext': 'mp4', @@ -218,9 +215,6 @@ class PBSIE(InfoExtractor): 'description': 'md5:4d3eaa01f94e61b3e73704735f1196d9', 'duration': 5050, }, - 'params': { - 'skip_download': True, # requires ffmpeg - } }, { 'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/', @@ -244,9 +238,6 @@ class PBSIE(InfoExtractor): 'duration': 6559, 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { 'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html', @@ -262,9 +253,6 @@ class PBSIE(InfoExtractor): 'upload_date': '20140122', 'age_limit': 10, }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { 'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/', @@ -290,6 +278,7 @@ class PBSIE(InfoExtractor): }, { 'url': 'http://www.pbs.org/video/2365245528/', + 'md5': '115223d41bd55cda8ae5cd5ed4e11497', 'info_dict': { 'id': '2365245528', 'display_id': '2365245528', @@ -299,15 +288,13 @@ class PBSIE(InfoExtractor): 'duration': 6851, 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { # Video embedded in iframe containing angle brackets as attribute's value (e.g. # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see # https://github.com/rg3/youtube-dl/issues/7059) 'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/', + 'md5': '84ced42850d78f1d4650297356e95e6f', 'info_dict': { 'id': '2365546844', 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business', @@ -317,9 +304,6 @@ class PBSIE(InfoExtractor): 'duration': 1480, 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { # Frontline video embedded via flp2012.js @@ -340,6 +324,7 @@ class PBSIE(InfoExtractor): { # Serves hd only via wigget/partnerplayer page 'url': 'http://www.pbs.org/video/2365641075/', + 'md5': 'acfd4c400b48149a44861cb16dd305cf', 'info_dict': { 'id': '2365641075', 'ext': 'mp4', @@ -348,9 +333,6 @@ class PBSIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg$', 'formats': 'mincount:8', }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, }, { 'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true', @@ -494,6 +476,7 @@ class PBSIE(InfoExtractor): info = video_info formats = [] + http_url = None for num, redirect in enumerate(redirects): redirect_id = redirect.get('eeid') @@ -514,13 +497,37 @@ class PBSIE(InfoExtractor): if determine_ext(format_url) == 'm3u8': formats.extend(self._extract_m3u8_formats( - format_url, display_id, 'mp4', preference=1, m3u8_id='hls')) + format_url, display_id, 'mp4', m3u8_id='hls', fatal=False)) else: formats.append({ 'url': format_url, 'format_id': redirect_id, }) + if re.search(r'^https?://.*(?:\d+k|baseline)', format_url): + http_url = format_url self._remove_duplicate_formats(formats) + m3u8_formats = list(filter( + lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + formats)) + if http_url: + for m3u8_format in m3u8_formats: + bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None) + # extract only the formats that we know that they will be available as http format. + # https://projects.pbs.org/confluence/display/coveapi/COVE+Video+Specifications + if not bitrate or bitrate not in ('400k', '800k', '1200k', '2500k'): + continue + f_url = re.sub(r'\d+k|baseline', bitrate, http_url) + # This may produce invalid links sometimes (e.g. + # http://www.pbs.org/wgbh/frontline/film/suicide-plan) + if not self._is_valid_url(f_url, display_id, 'http-%s video' % bitrate): + continue + f = m3u8_format.copy() + f.update({ + 'url': f_url, + 'format_id': m3u8_format['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) self._sort_formats(formats) rating_str = info.get('rating') @@ -535,6 +542,19 @@ class PBSIE(InfoExtractor): 'ext': 'ttml', 'url': closed_captions_url, }] + mobj = re.search(r'/(\d+)_Encoded\.dfxp', closed_captions_url) + if mobj: + ttml_caption_suffix, ttml_caption_id = mobj.group(0, 1) + ttml_caption_id = int(ttml_caption_id) + subtitles['en'].extend([{ + 'url': closed_captions_url.replace( + ttml_caption_suffix, '/%d_Encoded.srt' % (ttml_caption_id + 1)), + 'ext': 'srt', + }, { + 'url': closed_captions_url.replace( + ttml_caption_suffix, '/%d_Encoded.vtt' % (ttml_caption_id + 2)), + 'ext': 'vtt', + }]) # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc) # Try turning it to 'program - title' naming scheme if possible diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py index 514e9b433..75f5884a9 100644 --- a/youtube_dl/extractor/periscope.py +++ b/youtube_dl/extractor/periscope.py @@ -2,11 +2,15 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import parse_iso8601 +from ..utils import ( + parse_iso8601, + unescapeHTML, +) class PeriscopeIE(InfoExtractor): IE_DESC = 'Periscope' + IE_NAME = 'periscope' _VALID_URL = r'https?://(?:www\.)?periscope\.tv/[^/]+/(?P<id>[^/?#]+)' # Alive example URLs can be found here http://onperiscope.com/ _TESTS = [{ @@ -41,8 +45,11 @@ class PeriscopeIE(InfoExtractor): broadcast = broadcast_data['broadcast'] status = broadcast['status'] - uploader = broadcast.get('user_display_name') or broadcast_data.get('user', {}).get('display_name') - uploader_id = broadcast.get('user_id') or broadcast_data.get('user', {}).get('id') + user = broadcast_data.get('user', {}) + + uploader = broadcast.get('user_display_name') or user.get('display_name') + uploader_id = (broadcast.get('username') or user.get('username') or + broadcast.get('user_id') or user.get('id')) title = '%s - %s' % (uploader, status) if uploader else status state = broadcast.get('state').lower() @@ -79,3 +86,46 @@ class PeriscopeIE(InfoExtractor): 'thumbnails': thumbnails, 'formats': formats, } + + +class PeriscopeUserIE(InfoExtractor): + _VALID_URL = r'https?://www\.periscope\.tv/(?P<id>[^/]+)/?$' + IE_DESC = 'Periscope user videos' + IE_NAME = 'periscope:user' + + _TEST = { + 'url': 'https://www.periscope.tv/LularoeHusbandMike/', + 'info_dict': { + 'id': 'LularoeHusbandMike', + 'title': 'LULAROE HUSBAND MIKE', + 'description': 'md5:6cf4ec8047768098da58e446e82c82f0', + }, + # Periscope only shows videos in the last 24 hours, so it's possible to + # get 0 videos + 'playlist_mincount': 0, + } + + def _real_extract(self, url): + user_id = self._match_id(url) + + webpage = self._download_webpage(url, user_id) + + data_store = self._parse_json( + unescapeHTML(self._search_regex( + r'data-store=(["\'])(?P<data>.+?)\1', + webpage, 'data store', default='{}', group='data')), + user_id) + + user = data_store.get('User', {}).get('user', {}) + title = user.get('display_name') or user.get('username') + description = user.get('description') + + broadcast_ids = (data_store.get('UserBroadcastHistory', {}).get('broadcastIds') or + data_store.get('BroadcastCache', {}).get('broadcastIds', [])) + + entries = [ + self.url_result( + 'https://www.periscope.tv/%s/%s' % (user_id, broadcast_id)) + for broadcast_id in broadcast_ids] + + return self.playlist_result(entries, user_id, title, description) diff --git a/youtube_dl/extractor/pladform.py b/youtube_dl/extractor/pladform.py index bc559d1df..77e1211d6 100644 --- a/youtube_dl/extractor/pladform.py +++ b/youtube_dl/extractor/pladform.py @@ -49,7 +49,7 @@ class PladformIE(InfoExtractor): @staticmethod def _extract_url(webpage): mobj = re.search( - r'<iframe[^>]+src="(?P<url>(?:https?:)?//out\.pladform\.ru/player\?.+?)"', webpage) + r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//out\.pladform\.ru/player\?.+?)\1', webpage) if mobj: return mobj.group('url') diff --git a/youtube_dl/extractor/playwire.py b/youtube_dl/extractor/playwire.py index 6d138ef25..0bc743118 100644 --- a/youtube_dl/extractor/playwire.py +++ b/youtube_dl/extractor/playwire.py @@ -4,9 +4,8 @@ import re from .common import InfoExtractor from ..utils import ( - xpath_text, + dict_get, float_or_none, - int_or_none, ) @@ -23,6 +22,19 @@ class PlaywireIE(InfoExtractor): 'duration': 145.94, }, }, { + # m3u8 in f4m + 'url': 'http://config.playwire.com/21772/videos/v2/4840492/zeus.json', + 'info_dict': { + 'id': '4840492', + 'ext': 'mp4', + 'title': 'ITV EL SHOW FULL', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + # Multiple resolutions while bitrates missing 'url': 'http://cdn.playwire.com/11625/embed/85228.html', 'only_matching': True, }, { @@ -48,25 +60,10 @@ class PlaywireIE(InfoExtractor): thumbnail = content.get('poster') src = content['media']['f4m'] - f4m = self._download_xml(src, video_id) - base_url = xpath_text(f4m, './{http://ns.adobe.com/f4m/1.0}baseURL', 'base url', fatal=True) - formats = [] - for media in f4m.findall('./{http://ns.adobe.com/f4m/1.0}media'): - media_url = media.get('url') - if not media_url: - continue - tbr = int_or_none(media.get('bitrate')) - width = int_or_none(media.get('width')) - height = int_or_none(media.get('height')) - f = { - 'url': '%s/%s' % (base_url, media.attrib['url']), - 'tbr': tbr, - 'width': width, - 'height': height, - } - if not (tbr or width or height): - f['quality'] = 1 if '-hd.' in media_url else 0 - formats.append(f) + formats = self._extract_f4m_formats(src, video_id, m3u8_id='hls') + for a_format in formats: + if not dict_get(a_format, ['tbr', 'width', 'height']): + a_format['quality'] = 1 if '-hd.' in a_format['url'] else 0 self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/polskieradio.py b/youtube_dl/extractor/polskieradio.py new file mode 100644 index 000000000..f559b899f --- /dev/null +++ b/youtube_dl/extractor/polskieradio.py @@ -0,0 +1,99 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import ( + compat_str, + compat_urllib_parse_unquote, +) +from ..utils import ( + int_or_none, + strip_or_none, + unified_timestamp, +) + + +class PolskieRadioIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+/\d+/Artykul/(?P<id>[0-9]+)' + _TESTS = [{ + 'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943,Prof-Andrzej-Nowak-o-historii-nie-da-sie-myslec-beznamietnie', + 'info_dict': { + 'id': '1587943', + 'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie', + 'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5', + }, + 'playlist': [{ + 'md5': '2984ee6ce9046d91fc233bc1a864a09a', + 'info_dict': { + 'id': '1540576', + 'ext': 'mp3', + 'title': 'md5:d4623290d4ac983bf924061c75c23a0d', + 'timestamp': 1456594200, + 'upload_date': '20160227', + 'duration': 2364, + 'thumbnail': 're:^https?://static\.prsa\.pl/images/.*\.jpg$' + }, + }], + }, { + 'url': 'http://www.polskieradio.pl/265/5217/Artykul/1635803,Euro-2016-nie-ma-miejsca-na-blad-Polacy-graja-ze-Szwajcaria-o-cwiercfinal', + 'info_dict': { + 'id': '1635803', + 'title': 'Euro 2016: nie ma miejsca na błąd. Polacy grają ze Szwajcarią o ćwierćfinał', + 'description': 'md5:01cb7d0cad58664095d72b51a1ebada2', + }, + 'playlist_mincount': 12, + }, { + 'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis', + 'only_matching': True, + }, { + 'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943', + 'only_matching': True, + }, { + # with mp4 video + 'url': 'http://www.polskieradio.pl/9/299/Artykul/1634903,Brexit-Leszek-Miller-swiat-sie-nie-zawali-Europa-bedzie-trwac-dalej', + 'only_matching': True, + }] + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + content = self._search_regex( + r'(?s)<div[^>]+class="audio atarticle"[^>]*>(.+?)<script>', + webpage, 'content') + + timestamp = unified_timestamp(self._html_search_regex( + r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>', + webpage, 'timestamp', fatal=False)) + + thumbnail_url = self._og_search_thumbnail(webpage) + + entries = [] + + media_urls = set() + + for data_media in re.findall(r'<[^>]+data-media=({[^>]+})', content): + media = self._parse_json(data_media, playlist_id, fatal=False) + if not media.get('file') or not media.get('desc'): + continue + media_url = self._proto_relative_url(media['file'], 'http:') + if media_url in media_urls: + continue + media_urls.add(media_url) + entries.append({ + 'id': compat_str(media['id']), + 'url': media_url, + 'title': compat_urllib_parse_unquote(media['desc']), + 'duration': int_or_none(media.get('length')), + 'vcodec': 'none' if media.get('provider') == 'audio' else None, + 'timestamp': timestamp, + 'thumbnail': thumbnail_url + }) + + title = self._og_search_title(webpage).strip() + description = strip_or_none(self._og_search_description(webpage)) + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py index 39b53ecf6..8df12eec0 100644 --- a/youtube_dl/extractor/pornhd.py +++ b/youtube_dl/extractor/pornhd.py @@ -1,19 +1,32 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..utils import ( + ExtractorError, int_or_none, js_to_json, - qualities, ) class PornHdIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?' - _TEST = { + _TESTS = [{ + 'url': 'http://www.pornhd.com/videos/9864/selfie-restroom-masturbation-fun-with-chubby-cutie-hd-porn-video', + 'md5': 'c8b964b1f0a4b5f7f28ae3a5c9f86ad5', + 'info_dict': { + 'id': '9864', + 'display_id': 'selfie-restroom-masturbation-fun-with-chubby-cutie-hd-porn-video', + 'ext': 'mp4', + 'title': 'Restroom selfie masturbation', + 'description': 'md5:3748420395e03e31ac96857a8f125b2b', + 'thumbnail': 're:^https?://.*\.jpg', + 'view_count': int, + 'age_limit': 18, + } + }, { + # removed video 'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video', 'md5': '956b8ca569f7f4d8ec563e2c41598441', 'info_dict': { @@ -25,8 +38,9 @@ class PornHdIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg', 'view_count': int, 'age_limit': 18, - } - } + }, + 'skip': 'Not available anymore', + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -38,28 +52,38 @@ class PornHdIE(InfoExtractor): title = self._html_search_regex( [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)', r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title') - description = self._html_search_regex( - r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False) - view_count = int_or_none(self._html_search_regex( - r'(\d+) views\s*</span>', webpage, 'view count', fatal=False)) - thumbnail = self._search_regex( - r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False) - quality = qualities(['sd', 'hd']) - sources = json.loads(js_to_json(self._search_regex( + sources = self._parse_json(js_to_json(self._search_regex( r"(?s)'sources'\s*:\s*(\{.+?\})\s*\}[;,)]", - webpage, 'sources'))) + webpage, 'sources', default='{}')), video_id) + + if not sources: + message = self._html_search_regex( + r'(?s)<(div|p)[^>]+class="no-video"[^>]*>(?P<value>.+?)</\1', + webpage, 'error message', group='value') + raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True) + formats = [] - for qname, video_url in sources.items(): + for format_id, video_url in sources.items(): if not video_url: continue + height = int_or_none(self._search_regex( + r'^(\d+)[pP]', format_id, 'height', default=None)) formats.append({ 'url': video_url, - 'format_id': qname, - 'quality': quality(qname), + 'format_id': format_id, + 'height': height, }) self._sort_formats(formats) + description = self._html_search_regex( + r'<(div|p)[^>]+class="description"[^>]*>(?P<value>[^<]+)</\1', + webpage, 'description', fatal=False, group='value') + view_count = int_or_none(self._html_search_regex( + r'(\d+) views\s*<', webpage, 'view count', fatal=False)) + thumbnail = self._search_regex( + r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False) + return { 'id': video_id, 'display_id': display_id, diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 407ea08d4..d2c92531b 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -1,3 +1,4 @@ +# coding: utf-8 from __future__ import unicode_literals import itertools @@ -24,7 +25,15 @@ from ..aes import ( class PornHubIE(InfoExtractor): - _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)' + IE_DESC = 'PornHub and Thumbzilla' + _VALID_URL = r'''(?x) + https?:// + (?: + (?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)| + (?:www\.)?thumbzilla\.com/video/ + ) + (?P<id>[0-9a-z]+) + ''' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': '1e19b41231a02eba417839222ac9d58e', @@ -39,13 +48,47 @@ class PornHubIE(InfoExtractor): 'dislike_count': int, 'comment_count': int, 'age_limit': 18, - } + }, + }, { + # non-ASCII title + 'url': 'http://www.pornhub.com/view_video.php?viewkey=1331683002', + 'info_dict': { + 'id': '1331683002', + 'ext': 'mp4', + 'title': '重庆婷婷女王足交', + 'uploader': 'cj397186295', + 'duration': 1753, + 'view_count': int, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, + 'age_limit': 18, + }, + 'params': { + 'skip_download': True, + }, }, { 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d', 'only_matching': True, }, { + # removed at the request of cam4.com 'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862', 'only_matching': True, + }, { + # removed at the request of the copyright owner + 'url': 'http://www.pornhub.com/view_video.php?viewkey=788152859', + 'only_matching': True, + }, { + # removed by uploader + 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111', + 'only_matching': True, + }, { + # private video + 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7', + 'only_matching': True, + }, { + 'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex', + 'only_matching': True, }] @classmethod @@ -68,27 +111,33 @@ class PornHubIE(InfoExtractor): webpage = self._download_webpage(req, video_id) error_msg = self._html_search_regex( - r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>', - webpage, 'error message', default=None) + r'(?s)<div[^>]+class=(["\']).*?\b(?:removed|userMessageSection)\b.*?\1[^>]*>(?P<error>.+?)</div>', + webpage, 'error message', default=None, group='error') if error_msg: error_msg = re.sub(r'\s+', ' ', error_msg) raise ExtractorError( 'PornHub said: %s' % error_msg, expected=True, video_id=video_id) + # video_title from flashvars contains whitespace instead of non-ASCII (see + # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying + # on that anymore. + title = self._html_search_meta( + 'twitter:title', webpage, default=None) or self._search_regex( + (r'<h1[^>]+class=["\']title["\'][^>]*>(?P<title>[^<]+)', + r'<div[^>]+data-video-title=(["\'])(?P<title>.+?)\1', + r'shareTitle\s*=\s*(["\'])(?P<title>.+?)\1'), + webpage, 'title', group='title') + flashvars = self._parse_json( self._search_regex( r'var\s+flashvars_\d+\s*=\s*({.+?});', webpage, 'flashvars', default='{}'), video_id) if flashvars: - video_title = flashvars.get('video_title') thumbnail = flashvars.get('image_url') duration = int_or_none(flashvars.get('video_duration')) else: - video_title, thumbnail, duration = [None] * 3 - - if not video_title: - video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title') + title, thumbnail, duration = [None] * 3 video_uploader = self._html_search_regex( r'(?s)From: .+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<', @@ -137,7 +186,7 @@ class PornHubIE(InfoExtractor): return { 'id': video_id, 'uploader': video_uploader, - 'title': video_title, + 'title': title, 'thumbnail': thumbnail, 'duration': duration, 'view_count': view_count, diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py index 07d49d489..c6eee3b72 100644 --- a/youtube_dl/extractor/prosiebensat1.py +++ b/youtube_dl/extractor/prosiebensat1.py @@ -5,7 +5,7 @@ import re from hashlib import sha1 from .common import InfoExtractor -from ..compat import compat_urllib_parse_urlencode +from ..compat import compat_str from ..utils import ( ExtractorError, determine_ext, @@ -71,6 +71,7 @@ class ProSiebenSat1IE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'This video is unavailable', }, { 'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip', @@ -86,6 +87,7 @@ class ProSiebenSat1IE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'This video is unavailable', }, { 'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip', @@ -101,6 +103,7 @@ class ProSiebenSat1IE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'This video is unavailable', }, { 'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge', @@ -116,6 +119,7 @@ class ProSiebenSat1IE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'This video is unavailable', }, { 'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge', @@ -131,6 +135,7 @@ class ProSiebenSat1IE(InfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'This video is unavailable', }, { 'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip', @@ -227,70 +232,42 @@ class ProSiebenSat1IE(InfoExtractor): ] def _extract_clip(self, url, webpage): - clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id') + clip_id = self._html_search_regex( + self._CLIPID_REGEXES, webpage, 'clip id') access_token = 'prosieben' client_name = 'kolibri-2.0.19-splec4' client_location = url - videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse_urlencode({ - 'access_token': access_token, - 'client_location': client_location, - 'client_name': client_name, - 'ids': clip_id, - }) - - video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0] + video = self._download_json( + 'http://vas.sim-technik.de/vas/live/v2/videos', + clip_id, 'Downloading videos JSON', query={ + 'access_token': access_token, + 'client_location': client_location, + 'client_name': client_name, + 'ids': clip_id, + })[0] if video.get('is_protected') is True: raise ExtractorError('This video is DRM protected.', expected=True) duration = float_or_none(video.get('duration')) - source_ids = [source['id'] for source in video['sources']] - source_ids_str = ','.join(map(str, source_ids)) + source_ids = [compat_str(source['id']) for source in video['sources']] g = '01!8d8F_)r9]4s[qeuXfP%' + client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name]).encode('utf-8')).hexdigest() - client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name]) - .encode('utf-8')).hexdigest() - - sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse_urlencode({ - 'access_token': access_token, - 'client_id': client_id, - 'client_location': client_location, - 'client_name': client_name, - })) - - sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON') + sources = self._download_json( + 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id, + clip_id, 'Downloading sources JSON', query={ + 'access_token': access_token, + 'client_id': client_id, + 'client_location': client_location, + 'client_name': client_name, + }) server_id = sources['server_id'] - client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, - client_location, source_ids_str, g, client_name]) - .encode('utf-8')).hexdigest() - - url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse_urlencode({ - 'access_token': access_token, - 'client_id': client_id, - 'client_location': client_location, - 'client_name': client_name, - 'server_id': server_id, - 'source_ids': source_ids_str, - })) - - urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON') - title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title') - description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False) - thumbnail = self._og_search_thumbnail(webpage) - - upload_date = unified_strdate(self._html_search_regex( - self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None)) - - formats = [] - - urls_sources = urls['sources'] - if isinstance(urls_sources, dict): - urls_sources = urls_sources.values() def fix_bitrate(bitrate): bitrate = int_or_none(bitrate) @@ -298,37 +275,73 @@ class ProSiebenSat1IE(InfoExtractor): return None return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate - for source in urls_sources: - protocol = source['protocol'] - source_url = source['url'] - if protocol == 'rtmp' or protocol == 'rtmpe': - mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url) - if not mobj: - continue - path = mobj.group('path') - mp4colon_index = path.rfind('mp4:') - app = path[:mp4colon_index] - play_path = path[mp4colon_index:] - formats.append({ - 'url': '%s/%s' % (mobj.group('url'), app), - 'app': app, - 'play_path': play_path, - 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf', - 'page_url': 'http://www.prosieben.de', - 'vbr': fix_bitrate(source['bitrate']), - 'ext': 'mp4', - 'format_id': '%s_%s' % (source['cdn'], source['bitrate']), - }) - elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m': - formats.extend(self._extract_f4m_formats(source_url, clip_id)) - else: - formats.append({ - 'url': source_url, - 'vbr': fix_bitrate(source['bitrate']), + formats = [] + for source_id in source_ids: + client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, client_location, source_id, g, client_name]).encode('utf-8')).hexdigest() + urls = self._download_json( + 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id, + clip_id, 'Downloading urls JSON', fatal=False, query={ + 'access_token': access_token, + 'client_id': client_id, + 'client_location': client_location, + 'client_name': client_name, + 'server_id': server_id, + 'source_ids': source_id, }) - + if not urls: + continue + if urls.get('status_code') != 0: + raise ExtractorError('This video is unavailable', expected=True) + urls_sources = urls['sources'] + if isinstance(urls_sources, dict): + urls_sources = urls_sources.values() + for source in urls_sources: + source_url = source.get('url') + if not source_url: + continue + protocol = source.get('protocol') + mimetype = source.get('mimetype') + if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m': + formats.extend(self._extract_f4m_formats( + source_url, clip_id, f4m_id='hds', fatal=False)) + elif mimetype == 'application/x-mpegURL': + formats.extend(self._extract_m3u8_formats( + source_url, clip_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + else: + tbr = fix_bitrate(source['bitrate']) + if protocol in ('rtmp', 'rtmpe'): + mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url) + if not mobj: + continue + path = mobj.group('path') + mp4colon_index = path.rfind('mp4:') + app = path[:mp4colon_index] + play_path = path[mp4colon_index:] + formats.append({ + 'url': '%s/%s' % (mobj.group('url'), app), + 'app': app, + 'play_path': play_path, + 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf', + 'page_url': 'http://www.prosieben.de', + 'tbr': tbr, + 'ext': 'flv', + 'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''), + }) + else: + formats.append({ + 'url': source_url, + 'tbr': tbr, + 'format_id': 'http%s' % ('-%d' % tbr if tbr else ''), + }) self._sort_formats(formats) + description = self._html_search_regex( + self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) + upload_date = unified_strdate(self._html_search_regex( + self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None)) + return { 'id': clip_id, 'title': title, diff --git a/youtube_dl/extractor/r7.py b/youtube_dl/extractor/r7.py index 976c8feec..069dbfaed 100644 --- a/youtube_dl/extractor/r7.py +++ b/youtube_dl/extractor/r7.py @@ -2,22 +2,19 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - js_to_json, - unescapeHTML, - int_or_none, -) +from ..utils import int_or_none class R7IE(InfoExtractor): - _VALID_URL = r'''(?x)https?:// + _VALID_URL = r'''(?x) + https?:// (?: (?:[a-zA-Z]+)\.r7\.com(?:/[^/]+)+/idmedia/| noticias\.r7\.com(?:/[^/]+)+/[^/]+-| player\.r7\.com/video/i/ ) (?P<id>[\da-f]{24}) - ''' + ''' _TESTS = [{ 'url': 'http://videos.r7.com/policiais-humilham-suspeito-a-beira-da-morte-morre-com-dignidade-/idmedia/54e7050b0cf2ff57e0279389.html', 'md5': '403c4e393617e8e8ddc748978ee8efde', @@ -25,6 +22,7 @@ class R7IE(InfoExtractor): 'id': '54e7050b0cf2ff57e0279389', 'ext': 'mp4', 'title': 'Policiais humilham suspeito à beira da morte: "Morre com dignidade"', + 'description': 'md5:01812008664be76a6479aa58ec865b72', 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 98, 'like_count': int, @@ -44,45 +42,72 @@ class R7IE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage( - 'http://player.r7.com/video/i/%s' % video_id, video_id) + video = self._download_json( + 'http://player-api.r7.com/video/i/%s' % video_id, video_id) - item = self._parse_json(js_to_json(self._search_regex( - r'(?s)var\s+item\s*=\s*({.+?});', webpage, 'player')), video_id) - - title = unescapeHTML(item['title']) - thumbnail = item.get('init', {}).get('thumbUri') - duration = None - - statistics = item.get('statistics', {}) - like_count = int_or_none(statistics.get('likes')) - view_count = int_or_none(statistics.get('views')) + title = video['title'] formats = [] - for format_key, format_dict in item['playlist'][0].items(): - src = format_dict.get('src') - if not src: - continue - format_id = format_dict.get('format') or format_key - if duration is None: - duration = format_dict.get('duration') - if '.f4m' in src: - formats.extend(self._extract_f4m_formats(src, video_id, preference=-1)) - elif src.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(src, video_id, 'mp4', preference=-2)) - else: - formats.append({ - 'url': src, - 'format_id': format_id, - }) + media_url_hls = video.get('media_url_hls') + if media_url_hls: + formats.extend(self._extract_m3u8_formats( + media_url_hls, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + media_url = video.get('media_url') + if media_url: + f = { + 'url': media_url, + 'format_id': 'http', + } + # m3u8 format always matches the http format, let's copy metadata from + # one to another + m3u8_formats = list(filter( + lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + formats)) + if len(m3u8_formats) == 1: + f_copy = m3u8_formats[0].copy() + f_copy.update(f) + f_copy['protocol'] = 'http' + f = f_copy + formats.append(f) self._sort_formats(formats) + description = video.get('description') + thumbnail = video.get('thumb') + duration = int_or_none(video.get('media_duration')) + like_count = int_or_none(video.get('likes')) + view_count = int_or_none(video.get('views')) + return { 'id': video_id, 'title': title, + 'description': description, 'thumbnail': thumbnail, 'duration': duration, 'like_count': like_count, 'view_count': view_count, 'formats': formats, } + + +class R7ArticleIE(InfoExtractor): + _VALID_URL = r'https?://(?:[a-zA-Z]+)\.r7\.com/(?:[^/]+/)+[^/?#&]+-(?P<id>\d+)' + _TEST = { + 'url': 'http://tv.r7.com/record-play/balanco-geral/videos/policiais-humilham-suspeito-a-beira-da-morte-morre-com-dignidade-16102015', + 'only_matching': True, + } + + @classmethod + def suitable(cls, url): + return False if R7IE.suitable(url) else super(R7ArticleIE, cls).suitable(url) + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._search_regex( + r'<div[^>]+(?:id=["\']player-|class=["\']embed["\'][^>]+id=["\'])([\da-f]{24})', + webpage, 'video id') + + return self.url_result('http://player.r7.com/video/i/%s' % video_id, R7IE.ie_key()) diff --git a/youtube_dl/extractor/radiocanada.py b/youtube_dl/extractor/radiocanada.py new file mode 100644 index 000000000..8ec402646 --- /dev/null +++ b/youtube_dl/extractor/radiocanada.py @@ -0,0 +1,148 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + xpath_text, + find_xpath_attr, + determine_ext, + int_or_none, + unified_strdate, + xpath_element, + ExtractorError, + determine_protocol, +) + + +class RadioCanadaIE(InfoExtractor): + IE_NAME = 'radiocanada' + _VALID_URL = r'(?:radiocanada:|https?://ici\.radio-canada\.ca/widgets/mediaconsole/)(?P<app_code>[^:/]+)[:/](?P<id>[0-9]+)' + _TEST = { + 'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272', + 'info_dict': { + 'id': '7184272', + 'ext': 'mp4', + 'title': 'Le parcours du tireur capté sur vidéo', + 'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa', + 'upload_date': '20141023', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + app_code, video_id = re.match(self._VALID_URL, url).groups() + + device_types = ['ipad', 'android'] + if app_code != 'toutv': + device_types.append('flash') + + formats = [] + # TODO: extract f4m formats + # f4m formats can be extracted using flashhd device_type but they produce unplayable file + for device_type in device_types: + v_data = self._download_xml( + 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx', + video_id, note='Downloading %s XML' % device_type, query={ + 'appCode': app_code, + 'idMedia': video_id, + 'connectionType': 'broadband', + 'multibitrate': 'true', + 'deviceType': device_type, + # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction + 'paysJ391wsHjbOJwvCs26toz': 'CA', + 'bypasslock': 'NZt5K62gRqfc', + }, fatal=False) + v_url = xpath_text(v_data, 'url') + if not v_url: + continue + if v_url == 'null': + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, xpath_text(v_data, 'message')), expected=True) + ext = determine_ext(v_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + v_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + v_url, video_id, f4m_id='hds', fatal=False)) + else: + ext = determine_ext(v_url) + bitrates = xpath_element(v_data, 'bitrates') + for url_e in bitrates.findall('url'): + tbr = int_or_none(url_e.get('bitrate')) + if not tbr: + continue + f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url) + protocol = determine_protocol({'url': f_url}) + formats.append({ + 'format_id': '%s-%d' % (protocol, tbr), + 'url': f_url, + 'ext': 'flv' if protocol == 'rtmp' else ext, + 'protocol': protocol, + 'width': int_or_none(url_e.get('width')), + 'height': int_or_none(url_e.get('height')), + 'tbr': tbr, + }) + if protocol == 'rtsp': + base_url = self._search_regex( + r'rtsp://([^?]+)', f_url, 'base url', default=None) + if base_url: + base_url = 'http://' + base_url + formats.extend(self._extract_m3u8_formats( + base_url + '/playlist.m3u8', video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + base_url + '/manifest.f4m', video_id, + f4m_id='hds', fatal=False)) + self._sort_formats(formats) + + metadata = self._download_xml( + 'http://api.radio-canada.ca/metaMedia/v1/index.ashx', + video_id, note='Downloading metadata XML', query={ + 'appCode': app_code, + 'idMedia': video_id, + }) + + def get_meta(name): + el = find_xpath_attr(metadata, './/Meta', 'name', name) + return el.text if el is not None else None + + return { + 'id': video_id, + 'title': get_meta('Title'), + 'description': get_meta('Description') or get_meta('ShortDescription'), + 'thumbnail': get_meta('imageHR') or get_meta('imageMR') or get_meta('imageBR'), + 'duration': int_or_none(get_meta('length')), + 'series': get_meta('Emission'), + 'season_number': int_or_none('SrcSaison'), + 'episode_number': int_or_none('SrcEpisode'), + 'upload_date': unified_strdate(get_meta('Date')), + 'formats': formats, + } + + +class RadioCanadaAudioVideoIE(InfoExtractor): + 'radiocanada:audiovideo' + _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam', + 'info_dict': { + 'id': '7527184', + 'ext': 'mp4', + 'title': 'Barack Obama au Vietnam', + 'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam', + 'upload_date': '20160523', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + return self.url_result('radiocanada:medianet:%s' % self._match_id(url)) diff --git a/youtube_dl/extractor/radiojavan.py b/youtube_dl/extractor/radiojavan.py index 884c28420..ec4fa6e60 100644 --- a/youtube_dl/extractor/radiojavan.py +++ b/youtube_dl/extractor/radiojavan.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import( +from ..utils import ( unified_strdate, str_to_int, ) diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py index e36ce1aa1..dc640b1bc 100644 --- a/youtube_dl/extractor/rai.py +++ b/youtube_dl/extractor/rai.py @@ -1,47 +1,141 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( - ExtractorError, determine_ext, + ExtractorError, + find_xpath_attr, + fix_xml_ampersands, + int_or_none, parse_duration, unified_strdate, - int_or_none, + update_url_query, xpath_text, ) -class RaiTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+media/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' +class RaiBaseIE(InfoExtractor): + def _extract_relinker_formats(self, relinker_url, video_id): + formats = [] + + for platform in ('mon', 'flash', 'native'): + relinker = self._download_xml( + relinker_url, video_id, + note='Downloading XML metadata for platform %s' % platform, + transform_source=fix_xml_ampersands, + query={'output': 45, 'pl': platform}, + headers=self.geo_verification_headers()) + + media_url = find_xpath_attr(relinker, './url', 'type', 'content').text + if media_url == 'http://download.rai.it/video_no_available.mp4': + self.raise_geo_restricted() + + ext = determine_ext(media_url) + if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'): + continue + + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif ext == 'f4m': + manifest_url = update_url_query( + media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'), + {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'}) + formats.extend(self._extract_f4m_formats( + manifest_url, video_id, f4m_id='hds', fatal=False)) + else: + bitrate = int_or_none(xpath_text(relinker, 'bitrate')) + formats.append({ + 'url': media_url, + 'tbr': bitrate if bitrate > 0 else None, + 'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http', + }) + + return formats + + def _extract_from_content_id(self, content_id, base_url): + media = self._download_json( + 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id, + content_id, 'Downloading video JSON') + + thumbnails = [] + for image_type in ('image', 'image_medium', 'image_300'): + thumbnail_url = media.get(image_type) + if thumbnail_url: + thumbnails.append({ + 'url': compat_urlparse.urljoin(base_url, thumbnail_url), + }) + + formats = [] + media_type = media['type'] + if 'Audio' in media_type: + formats.append({ + 'format_id': media.get('formatoAudio'), + 'url': media['audioUrl'], + 'ext': media.get('formatoAudio'), + }) + elif 'Video' in media_type: + formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id)) + self._sort_formats(formats) + else: + raise ExtractorError('not a media file') + + subtitles = {} + captions = media.get('subtitlesUrl') + if captions: + STL_EXT = '.stl' + SRT_EXT = '.srt' + if captions.endswith(STL_EXT): + captions = captions[:-len(STL_EXT)] + SRT_EXT + subtitles['it'] = [{ + 'ext': 'srt', + 'url': captions, + }] + + return { + 'id': content_id, + 'title': media['name'], + 'description': media.get('desc'), + 'thumbnails': thumbnails, + 'uploader': media.get('author'), + 'upload_date': unified_strdate(media.get('date')), + 'duration': parse_duration(media.get('length')), + 'formats': formats, + 'subtitles': subtitles, + } + + +class RaiTVIE(RaiBaseIE): + _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' _TESTS = [ { 'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html', - 'md5': '96382709b61dd64a6b88e0f791e6df4c', + 'md5': '8970abf8caf8aef4696e7b1f2adfc696', 'info_dict': { 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Report del 07/04/2014', 'description': 'md5:f27c544694cacb46a078db84ec35d2d9', 'upload_date': '20140407', 'duration': 6160, + 'thumbnail': 're:^https?://.*\.jpg$', } }, { + # no m3u8 stream 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html', - 'md5': 'd9751b78eac9710d62c2447b224dea39', + # HDS download, MD5 is unstable 'info_dict': { 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9', 'ext': 'flv', 'title': 'TG PRIMO TEMPO', 'upload_date': '20140612', 'duration': 1758, + 'thumbnail': 're:^https?://.*\.jpg$', }, + 'skip': 'Geo-restricted to Italy', }, { 'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html', @@ -67,127 +161,70 @@ class RaiTVIE(InfoExtractor): }, { 'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html', - 'md5': '496ab63e420574447f70d02578333437', + 'md5': 'e57493e1cb8bc7c564663f363b171847', 'info_dict': { 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Il Candidato - Primo episodio: "Le Primarie"', 'description': 'md5:364b604f7db50594678f483353164fb8', 'upload_date': '20140923', 'duration': 386, + 'thumbnail': 're:^https?://.*\.jpg$', } }, ] def _real_extract(self, url): video_id = self._match_id(url) - media = self._download_json( - 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % video_id, - video_id, 'Downloading video JSON') - - thumbnails = [] - for image_type in ('image', 'image_medium', 'image_300'): - thumbnail_url = media.get(image_type) - if thumbnail_url: - thumbnails.append({ - 'url': thumbnail_url, - }) - - subtitles = [] - formats = [] - media_type = media['type'] - if 'Audio' in media_type: - formats.append({ - 'format_id': media.get('formatoAudio'), - 'url': media['audioUrl'], - 'ext': media.get('formatoAudio'), - }) - elif 'Video' in media_type: - def fix_xml(xml): - return xml.replace(' tag elementi', '').replace('>/', '</') - - relinker = self._download_xml( - media['mediaUri'] + '&output=43', - video_id, transform_source=fix_xml) - - has_subtitle = False - - for element in relinker.findall('element'): - media_url = xpath_text(element, 'url') - ext = determine_ext(media_url) - content_type = xpath_text(element, 'content-type') - if ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - media_url, video_id, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) - elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - media_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', - video_id, f4m_id='hds', fatal=False)) - elif ext == 'stl': - has_subtitle = True - elif content_type.startswith('video/'): - bitrate = int_or_none(xpath_text(element, 'bitrate')) - formats.append({ - 'url': media_url, - 'tbr': bitrate if bitrate > 0 else None, - 'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http', - }) - elif content_type.startswith('image/'): - thumbnails.append({ - 'url': media_url, - }) - - self._sort_formats(formats) - if has_subtitle: - webpage = self._download_webpage(url, video_id) - subtitles = self._get_subtitles(video_id, webpage) - else: - raise ExtractorError('not a media file') + return self._extract_from_content_id(video_id, url) - return { - 'id': video_id, - 'title': media['name'], - 'description': media.get('desc'), - 'thumbnails': thumbnails, - 'uploader': media.get('author'), - 'upload_date': unified_strdate(media.get('date')), - 'duration': parse_duration(media.get('length')), - 'formats': formats, - 'subtitles': subtitles, - } - def _get_subtitles(self, video_id, webpage): - subtitles = {} - m = re.search(r'<meta name="closedcaption" content="(?P<captions>[^"]+)"', webpage) - if m: - captions = m.group('captions') - STL_EXT = '.stl' - SRT_EXT = '.srt' - if captions.endswith(STL_EXT): - captions = captions[:-len(STL_EXT)] + SRT_EXT - subtitles['it'] = [{ - 'ext': 'srt', - 'url': 'http://www.rai.tv%s' % compat_urllib_parse.quote(captions), - }] - return subtitles - - -class RaiIE(InfoExtractor): +class RaiIE(RaiBaseIE): _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' _TESTS = [ { 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', - 'md5': 'e0e7a8a131e249d1aa0ebf270d1d8db7', + 'md5': '2dd727e61114e1ee9c47f0da6914e178', 'info_dict': { 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Il pacco', 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', 'upload_date': '20141221', }, - } + }, + { + # Direct relinker URL + 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc', + 'ext': 'flv', + 'title': 'EuroNews', + }, + 'skip': 'Geo-restricted to Italy', + }, + { + # Embedded content item ID + 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined', + 'md5': '84c1135ce960e8822ae63cec34441d63', + 'info_dict': { + 'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8', + 'ext': 'mp4', + 'title': 'TG1 ore 20:00 del 02/07/2016', + 'upload_date': '20160702', + }, + }, + { + 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9', + 'ext': 'flv', + 'title': 'La diretta di Rainews24', + }, + }, ] @classmethod @@ -201,7 +238,30 @@ class RaiIE(InfoExtractor): iframe_url = self._search_regex( [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"', r'drawMediaRaiTV\(["\'](.+?)["\']'], - webpage, 'iframe') - if not iframe_url.startswith('http'): - iframe_url = compat_urlparse.urljoin(url, iframe_url) - return self.url_result(iframe_url) + webpage, 'iframe', default=None) + if iframe_url: + if not iframe_url.startswith('http'): + iframe_url = compat_urlparse.urljoin(url, iframe_url) + return self.url_result(iframe_url) + + content_item_id = self._search_regex( + r'initEdizione\((?P<q1>[\'"])ContentItem-(?P<content_id>[^\'"]+)(?P=q1)', + webpage, 'content item ID', group='content_id', default=None) + if content_item_id: + return self._extract_from_content_id(content_item_id, url) + + relinker_url = compat_urlparse.urljoin(url, self._search_regex( + r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)', + webpage, 'relinker URL', group='url')) + formats = self._extract_relinker_formats(relinker_url, video_id) + self._sort_formats(formats) + + title = self._search_regex( + r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1', + webpage, 'title', group='title', default=None) or self._og_search_title(webpage) + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + } diff --git a/youtube_dl/extractor/rds.py b/youtube_dl/extractor/rds.py index 796adfdf9..bf200ea4d 100644 --- a/youtube_dl/extractor/rds.py +++ b/youtube_dl/extractor/rds.py @@ -1,23 +1,23 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( parse_duration, parse_iso8601, + js_to_json, ) +from ..compat import compat_str class RDSIE(InfoExtractor): IE_DESC = 'RDS.ca' - _VALID_URL = r'https?://(?:www\.)?rds\.ca/vid(?:[eé]|%C3%A9)os/(?:[^/]+/)*(?P<display_id>[^/]+)-(?P<id>\d+\.\d+)' + _VALID_URL = r'https?://(?:www\.)?rds\.ca/vid(?:[eé]|%C3%A9)os/(?:[^/]+/)*(?P<id>[^/]+)-\d+\.\d+' _TESTS = [{ 'url': 'http://www.rds.ca/videos/football/nfl/fowler-jr-prend-la-direction-de-jacksonville-3.1132799', 'info_dict': { - 'id': '3.1132799', + 'id': '604333', 'display_id': 'fowler-jr-prend-la-direction-de-jacksonville', 'ext': 'mp4', 'title': 'Fowler Jr. prend la direction de Jacksonville', @@ -33,22 +33,17 @@ class RDSIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') + display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - # TODO: extract f4m from 9c9media.com - video_url = self._search_regex( - r'<span[^>]+itemprop="contentURL"[^>]+content="([^"]+)"', - webpage, 'video url') - - title = self._og_search_title(webpage) or self._html_search_meta( + item = self._parse_json(self._search_regex(r'(?s)itemToPush\s*=\s*({.+?});', webpage, 'item'), display_id, js_to_json) + video_id = compat_str(item['id']) + title = item.get('title') or self._og_search_title(webpage) or self._html_search_meta( 'title', webpage, 'title', fatal=True) description = self._og_search_description(webpage) or self._html_search_meta( 'description', webpage, 'description') - thumbnail = self._og_search_thumbnail(webpage) or self._search_regex( + thumbnail = item.get('urlImageBig') or self._og_search_thumbnail(webpage) or self._search_regex( [r'<link[^>]+itemprop="thumbnailUrl"[^>]+href="([^"]+)"', r'<span[^>]+itemprop="thumbnailUrl"[^>]+content="([^"]+)"'], webpage, 'thumbnail', fatal=False) @@ -61,13 +56,15 @@ class RDSIE(InfoExtractor): age_limit = self._family_friendly_search(webpage) return { + '_type': 'url_transparent', 'id': video_id, 'display_id': display_id, - 'url': video_url, + 'url': '9c9media:rds_web:%s' % video_id, 'title': title, 'description': description, 'thumbnail': thumbnail, 'timestamp': timestamp, 'duration': duration, 'age_limit': age_limit, + 'ie_key': 'NineCNineMedia', } diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py index 7ba41ba59..721fc3a9e 100644 --- a/youtube_dl/extractor/redtube.py +++ b/youtube_dl/extractor/redtube.py @@ -1,7 +1,12 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + int_or_none, + str_to_int, + unified_strdate, +) class RedTubeIE(InfoExtractor): @@ -13,6 +18,9 @@ class RedTubeIE(InfoExtractor): 'id': '66418', 'ext': 'mp4', 'title': 'Sucked on a toilet', + 'upload_date': '20120831', + 'duration': 596, + 'view_count': int, 'age_limit': 18, } } @@ -24,12 +32,39 @@ class RedTubeIE(InfoExtractor): if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']): raise ExtractorError('Video %s has been removed' % video_id, expected=True) - video_url = self._html_search_regex( - r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL') - video_title = self._html_search_regex( - r'<h1 class="videoTitle[^"]*">(.+?)</h1>', - webpage, 'title') - video_thumbnail = self._og_search_thumbnail(webpage) + title = self._html_search_regex( + (r'<h1 class="videoTitle[^"]*">(?P<title>.+?)</h1>', + r'videoTitle\s*:\s*(["\'])(?P<title>)\1'), + webpage, 'title', group='title') + + formats = [] + sources = self._parse_json( + self._search_regex( + r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'), + video_id, fatal=False) + if sources and isinstance(sources, dict): + for format_id, format_url in sources.items(): + if format_url: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'height': int_or_none(format_id), + }) + else: + video_url = self._html_search_regex( + r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL') + formats.append({'url': video_url}) + self._sort_formats(formats) + + thumbnail = self._og_search_thumbnail(webpage) + upload_date = unified_strdate(self._search_regex( + r'<span[^>]+class="added-time"[^>]*>ADDED ([^<]+)<', + webpage, 'upload date', fatal=False)) + duration = int_or_none(self._search_regex( + r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False)) + view_count = str_to_int(self._search_regex( + r'<span[^>]*>VIEWS</span></td>\s*<td>([\d,.]+)', + webpage, 'view count', fatal=False)) # No self-labeling, but they describe themselves as # "Home of Videos Porno" @@ -37,9 +72,12 @@ class RedTubeIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, 'ext': 'mp4', - 'title': video_title, - 'thumbnail': video_thumbnail, + 'title': title, + 'thumbnail': thumbnail, + 'upload_date': upload_date, + 'duration': duration, + 'view_count': view_count, 'age_limit': age_limit, + 'formats': formats, } diff --git a/youtube_dl/extractor/reuters.py b/youtube_dl/extractor/reuters.py new file mode 100644 index 000000000..961d504eb --- /dev/null +++ b/youtube_dl/extractor/reuters.py @@ -0,0 +1,69 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + js_to_json, + int_or_none, + unescapeHTML, +) + + +class ReutersIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?reuters\.com/.*?\?.*?videoId=(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.reuters.com/video/2016/05/20/san-francisco-police-chief-resigns?videoId=368575562', + 'md5': '8015113643a0b12838f160b0b81cc2ee', + 'info_dict': { + 'id': '368575562', + 'ext': 'mp4', + 'title': 'San Francisco police chief resigns', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage( + 'http://www.reuters.com/assets/iframe/yovideo?videoId=%s' % video_id, video_id) + video_data = js_to_json(self._search_regex( + r'(?s)Reuters\.yovideo\.drawPlayer\(({.*?})\);', + webpage, 'video data')) + + def get_json_value(key, fatal=False): + return self._search_regex('"%s"\s*:\s*"([^"]+)"' % key, video_data, key, fatal=fatal) + + title = unescapeHTML(get_json_value('title', fatal=True)) + mmid, fid = re.search(r',/(\d+)\?f=(\d+)', get_json_value('flv', fatal=True)).groups() + + mas_data = self._download_json( + 'http://mas-e.cds1.yospace.com/mas/%s/%s?trans=json' % (mmid, fid), + video_id, transform_source=js_to_json) + formats = [] + for f in mas_data: + f_url = f.get('url') + if not f_url: + continue + method = f.get('method') + if method == 'hls': + formats.extend(self._extract_m3u8_formats( + f_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + else: + container = f.get('container') + ext = '3gp' if method == 'mobile' else container + formats.append({ + 'format_id': ext, + 'url': f_url, + 'ext': ext, + 'container': container if method != 'mobile' else None, + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'thumbnail': get_json_value('thumb'), + 'duration': int_or_none(get_json_value('seconds')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/revision3.py b/youtube_dl/extractor/revision3.py index 99979ebe1..833d8a2f0 100644 --- a/youtube_dl/extractor/revision3.py +++ b/youtube_dl/extractor/revision3.py @@ -13,8 +13,64 @@ from ..utils import ( ) +class Revision3EmbedIE(InfoExtractor): + IE_NAME = 'revision3:embed' + _VALID_URL = r'(?:revision3:(?:(?P<playlist_type>[^:]+):)?|https?://(?:(?:(?:www|embed)\.)?(?:revision3|animalist)|(?:(?:api|embed)\.)?seekernetwork)\.com/player/embed\?videoId=)(?P<playlist_id>\d+)' + _TEST = { + 'url': 'http://api.seekernetwork.com/player/embed?videoId=67558', + 'md5': '83bcd157cab89ad7318dd7b8c9cf1306', + 'info_dict': { + 'id': '67558', + 'ext': 'mp4', + 'title': 'The Pros & Cons Of Zoos', + 'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?', + 'uploader_id': 'dnews', + 'uploader': 'DNews', + } + } + _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + playlist_id = mobj.group('playlist_id') + playlist_type = mobj.group('playlist_type') or 'video_id' + video_data = self._download_json( + 'http://revision3.com/api/getPlaylist.json', playlist_id, query={ + 'api_key': self._API_KEY, + 'codecs': 'h264,vp8,theora', + playlist_type: playlist_id, + })['items'][0] + + formats = [] + for vcodec, media in video_data['media'].items(): + for quality_id, quality in media.items(): + if quality_id == 'hls': + formats.extend(self._extract_m3u8_formats( + quality['url'], playlist_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + else: + formats.append({ + 'url': quality['url'], + 'format_id': '%s-%s' % (vcodec, quality_id), + 'tbr': int_or_none(quality.get('bitrate')), + 'vcodec': vcodec, + }) + self._sort_formats(formats) + + return { + 'id': playlist_id, + 'title': unescapeHTML(video_data['title']), + 'description': unescapeHTML(video_data.get('summary')), + 'uploader': video_data.get('show', {}).get('name'), + 'uploader_id': video_data.get('show', {}).get('slug'), + 'duration': int_or_none(video_data.get('duration')), + 'formats': formats, + } + + class Revision3IE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|testtube|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)' + IE_NAME = 'revision' + _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)' _TESTS = [{ 'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016', 'md5': 'd94a72d85d0a829766de4deb8daaf7df', @@ -32,52 +88,14 @@ class Revision3IE(InfoExtractor): } }, { # Show - 'url': 'http://testtube.com/brainstuff', - 'info_dict': { - 'id': '251', - 'title': 'BrainStuff', - 'description': 'Whether the topic is popcorn or particle physics, you can count on the HowStuffWorks team to explore-and explain-the everyday science in the world around us on BrainStuff.', - }, - 'playlist_mincount': 93, - }, { - 'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial', - 'info_dict': { - 'id': '58227', - 'display_id': 'dnews/5-weird-ways-plants-can-eat-animals', - 'duration': 275, - 'ext': 'webm', - 'title': '5 Weird Ways Plants Can Eat Animals', - 'description': 'Why have some plants evolved to eat meat?', - 'upload_date': '20150120', - 'timestamp': 1421763300, - 'uploader': 'DNews', - 'uploader_id': 'dnews', - }, - }, { - 'url': 'http://testtube.com/tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min', - 'info_dict': { - 'id': '71618', - 'ext': 'mp4', - 'display_id': 'tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min', - 'title': 'The Israel-Palestine Conflict Explained in Ten Minutes', - 'description': 'If you\'d like to learn about the struggle between Israelis and Palestinians, this video is a great place to start', - 'uploader': 'Editors\' Picks', - 'uploader_id': 'tt-editors-picks', - 'timestamp': 1453309200, - 'upload_date': '20160120', - }, - 'add_ie': ['Youtube'], + 'url': 'http://revision3.com/variant', + 'only_matching': True, }, { # Tag - 'url': 'http://testtube.com/tech-news', - 'info_dict': { - 'id': '21018', - 'title': 'tech news', - }, - 'playlist_mincount': 9, + 'url': 'http://revision3.com/vr', + 'only_matching': True, }] _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s' - _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62' def _real_extract(self, url): domain, display_id = re.match(self._VALID_URL, url).groups() @@ -119,33 +137,9 @@ class Revision3IE(InfoExtractor): }) return info - video_data = self._download_json( - 'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id), - video_id)['items'][0] - - formats = [] - for vcodec, media in video_data['media'].items(): - for quality_id, quality in media.items(): - if quality_id == 'hls': - formats.extend(self._extract_m3u8_formats( - quality['url'], video_id, 'mp4', - 'm3u8_native', m3u8_id='hls', fatal=False)) - else: - formats.append({ - 'url': quality['url'], - 'format_id': '%s-%s' % (vcodec, quality_id), - 'tbr': int_or_none(quality.get('bitrate')), - 'vcodec': vcodec, - }) - self._sort_formats(formats) - info.update({ - 'title': unescapeHTML(video_data['title']), - 'description': unescapeHTML(video_data.get('summary')), - 'uploader': video_data.get('show', {}).get('name'), - 'uploader_id': video_data.get('show', {}).get('slug'), - 'duration': int_or_none(video_data.get('duration')), - 'formats': formats, + '_type': 'url_transparent', + 'url': 'revision3:%s' % video_id, }) return info else: diff --git a/youtube_dl/extractor/rockstargames.py b/youtube_dl/extractor/rockstargames.py new file mode 100644 index 000000000..48128e219 --- /dev/null +++ b/youtube_dl/extractor/rockstargames.py @@ -0,0 +1,69 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_iso8601, +) + + +class RockstarGamesIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?rockstargames\.com/videos(?:/video/|#?/?\?.*\bvideo=)(?P<id>\d+)' + _TESTS = [{ + 'url': 'https://www.rockstargames.com/videos/video/11544/', + 'md5': '03b5caa6e357a4bd50e3143fc03e5733', + 'info_dict': { + 'id': '11544', + 'ext': 'mp4', + 'title': 'Further Adventures in Finance and Felony Trailer', + 'description': 'md5:6d31f55f30cb101b5476c4a379e324a3', + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1464876000, + 'upload_date': '20160602', + } + }, { + 'url': 'http://www.rockstargames.com/videos#/?video=48', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'https://www.rockstargames.com/videoplayer/videos/get-video.json', + video_id, query={ + 'id': video_id, + 'locale': 'en_us', + })['video'] + + title = video['title'] + + formats = [] + for video in video['files_processed']['video/mp4']: + if not video.get('src'): + continue + resolution = video.get('resolution') + height = int_or_none(self._search_regex( + r'^(\d+)[pP]$', resolution or '', 'height', default=None)) + formats.append({ + 'url': self._proto_relative_url(video['src']), + 'format_id': resolution, + 'height': height, + }) + + if not formats: + youtube_id = video.get('youtube_id') + if youtube_id: + return self.url_result(youtube_id, 'Youtube') + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': video.get('description'), + 'thumbnail': self._proto_relative_url(video.get('screencap')), + 'timestamp': parse_iso8601(video.get('created')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/rtlnl.py b/youtube_dl/extractor/rtlnl.py index 543d94417..4d612b5e3 100644 --- a/youtube_dl/extractor/rtlnl.py +++ b/youtube_dl/extractor/rtlnl.py @@ -20,18 +20,19 @@ class RtlNlIE(InfoExtractor): (?P<id>[0-9a-f-]+)''' _TESTS = [{ - 'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677', - 'md5': 'cc16baa36a6c169391f0764fa6b16654', + 'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/82b1aad1-4a14-3d7b-b554-b0aed1b2c416', + 'md5': '473d1946c1fdd050b2c0161a4b13c373', 'info_dict': { - 'id': '6e4203a6-0a5e-3596-8424-c599a59e0677', + 'id': '82b1aad1-4a14-3d7b-b554-b0aed1b2c416', 'ext': 'mp4', - 'title': 'RTL Nieuws - Laat', - 'description': 'md5:6b61f66510c8889923b11f2778c72dc5', - 'timestamp': 1408051800, - 'upload_date': '20140814', - 'duration': 576.880, + 'title': 'RTL Nieuws', + 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e', + 'timestamp': 1461951000, + 'upload_date': '20160429', + 'duration': 1167.96, }, }, { + # best format avaialble a3t 'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed/autoplay=false', 'md5': 'dea7474214af1271d91ef332fb8be7ea', 'info_dict': { @@ -39,18 +40,19 @@ class RtlNlIE(InfoExtractor): 'ext': 'mp4', 'timestamp': 1424039400, 'title': 'RTL Nieuws - Nieuwe beelden Kopenhagen: chaos direct na aanslag', - 'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$', + 'thumbnail': 're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$', 'upload_date': '20150215', 'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.', } }, { # empty synopsis and missing episodes (see https://github.com/rg3/youtube-dl/issues/6275) + # best format available nettv 'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a/autoplay=false', 'info_dict': { 'id': 'f536aac0-1dc3-4314-920e-3bd1c5b3811a', 'ext': 'mp4', 'title': 'RTL Nieuws - Meer beelden van overval juwelier', - 'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$', + 'thumbnail': 're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$', 'timestamp': 1437233400, 'upload_date': '20150718', 'duration': 30.474, @@ -94,22 +96,46 @@ class RtlNlIE(InfoExtractor): videopath = material['videopath'] m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath - formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4') + formats = self._extract_m3u8_formats( + m3u8_url, uuid, 'mp4', m3u8_id='hls', fatal=False) video_urlpart = videopath.split('/adaptive/')[1][:-5] PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4' - formats.extend([ - { - 'url': PG_URL_TEMPLATE % ('a2m', video_urlpart), - 'format_id': 'pg-sd', - }, - { - 'url': PG_URL_TEMPLATE % ('a3m', video_urlpart), - 'format_id': 'pg-hd', - 'quality': 0, + PG_FORMATS = ( + ('a2t', 512, 288), + ('a3t', 704, 400), + ('nettv', 1280, 720), + ) + + def pg_format(format_id, width, height): + return { + 'url': PG_URL_TEMPLATE % (format_id, video_urlpart), + 'format_id': 'pg-%s' % format_id, + 'protocol': 'http', + 'width': width, + 'height': height, } - ]) + + if not formats: + formats = [pg_format(*pg_tuple) for pg_tuple in PG_FORMATS] + else: + pg_formats = [] + for format_id, width, height in PG_FORMATS: + try: + # Find hls format with the same width and height corresponding + # to progressive format and copy metadata from it. + f = next(f for f in formats if f.get('height') == height) + # hls formats may have invalid width + f['width'] = width + f_copy = f.copy() + f_copy.update(pg_format(format_id, width, height)) + pg_formats.append(f_copy) + except StopIteration: + # Missing hls format does mean that no progressive format with + # such width and height exists either. + pass + formats.extend(pg_formats) self._sort_formats(formats) thumbnails = [] diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py index 79af47715..f11e3588b 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -6,6 +6,9 @@ import re import time from .common import InfoExtractor +from ..compat import ( + compat_struct_unpack, +) from ..utils import ( ExtractorError, float_or_none, @@ -13,7 +16,6 @@ from ..utils import ( remove_start, sanitized_Request, std_headers, - struct_unpack, ) @@ -21,7 +23,7 @@ def _decrypt_url(png): encrypted_data = base64.b64decode(png.encode('utf-8')) text_index = encrypted_data.find(b'tEXt') text_chunk = encrypted_data[text_index - 4:] - length = struct_unpack('!I', text_chunk[:4])[0] + length = compat_struct_unpack('!I', text_chunk[:4])[0] # Use bytearray to get integers when iterating in both python 2.x and 3.x data = bytearray(text_chunk[8:8 + length]) data = [chr(b) for b in data if b != 0] @@ -62,7 +64,7 @@ def _decrypt_url(png): class RTVEALaCartaIE(InfoExtractor): IE_NAME = 'rtve.es:alacarta' IE_DESC = 'RTVE a la carta' - _VALID_URL = r'https?://www\.rtve\.es/(m/)?alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)' + _VALID_URL = r'https?://www\.rtve\.es/(m/)?(alacarta/videos|filmoteca)/[^/]+/[^/]+/(?P<id>\d+)' _TESTS = [{ 'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/', @@ -85,6 +87,9 @@ class RTVEALaCartaIE(InfoExtractor): }, { 'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve', 'only_matching': True, + }, { + 'url': 'http://www.rtve.es/filmoteca/no-do/not-1-introduccion-primer-noticiario-espanol/1465256/', + 'only_matching': True, }] def _real_initialize(self): diff --git a/youtube_dl/extractor/rtvnh.py b/youtube_dl/extractor/rtvnh.py index 4896d09d6..f6454c6b0 100644 --- a/youtube_dl/extractor/rtvnh.py +++ b/youtube_dl/extractor/rtvnh.py @@ -9,7 +9,7 @@ class RTVNHIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?rtvnh\.nl/video/(?P<id>[0-9]+)' _TEST = { 'url': 'http://www.rtvnh.nl/video/131946', - 'md5': '6e1d0ab079e2a00b6161442d3ceacfc1', + 'md5': 'cdbec9f44550763c8afc96050fa747dc', 'info_dict': { 'id': '131946', 'ext': 'mp4', @@ -29,15 +29,29 @@ class RTVNHIE(InfoExtractor): raise ExtractorError( '%s returned error code %d' % (self.IE_NAME, status), expected=True) - formats = self._extract_smil_formats( - 'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id, fatal=False) - - for item in meta['source']['fb']: - if item.get('type') == 'hls': - formats.extend(self._extract_m3u8_formats( - item['file'], video_id, ext='mp4', entry_protocol='m3u8_native')) - elif item.get('type') == '': - formats.append({'url': item['file']}) + formats = [] + rtmp_formats = self._extract_smil_formats( + 'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id) + formats.extend(rtmp_formats) + + for rtmp_format in rtmp_formats: + rtmp_url = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path']) + rtsp_format = rtmp_format.copy() + del rtsp_format['play_path'] + del rtsp_format['ext'] + rtsp_format.update({ + 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'), + 'url': rtmp_url.replace('rtmp://', 'rtsp://'), + 'protocol': 'rtsp', + }) + formats.append(rtsp_format) + http_base_url = rtmp_url.replace('rtmp://', 'http://') + formats.extend(self._extract_m3u8_formats( + http_base_url + '/playlist.m3u8', video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + http_base_url + '/manifest.f4m', + video_id, f4m_id='hds', fatal=False)) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/sandia.py b/youtube_dl/extractor/sandia.py index 759898a49..96e43af84 100644 --- a/youtube_dl/extractor/sandia.py +++ b/youtube_dl/extractor/sandia.py @@ -1,18 +1,12 @@ # coding: utf-8 from __future__ import unicode_literals -import itertools import json -import re from .common import InfoExtractor -from ..compat import compat_urlparse from ..utils import ( int_or_none, - js_to_json, mimetype2ext, - sanitized_Request, - unified_strdate, ) @@ -27,7 +21,8 @@ class SandiaIE(InfoExtractor): 'ext': 'mp4', 'title': 'Xyce Software Training - Section 1', 'description': 're:(?s)SAND Number: SAND 2013-7800.{200,}', - 'upload_date': '20120904', + 'upload_date': '20120409', + 'timestamp': 1333983600, 'duration': 7794, } } @@ -35,81 +30,36 @@ class SandiaIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - req = sanitized_Request(url) - req.add_header('Cookie', 'MediasitePlayerCaps=ClientPlugins=4') - webpage = self._download_webpage(req, video_id) + presentation_data = self._download_json( + 'http://digitalops.sandia.gov/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions', + video_id, data=json.dumps({ + 'getPlayerOptionsRequest': { + 'ResourceId': video_id, + 'QueryString': '', + } + }), headers={ + 'Content-Type': 'application/json; charset=utf-8', + })['d']['Presentation'] - js_path = self._search_regex( - r'<script type="text/javascript" src="(/Mediasite/FileServer/Presentation/[^"]+)"', - webpage, 'JS code URL') - js_url = compat_urlparse.urljoin(url, js_path) - - js_code = self._download_webpage( - js_url, video_id, note='Downloading player') - - def extract_str(key, **args): - return self._search_regex( - r'Mediasite\.PlaybackManifest\.%s\s*=\s*(.+);\s*?\n' % re.escape(key), - js_code, key, **args) - - def extract_data(key, **args): - data_json = extract_str(key, **args) - if data_json is None: - return data_json - return self._parse_json( - data_json, video_id, transform_source=js_to_json) + title = presentation_data['Title'] formats = [] - for i in itertools.count(): - fd = extract_data('VideoUrls[%d]' % i, default=None) - if fd is None: - break - formats.append({ - 'format_id': '%s' % i, - 'format_note': fd['MimeType'].partition('/')[2], - 'ext': mimetype2ext(fd['MimeType']), - 'url': fd['Location'], - 'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None, - }) + for stream in presentation_data.get('Streams', []): + for fd in stream.get('VideoUrls', []): + formats.append({ + 'format_id': fd['MediaType'], + 'format_note': fd['MimeType'].partition('/')[2], + 'ext': mimetype2ext(fd['MimeType']), + 'url': fd['Location'], + 'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None, + }) self._sort_formats(formats) - slide_baseurl = compat_urlparse.urljoin( - url, extract_data('SlideBaseUrl')) - slide_template = slide_baseurl + re.sub( - r'\{0:D?([0-9+])\}', r'%0\1d', extract_data('SlideImageFileNameTemplate')) - slides = [] - last_slide_time = 0 - for i in itertools.count(1): - sd = extract_str('Slides[%d]' % i, default=None) - if sd is None: - break - timestamp = int_or_none(self._search_regex( - r'^Mediasite\.PlaybackManifest\.CreateSlide\("[^"]*"\s*,\s*([0-9]+),', - sd, 'slide %s timestamp' % i, fatal=False)) - slides.append({ - 'url': slide_template % i, - 'duration': timestamp - last_slide_time, - }) - last_slide_time = timestamp - formats.append({ - 'format_id': 'slides', - 'protocol': 'slideshow', - 'url': json.dumps(slides), - 'preference': -10000, # Downloader not yet written - }) - self._sort_formats(formats) - - title = extract_data('Title') - description = extract_data('Description', fatal=False) - duration = int_or_none(extract_data( - 'Duration', fatal=False), scale=1000) - upload_date = unified_strdate(extract_data('AirDate', fatal=False)) - return { 'id': video_id, 'title': title, - 'description': description, + 'description': presentation_data.get('Description'), 'formats': formats, - 'upload_date': upload_date, - 'duration': duration, + 'timestamp': int_or_none(presentation_data.get('UnixTime'), 1000), + 'duration': int_or_none(presentation_data.get('Duration'), 1000), } diff --git a/youtube_dl/extractor/scivee.py b/youtube_dl/extractor/scivee.py index 3bf93c870..b1ca12fde 100644 --- a/youtube_dl/extractor/scivee.py +++ b/youtube_dl/extractor/scivee.py @@ -18,6 +18,7 @@ class SciVeeIE(InfoExtractor): 'title': 'Adam Arkin at the 2014 DOE JGI Genomics of Energy & Environment Meeting', 'description': 'md5:81f1710638e11a481358fab1b11059d7', }, + 'skip': 'Not accessible from Travis CI server', } def _real_extract(self, url): diff --git a/youtube_dl/extractor/screenwavemedia.py b/youtube_dl/extractor/screenwavemedia.py index 44b0bbee6..40333c825 100644 --- a/youtube_dl/extractor/screenwavemedia.py +++ b/youtube_dl/extractor/screenwavemedia.py @@ -12,7 +12,7 @@ from ..utils import ( class ScreenwaveMediaIE(InfoExtractor): - _VALID_URL = r'https?://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P<id>[A-Za-z0-9-]+)' + _VALID_URL = r'(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P<id>[A-Za-z0-9-]+)' EMBED_PATTERN = r'src=(["\'])(?P<url>(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=.+?)\1' _TESTS = [{ 'url': 'http://player.screenwavemedia.com/play/play.php?playerdiv=videoarea&companiondiv=squareAd&id=Cinemassacre-19911', diff --git a/youtube_dl/extractor/seeker.py b/youtube_dl/extractor/seeker.py new file mode 100644 index 000000000..3b9c65e7e --- /dev/null +++ b/youtube_dl/extractor/seeker.py @@ -0,0 +1,57 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class SeekerIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?seeker\.com/(?P<display_id>.*)-(?P<article_id>\d+)\.html' + _TESTS = [{ + # player.loadRevision3Item + 'url': 'http://www.seeker.com/should-trump-be-required-to-release-his-tax-returns-1833805621.html', + 'md5': '30c1dc4030cc715cf05b423d0947ac18', + 'info_dict': { + 'id': '76243', + 'ext': 'webm', + 'title': 'Should Trump Be Required To Release His Tax Returns?', + 'description': 'Donald Trump has been secretive about his "big," "beautiful" tax returns. So what can we learn if he decides to release them?', + 'uploader': 'Seeker Daily', + 'uploader_id': 'seekerdaily', + } + }, { + 'url': 'http://www.seeker.com/changes-expected-at-zoos-following-recent-gorilla-lion-shootings-1834116536.html', + 'playlist': [ + { + 'md5': '83bcd157cab89ad7318dd7b8c9cf1306', + 'info_dict': { + 'id': '67558', + 'ext': 'mp4', + 'title': 'The Pros & Cons Of Zoos', + 'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?', + 'uploader': 'DNews', + 'uploader_id': 'dnews', + }, + } + ], + 'info_dict': { + 'id': '1834116536', + 'title': 'After Gorilla Killing, Changes Ahead for Zoos', + 'description': 'The largest association of zoos and others are hoping to learn from recent incidents that led to the shooting deaths of a gorilla and two lions.', + }, + }] + + def _real_extract(self, url): + display_id, article_id = re.match(self._VALID_URL, url).groups() + webpage = self._download_webpage(url, display_id) + mobj = re.search(r"player\.loadRevision3Item\('([^']+)'\s*,\s*(\d+)\);", webpage) + if mobj: + playlist_type, playlist_id = mobj.groups() + return self.url_result( + 'revision3:%s:%s' % (playlist_type, playlist_id), 'Revision3Embed', playlist_id) + else: + entries = [self.url_result('revision3:video_id:%s' % video_id, 'Revision3Embed', video_id) for video_id in re.findall( + r'<iframe[^>]+src=[\'"](?:https?:)?//api\.seekernetwork\.com/player/embed\?videoId=(\d+)', webpage)] + return self.playlist_result( + entries, article_id, self._og_search_title(webpage), self._og_search_description(webpage)) diff --git a/youtube_dl/extractor/sendtonews.py b/youtube_dl/extractor/sendtonews.py new file mode 100644 index 000000000..1c636f672 --- /dev/null +++ b/youtube_dl/extractor/sendtonews.py @@ -0,0 +1,86 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .jwplatform import JWPlatformBaseIE +from ..compat import compat_parse_qs +from ..utils import ( + ExtractorError, + parse_duration, +) + + +class SendtoNewsIE(JWPlatformBaseIE): + _VALID_URL = r'https?://embed\.sendtonews\.com/player/embed\.php\?(?P<query>[^#]+)' + + _TEST = { + # From http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/ + 'url': 'http://embed.sendtonews.com/player/embed.php?SK=GxfCe0Zo7D&MK=175909&PK=5588&autoplay=on&sound=yes', + 'info_dict': { + 'id': 'GxfCe0Zo7D-175909-5588', + 'ext': 'mp4', + 'title': 'Recap: CLE 15, CIN 6', + 'description': '5/16/16: Indians\' bats explode for 15 runs in a win', + 'duration': 49, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + _URL_TEMPLATE = '//embed.sendtonews.com/player/embed.php?SK=%s&MK=%s&PK=%s' + + @classmethod + def _extract_url(cls, webpage): + mobj = re.search(r'''(?x)<script[^>]+src=([\'"]) + (?:https?:)?//embed\.sendtonews\.com/player/responsiveembed\.php\? + .*\bSC=(?P<SC>[0-9a-zA-Z-]+).* + \1>''', webpage) + if mobj: + sk, mk, pk = mobj.group('SC').split('-') + return cls._URL_TEMPLATE % (sk, mk, pk) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + params = compat_parse_qs(mobj.group('query')) + + if 'SK' not in params or 'MK' not in params or 'PK' not in params: + raise ExtractorError('Invalid URL', expected=True) + + video_id = '-'.join([params['SK'][0], params['MK'][0], params['PK'][0]]) + + webpage = self._download_webpage(url, video_id) + + jwplayer_data_str = self._search_regex( + r'jwplayer\("[^"]+"\)\.setup\((.+?)\);', webpage, 'JWPlayer data') + js_vars = { + 'w': 1024, + 'h': 768, + 'modeVar': 'html5', + } + for name, val in js_vars.items(): + js_val = '%d' % val if isinstance(val, int) else '"%s"' % val + jwplayer_data_str = jwplayer_data_str.replace(':%s,' % name, ':%s,' % js_val) + + info_dict = self._parse_jwplayer_data( + self._parse_json(jwplayer_data_str, video_id), + video_id, require_title=False, rtmp_params={'no_resume': True}) + + title = self._html_search_regex( + r'<div[^>]+class="embedTitle">([^<]+)</div>', webpage, 'title') + description = self._html_search_regex( + r'<div[^>]+class="embedSubTitle">([^<]+)</div>', webpage, + 'description', fatal=False) + duration = parse_duration(self._html_search_regex( + r'<div[^>]+class="embedDetails">([0-9:]+)', webpage, + 'duration', fatal=False)) + + info_dict.update({ + 'title': title, + 'description': description, + 'duration': duration, + }) + + return info_dict diff --git a/youtube_dl/extractor/sina.py b/youtube_dl/extractor/sina.py index d03f1b1d4..8fc66732a 100644 --- a/youtube_dl/extractor/sina.py +++ b/youtube_dl/extractor/sina.py @@ -4,28 +4,35 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_parse_urlencode -from ..utils import sanitized_Request +from ..utils import ( + HEADRequest, + ExtractorError, + int_or_none, + update_url_query, + qualities, + get_element_by_attribute, + clean_html, +) class SinaIE(InfoExtractor): - _VALID_URL = r'''(?x)https?://(.*?\.)?video\.sina\.com\.cn/ - ( - (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-)))) - | + _VALID_URL = r'''(?x)https?://(?:.*?\.)?video\.sina\.com\.cn/ + (?: + (?:view/|.*\#)(?P<video_id>\d+)| + .+?/(?P<pseudo_id>[^/?#]+)(?:\.s?html)| # This is used by external sites like Weibo - (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf) + api/sinawebApi/outplay.php/(?P<token>.+?)\.swf ) ''' _TESTS = [ { - 'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898', - 'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f', + 'url': 'http://video.sina.com.cn/news/spj/topvideoes20160504/?opsubject_id=top1#250576622', + 'md5': 'd38433e2fc886007729735650ae4b3e9', 'info_dict': { - 'id': '110028898', - 'ext': 'flv', - 'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员', + 'id': '250576622', + 'ext': 'mp4', + 'title': '现场:克鲁兹宣布退选 特朗普将稳获提名', } }, { @@ -35,37 +42,74 @@ class SinaIE(InfoExtractor): 'ext': 'flv', 'title': '军方提高对朝情报监视级别', }, + 'skip': 'the page does not exist or has been deleted', + }, + { + 'url': 'http://video.sina.com.cn/view/250587748.html', + 'md5': '3d1807a25c775092aab3bc157fff49b4', + 'info_dict': { + 'id': '250587748', + 'ext': 'mp4', + 'title': '瞬间泪目：8年前汶川地震珍贵视频首曝光', + }, }, ] - def _extract_video(self, video_id): - data = compat_urllib_parse_urlencode({'vid': video_id}) - url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data, - video_id, 'Downloading video url') - image_page = self._download_webpage( - 'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data, - video_id, 'Downloading thumbnail info') - - return {'id': video_id, - 'url': url_doc.find('./durl/url').text, - 'ext': 'flv', - 'title': url_doc.find('./vname').text, - 'thumbnail': image_page.split('=')[1], - } - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - if mobj.group('token') is not None: - # The video id is in the redirected url - self.to_screen('Getting video id') - request = sanitized_Request(url) - request.get_method = lambda: 'HEAD' - (_, urlh) = self._download_webpage_handle(request, 'NA', False) - return self._real_extract(urlh.geturl()) - elif video_id is None: - pseudo_id = mobj.group('pseudo_id') - webpage = self._download_webpage(url, pseudo_id) - video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id') - return self._extract_video(video_id) + video_id = mobj.group('video_id') + if not video_id: + if mobj.group('token') is not None: + # The video id is in the redirected url + self.to_screen('Getting video id') + request = HEADRequest(url) + (_, urlh) = self._download_webpage_handle(request, 'NA', False) + return self._real_extract(urlh.geturl()) + else: + pseudo_id = mobj.group('pseudo_id') + webpage = self._download_webpage(url, pseudo_id) + error = get_element_by_attribute('class', 'errtitle', webpage) + if error: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, clean_html(error)), expected=True) + video_id = self._search_regex( + r"video_id\s*:\s*'(\d+)'", webpage, 'video id') + + video_data = self._download_json( + 'http://s.video.sina.com.cn/video/h5play', + video_id, query={'video_id': video_id}) + if video_data['code'] != 1: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, video_data['message']), expected=True) + else: + video_data = video_data['data'] + title = video_data['title'] + description = video_data.get('description') + if description: + description = description.strip() + + preference = qualities(['cif', 'sd', 'hd', 'fhd', 'ffd']) + formats = [] + for quality_id, quality in video_data.get('videos', {}).get('mp4', {}).items(): + file_api = quality.get('file_api') + file_id = quality.get('file_id') + if not file_api or not file_id: + continue + formats.append({ + 'format_id': quality_id, + 'url': update_url_query(file_api, {'vid': file_id}), + 'preference': preference(quality_id), + 'ext': 'mp4', + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': video_data.get('image'), + 'duration': int_or_none(video_data.get('length')), + 'timestamp': int_or_none(video_data.get('create_time')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/sixplay.py b/youtube_dl/extractor/sixplay.py new file mode 100644 index 000000000..d3aba58a2 --- /dev/null +++ b/youtube_dl/extractor/sixplay.py @@ -0,0 +1,64 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + qualities, + int_or_none, + mimetype2ext, + determine_ext, +) + + +class SixPlayIE(InfoExtractor): + _VALID_URL = r'(?:6play:|https?://(?:www\.)?6play\.fr/.+?-c_)(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.6play.fr/jamel-et-ses-amis-au-marrakech-du-rire-p_1316/jamel-et-ses-amis-au-marrakech-du-rire-2015-c_11495320', + 'md5': '42310bffe4ba3982db112b9cd3467328', + 'info_dict': { + 'id': '11495320', + 'ext': 'mp4', + 'title': 'Jamel et ses amis au Marrakech du rire 2015', + 'description': 'md5:ba2149d5c321d5201b78070ee839d872', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + clip_data = self._download_json( + 'https://player.m6web.fr/v2/video/config/6play-auth/FR/%s.json' % video_id, + video_id) + video_data = clip_data['videoInfo'] + + quality_key = qualities(['lq', 'sd', 'hq', 'hd']) + formats = [] + for source in clip_data['sources']: + source_type, source_url = source.get('type'), source.get('src') + if not source_url or source_type == 'hls/primetime': + continue + ext = mimetype2ext(source_type) or determine_ext(source_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + source_url, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + source_url.replace('.m3u8', '.f4m'), + video_id, f4m_id='hds', fatal=False)) + elif ext == 'mp4': + quality = source.get('quality') + formats.append({ + 'url': source_url, + 'format_id': quality, + 'quality': quality_key(quality), + 'ext': ext, + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video_data['title'].strip(), + 'description': video_data.get('description'), + 'duration': int_or_none(video_data.get('duration')), + 'series': video_data.get('titlePgm'), + 'formats': formats, + } diff --git a/youtube_dl/extractor/skynewsarabia.py b/youtube_dl/extractor/skynewsarabia.py index 05e1b02ad..fffc9aa22 100644 --- a/youtube_dl/extractor/skynewsarabia.py +++ b/youtube_dl/extractor/skynewsarabia.py @@ -67,7 +67,7 @@ class SkyNewsArabiaIE(SkyNewsArabiaBaseIE): class SkyNewsArabiaArticleIE(SkyNewsArabiaBaseIE): - IE_NAME = 'skynewsarabia:video' + IE_NAME = 'skynewsarabia:article' _VALID_URL = r'https?://(?:www\.)?skynewsarabia\.com/web/article/(?P<id>[0-9]+)' _TESTS = [{ 'url': 'http://www.skynewsarabia.com/web/article/794549/%D8%A7%D9%94%D8%AD%D8%AF%D8%A7%D8%AB-%D8%A7%D9%84%D8%B4%D8%B1%D9%82-%D8%A7%D9%84%D8%A7%D9%94%D9%88%D8%B3%D8%B7-%D8%AE%D8%B1%D9%8A%D8%B7%D8%A9-%D8%A7%D9%84%D8%A7%D9%94%D9%84%D8%B9%D8%A7%D8%A8-%D8%A7%D9%84%D8%B0%D9%83%D9%8A%D8%A9', diff --git a/youtube_dl/extractor/skysports.py b/youtube_dl/extractor/skysports.py new file mode 100644 index 000000000..9dc78c7d2 --- /dev/null +++ b/youtube_dl/extractor/skysports.py @@ -0,0 +1,33 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class SkySportsIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?skysports\.com/watch/video/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.skysports.com/watch/video/10328419/bale-its-our-time-to-shine', + 'md5': 'c44a1db29f27daf9a0003e010af82100', + 'info_dict': { + 'id': '10328419', + 'ext': 'flv', + 'title': 'Bale: Its our time to shine', + 'description': 'md5:9fd1de3614d525f5addda32ac3c482c9', + }, + 'add_ie': ['Ooyala'], + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + return { + '_type': 'url_transparent', + 'id': video_id, + 'url': 'ooyala:%s' % self._search_regex( + r'data-video-id="([^"]+)"', webpage, 'ooyala id'), + 'title': self._og_search_title(webpage), + 'description': self._og_search_description(webpage), + 'ie_key': 'Ooyala', + } diff --git a/youtube_dl/extractor/slideshare.py b/youtube_dl/extractor/slideshare.py index 0b717a1e4..4967c1b77 100644 --- a/youtube_dl/extractor/slideshare.py +++ b/youtube_dl/extractor/slideshare.py @@ -9,6 +9,7 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + get_element_by_id, ) @@ -40,7 +41,7 @@ class SlideshareIE(InfoExtractor): bucket = info['jsplayer']['video_bucket'] ext = info['jsplayer']['video_extension'] video_url = compat_urlparse.urljoin(bucket, doc + '-SD.' + ext) - description = self._html_search_regex( + description = get_element_by_id('slideshow-description-paragraph', webpage) or self._html_search_regex( r'(?s)<p[^>]+itemprop="description"[^>]*>(.+?)</p>', webpage, 'description', fatal=False) @@ -51,5 +52,5 @@ class SlideshareIE(InfoExtractor): 'ext': ext, 'url': video_url, 'thumbnail': info['slideshow']['pin_image_url'], - 'description': description, + 'description': description.strip() if description else None, } diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py index 49e5d09ae..72fe66142 100644 --- a/youtube_dl/extractor/sohu.py +++ b/youtube_dl/extractor/sohu.py @@ -8,10 +8,7 @@ from ..compat import ( compat_str, compat_urllib_parse_urlencode, ) -from ..utils import ( - ExtractorError, - sanitized_Request, -) +from ..utils import ExtractorError class SohuIE(InfoExtractor): @@ -96,15 +93,10 @@ class SohuIE(InfoExtractor): else: base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' - req = sanitized_Request(base_data_url + vid_id) - - cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') - if cn_verification_proxy: - req.add_header('Ytdl-request-proxy', cn_verification_proxy) - return self._download_json( - req, video_id, - 'Downloading JSON data for %s' % vid_id) + base_data_url + vid_id, video_id, + 'Downloading JSON data for %s' % vid_id, + headers=self.geo_verification_headers()) mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py index 692fd78e8..92a7120a3 100644 --- a/youtube_dl/extractor/spankwire.py +++ b/youtube_dl/extractor/spankwire.py @@ -96,20 +96,18 @@ class SpankwireIE(InfoExtractor): formats = [] for height, video_url in zip(heights, video_urls): path = compat_urllib_parse_urlparse(video_url).path - _, quality = path.split('/')[4].split('_')[:2] - f = { + m = re.search(r'/(?P<height>\d+)[pP]_(?P<tbr>\d+)[kK]', path) + if m: + tbr = int(m.group('tbr')) + height = int(m.group('height')) + else: + tbr = None + formats.append({ 'url': video_url, + 'format_id': '%dp' % height, 'height': height, - } - tbr = self._search_regex(r'^(\d+)[Kk]$', quality, 'tbr', default=None) - if tbr: - f.update({ - 'tbr': int(tbr), - 'format_id': '%dp' % height, - }) - else: - f['format_id'] = quality - formats.append(f) + 'tbr': tbr, + }) self._sort_formats(formats) age_limit = self._rta_search(webpage) diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 39a7aaf9d..3c552807e 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -4,8 +4,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urlparse from .spiegeltv import SpiegeltvIE +from ..compat import compat_urlparse +from ..utils import ( + extract_attributes, + unified_strdate, + get_element_by_attribute, +) class SpiegelIE(InfoExtractor): @@ -19,6 +24,7 @@ class SpiegelIE(InfoExtractor): 'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv', 'description': 'md5:8029d8310232196eb235d27575a8b9f4', 'duration': 49, + 'upload_date': '20130311', }, }, { 'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html', @@ -29,6 +35,7 @@ class SpiegelIE(InfoExtractor): 'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers', 'description': 'md5:c2322b65e58f385a820c10fa03b2d088', 'duration': 983, + 'upload_date': '20131115', }, }, { 'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html', @@ -38,6 +45,7 @@ class SpiegelIE(InfoExtractor): 'ext': 'mp4', 'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.', 'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"', + 'upload_date': '20140904', } }, { 'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html', @@ -52,10 +60,10 @@ class SpiegelIE(InfoExtractor): if SpiegeltvIE.suitable(handle.geturl()): return self.url_result(handle.geturl(), 'Spiegeltv') - title = re.sub(r'\s+', ' ', self._html_search_regex( - r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>', - webpage, 'title')) - description = self._html_search_meta('description', webpage, 'description') + video_data = extract_attributes(self._search_regex(r'(<div[^>]+id="spVideoElements"[^>]+>)', webpage, 'video element', default='')) + + title = video_data.get('data-video-title') or get_element_by_attribute('class', 'module-title', webpage) + description = video_data.get('data-video-teaser') or self._html_search_meta('description', webpage, 'description') base_url = self._search_regex( [r'server\s*:\s*(["\'])(?P<url>.+?)\1', r'var\s+server\s*=\s*"(?P<url>[^"]+)\"'], @@ -87,8 +95,9 @@ class SpiegelIE(InfoExtractor): return { 'id': video_id, 'title': title, - 'description': description, + 'description': description.strip() if description else None, 'duration': duration, + 'upload_date': unified_strdate(video_data.get('data-video-date')), 'formats': formats, } @@ -104,6 +113,7 @@ class SpiegelArticleIE(InfoExtractor): 'ext': 'mp4', 'title': 'Faszination Badminton: Nennt es bloß nicht Federball', 'description': 're:^Patrick Kämnitz gehört.{100,}', + 'upload_date': '20140825', }, }, { 'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html', diff --git a/youtube_dl/extractor/sportschau.py b/youtube_dl/extractor/sportschau.py new file mode 100644 index 000000000..0d7925a08 --- /dev/null +++ b/youtube_dl/extractor/sportschau.py @@ -0,0 +1,38 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .wdr import WDRBaseIE +from ..utils import get_element_by_attribute + + +class SportschauIE(WDRBaseIE): + IE_NAME = 'Sportschau' + _VALID_URL = r'https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video-?(?P<id>[^/#?]+)\.html' + _TEST = { + 'url': 'http://www.sportschau.de/uefaeuro2016/videos/video-dfb-team-geht-gut-gelaunt-ins-spiel-gegen-polen-100.html', + 'info_dict': { + 'id': 'mdb-1140188', + 'display_id': 'dfb-team-geht-gut-gelaunt-ins-spiel-gegen-polen-100', + 'ext': 'mp4', + 'title': 'DFB-Team geht gut gelaunt ins Spiel gegen Polen', + 'description': 'Vor dem zweiten Gruppenspiel gegen Polen herrscht gute Stimmung im deutschen Team. Insbesondere Bastian Schweinsteiger strotzt vor Optimismus nach seinem Tor gegen die Ukraine.', + 'upload_date': '20160615', + }, + 'skip': 'Geo-restricted to Germany', + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + title = get_element_by_attribute('class', 'headline', webpage) + description = self._html_search_meta('description', webpage, 'description') + + info = self._extract_wdr_video(webpage, video_id) + + info.update({ + 'title': title, + 'description': description, + }) + + return info diff --git a/youtube_dl/extractor/srmediathek.py b/youtube_dl/extractor/srmediathek.py index 74d01183f..409d50304 100644 --- a/youtube_dl/extractor/srmediathek.py +++ b/youtube_dl/extractor/srmediathek.py @@ -9,8 +9,9 @@ from ..utils import ( class SRMediathekIE(ARDMediathekIE): + IE_NAME = 'sr:mediathek' IE_DESC = 'Saarländischer Rundfunk' - _VALID_URL = r'https?://sr-mediathek\.sr-online\.de/index\.php\?.*?&id=(?P<id>[0-9]+)' + _VALID_URL = r'https?://sr-mediathek(?:\.sr-online)?\.de/index\.php\?.*?&id=(?P<id>[0-9]+)' _TESTS = [{ 'url': 'http://sr-mediathek.sr-online.de/index.php?seite=7&id=28455', @@ -34,7 +35,9 @@ class SRMediathekIE(ARDMediathekIE): # m3u8 download 'skip_download': True, }, - 'expected_warnings': ['Unable to download f4m manifest'] + }, { + 'url': 'http://sr-mediathek.de/index.php?seite=7&id=7480', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/stitcher.py b/youtube_dl/extractor/stitcher.py index d5c852f52..0f8782d03 100644 --- a/youtube_dl/extractor/stitcher.py +++ b/youtube_dl/extractor/stitcher.py @@ -56,7 +56,7 @@ class StitcherIE(InfoExtractor): episode = self._parse_json( js_to_json(self._search_regex( - r'(?s)var\s+stitcher\s*=\s*({.+?});\n', webpage, 'episode config')), + r'(?s)var\s+stitcher(?:Config)?\s*=\s*({.+?});\n', webpage, 'episode config')), display_id)['config']['episode'] title = unescapeHTML(episode['title']) diff --git a/youtube_dl/extractor/streamcloud.py b/youtube_dl/extractor/streamcloud.py index 712359885..6a6bb90c4 100644 --- a/youtube_dl/extractor/streamcloud.py +++ b/youtube_dl/extractor/streamcloud.py @@ -5,7 +5,7 @@ import re from .common import InfoExtractor from ..utils import ( - sanitized_Request, + ExtractorError, urlencode_postdata, ) @@ -14,7 +14,7 @@ class StreamcloudIE(InfoExtractor): IE_NAME = 'streamcloud.eu' _VALID_URL = r'https?://streamcloud\.eu/(?P<id>[a-zA-Z0-9_-]+)(?:/(?P<fname>[^#?]*)\.html)?' - _TEST = { + _TESTS = [{ 'url': 'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html', 'md5': '6bea4c7fa5daaacc2a946b7146286686', 'info_dict': { @@ -23,7 +23,10 @@ class StreamcloudIE(InfoExtractor): 'title': 'youtube-dl test video \'/\\ ä ↭', }, 'skip': 'Only available from the EU' - } + }, { + 'url': 'http://streamcloud.eu/ua8cmfh1nbe6/NSHIP-148--KUC-NG--H264-.mp4.html', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -31,26 +34,36 @@ class StreamcloudIE(InfoExtractor): orig_webpage = self._download_webpage(url, video_id) + if '>File Not Found<' in orig_webpage: + raise ExtractorError( + 'Video %s does not exist' % video_id, expected=True) + fields = re.findall(r'''(?x)<input\s+ type="(?:hidden|submit)"\s+ name="([^"]+)"\s+ (?:id="[^"]+"\s+)? value="([^"]*)" ''', orig_webpage) - post = urlencode_postdata(fields) self._sleep(12, video_id) - headers = { - b'Content-Type': b'application/x-www-form-urlencoded', - } - req = sanitized_Request(url, post, headers) webpage = self._download_webpage( - req, video_id, note='Downloading video page ...') - title = self._html_search_regex( - r'<h1[^>]*>([^<]+)<', webpage, 'title') - video_url = self._search_regex( - r'file:\s*"([^"]+)"', webpage, 'video URL') + url, video_id, data=urlencode_postdata(fields), headers={ + b'Content-Type': b'application/x-www-form-urlencoded', + }) + + try: + title = self._html_search_regex( + r'<h1[^>]*>([^<]+)<', webpage, 'title') + video_url = self._search_regex( + r'file:\s*"([^"]+)"', webpage, 'video URL') + except ExtractorError: + message = self._html_search_regex( + r'(?s)<div[^>]+class=(["\']).*?msgboxinfo.*?\1[^>]*>(?P<message>.+?)</div>', + webpage, 'message', default=None, group='message') + if message: + raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True) + raise thumbnail = self._search_regex( r'image:\s*"([^"]+)"', webpage, 'thumbnail URL', fatal=False) diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index 2ab30e45f..1c04dfb7b 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -6,17 +6,14 @@ import re from .common import InfoExtractor from ..utils import ( determine_ext, + dict_get, + int_or_none, + try_get, ) class SVTBaseIE(InfoExtractor): - def _extract_video(self, url, video_id): - info = self._download_json(url, video_id) - - title = info['context']['title'] - thumbnail = info['context'].get('thumbnailImage') - - video_info = info['video'] + def _extract_video(self, video_info, video_id): formats = [] for vr in video_info['videoReferences']: player_type = vr.get('playerType') @@ -40,27 +37,49 @@ class SVTBaseIE(InfoExtractor): 'format_id': player_type, 'url': vurl, }) + if not formats and video_info.get('rights', {}).get('geoBlockedSweden'): + self.raise_geo_restricted('This video is only available in Sweden') self._sort_formats(formats) subtitles = {} - subtitle_references = video_info.get('subtitleReferences') + subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences')) if isinstance(subtitle_references, list): for sr in subtitle_references: subtitle_url = sr.get('url') + subtitle_lang = sr.get('language', 'sv') if subtitle_url: - subtitles.setdefault('sv', []).append({'url': subtitle_url}) + if determine_ext(subtitle_url) == 'm3u8': + # TODO(yan12125): handle WebVTT in m3u8 manifests + continue + + subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url}) - duration = video_info.get('materialLength') - age_limit = 18 if video_info.get('inappropriateForChildren') else 0 + title = video_info.get('title') + + series = video_info.get('programTitle') + season_number = int_or_none(video_info.get('season')) + episode = video_info.get('episodeTitle') + episode_number = int_or_none(video_info.get('episodeNumber')) + + duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration'))) + age_limit = None + adult = dict_get( + video_info, ('inappropriateForChildren', 'blockedForChildren'), + skip_false_values=False) + if adult is not None: + age_limit = 18 if adult else 0 return { 'id': video_id, 'title': title, 'formats': formats, 'subtitles': subtitles, - 'thumbnail': thumbnail, 'duration': duration, 'age_limit': age_limit, + 'series': series, + 'season_number': season_number, + 'episode': episode, + 'episode_number': episode_number, } @@ -68,11 +87,11 @@ class SVTIE(SVTBaseIE): _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)' _TEST = { 'url': 'http://www.svt.se/wd?widgetId=23991§ionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false', - 'md5': '9648197555fc1b49e3dc22db4af51d46', + 'md5': '33e9a5d8f646523ce0868ecfb0eed77d', 'info_dict': { 'id': '2900353', - 'ext': 'flv', - 'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)', + 'ext': 'mp4', + 'title': 'Stjärnorna skojar till det - under SVT-intervjun', 'duration': 27, 'age_limit': 0, }, @@ -89,15 +108,20 @@ class SVTIE(SVTBaseIE): mobj = re.match(self._VALID_URL, url) widget_id = mobj.group('widget_id') article_id = mobj.group('id') - return self._extract_video( + + info = self._download_json( 'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id), article_id) + info_dict = self._extract_video(info['video'], article_id) + info_dict['title'] = info['context']['title'] + return info_dict + class SVTPlayIE(SVTBaseIE): IE_DESC = 'SVT Play and Öppet arkiv' - _VALID_URL = r'https?://(?:www\.)?(?P<host>svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)' + _TESTS = [{ 'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2', 'md5': '2b6704fe4a28801e1a098bbf3c5ac611', 'info_dict': { @@ -113,12 +137,50 @@ class SVTPlayIE(SVTBaseIE): }] }, }, - } + }, { + # geo restricted to Sweden + 'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten', + 'only_matching': True, + }, { + 'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - host = mobj.group('host') - return self._extract_video( - 'http://www.%s.se/video/%s?output=json' % (host, video_id), - video_id) + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + data = self._parse_json( + self._search_regex( + r'root\["__svtplay"\]\s*=\s*([^;]+);', + webpage, 'embedded data', default='{}'), + video_id, fatal=False) + + thumbnail = self._og_search_thumbnail(webpage) + + if data: + video_info = try_get( + data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'], + dict) + if video_info: + info_dict = self._extract_video(video_info, video_id) + info_dict.update({ + 'title': data['context']['dispatcher']['stores']['MetaStore']['title'], + 'thumbnail': thumbnail, + }) + return info_dict + + video_id = self._search_regex( + r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)', + webpage, 'video id', default=None) + + if video_id: + data = self._download_json( + 'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id) + info_dict = self._extract_video(data, video_id) + if not info_dict.get('title'): + info_dict['title'] = re.sub( + r'\s*\|\s*.+?$', '', + info_dict.get('episode') or self._og_search_title(webpage)) + return info_dict diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py index 73e7657d4..136e18f96 100644 --- a/youtube_dl/extractor/tagesschau.py +++ b/youtube_dl/extractor/tagesschau.py @@ -4,43 +4,179 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import parse_filesize +from ..utils import ( + determine_ext, + js_to_json, + parse_iso8601, + parse_filesize, +) + + +class TagesschauPlayerIE(InfoExtractor): + IE_NAME = 'tagesschau:player' + _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?P<kind>audio|video)/(?P=kind)-(?P<id>\d+)~player(?:_[^/?#&]+)?\.html' + + _TESTS = [{ + 'url': 'http://www.tagesschau.de/multimedia/video/video-179517~player.html', + 'md5': '8d09548d5c15debad38bee3a4d15ca21', + 'info_dict': { + 'id': '179517', + 'ext': 'mp4', + 'title': 'Marie Kristin Boese, ARD Berlin, über den zukünftigen Kurs der AfD', + 'thumbnail': 're:^https?:.*\.jpg$', + 'formats': 'mincount:6', + }, + }, { + 'url': 'https://www.tagesschau.de/multimedia/audio/audio-29417~player.html', + 'md5': '76e6eec6ebd40740671cf0a2c88617e5', + 'info_dict': { + 'id': '29417', + 'ext': 'mp3', + 'title': 'Trabi - Bye, bye Rennpappe', + 'thumbnail': 're:^https?:.*\.jpg$', + 'formats': 'mincount:2', + }, + }, { + 'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417~player_autoplay-true.html', + 'only_matching': True, + }] + + _FORMATS = { + 'xs': {'quality': 0}, + 's': {'width': 320, 'height': 180, 'quality': 1}, + 'm': {'width': 512, 'height': 288, 'quality': 2}, + 'l': {'width': 960, 'height': 540, 'quality': 3}, + 'xl': {'width': 1280, 'height': 720, 'quality': 4}, + 'xxl': {'quality': 5}, + } + + def _extract_via_api(self, kind, video_id): + info = self._download_json( + 'https://www.tagesschau.de/api/multimedia/{0}/{0}-{1}.json'.format(kind, video_id), + video_id) + title = info['headline'] + formats = [] + for media in info['mediadata']: + for format_id, format_url in media.items(): + if determine_ext(format_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls')) + else: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'vcodec': 'none' if kind == 'audio' else None, + }) + self._sort_formats(formats) + timestamp = parse_iso8601(info.get('date')) + return { + 'id': video_id, + 'title': title, + 'timestamp': timestamp, + 'formats': formats, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + # kind = mobj.group('kind').lower() + # if kind == 'video': + # return self._extract_via_api(kind, video_id) + + # JSON api does not provide some audio formats (e.g. ogg) thus + # extractiong audio via webpage + + webpage = self._download_webpage(url, video_id) + + title = self._og_search_title(webpage).strip() + formats = [] + + for media_json in re.findall(r'({src\s*:\s*["\']http[^}]+type\s*:[^}]+})', webpage): + media = self._parse_json(js_to_json(media_json), video_id, fatal=False) + if not media: + continue + src = media.get('src') + if not src: + return + quality = media.get('quality') + kind = media.get('type', '').split('/')[0] + ext = determine_ext(src) + f = { + 'url': src, + 'format_id': '%s_%s' % (quality, ext) if quality else ext, + 'ext': ext, + 'vcodec': 'none' if kind == 'audio' else None, + } + f.update(self._FORMATS.get(quality, {})) + formats.append(f) + + self._sort_formats(formats) + + thumbnail = self._og_search_thumbnail(webpage) + + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'formats': formats, + } class TagesschauIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_[^/#?]+?)?\.html' + _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/(?P<path>[^/]+/(?:[^/]+/)*?(?P<id>[^/#?]+?(?:-?[0-9]+)?))(?:~_?[^/#?]+?)?\.html' _TESTS = [{ 'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html', - 'md5': '917a228bc7df7850783bc47979673a09', + 'md5': 'f7c27a0eff3bfe8c7727e65f8fe1b1e6', 'info_dict': { - 'id': '102143', + 'id': 'video-102143', 'ext': 'mp4', 'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt', - 'description': 'md5:171feccd9d9b3dd54d05d501568f6359', + 'description': '18.07.2015 20:10 Uhr', 'thumbnail': 're:^https?:.*\.jpg$', }, }, { 'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html', 'md5': '3c54c1f6243d279b706bde660ceec633', 'info_dict': { - 'id': '5727', + 'id': 'ts-5727', 'ext': 'mp4', - 'description': 'md5:695c01bfd98b7e313c501386327aea59', 'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr', + 'description': 'md5:695c01bfd98b7e313c501386327aea59', + 'thumbnail': 're:^https?:.*\.jpg$', + }, + }, { + # exclusive audio + 'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417.html', + 'md5': '76e6eec6ebd40740671cf0a2c88617e5', + 'info_dict': { + 'id': 'audio-29417', + 'ext': 'mp3', + 'title': 'Trabi - Bye, bye Rennpappe', + 'description': 'md5:8687dda862cbbe2cfb2df09b56341317', 'thumbnail': 're:^https?:.*\.jpg$', }, }, { - 'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html', - 'md5': 'aef45de271c4bf0a5db834aa40bf774c', + # audio in article + 'url': 'http://www.tagesschau.de/inland/bnd-303.html', + 'md5': 'e0916c623e85fc1d2b26b78f299d3958', 'info_dict': { - 'id': '18407', + 'id': 'bnd-303', 'ext': 'mp3', - 'title': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', - 'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', + 'title': 'Viele Baustellen für neuen BND-Chef', + 'description': 'md5:1e69a54be3e1255b2b07cdbce5bcd8b4', 'thumbnail': 're:^https?:.*\.jpg$', }, }, { + 'url': 'http://www.tagesschau.de/inland/afd-parteitag-135.html', + 'info_dict': { + 'id': 'afd-parteitag-135', + 'title': 'Möchtegern-Underdog mit Machtanspruch', + }, + 'playlist_count': 2, + }, { 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html', 'only_matching': True, }, { @@ -61,88 +197,108 @@ class TagesschauIE(InfoExtractor): }, { 'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html', 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/100sekunden/index.html', + 'only_matching': True, + }, { + # playlist article with collapsing sections + 'url': 'http://www.tagesschau.de/wirtschaft/faq-freihandelszone-eu-usa-101.html', + 'only_matching': True, }] - _FORMATS = { - 's': {'width': 256, 'height': 144, 'quality': 1}, - 'm': {'width': 512, 'height': 288, 'quality': 2}, - 'l': {'width': 960, 'height': 544, 'quality': 3}, - } + @classmethod + def suitable(cls, url): + return False if TagesschauPlayerIE.suitable(url) else super(TagesschauIE, cls).suitable(url) + + def _extract_formats(self, download_text, media_kind): + links = re.finditer( + r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>', + download_text) + formats = [] + for l in links: + link_url = l.group('url') + if not link_url: + continue + format_id = self._search_regex( + r'.*/[^/.]+\.([^/]+)\.[^/.]+$', link_url, 'format ID', + default=determine_ext(link_url)) + format = { + 'format_id': format_id, + 'url': l.group('url'), + 'format_name': l.group('name'), + } + title = l.group('title') + if title: + if media_kind.lower() == 'video': + m = re.match( + r'''(?x) + Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10; + (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10; + (?P<vbr>[0-9]+)kbps&\#10; + Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10; + Größe:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''', + title) + if m: + format.update({ + 'format_note': m.group('audio_desc'), + 'vcodec': m.group('vcodec'), + 'width': int(m.group('width')), + 'height': int(m.group('height')), + 'abr': int(m.group('abr')), + 'vbr': int(m.group('vbr')), + 'filesize_approx': parse_filesize(m.group('filesize_approx')), + }) + else: + m = re.match( + r'(?P<format>.+?)-Format\s*:\s*(?P<abr>\d+)kbps\s*,\s*(?P<note>.+)', + title) + if m: + format.update({ + 'format_note': '%s, %s' % (m.group('format'), m.group('note')), + 'vcodec': 'none', + 'abr': int(m.group('abr')), + }) + formats.append(format) + self._sort_formats(formats) + return formats def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') or mobj.group('path') display_id = video_id.lstrip('-') + webpage = self._download_webpage(url, display_id) - player_url = self._html_search_meta( - 'twitter:player', webpage, 'player URL', default=None) - if player_url: - playerpage = self._download_webpage( - player_url, display_id, 'Downloading player page') - - formats = [] - for media in re.finditer( - r'''(?x) - (?P<q_url>["\'])(?P<url>http://media.+?)(?P=q_url) - ,\s*type:(?P<q_type>["\'])(?P<type>video|audio)/(?P<ext>.+?)(?P=q_type) - (?:,\s*quality:(?P<q_quality>["\'])(?P<quality>.+?)(?P=q_quality))? - ''', playerpage): - url = media.group('url') - type_ = media.group('type') - ext = media.group('ext') - res = media.group('quality') - f = { - 'format_id': '%s_%s' % (res, ext) if res else ext, - 'url': url, - 'ext': ext, - 'vcodec': 'none' if type_ == 'audio' else None, - } - f.update(self._FORMATS.get(res, {})) - formats.append(f) - thumbnail = self._og_search_thumbnail(playerpage) - title = self._og_search_title(webpage).strip() - description = self._og_search_description(webpage).strip() - else: + title = self._html_search_regex( + r'<span[^>]*class="headline"[^>]*>(.+?)</span>', + webpage, 'title', default=None) or self._og_search_title(webpage) + + DOWNLOAD_REGEX = r'(?s)<p>Wir bieten dieses (?P<kind>Video|Audio) in folgenden Formaten zum Download an:</p>\s*<div class="controls">(?P<links>.*?)</div>\s*<p>' + + webpage_type = self._og_search_property('type', webpage, default=None) + if webpage_type == 'website': # Article + entries = [] + for num, (entry_title, media_kind, download_text) in enumerate(re.findall( + r'(?s)<p[^>]+class="infotext"[^>]*>\s*(?:<a[^>]+>)?\s*<strong>(.+?)</strong>.*?</p>.*?%s' % DOWNLOAD_REGEX, + webpage), 1): + entries.append({ + 'id': '%s-%d' % (display_id, num), + 'title': '%s' % entry_title, + 'formats': self._extract_formats(download_text, media_kind), + }) + if len(entries) > 1: + return self.playlist_result(entries, display_id, title) + formats = entries[0]['formats'] + else: # Assume single video download_text = self._search_regex( - r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>', - webpage, 'download links') - links = re.finditer( - r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>', - download_text) - formats = [] - for l in links: - format_id = self._search_regex( - r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID') - format = { - 'format_id': format_id, - 'url': l.group('url'), - 'format_name': l.group('name'), - } - m = re.match( - r'''(?x) - Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10; - (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10; - (?P<vbr>[0-9]+)kbps&\#10; - Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10; - Größe:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''', - l.group('title')) - if m: - format.update({ - 'format_note': m.group('audio_desc'), - 'vcodec': m.group('vcodec'), - 'width': int(m.group('width')), - 'height': int(m.group('height')), - 'abr': int(m.group('abr')), - 'vbr': int(m.group('vbr')), - 'filesize_approx': parse_filesize(m.group('filesize_approx')), - }) - formats.append(format) - thumbnail = self._og_search_thumbnail(webpage) - description = self._html_search_regex( - r'(?s)<p class="teasertext">(.*?)</p>', - webpage, 'description', default=None) - title = self._html_search_regex( - r'<span class="headline".*?>(.*?)</span>', webpage, 'title') + DOWNLOAD_REGEX, webpage, 'download links', group='links') + media_kind = self._search_regex( + DOWNLOAD_REGEX, webpage, 'media kind', default='Video', group='kind') + formats = self._extract_formats(download_text, media_kind) + thumbnail = self._og_search_thumbnail(webpage) + description = self._html_search_regex( + r'(?s)<p class="teasertext">(.*?)</p>', + webpage, 'description', default=None) self._sort_formats(formats) diff --git a/youtube_dl/extractor/teachingchannel.py b/youtube_dl/extractor/teachingchannel.py index e0477382c..d14d93e3a 100644 --- a/youtube_dl/extractor/teachingchannel.py +++ b/youtube_dl/extractor/teachingchannel.py @@ -11,6 +11,7 @@ class TeachingChannelIE(InfoExtractor): _TEST = { 'url': 'https://www.teachingchannel.org/videos/teacher-teaming-evolution', + 'md5': '3d6361864d7cac20b57c8784da17166f', 'info_dict': { 'id': 'F3bnlzbToeI6pLEfRyrlfooIILUjz4nM', 'ext': 'mp4', @@ -19,9 +20,9 @@ class TeachingChannelIE(InfoExtractor): 'duration': 422.255, }, 'params': { - # m3u8 download 'skip_download': True, }, + 'add_ie': ['Ooyala'], } def _real_extract(self, url): diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py index b49ab5f5b..79a778920 100644 --- a/youtube_dl/extractor/teamcoco.py +++ b/youtube_dl/extractor/teamcoco.py @@ -88,7 +88,7 @@ class TeamcocoIE(InfoExtractor): preload_codes = self._html_search_regex( r'(function.+)setTimeout\(function\(\)\{playlist', webpage, 'preload codes') - base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes) + base64_fragments = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes) base64_fragments.remove('init') def _check_sequence(cur_fragments): diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index cf8851438..451cde76d 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -27,7 +27,7 @@ class TEDIE(InfoExtractor): ''' _TESTS = [{ 'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html', - 'md5': 'fc94ac279feebbce69f21c0c6ee82810', + 'md5': '0de43ac406aa3e4ea74b66c9c7789b13', 'info_dict': { 'id': '102', 'ext': 'mp4', @@ -37,21 +37,26 @@ class TEDIE(InfoExtractor): 'consciousness, but that half the time our brains are ' 'actively fooling us.'), 'uploader': 'Dan Dennett', - 'width': 854, + 'width': 853, 'duration': 1308, } }, { 'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms', - 'md5': '226f4fb9c62380d11b7995efa4c87994', + 'md5': 'b899ac15e345fb39534d913f7606082b', 'info_dict': { - 'id': 'vishal-sikka-the-beauty-and-power-of-algorithms', + 'id': 'tSVI8ta_P4w', 'ext': 'mp4', 'title': 'Vishal Sikka: The beauty and power of algorithms', 'thumbnail': 're:^https?://.+\.jpg', - 'description': 'Adaptive, intelligent, and consistent, algorithms are emerging as the ultimate app for everything from matching consumers to products to assessing medical diagnoses. Vishal Sikka shares his appreciation for the algorithm, charting both its inherent beauty and its growing power.', - } + 'description': 'md5:6261fdfe3e02f4f579cbbfc00aff73f4', + 'upload_date': '20140122', + 'uploader_id': 'TEDInstitute', + 'uploader': 'TED Institute', + }, + 'add_ie': ['Youtube'], }, { 'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best', + 'md5': '71b3ab2f4233012dce09d515c9c39ce2', 'info_dict': { 'id': '1972', 'ext': 'mp4', @@ -102,9 +107,9 @@ class TEDIE(InfoExtractor): }] _NATIVE_FORMATS = { - 'low': {'preference': 1, 'width': 320, 'height': 180}, - 'medium': {'preference': 2, 'width': 512, 'height': 288}, - 'high': {'preference': 3, 'width': 854, 'height': 480}, + 'low': {'width': 320, 'height': 180}, + 'medium': {'width': 512, 'height': 288}, + 'high': {'width': 854, 'height': 480}, } def _extract_info(self, webpage): @@ -171,15 +176,21 @@ class TEDIE(InfoExtractor): if finfo: f.update(finfo) + http_url = None for format_id, resources in talk_info['resources'].items(): if format_id == 'h264': for resource in resources: + h264_url = resource.get('file') + if not h264_url: + continue bitrate = int_or_none(resource.get('bitrate')) formats.append({ - 'url': resource['file'], + 'url': h264_url, 'format_id': '%s-%sk' % (format_id, bitrate), 'tbr': bitrate, }) + if re.search('\d+k', h264_url): + http_url = h264_url elif format_id == 'rtmp': streamer = talk_info.get('streamer') if not streamer: @@ -195,16 +206,24 @@ class TEDIE(InfoExtractor): 'tbr': int_or_none(resource.get('bitrate')), }) elif format_id == 'hls': - hls_formats = self._extract_m3u8_formats( - resources.get('stream'), video_name, 'mp4', m3u8_id=format_id) - for f in hls_formats: - if f.get('format_id') == 'hls-meta': - continue - if not f.get('height'): - f['vcodec'] = 'none' - else: - f['acodec'] = 'none' - formats.extend(hls_formats) + formats.extend(self._extract_m3u8_formats( + resources.get('stream'), video_name, 'mp4', m3u8_id=format_id, fatal=False)) + + m3u8_formats = list(filter( + lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + formats)) + if http_url: + for m3u8_format in m3u8_formats: + bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None) + if not bitrate: + continue + f = m3u8_format.copy() + f.update({ + 'url': re.sub(r'\d+k', bitrate, http_url), + 'format_id': m3u8_format['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) audio_download = talk_info.get('audioDownload') if audio_download: @@ -212,7 +231,6 @@ class TEDIE(InfoExtractor): 'url': audio_download, 'format_id': 'audio', 'vcodec': 'none', - 'preference': -0.5, }) self._sort_formats(formats) @@ -254,7 +272,11 @@ class TEDIE(InfoExtractor): config_json = self._html_search_regex( r'"pages\.jwplayer"\s*,\s*({.+?})\s*\)\s*</script>', - webpage, 'config') + webpage, 'config', default=None) + if not config_json: + embed_url = self._search_regex( + r"<iframe[^>]+class='pages-video-embed__video__object'[^>]+src='([^']+)'", webpage, 'embed url') + return self.url_result(self._proto_relative_url(embed_url)) config = json.loads(config_json)['config'] video_url = config['video']['url'] thumbnail = config.get('image', {}).get('url') diff --git a/youtube_dl/extractor/telecinco.py b/youtube_dl/extractor/telecinco.py index 4b4b740b4..2ecfd0405 100644 --- a/youtube_dl/extractor/telecinco.py +++ b/youtube_dl/extractor/telecinco.py @@ -1,50 +1,41 @@ # coding: utf-8 from __future__ import unicode_literals -import json +from .mitele import MiTeleBaseIE -from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_unquote, - compat_urllib_parse_urlencode, - compat_urlparse, -) -from ..utils import ( - get_element_by_attribute, - parse_duration, - strip_jsonp, -) - -class TelecincoIE(InfoExtractor): +class TelecincoIE(MiTeleBaseIE): IE_DESC = 'telecinco.es, cuatro.com and mediaset.es' _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html' _TESTS = [{ 'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html', - 'md5': '5cbef3ad5ef17bf0d21570332d140729', + 'md5': '8d7b2d5f699ee2709d992a63d5cd1712', 'info_dict': { - 'id': 'MDSVID20141015_0058', + 'id': 'JEA5ijCnF6p5W08A1rNKn7', 'ext': 'mp4', - 'title': 'Con Martín Berasategui, hacer un bacalao al ...', + 'title': 'Bacalao con kokotxas al pil-pil', + 'description': 'md5:1382dacd32dd4592d478cbdca458e5bb', 'duration': 662, }, }, { 'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html', - 'md5': '0a5b9f3cc8b074f50a0578f823a12694', + 'md5': '284393e5387b3b947b77c613ef04749a', 'info_dict': { - 'id': 'MDSVID20150916_0128', + 'id': 'jn24Od1zGLG4XUZcnUnZB6', 'ext': 'mp4', - 'title': '¿Quién es este ex futbolista con el que hablan ...', + 'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?', + 'description': 'md5:a62ecb5f1934fc787107d7b9a2262805', 'duration': 79, }, }, { 'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html', - 'md5': 'ad1bfaaba922dd4a295724b05b68f86a', + 'md5': '749afab6ea5a136a8806855166ae46a2', 'info_dict': { - 'id': 'MDSVID20150513_0220', + 'id': 'aywerkD2Sv1vGNqq9b85Q2', 'ext': 'mp4', 'title': '#DOYLACARA. Con la trata no hay trato', + 'description': 'md5:2771356ff7bfad9179c5f5cd954f1477', 'duration': 50, }, }, { @@ -56,40 +47,16 @@ class TelecincoIE(InfoExtractor): }] def _real_extract(self, url): - episode = self._match_id(url) - webpage = self._download_webpage(url, episode) - embed_data_json = self._search_regex( - r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data', - ).replace('\'', '"') - embed_data = json.loads(embed_data_json) - - domain = embed_data['mediaUrl'] - if not domain.startswith('http'): - # only happens in telecinco.es videos - domain = 'http://' + domain - info_url = compat_urlparse.urljoin( - domain, - compat_urllib_parse_unquote(embed_data['flashvars']['host']) - ) - info_el = self._download_xml(info_url, episode).find('./video/info') - - video_link = info_el.find('videoUrl/link').text - token_query = compat_urllib_parse_urlencode({'id': video_link}) - token_info = self._download_json( - embed_data['flashvars']['ov_tk'] + '?' + token_query, - episode, - transform_source=strip_jsonp - ) - formats = self._extract_m3u8_formats( - token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native') - self._sort_formats(formats) - - return { - 'id': embed_data['videoId'], - 'display_id': episode, - 'title': info_el.find('title').text, - 'formats': formats, - 'description': get_element_by_attribute('class', 'text', webpage), - 'thumbnail': info_el.find('thumb').text, - 'duration': parse_duration(info_el.find('duration').text), - } + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + title = self._html_search_meta( + ['og:title', 'twitter:title'], webpage, 'title') + info = self._get_player_info(url, webpage) + info.update({ + 'display_id': display_id, + 'title': title, + 'description': self._html_search_meta( + ['og:description', 'twitter:description'], + webpage, 'title', fatal=False), + }) + return info diff --git a/youtube_dl/extractor/telegraaf.py b/youtube_dl/extractor/telegraaf.py index 6f8333cfc..9092e9b85 100644 --- a/youtube_dl/extractor/telegraaf.py +++ b/youtube_dl/extractor/telegraaf.py @@ -2,14 +2,16 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import remove_end +from ..utils import ( + determine_ext, + remove_end, +) class TelegraafIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/tv/(?:[^/]+/)+(?P<id>\d+)/[^/]+\.html' _TEST = { 'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html', - 'md5': '83245a9779bcc4a24454bfd53c65b6dc', 'info_dict': { 'id': '24353229', 'ext': 'mp4', @@ -18,18 +20,60 @@ class TelegraafIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 33, }, + 'params': { + # m3u8 download + 'skip_download': True, + }, } def _real_extract(self, url): - playlist_id = self._match_id(url) + video_id = self._match_id(url) - webpage = self._download_webpage(url, playlist_id) + webpage = self._download_webpage(url, video_id) + player_url = self._html_search_regex( + r'<iframe[^>]+src="([^"]+")', webpage, 'player URL') + player_page = self._download_webpage( + player_url, video_id, note='Download player webpage') playlist_url = self._search_regex( - r"iframe\.loadPlayer\('([^']+)'", webpage, 'player') + r'playlist\s*:\s*"([^"]+)"', player_page, 'playlist URL') + playlist_data = self._download_json(playlist_url, video_id) + + item = playlist_data['items'][0] + formats = [] + locations = item['locations'] + for location in locations.get('adaptive', []): + manifest_url = location['src'] + ext = determine_ext(manifest_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + manifest_url, video_id, ext='mp4', m3u8_id='hls')) + elif ext == 'mpd': + # TODO: Current DASH formats are broken - $Time$ pattern in + # <SegmentTemplate> not implemented yet + continue + else: + self.report_warning('Unknown adaptive format %s' % ext) + for location in locations.get('progressive', []): + formats.append({ + 'url': location['sources'][0]['src'], + 'width': location.get('width'), + 'height': location.get('height'), + 'format_id': 'http-%s' % location['label'], + }) + + self._sort_formats(formats) - entries = self._extract_xspf_playlist(playlist_url, playlist_id) title = remove_end(self._og_search_title(webpage), ' - VIDEO') description = self._og_search_description(webpage) + duration = item.get('duration') + thumbnail = item.get('poster') - return self.playlist_result(entries, playlist_id, title, description) + return { + 'id': video_id, + 'title': title, + 'description': description, + 'formats': formats, + 'duration': duration, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/telewebion.py b/youtube_dl/extractor/telewebion.py new file mode 100644 index 000000000..77916c601 --- /dev/null +++ b/youtube_dl/extractor/telewebion.py @@ -0,0 +1,55 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class TelewebionIE(InfoExtractor): + _VALID_URL = r'https?://www\.telewebion\.com/#!/episode/(?P<id>\d+)' + + _TEST = { + 'url': 'http://www.telewebion.com/#!/episode/1263668/', + 'info_dict': { + 'id': '1263668', + 'ext': 'mp4', + 'title': 'قرعه\u200cکشی لیگ قهرمانان اروپا', + 'thumbnail': 're:^https?://.*\.jpg', + 'view_count': int, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + secure_token = self._download_webpage( + 'http://m.s2.telewebion.com/op/op?action=getSecurityToken', video_id) + episode_details = self._download_json( + 'http://m.s2.telewebion.com/op/op', video_id, + query={'action': 'getEpisodeDetails', 'episode_id': video_id}) + + m3u8_url = 'http://m.s1.telewebion.com/smil/%s.m3u8?filepath=%s&m3u8=1&secure_token=%s' % ( + video_id, episode_details['file_path'], secure_token) + formats = self._extract_m3u8_formats( + m3u8_url, video_id, ext='mp4', m3u8_id='hls') + + picture_paths = [ + episode_details.get('picture_path'), + episode_details.get('large_picture_path'), + ] + + thumbnails = [{ + 'url': picture_path, + 'preference': idx, + } for idx, picture_path in enumerate(picture_paths) if picture_path is not None] + + return { + 'id': video_id, + 'title': episode_details['title'], + 'formats': formats, + 'thumbnails': thumbnails, + 'view_count': episode_details.get('view_count'), + } diff --git a/youtube_dl/extractor/tf1.py b/youtube_dl/extractor/tf1.py index 3f54b2744..e595c4a69 100644 --- a/youtube_dl/extractor/tf1.py +++ b/youtube_dl/extractor/tf1.py @@ -6,7 +6,7 @@ from .common import InfoExtractor class TF1IE(InfoExtractor): """TF1 uses the wat.tv player.""" - _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|www\.tfou)\.fr/(?:[^/]+/)*(?P<id>.+?)\.html' + _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|(?:www\.)?(?:tfou|ushuaiatv|histoire|tvbreizh))\.fr/(?:[^/]+/)*(?P<id>[^/?#.]+)' _TESTS = [{ 'url': 'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html', 'info_dict': { @@ -48,6 +48,6 @@ class TF1IE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) wat_id = self._html_search_regex( - r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})(?:#.*?)?\1', + r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})\1', webpage, 'wat id', group='id') return self.url_result('wat:%s' % wat_id, 'Wat') diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 7a5a533b7..bb3efc4ea 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -6,6 +6,7 @@ import time import hmac import binascii import hashlib +import netrc from .once import OnceIE @@ -14,14 +15,19 @@ from ..compat import ( compat_urllib_parse_urlparse, ) from ..utils import ( + determine_ext, ExtractorError, float_or_none, int_or_none, sanitized_Request, unsmuggle_url, + update_url_query, xpath_with_ns, mimetype2ext, find_xpath_attr, + unescapeHTML, + urlencode_postdata, + unified_timestamp, ) default_ns = 'http://www.w3.org/2005/SMIL21/Language' @@ -48,16 +54,23 @@ class ThePlatformBaseIE(OnceIE): if OnceIE.suitable(_format['url']): formats.extend(self._extract_once_formats(_format['url'])) else: + media_url = _format['url'] + if determine_ext(media_url) == 'm3u8': + hdnea2 = self._get_cookies(media_url).get('hdnea2') + if hdnea2: + _format['url'] = update_url_query(media_url, {'hdnea3': hdnea2.value}) + formats.append(_format) subtitles = self._parse_smil_subtitles(meta, default_ns) return formats, subtitles - def get_metadata(self, path, video_id): + def _download_theplatform_metadata(self, path, video_id): info_url = 'http://link.theplatform.com/s/%s?format=preview' % path - info = self._download_json(info_url, video_id) + return self._download_json(info_url, video_id) + def _parse_theplatform_metadata(self, info): subtitles = {} captions = info.get('captions') if isinstance(captions, list): @@ -78,6 +91,10 @@ class ThePlatformBaseIE(OnceIE): 'uploader': info.get('billingCode'), } + def _extract_theplatform_metadata(self, path, video_id): + info = self._download_theplatform_metadata(path, video_id) + return self._parse_theplatform_metadata(info) + class ThePlatformIE(ThePlatformBaseIE): _VALID_URL = r'''(?x) @@ -150,6 +167,23 @@ class ThePlatformIE(ThePlatformBaseIE): 'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781', 'only_matching': True, }] + _SERVICE_PROVIDER_TEMPLATE = 'https://sp.auth.adobe.com/adobe-services/%s' + + @classmethod + def _extract_urls(cls, webpage): + m = re.search( + r'''(?x) + <meta\s+ + property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+ + content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2 + ''', webpage) + if m: + return [m.group('url')] + + matches = re.findall( + r'<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage) + if matches: + return list(zip(*matches))[1] @staticmethod def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False): @@ -159,15 +193,105 @@ class ThePlatformIE(ThePlatformBaseIE): def str_to_hex(str): return binascii.b2a_hex(str.encode('ascii')).decode('ascii') - def hex_to_str(hex): - return binascii.a2b_hex(hex) + def hex_to_bytes(hex): + return binascii.a2b_hex(hex.encode('ascii')) relative_path = re.match(r'https?://link.theplatform.com/s/([^?]+)', url).group(1) - clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path)) + clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path)) checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest() sig = flags + expiration_date + checksum + str_to_hex(sig_secret) return '%s&sig=%s' % (url, sig) + def _extract_mvpd_auth(self, url, video_id, requestor_id, resource): + def xml_text(xml_str, tag): + return self._search_regex( + '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag) + + mvpd_headers = { + 'ap_42': 'anonymous', + 'ap_11': 'Linux i686', + 'ap_z': 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0', + 'User-Agent': 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0', + } + + guid = xml_text(resource, 'guid') + requestor_info = self._downloader.cache.load('mvpd', requestor_id) or {} + authn_token = requestor_info.get('authn_token') + if authn_token: + token_expires = unified_timestamp(xml_text(authn_token, 'simpleTokenExpires').replace('_GMT', '')) + if token_expires and token_expires >= time.time(): + authn_token = None + if not authn_token: + # TODO add support for other TV Providers + mso_id = 'DTV' + login_info = netrc.netrc().authenticators(mso_id) + if not login_info: + return None + + def post_form(form_page, note, data={}): + post_url = self._html_search_regex(r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_page, 'post url', group='url') + return self._download_webpage( + post_url, video_id, note, data=urlencode_postdata(data or self._hidden_inputs(form_page)), headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + }) + + provider_redirect_page = self._download_webpage( + self._SERVICE_PROVIDER_TEMPLATE % 'authenticate/saml', video_id, + 'Downloading Provider Redirect Page', query={ + 'noflash': 'true', + 'mso_id': mso_id, + 'requestor_id': requestor_id, + 'no_iframe': 'false', + 'domain_name': 'adobe.com', + 'redirect_url': url, + }) + provider_login_page = post_form( + provider_redirect_page, 'Downloading Provider Login Page') + mvpd_confirm_page = post_form(provider_login_page, 'Logging in', { + 'username': login_info[0], + 'password': login_info[2], + }) + post_form(mvpd_confirm_page, 'Confirming Login') + + session = self._download_webpage( + self._SERVICE_PROVIDER_TEMPLATE % 'session', video_id, + 'Retrieving Session', data=urlencode_postdata({ + '_method': 'GET', + 'requestor_id': requestor_id, + }), headers=mvpd_headers) + authn_token = unescapeHTML(xml_text(session, 'authnToken')) + requestor_info['authn_token'] = authn_token + self._downloader.cache.store('mvpd', requestor_id, requestor_info) + + authz_token = requestor_info.get(guid) + if not authz_token: + authorize = self._download_webpage( + self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id, + 'Retrieving Authorization Token', data=urlencode_postdata({ + 'resource_id': resource, + 'requestor_id': requestor_id, + 'authentication_token': authn_token, + 'mso_id': xml_text(authn_token, 'simpleTokenMsoID'), + 'userMeta': '1', + }), headers=mvpd_headers) + authz_token = unescapeHTML(xml_text(authorize, 'authzToken')) + requestor_info[guid] = authz_token + self._downloader.cache.store('mvpd', requestor_id, requestor_info) + + mvpd_headers.update({ + 'ap_19': xml_text(authn_token, 'simpleSamlNameID'), + 'ap_23': xml_text(authn_token, 'simpleSamlSessionIndex'), + }) + + return self._download_webpage( + self._SERVICE_PROVIDER_TEMPLATE % 'shortAuthorize', + video_id, 'Retrieving Media Token', data=urlencode_postdata({ + 'authz_token': authz_token, + 'requestor_id': requestor_id, + 'session_guid': xml_text(authn_token, 'simpleTokenAuthenticationGuid'), + 'hashed_guid': 'false', + }), headers=mvpd_headers) + def _real_extract(self, url): url, smuggled_data = unsmuggle_url(url, {}) @@ -241,7 +365,7 @@ class ThePlatformIE(ThePlatformBaseIE): formats, subtitles = self._extract_theplatform_smil(smil_url, video_id) self._sort_formats(formats) - ret = self.get_metadata(path, video_id) + ret = self._extract_theplatform_metadata(path, video_id) combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles) ret.update({ 'id': video_id, @@ -253,9 +377,9 @@ class ThePlatformIE(ThePlatformBaseIE): class ThePlatformFeedIE(ThePlatformBaseIE): - _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&byGuid=%s' - _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*byGuid=(?P<id>[a-zA-Z0-9_]+)' - _TEST = { + _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&%s' + _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*(?P<filter>by(?:Gui|I)d=(?P<id>[\w-]+))' + _TESTS = [{ # From http://player.theplatform.com/p/7wvmTC/MSNBCEmbeddedOffSite?guid=n_hardball_5biden_140207 'url': 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207', 'md5': '6e32495b5073ab414471b615c5ded394', @@ -269,33 +393,40 @@ class ThePlatformFeedIE(ThePlatformBaseIE): 'timestamp': 1391824260, 'duration': 467.0, 'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'], + 'uploader': 'NBCU-NEWS', }, - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - provider_id = mobj.group('provider_id') - feed_id = mobj.group('feed_id') + }] - real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, video_id) - feed = self._download_json(real_url, video_id) - entry = feed['entries'][0] + def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}): + real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query) + entry = self._download_json(real_url, video_id)['entries'][0] formats = [] subtitles = {} first_video_id = None duration = None + asset_types = [] for item in entry['media$content']: - smil_url = item['plfile$url'] + '&mbr=true' + smil_url = item['plfile$url'] cur_video_id = ThePlatformIE._match_id(smil_url) if first_video_id is None: first_video_id = cur_video_id duration = float_or_none(item.get('plfile$duration')) - cur_formats, cur_subtitles = self._extract_theplatform_smil(smil_url, video_id, 'Downloading SMIL data for %s' % cur_video_id) - formats.extend(cur_formats) - subtitles = self._merge_subtitles(subtitles, cur_subtitles) + for asset_type in item['plfile$assetTypes']: + if asset_type in asset_types: + continue + asset_types.append(asset_type) + query = { + 'mbr': 'true', + 'formats': item['plfile$format'], + 'assetTypes': asset_type, + } + if asset_type in asset_types_query: + query.update(asset_types_query[asset_type]) + cur_formats, cur_subtitles = self._extract_theplatform_smil(update_url_query( + smil_url, query), video_id, 'Downloading SMIL data for %s' % asset_type) + formats.extend(cur_formats) + subtitles = self._merge_subtitles(subtitles, cur_subtitles) self._sort_formats(formats) @@ -308,7 +439,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE): timestamp = int_or_none(entry.get('media$availableDate'), scale=1000) categories = [item['media$name'] for item in entry.get('media$categories', [])] - ret = self.get_metadata('%s/%s' % (provider_id, first_video_id), video_id) + ret = self._extract_theplatform_metadata('%s/%s' % (provider_id, first_video_id), video_id) subtitles = self._merge_subtitles(subtitles, ret['subtitles']) ret.update({ 'id': video_id, @@ -319,5 +450,17 @@ class ThePlatformFeedIE(ThePlatformBaseIE): 'timestamp': timestamp, 'categories': categories, }) + if custom_fields: + ret.update(custom_fields(entry)) return ret + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('id') + provider_id = mobj.group('provider_id') + feed_id = mobj.group('feed_id') + filter_query = mobj.group('filter') + + return self._extract_feed_info(provider_id, feed_id, filter_query, video_id) diff --git a/youtube_dl/extractor/thesixtyone.py b/youtube_dl/extractor/thesixtyone.py index d8b1fd281..d63aef5de 100644 --- a/youtube_dl/extractor/thesixtyone.py +++ b/youtube_dl/extractor/thesixtyone.py @@ -12,7 +12,7 @@ class TheSixtyOneIE(InfoExtractor): s| song/comments/list| song - )/(?P<id>[A-Za-z0-9]+)/?$''' + )/(?:[^/]+/)?(?P<id>[A-Za-z0-9]+)/?$''' _SONG_URL_TEMPLATE = 'http://thesixtyone.com/s/{0:}' _SONG_FILE_URL_TEMPLATE = 'http://{audio_server:}/thesixtyone_production/audio/{0:}_stream' _THUMBNAIL_URL_TEMPLATE = '{photo_base_url:}_desktop' @@ -45,6 +45,10 @@ class TheSixtyOneIE(InfoExtractor): 'url': 'http://www.thesixtyone.com/song/SrE3zD7s1jt/', 'only_matching': True, }, + { + 'url': 'http://www.thesixtyone.com/maryatmidnight/song/StrawberriesandCream/yvWtLp0c4GQ/', + 'only_matching': True, + }, ] _DECODE_MAP = { diff --git a/youtube_dl/extractor/threeqsdn.py b/youtube_dl/extractor/threeqsdn.py new file mode 100644 index 000000000..a0bc12c81 --- /dev/null +++ b/youtube_dl/extractor/threeqsdn.py @@ -0,0 +1,138 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + js_to_json, + mimetype2ext, +) + + +class ThreeQSDNIE(InfoExtractor): + IE_NAME = '3qsdn' + IE_DESC = '3Q SDN' + _VALID_URL = r'https?://playout\.3qsdn\.com/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _TESTS = [{ + # ondemand from http://www.philharmonie.tv/veranstaltung/26/ + 'url': 'http://playout.3qsdn.com/0280d6b9-1215-11e6-b427-0cc47a188158?protocol=http', + 'md5': 'ab040e37bcfa2e0c079f92cb1dd7f6cd', + 'info_dict': { + 'id': '0280d6b9-1215-11e6-b427-0cc47a188158', + 'ext': 'mp4', + 'title': '0280d6b9-1215-11e6-b427-0cc47a188158', + 'is_live': False, + }, + 'expected_warnings': ['Failed to download MPD manifest'], + }, { + # live video stream + 'url': 'https://playout.3qsdn.com/d755d94b-4ab9-11e3-9162-0025907ad44f?js=true', + 'info_dict': { + 'id': 'd755d94b-4ab9-11e3-9162-0025907ad44f', + 'ext': 'mp4', + 'title': 'd755d94b-4ab9-11e3-9162-0025907ad44f', + 'is_live': False, + }, + }, { + # live audio stream + 'url': 'http://playout.3qsdn.com/9edf36e0-6bf2-11e2-a16a-9acf09e2db48', + 'only_matching': True, + }, { + # live audio stream with some 404 URLs + 'url': 'http://playout.3qsdn.com/ac5c3186-777a-11e2-9c30-9acf09e2db48', + 'only_matching': True, + }, { + # geo restricted with 'This content is not available in your country' + 'url': 'http://playout.3qsdn.com/d63a3ffe-75e8-11e2-9c30-9acf09e2db48', + 'only_matching': True, + }, { + # geo restricted with 'playout.3qsdn.com/forbidden' + 'url': 'http://playout.3qsdn.com/8e330f26-6ae2-11e2-a16a-9acf09e2db48', + 'only_matching': True, + }, { + # live video with rtmp link + 'url': 'https://playout.3qsdn.com/6092bb9e-8f72-11e4-a173-002590c750be', + 'only_matching': True, + }] + + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r'<iframe[^>]+\b(?:data-)?src=(["\'])(?P<url>%s.*?)\1' % ThreeQSDNIE._VALID_URL, webpage) + if mobj: + return mobj.group('url') + + def _real_extract(self, url): + video_id = self._match_id(url) + + js = self._download_webpage( + 'http://playout.3qsdn.com/%s' % video_id, video_id, + query={'js': 'true'}) + + if any(p in js for p in ( + '>This content is not available in your country', + 'playout.3qsdn.com/forbidden')): + self.raise_geo_restricted() + + stream_content = self._search_regex( + r'streamContent\s*:\s*(["\'])(?P<content>.+?)\1', js, + 'stream content', default='demand', group='content') + + live = stream_content == 'live' + + stream_type = self._search_regex( + r'streamType\s*:\s*(["\'])(?P<type>audio|video)\1', js, + 'stream type', default='video', group='type') + + formats = [] + urls = set() + + def extract_formats(item_url, item={}): + if not item_url or item_url in urls: + return + urls.add(item_url) + ext = mimetype2ext(item.get('type')) or determine_ext(item_url, default_ext=None) + if ext == 'mpd': + formats.extend(self._extract_mpd_formats( + item_url, video_id, mpd_id='mpd', fatal=False)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + item_url, video_id, 'mp4', + entry_protocol='m3u8' if live else 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + item_url, video_id, f4m_id='hds', fatal=False)) + else: + if not self._is_valid_url(item_url, video_id): + return + formats.append({ + 'url': item_url, + 'format_id': item.get('quality'), + 'ext': 'mp4' if item_url.startswith('rtsp') else ext, + 'vcodec': 'none' if stream_type == 'audio' else None, + }) + + for item_js in re.findall(r'({.*?\b(?:src|source)\s*:\s*["\'].+?})', js): + f = self._parse_json( + item_js, video_id, transform_source=js_to_json, fatal=False) + if not f: + continue + extract_formats(f.get('src'), f) + + # More relaxed version to collect additional URLs and acting + # as a future-proof fallback + for _, src in re.findall(r'\b(?:src|source)\s*:\s*(["\'])((?:https?|rtsp)://.+?)\1', js): + extract_formats(src) + + self._sort_formats(formats) + + title = self._live_title(video_id) if live else video_id + + return { + 'id': video_id, + 'title': title, + 'is_live': live, + 'formats': formats, + } diff --git a/youtube_dl/extractor/toutv.py b/youtube_dl/extractor/toutv.py index 4797d1310..54c2d0aa6 100644 --- a/youtube_dl/extractor/toutv.py +++ b/youtube_dl/extractor/toutv.py @@ -1,74 +1,41 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..utils import ( - ExtractorError, - unified_strdate, -) +from ..utils import int_or_none class TouTvIE(InfoExtractor): IE_NAME = 'tou.tv' - _VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))' + _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+/S[0-9]+E[0-9]+)' _TEST = { - 'url': 'http://www.tou.tv/30-vies/S04E41', + 'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17', 'info_dict': { - 'id': '30-vies_S04E41', + 'id': '122017', 'ext': 'mp4', - 'title': '30 vies Saison 4 / Épisode 41', - 'description': 'md5:da363002db82ccbe4dafeb9cab039b09', - 'age_limit': 8, - 'uploader': 'Groupe des Nouveaux Médias', - 'duration': 1296, - 'upload_date': '20131118', - 'thumbnail': 'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg', + 'title': 'Saison 2015 Épisode 17', + 'description': 'La photo de famille 2', + 'upload_date': '20100717', }, 'params': { - 'skip_download': True, # Requires rtmpdump + # m3u8 download + 'skip_download': True, }, - 'skip': 'Only available in Canada' } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) - - mediaId = self._search_regex( - r'"idMedia":\s*"([^"]+)"', webpage, 'media ID') - - streams_url = 'http://release.theplatform.com/content.select?pid=' + mediaId - streams_doc = self._download_xml( - streams_url, video_id, note='Downloading stream list') - - video_url = next(n.text - for n in streams_doc.findall('.//choice/url') - if '//ad.doubleclick' not in n.text) - if video_url.endswith('/Unavailable.flv'): - raise ExtractorError( - 'Access to this video is blocked from outside of Canada', - expected=True) - - duration_str = self._html_search_meta( - 'video:duration', webpage, 'duration') - duration = int(duration_str) if duration_str else None - upload_date_str = self._html_search_meta( - 'video:release_date', webpage, 'upload date') - upload_date = unified_strdate(upload_date_str) if upload_date_str else None + path = self._match_id(url) + metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path) + video_id = metadata['IdMedia'] + details = metadata['Details'] + title = details['OriginalTitle'] return { + '_type': 'url_transparent', + 'url': 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id), 'id': video_id, - 'title': self._og_search_title(webpage), - 'url': video_url, - 'description': self._og_search_description(webpage), - 'uploader': self._dc_search_uploader(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), - 'age_limit': self._media_rating_search(webpage), - 'duration': duration, - 'upload_date': upload_date, - 'ext': 'mp4', + 'title': title, + 'thumbnail': details.get('ImageUrl'), + 'duration': int_or_none(details.get('LengthInSeconds')), } diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index f57d609d4..5070082da 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -1,25 +1,24 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import ( + determine_ext, + clean_html, + get_element_by_attribute, + ExtractorError, +) -class TvpIE(InfoExtractor): - IE_NAME = 'tvp.pl' - _VALID_URL = r'https?://(?:vod|www)\.tvp\.pl/.*/(?P<id>\d+)$' +class TVPIE(InfoExtractor): + IE_NAME = 'tvp' + IE_DESC = 'Telewizja Polska' + _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)' _TESTS = [{ - 'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem/wideo/odc-2/4278035', - 'md5': 'cdd98303338b8a7f7abab5cd14092bf2', - 'info_dict': { - 'id': '4278035', - 'ext': 'wmv', - 'title': 'Ogniem i mieczem, odc. 2', - }, - }, { - 'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536', + 'url': 'http://vod.tvp.pl/194536/i-seria-odc-13', 'md5': '8aa518c15e5cc32dfe8db400dc921fbb', 'info_dict': { 'id': '194536', @@ -28,7 +27,7 @@ class TvpIE(InfoExtractor): }, }, { 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176', - 'md5': 'c3b15ed1af288131115ff17a17c19dda', + 'md5': 'b0005b542e5b4de643a9690326ab1257', 'info_dict': { 'id': '17916176', 'ext': 'mp4', @@ -36,12 +35,22 @@ class TvpIE(InfoExtractor): }, }, { 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272', - 'md5': 'c3b15ed1af288131115ff17a17c19dda', - 'info_dict': { - 'id': '17834272', - 'ext': 'mp4', - 'title': 'Na sygnale, odc. 39', - }, + 'only_matching': True, + }, { + 'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200', + 'only_matching': True, + }, { + 'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa', + 'only_matching': True, + }, { + 'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach', + 'only_matching': True, + }, { + 'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum', + 'only_matching': True, + }, { + 'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji', + 'only_matching': True, }] def _real_extract(self, url): @@ -50,6 +59,11 @@ class TvpIE(InfoExtractor): webpage = self._download_webpage( 'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id) + error_massage = get_element_by_attribute('class', 'msg error', webpage) + if error_massage: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, clean_html(error_massage)), expected=True) + title = self._search_regex( r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1', webpage, 'title', group='title') @@ -63,24 +77,50 @@ class TvpIE(InfoExtractor): r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None) video_url = self._search_regex( - r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None) - if not video_url: + r'0:{src:([\'"])(?P<url>.*?)\1', webpage, + 'formats', group='url', default=None) + if not video_url or 'material_niedostepny.mp4' in video_url: video_url = self._download_json( 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id, video_id)['video_url'] - ext = video_url.rsplit('.', 1)[-1] - if ext != 'ism/manifest': - if '/' in ext: - ext = 'mp4' + formats = [] + video_url_base = self._search_regex( + r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)', + video_url, 'video base url', default=None) + if video_url_base: + # TODO: Current DASH formats are broken - $Time$ pattern in + # <SegmentTemplate> not implemented yet + # formats.extend(self._extract_mpd_formats( + # video_url_base + '.ism/video.mpd', + # video_id, mpd_id='dash', fatal=False)) + formats.extend(self._extract_f4m_formats( + video_url_base + '.ism/video.f4m', + video_id, f4m_id='hds', fatal=False)) + m3u8_formats = self._extract_m3u8_formats( + video_url_base + '.ism/video.m3u8', video_id, + 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + self._sort_formats(m3u8_formats) + m3u8_formats = list(filter( + lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + m3u8_formats)) + formats.extend(m3u8_formats) + for i, m3u8_format in enumerate(m3u8_formats, 2): + http_url = '%s-%d.mp4' % (video_url_base, i) + if self._is_valid_url(http_url, video_id): + f = m3u8_format.copy() + f.update({ + 'url': http_url, + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) + else: formats = [{ 'format_id': 'direct', 'url': video_url, - 'ext': ext, + 'ext': determine_ext(video_url, 'mp4'), }] - else: - m3u8_url = re.sub('([^/]*)\.ism/manifest', r'\1.ism/\1.m3u8', video_url) - formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4') self._sort_formats(formats) @@ -92,8 +132,8 @@ class TvpIE(InfoExtractor): } -class TvpSeriesIE(InfoExtractor): - IE_NAME = 'tvp.pl:Series' +class TVPSeriesIE(InfoExtractor): + IE_NAME = 'tvp:series' _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$' _TESTS = [{ @@ -127,7 +167,7 @@ class TvpSeriesIE(InfoExtractor): videos_paths = re.findall( '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist) entries = [ - self.url_result('http://vod.tvp.pl%s' % v_path, ie=TvpIE.ie_key()) + self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key()) for v_path in videos_paths] return { diff --git a/youtube_dl/extractor/tweakers.py b/youtube_dl/extractor/tweakers.py index f3198fb85..7a9386cde 100644 --- a/youtube_dl/extractor/tweakers.py +++ b/youtube_dl/extractor/tweakers.py @@ -1,25 +1,62 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import ( + int_or_none, + determine_ext, + mimetype2ext, +) class TweakersIE(InfoExtractor): _VALID_URL = r'https?://tweakers\.net/video/(?P<id>\d+)' _TEST = { 'url': 'https://tweakers.net/video/9926/new-nintendo-3ds-xl-op-alle-fronten-beter.html', - 'md5': '3147e4ddad366f97476a93863e4557c8', + 'md5': 'fe73e417c093a788e0160c4025f88b15', 'info_dict': { 'id': '9926', 'ext': 'mp4', 'title': 'New Nintendo 3DS XL - Op alle fronten beter', - 'description': 'md5:f97324cc71e86e11c853f0763820e3ba', + 'description': 'md5:3789b21fed9c0219e9bcaacd43fab280', 'thumbnail': 're:^https?://.*\.jpe?g$', 'duration': 386, + 'uploader_id': 's7JeEm', } } def _real_extract(self, url): - playlist_id = self._match_id(url) - entries = self._extract_xspf_playlist( - 'https://tweakers.net/video/s1playlist/%s/playlist.xspf' % playlist_id, playlist_id) - return self.playlist_result(entries, playlist_id) + video_id = self._match_id(url) + video_data = self._download_json( + 'https://tweakers.net/video/s1playlist/%s/1920/1080/playlist.json' % video_id, + video_id)['items'][0] + + title = video_data['title'] + + formats = [] + for location in video_data.get('locations', {}).get('progressive', []): + format_id = location.get('label') + width = int_or_none(location.get('width')) + height = int_or_none(location.get('height')) + for source in location.get('sources', []): + source_url = source.get('src') + if not source_url: + continue + ext = mimetype2ext(source.get('type')) or determine_ext(source_url) + formats.append({ + 'format_id': format_id, + 'url': source_url, + 'width': width, + 'height': height, + 'ext': ext, + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': video_data.get('description'), + 'thumbnail': video_data.get('poster'), + 'duration': int_or_none(video_data.get('duration')), + 'uploader_id': video_data.get('account'), + 'formats': formats, + } diff --git a/youtube_dl/extractor/twentyfourvideo.py b/youtube_dl/extractor/twentyfourvideo.py index e03e2dbaa..4025edf02 100644 --- a/youtube_dl/extractor/twentyfourvideo.py +++ b/youtube_dl/extractor/twentyfourvideo.py @@ -47,7 +47,8 @@ class TwentyFourVideoIE(InfoExtractor): title = self._og_search_title(webpage) description = self._html_search_regex( - r'<span itemprop="description">([^<]+)</span>', webpage, 'description', fatal=False) + r'<(p|span)[^>]+itemprop="description"[^>]*>(?P<description>[^<]+)</\1>', + webpage, 'description', fatal=False, group='description') thumbnail = self._og_search_thumbnail(webpage) duration = int_or_none(self._og_search_property( 'duration', webpage, 'duration', fatal=False)) diff --git a/youtube_dl/extractor/twentymin.py b/youtube_dl/extractor/twentymin.py index ca7d953b8..b721ecb0a 100644 --- a/youtube_dl/extractor/twentymin.py +++ b/youtube_dl/extractor/twentymin.py @@ -32,7 +32,22 @@ class TwentyMinutenIE(InfoExtractor): 'title': '«Wir müssen mutig nach vorne schauen»', 'description': 'Kein Land sei innovativer als die Schweiz, sagte Johann Schneider-Ammann in seiner Neujahrsansprache. Das Land müsse aber seine Hausaufgaben machen.', 'thumbnail': 'http://www.20min.ch/images/content/2/2/0/22050469/10/teaserbreit.jpg' - } + }, + 'skip': '"This video is no longer available" is shown both on the web page and in the downloaded file.', + }, { + # YouTube embed + 'url': 'http://www.20min.ch/ro/sports/football/story/Il-marque-une-bicyclette-de-plus-de-30-metres--21115184', + 'md5': 'cec64d59aa01c0ed9dbba9cf639dd82f', + 'info_dict': { + 'id': 'ivM7A7SpDOs', + 'ext': 'mp4', + 'title': 'GOLAZO DE CHILENA DE JAVI GÓMEZ, FINALISTA AL BALÓN DE CLM 2016', + 'description': 'md5:903c92fbf2b2f66c09de514bc25e9f5a', + 'upload_date': '20160424', + 'uploader': 'RTVCM Castilla-La Mancha', + 'uploader_id': 'RTVCM', + }, + 'add_ie': ['Youtube'], }, { 'url': 'http://www.20min.ch/videotv/?cid=44&vid=468738', 'only_matching': True, @@ -48,6 +63,12 @@ class TwentyMinutenIE(InfoExtractor): webpage = self._download_webpage(url, display_id) + youtube_url = self._html_search_regex( + r'<iframe[^>]+src="((?:https?:)?//www\.youtube\.com/embed/[^"]+)"', + webpage, 'YouTube embed URL', default=None) + if youtube_url is not None: + return self.url_result(youtube_url, 'Youtube') + title = self._html_search_regex( r'<h1>.*?<span>(.+?)</span></h1>', webpage, 'title', default=None) diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index 36ee1adff..67b1277cc 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -16,6 +16,7 @@ from ..compat import ( from ..utils import ( ExtractorError, int_or_none, + js_to_json, orderedSet, parse_duration, parse_iso8601, @@ -28,7 +29,7 @@ class TwitchBaseIE(InfoExtractor): _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv' _API_BASE = 'https://api.twitch.tv' - _USHER_BASE = 'http://usher.twitch.tv' + _USHER_BASE = 'https://usher.ttvnw.net' _LOGIN_URL = 'http://www.twitch.tv/login' _NETRC_MACHINE = 'twitch' @@ -171,6 +172,7 @@ class TwitchVideoIE(TwitchItemBaseIE): 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', }, 'playlist_mincount': 12, + 'skip': 'HTTP Error 404: Not Found', } @@ -187,6 +189,7 @@ class TwitchChapterIE(TwitchItemBaseIE): 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', }, 'playlist_mincount': 3, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361', 'only_matching': True, @@ -258,7 +261,7 @@ class TwitchVodIE(TwitchItemBaseIE): 'nauth': access_token['token'], 'nauthsig': access_token['sig'], })), - item_id, 'mp4') + item_id, 'mp4', entry_protocol='m3u8_native') self._prefer_source(formats) info['formats'] = formats @@ -355,31 +358,6 @@ class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE): } -class TwitchBookmarksIE(TwitchPlaylistBaseIE): - IE_NAME = 'twitch:bookmarks' - _VALID_URL = r'%s/(?P<id>[^/]+)/profile/bookmarks/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE - _PLAYLIST_URL = '%s/api/bookmark/?user=%%s&offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE - _PLAYLIST_TYPE = 'bookmarks' - - _TEST = { - 'url': 'http://www.twitch.tv/ognos/profile/bookmarks', - 'info_dict': { - 'id': 'ognos', - 'title': 'Ognos', - }, - 'playlist_mincount': 3, - } - - def _extract_playlist_page(self, response): - entries = [] - for bookmark in response.get('bookmarks', []): - video = bookmark.get('video') - if not video: - continue - entries.append(video['url']) - return entries - - class TwitchStreamIE(TwitchBaseIE): IE_NAME = 'twitch:stream' _VALID_URL = r'%s/(?P<id>[^/#?]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE @@ -477,3 +455,45 @@ class TwitchStreamIE(TwitchBaseIE): 'formats': formats, 'is_live': True, } + + +class TwitchClipsIE(InfoExtractor): + IE_NAME = 'twitch:clips' + _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)' + + _TEST = { + 'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound', + 'md5': '761769e1eafce0ffebfb4089cb3847cd', + 'info_dict': { + 'id': 'AggressiveCobraPoooound', + 'ext': 'mp4', + 'title': 'EA Play 2016 Live from the Novo Theatre', + 'thumbnail': 're:^https?://.*\.jpg', + 'creator': 'EA', + 'uploader': 'stereotype_', + 'uploader_id': 'stereotype_', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + clip = self._parse_json( + self._search_regex( + r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'), + video_id, transform_source=js_to_json) + + video_url = clip['clip_video_url'] + title = clip['channel_title'] + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'thumbnail': self._og_search_thumbnail(webpage), + 'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'), + 'uploader': clip.get('curator_login'), + 'uploader_id': clip.get('curator_display_name'), + } diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index ea673054f..b73842986 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -5,6 +5,7 @@ import re from .common import InfoExtractor from ..utils import ( + determine_ext, float_or_none, xpath_text, remove_end, @@ -52,7 +53,7 @@ class TwitterCardIE(TwitterBaseIE): 'id': 'dq4Oj5quskI', 'ext': 'mp4', 'title': 'Ubuntu 11.10 Overview', - 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/', + 'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10...', 'upload_date': '20111013', 'uploader': 'OMG! Ubuntu!', 'uploader_id': 'omgubuntu', @@ -116,13 +117,16 @@ class TwitterCardIE(TwitterBaseIE): video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source') if video_url: - f = { - 'url': video_url, - } + if determine_ext(video_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls')) + else: + f = { + 'url': video_url, + } - _search_dimensions_in_video_url(f, video_url) + _search_dimensions_in_video_url(f, video_url) - formats.append(f) + formats.append(f) vmap_url = config.get('vmapUrl') or config.get('vmap_url') if vmap_url: @@ -207,6 +211,7 @@ class TwitterIE(InfoExtractor): 'uploader_id': 'giphz', }, 'expected_warnings': ['height', 'width'], + 'skip': 'Account suspended', }, { 'url': 'https://twitter.com/starwars/status/665052190608723968', 'md5': '39b7199856dee6cd4432e72c74bc69d4', @@ -239,10 +244,10 @@ class TwitterIE(InfoExtractor): 'info_dict': { 'id': '700207533655363584', 'ext': 'mp4', - 'title': 'jay - BEAT PROD: @suhmeduh #Damndaniel', - 'description': 'jay on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"', + 'title': 'Donte The Dumbass - BEAT PROD: @suhmeduh #Damndaniel', + 'description': 'Donte The Dumbass on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"', 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'jay', + 'uploader': 'Donte The Dumbass', 'uploader_id': 'jaydingeer', }, 'params': { @@ -262,7 +267,6 @@ class TwitterIE(InfoExtractor): 'add_ie': ['Vine'], }, { 'url': 'https://twitter.com/captainamerica/status/719944021058060289', - # md5 constantly changes 'info_dict': { 'id': '719944021058060289', 'ext': 'mp4', @@ -271,6 +275,9 @@ class TwitterIE(InfoExtractor): 'uploader_id': 'captainamerica', 'uploader': 'Captain America', }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, }] def _real_extract(self, url): @@ -278,7 +285,11 @@ class TwitterIE(InfoExtractor): user_id = mobj.group('user_id') twid = mobj.group('id') - webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid) + webpage, urlh = self._download_webpage_handle( + self._TEMPLATE_URL % (user_id, twid), twid) + + if 'twitter.com/account/suspended' in urlh.geturl(): + raise ExtractorError('Account suspended by Twitter.', expected=True) username = remove_end(self._og_search_title(webpage), ' on Twitter') diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index d1e6f2703..89b869559 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_parse_urlencode, compat_urllib_request, compat_urlparse, ) @@ -84,18 +83,19 @@ class UdemyIE(InfoExtractor): if enroll_url: webpage = self._download_webpage( combine_url(base_url, enroll_url), - course_id, 'Enrolling in the course') + course_id, 'Enrolling in the course', + headers={'Referer': base_url}) if '>You have enrolled in' in webpage: self.to_screen('%s: Successfully enrolled in the course' % course_id) def _download_lecture(self, course_id, lecture_id): return self._download_json( - 'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % ( - course_id, lecture_id, compat_urllib_parse_urlencode({ - 'fields[lecture]': 'title,description,view_html,asset', - 'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data', - })), - lecture_id, 'Downloading lecture JSON') + 'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?' + % (course_id, lecture_id), + lecture_id, 'Downloading lecture JSON', query={ + 'fields[lecture]': 'title,description,view_html,asset', + 'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data', + }) def _handle_error(self, response): if not isinstance(response, dict): @@ -142,7 +142,9 @@ class UdemyIE(InfoExtractor): self._LOGIN_URL, None, 'Downloading login popup') def is_logged(webpage): - return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<']) + return any(re.search(p, webpage) for p in ( + r'href=["\'](?:https://www\.udemy\.com)?/user/logout/', + r'>Logout<')) # already logged in if is_logged(login_popup): @@ -155,13 +157,13 @@ class UdemyIE(InfoExtractor): 'password': password, }) - request = sanitized_Request( - self._LOGIN_URL, urlencode_postdata(login_form)) - request.add_header('Referer', self._ORIGIN_URL) - request.add_header('Origin', self._ORIGIN_URL) - response = self._download_webpage( - request, None, 'Logging in as %s' % username) + self._LOGIN_URL, None, 'Logging in as %s' % username, + data=urlencode_postdata(login_form), + headers={ + 'Referer': self._ORIGIN_URL, + 'Origin': self._ORIGIN_URL, + }) if not is_logged(response): error = self._html_search_regex( diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py index ee35b7227..57dd73aef 100644 --- a/youtube_dl/extractor/udn.py +++ b/youtube_dl/extractor/udn.py @@ -2,10 +2,13 @@ from __future__ import unicode_literals import json +import re + from .common import InfoExtractor from ..utils import ( + determine_ext, + int_or_none, js_to_json, - ExtractorError, ) from ..compat import compat_urlparse @@ -16,13 +19,16 @@ class UDNEmbedIE(InfoExtractor): _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL _TESTS = [{ 'url': 'http://video.udn.com/embed/news/300040', - 'md5': 'de06b4c90b042c128395a88f0384817e', 'info_dict': { 'id': '300040', 'ext': 'mp4', 'title': '生物老師男變女 全校挺"做自己"', 'thumbnail': 're:^https?://.*\.jpg$', - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'https://video.udn.com/embed/news/300040', 'only_matching': True, @@ -38,39 +44,53 @@ class UDNEmbedIE(InfoExtractor): page = self._download_webpage(url, video_id) options = json.loads(js_to_json(self._html_search_regex( - r'var options\s*=\s*([^;]+);', page, 'video urls dictionary'))) + r'var\s+options\s*=\s*([^;]+);', page, 'video urls dictionary'))) video_urls = options['video'] if video_urls.get('youtube'): return self.url_result(video_urls.get('youtube'), 'Youtube') - try: - del video_urls['youtube'] - except KeyError: - pass + formats = [] + for video_type, api_url in video_urls.items(): + if not api_url: + continue - formats = [{ - 'url': self._download_webpage( + video_url = self._download_webpage( compat_urlparse.urljoin(url, api_url), video_id, - 'retrieve url for %s video' % video_type), - 'format_id': video_type, - 'preference': 0 if video_type == 'mp4' else -1, - } for video_type, api_url in video_urls.items() if api_url] + note='retrieve url for %s video' % video_type) - if not formats: - raise ExtractorError('No videos found', expected=True) + ext = determine_ext(video_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, ext='mp4', m3u8_id='hls')) + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + video_url, video_id, f4m_id='hds')) + else: + mobj = re.search(r'_(?P<height>\d+)p_(?P<tbr>\d+).mp4', video_url) + a_format = { + 'url': video_url, + # video_type may be 'mp4', which confuses YoutubeDL + 'format_id': 'http-' + video_type, + } + if mobj: + a_format.update({ + 'height': int_or_none(mobj.group('height')), + 'tbr': int_or_none(mobj.group('tbr')), + }) + formats.append(a_format) self._sort_formats(formats) - thumbnail = None - - if options.get('gallery') and len(options['gallery']): - thumbnail = options['gallery'][0].get('original') + thumbnails = [{ + 'url': img_url, + 'id': img_type, + } for img_type, img_url in options.get('gallery', [{}])[0].items() if img_url] return { 'id': video_id, 'formats': formats, 'title': options['title'], - 'thumbnail': thumbnail + 'thumbnails': thumbnails, } diff --git a/youtube_dl/extractor/unistra.py b/youtube_dl/extractor/unistra.py index 66d9f1bf3..a724cdbef 100644 --- a/youtube_dl/extractor/unistra.py +++ b/youtube_dl/extractor/unistra.py @@ -49,6 +49,7 @@ class UnistraIE(InfoExtractor): 'format_id': format_id, 'quality': quality(format_id) }) + self._sort_formats(formats) title = self._html_search_regex( r'<title>UTV - (.*?)</', webpage, 'title') diff --git a/youtube_dl/extractor/urplay.py b/youtube_dl/extractor/urplay.py new file mode 100644 index 000000000..ce3bf6b02 --- /dev/null +++ b/youtube_dl/extractor/urplay.py @@ -0,0 +1,67 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class URPlayIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?urplay\.se/program/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://urplay.se/program/190031-tripp-trapp-trad-sovkudde', + 'md5': '15ca67b63fd8fb320ac2bcd854bad7b6', + 'info_dict': { + 'id': '190031', + 'ext': 'mp4', + 'title': 'Tripp, Trapp, Träd : Sovkudde', + 'description': 'md5:b86bffdae04a7e9379d1d7e5947df1d1', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + urplayer_data = self._parse_json(self._search_regex( + r'urPlayer\.init\(({.+?})\);', webpage, 'urplayer data'), video_id) + host = self._download_json('http://streaming-loadbalancer.ur.se/loadbalancer.json', video_id)['redirect'] + + formats = [] + for quality_attr, quality, preference in (('', 'sd', 0), ('_hd', 'hd', 1)): + file_rtmp = urplayer_data.get('file_rtmp' + quality_attr) + if file_rtmp: + formats.append({ + 'url': 'rtmp://%s/urplay/mp4:%s' % (host, file_rtmp), + 'format_id': quality + '-rtmp', + 'ext': 'flv', + 'preference': preference, + }) + file_http = urplayer_data.get('file_http' + quality_attr) or urplayer_data.get('file_http_sub' + quality_attr) + if file_http: + file_http_base_url = 'http://%s/%s' % (host, file_http) + formats.extend(self._extract_f4m_formats( + file_http_base_url + 'manifest.f4m', video_id, + preference, '%s-hds' % quality, fatal=False)) + formats.extend(self._extract_m3u8_formats( + file_http_base_url + 'playlist.m3u8', video_id, 'mp4', + 'm3u8_native', preference, '%s-hls' % quality, fatal=False)) + self._sort_formats(formats) + + subtitles = {} + for subtitle in urplayer_data.get('subtitles', []): + subtitle_url = subtitle.get('file') + kind = subtitle.get('kind') + if subtitle_url or kind and kind != 'captions': + continue + subtitles.setdefault(subtitle.get('label', 'Svenska'), []).append({ + 'url': subtitle_url, + }) + + return { + 'id': video_id, + 'title': urplayer_data['title'], + 'description': self._og_search_description(webpage), + 'thumbnail': urplayer_data.get('image'), + 'series': urplayer_data.get('series_title'), + 'subtitles': subtitles, + 'formats': formats, + } diff --git a/youtube_dl/extractor/ustudio.py b/youtube_dl/extractor/ustudio.py index cafc082b6..3484a2046 100644 --- a/youtube_dl/extractor/ustudio.py +++ b/youtube_dl/extractor/ustudio.py @@ -6,10 +6,12 @@ from .common import InfoExtractor from ..utils import ( int_or_none, unified_strdate, + unescapeHTML, ) class UstudioIE(InfoExtractor): + IE_NAME = 'ustudio' _VALID_URL = r'https?://(?:(?:www|v1)\.)?ustudio\.com/video/(?P<id>[^/]+)/(?P<display_id>[^/?#&]+)' _TEST = { 'url': 'http://ustudio.com/video/Uxu2my9bgSph/san_francisco_golden_gate_bridge', @@ -27,9 +29,7 @@ class UstudioIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') + video_id, display_id = re.match(self._VALID_URL, url).groups() config = self._download_xml( 'http://v1.ustudio.com/embed/%s/ustudio/config.xml' % video_id, @@ -37,7 +37,7 @@ class UstudioIE(InfoExtractor): def extract(kind): return [{ - 'url': item.attrib['url'], + 'url': unescapeHTML(item.attrib['url']), 'width': int_or_none(item.get('width')), 'height': int_or_none(item.get('height')), } for item in config.findall('./qualities/quality/%s' % kind) if item.get('url')] @@ -65,3 +65,61 @@ class UstudioIE(InfoExtractor): 'uploader': uploader, 'formats': formats, } + + +class UstudioEmbedIE(InfoExtractor): + IE_NAME = 'ustudio:embed' + _VALID_URL = r'https?://(?:(?:app|embed)\.)?ustudio\.com/embed/(?P<uid>[^/]+)/(?P<id>[^/]+)' + _TEST = { + 'url': 'http://app.ustudio.com/embed/DeN7VdYRDKhP/Uw7G1kMCe65T', + 'md5': '47c0be52a09b23a7f40de9469cec58f4', + 'info_dict': { + 'id': 'Uw7G1kMCe65T', + 'ext': 'mp4', + 'title': '5 Things IT Should Know About Video', + 'description': 'md5:93d32650884b500115e158c5677d25ad', + 'uploader_id': 'DeN7VdYRDKhP', + } + } + + def _real_extract(self, url): + uploader_id, video_id = re.match(self._VALID_URL, url).groups() + video_data = self._download_json( + 'http://app.ustudio.com/embed/%s/%s/config.json' % (uploader_id, video_id), + video_id)['videos'][0] + title = video_data['name'] + + formats = [] + for ext, qualities in video_data.get('transcodes', {}).items(): + for quality in qualities: + quality_url = quality.get('url') + if not quality_url: + continue + height = int_or_none(quality.get('height')) + formats.append({ + 'format_id': '%s-%dp' % (ext, height) if height else ext, + 'url': quality_url, + 'width': int_or_none(quality.get('width')), + 'height': height, + }) + self._sort_formats(formats) + + thumbnails = [] + for image in video_data.get('images', []): + image_url = image.get('url') + if not image_url: + continue + thumbnails.append({ + 'url': image_url, + }) + + return { + 'id': video_id, + 'title': title, + 'description': video_data.get('description'), + 'duration': int_or_none(video_data.get('duration')), + 'uploader_id': uploader_id, + 'tags': video_data.get('keywords'), + 'thumbnails': thumbnails, + 'formats': formats, + } diff --git a/youtube_dl/extractor/veoh.py b/youtube_dl/extractor/veoh.py index 23ce0a0d1..0f5d68738 100644 --- a/youtube_dl/extractor/veoh.py +++ b/youtube_dl/extractor/veoh.py @@ -37,6 +37,7 @@ class VeohIE(InfoExtractor): 'uploader': 'afp-news', 'duration': 123, }, + 'skip': 'This video has been deleted.', }, { 'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX', diff --git a/youtube_dl/extractor/vessel.py b/youtube_dl/extractor/vessel.py index 1a0ff3395..2cd617b91 100644 --- a/youtube_dl/extractor/vessel.py +++ b/youtube_dl/extractor/vessel.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import json +import re from .common import InfoExtractor from ..utils import ( @@ -12,11 +13,11 @@ from ..utils import ( class VesselIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vessel\.com/videos/(?P<id>[0-9a-zA-Z]+)' + _VALID_URL = r'https?://(?:www\.)?vessel\.com/(?:videos|embed)/(?P<id>[0-9a-zA-Z]+)' _API_URL_TEMPLATE = 'https://www.vessel.com/api/view/items/%s' _LOGIN_URL = 'https://www.vessel.com/api/account/login' _NETRC_MACHINE = 'vessel' - _TEST = { + _TESTS = [{ 'url': 'https://www.vessel.com/videos/HDN7G5UMs', 'md5': '455cdf8beb71c6dd797fd2f3818d05c4', 'info_dict': { @@ -28,7 +29,16 @@ class VesselIE(InfoExtractor): 'description': 'Did Nvidia pull out all the stops on the Titan X, or does its performance leave something to be desired?', 'timestamp': int, }, - } + }, { + 'url': 'https://www.vessel.com/embed/G4U7gUJ6a?w=615&h=346', + 'only_matching': True, + }] + + @staticmethod + def _extract_urls(webpage): + return [url for _, url in re.findall( + r'<iframe[^>]+src=(["\'])((?:https?:)?//(?:www\.)?vessel\.com/embed/[0-9a-zA-Z]+.*?)\1', + webpage)] @staticmethod def make_json_request(url, data): @@ -98,16 +108,24 @@ class VesselIE(InfoExtractor): formats = [] for f in video_asset.get('sources', []): - if f['name'] == 'hls-index': + location = f.get('location') + if not location: + continue + name = f.get('name') + if name == 'hls-index': formats.extend(self._extract_m3u8_formats( - f['location'], video_id, ext='mp4', m3u8_id='m3u8')) + location, video_id, ext='mp4', + entry_protocol='m3u8_native', m3u8_id='m3u8', fatal=False)) + elif name == 'dash-index': + formats.extend(self._extract_mpd_formats( + location, video_id, mpd_id='dash', fatal=False)) else: formats.append({ - 'format_id': f['name'], + 'format_id': name, 'tbr': f.get('bitrate'), 'height': f.get('height'), 'width': f.get('width'), - 'url': f['location'], + 'url': location, }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 147480f64..388b4debe 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -3,7 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_etree_fromstring +from ..compat import ( + compat_etree_fromstring, + compat_str, + compat_urlparse, +) from ..utils import ( ExtractorError, int_or_none, @@ -12,13 +16,22 @@ from ..utils import ( ) -class VevoIE(InfoExtractor): +class VevoBaseIE(InfoExtractor): + def _extract_json(self, webpage, video_id, item): + return self._parse_json( + self._search_regex( + r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>', + webpage, 'initial store'), + video_id)['default'][item] + + +class VevoIE(VevoBaseIE): ''' Accepts urls from vevo.com or in the format 'vevo:{id}' (currently used by MTVIE and MySpaceIE) ''' _VALID_URL = r'''(?x) - (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?| + (?:https?://www\.vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?| https?://cache\.vevo\.com/m/html/embed\.html\?video=| https?://videoplayer\.vevo\.com/embed/embedded\?videoId=| vevo:) @@ -30,11 +43,15 @@ class VevoIE(InfoExtractor): 'info_dict': { 'id': 'GB1101300280', 'ext': 'mp4', - 'title': 'Somebody to Die For', + 'title': 'Hurts - Somebody to Die For', + 'timestamp': 1372057200, 'upload_date': '20130624', 'uploader': 'Hurts', - 'timestamp': 1372057200, + 'track': 'Somebody to Die For', + 'artist': 'Hurts', + 'genre': 'Pop', }, + 'expected_warnings': ['Unable to download SMIL file'], }, { 'note': 'v3 SMIL format', 'url': 'http://www.vevo.com/watch/cassadee-pope/i-wish-i-could-break-your-heart/USUV71302923', @@ -42,23 +59,31 @@ class VevoIE(InfoExtractor): 'info_dict': { 'id': 'USUV71302923', 'ext': 'mp4', - 'title': 'I Wish I Could Break Your Heart', + 'title': 'Cassadee Pope - I Wish I Could Break Your Heart', + 'timestamp': 1392796919, 'upload_date': '20140219', 'uploader': 'Cassadee Pope', - 'timestamp': 1392796919, + 'track': 'I Wish I Could Break Your Heart', + 'artist': 'Cassadee Pope', + 'genre': 'Country', }, + 'expected_warnings': ['Unable to download SMIL file'], }, { 'note': 'Age-limited video', 'url': 'https://www.vevo.com/watch/justin-timberlake/tunnel-vision-explicit/USRV81300282', 'info_dict': { 'id': 'USRV81300282', 'ext': 'mp4', - 'title': 'Tunnel Vision (Explicit)', - 'upload_date': '20130703', + 'title': 'Justin Timberlake - Tunnel Vision (Explicit)', 'age_limit': 18, - 'uploader': 'Justin Timberlake', 'timestamp': 1372888800, + 'upload_date': '20130703', + 'uploader': 'Justin Timberlake', + 'track': 'Tunnel Vision (Explicit)', + 'artist': 'Justin Timberlake', + 'genre': 'Pop', }, + 'expected_warnings': ['Unable to download SMIL file'], }, { 'note': 'No video_info', 'url': 'http://www.vevo.com/watch/k-camp-1/Till-I-Die/USUV71503000', @@ -66,12 +91,36 @@ class VevoIE(InfoExtractor): 'info_dict': { 'id': 'USUV71503000', 'ext': 'mp4', - 'title': 'Till I Die', - 'upload_date': '20151207', + 'title': 'K Camp - Till I Die', 'age_limit': 18, - 'uploader': 'K Camp', 'timestamp': 1449468000, + 'upload_date': '20151207', + 'uploader': 'K Camp', + 'track': 'Till I Die', + 'artist': 'K Camp', + 'genre': 'Rap/Hip-Hop', + }, + }, { + 'note': 'Only available via webpage', + 'url': 'http://www.vevo.com/watch/GBUV71600656', + 'md5': '67e79210613865b66a47c33baa5e37fe', + 'info_dict': { + 'id': 'GBUV71600656', + 'ext': 'mp4', + 'title': 'ABC - Viva Love', + 'age_limit': 0, + 'timestamp': 1461830400, + 'upload_date': '20160428', + 'uploader': 'ABC', + 'track': 'Viva Love', + 'artist': 'ABC', + 'genre': 'Pop', }, + 'expected_warnings': ['Failed to download video versions info'], + }, { + # no genres available + 'url': 'http://www.vevo.com/watch/INS171400764', + 'only_matching': True, }] _SMIL_BASE_URL = 'http://smil.lvl3.vevo.com' _SOURCE_TYPES = { @@ -140,42 +189,41 @@ class VevoIE(InfoExtractor): errnote='Unable to retrieve oauth token') if 'THIS PAGE IS CURRENTLY UNAVAILABLE IN YOUR REGION' in webpage: - raise ExtractorError( - '%s said: This page is currently unavailable in your region.' % self.IE_NAME, expected=True) + self.raise_geo_restricted( + '%s said: This page is currently unavailable in your region' % self.IE_NAME) auth_info = self._parse_json(webpage, video_id) self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token'] - def _call_api(self, path, video_id, note, errnote, fatal=True): - return self._download_json(self._api_url_template % path, video_id, note, errnote) + def _call_api(self, path, *args, **kwargs): + return self._download_json(self._api_url_template % path, *args, **kwargs) def _real_extract(self, url): video_id = self._match_id(url) json_url = 'http://api.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id response = self._download_json( - json_url, video_id, 'Downloading video info', 'Unable to download info') + json_url, video_id, 'Downloading video info', + 'Unable to download info', fatal=False) or {} video_info = response.get('video') or {} - video_versions = video_info.get('videoVersions') + artist = None + featured_artist = None uploader = None - timestamp = None view_count = None formats = [] if not video_info: - if response.get('statusCode') != 909: + try: + self._initialize_api(video_id) + except ExtractorError: ytid = response.get('errorInfo', {}).get('ytid') if ytid: self.report_warning( 'Video is geoblocked, trying with the YouTube video %s' % ytid) return self.url_result(ytid, 'Youtube', ytid) - if 'statusMessage' in response: - raise ExtractorError('%s said: %s' % ( - self.IE_NAME, response['statusMessage']), expected=True) - raise ExtractorError('Unable to extract videos') + raise - self._initialize_api(video_id) video_info = self._call_api( 'video/%s' % video_id, video_id, 'Downloading api video info', 'Failed to download video info') @@ -183,12 +231,19 @@ class VevoIE(InfoExtractor): video_versions = self._call_api( 'video/%s/streams' % video_id, video_id, 'Downloading video versions info', - 'Failed to download video versions info') + 'Failed to download video versions info', + fatal=False) + + # Some videos are only available via webpage (e.g. + # https://github.com/rg3/youtube-dl/issues/9366) + if not video_versions: + webpage = self._download_webpage(url, video_id) + video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0] timestamp = parse_iso8601(video_info.get('releaseDate')) artists = video_info.get('artists') if artists: - uploader = artists[0]['name'] + artist = uploader = artists[0]['name'] view_count = int_or_none(video_info.get('views', {}).get('total')) for video_version in video_versions: @@ -241,7 +296,11 @@ class VevoIE(InfoExtractor): scale=1000) artists = video_info.get('mainArtists') if artists: - uploader = artists[0]['artistName'] + artist = uploader = artists[0]['artistName'] + + featured_artists = video_info.get('featuredArtists') + if featured_artists: + featured_artist = featured_artists[0]['artistName'] smil_parsed = False for video_version in video_info['videoVersions']: @@ -278,7 +337,15 @@ class VevoIE(InfoExtractor): smil_parsed = True self._sort_formats(formats) - title = video_info['title'] + track = video_info['title'] + if featured_artist: + artist = '%s ft. %s' % (artist, featured_artist) + title = '%s - %s' % (artist, track) if artist else track + + genres = video_info.get('genres') + genre = ( + genres[0] if genres and isinstance(genres, list) and + isinstance(genres[0], compat_str) else None) is_explicit = video_info.get('isExplicit') if is_explicit is True: @@ -300,4 +367,75 @@ class VevoIE(InfoExtractor): 'duration': duration, 'view_count': view_count, 'age_limit': age_limit, + 'track': track, + 'artist': uploader, + 'genre': genre, } + + +class VevoPlaylistIE(VevoBaseIE): + _VALID_URL = r'https?://www\.vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29', + 'info_dict': { + 'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29', + 'title': 'Best-Of: Birdman', + }, + 'playlist_count': 10, + }, { + 'url': 'http://www.vevo.com/watch/genre/rock', + 'info_dict': { + 'id': 'rock', + 'title': 'Rock', + }, + 'playlist_count': 20, + }, { + 'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0', + 'md5': '32dcdfddddf9ec6917fc88ca26d36282', + 'info_dict': { + 'id': 'USCMV1100073', + 'ext': 'mp4', + 'title': 'Birdman - Y.U. MAD', + 'timestamp': 1323417600, + 'upload_date': '20111209', + 'uploader': 'Birdman', + 'track': 'Y.U. MAD', + 'artist': 'Birdman', + 'genre': 'Rap/Hip-Hop', + }, + 'expected_warnings': ['Unable to download SMIL file'], + }, { + 'url': 'http://www.vevo.com/watch/genre/rock?index=0', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + playlist_id = mobj.group('id') + playlist_kind = mobj.group('kind') + + webpage = self._download_webpage(url, playlist_id) + + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + index = qs.get('index', [None])[0] + + if index: + video_id = self._search_regex( + r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>', + webpage, 'video id', default=None, group='id') + if video_id: + return self.url_result('vevo:%s' % video_id, VevoIE.ie_key()) + + playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind) + + playlist = (list(playlists.values())[0] + if playlist_kind == 'playlist' else playlists[playlist_id]) + + entries = [ + self.url_result('vevo:%s' % src, VevoIE.ie_key()) + for src in playlist['isrcs']] + + return self.playlist_result( + entries, playlist.get('playlistId') or playlist_id, + playlist.get('name'), playlist.get('description')) diff --git a/youtube_dl/extractor/vice.py b/youtube_dl/extractor/vice.py index 95daf4dfd..e2b2ce098 100644 --- a/youtube_dl/extractor/vice.py +++ b/youtube_dl/extractor/vice.py @@ -11,12 +11,14 @@ class ViceIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.vice.com/video/cowboy-capitalists-part-1', + 'md5': 'e9d77741f9e42ba583e683cd170660f7', 'info_dict': { 'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp', 'ext': 'flv', 'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', 'duration': 725.983, }, + 'add_ie': ['Ooyala'], }, { 'url': 'http://www.vice.com/video/how-to-hack-a-car', 'md5': '6fb2989a3fed069fb8eab3401fc2d3c9', @@ -29,6 +31,7 @@ class ViceIE(InfoExtractor): 'uploader': 'Motherboard', 'upload_date': '20140529', }, + 'add_ie': ['Youtube'], }, { 'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab', 'only_matching': True, diff --git a/youtube_dl/extractor/vidbit.py b/youtube_dl/extractor/vidbit.py new file mode 100644 index 000000000..e7ac5a842 --- /dev/null +++ b/youtube_dl/extractor/vidbit.py @@ -0,0 +1,84 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + int_or_none, + js_to_json, + remove_end, + unified_strdate, +) + + +class VidbitIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?vidbit\.co/(?:watch|embed)\?.*?\bv=(?P<id>[\da-zA-Z]+)' + _TESTS = [{ + 'url': 'http://www.vidbit.co/watch?v=jkL2yDOEq2', + 'md5': '1a34b7f14defe3b8fafca9796892924d', + 'info_dict': { + 'id': 'jkL2yDOEq2', + 'ext': 'mp4', + 'title': 'Intro to VidBit', + 'description': 'md5:5e0d6142eec00b766cbf114bfd3d16b7', + 'thumbnail': 're:https?://.*\.jpg$', + 'upload_date': '20160618', + 'view_count': int, + 'comment_count': int, + } + }, { + 'url': 'http://www.vidbit.co/embed?v=jkL2yDOEq2&auto=0&water=0', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage( + compat_urlparse.urljoin(url, '/watch?v=%s' % video_id), video_id) + + video_url, title = [None] * 2 + + config = self._parse_json(self._search_regex( + r'(?s)\.setup\(({.+?})\);', webpage, 'setup', default='{}'), + video_id, transform_source=js_to_json) + if config: + if config.get('file'): + video_url = compat_urlparse.urljoin(url, config['file']) + title = config.get('title') + + if not video_url: + video_url = compat_urlparse.urljoin(url, self._search_regex( + r'file\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', + webpage, 'video URL', group='url')) + + if not title: + title = remove_end( + self._html_search_regex( + (r'<h1>(.+?)</h1>', r'<title>(.+?)</title>'), + webpage, 'title', default=None) or self._og_search_title(webpage), + ' - VidBit') + + description = self._html_search_meta( + ('description', 'og:description', 'twitter:description'), + webpage, 'description') + + upload_date = unified_strdate(self._html_search_meta( + 'datePublished', webpage, 'upload date')) + + view_count = int_or_none(self._search_regex( + r'<strong>(\d+)</strong> views', + webpage, 'view count', fatal=False)) + comment_count = int_or_none(self._search_regex( + r'id=["\']cmt_num["\'][^>]*>\((\d+)\)', + webpage, 'comment count', fatal=False)) + + return { + 'id': video_id, + 'url': video_url, + 'title': title, + 'description': description, + 'thumbnail': self._og_search_thumbnail(webpage), + 'upload_date': upload_date, + 'view_count': view_count, + 'comment_count': comment_count, + } diff --git a/youtube_dl/extractor/vidio.py b/youtube_dl/extractor/vidio.py new file mode 100644 index 000000000..6898042de --- /dev/null +++ b/youtube_dl/extractor/vidio.py @@ -0,0 +1,73 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import int_or_none + + +class VidioIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)' + _TESTS = [{ + 'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015', + 'md5': 'cd2801394afc164e9775db6a140b91fe', + 'info_dict': { + 'id': '165683', + 'display_id': 'dj_ambred-booyah-live-2015', + 'ext': 'mp4', + 'title': 'DJ_AMBRED - Booyah (Live 2015)', + 'description': 'md5:27dc15f819b6a78a626490881adbadf8', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 149, + 'like_count': int, + }, + }, { + 'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id, display_id = mobj.group('id', 'display_id') + + webpage = self._download_webpage(url, display_id) + + title = self._og_search_title(webpage) + + m3u8_url, duration, thumbnail = [None] * 3 + + clips = self._parse_json( + self._html_search_regex( + r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1', + webpage, 'video data', default='[]', group='data'), + display_id, fatal=False) + if clips: + clip = clips[0] + m3u8_url = clip.get('sources', [{}])[0].get('file') + duration = clip.get('clip_duration') + thumbnail = clip.get('image') + + m3u8_url = m3u8_url or self._search_regex( + r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>.+?)\1', webpage, 'hls url') + formats = self._extract_m3u8_formats(m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native') + + duration = int_or_none(duration or self._search_regex( + r'data-video-duration=(["\'])(?P<duartion>\d+)\1', webpage, 'duration')) + thumbnail = thumbnail or self._og_search_thumbnail(webpage) + + like_count = int_or_none(self._search_regex( + (r'<span[^>]+data-comment-vote-count=["\'](\d+)', + r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'), + webpage, 'like count', fatal=False)) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'description': self._og_search_description(webpage), + 'thumbnail': thumbnail, + 'duration': duration, + 'like_count': like_count, + 'formats': formats, + } diff --git a/youtube_dl/extractor/snagfilms.py b/youtube_dl/extractor/viewlift.py index 6977afb27..19500eba8 100644 --- a/youtube_dl/extractor/snagfilms.py +++ b/youtube_dl/extractor/viewlift.py @@ -13,8 +13,12 @@ from ..utils import ( ) -class SnagFilmsEmbedIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?:www|embed)\.)?snagfilms\.com/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})' +class ViewLiftBaseIE(InfoExtractor): + _DOMAINS_REGEX = '(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|monumentalsportsnetwork|vayafilm)\.com|kesari\.tv' + + +class ViewLiftEmbedIE(ViewLiftBaseIE): + _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})' % ViewLiftBaseIE._DOMAINS_REGEX _TESTS = [{ 'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500', 'md5': '2924e9215c6eff7a55ed35b72276bd93', @@ -40,7 +44,7 @@ class SnagFilmsEmbedIE(InfoExtractor): @staticmethod def _extract_url(webpage): mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:embed\.)?snagfilms\.com/embed/player.+?)\1', + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:embed\.)?(?:%s)/embed/player.+?)\1' % ViewLiftBaseIE._DOMAINS_REGEX, webpage) if mobj: return mobj.group('url') @@ -55,6 +59,7 @@ class SnagFilmsEmbedIE(InfoExtractor): 'Film %s is not playable in your area.' % video_id, expected=True) formats = [] + has_bitrate = False for source in self._parse_json(js_to_json(self._search_regex( r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id): file_ = source.get('file') @@ -63,22 +68,25 @@ class SnagFilmsEmbedIE(InfoExtractor): type_ = source.get('type') ext = determine_ext(file_) format_id = source.get('label') or ext - if all(v == 'm3u8' for v in (type_, ext)): + if all(v == 'm3u8' or v == 'hls' for v in (type_, ext)): formats.extend(self._extract_m3u8_formats( file_, video_id, 'mp4', m3u8_id='hls')) else: bitrate = int_or_none(self._search_regex( [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext], file_, 'bitrate', default=None)) + if not has_bitrate and bitrate: + has_bitrate = True height = int_or_none(self._search_regex( r'^(\d+)[pP]$', format_id, 'height', default=None)) formats.append({ 'url': file_, - 'format_id': format_id, + 'format_id': 'http-%s%s' % (format_id, ('-%dk' % bitrate if bitrate else '')), 'tbr': bitrate, 'height': height, }) - self._sort_formats(formats) + field_preference = None if has_bitrate else ('height', 'tbr', 'format_id') + self._sort_formats(formats, field_preference) title = self._search_regex( [r"title\s*:\s*'([^']+)'", r'<title>([^<]+)</title>'], @@ -91,8 +99,8 @@ class SnagFilmsEmbedIE(InfoExtractor): } -class SnagFilmsIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?snagfilms\.com/(?:films/title|show)/(?P<id>[^?#]+)' +class ViewLiftIE(ViewLiftBaseIE): + _VALID_URL = r'https?://(?:www\.)?(?P<domain>%s)/(?:films/title|show|(?:news/)?videos?)/(?P<id>[^?#]+)' % ViewLiftBaseIE._DOMAINS_REGEX _TESTS = [{ 'url': 'http://www.snagfilms.com/films/title/lost_for_life', 'md5': '19844f897b35af219773fd63bdec2942', @@ -127,10 +135,20 @@ class SnagFilmsIE(InfoExtractor): # Film is not available. 'url': 'http://www.snagfilms.com/show/augie_alone/flirting', 'only_matching': True, + }, { + 'url': 'http://www.winnersview.com/videos/the-good-son', + 'only_matching': True, + }, { + 'url': 'http://www.kesari.tv/news/video/1461919076414', + 'only_matching': True, + }, { + # Was once Kaltura embed + 'url': 'https://www.monumentalsportsnetwork.com/videos/john-carlson-postgame-2-25-15', + 'only_matching': True, }] def _real_extract(self, url): - display_id = self._match_id(url) + domain, display_id = re.match(self._VALID_URL, url).groups() webpage = self._download_webpage(url, display_id) @@ -170,7 +188,7 @@ class SnagFilmsIE(InfoExtractor): return { '_type': 'url_transparent', - 'url': 'http://embed.snagfilms.com/embed/player?filmId=%s' % film_id, + 'url': 'http://%s/embed/player?filmId=%s' % (domain, film_id), 'id': film_id, 'display_id': display_id, 'title': title, @@ -178,4 +196,5 @@ class SnagFilmsIE(InfoExtractor): 'thumbnail': thumbnail, 'duration': duration, 'categories': categories, + 'ie_key': 'ViewLiftEmbed', } diff --git a/youtube_dl/extractor/viewster.py b/youtube_dl/extractor/viewster.py index fe94a4793..a93196a07 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -1,10 +1,11 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..compat import ( compat_HTTPError, - compat_urllib_parse, compat_urllib_parse_unquote, ) from ..utils import ( @@ -14,6 +15,7 @@ from ..utils import ( parse_iso8601, sanitized_Request, HEADRequest, + url_basename, ) @@ -75,11 +77,11 @@ class ViewsterIE(InfoExtractor): _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' - def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True): + def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True, query={}): request = sanitized_Request(url) request.add_header('Accept', self._ACCEPT_HEADER) request.add_header('Auth-token', self._AUTH_TOKEN) - return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal) + return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal, query=query) def _real_extract(self, url): video_id = self._match_id(url) @@ -114,43 +116,85 @@ class ViewsterIE(InfoExtractor): return self.playlist_result(entries, video_id, title, description) formats = [] - for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'): - media = self._download_json( - 'https://public-api.viewster.com/movies/%s/video?mediaType=%s' - % (entry_id, compat_urllib_parse.quote(media_type)), - video_id, 'Downloading %s JSON' % media_type, fatal=False) - if not media: - continue - video_url = media.get('Uri') - if not video_url: - continue - ext = determine_ext(video_url) - if ext == 'f4m': - video_url += '&' if '?' in video_url else '?' - video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1' - formats.extend(self._extract_f4m_formats( - video_url, video_id, f4m_id='hds')) - elif ext == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - video_url, video_id, 'mp4', m3u8_id='hls', - fatal=False) # m3u8 sometimes fail - if m3u8_formats: - formats.extend(m3u8_formats) - else: - format_id = media.get('Bitrate') - f = { - 'url': video_url, - 'format_id': 'mp4-%s' % format_id, - 'height': int_or_none(media.get('Height')), - 'width': int_or_none(media.get('Width')), - 'preference': 1, - } - if format_id and not f['height']: - f['height'] = int_or_none(self._search_regex( - r'^(\d+)[pP]$', format_id, 'height', default=None)) - formats.append(f) - - if not formats and not info.get('LanguageSets') and not info.get('VODSettings'): + for language_set in info.get('LanguageSets', []): + manifest_url = None + m3u8_formats = [] + audio = language_set.get('Audio') or '' + subtitle = language_set.get('Subtitle') or '' + base_format_id = audio + if subtitle: + base_format_id += '-%s' % subtitle + + def concat(suffix, sep='-'): + return (base_format_id + '%s%s' % (sep, suffix)) if base_format_id else suffix + + for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'): + media = self._download_json( + 'https://public-api.viewster.com/movies/%s/video' % entry_id, + video_id, 'Downloading %s JSON' % concat(media_type, ' '), fatal=False, query={ + 'mediaType': media_type, + 'language': audio, + 'subtitle': subtitle, + }) + if not media: + continue + video_url = media.get('Uri') + if not video_url: + continue + ext = determine_ext(video_url) + if ext == 'f4m': + manifest_url = video_url + video_url += '&' if '?' in video_url else '?' + video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1' + formats.extend(self._extract_f4m_formats( + video_url, video_id, f4m_id=concat('hds'))) + elif ext == 'm3u8': + manifest_url = video_url + m3u8_formats = self._extract_m3u8_formats( + video_url, video_id, 'mp4', m3u8_id=concat('hls'), + fatal=False) # m3u8 sometimes fail + if m3u8_formats: + formats.extend(m3u8_formats) + else: + qualities_basename = self._search_regex( + '/([^/]+)\.csmil/', + manifest_url, 'qualities basename', default=None) + if not qualities_basename: + continue + QUALITIES_RE = r'((,\d+k)+,?)' + qualities = self._search_regex( + QUALITIES_RE, qualities_basename, + 'qualities', default=None) + if not qualities: + continue + qualities = list(map(lambda q: int(q[:-1]), qualities.strip(',').split(','))) + qualities.sort() + http_template = re.sub(QUALITIES_RE, r'%dk', qualities_basename) + http_url_basename = url_basename(video_url) + if m3u8_formats: + self._sort_formats(m3u8_formats) + m3u8_formats = list(filter( + lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple', + m3u8_formats)) + if len(qualities) == len(m3u8_formats): + for q, m3u8_format in zip(qualities, m3u8_formats): + f = m3u8_format.copy() + f.update({ + 'url': video_url.replace(http_url_basename, http_template % q), + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) + else: + for q in qualities: + formats.append({ + 'url': video_url.replace(http_url_basename, http_template % q), + 'ext': 'mp4', + 'format_id': 'http-%d' % q, + 'tbr': q, + }) + + if not formats and not info.get('VODSettings'): self.raise_geo_restricted() self._sort_formats(formats) diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py index e04b814c8..efa15e0b6 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -101,10 +101,13 @@ class VikiBaseIE(InfoExtractor): self.report_warning('Unable to get session token, login has probably failed') @staticmethod - def dict_selection(dict_obj, preferred_key): + def dict_selection(dict_obj, preferred_key, allow_fallback=True): if preferred_key in dict_obj: return dict_obj.get(preferred_key) + if not allow_fallback: + return + filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()])) return filtered_dict[0] if filtered_dict else None @@ -127,7 +130,7 @@ class VikiIE(VikiBaseIE): }, { # clip 'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference', - 'md5': '86c0b5dbd4d83a6611a79987cc7a1989', + 'md5': 'feea2b1d7b3957f70886e6dfd8b8be84', 'info_dict': { 'id': '1067139v', 'ext': 'mp4', @@ -156,17 +159,18 @@ class VikiIE(VikiBaseIE): 'params': { # m3u8 download 'skip_download': True, - } + }, + 'skip': 'Blocked in the US', }, { # episode 'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1', - 'md5': '190f3ef426005ba3a080a63325955bc3', + 'md5': '1f54697dabc8f13f31bf06bb2e4de6db', 'info_dict': { 'id': '44699v', 'ext': 'mp4', 'title': 'Boys Over Flowers - Episode 1', - 'description': 'md5:52617e4f729c7d03bfd4bcbbb6e946f2', - 'duration': 4155, + 'description': 'md5:b89cf50038b480b88b5b3c93589a9076', + 'duration': 4204, 'timestamp': 1270496524, 'upload_date': '20100405', 'uploader': 'group8', @@ -196,7 +200,7 @@ class VikiIE(VikiBaseIE): }, { # non-English description 'url': 'http://www.viki.com/videos/158036v-love-in-magic', - 'md5': '1713ae35df5a521b31f6dc40730e7c9c', + 'md5': '013dc282714e22acf9447cad14ff1208', 'info_dict': { 'id': '158036v', 'ext': 'mp4', @@ -217,7 +221,7 @@ class VikiIE(VikiBaseIE): self._check_errors(video) - title = self.dict_selection(video.get('titles', {}), 'en') + title = self.dict_selection(video.get('titles', {}), 'en', allow_fallback=False) if not title: title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id container_titles = video.get('container', {}).get('titles', {}) @@ -302,7 +306,7 @@ class VikiChannelIE(VikiBaseIE): 'title': 'Boys Over Flowers', 'description': 'md5:ecd3cff47967fe193cff37c0bec52790', }, - 'playlist_count': 70, + 'playlist_mincount': 71, }, { 'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete', 'info_dict': { diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 59f9cb1ae..d9c9852d4 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -8,6 +8,7 @@ import itertools from .common import InfoExtractor from ..compat import ( compat_HTTPError, + compat_str, compat_urlparse, ) from ..utils import ( @@ -15,6 +16,7 @@ from ..utils import ( ExtractorError, InAdvancePagedList, int_or_none, + NO_DEFAULT, RegexNotFoundError, sanitized_Request, smuggle_url, @@ -24,6 +26,7 @@ from ..utils import ( urlencode_postdata, unescapeHTML, parse_filesize, + try_get, ) @@ -54,6 +57,26 @@ class VimeoBaseInfoExtractor(InfoExtractor): self._set_vimeo_cookie('vuid', vuid) self._download_webpage(login_request, None, False, 'Wrong login info') + def _verify_video_password(self, url, video_id, webpage): + password = self._downloader.params.get('videopassword') + if password is None: + raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + token, vuid = self._extract_xsrft_and_vuid(webpage) + data = urlencode_postdata({ + 'password': password, + 'token': token, + }) + if url.startswith('http://'): + # vimeo only supports https now, but the user can give an http url + url = url.replace('http://', 'https://') + password_request = sanitized_Request(url + '/password', data) + password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + password_request.add_header('Referer', url) + self._set_vimeo_cookie('vuid', vuid) + return self._download_webpage( + password_request, video_id, + 'Verifying the password', 'Wrong password') + def _extract_xsrft_and_vuid(self, webpage): xsrft = self._search_regex( r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)', @@ -66,6 +89,69 @@ class VimeoBaseInfoExtractor(InfoExtractor): def _set_vimeo_cookie(self, name, value): self._set_cookie('vimeo.com', name, value) + def _vimeo_sort_formats(self, formats): + # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps + # at the same time without actual units specified. This lead to wrong sorting. + self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'format_id')) + + def _parse_config(self, config, video_id): + # Extract title + video_title = config['video']['title'] + + # Extract uploader, uploader_url and uploader_id + video_uploader = config['video'].get('owner', {}).get('name') + video_uploader_url = config['video'].get('owner', {}).get('url') + video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None + + # Extract video thumbnail + video_thumbnail = config['video'].get('thumbnail') + if video_thumbnail is None: + video_thumbs = config['video'].get('thumbs') + if video_thumbs and isinstance(video_thumbs, dict): + _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1] + + # Extract video duration + video_duration = int_or_none(config['video'].get('duration')) + + formats = [] + config_files = config['video'].get('files') or config['request'].get('files', {}) + for f in config_files.get('progressive', []): + video_url = f.get('url') + if not video_url: + continue + formats.append({ + 'url': video_url, + 'format_id': 'http-%s' % f.get('quality'), + 'width': int_or_none(f.get('width')), + 'height': int_or_none(f.get('height')), + 'fps': int_or_none(f.get('fps')), + 'tbr': int_or_none(f.get('bitrate')), + }) + m3u8_url = config_files.get('hls', {}).get('url') + if m3u8_url: + formats.extend(self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + + subtitles = {} + text_tracks = config['request'].get('text_tracks') + if text_tracks: + for tt in text_tracks: + subtitles[tt['lang']] = [{ + 'ext': 'vtt', + 'url': 'https://vimeo.com' + tt['url'], + }] + + return { + 'title': video_title, + 'uploader': video_uploader, + 'uploader_id': video_uploader_id, + 'uploader_url': video_uploader_url, + 'thumbnail': video_thumbnail, + 'duration': video_duration, + 'formats': formats, + 'subtitles': subtitles, + } + class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" @@ -81,7 +167,7 @@ class VimeoIE(VimeoBaseInfoExtractor): \. )? vimeo(?P<pro>pro)?\.com/ - (?!channels/[^/?#]+/?(?:$|[?#])|[^/]+/review/|(?:album|ondemand)/) + (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) (?:.*?/)? (?: (?: @@ -153,7 +239,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user18948128', 'uploader': 'Jaime Marquínez Ferrándiz', 'duration': 10, - 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people\u2026', + 'description': 'This is "youtube-dl password protected test video" by on Vimeo, the home for high quality videos and the people who love them.', }, 'params': { 'videopassword': 'youtube-dl', @@ -162,8 +248,6 @@ class VimeoIE(VimeoBaseInfoExtractor): { 'url': 'http://vimeo.com/channels/keypeele/75629013', 'md5': '2f86a05afe9d7abc0b9126d229bbe15d', - 'note': 'Video is freely available via original URL ' - 'and protected with password when accessed via http://vimeo.com/75629013', 'info_dict': { 'id': '75629013', 'ext': 'mp4', @@ -207,7 +291,7 @@ class VimeoIE(VimeoBaseInfoExtractor): { # contains original format 'url': 'https://vimeo.com/33951933', - 'md5': '53c688fa95a55bf4b7293d37a89c5c53', + 'md5': '2d9f5475e0537f013d0073e812ab89e6', 'info_dict': { 'id': '33951933', 'ext': 'mp4', @@ -220,6 +304,29 @@ class VimeoIE(VimeoBaseInfoExtractor): }, }, { + # only available via https://vimeo.com/channels/tributes/6213729 and + # not via https://vimeo.com/6213729 + 'url': 'https://vimeo.com/channels/tributes/6213729', + 'info_dict': { + 'id': '6213729', + 'ext': 'mp4', + 'title': 'Vimeo Tribute: The Shining', + 'uploader': 'Casey Donahue', + 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue', + 'uploader_id': 'caseydonahue', + 'upload_date': '20090821', + 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6', + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Unable to download JSON metadata'], + }, + { + 'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741', + 'only_matching': True, + }, + { 'url': 'https://vimeo.com/109815029', 'note': 'Video not completely processed, "failed" seed status', 'only_matching': True, @@ -229,6 +336,10 @@ class VimeoIE(VimeoBaseInfoExtractor): 'only_matching': True, }, { + 'url': 'https://vimeo.com/album/2632481/video/79010983', + 'only_matching': True, + }, + { # source file returns 403: Forbidden 'url': 'https://vimeo.com/7809605', 'only_matching': True, @@ -254,26 +365,6 @@ class VimeoIE(VimeoBaseInfoExtractor): if mobj: return mobj.group(1) - def _verify_video_password(self, url, video_id, webpage): - password = self._downloader.params.get('videopassword') - if password is None: - raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata({ - 'password': password, - 'token': token, - }) - if url.startswith('http://'): - # vimeo only supports https now, but the user can give an http url - url = url.replace('http://', 'https://') - password_request = sanitized_Request(url + '/password', data) - password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Referer', url) - self._set_vimeo_cookie('vuid', vuid) - return self._download_webpage( - password_request, video_id, - 'Verifying the password', 'Wrong password') - def _verify_player_video_password(self, url, video_id): password = self._downloader.params.get('videopassword') if password is None: @@ -304,7 +395,7 @@ class VimeoIE(VimeoBaseInfoExtractor): orig_url = url if mobj.group('pro') or mobj.group('player'): url = 'https://player.vimeo.com/video/' + video_id - else: + elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information @@ -382,28 +473,24 @@ class VimeoIE(VimeoBaseInfoExtractor): if config.get('view') == 4: config = self._verify_player_video_password(url, video_id) - if '>You rented this title.<' in webpage: + def is_rented(): + if '>You rented this title.<' in webpage: + return True + if config.get('user', {}).get('purchased'): + return True + label = try_get( + config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str) + if label and label.startswith('You rented this'): + return True + return False + + if is_rented(): feature_id = config.get('video', {}).get('vod', {}).get('feature_id') if feature_id and not data.get('force_feature_id', False): return self.url_result(smuggle_url( 'https://player.vimeo.com/player/%s' % feature_id, {'force_feature_id': True}), 'Vimeo') - # Extract title - video_title = config['video']['title'] - - # Extract uploader, uploader_url and uploader_id - video_uploader = config['video'].get('owner', {}).get('name') - video_uploader_url = config['video'].get('owner', {}).get('url') - video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None - - # Extract video thumbnail - video_thumbnail = config['video'].get('thumbnail') - if video_thumbnail is None: - video_thumbs = config['video'].get('thumbs') - if video_thumbs and isinstance(video_thumbs, dict): - _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1] - # Extract video description video_description = self._html_search_regex( @@ -423,9 +510,6 @@ class VimeoIE(VimeoBaseInfoExtractor): if not video_description and not mobj.group('player'): self._downloader.report_warning('Cannot find video description') - # Extract video duration - video_duration = int_or_none(config['video'].get('duration')) - # Extract upload date video_upload_date = None mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage) @@ -463,53 +547,22 @@ class VimeoIE(VimeoBaseInfoExtractor): 'format_id': source_name, 'preference': 1, }) - config_files = config['video'].get('files') or config['request'].get('files', {}) - for f in config_files.get('progressive', []): - video_url = f.get('url') - if not video_url: - continue - formats.append({ - 'url': video_url, - 'format_id': 'http-%s' % f.get('quality'), - 'width': int_or_none(f.get('width')), - 'height': int_or_none(f.get('height')), - 'fps': int_or_none(f.get('fps')), - 'tbr': int_or_none(f.get('bitrate')), - }) - m3u8_url = config_files.get('hls', {}).get('url') - if m3u8_url: - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) - # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps - # at the same time without actual units specified. This lead to wrong sorting. - self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'format_id')) - - subtitles = {} - text_tracks = config['request'].get('text_tracks') - if text_tracks: - for tt in text_tracks: - subtitles[tt['lang']] = [{ - 'ext': 'vtt', - 'url': 'https://vimeo.com' + tt['url'], - }] - return { + info_dict = self._parse_config(config, video_id) + formats.extend(info_dict['formats']) + self._vimeo_sort_formats(formats) + info_dict.update({ 'id': video_id, - 'uploader': video_uploader, - 'uploader_url': video_uploader_url, - 'uploader_id': video_uploader_id, + 'formats': formats, 'upload_date': video_upload_date, - 'title': video_title, - 'thumbnail': video_thumbnail, 'description': video_description, - 'duration': video_duration, - 'formats': formats, 'webpage_url': url, 'view_count': view_count, 'like_count': like_count, 'comment_count': comment_count, - 'subtitles': subtitles, - } + }) + + return info_dict class VimeoOndemandIE(VimeoBaseInfoExtractor): @@ -603,8 +656,21 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): webpage = self._login_list_password(page_url, list_id, webpage) yield self._extract_list_title(webpage) - for video_id in re.findall(r'id="clip_(\d+?)"', webpage): - yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') + # Try extracting href first since not all videos are available via + # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729) + clips = re.findall( + r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)', webpage) + if clips: + for video_id, video_url in clips: + yield self.url_result( + compat_urlparse.urljoin(base_url, video_url), + VimeoIE.ie_key(), video_id=video_id) + # More relaxed fallback + else: + for video_id in re.findall(r'id=["\']clip_(\d+)', webpage): + yield self.url_result( + 'https://vimeo.com/%s' % video_id, + VimeoIE.ie_key(), video_id=video_id) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break @@ -641,7 +707,7 @@ class VimeoUserIE(VimeoChannelIE): class VimeoAlbumIE(VimeoChannelIE): IE_NAME = 'vimeo:album' - _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)' + _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))' _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>' _TESTS = [{ 'url': 'https://vimeo.com/album/2632481', @@ -661,6 +727,13 @@ class VimeoAlbumIE(VimeoChannelIE): 'params': { 'videopassword': 'youtube-dl', } + }, { + 'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail', + 'only_matching': True, + }, { + # TODO: respect page number + 'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail', + 'only_matching': True, }] def _page_url(self, base_url, pagenum): @@ -692,7 +765,7 @@ class VimeoGroupsIE(VimeoAlbumIE): return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name) -class VimeoReviewIE(InfoExtractor): +class VimeoReviewIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:review' IE_DESC = 'Review pages on vimeo' _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)' @@ -704,6 +777,7 @@ class VimeoReviewIE(InfoExtractor): 'ext': 'mp4', 'title': "DICK HARDWICK 'Comedian'", 'uploader': 'Richard Hardwick', + 'uploader_id': 'user21297594', } }, { 'note': 'video player needs Referer', @@ -716,14 +790,45 @@ class VimeoReviewIE(InfoExtractor): 'uploader': 'DevWeek Events', 'duration': 2773, 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader_id': 'user22258446', } + }, { + 'note': 'Password protected', + 'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde', + 'info_dict': { + 'id': '138823582', + 'ext': 'mp4', + 'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1', + 'uploader': 'TMB', + 'uploader_id': 'user37284429', + }, + 'params': { + 'videopassword': 'holygrail', + }, }] + def _real_initialize(self): + self._login() + + def _get_config_url(self, webpage_url, video_id, video_password_verified=False): + webpage = self._download_webpage(webpage_url, video_id) + config_url = self._html_search_regex( + r'data-config-url="([^"]+)"', webpage, 'config URL', + default=NO_DEFAULT if video_password_verified else None) + if config_url is None: + self._verify_video_password(webpage_url, video_id, webpage) + config_url = self._get_config_url( + webpage_url, video_id, video_password_verified=True) + return config_url + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - player_url = 'https://player.vimeo.com/player/' + video_id - return self.url_result(player_url, 'Vimeo', video_id) + video_id = self._match_id(url) + config_url = self._get_config_url(url, video_id) + config = self._download_json(config_url, video_id) + info_dict = self._parse_config(config, video_id) + self._vimeo_sort_formats(info_dict['formats']) + info_dict['id'] = video_id + return info_dict class VimeoWatchLaterIE(VimeoChannelIE): diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py index a6a6cc479..0183f052a 100644 --- a/youtube_dl/extractor/vine.py +++ b/youtube_dl/extractor/vine.py @@ -24,6 +24,7 @@ class VineIE(InfoExtractor): 'upload_date': '20130519', 'uploader': 'Jack Dorsey', 'uploader_id': '76', + 'view_count': int, 'like_count': int, 'comment_count': int, 'repost_count': int, @@ -39,6 +40,7 @@ class VineIE(InfoExtractor): 'upload_date': '20140815', 'uploader': 'Mars Ruiz', 'uploader_id': '1102363502380728320', + 'view_count': int, 'like_count': int, 'comment_count': int, 'repost_count': int, @@ -54,6 +56,7 @@ class VineIE(InfoExtractor): 'upload_date': '20130430', 'uploader': 'Z3k3', 'uploader_id': '936470460173008896', + 'view_count': int, 'like_count': int, 'comment_count': int, 'repost_count': int, @@ -71,6 +74,7 @@ class VineIE(InfoExtractor): 'upload_date': '20150705', 'uploader': 'Pimry_zaa', 'uploader_id': '1135760698325307392', + 'view_count': int, 'like_count': int, 'comment_count': int, 'repost_count': int, @@ -86,10 +90,12 @@ class VineIE(InfoExtractor): data = self._parse_json( self._search_regex( - r'window\.POST_DATA\s*=\s*{\s*%s\s*:\s*({.+?})\s*};\s*</script>' % video_id, + r'window\.POST_DATA\s*=\s*({.+?});\s*</script>', webpage, 'vine data'), video_id) + data = data[list(data.keys())[0]] + formats = [{ 'format_id': '%(format)s-%(rate)s' % f, 'vcodec': f.get('format'), @@ -109,6 +115,7 @@ class VineIE(InfoExtractor): 'upload_date': unified_strdate(data.get('created')), 'uploader': username, 'uploader_id': data.get('userIdStr'), + 'view_count': int_or_none(data.get('loops', {}).get('count')), 'like_count': int_or_none(data.get('likes', {}).get('count')), 'comment_count': int_or_none(data.get('comments', {}).get('count')), 'repost_count': int_or_none(data.get('reposts', {}).get('count')), diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 67220f1b7..758d9c86b 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -3,6 +3,7 @@ from __future__ import unicode_literals import re import json +import sys from .common import InfoExtractor from ..compat import compat_str @@ -10,7 +11,6 @@ from ..utils import ( ExtractorError, int_or_none, orderedSet, - sanitized_Request, str_to_int, unescapeHTML, unified_strdate, @@ -26,12 +26,16 @@ class VKIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// (?: - (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)| (?: - (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video| - (?:www\.)?biqle\.ru/watch/ + (?:(?:m|new)\.)?vk\.com/video_| + (?:www\.)?daxab.com/ ) - (?P<videoid>[^s].*?)(?:\?(?:.*\blist=(?P<list_id>[\da-f]+))?|%2F|$) + ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)| + (?: + (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video| + (?:www\.)?daxab.com/embed/ + ) + (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))? ) ''' _NETRC_MACHINE = 'vk' @@ -75,7 +79,8 @@ class VKIE(InfoExtractor): 'duration': 101, 'upload_date': '20120730', 'view_count': int, - } + }, + 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -142,7 +147,7 @@ class VKIE(InfoExtractor): 'id': 'V3K4mi0SYkc', 'ext': 'webm', 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", - 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', + 'description': 'md5:d9903938abdc74c738af77f527ca0596', 'duration': 178, 'upload_date': '20130116', 'uploader': "Children's Joy Foundation", @@ -174,13 +179,12 @@ class VKIE(InfoExtractor): 'only_matching': True, }, { - # vk wrapper - 'url': 'http://www.biqle.ru/watch/847655_160197695', + # pladform embed + 'url': 'https://vk.com/video-76116461_171554880', 'only_matching': True, }, { - # pladform embed - 'url': 'https://vk.com/video-76116461_171554880', + 'url': 'http://new.vk.com/video205387401_165548505', 'only_matching': True, } ] @@ -190,7 +194,7 @@ class VKIE(InfoExtractor): if username is None: return - login_page = self._download_webpage( + login_page, url_handle = self._download_webpage_handle( 'https://vk.com', None, 'Downloading login page') login_form = self._hidden_inputs(login_page) @@ -200,11 +204,26 @@ class VKIE(InfoExtractor): 'pass': password.encode('cp1251'), }) - request = sanitized_Request( - 'https://login.vk.com/?act=login', - urlencode_postdata(login_form)) + # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header + # and expects the first one to be set rather than second (see + # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201). + # As of RFC6265 the newer one cookie should be set into cookie store + # what actually happens. + # We will workaround this VK issue by resetting the remixlhk cookie to + # the first one manually. + cookies = url_handle.headers.get('Set-Cookie') + if sys.version_info[0] >= 3: + cookies = cookies.encode('iso-8859-1') + cookies = cookies.decode('utf-8') + remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies) + if remixlhk: + value, domain = remixlhk.groups() + self._set_cookie(domain, 'remixlhk', value) + login_page = self._download_webpage( - request, None, note='Logging in as %s' % username) + 'https://login.vk.com/?act=login', None, + note='Logging in as %s' % username, + data=urlencode_postdata(login_form)) if re.search(r'onLoginFailed', login_page): raise ExtractorError( @@ -217,20 +236,21 @@ class VKIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') - if not video_id: + if video_id: + info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id + # Some videos (removed?) can only be downloaded with list id specified + list_id = mobj.group('list_id') + if list_id: + info_url += '&list=%s' % list_id + else: + info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query') video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id')) - info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id - - # Some videos (removed?) can only be downloaded with list id specified - list_id = mobj.group('list_id') - if list_id: - info_url += '&list=%s' % list_id - info_page = self._download_webpage(info_url, video_id) error_message = self._html_search_regex( - r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>', + [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>', + r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'], info_page, 'error message', default=None) if error_message: raise ExtractorError(error_message, expected=True) @@ -305,17 +325,17 @@ class VKIE(InfoExtractor): view_count = None views = self._html_search_regex( r'"mv_views_count_number"[^>]*>(.+?\bviews?)<', - info_page, 'view count', fatal=False) + info_page, 'view count', default=None) if views: view_count = str_to_int(self._search_regex( r'([\d,.]+)', views, 'view count', fatal=False)) formats = [] for k, v in data.items(): - if not k.startswith('url') and k != 'extra_data' or not v: + if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v: continue height = int_or_none(self._search_regex( - r'^url(\d+)', k, 'height', default=None)) + r'^(?:url|cache)(\d+)', k, 'height', default=None)) formats.append({ 'format_id': k, 'url': v, @@ -338,7 +358,7 @@ class VKIE(InfoExtractor): class VKUserVideosIE(InfoExtractor): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ 'url': 'http://vk.com/videos205387401', @@ -353,6 +373,12 @@ class VKUserVideosIE(InfoExtractor): }, { 'url': 'http://vk.com/videos-97664626?section=all', 'only_matching': True, + }, { + 'url': 'http://m.vk.com/videos205387401', + 'only_matching': True, + }, { + 'url': 'http://new.vk.com/videos205387401', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index baf39bb2c..8d671cca7 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -1,11 +1,15 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..utils import ( dict_get, + ExtractorError, float_or_none, int_or_none, + remove_start, ) from ..compat import compat_urllib_parse_urlencode @@ -19,7 +23,7 @@ class VLiveIE(InfoExtractor): 'info_dict': { 'id': '1326', 'ext': 'mp4', - 'title': "[V] Girl's Day's Broadcast", + 'title': "[V LIVE] Girl's Day's Broadcast", 'creator': "Girl's Day", 'view_count': int, }, @@ -31,16 +35,62 @@ class VLiveIE(InfoExtractor): webpage = self._download_webpage( 'http://www.vlive.tv/video/%s' % video_id, video_id) - long_video_id = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"', - webpage, 'long video id') + video_params = self._search_regex( + r'\bvlive\.video\.init\(([^)]+)\)', + webpage, 'video params') + status, _, _, live_params, long_video_id, key = re.split( + r'"\s*,\s*"', video_params)[2:8] + status = remove_start(status, 'PRODUCT_') + + if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR': + live_params = self._parse_json('"%s"' % live_params, video_id) + live_params = self._parse_json(live_params, video_id) + return self._live(video_id, webpage, live_params) + elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO': + if long_video_id and key: + return self._replay(video_id, webpage, long_video_id, key) + else: + status = 'COMING_SOON' - key = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"', - webpage, 'key') + if status == 'LIVE_END': + raise ExtractorError('Uploading for replay. Please wait...', + expected=True) + elif status == 'COMING_SOON': + raise ExtractorError('Coming soon!', expected=True) + elif status == 'CANCELED': + raise ExtractorError('We are sorry, ' + 'but the live broadcast has been canceled.', + expected=True) + else: + raise ExtractorError('Unknown status %s' % status) + def _get_common_fields(self, webpage): title = self._og_search_title(webpage) + creator = self._html_search_regex( + r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)', + webpage, 'creator', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) + return { + 'title': title, + 'creator': creator, + 'thumbnail': thumbnail, + } + + def _live(self, video_id, webpage, live_params): + formats = [] + for vid in live_params.get('resolutions', []): + formats.extend(self._extract_m3u8_formats( + vid['cdnUrl'], video_id, 'mp4', + m3u8_id=vid.get('name'), + fatal=False, live=True)) + self._sort_formats(formats) + + return dict(self._get_common_fields(webpage), + id=video_id, + formats=formats, + is_live=True) + def _replay(self, video_id, webpage, long_video_id, key): playinfo = self._download_json( 'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s' % compat_urllib_parse_urlencode({ @@ -62,11 +112,6 @@ class VLiveIE(InfoExtractor): } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')] self._sort_formats(formats) - thumbnail = self._og_search_thumbnail(webpage) - creator = self._html_search_regex( - r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)', - webpage, 'creator', fatal=False) - view_count = int_or_none(playinfo.get('meta', {}).get('count')) subtitles = {} @@ -77,12 +122,8 @@ class VLiveIE(InfoExtractor): 'ext': 'vtt', 'url': caption['source']}] - return { - 'id': video_id, - 'title': title, - 'creator': creator, - 'thumbnail': thumbnail, - 'view_count': view_count, - 'formats': formats, - 'subtitles': subtitles, - } + return dict(self._get_common_fields(webpage), + id=video_id, + formats=formats, + view_count=view_count, + subtitles=subtitles) diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py index 93d15a556..4f1a99a89 100644 --- a/youtube_dl/extractor/voicerepublic.py +++ b/youtube_dl/extractor/voicerepublic.py @@ -3,7 +3,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urlparse +from ..compat import ( + compat_str, + compat_urlparse, +) from ..utils import ( ExtractorError, determine_ext, @@ -16,13 +19,13 @@ class VoiceRepublicIE(InfoExtractor): _VALID_URL = r'https?://voicerepublic\.com/(?:talks|embed)/(?P<id>[0-9a-z-]+)' _TESTS = [{ 'url': 'http://voicerepublic.com/talks/watching-the-watchers-building-a-sousveillance-state', - 'md5': '0554a24d1657915aa8e8f84e15dc9353', + 'md5': 'b9174d651323f17783000876347116e3', 'info_dict': { 'id': '2296', 'display_id': 'watching-the-watchers-building-a-sousveillance-state', 'ext': 'm4a', 'title': 'Watching the Watchers: Building a Sousveillance State', - 'description': 'md5:715ba964958afa2398df615809cfecb1', + 'description': 'Secret surveillance programs have metadata too. The people and companies that operate secret surveillance programs can be surveilled.', 'thumbnail': 're:^https?://.*\.(?:png|jpg)$', 'duration': 1800, 'view_count': int, @@ -52,7 +55,7 @@ class VoiceRepublicIE(InfoExtractor): if data: title = data['title'] description = data.get('teaser') - talk_id = data.get('talk_id') or display_id + talk_id = compat_str(data.get('talk_id') or display_id) talk = data['talk'] duration = int_or_none(talk.get('duration')) formats = [{ diff --git a/youtube_dl/extractor/voxmedia.py b/youtube_dl/extractor/voxmedia.py index 0c6b1f030..b1b32ad44 100644 --- a/youtube_dl/extractor/voxmedia.py +++ b/youtube_dl/extractor/voxmedia.py @@ -15,7 +15,8 @@ class VoxMediaIE(InfoExtractor): 'ext': 'mp4', 'title': 'Google\'s new material design direction', 'description': 'md5:2f44f74c4d14a1f800ea73e1c6832ad2', - } + }, + 'add_ie': ['Ooyala'], }, { # data-ooyala-id 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet', @@ -25,7 +26,8 @@ class VoxMediaIE(InfoExtractor): 'ext': 'mp4', 'title': 'The Nexus 6: hands-on with Google\'s phablet', 'description': 'md5:87a51fe95ff8cea8b5bdb9ac7ae6a6af', - } + }, + 'add_ie': ['Ooyala'], }, { # volume embed 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill', @@ -35,7 +37,8 @@ class VoxMediaIE(InfoExtractor): 'ext': 'mp4', 'title': 'The new frontier of LGBTQ civil rights, explained', 'description': 'md5:0dc58e94a465cbe91d02950f770eb93f', - } + }, + 'add_ie': ['Ooyala'], }, { # youtube embed 'url': 'http://www.vox.com/2016/3/24/11291692/robot-dance', @@ -48,7 +51,8 @@ class VoxMediaIE(InfoExtractor): 'upload_date': '20160324', 'uploader_id': 'voxdotcom', 'uploader': 'Vox', - } + }, + 'add_ie': ['Youtube'], }, { # SBN.VideoLinkset.entryGroup multiple ooyala embeds 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', @@ -117,7 +121,7 @@ class VoxMediaIE(InfoExtractor): volume_webpage = self._download_webpage( 'http://volume.vox-cdn.com/embed/%s' % volume_uuid, volume_uuid) video_data = self._parse_json(self._search_regex( - r'Volume\.createVideo\(({.+})\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid) + r'Volume\.createVideo\(({.+})\s*,\s*{.*}\s*,\s*\[.*\]\s*,\s*{.*}\);', volume_webpage, 'video data'), volume_uuid) for provider_video_type in ('ooyala', 'youtube'): provider_video_id = video_data.get('%s_id' % provider_video_type) if provider_video_id: diff --git a/youtube_dl/extractor/vporn.py b/youtube_dl/extractor/vporn.py index 92c90e517..1557a0e04 100644 --- a/youtube_dl/extractor/vporn.py +++ b/youtube_dl/extractor/vporn.py @@ -4,6 +4,7 @@ import re from .common import InfoExtractor from ..utils import ( + ExtractorError, parse_duration, str_to_int, ) @@ -27,7 +28,8 @@ class VpornIE(InfoExtractor): 'duration': 393, 'age_limit': 18, 'view_count': int, - } + }, + 'skip': 'video removed', }, { 'url': 'http://www.vporn.com/female/hana-shower/523564/', @@ -40,7 +42,7 @@ class VpornIE(InfoExtractor): 'description': 'Hana showers at the bathroom.', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Hmmmmm', - 'categories': ['Big Boobs', 'Erotic', 'Teen', 'Female'], + 'categories': ['Big Boobs', 'Erotic', 'Teen', 'Female', '720p'], 'duration': 588, 'age_limit': 18, 'view_count': int, @@ -55,6 +57,10 @@ class VpornIE(InfoExtractor): webpage = self._download_webpage(url, display_id) + errmsg = 'This video has been deleted due to Copyright Infringement or by the account owner!' + if errmsg in webpage: + raise ExtractorError('%s said: %s' % (self.IE_NAME, errmsg), expected=True) + title = self._html_search_regex( r'videoname\s*=\s*\'([^\']+)\'', webpage, 'title').strip() description = self._html_search_regex( diff --git a/youtube_dl/extractor/vrt.py b/youtube_dl/extractor/vrt.py index 8e35f24e8..bec7ab327 100644 --- a/youtube_dl/extractor/vrt.py +++ b/youtube_dl/extractor/vrt.py @@ -25,7 +25,8 @@ class VRTIE(InfoExtractor): 'timestamp': 1414271750.949, 'upload_date': '20141025', 'duration': 929, - } + }, + 'skip': 'HTTP Error 404: Not Found', }, # sporza.be { @@ -39,7 +40,8 @@ class VRTIE(InfoExtractor): 'timestamp': 1413835980.560, 'upload_date': '20141020', 'duration': 3238, - } + }, + 'skip': 'HTTP Error 404: Not Found', }, # cobra.be { @@ -53,16 +55,39 @@ class VRTIE(InfoExtractor): 'timestamp': 1413967500.494, 'upload_date': '20141022', 'duration': 661, - } + }, + 'skip': 'HTTP Error 404: Not Found', }, { # YouTube video 'url': 'http://deredactie.be/cm/vrtnieuws/videozone/nieuws/cultuurenmedia/1.2622957', - 'only_matching': True, + 'md5': 'b8b93da1df1cea6c8556255a796b7d61', + 'info_dict': { + 'id': 'Wji-BZ0oCwg', + 'ext': 'mp4', + 'title': 'ROGUE ONE: A STAR WARS STORY Official Teaser Trailer', + 'description': 'md5:8e468944dce15567a786a67f74262583', + 'uploader': 'Star Wars', + 'uploader_id': 'starwars', + 'upload_date': '20160407', + }, + 'add_ie': ['Youtube'], }, { 'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055', - 'only_matching': True, + 'md5': '', + 'info_dict': { + 'id': '2377055', + 'ext': 'mp4', + 'title': 'Cafe Derby', + 'description': 'Lenny Van Wesemael debuteert met de langspeelfilm Café Derby. Een waar gebeurd maar ook verzonnen verhaal.', + 'upload_date': '20150626', + 'timestamp': 1435305240.769, + }, + 'params': { + # m3u8 download + 'skip_download': True, + } } ] @@ -98,6 +123,32 @@ class VRTIE(InfoExtractor): formats.extend(self._extract_m3u8_formats( src, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_f4m_formats( + src.replace('playlist.m3u8', 'manifest.f4m'), + video_id, f4m_id='hds', fatal=False)) + if 'data-video-geoblocking="true"' not in webpage: + rtmp_formats = self._extract_smil_formats( + src.replace('playlist.m3u8', 'jwplayer.smil'), + video_id, fatal=False) + formats.extend(rtmp_formats) + for rtmp_format in rtmp_formats: + rtmp_format_c = rtmp_format.copy() + rtmp_format_c['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path']) + del rtmp_format_c['play_path'] + del rtmp_format_c['ext'] + http_format = rtmp_format_c.copy() + http_format.update({ + 'url': rtmp_format_c['url'].replace('rtmp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''), + 'format_id': rtmp_format['format_id'].replace('rtmp', 'http'), + 'protocol': 'http', + }) + rtsp_format = rtmp_format_c.copy() + rtsp_format.update({ + 'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'), + 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'), + 'protocol': 'rtsp', + }) + formats.extend([http_format, rtsp_format]) else: formats.extend(self._extract_f4m_formats( '%s/manifest.f4m' % src, video_id, f4m_id='hds', fatal=False)) diff --git a/youtube_dl/extractor/vulture.py b/youtube_dl/extractor/vulture.py deleted file mode 100644 index faa167e65..000000000 --- a/youtube_dl/extractor/vulture.py +++ /dev/null @@ -1,69 +0,0 @@ -from __future__ import unicode_literals - -import json -import os.path -import re - -from .common import InfoExtractor -from ..utils import ( - int_or_none, - parse_iso8601, -) - - -class VultureIE(InfoExtractor): - IE_NAME = 'vulture.com' - _VALID_URL = r'https?://video\.vulture\.com/video/(?P<display_id>[^/]+)/' - _TEST = { - 'url': 'http://video.vulture.com/video/Mindy-Kaling-s-Harvard-Speech/player?layout=compact&read_more=1', - 'md5': '8d997845642a2b5152820f7257871bc8', - 'info_dict': { - 'id': '6GHRQL3RV7MSD1H4', - 'ext': 'mp4', - 'title': 'kaling-speech-2-MAGNIFY STANDARD CONTAINER REVISED', - 'uploader_id': 'Sarah', - 'thumbnail': 're:^http://.*\.jpg$', - 'timestamp': 1401288564, - 'upload_date': '20140528', - 'description': 'Uplifting and witty, as predicted.', - 'duration': 1015, - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') - - webpage = self._download_webpage(url, display_id) - query_string = self._search_regex( - r"queryString\s*=\s*'([^']+)'", webpage, 'query string') - video_id = self._search_regex( - r'content=([^&]+)', query_string, 'video ID') - query_url = 'http://video.vulture.com/embed/player/container/1000/1000/?%s' % query_string - - query_webpage = self._download_webpage( - query_url, display_id, note='Downloading query page') - params_json = self._search_regex( - r'(?sm)new MagnifyEmbeddablePlayer\({.*?contentItem:\s*(\{.*?\})\n?,\n', - query_webpage, - 'player params') - params = json.loads(params_json) - - upload_timestamp = parse_iso8601(params['posted'].replace(' ', 'T')) - uploader_id = params.get('user', {}).get('handle') - - media_item = params['media_item'] - title = os.path.splitext(media_item['title'])[0] - duration = int_or_none(media_item.get('duration_seconds')) - - return { - 'id': video_id, - 'display_id': display_id, - 'url': media_item['pipeline_xid'], - 'title': title, - 'timestamp': upload_timestamp, - 'thumbnail': params.get('thumbnail_url'), - 'uploader_id': uploader_id, - 'description': params.get('description'), - 'duration': duration, - } diff --git a/youtube_dl/extractor/washingtonpost.py b/youtube_dl/extractor/washingtonpost.py index ec8b99998..839cad986 100644 --- a/youtube_dl/extractor/washingtonpost.py +++ b/youtube_dl/extractor/washingtonpost.py @@ -11,7 +11,96 @@ from ..utils import ( class WashingtonPostIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])' + IE_NAME = 'washingtonpost' + _VALID_URL = r'(?:washingtonpost:|https?://(?:www\.)?washingtonpost\.com/video/(?:[^/]+/)*)(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _TEST = { + 'url': 'https://www.washingtonpost.com/video/c/video/480ba4ee-1ec7-11e6-82c2-a7dcb313287d', + 'md5': '6f537e1334b714eb15f9563bd4b9cdfa', + 'info_dict': { + 'id': '480ba4ee-1ec7-11e6-82c2-a7dcb313287d', + 'ext': 'mp4', + 'title': 'Egypt finds belongings, debris from plane crash', + 'description': 'md5:a17ceee432f215a5371388c1f680bd86', + 'upload_date': '20160520', + 'uploader': 'Reuters', + 'timestamp': 1463778452, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + video_data = self._download_json( + 'http://www.washingtonpost.com/posttv/c/videojson/%s?resType=jsonp' % video_id, + video_id, transform_source=strip_jsonp)[0]['contentConfig'] + title = video_data['title'] + + urls = [] + formats = [] + for s in video_data.get('streams', []): + s_url = s.get('url') + if not s_url or s_url in urls: + continue + urls.append(s_url) + video_type = s.get('type') + if video_type == 'smil': + continue + elif video_type in ('ts', 'hls') and ('_master.m3u8' in s_url or '_mobile.m3u8' in s_url): + m3u8_formats = self._extract_m3u8_formats( + s_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + for m3u8_format in m3u8_formats: + width = m3u8_format.get('width') + if not width: + continue + vbr = self._search_regex( + r'%d_%d_(\d+)' % (width, m3u8_format['height']), m3u8_format['url'], 'vbr', default=None) + if vbr: + m3u8_format.update({ + 'vbr': int_or_none(vbr), + }) + formats.extend(m3u8_formats) + else: + width = int_or_none(s.get('width')) + vbr = int_or_none(s.get('bitrate')) + has_width = width != 0 + formats.append({ + 'format_id': ( + '%s-%d-%d' % (video_type, width, vbr) + if width + else video_type), + 'vbr': vbr if has_width else None, + 'width': width, + 'height': int_or_none(s.get('height')), + 'acodec': s.get('audioCodec'), + 'vcodec': s.get('videoCodec') if has_width else 'none', + 'filesize': int_or_none(s.get('fileSize')), + 'url': s_url, + 'ext': 'mp4', + 'protocol': 'm3u8_native' if video_type in ('ts', 'hls') else None, + }) + source_media_url = video_data.get('sourceMediaURL') + if source_media_url: + formats.append({ + 'format_id': 'source_media', + 'url': source_media_url, + }) + self._sort_formats( + formats, ('width', 'height', 'vbr', 'filesize', 'tbr', 'format_id')) + + return { + 'id': video_id, + 'title': title, + 'description': video_data.get('blurb'), + 'uploader': video_data.get('credits', {}).get('source'), + 'formats': formats, + 'duration': int_or_none(video_data.get('videoDuration'), 100), + 'timestamp': int_or_none( + video_data.get('dateConfig', {}).get('dateFirstPublished'), 1000), + } + + +class WashingtonPostArticleIE(InfoExtractor): + IE_NAME = 'washingtonpost:article' + _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/(?:[^/]+/)*(?P<id>[^/?#]+)' _TESTS = [{ 'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/', 'info_dict': { @@ -63,6 +152,10 @@ class WashingtonPostIE(InfoExtractor): }] }] + @classmethod + def suitable(cls, url): + return False if WashingtonPostIE.suitable(url) else super(WashingtonPostArticleIE, cls).suitable(url) + def _real_extract(self, url): page_id = self._match_id(url) webpage = self._download_webpage(url, page_id) @@ -74,54 +167,7 @@ class WashingtonPostIE(InfoExtractor): <div\s+class="posttv-video-embed[^>]*?data-uuid=| data-video-uuid= )"([^"]+)"''', webpage) - entries = [] - for i, uuid in enumerate(uuids, start=1): - vinfo_all = self._download_json( - 'http://www.washingtonpost.com/posttv/c/videojson/%s?resType=jsonp' % uuid, - page_id, - transform_source=strip_jsonp, - note='Downloading information of video %d/%d' % (i, len(uuids)) - ) - vinfo = vinfo_all[0]['contentConfig'] - uploader = vinfo.get('credits', {}).get('source') - timestamp = int_or_none( - vinfo.get('dateConfig', {}).get('dateFirstPublished'), 1000) - - formats = [{ - 'format_id': ( - '%s-%s-%s' % (s.get('type'), s.get('width'), s.get('bitrate')) - if s.get('width') - else s.get('type')), - 'vbr': s.get('bitrate') if s.get('width') != 0 else None, - 'width': s.get('width'), - 'height': s.get('height'), - 'acodec': s.get('audioCodec'), - 'vcodec': s.get('videoCodec') if s.get('width') != 0 else 'none', - 'filesize': s.get('fileSize'), - 'url': s.get('url'), - 'ext': 'mp4', - 'preference': -100 if s.get('type') == 'smil' else None, - 'protocol': { - 'MP4': 'http', - 'F4F': 'f4m', - }.get(s.get('type')), - } for s in vinfo.get('streams', [])] - source_media_url = vinfo.get('sourceMediaURL') - if source_media_url: - formats.append({ - 'format_id': 'source_media', - 'url': source_media_url, - }) - self._sort_formats(formats) - entries.append({ - 'id': uuid, - 'title': vinfo['title'], - 'description': vinfo.get('blurb'), - 'uploader': uploader, - 'formats': formats, - 'duration': int_or_none(vinfo.get('videoDuration'), 100), - 'timestamp': timestamp, - }) + entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids] return { '_type': 'playlist', diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index 5227bb5ad..de7d6b559 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -2,25 +2,26 @@ from __future__ import unicode_literals import re -import hashlib from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( ExtractorError, unified_strdate, + HEADRequest, + float_or_none, ) class WatIE(InfoExtractor): - _VALID_URL = r'(?:wat:(?P<real_id>\d{8})|https?://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html)' + _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P<id>[0-9a-z]+)' IE_NAME = 'wat.tv' _TESTS = [ { 'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html', - 'md5': 'ce70e9223945ed26a8056d413ca55dc9', + 'md5': '83d882d9de5c9d97f0bb2c6273cde56a', 'info_dict': { 'id': '11713067', - 'display_id': 'soupe-figues-l-orange-aux-epices', 'ext': 'mp4', 'title': 'Soupe de figues à l\'orange et aux épices', 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.', @@ -33,7 +34,6 @@ class WatIE(InfoExtractor): 'md5': 'fbc84e4378165278e743956d9c1bf16b', 'info_dict': { 'id': '11713075', - 'display_id': 'gregory-lemarchal-voix-ange', 'ext': 'mp4', 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)', 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3', @@ -44,96 +44,85 @@ class WatIE(InfoExtractor): }, ] - def download_video_info(self, real_id): - # 'contentv4' is used in the website, but it also returns the related - # videos, we don't need them - info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id) - return info['media'] - def _real_extract(self, url): - def real_id_for_chapter(chapter): - return chapter['tc_start'].split('-')[0] - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') - real_id = mobj.group('real_id') - if not real_id: - short_id = mobj.group('short_id') - webpage = self._download_webpage(url, display_id or short_id) - real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id') + video_id = self._match_id(url) + video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36)) - video_info = self.download_video_info(real_id) + # 'contentv4' is used in the website, but it also returns the related + # videos, we don't need them + video_info = self._download_json( + 'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media'] error_desc = video_info.get('error_desc') if error_desc: raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True) - geo_list = video_info.get('geoList') - country = geo_list[0] if geo_list else '' - chapters = video_info['chapters'] first_chapter = chapters[0] - files = video_info['files'] - first_file = files[0] - if real_id_for_chapter(first_chapter) != real_id: - self.to_screen('Multipart video detected') - chapter_urls = [] - for chapter in chapters: - chapter_id = real_id_for_chapter(chapter) - # Yes, when we this chapter is processed by WatIE, - # it will download the info again - chapter_info = self.download_video_info(chapter_id) - chapter_urls.append(chapter_info['url']) - entries = [self.url_result(chapter_url) for chapter_url in chapter_urls] - return self.playlist_result(entries, real_id, video_info['title']) + def video_id_for_chapter(chapter): + return chapter['tc_start'].split('-')[0] - upload_date = None - if 'date_diffusion' in first_chapter: - upload_date = unified_strdate(first_chapter['date_diffusion']) + if video_id_for_chapter(first_chapter) != video_id: + self.to_screen('Multipart video detected') + entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters] + return self.playlist_result(entries, video_id, video_info['title']) # Otherwise we can continue and extract just one part, we have to use - # the short id for getting the video url - - formats = [{ - 'url': 'http://wat.tv/get/android5/%s.mp4' % real_id, - 'format_id': 'Mobile', - }] - - fmts = [('SD', 'web')] - if first_file.get('hasHD'): - fmts.append(('HD', 'webhd')) - - def compute_token(param): - timestamp = '%08x' % int(self._download_webpage( - 'http://www.wat.tv/servertime', real_id, - 'Downloading server time').split('|')[0]) - magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564' - return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp) - - for fmt in fmts: - webid = '/%s/%s' % (fmt[1], real_id) - video_url = self._download_webpage( - 'http://www.wat.tv/get%s?token=%s&getURL=1&country=%s' % (webid, compute_token(webid), country), - real_id, - 'Downloading %s video URL' % fmt[0], - 'Failed to download %s video URL' % fmt[0], - False) - if not video_url: + # the video id for getting the video url + + date_diffusion = first_chapter.get('date_diffusion') + upload_date = unified_strdate(date_diffusion) if date_diffusion else None + + def extract_url(path_template, url_type): + req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id) + head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type) + red_url = head.geturl() + if req_url == red_url: + raise ExtractorError( + '%s said: Sorry, this video is not available from your country.' % self.IE_NAME, + expected=True) + return red_url + + m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8') + http_url = extract_url('android5/%s.mp4', 'http') + + formats = [] + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls') + formats.extend(m3u8_formats) + formats.extend(self._extract_f4m_formats( + m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'), + video_id, f4m_id='hds', fatal=False)) + for m3u8_format in m3u8_formats: + mobj = re.search( + r'audio.*?%3D(\d+)(?:-video.*?%3D(\d+))?', m3u8_format['url']) + if not mobj: continue - formats.append({ - 'url': video_url, - 'ext': 'mp4', - 'format_id': fmt[0], + abr, vbr = mobj.groups() + abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000) + m3u8_format.update({ + 'vbr': vbr, + 'abr': abr, + }) + if not vbr or not abr: + continue + f = m3u8_format.copy() + f.update({ + 'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url), + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', }) + formats.append(f) + self._sort_formats(formats) return { - 'id': real_id, - 'display_id': display_id, + 'id': video_id, 'title': first_chapter['title'], 'thumbnail': first_chapter['preview'], 'description': first_chapter['description'], 'view_count': video_info['views'], 'upload_date': upload_date, - 'duration': first_file['duration'], + 'duration': video_info['files'][0]['duration'], 'formats': formats, } diff --git a/youtube_dl/extractor/sexykarma.py b/youtube_dl/extractor/watchindianporn.py index e33483674..5d3b5bdb4 100644 --- a/youtube_dl/extractor/sexykarma.py +++ b/youtube_dl/extractor/watchindianporn.py @@ -11,61 +11,27 @@ from ..utils import ( ) -class SexyKarmaIE(InfoExtractor): - IE_DESC = 'Sexy Karma and Watch Indian Porn' - _VALID_URL = r'https?://(?:www\.)?(?:sexykarma\.com|watchindianporn\.net)/(?:[^/]+/)*video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html' - _TESTS = [{ - 'url': 'http://www.sexykarma.com/gonewild/video/taking-a-quick-pee-yHI70cOyIHt.html', - 'md5': 'b9798e7d1ef1765116a8f516c8091dbd', +class WatchIndianPornIE(InfoExtractor): + IE_DESC = 'Watch Indian Porn' + _VALID_URL = r'https?://(?:www\.)?watchindianporn\.net/(?:[^/]+/)*video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html' + _TEST = { + 'url': 'http://www.watchindianporn.net/video/hot-milf-from-kerala-shows-off-her-gorgeous-large-breasts-on-camera-RZa2avywNPa.html', + 'md5': '249589a164dde236ec65832bfce17440', 'info_dict': { - 'id': 'yHI70cOyIHt', - 'display_id': 'taking-a-quick-pee', + 'id': 'RZa2avywNPa', + 'display_id': 'hot-milf-from-kerala-shows-off-her-gorgeous-large-breasts-on-camera', 'ext': 'mp4', - 'title': 'Taking a quick pee.', + 'title': 'Hot milf from kerala shows off her gorgeous large breasts on camera', 'thumbnail': 're:^https?://.*\.jpg$', - 'uploader': 'wildginger7', - 'upload_date': '20141008', - 'duration': 22, + 'uploader': 'LoveJay', + 'upload_date': '20160428', + 'duration': 226, 'view_count': int, 'comment_count': int, 'categories': list, 'age_limit': 18, } - }, { - 'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html', - 'md5': 'dd216c68d29b49b12842b9babe762a5d', - 'info_dict': { - 'id': '8Id6EZPbuHf', - 'display_id': 'pot-pixie-tribute', - 'ext': 'mp4', - 'title': 'pot_pixie tribute', - 'thumbnail': 're:^https?://.*\.jpg$', - 'uploader': 'banffite', - 'upload_date': '20141013', - 'duration': 16, - 'view_count': int, - 'comment_count': int, - 'categories': list, - 'age_limit': 18, - } - }, { - 'url': 'http://www.watchindianporn.net/video/desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number-dW2mtctxJfs.html', - 'md5': '9afb80675550406ed9a63ac2819ef69d', - 'info_dict': { - 'id': 'dW2mtctxJfs', - 'display_id': 'desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number', - 'ext': 'mp4', - 'title': 'Desi dancer namrata stripping completely nude and dancing on a hot number', - 'thumbnail': 're:^https?://.*\.jpg$', - 'uploader': 'Don', - 'upload_date': '20140213', - 'duration': 83, - 'view_count': int, - 'comment_count': int, - 'categories': list, - 'age_limit': 18, - } - }] + } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -109,6 +75,9 @@ class SexyKarmaIE(InfoExtractor): 'id': video_id, 'display_id': display_id, 'url': video_url, + 'http_headers': { + 'Referer': url, + }, 'title': title, 'thumbnail': thumbnail, 'uploader': uploader, diff --git a/youtube_dl/extractor/wdr.py b/youtube_dl/extractor/wdr.py index 31c904303..390f9e830 100644 --- a/youtube_dl/extractor/wdr.py +++ b/youtube_dl/extractor/wdr.py @@ -1,215 +1,236 @@ # -*- coding: utf-8 -*- from __future__ import unicode_literals -import itertools import re from .common import InfoExtractor -from ..compat import ( - compat_parse_qs, - compat_urlparse, -) from ..utils import ( + determine_ext, + ExtractorError, + js_to_json, + strip_jsonp, unified_strdate, - qualities, + update_url_query, + urlhandle_detect_ext, ) -class WDRIE(InfoExtractor): - _PLAYER_REGEX = '-(?:video|audio)player(?:_size-[LMS])?' - _VALID_URL = r'(?P<url>https?://www\d?\.(?:wdr\d?|funkhauseuropa)\.de/)(?P<id>.+?)(?P<player>%s)?\.html' % _PLAYER_REGEX +class WDRBaseIE(InfoExtractor): + def _extract_wdr_video(self, webpage, display_id): + # for wdr.de the data-extension is in a tag with the class "mediaLink" + # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn" + # for wdrmaus its in a link to the page in a multiline "videoLink"-tag + json_metadata = self._html_search_regex( + r'class=(?:"(?:mediaLink|wdrrPlayerPlayBtn)\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"', + webpage, 'media link', default=None, flags=re.MULTILINE) + + if not json_metadata: + return + + media_link_obj = self._parse_json(json_metadata, display_id, + transform_source=js_to_json) + jsonp_url = media_link_obj['mediaObj']['url'] + + metadata = self._download_json( + jsonp_url, 'metadata', transform_source=strip_jsonp) + + metadata_tracker_data = metadata['trackerData'] + metadata_media_resource = metadata['mediaResource'] + + formats = [] + + # check if the metadata contains a direct URL to a file + for kind, media_resource in metadata_media_resource.items(): + if kind not in ('dflt', 'alt'): + continue + + for tag_name, medium_url in media_resource.items(): + if tag_name not in ('videoURL', 'audioURL'): + continue + + ext = determine_ext(medium_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + medium_url, display_id, 'mp4', 'm3u8_native', + m3u8_id='hls')) + elif ext == 'f4m': + manifest_url = update_url_query( + medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'}) + formats.extend(self._extract_f4m_formats( + manifest_url, display_id, f4m_id='hds', fatal=False)) + elif ext == 'smil': + formats.extend(self._extract_smil_formats( + medium_url, 'stream', fatal=False)) + else: + a_format = { + 'url': medium_url + } + if ext == 'unknown_video': + urlh = self._request_webpage( + medium_url, display_id, note='Determining extension') + ext = urlhandle_detect_ext(urlh) + a_format['ext'] = ext + formats.append(a_format) + + self._sort_formats(formats) + + subtitles = {} + caption_url = metadata_media_resource.get('captionURL') + if caption_url: + subtitles['de'] = [{ + 'url': caption_url, + 'ext': 'ttml', + }] + + title = metadata_tracker_data['trackerClipTitle'] + + return { + 'id': metadata_tracker_data.get('trackerClipId', display_id), + 'display_id': display_id, + 'title': title, + 'alt_title': metadata_tracker_data.get('trackerClipSubcategory'), + 'formats': formats, + 'subtitles': subtitles, + 'upload_date': unified_strdate(metadata_tracker_data.get('trackerClipAirTime')), + } + + +class WDRIE(WDRBaseIE): + _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5' + _PAGE_REGEX = r'/(?:mediathek/)?[^/]+/(?P<type>[^/]+)/(?P<display_id>.+)\.html' + _VALID_URL = r'(?P<page_url>https?://(?:www\d\.)?wdr\d?\.de)' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL _TESTS = [ { - 'url': 'http://www1.wdr.de/mediathek/video/sendungen/servicezeit/videoservicezeit560-videoplayer_size-L.html', + 'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html', + # HDS download, MD5 is unstable 'info_dict': { - 'id': 'mdb-362427', + 'id': 'mdb-1058683', 'ext': 'flv', - 'title': 'Servicezeit', - 'description': 'md5:c8f43e5e815eeb54d0b96df2fba906cb', - 'upload_date': '20140310', - 'is_live': False + 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100', + 'title': 'Geheimnis Aachener Dom', + 'alt_title': 'Doku am Freitag', + 'upload_date': '20160304', + 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318', + 'is_live': False, + 'subtitles': {'de': [{ + 'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml', + 'ext': 'ttml', + }]}, }, - 'params': { - 'skip_download': True, + }, + { + 'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html', + 'md5': 'f4c1f96d01cf285240f53ea4309663d8', + 'info_dict': { + 'id': 'mdb-1072000', + 'ext': 'mp3', + 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100', + 'title': 'Schriftstellerin Juli Zeh', + 'alt_title': 'WDR 3 Gespräch am Samstag', + 'upload_date': '20160312', + 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7', + 'is_live': False, + 'subtitles': {} }, - 'skip': 'Page Not Found', }, { - 'url': 'http://www1.wdr.de/themen/av/videomargaspiegelisttot101-videoplayer.html', + 'url': 'http://www1.wdr.de/mediathek/video/live/index.html', 'info_dict': { - 'id': 'mdb-363194', - 'ext': 'flv', - 'title': 'Marga Spiegel ist tot', - 'description': 'md5:2309992a6716c347891c045be50992e4', - 'upload_date': '20140311', - 'is_live': False + 'id': 'mdb-103364', + 'ext': 'mp4', + 'display_id': 'index', + 'title': r're:^WDR Fernsehen im Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'alt_title': 'WDR Fernsehen Live', + 'upload_date': None, + 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9', + 'is_live': True, + 'subtitles': {} }, 'params': { - 'skip_download': True, + 'skip_download': True, # m3u8 download }, - 'skip': 'Page Not Found', }, { - 'url': 'http://www1.wdr.de/themen/kultur/audioerlebtegeschichtenmargaspiegel100-audioplayer.html', - 'md5': '83e9e8fefad36f357278759870805898', + 'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html', + 'playlist_mincount': 8, 'info_dict': { - 'id': 'mdb-194332', - 'ext': 'mp3', - 'title': 'Erlebte Geschichten: Marga Spiegel (29.11.2009)', - 'description': 'md5:2309992a6716c347891c045be50992e4', - 'upload_date': '20091129', - 'is_live': False + 'id': 'aktuelle-stunde/aktuelle-stunde-120', }, }, { - 'url': 'http://www.funkhauseuropa.de/av/audioflaviacoelhoamaramar100-audioplayer.html', - 'md5': '99a1443ff29af19f6c52cf6f4dc1f4aa', + 'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5', 'info_dict': { - 'id': 'mdb-478135', - 'ext': 'mp3', - 'title': 'Flavia Coelho: Amar é Amar', - 'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a', - 'upload_date': '20140717', - 'is_live': False + 'id': 'mdb-1096487', + 'ext': 'flv', + 'upload_date': 're:^[0-9]{8}$', + 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$', + 'description': '- Die Sendung mit der Maus -', }, - 'skip': 'Page Not Found', + 'skip': 'The id changes from week to week because of the new episode' }, { - 'url': 'http://www1.wdr.de/mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100.html', - 'playlist_mincount': 146, + 'url': 'http://www.wdrmaus.de/sachgeschichten/sachgeschichten/achterbahn.php5', + 'md5': '803138901f6368ee497b4d195bb164f2', 'info_dict': { - 'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100', - } + 'id': 'mdb-186083', + 'ext': 'mp4', + 'upload_date': '20130919', + 'title': 'Sachgeschichte - Achterbahn ', + 'description': '- Die Sendung mit der Maus -', + }, }, { - 'url': 'http://www1.wdr.de/mediathek/video/livestream/index.html', + 'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html', + # Live stream, MD5 unstable 'info_dict': { - 'id': 'mdb-103364', - 'title': 're:^WDR Fernsehen Live [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', - 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9', + 'id': 'mdb-869971', 'ext': 'flv', - 'upload_date': '20150101', - 'is_live': True - }, - 'params': { - 'skip_download': True, + 'title': 'Funkhaus Europa Livestream', + 'description': 'md5:2309992a6716c347891c045be50992e4', + 'upload_date': '20160101', }, } ] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - page_url = mobj.group('url') - page_id = mobj.group('id') + url_type = mobj.group('type') + page_url = mobj.group('page_url') + display_id = mobj.group('display_id') + webpage = self._download_webpage(url, display_id) - webpage = self._download_webpage(url, page_id) + info_dict = self._extract_wdr_video(webpage, display_id) - if mobj.group('player') is None: + if not info_dict: entries = [ - self.url_result(page_url + href, 'WDR') + self.url_result(page_url + href[0], 'WDR') for href in re.findall( - r'<a href="/?(.+?%s\.html)" rel="nofollow"' % self._PLAYER_REGEX, + r'<a href="(%s)"[^>]+data-extension=' % self._PAGE_REGEX, webpage) ] if entries: # Playlist page - return self.playlist_result(entries, page_id) + return self.playlist_result(entries, playlist_id=display_id) - # Overview page - entries = [] - for page_num in itertools.count(2): - hrefs = re.findall( - r'<li class="mediathekvideo"\s*>\s*<img[^>]*>\s*<a href="(/mediathek/video/[^"]+)"', - webpage) - entries.extend( - self.url_result(page_url + href, 'WDR') - for href in hrefs) - next_url_m = re.search( - r'<li class="nextToLast">\s*<a href="([^"]+)"', webpage) - if not next_url_m: - break - next_url = page_url + next_url_m.group(1) - webpage = self._download_webpage( - next_url, page_id, - note='Downloading playlist page %d' % page_num) - return self.playlist_result(entries, page_id) - - flashvars = compat_parse_qs(self._html_search_regex( - r'<param name="flashvars" value="([^"]+)"', webpage, 'flashvars')) + raise ExtractorError('No downloadable streams found', expected=True) - page_id = flashvars['trackerClipId'][0] - video_url = flashvars['dslSrc'][0] - title = flashvars['trackerClipTitle'][0] - thumbnail = flashvars['startPicture'][0] if 'startPicture' in flashvars else None - is_live = flashvars.get('isLive', ['0'])[0] == '1' + is_live = url_type == 'live' if is_live: - title = self._live_title(title) - - if 'trackerClipAirTime' in flashvars: - upload_date = flashvars['trackerClipAirTime'][0] - else: - upload_date = self._html_search_meta( - 'DC.Date', webpage, 'upload date') - - if upload_date: - upload_date = unified_strdate(upload_date) - - formats = [] - preference = qualities(['S', 'M', 'L', 'XL']) - - if video_url.endswith('.f4m'): - formats.extend(self._extract_f4m_formats( - video_url + '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18', page_id, - f4m_id='hds', fatal=False)) - elif video_url.endswith('.smil'): - formats.extend(self._extract_smil_formats( - video_url, page_id, False, { - 'hdcore': '3.3.0', - 'plugin': 'aasp-3.3.0.99.43', - })) - else: - formats.append({ - 'url': video_url, - 'http_headers': { - 'User-Agent': 'mobile', - }, + info_dict.update({ + 'title': self._live_title(info_dict['title']), + 'upload_date': None, }) + elif 'upload_date' not in info_dict: + info_dict['upload_date'] = unified_strdate(self._html_search_meta('DC.Date', webpage, 'upload date')) - m3u8_url = self._search_regex( - r'rel="adaptiv"[^>]+href="([^"]+)"', - webpage, 'm3u8 url', default=None) - if m3u8_url: - formats.extend(self._extract_m3u8_formats( - m3u8_url, page_id, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) + info_dict.update({ + 'description': self._html_search_meta('Description', webpage), + 'is_live': is_live, + }) - direct_urls = re.findall( - r'rel="web(S|M|L|XL)"[^>]+href="([^"]+)"', webpage) - if direct_urls: - for quality, video_url in direct_urls: - formats.append({ - 'url': video_url, - 'preference': preference(quality), - 'http_headers': { - 'User-Agent': 'mobile', - }, - }) - - self._sort_formats(formats) - - description = self._html_search_meta('Description', webpage, 'description') - - return { - 'id': page_id, - 'formats': formats, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'is_live': is_live - } + return info_dict class WDRMobileIE(InfoExtractor): @@ -241,81 +262,3 @@ class WDRMobileIE(InfoExtractor): 'User-Agent': 'mobile', }, } - - -class WDRMausIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?wdrmaus\.de/(?:[^/]+/){,2}(?P<id>[^/?#]+)(?:/index\.php5|(?<!index)\.php5|/(?:$|[?#]))' - IE_DESC = 'Sendung mit der Maus' - _TESTS = [{ - 'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5', - 'info_dict': { - 'id': 'aktuelle-sendung', - 'ext': 'mp4', - 'thumbnail': 're:^http://.+\.jpg', - 'upload_date': 're:^[0-9]{8}$', - 'title': 're:^[0-9.]{10} - Aktuelle Sendung$', - } - }, { - 'url': 'http://www.wdrmaus.de/sachgeschichten/sachgeschichten/40_jahre_maus.php5', - 'md5': '3b1227ca3ed28d73ec5737c65743b2a3', - 'info_dict': { - 'id': '40_jahre_maus', - 'ext': 'mp4', - 'thumbnail': 're:^http://.+\.jpg', - 'upload_date': '20131007', - 'title': '12.03.2011 - 40 Jahre Maus', - } - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - param_code = self._html_search_regex( - r'<a href="\?startVideo=1&([^"]+)"', webpage, 'parameters') - - title_date = self._search_regex( - r'<div class="sendedatum"><p>Sendedatum:\s*([0-9\.]+)</p>', - webpage, 'air date') - title_str = self._html_search_regex( - r'<h1>(.*?)</h1>', webpage, 'title') - title = '%s - %s' % (title_date, title_str) - upload_date = unified_strdate( - self._html_search_meta('dc.date', webpage)) - - fields = compat_parse_qs(param_code) - video_url = fields['firstVideo'][0] - thumbnail = compat_urlparse.urljoin(url, fields['startPicture'][0]) - - formats = [{ - 'format_id': 'rtmp', - 'url': video_url, - }] - - jscode = self._download_webpage( - 'http://www.wdrmaus.de/codebase/js/extended-medien.min.js', - video_id, fatal=False, - note='Downloading URL translation table', - errnote='Could not download URL translation table') - if jscode: - for m in re.finditer( - r"stream:\s*'dslSrc=(?P<stream>[^']+)',\s*download:\s*'(?P<dl>[^']+)'\s*\}", - jscode): - if video_url.startswith(m.group('stream')): - http_url = video_url.replace( - m.group('stream'), m.group('dl')) - formats.append({ - 'format_id': 'http', - 'url': http_url, - }) - break - - self._sort_formats(formats) - - return { - 'id': video_id, - 'title': title, - 'formats': formats, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - } diff --git a/youtube_dl/extractor/weibo.py b/youtube_dl/extractor/weibo.py deleted file mode 100644 index 20bb039d3..000000000 --- a/youtube_dl/extractor/weibo.py +++ /dev/null @@ -1,49 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor - - -class WeiboIE(InfoExtractor): - """ - The videos in Weibo come from different sites, this IE just finds the link - to the external video and returns it. - """ - _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm' - - _TEST = { - 'url': 'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm', - 'info_dict': { - 'id': '98322879', - 'ext': 'flv', - 'title': '魔声耳机最新广告“All Eyes On Us”', - }, - 'params': { - 'skip_download': True, - }, - 'add_ie': ['Sina'], - } - - # Additional example videos from different sites - # Youku: http://video.weibo.com/v/weishipin/t_zQGDWQ8.htm - # 56.com: http://video.weibo.com/v/weishipin/t_zQ44HxN.htm - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE) - video_id = mobj.group('id') - info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id - info = self._download_json(info_url, video_id) - - videos_urls = map(lambda v: v['play_page_url'], info['result']['data']) - # Prefer sina video since they have thumbnails - videos_urls = sorted(videos_urls, key=lambda u: 'video.sina.com' in u) - player_url = videos_urls[-1] - m_sina = re.match(r'https?://video\.sina\.com\.cn/v/b/(\d+)-\d+\.html', - player_url) - if m_sina is not None: - self.to_screen('Sina video detected') - sina_id = m_sina.group(1) - player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id - return self.url_result(player_url) diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py index 828c03dc3..54eb51427 100644 --- a/youtube_dl/extractor/wimp.py +++ b/youtube_dl/extractor/wimp.py @@ -1,29 +1,33 @@ from __future__ import unicode_literals -from .common import InfoExtractor from .youtube import YoutubeIE +from .jwplatform import JWPlatformBaseIE -class WimpIE(InfoExtractor): +class WimpIE(JWPlatformBaseIE): _VALID_URL = r'https?://(?:www\.)?wimp\.com/(?P<id>[^/]+)' _TESTS = [{ - 'url': 'http://www.wimp.com/maruexhausted/', + 'url': 'http://www.wimp.com/maru-is-exhausted/', 'md5': 'ee21217ffd66d058e8b16be340b74883', 'info_dict': { - 'id': 'maruexhausted', + 'id': 'maru-is-exhausted', 'ext': 'mp4', 'title': 'Maru is exhausted.', 'description': 'md5:57e099e857c0a4ea312542b684a869b8', } }, { 'url': 'http://www.wimp.com/clowncar/', - 'md5': '4e2986c793694b55b37cf92521d12bb4', + 'md5': '5c31ad862a90dc5b1f023956faec13fe', 'info_dict': { - 'id': 'clowncar', + 'id': 'cG4CEr2aiSg', 'ext': 'webm', - 'title': 'It\'s like a clown car.', - 'description': 'md5:0e56db1370a6e49c5c1d19124c0d2fb2', + 'title': 'Basset hound clown car...incredible!', + 'description': '5 of my Bassets crawled in this dog loo! www.bellinghambassets.com\n\nFor licensing/usage please contact: licensing(at)jukinmediadotcom', + 'upload_date': '20140303', + 'uploader': 'Gretchen Hoey', + 'uploader_id': 'gretchenandjeff1', }, + 'add_ie': ['Youtube'], }] def _real_extract(self, url): @@ -41,14 +45,13 @@ class WimpIE(InfoExtractor): 'ie_key': YoutubeIE.ie_key(), } - video_url = self._search_regex( - r'<video[^>]+>\s*<source[^>]+src=(["\'])(?P<url>.+?)\1', - webpage, 'video URL', group='url') + info_dict = self._extract_jwplayer_data( + webpage, video_id, require_title=False) - return { + info_dict.update({ 'id': video_id, - 'url': video_url, 'title': self._og_search_title(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), 'description': self._og_search_description(webpage), - } + }) + + return info_dict diff --git a/youtube_dl/extractor/wistia.py b/youtube_dl/extractor/wistia.py index 8b14840a2..c634b8dec 100644 --- a/youtube_dl/extractor/wistia.py +++ b/youtube_dl/extractor/wistia.py @@ -3,16 +3,17 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( ExtractorError, - sanitized_Request, int_or_none, + float_or_none, ) class WistiaIE(InfoExtractor): - _VALID_URL = r'https?://(?:fast\.)?wistia\.net/embed/iframe/(?P<id>[a-z0-9]+)' - _API_URL = 'http://fast.wistia.com/embed/medias/{0:}.json' + _VALID_URL = r'(?:wistia:|https?://(?:fast\.)?wistia\.net/embed/iframe/)(?P<id>[a-z0-9]+)' + _API_URL = 'http://fast.wistia.com/embed/medias/%s.json' + _IFRAME_URL = 'http://fast.wistia.net/embed/iframe/%s' - _TEST = { + _TESTS = [{ 'url': 'http://fast.wistia.net/embed/iframe/sh7fpupwlt', 'md5': 'cafeb56ec0c53c18c97405eecb3133df', 'info_dict': { @@ -24,36 +25,54 @@ class WistiaIE(InfoExtractor): 'timestamp': 1386185018, 'duration': 117, }, - } + }, { + 'url': 'wistia:sh7fpupwlt', + 'only_matching': True, + }, { + # with hls video + 'url': 'wistia:807fafadvk', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - request = sanitized_Request(self._API_URL.format(video_id)) - request.add_header('Referer', url) # Some videos require this. - data_json = self._download_json(request, video_id) + data_json = self._download_json( + self._API_URL % video_id, video_id, + # Some videos require this. + headers={ + 'Referer': url if url.startswith('http') else self._IFRAME_URL % video_id, + }) + if data_json.get('error'): - raise ExtractorError('Error while getting the playlist', - expected=True) + raise ExtractorError( + 'Error while getting the playlist', expected=True) + data = data_json['media'] title = data['name'] formats = [] thumbnails = [] for a in data['assets']: + aurl = a.get('url') + if not aurl: + continue astatus = a.get('status') atype = a.get('type') - if (astatus is not None and astatus != 2) or atype == 'preview': + if (astatus is not None and astatus != 2) or atype in ('preview', 'storyboard'): continue elif atype in ('still', 'still_image'): thumbnails.append({ - 'url': a['url'], - 'resolution': '%dx%d' % (a['width'], a['height']), + 'url': aurl, + 'width': int_or_none(a.get('width')), + 'height': int_or_none(a.get('height')), }) else: + aext = a.get('ext') + is_m3u8 = a.get('container') == 'm3u8' or aext == 'm3u8' formats.append({ 'format_id': atype, - 'url': a['url'], + 'url': aurl, 'tbr': int_or_none(a.get('bitrate')), 'vbr': int_or_none(a.get('opt_vbitrate')), 'width': int_or_none(a.get('width')), @@ -61,7 +80,8 @@ class WistiaIE(InfoExtractor): 'filesize': int_or_none(a.get('size')), 'vcodec': a.get('codec'), 'container': a.get('container'), - 'ext': a.get('ext'), + 'ext': 'mp4' if is_m3u8 else aext, + 'protocol': 'm3u8' if is_m3u8 else None, 'preference': 1 if atype == 'original' else None, }) @@ -73,6 +93,6 @@ class WistiaIE(InfoExtractor): 'description': data.get('seoDescription'), 'formats': formats, 'thumbnails': thumbnails, - 'duration': int_or_none(data.get('duration')), + 'duration': float_or_none(data.get('duration')), 'timestamp': int_or_none(data.get('createdAt')), } diff --git a/youtube_dl/extractor/wrzuta.py b/youtube_dl/extractor/wrzuta.py index c42764921..bdd7097ba 100644 --- a/youtube_dl/extractor/wrzuta.py +++ b/youtube_dl/extractor/wrzuta.py @@ -5,8 +5,10 @@ import re from .common import InfoExtractor from ..utils import ( + ExtractorError, int_or_none, qualities, + remove_start, ) @@ -26,16 +28,17 @@ class WrzutaIE(InfoExtractor): 'uploader_id': 'laboratoriumdextera', 'description': 'md5:7fb5ef3c21c5893375fda51d9b15d9cd', }, + 'skip': 'Redirected to wrzuta.pl', }, { - 'url': 'http://jolka85.wrzuta.pl/audio/063jOPX5ue2/liber_natalia_szroeder_-_teraz_ty', - 'md5': 'bc78077859bea7bcfe4295d7d7fc9025', + 'url': 'http://vexling.wrzuta.pl/audio/01xBFabGXu6/james_horner_-_into_the_na_39_vi_world_bonus', + 'md5': 'f80564fb5a2ec6ec59705ae2bf2ba56d', 'info_dict': { - 'id': '063jOPX5ue2', - 'ext': 'ogg', - 'title': 'Liber & Natalia Szroeder - Teraz Ty', - 'duration': 203, - 'uploader_id': 'jolka85', - 'description': 'md5:2d2b6340f9188c8c4cd891580e481096', + 'id': '01xBFabGXu6', + 'ext': 'mp3', + 'title': 'James Horner - Into The Na\'vi World [Bonus]', + 'description': 'md5:30a70718b2cd9df3120fce4445b0263b', + 'duration': 95, + 'uploader_id': 'vexling', }, }] @@ -45,7 +48,10 @@ class WrzutaIE(InfoExtractor): typ = mobj.group('typ') uploader = mobj.group('uploader') - webpage = self._download_webpage(url, video_id) + webpage, urlh = self._download_webpage_handle(url, video_id) + + if urlh.geturl() == 'http://www.wrzuta.pl/': + raise ExtractorError('Video removed', expected=True) quality = qualities(['SD', 'MQ', 'HQ', 'HD']) @@ -80,3 +86,73 @@ class WrzutaIE(InfoExtractor): 'description': self._og_search_description(webpage), 'age_limit': embedpage.get('minimalAge', 0), } + + +class WrzutaPlaylistIE(InfoExtractor): + """ + this class covers extraction of wrzuta playlist entries + the extraction process bases on following steps: + * collect information of playlist size + * download all entries provided on + the playlist webpage (the playlist is split + on two pages: first directly reached from webpage + second: downloaded on demand by ajax call and rendered + using the ajax call response) + * in case size of extracted entries not reached total number of entries + use the ajax call to collect the remaining entries + """ + + IE_NAME = 'wrzuta.pl:playlist' + _VALID_URL = r'https?://(?P<uploader>[0-9a-zA-Z]+)\.wrzuta\.pl/playlista/(?P<id>[0-9a-zA-Z]+)' + _TESTS = [{ + 'url': 'http://miromak71.wrzuta.pl/playlista/7XfO4vE84iR/moja_muza', + 'playlist_mincount': 14, + 'info_dict': { + 'id': '7XfO4vE84iR', + 'title': 'Moja muza', + }, + }, { + 'url': 'http://heroesf70.wrzuta.pl/playlista/6Nj3wQHx756/lipiec_-_lato_2015_muzyka_swiata', + 'playlist_mincount': 144, + 'info_dict': { + 'id': '6Nj3wQHx756', + 'title': 'Lipiec - Lato 2015 Muzyka Świata', + }, + }, { + 'url': 'http://miromak71.wrzuta.pl/playlista/7XfO4vE84iR', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + playlist_id = mobj.group('id') + uploader = mobj.group('uploader') + + webpage = self._download_webpage(url, playlist_id) + + playlist_size = int_or_none(self._html_search_regex( + (r'<div[^>]+class=["\']playlist-counter["\'][^>]*>\d+/(\d+)', + r'<div[^>]+class=["\']all-counter["\'][^>]*>(.+?)</div>'), + webpage, 'playlist size', default=None)) + + playlist_title = remove_start( + self._og_search_title(webpage), 'Playlista: ') + + entries = [] + if playlist_size: + entries = [ + self.url_result(entry_url) + for _, entry_url in re.findall( + r'<a[^>]+href=(["\'])(http.+?)\1[^>]+class=["\']playlist-file-page', + webpage)] + if playlist_size > len(entries): + playlist_content = self._download_json( + 'http://%s.wrzuta.pl/xhr/get_playlist_offset/%s' % (uploader, playlist_id), + playlist_id, + 'Downloading playlist JSON', + 'Unable to download playlist JSON') + entries.extend([ + self.url_result(entry['filelink']) + for entry in playlist_content.get('files', []) if entry.get('filelink')]) + + return self.playlist_result(entries, playlist_id, playlist_title) diff --git a/youtube_dl/extractor/wsj.py b/youtube_dl/extractor/wsj.py index 5a897371d..a83e68b17 100644 --- a/youtube_dl/extractor/wsj.py +++ b/youtube_dl/extractor/wsj.py @@ -4,16 +4,22 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( int_or_none, + float_or_none, unified_strdate, ) class WSJIE(InfoExtractor): - _VALID_URL = r'https?://video-api\.wsj\.com/api-video/player/iframe\.html\?guid=(?P<id>[a-zA-Z0-9-]+)' + _VALID_URL = r'''(?x)https?:// + (?: + video-api\.wsj\.com/api-video/player/iframe\.html\?guid=| + (?:www\.)?wsj\.com/video/[^/]+/ + ) + (?P<id>[a-zA-Z0-9-]+)''' IE_DESC = 'Wall Street Journal' - _TEST = { + _TESTS = [{ 'url': 'http://video-api.wsj.com/api-video/player/iframe.html?guid=1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A', - 'md5': '9747d7a6ebc2f4df64b981e1dde9efa9', + 'md5': 'e230a5bb249075e40793b655a54a02e4', 'info_dict': { 'id': '1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A', 'ext': 'mp4', @@ -24,65 +30,60 @@ class WSJIE(InfoExtractor): 'duration': 90, 'title': 'Bills Coach Rex Ryan Updates His Old Jets Tattoo', }, - } + }, { + 'url': 'http://www.wsj.com/video/can-alphabet-build-a-smarter-city/359DDAA8-9AC1-489C-82E6-0429C1E430E0.html', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - bitrates = [128, 174, 264, 320, 464, 664, 1264] api_url = ( 'http://video-api.wsj.com/api-video/find_all_videos.asp?' - 'type=guid&count=1&query=%s&' - 'fields=hls,adZone,thumbnailList,guid,state,secondsUntilStartTime,' - 'author,description,name,linkURL,videoStillURL,duration,videoURL,' - 'adCategory,catastrophic,linkShortURL,doctypeID,youtubeID,' - 'titletag,rssURL,wsj-section,wsj-subsection,allthingsd-section,' - 'allthingsd-subsection,sm-section,sm-subsection,provider,' - 'formattedCreationDate,keywords,keywordsOmniture,column,editor,' - 'emailURL,emailPartnerID,showName,omnitureProgramName,' - 'omnitureVideoFormat,linkRelativeURL,touchCastID,' - 'omniturePublishDate,%s') % ( - video_id, ','.join('video%dkMP4Url' % br for br in bitrates)) + 'type=guid&count=1&query=%s&fields=type,hls,videoMP4List,' + 'thumbnailList,author,description,name,duration,videoURL,' + 'titletag,formattedCreationDate,keywords,editor' % video_id) info = self._download_json(api_url, video_id)['items'][0] - - # Thumbnails are conveniently in the correct format already - thumbnails = info.get('thumbnailList') - creator = info.get('author') - uploader_id = info.get('editor') - categories = info.get('keywords') - duration = int_or_none(info.get('duration')) - upload_date = unified_strdate( - info.get('formattedCreationDate'), day_first=False) title = info.get('name', info.get('titletag')) - formats = [{ - 'format_id': 'f4m', - 'format_note': 'f4m (meta URL)', - 'url': info['videoURL'], - }] - if info.get('hls'): + formats = [] + + f4m_url = info.get('videoURL') + if f4m_url: + formats.extend(self._extract_f4m_formats( + f4m_url, video_id, f4m_id='hds', fatal=False)) + + m3u8_url = info.get('hls') + if m3u8_url: formats.extend(self._extract_m3u8_formats( info['hls'], video_id, ext='mp4', - preference=0, entry_protocol='m3u8_native')) - for br in bitrates: - field = 'video%dkMP4Url' % br - if info.get(field): - formats.append({ - 'format_id': 'mp4-%d' % br, - 'container': 'mp4', - 'tbr': br, - 'url': info[field], - }) + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) + + for v in info.get('videoMP4List', []): + mp4_url = v.get('url') + if not mp4_url: + continue + tbr = int_or_none(v.get('bitrate')) + formats.append({ + 'url': mp4_url, + 'format_id': 'http' + ('-%d' % tbr if tbr else ''), + 'tbr': tbr, + 'width': int_or_none(v.get('width')), + 'height': int_or_none(v.get('height')), + 'fps': float_or_none(v.get('fps')), + }) self._sort_formats(formats) return { 'id': video_id, 'formats': formats, - 'thumbnails': thumbnails, - 'creator': creator, - 'uploader_id': uploader_id, - 'duration': duration, - 'upload_date': upload_date, + # Thumbnails are conveniently in the correct format already + 'thumbnails': info.get('thumbnailList'), + 'creator': info.get('author'), + 'uploader_id': info.get('editor'), + 'duration': int_or_none(info.get('duration')), + 'upload_date': unified_strdate(info.get( + 'formattedCreationDate'), day_first=False), 'title': title, - 'categories': categories, + 'categories': info.get('keywords'), } diff --git a/youtube_dl/extractor/xfileshare.py b/youtube_dl/extractor/xfileshare.py index 2d1504eaa..995aada0d 100644 --- a/youtube_dl/extractor/xfileshare.py +++ b/youtube_dl/extractor/xfileshare.py @@ -5,29 +5,44 @@ import re from .common import InfoExtractor from ..utils import ( + decode_packed_codes, ExtractorError, int_or_none, + NO_DEFAULT, sanitized_Request, urlencode_postdata, ) class XFileShareIE(InfoExtractor): - IE_DESC = 'XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me' - _VALID_URL = r'''(?x) - https?://(?P<host>(?:www\.)? - (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com|vidto\.me|powerwatch\.pw))/ - (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)? - ''' - - _FILE_NOT_FOUND_REGEX = r'>(?:404 - )?File Not Found<' + _SITES = ( + ('daclips.in', 'DaClips'), + ('filehoot.com', 'FileHoot'), + ('gorillavid.in', 'GorillaVid'), + ('movpod.in', 'MovPod'), + ('powerwatch.pw', 'PowerWatch'), + ('rapidvideo.ws', 'Rapidvideo.ws'), + ('thevideobee.to', 'TheVideoBee'), + ('vidto.me', 'Vidto'), + ('streamin.to', 'Streamin.To'), + ('xvidstage.com', 'XVIDSTAGE'), + ) + + IE_DESC = 'XFileShare based sites: %s' % ', '.join(list(zip(*_SITES))[1]) + _VALID_URL = (r'https?://(?P<host>(?:www\.)?(?:%s))/(?:embed-)?(?P<id>[0-9a-zA-Z]+)' + % '|'.join(re.escape(site) for site in list(zip(*_SITES))[0])) + + _FILE_NOT_FOUND_REGEXES = ( + r'>(?:404 - )?File Not Found<', + r'>The file was removed by administrator<', + ) _TESTS = [{ 'url': 'http://gorillavid.in/06y9juieqpmi', 'md5': '5ae4a3580620380619678ee4875893ba', 'info_dict': { 'id': '06y9juieqpmi', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Rebecca Black My Moment Official Music Video Reaction-6GK87Rc8bzQ', 'thumbnail': 're:http://.*\.jpg', }, @@ -44,25 +59,6 @@ class XFileShareIE(InfoExtractor): 'thumbnail': 're:http://.*\.jpg', } }, { - # video with countdown timeout - 'url': 'http://fastvideo.in/1qmdn1lmsmbw', - 'md5': '8b87ec3f6564a3108a0e8e66594842ba', - 'info_dict': { - 'id': '1qmdn1lmsmbw', - 'ext': 'mp4', - 'title': 'Man of Steel - Trailer', - 'thumbnail': 're:http://.*\.jpg', - }, - }, { - 'url': 'http://realvid.net/ctn2y6p2eviw', - 'md5': 'b2166d2cf192efd6b6d764c18fd3710e', - 'info_dict': { - 'id': 'ctn2y6p2eviw', - 'ext': 'flv', - 'title': 'rdx 1955', - 'thumbnail': 're:http://.*\.jpg', - }, - }, { 'url': 'http://movpod.in/0wguyyxi1yca', 'only_matching': True, }, { @@ -72,7 +68,8 @@ class XFileShareIE(InfoExtractor): 'ext': 'mp4', 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', 'thumbnail': 're:http://.*\.jpg', - } + }, + 'skip': 'Video removed', }, { 'url': 'http://vidto.me/ku5glz52nqe1.html', 'info_dict': { @@ -87,6 +84,17 @@ class XFileShareIE(InfoExtractor): 'ext': 'mp4', 'title': 'Big Buck Bunny trailer', }, + }, { + 'url': 'http://xvidstage.com/e0qcnl03co6z', + 'info_dict': { + 'id': 'e0qcnl03co6z', + 'ext': 'mp4', + 'title': 'Chucky Prank 2015.mp4', + }, + }, { + # removed by administrator + 'url': 'http://xvidstage.com/amfy7atlkx25', + 'only_matching': True, }] def _real_extract(self, url): @@ -96,7 +104,7 @@ class XFileShareIE(InfoExtractor): url = 'http://%s/%s' % (mobj.group('host'), video_id) webpage = self._download_webpage(url, video_id) - if re.search(self._FILE_NOT_FOUND_REGEX, webpage) is not None: + if any(re.search(p, webpage) for p in self._FILE_NOT_FOUND_REGEXES): raise ExtractorError('Video %s does not exist' % video_id, expected=True) fields = self._hidden_inputs(webpage) @@ -122,10 +130,23 @@ class XFileShareIE(InfoExtractor): r'>Watch (.+) ', r'<h2 class="video-page-head">([^<]+)</h2>'], webpage, 'title', default=None) or self._og_search_title(webpage)).strip() - video_url = self._search_regex( - [r'file\s*:\s*["\'](http[^"\']+)["\'],', - r'file_link\s*=\s*\'(https?:\/\/[0-9a-zA-z.\/\-_]+)'], - webpage, 'file url') + + def extract_video_url(default=NO_DEFAULT): + return self._search_regex( + (r'file\s*:\s*(["\'])(?P<url>http.+?)\1,', + r'file_link\s*=\s*(["\'])(?P<url>http.+?)\1', + r'addVariable\((\\?["\'])file\1\s*,\s*(\\?["\'])(?P<url>http.+?)\2\)', + r'<embed[^>]+src=(["\'])(?P<url>http.+?)\1'), + webpage, 'file url', default=default, group='url') + + video_url = extract_video_url(default=None) + + if not video_url: + webpage = decode_packed_codes(self._search_regex( + r"(}\('(.+)',(\d+),(\d+),'[^']*\b(?:file|embed)\b[^']*'\.split\('\|'\))", + webpage, 'packed code')) + video_url = extract_video_url() + thumbnail = self._search_regex( r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None) diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index b3547174d..bd8e1af2e 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -12,37 +12,52 @@ from ..utils import ( class XHamsterIE(InfoExtractor): - _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.+?)\.html(?:\?.*)?' - _TESTS = [ - { - 'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html', - 'info_dict': { - 'id': '1509445', - 'ext': 'mp4', - 'title': 'FemaleAgent Shy beauty takes the bait', - 'upload_date': '20121014', - 'uploader': 'Ruseful2011', - 'duration': 893.52, - 'age_limit': 18, - } + _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.*?)\.html(?:\?.*)?' + _TESTS = [{ + 'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html', + 'md5': '8281348b8d3c53d39fffb377d24eac4e', + 'info_dict': { + 'id': '1509445', + 'ext': 'mp4', + 'title': 'FemaleAgent Shy beauty takes the bait', + 'upload_date': '20121014', + 'uploader': 'Ruseful2011', + 'duration': 893.52, + 'age_limit': 18, }, - { - 'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd', - 'info_dict': { - 'id': '2221348', - 'ext': 'mp4', - 'title': 'Britney Spears Sexy Booty', - 'upload_date': '20130914', - 'uploader': 'jojo747400', - 'duration': 200.48, - 'age_limit': 18, - } + }, { + 'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd', + 'info_dict': { + 'id': '2221348', + 'ext': 'mp4', + 'title': 'Britney Spears Sexy Booty', + 'upload_date': '20130914', + 'uploader': 'jojo747400', + 'duration': 200.48, + 'age_limit': 18, + }, + 'params': { + 'skip_download': True, + }, + }, { + # empty seo + 'url': 'http://xhamster.com/movies/5667973/.html', + 'info_dict': { + 'id': '5667973', + 'ext': 'mp4', + 'title': '....', + 'upload_date': '20160208', + 'uploader': 'parejafree', + 'duration': 72.0, + 'age_limit': 18, }, - { - 'url': 'https://xhamster.com/movies/2272726/amber_slayed_by_the_knight.html', - 'only_matching': True, + 'params': { + 'skip_download': True, }, - ] + }, { + 'url': 'https://xhamster.com/movies/2272726/amber_slayed_by_the_knight.html', + 'only_matching': True, + }] def _real_extract(self, url): def extract_video_url(webpage, name): @@ -170,7 +185,7 @@ class XHamsterEmbedIE(InfoExtractor): webpage = self._download_webpage(url, video_id) video_url = self._search_regex( - r'href="(https?://xhamster\.com/movies/%s/[^"]+\.html[^"]*)"' % video_id, + r'href="(https?://xhamster\.com/movies/%s/[^"]*\.html[^"]*)"' % video_id, webpage, 'xhamster url', default=None) if not video_url: diff --git a/youtube_dl/extractor/xiami.py b/youtube_dl/extractor/xiami.py new file mode 100644 index 000000000..a6dfc4af9 --- /dev/null +++ b/youtube_dl/extractor/xiami.py @@ -0,0 +1,168 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_urllib_parse_unquote +from ..utils import int_or_none + + +class XiamiBaseIE(InfoExtractor): + _API_BASE_URL = 'http://www.xiami.com/song/playlist/cat/json/id' + + def _download_webpage(self, *args, **kwargs): + webpage = super(XiamiBaseIE, self)._download_webpage(*args, **kwargs) + if '>Xiami is currently not available in your country.<' in webpage: + self.raise_geo_restricted('Xiami is currently not available in your country') + + def _extract_track(self, track, track_id=None): + title = track['title'] + track_url = self._decrypt(track['location']) + + subtitles = {} + lyrics_url = track.get('lyric_url') or track.get('lyric') + if lyrics_url and lyrics_url.startswith('http'): + subtitles['origin'] = [{'url': lyrics_url}] + + return { + 'id': track.get('song_id') or track_id, + 'url': track_url, + 'title': title, + 'thumbnail': track.get('pic') or track.get('album_pic'), + 'duration': int_or_none(track.get('length')), + 'creator': track.get('artist', '').split(';')[0], + 'track': title, + 'album': track.get('album_name'), + 'artist': track.get('artist'), + 'subtitles': subtitles, + } + + def _extract_tracks(self, item_id, typ=None): + playlist = self._download_json( + '%s/%s%s' % (self._API_BASE_URL, item_id, '/type/%s' % typ if typ else ''), item_id) + return [ + self._extract_track(track, item_id) + for track in playlist['data']['trackList']] + + @staticmethod + def _decrypt(origin): + n = int(origin[0]) + origin = origin[1:] + short_lenth = len(origin) // n + long_num = len(origin) - short_lenth * n + l = tuple() + for i in range(0, n): + length = short_lenth + if i < long_num: + length += 1 + l += (origin[0:length], ) + origin = origin[length:] + ans = '' + for i in range(0, short_lenth + 1): + for j in range(0, n): + if len(l[j]) > i: + ans += l[j][i] + return compat_urllib_parse_unquote(ans).replace('^', '0') + + +class XiamiSongIE(XiamiBaseIE): + IE_NAME = 'xiami:song' + IE_DESC = '虾米音乐' + _VALID_URL = r'https?://(?:www\.)?xiami\.com/song/(?P<id>[0-9]+)' + _TESTS = [{ + 'url': 'http://www.xiami.com/song/1775610518', + 'md5': '521dd6bea40fd5c9c69f913c232cb57e', + 'info_dict': { + 'id': '1775610518', + 'ext': 'mp3', + 'title': 'Woman', + 'thumbnail': r're:http://img\.xiami\.net/images/album/.*\.jpg', + 'duration': 265, + 'creator': 'HONNE', + 'track': 'Woman', + 'album': 'Woman', + 'artist': 'HONNE', + 'subtitles': { + 'origin': [{ + 'ext': 'lrc', + }], + }, + }, + 'skip': 'Georestricted', + }, { + 'url': 'http://www.xiami.com/song/1775256504', + 'md5': '932a3abd45c6aa2b1fdbe028fcb4c4fc', + 'info_dict': { + 'id': '1775256504', + 'ext': 'mp3', + 'title': '悟空', + 'thumbnail': r're:http://img\.xiami\.net/images/album/.*\.jpg', + 'duration': 200, + 'creator': '戴荃', + 'track': '悟空', + 'album': '悟空', + 'artist': '戴荃', + 'subtitles': { + 'origin': [{ + 'ext': 'lrc', + }], + }, + }, + 'skip': 'Georestricted', + }] + + def _real_extract(self, url): + return self._extract_tracks(self._match_id(url))[0] + + +class XiamiPlaylistBaseIE(XiamiBaseIE): + def _real_extract(self, url): + item_id = self._match_id(url) + return self.playlist_result(self._extract_tracks(item_id, self._TYPE), item_id) + + +class XiamiAlbumIE(XiamiPlaylistBaseIE): + IE_NAME = 'xiami:album' + IE_DESC = '虾米音乐 - 专辑' + _VALID_URL = r'https?://(?:www\.)?xiami\.com/album/(?P<id>[0-9]+)' + _TYPE = '1' + _TESTS = [{ + 'url': 'http://www.xiami.com/album/2100300444', + 'info_dict': { + 'id': '2100300444', + }, + 'playlist_count': 10, + 'skip': 'Georestricted', + }, { + 'url': 'http://www.xiami.com/album/512288?spm=a1z1s.6843761.1110925389.6.hhE9p9', + 'only_matching': True, + }] + + +class XiamiArtistIE(XiamiPlaylistBaseIE): + IE_NAME = 'xiami:artist' + IE_DESC = '虾米音乐 - 歌手' + _VALID_URL = r'https?://(?:www\.)?xiami\.com/artist/(?P<id>[0-9]+)' + _TYPE = '2' + _TEST = { + 'url': 'http://www.xiami.com/artist/2132?spm=0.0.0.0.dKaScp', + 'info_dict': { + 'id': '2132', + }, + 'playlist_count': 20, + 'skip': 'Georestricted', + } + + +class XiamiCollectionIE(XiamiPlaylistBaseIE): + IE_NAME = 'xiami:collection' + IE_DESC = '虾米音乐 - 精选集' + _VALID_URL = r'https?://(?:www\.)?xiami\.com/collect/(?P<id>[0-9]+)' + _TYPE = '3' + _TEST = { + 'url': 'http://www.xiami.com/collect/156527391?spm=a1z1s.2943601.6856193.12.4jpBnr', + 'info_dict': { + 'id': '156527391', + }, + 'playlist_mincount': 29, + 'skip': 'Georestricted', + } diff --git a/youtube_dl/extractor/xminus.py b/youtube_dl/extractor/xminus.py index 7c9d8af6f..36e5ead1e 100644 --- a/youtube_dl/extractor/xminus.py +++ b/youtube_dl/extractor/xminus.py @@ -2,15 +2,15 @@ from __future__ import unicode_literals import re +import time from .common import InfoExtractor from ..compat import ( - compat_chr, compat_ord, ) from ..utils import ( int_or_none, - parse_filesize, + parse_duration, ) @@ -22,7 +22,7 @@ class XMinusIE(InfoExtractor): 'info_dict': { 'id': '4542', 'ext': 'mp3', - 'title': 'Леонид Агутин-Песенка шофера', + 'title': 'Леонид Агутин-Песенка шофёра', 'duration': 156, 'tbr': 320, 'filesize_approx': 5900000, @@ -36,38 +36,41 @@ class XMinusIE(InfoExtractor): webpage = self._download_webpage(url, video_id) artist = self._html_search_regex( - r'minus_track\.artist="(.+?)"', webpage, 'artist') + r'<a[^>]+href="/artist/\d+">([^<]+)</a>', webpage, 'artist') title = artist + '-' + self._html_search_regex( - r'minus_track\.title="(.+?)"', webpage, 'title') - duration = int_or_none(self._html_search_regex( - r'minus_track\.dur_sec=\'([0-9]*?)\'', + r'<span[^>]+class="minustrack-full-title(?:\s+[^"]+)?"[^>]*>([^<]+)', webpage, 'title') + duration = parse_duration(self._html_search_regex( + r'<span[^>]+class="player-duration(?:\s+[^"]+)?"[^>]*>([^<]+)', webpage, 'duration', fatal=False)) - filesize_approx = parse_filesize(self._html_search_regex( - r'<div id="finfo"[^>]*>\s*↓\s*([0-9.]+\s*[a-zA-Z][bB])', - webpage, 'approximate filesize', fatal=False)) - tbr = int_or_none(self._html_search_regex( - r'<div class="quality[^"]*"></div>\s*([0-9]+)\s*kbps', - webpage, 'bitrate', fatal=False)) + mobj = re.search( + r'<div[^>]+class="dw-info(?:\s+[^"]+)?"[^>]*>(?P<tbr>\d+)\s*кбит/c\s+(?P<filesize>[0-9.]+)\s*мб</div>', + webpage) + tbr = filesize_approx = None + if mobj: + filesize_approx = float(mobj.group('filesize')) * 1000000 + tbr = float(mobj.group('tbr')) view_count = int_or_none(self._html_search_regex( - r'<div class="quality.*?► ([0-9]+)', + r'<span><[^>]+class="icon-chart-bar".*?>(\d+)</span>', webpage, 'view count', fatal=False)) description = self._html_search_regex( - r'(?s)<div id="song_texts">(.*?)</div><br', + r'(?s)<pre[^>]+id="lyrics-original"[^>]*>(.*?)</pre>', webpage, 'song lyrics', fatal=False) if description: description = re.sub(' *\r *', '\n', description) - enc_token = self._html_search_regex( - r'minus_track\.s?tkn="(.+?)"', webpage, 'enc_token') - token = ''.join( - c if pos == 3 else compat_chr(compat_ord(c) - 1) - for pos, c in enumerate(reversed(enc_token))) - video_url = 'http://x-minus.org/dwlf/%s/%s.mp3' % (video_id, token) + k = self._search_regex( + r'<div[^>]+id="player-bottom"[^>]+data-k="([^"]+)">', webpage, + 'encoded data') + h = time.time() / 3600 + a = sum(map(int, [compat_ord(c) for c in k])) + int(video_id) + h + video_url = 'http://x-minus.me/dl/minus?id=%s&tkn2=%df%d' % (video_id, a, h) return { 'id': video_id, 'title': title, 'url': video_url, + # The extension is unknown until actual downloading + 'ext': 'mp3', 'duration': duration, 'filesize_approx': filesize_approx, 'tbr': tbr, diff --git a/youtube_dl/extractor/xnxx.py b/youtube_dl/extractor/xnxx.py index 5a41f8ffa..bcb140305 100644 --- a/youtube_dl/extractor/xnxx.py +++ b/youtube_dl/extractor/xnxx.py @@ -6,17 +6,23 @@ from ..compat import compat_urllib_parse_unquote class XNXXIE(InfoExtractor): - _VALID_URL = r'^https?://(?:video|www)\.xnxx\.com/video(?P<id>[0-9]+)/(.*)' - _TEST = { - 'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_', - 'md5': '0831677e2b4761795f68d417e0b7b445', + _VALID_URL = r'https?://(?:video|www)\.xnxx\.com/video-?(?P<id>[0-9a-z]+)/' + _TESTS = [{ + 'url': 'http://www.xnxx.com/video-55awb78/skyrim_test_video', + 'md5': 'ef7ecee5af78f8b03dca2cf31341d3a0', 'info_dict': { - 'id': '1135332', + 'id': '55awb78', 'ext': 'flv', - 'title': 'lida » Naked Funny Actress (5)', + 'title': 'Skyrim Test Video', 'age_limit': 18, - } - } + }, + }, { + 'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_', + 'only_matching': True, + }, { + 'url': 'http://www.xnxx.com/video-55awb78/', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 4075b8a4f..83bc1fef2 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -4,17 +4,23 @@ import itertools import re from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote from ..utils import ( int_or_none, orderedSet, + parse_duration, sanitized_Request, str_to_int, ) class XTubeIE(InfoExtractor): - _VALID_URL = r'(?:xtube:|https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-))(?P<id>[^/?&#]+)' + _VALID_URL = r'''(?x) + (?: + xtube:| + https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-) + ) + (?P<id>[^/?&#]+) + ''' _TESTS = [{ # old URL schema @@ -27,6 +33,8 @@ class XTubeIE(InfoExtractor): 'description': 'contains:an ET kind of thing', 'uploader': 'greenshowers', 'duration': 450, + 'view_count': int, + 'comment_count': int, 'age_limit': 18, } }, { @@ -51,21 +59,30 @@ class XTubeIE(InfoExtractor): req.add_header('Cookie', 'age_verified=1; cookiesAccepted=1') webpage = self._download_webpage(req, display_id) - flashvars = self._parse_json( - self._search_regex( - r'xt\.playerOps\s*=\s*({.+?});', webpage, 'player ops'), - video_id)['flashvars'] - - title = flashvars.get('title') or self._search_regex( - r'<h1>([^<]+)</h1>', webpage, 'title') - video_url = compat_urllib_parse_unquote(flashvars['video_url']) - duration = int_or_none(flashvars.get('video_duration')) - - uploader = self._search_regex( - r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"', - webpage, 'uploader', fatal=False) + sources = self._parse_json(self._search_regex( + r'sources\s*:\s*({.+?}),', webpage, 'sources'), video_id) + + formats = [] + for format_id, format_url in sources.items(): + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'height': int_or_none(format_id), + }) + self._sort_formats(formats) + + title = self._search_regex( + (r'<h1>(?P<title>[^<]+)</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'), + webpage, 'title', group='title') description = self._search_regex( r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False) + uploader = self._search_regex( + (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"', + r'<span[^>]+class="nickname"[^>]*>([^<]+)'), + webpage, 'uploader', fatal=False) + duration = parse_duration(self._search_regex( + r'<dt>Runtime:</dt>\s*<dd>([^<]+)</dd>', + webpage, 'duration', fatal=False)) view_count = str_to_int(self._search_regex( r'<dt>Views:</dt>\s*<dd>([\d,\.]+)</dd>', webpage, 'view count', fatal=False)) @@ -76,7 +93,6 @@ class XTubeIE(InfoExtractor): return { 'id': video_id, 'display_id': display_id, - 'url': video_url, 'title': title, 'description': description, 'uploader': uploader, @@ -84,6 +100,7 @@ class XTubeIE(InfoExtractor): 'view_count': view_count, 'comment_count': comment_count, 'age_limit': 18, + 'formats': formats, } diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index 2466410fa..a66daee46 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -66,6 +66,21 @@ class XuiteIE(InfoExtractor): 'uploader_id': '242127761', 'categories': ['電玩動漫'], }, + 'skip': 'Video removed', + }, { + # Video with encoded media id + # from http://forgetfulbc.blogspot.com/2016/06/date.html + 'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0', + 'info_dict': { + 'id': 'cE1xbENoLTI3NDQ3MzM2LmZsdg==', + 'ext': 'mp4', + 'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)', + 'description': 'md5:f0abdcb69df300f522a5442ef3146f2a', + 'timestamp': 1466160960, + 'upload_date': '20160617', + 'uploader': 'B.C. & Lowy', + 'uploader_id': '232279340', + }, }, { 'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9', 'only_matching': True, @@ -79,10 +94,9 @@ class XuiteIE(InfoExtractor): def base64_encode_utf8(data): return base64.b64encode(data.encode('utf-8')).decode('utf-8') - def _extract_flv_config(self, media_id): - base64_media_id = self.base64_encode_utf8(media_id) + def _extract_flv_config(self, encoded_media_id): flv_config = self._download_xml( - 'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id, + 'http://vlog.xuite.net/flash/player?media=%s' % encoded_media_id, 'flv config') prop_dict = {} for prop in flv_config.findall('./property'): @@ -107,9 +121,14 @@ class XuiteIE(InfoExtractor): '%s returned error: %s' % (self.IE_NAME, error_msg), expected=True) - video_id = self._html_search_regex( - r'data-mediaid="(\d+)"', webpage, 'media id') - flv_config = self._extract_flv_config(video_id) + encoded_media_id = self._search_regex( + r'attributes\.name\s*=\s*"([^"]+)"', webpage, + 'encoded media id', default=None) + if encoded_media_id is None: + video_id = self._html_search_regex( + r'data-mediaid="(\d+)"', webpage, 'media id') + encoded_media_id = self.base64_encode_utf8(video_id) + flv_config = self._extract_flv_config(encoded_media_id) FORMATS = { 'audio': 'mp3', diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index 710ad5041..1dfe031ca 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -8,7 +8,6 @@ from ..utils import ( clean_html, ExtractorError, determine_ext, - sanitized_Request, ) @@ -25,8 +24,6 @@ class XVideosIE(InfoExtractor): } } - _ANDROID_USER_AGENT = 'Mozilla/5.0 (Linux; Android 4.0.4; Galaxy Nexus Build/IMM76B) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.133 Mobile Safari/535.19' - def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) @@ -35,31 +32,34 @@ class XVideosIE(InfoExtractor): if mobj: raise ExtractorError('%s said: %s' % (self.IE_NAME, clean_html(mobj.group(1))), expected=True) - video_url = compat_urllib_parse_unquote( - self._search_regex(r'flv_url=(.+?)&', webpage, 'video URL')) video_title = self._html_search_regex( r'<title>(.*?)\s+-\s+XVID', webpage, 'title') video_thumbnail = self._search_regex( r'url_bigthumb=(.+?)&', webpage, 'thumbnail', fatal=False) - formats = [{ - 'url': video_url, - }] + formats = [] - android_req = sanitized_Request(url) - android_req.add_header('User-Agent', self._ANDROID_USER_AGENT) - android_webpage = self._download_webpage(android_req, video_id, fatal=False) + video_url = compat_urllib_parse_unquote(self._search_regex( + r'flv_url=(.+?)&', webpage, 'video URL', default='')) + if video_url: + formats.append({'url': video_url}) - if android_webpage is not None: - player_params_str = self._search_regex( - 'mobileReplacePlayerDivTwoQual\(([^)]+)\)', - android_webpage, 'player parameters', default='') - player_params = list(map(lambda s: s.strip(' \''), player_params_str.split(','))) - if player_params: - formats.extend([{ - 'url': param, - 'preference': -10, - } for param in player_params if determine_ext(param) == 'mp4']) + player_args = self._search_regex( + r'(?s)new\s+HTML5Player\((.+?)\)', webpage, ' html5 player', default=None) + if player_args: + for arg in player_args.split(','): + format_url = self._search_regex( + r'(["\'])(?P<url>https?://.+?)\1', arg, 'url', + default=None, group='url') + if not format_url: + continue + ext = determine_ext(format_url) + if ext == 'mp4': + formats.append({'url': format_url}) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) self._sort_formats(formats) @@ -67,7 +67,6 @@ class XVideosIE(InfoExtractor): 'id': video_id, 'formats': formats, 'title': video_title, - 'ext': 'flv', 'thumbnail': video_thumbnail, 'age_limit': 18, } diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index b2d8f4b48..b0679dfb7 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -19,12 +19,13 @@ from ..utils import ( mimetype2ext, ) +from .brightcove import BrightcoveNewIE from .nbc import NBCSportsVPlayerIE class YahooIE(InfoExtractor): IE_DESC = 'Yahoo screen and movies' - _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+)?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)' + _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+)?-(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?)' _TESTS = [ { 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', @@ -38,7 +39,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - 'md5': 'd6e6fc6e1313c608f316ddad7b82b306', + 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23', 'info_dict': { 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', 'ext': 'mp4', @@ -49,7 +50,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', - 'md5': '60e8ac193d8fb71997caa8fce54c6460', + 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136', 'info_dict': { 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', 'ext': 'mp4', @@ -59,15 +60,15 @@ class YahooIE(InfoExtractor): } }, { - 'url': 'https://tw.screen.yahoo.com/election-2014-askmayor/敢問市長-黃秀霜批賴清德-非常高傲-033009720.html', - 'md5': '3a09cf59349cfaddae1797acc3c087fc', + 'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html', + 'md5': '9035d38f88b1782682a3e89f985be5bb', 'info_dict': { 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', 'ext': 'mp4', 'title': '敢問市長／黃秀霜批賴清德「非常高傲」', 'description': '直言台南沒捷運 交通居五都之末', 'duration': 396, - } + }, }, { 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', @@ -89,17 +90,32 @@ class YahooIE(InfoExtractor): 'title': 'Program that makes hockey more affordable not offered in Manitoba', 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4', 'duration': 121, - } + }, + 'skip': 'Video gone', }, { 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html', - 'md5': '226a895aae7e21b0129e2a2006fe9690', 'info_dict': { - 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', - 'ext': 'mp4', - 'title': '\'The Interview\' TV Spot: War', - 'description': 'The Interview', - 'duration': 30, - } + 'id': '154609075', + }, + 'playlist': [{ + 'md5': 'f8e336c6b66f503282e5f719641d6565', + 'info_dict': { + 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', + 'ext': 'mp4', + 'title': '\'The Interview\' TV Spot: War', + 'description': 'The Interview', + 'duration': 30, + }, + }, { + 'md5': '958bcb90b4d6df71c56312137ee1cd5a', + 'info_dict': { + 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', + 'ext': 'mp4', + 'title': '\'The Interview\' TV Spot: Guys', + 'description': 'The Interview', + 'duration': 30, + }, + }], }, { 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', 'md5': '88e209b417f173d86186bef6e4d1f160', @@ -119,10 +135,11 @@ class YahooIE(InfoExtractor): 'title': 'Connect the Dots: Dark Side of Virgo', 'description': 'md5:1428185051cfd1949807ad4ff6d3686a', 'duration': 201, - } + }, + 'skip': 'Domain name in.lifestyle.yahoo.com gone', }, { 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', - 'md5': '989396ae73d20c6f057746fb226aa215', + 'md5': 'b17ac378b1134fa44370fb27db09a744', 'info_dict': { 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', 'ext': 'mp4', @@ -141,6 +158,9 @@ class YahooIE(InfoExtractor): 'ext': 'flv', 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', + 'upload_date': '20150313', + 'uploader': 'NBCU-SPORTS', + 'timestamp': 1426270238, } }, { 'url': 'https://tw.news.yahoo.com/-100120367.html', @@ -148,7 +168,7 @@ class YahooIE(InfoExtractor): }, { # Query result is embedded in webpage, but explicit request to video API fails with geo restriction 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', - 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', + 'md5': '1ddbf7c850777548438e5c4f147c7b8c', 'info_dict': { 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', 'ext': 'mp4', @@ -166,6 +186,17 @@ class YahooIE(InfoExtractor): 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.', }, }, + { + # config['models']['applet_model']['data']['sapi'] has no query + 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016', + 'md5': 'dac0c72d502bc5facda80c9e6d5c98db', + 'info_dict': { + 'id': 'a6015640-e9e5-3efb-bb60-05589a183919', + 'ext': 'mp4', + 'description': 'Galactic', + 'title': 'Dolla Diva (feat. Maggie Koerner)', + }, + }, ] def _real_extract(self, url): @@ -174,23 +205,35 @@ class YahooIE(InfoExtractor): page_id = mobj.group('id') url = mobj.group('url') host = mobj.group('host') - webpage = self._download_webpage(url, display_id) + webpage, urlh = self._download_webpage_handle(url, display_id) + if 'err=404' in urlh.geturl(): + raise ExtractorError('Video gone', expected=True) # Look for iframed media first - iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) - if iframe_m: + entries = [] + iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) + for idx, iframe_url in enumerate(iframe_urls): iframepage = self._download_webpage( - host + iframe_m.group(1), display_id, 'Downloading iframe webpage') + host + iframe_url, display_id, + note='Downloading iframe webpage for video #%d' % idx) items_json = self._search_regex( r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None) if items_json: items = json.loads(items_json) video_id = items[0]['id'] - return self._get_info(video_id, display_id, webpage) + entries.append(self._get_info(video_id, display_id, webpage)) + if entries: + return self.playlist_result(entries, page_id) + # Look for NBCSports iframes nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage) if nbc_sports_url: - return self.url_result(nbc_sports_url, 'NBCSportsVPlayer') + return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key()) + + # Look for Brightcove New Studio embeds + bc_url = BrightcoveNewIE._extract_url(webpage) + if bc_url: + return self.url_result(bc_url, BrightcoveNewIE.ie_key()) # Query result is often embedded in webpage as JSON. Sometimes explicit requests # to video API results in a failure with geo restriction reason therefore using @@ -202,7 +245,7 @@ class YahooIE(InfoExtractor): config = self._parse_json(config_json, display_id, fatal=False) if config: sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') - if sapi: + if sapi and 'query' in sapi: return self._extract_info(display_id, sapi, webpage) items_json = self._search_regex( @@ -306,7 +349,7 @@ class YahooIE(InfoExtractor): webpage, 'region', fatal=False, default='US') data = compat_urllib_parse_urlencode({ 'protocol': 'http', - 'region': region, + 'region': region.upper(), }) query_url = ( 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index 7a90cc60c..b37d0eab6 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -10,17 +10,35 @@ from ..utils import ( ExtractorError, int_or_none, float_or_none, - sanitized_Request, - urlencode_postdata, ) class YandexMusicBaseIE(InfoExtractor): @staticmethod def _handle_error(response): - error = response.get('error') - if error: - raise ExtractorError(error, expected=True) + if isinstance(response, dict): + error = response.get('error') + if error: + raise ExtractorError(error, expected=True) + if response.get('type') == 'captcha' or 'captcha' in response: + YandexMusicBaseIE._raise_captcha() + + @staticmethod + def _raise_captcha(): + raise ExtractorError( + 'YandexMusic has considered youtube-dl requests automated and ' + 'asks you to solve a CAPTCHA. You can either wait for some ' + 'time until unblocked and optionally use --sleep-interval ' + 'in future or alternatively you can go to https://music.yandex.ru/ ' + 'solve CAPTCHA, then export cookies and pass cookie file to ' + 'youtube-dl with --cookies', + expected=True) + + def _download_webpage(self, *args, **kwargs): + webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs) + if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage: + self._raise_captcha() + return webpage def _download_json(self, *args, **kwargs): response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs) @@ -47,7 +65,8 @@ class YandexMusicTrackIE(YandexMusicBaseIE): 'album_artist': 'Carlo Ambrosio', 'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio', 'release_year': '2009', - } + }, + 'skip': 'Travis CI servers blocked by YandexMusic', } def _get_track_url(self, storage_dir, track_id): @@ -139,6 +158,7 @@ class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE): 'title': 'Carlo Ambrosio - Gypsy Soul (2009)', }, 'playlist_count': 50, + 'skip': 'Travis CI servers blocked by YandexMusic', } def _real_extract(self, url): @@ -161,7 +181,7 @@ class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE): class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE): IE_NAME = 'yandexmusic:playlist' IE_DESC = 'Яндекс.Музыка - Плейлист' - _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)' + _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)' _TESTS = [{ 'url': 'http://music.yandex.ru/users/music.partners/playlists/1245', @@ -171,6 +191,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE): 'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9', }, 'playlist_count': 6, + 'skip': 'Travis CI servers blocked by YandexMusic', }, { # playlist exceeding the limit of 150 tracks shipped with webpage (see # https://github.com/rg3/youtube-dl/issues/6666) @@ -179,46 +200,64 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE): 'id': '1036', 'title': 'Музыка 90-х', }, - 'playlist_count': 310, + 'playlist_mincount': 300, + 'skip': 'Travis CI servers blocked by YandexMusic', }] def _real_extract(self, url): - playlist_id = self._match_id(url) - - webpage = self._download_webpage(url, playlist_id) - - mu = self._parse_json( - self._search_regex( - r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'), - playlist_id) - - playlist = mu['pageData']['playlist'] - tracks, track_ids = playlist['tracks'], playlist['trackIds'] - - # tracks dictionary shipped with webpage is limited to 150 tracks, + mobj = re.match(self._VALID_URL, url) + tld = mobj.group('tld') + user = mobj.group('user') + playlist_id = mobj.group('id') + + playlist = self._download_json( + 'https://music.yandex.%s/handlers/playlist.jsx' % tld, + playlist_id, 'Downloading missing tracks JSON', + fatal=False, + headers={ + 'Referer': url, + 'X-Requested-With': 'XMLHttpRequest', + 'X-Retpath-Y': url, + }, + query={ + 'owner': user, + 'kinds': playlist_id, + 'light': 'true', + 'lang': tld, + 'external-domain': 'music.yandex.%s' % tld, + 'overembed': 'false', + })['playlist'] + + tracks, track_ids = playlist['tracks'], map(compat_str, playlist['trackIds']) + + # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks, # missing tracks should be retrieved manually. if len(tracks) < len(track_ids): - present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')]) - missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids) - request = sanitized_Request( - 'https://music.yandex.ru/handlers/track-entries.jsx', - urlencode_postdata({ + present_track_ids = set([ + compat_str(track['id']) + for track in tracks if track.get('id')]) + missing_track_ids = [ + track_id for track_id in track_ids + if track_id not in present_track_ids] + missing_tracks = self._download_json( + 'https://music.yandex.%s/handlers/track-entries.jsx' % tld, + playlist_id, 'Downloading missing tracks JSON', + fatal=False, + headers={ + 'Referer': url, + 'X-Requested-With': 'XMLHttpRequest', + }, + query={ 'entries': ','.join(missing_track_ids), - 'lang': mu.get('settings', {}).get('lang', 'en'), - 'external-domain': 'music.yandex.ru', + 'lang': tld, + 'external-domain': 'music.yandex.%s' % tld, 'overembed': 'false', - 'sign': mu.get('authData', {}).get('user', {}).get('sign'), 'strict': 'true', - })) - request.add_header('Referer', url) - request.add_header('X-Requested-With', 'XMLHttpRequest') - - missing_tracks = self._download_json( - request, playlist_id, 'Downloading missing tracks JSON', fatal=False) + }) if missing_tracks: tracks.extend(missing_tracks) return self.playlist_result( self._build_playlist(tracks), compat_str(playlist_id), - playlist['title'], playlist.get('description')) + playlist.get('title'), playlist.get('description')) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 349ce0941..e37f237c7 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -2,7 +2,9 @@ from __future__ import unicode_literals import base64 +import itertools import random +import re import string import time @@ -13,7 +15,7 @@ from ..compat import ( ) from ..utils import ( ExtractorError, - sanitized_Request, + get_element_by_attribute, ) @@ -215,14 +217,10 @@ class YoukuIE(InfoExtractor): headers = { 'Referer': req_url, } + headers.update(self.geo_verification_headers()) self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com') - req = sanitized_Request(req_url, headers=headers) - cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') - if cn_verification_proxy: - req.add_header('Ytdl-request-proxy', cn_verification_proxy) - - raw_data = self._download_json(req, video_id, note=note) + raw_data = self._download_json(req_url, video_id, note=note, headers=headers) return raw_data['data'] @@ -275,6 +273,8 @@ class YoukuIE(InfoExtractor): 'format_id': self.get_format_name(fm), 'ext': self.parse_ext_l(fm), 'filesize': int(seg['size']), + 'width': stream.get('width'), + 'height': stream.get('height'), }) return { @@ -283,3 +283,52 @@ class YoukuIE(InfoExtractor): 'title': title, 'entries': entries, } + + +class YoukuShowIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?youku\.com/show_page/id_(?P<id>[0-9a-z]+)\.html' + IE_NAME = 'youku:show' + + _TEST = { + 'url': 'http://www.youku.com/show_page/id_zc7c670be07ff11e48b3f.html', + 'info_dict': { + 'id': 'zc7c670be07ff11e48b3f', + 'title': '花千骨 未删减版', + 'description': 'md5:578d4f2145ae3f9128d9d4d863312910', + }, + 'playlist_count': 50, + } + + _PAGE_SIZE = 40 + + def _find_videos_in_page(self, webpage): + videos = re.findall( + r'<li><a[^>]+href="(?P<url>https?://v\.youku\.com/[^"]+)"[^>]+title="(?P<title>[^"]+)"', webpage) + return [ + self.url_result(video_url, YoukuIE.ie_key(), title) + for video_url, title in videos] + + def _real_extract(self, url): + show_id = self._match_id(url) + webpage = self._download_webpage(url, show_id) + + entries = self._find_videos_in_page(webpage) + + playlist_title = self._html_search_regex( + r'<span[^>]+class="name">([^<]+)</span>', webpage, 'playlist title', fatal=False) + detail_div = get_element_by_attribute('class', 'detail', webpage) or '' + playlist_description = self._html_search_regex( + r'<span[^>]+style="display:none"[^>]*>([^<]+)</span>', + detail_div, 'playlist description', fatal=False) + + for idx in itertools.count(1): + episodes_page = self._download_webpage( + 'http://www.youku.com/show_episode/id_%s.html' % show_id, + show_id, query={'divid': 'reload_%d' % (idx * self._PAGE_SIZE + 1)}, + note='Downloading episodes page %d' % idx) + new_entries = self._find_videos_in_page(episodes_page) + entries.extend(new_entries) + if len(new_entries) < self._PAGE_SIZE: + break + + return self.playlist_result(entries, show_id, playlist_title, playlist_description) diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 1124fe6c2..0df2d76ee 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -17,7 +17,7 @@ class YouPornIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?youporn\.com/watch/(?P<id>\d+)/(?P<display_id>[^/?#&]+)' _TESTS = [{ 'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/', - 'md5': '71ec5fcfddacf80f495efa8b6a8d9a89', + 'md5': '3744d24c50438cf5b6f6d59feb5055c2', 'info_dict': { 'id': '505835', 'display_id': 'sex-ed-is-it-safe-to-masturbate-daily', @@ -121,21 +121,21 @@ class YouPornIE(InfoExtractor): webpage, 'thumbnail', fatal=False, group='thumbnail') uploader = self._html_search_regex( - r'(?s)<div[^>]+class=["\']videoInfoBy(?:\s+[^"\']+)?["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', + r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>', webpage, 'uploader', fatal=False) upload_date = unified_strdate(self._html_search_regex( - r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>', + r'(?s)<div[^>]+class=["\']videoInfo(?:Date|Time)["\'][^>]*>(.+?)</div>', webpage, 'upload date', fatal=False)) age_limit = self._rta_search(webpage) average_rating = int_or_none(self._search_regex( - r'<div[^>]+class=["\']videoInfoRating["\'][^>]*>\s*<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>', + r'<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>', webpage, 'average rating', fatal=False)) view_count = str_to_int(self._search_regex( - r'(?s)<div[^>]+class=["\']videoInfoViews["\'][^>]*>.*?([\d,.]+)\s*</div>', - webpage, 'view count', fatal=False)) + r'(?s)<div[^>]+class=(["\']).*?\bvideoInfoViews\b.*?\1[^>]*>.*?(?P<count>[\d,.]+)<', + webpage, 'view count', fatal=False, group='count')) comment_count = str_to_int(self._search_regex( r'>All [Cc]omments? \(([\d,.]+)\)', webpage, 'comment count', fatal=False)) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 44f98d294..8aa7dfc41 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -344,6 +344,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, + '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'}, + '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'}, # Dash webm '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, @@ -499,6 +501,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'youtube_include_dash_manifest': True, 'format': '141', }, + 'skip': 'format 141 not served anymore', }, # DASH manifest with encrypted signature { @@ -515,7 +518,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, 'params': { 'youtube_include_dash_manifest': True, - 'format': '141', + 'format': '141/bestaudio[ext=m4a]', }, }, # JS player signature function name containing $ @@ -535,7 +538,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, 'params': { 'youtube_include_dash_manifest': True, - 'format': '141', + 'format': '141/bestaudio[ext=m4a]', }, }, # Controversy video @@ -616,7 +619,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic', 'license': 'Standard YouTube License', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', - 'uploader': 'Olympics', + 'uploader': 'Olympic', 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', }, 'params': { @@ -669,7 +672,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000', 'uploader': 'dorappi2000', 'license': 'Standard YouTube License', - 'formats': 'mincount:33', + 'formats': 'mincount:32', }, }, # DASH manifest with segment_list @@ -689,7 +692,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'youtube_include_dash_manifest': True, 'format': '135', # bestvideo - } + }, + 'skip': 'This live event has ended.', }, { # Multifeed videos (multiple cameras), URL is for Main Camera @@ -760,6 +764,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30', }, 'playlist_count': 2, + 'skip': 'Not multifeed anymore', }, { 'url': 'http://vid.plus/FlRa-iH7PGw', @@ -812,6 +817,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'This video does not exist.', }, { # Video licensed under Creative Commons @@ -1326,10 +1332,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if video_description: video_description = re.sub(r'''(?x) <a\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? + (?:[a-zA-Z-]+="[^"]*"\s+)*? (?:title|href)="([^"]+)"\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? - class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*> + (?:[a-zA-Z-]+="[^"]*"\s+)*? + class="[^"]*"[^>]*> [^<]+\.{3}\s* </a> ''', r'\1', video_description) @@ -1724,6 +1730,39 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } +class YoutubeSharedVideoIE(InfoExtractor): + _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?ci=(?P<id>[0-9A-Za-z_-]{11})' + IE_NAME = 'youtube:shared' + + _TEST = { + 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU', + 'info_dict': { + 'id': 'uPDB5I9wfp8', + 'ext': 'webm', + 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3', + 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d', + 'upload_date': '20160219', + 'uploader': 'Pocoyo - Português (BR)', + 'uploader_id': 'PocoyoBrazil', + }, + 'add_ie': ['Youtube'], + 'params': { + # There are already too many Youtube downloads + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + real_video_id = self._html_search_meta( + 'videoId', webpage, 'YouTube video id', fatal=True) + + return self.url_result(real_video_id, YoutubeIE.ie_key()) + + class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: @@ -1939,10 +1978,13 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)) + def _build_template_url(self, url, channel_id): + return self._TEMPLATE_URL % channel_id + def _real_extract(self, url): channel_id = self._match_id(url) - url = self._TEMPLATE_URL % channel_id + url = self._build_template_url(url, channel_id) # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778) # Workaround by extracting as a playlist if managed to obtain channel playlist URL @@ -1956,9 +1998,13 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): channel_playlist_id = self._html_search_meta( 'channelId', channel_page, 'channel id', default=None) if not channel_playlist_id: - channel_playlist_id = self._search_regex( - r'data-(?:channel-external-|yt)id="([^"]+)"', - channel_page, 'channel id', default=None) + channel_url = self._html_search_meta( + ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'), + channel_page, 'channel url', default=None) + if channel_url: + channel_playlist_id = self._search_regex( + r'vnd\.youtube://user/([0-9A-Za-z_-]+)', + channel_url, 'channel id', default=None) if channel_playlist_id and channel_playlist_id.startswith('UC'): playlist_id = 'UU' + channel_playlist_id[2:] return self.url_result( @@ -1981,24 +2027,53 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): for video_id, video_title in self.extract_videos_from_page(channel_page)] return self.playlist_result(entries, channel_id) + try: + next(self._entries(channel_page, channel_id)) + except StopIteration: + alert_message = self._html_search_regex( + r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>', + channel_page, 'alert', default=None, group='alert') + if alert_message: + raise ExtractorError('Youtube said: %s' % alert_message, expected=True) + return self.playlist_result(self._entries(channel_page, channel_id), channel_id) class YoutubeUserIE(YoutubeChannelIE): IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' - _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' - _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' + _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos' IE_NAME = 'youtube:user' _TESTS = [{ 'url': 'https://www.youtube.com/user/TheLinuxFoundation', 'playlist_mincount': 320, 'info_dict': { - 'title': 'TheLinuxFoundation', + 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ', + 'title': 'Uploads from The Linux Foundation', + } + }, { + # Only available via https://www.youtube.com/c/12minuteathlete/videos + # but not https://www.youtube.com/user/12minuteathlete/videos + 'url': 'https://www.youtube.com/c/12minuteathlete/videos', + 'playlist_mincount': 249, + 'info_dict': { + 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ', + 'title': 'Uploads from 12 Minute Athlete', } }, { 'url': 'ytuser:phihag', 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/c/gametrailers', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/gametrailers', + 'only_matching': True, + }, { + # This channel is not available. + 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos', + 'only_matching': True, }] @classmethod @@ -2011,6 +2086,10 @@ class YoutubeUserIE(YoutubeChannelIE): else: return super(YoutubeUserIE, cls).suitable(url) + def _build_template_url(self, url, channel_id): + mobj = re.match(self._VALID_URL, url) + return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id')) + class YoutubeLiveIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com live streams' @@ -2139,10 +2218,11 @@ class YoutubeSearchDateIE(YoutubeSearchIE): _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'} -class YoutubeSearchURLIE(InfoExtractor): +class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com search URLs' IE_NAME = 'youtube:search_url' _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?' _TESTS = [{ 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', 'playlist_mincount': 5, @@ -2157,32 +2237,8 @@ class YoutubeSearchURLIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) query = compat_urllib_parse_unquote_plus(mobj.group('query')) - webpage = self._download_webpage(url, query) - result_code = self._search_regex( - r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML') - - part_codes = re.findall( - r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code) - entries = [] - for part_code in part_codes: - part_title = self._html_search_regex( - [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) - part_url_snippet = self._html_search_regex( - r'(?s)href="([^"]+)"', part_code, 'item URL') - part_url = compat_urlparse.urljoin( - 'https://www.youtube.com/', part_url_snippet) - entries.append({ - '_type': 'url', - 'url': part_url, - 'title': part_title, - }) - - return { - '_type': 'playlist', - 'entries': entries, - 'title': query, - } + return self.playlist_result(self._process_page(webpage), playlist_title=query) class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor): diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index a7440c582..9737f7002 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -232,7 +232,7 @@ class JSInterpreter(object): def extract_function(self, funcname): func_m = re.search( r'''(?x) - (?:function\s+%s|[{;,]%s\s*=\s*function|var\s+%s\s*=\s*function)\s* + (?:function\s+%s|[{;,]\s*%s\s*=\s*function|var\s+%s\s*=\s*function)\s* \((?P<args>[^)]*)\)\s* \{(?P<code>[^}]+)\}''' % ( re.escape(funcname), re.escape(funcname), re.escape(funcname)), diff --git a/youtube_dl/options.py b/youtube_dl/options.py index d1f8d1331..c4a85b2c0 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -26,9 +26,11 @@ def parseOpts(overrideArguments=None): except IOError: return default # silently skip if file is not present try: - res = [] - for l in optionf: - res += compat_shlex_split(l, comments=True) + # FIXME: https://github.com/rg3/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56 + contents = optionf.read() + if sys.version_info < (3,): + contents = contents.decode(preferredencoding()) + res = compat_shlex_split(contents, comments=True) finally: optionf.close() return res @@ -188,7 +190,10 @@ def parseOpts(overrideArguments=None): network.add_option( '--proxy', dest='proxy', default=None, metavar='URL', - help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection') + help='Use the specified HTTP/HTTPS/SOCKS proxy. To enable experimental ' + 'SOCKS proxy, specify a proper scheme. For example ' + 'socks5://127.0.0.1:1080/. Pass in an empty string (--proxy "") ' + 'for direct connection') network.add_option( '--socket-timeout', dest='socket_timeout', type=float, default=None, metavar='SECONDS', @@ -209,10 +214,15 @@ def parseOpts(overrideArguments=None): help='Make all connections via IPv6 (experimental)', ) network.add_option( + '--geo-verification-proxy', + dest='geo_verification_proxy', default=None, metavar='URL', + help='Use this proxy to verify the IP address for some geo-restricted sites. ' + 'The default proxy specified by --proxy (or none, if the options is not present) is used for the actual downloading. (experimental)' + ) + network.add_option( '--cn-verification-proxy', dest='cn_verification_proxy', default=None, metavar='URL', - help='Use this proxy to verify the IP address for some Chinese sites. ' - 'The default proxy specified by --proxy (or none, if the options is not present) is used for the actual downloading. (experimental)' + help=optparse.SUPPRESS_HELP, ) selection = optparse.OptionGroup(parser, 'Video Selection') @@ -392,8 +402,8 @@ def parseOpts(overrideArguments=None): downloader = optparse.OptionGroup(parser, 'Download Options') downloader.add_option( - '-r', '--rate-limit', - dest='ratelimit', metavar='LIMIT', + '-r', '--limit-rate', '--rate-limit', + dest='ratelimit', metavar='RATE', help='Maximum download rate in bytes per second (e.g. 50K or 4.2M)') downloader.add_option( '-R', '--retries', @@ -665,7 +675,7 @@ def parseOpts(overrideArguments=None): action='store_true', dest='writeannotations', default=False, help='Write video annotations to a .annotations.xml file') filesystem.add_option( - '--load-info', + '--load-info-json', '--load-info', dest='load_info_filename', metavar='FILE', help='JSON file containing the video information (created with the "--write-info-json" option)') filesystem.add_option( @@ -806,11 +816,11 @@ def parseOpts(overrideArguments=None): system_conf = [] user_conf = [] else: - system_conf = compat_conf(_readOptions('/etc/youtube-dl.conf')) + system_conf = _readOptions('/etc/youtube-dl.conf') if '--ignore-config' in system_conf: user_conf = [] else: - user_conf = compat_conf(_readUserConf()) + user_conf = _readUserConf() argv = system_conf + user_conf + command_line_conf opts, args = parser.parse_args(argv) diff --git a/youtube_dl/postprocessor/execafterdownload.py b/youtube_dl/postprocessor/execafterdownload.py index 74f66d669..90630c2d7 100644 --- a/youtube_dl/postprocessor/execafterdownload.py +++ b/youtube_dl/postprocessor/execafterdownload.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import subprocess from .common import PostProcessor -from ..compat import shlex_quote +from ..compat import compat_shlex_quote from ..utils import PostProcessingError @@ -17,7 +17,7 @@ class ExecAfterDownloadPP(PostProcessor): if '{}' not in cmd: cmd += ' {}' - cmd = cmd.replace('{}', shlex_quote(information['filepath'])) + cmd = cmd.replace('{}', compat_shlex_quote(information['filepath'])) self._downloader.to_screen('[exec] Executing command: %s' % cmd) retCode = subprocess.call(cmd, shell=True) diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index ca2d401f8..c1e9eb159 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -391,23 +391,30 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor): class FFmpegMetadataPP(FFmpegPostProcessor): def run(self, info): metadata = {} - if info.get('title') is not None: - metadata['title'] = info['title'] - if info.get('upload_date') is not None: - metadata['date'] = info['upload_date'] - if info.get('artist') is not None: - metadata['artist'] = info['artist'] - elif info.get('uploader') is not None: - metadata['artist'] = info['uploader'] - elif info.get('uploader_id') is not None: - metadata['artist'] = info['uploader_id'] - if info.get('description') is not None: - metadata['description'] = info['description'] - metadata['comment'] = info['description'] - if info.get('webpage_url') is not None: - metadata['purl'] = info['webpage_url'] - if info.get('album') is not None: - metadata['album'] = info['album'] + + def add(meta_list, info_list=None): + if not info_list: + info_list = meta_list + if not isinstance(meta_list, (list, tuple)): + meta_list = (meta_list,) + if not isinstance(info_list, (list, tuple)): + info_list = (info_list,) + for info_f in info_list: + if info.get(info_f) is not None: + for meta_f in meta_list: + metadata[meta_f] = info[info_f] + break + + add('title', ('track', 'title')) + add('date', 'upload_date') + add(('description', 'comment'), 'description') + add('purl', 'webpage_url') + add('track', 'track_number') + add('artist', ('artist', 'creator', 'uploader', 'uploader_id')) + add('genre') + add('album') + add('album_artist') + add('disc', 'disc_number') if not metadata: self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add') diff --git a/youtube_dl/socks.py b/youtube_dl/socks.py new file mode 100644 index 000000000..104807242 --- /dev/null +++ b/youtube_dl/socks.py @@ -0,0 +1,271 @@ +# Public Domain SOCKS proxy protocol implementation +# Adapted from https://gist.github.com/bluec0re/cafd3764412967417fd3 + +from __future__ import unicode_literals + +# References: +# SOCKS4 protocol http://www.openssh.com/txt/socks4.protocol +# SOCKS4A protocol http://www.openssh.com/txt/socks4a.protocol +# SOCKS5 protocol https://tools.ietf.org/html/rfc1928 +# SOCKS5 username/password authentication https://tools.ietf.org/html/rfc1929 + +import collections +import socket + +from .compat import ( + compat_ord, + compat_struct_pack, + compat_struct_unpack, +) + +__author__ = 'Timo Schmid <coding@timoschmid.de>' + +SOCKS4_VERSION = 4 +SOCKS4_REPLY_VERSION = 0x00 +# Excerpt from SOCKS4A protocol: +# if the client cannot resolve the destination host's domain name to find its +# IP address, it should set the first three bytes of DSTIP to NULL and the last +# byte to a non-zero value. +SOCKS4_DEFAULT_DSTIP = compat_struct_pack('!BBBB', 0, 0, 0, 0xFF) + +SOCKS5_VERSION = 5 +SOCKS5_USER_AUTH_VERSION = 0x01 +SOCKS5_USER_AUTH_SUCCESS = 0x00 + + +class Socks4Command(object): + CMD_CONNECT = 0x01 + CMD_BIND = 0x02 + + +class Socks5Command(Socks4Command): + CMD_UDP_ASSOCIATE = 0x03 + + +class Socks5Auth(object): + AUTH_NONE = 0x00 + AUTH_GSSAPI = 0x01 + AUTH_USER_PASS = 0x02 + AUTH_NO_ACCEPTABLE = 0xFF # For server response + + +class Socks5AddressType(object): + ATYP_IPV4 = 0x01 + ATYP_DOMAINNAME = 0x03 + ATYP_IPV6 = 0x04 + + +class ProxyError(IOError): + ERR_SUCCESS = 0x00 + + def __init__(self, code=None, msg=None): + if code is not None and msg is None: + msg = self.CODES.get(code) and 'unknown error' + super(ProxyError, self).__init__(code, msg) + + +class InvalidVersionError(ProxyError): + def __init__(self, expected_version, got_version): + msg = ('Invalid response version from server. Expected {0:02x} got ' + '{1:02x}'.format(expected_version, got_version)) + super(InvalidVersionError, self).__init__(0, msg) + + +class Socks4Error(ProxyError): + ERR_SUCCESS = 90 + + CODES = { + 91: 'request rejected or failed', + 92: 'request rejected because SOCKS server cannot connect to identd on the client', + 93: 'request rejected because the client program and identd report different user-ids' + } + + +class Socks5Error(ProxyError): + ERR_GENERAL_FAILURE = 0x01 + + CODES = { + 0x01: 'general SOCKS server failure', + 0x02: 'connection not allowed by ruleset', + 0x03: 'Network unreachable', + 0x04: 'Host unreachable', + 0x05: 'Connection refused', + 0x06: 'TTL expired', + 0x07: 'Command not supported', + 0x08: 'Address type not supported', + 0xFE: 'unknown username or invalid password', + 0xFF: 'all offered authentication methods were rejected' + } + + +class ProxyType(object): + SOCKS4 = 0 + SOCKS4A = 1 + SOCKS5 = 2 + +Proxy = collections.namedtuple('Proxy', ( + 'type', 'host', 'port', 'username', 'password', 'remote_dns')) + + +class sockssocket(socket.socket): + def __init__(self, *args, **kwargs): + self._proxy = None + super(sockssocket, self).__init__(*args, **kwargs) + + def setproxy(self, proxytype, addr, port, rdns=True, username=None, password=None): + assert proxytype in (ProxyType.SOCKS4, ProxyType.SOCKS4A, ProxyType.SOCKS5) + + self._proxy = Proxy(proxytype, addr, port, username, password, rdns) + + def recvall(self, cnt): + data = b'' + while len(data) < cnt: + cur = self.recv(cnt - len(data)) + if not cur: + raise IOError('{0} bytes missing'.format(cnt - len(data))) + data += cur + return data + + def _recv_bytes(self, cnt): + data = self.recvall(cnt) + return compat_struct_unpack('!{0}B'.format(cnt), data) + + @staticmethod + def _len_and_data(data): + return compat_struct_pack('!B', len(data)) + data + + def _check_response_version(self, expected_version, got_version): + if got_version != expected_version: + self.close() + raise InvalidVersionError(expected_version, got_version) + + def _resolve_address(self, destaddr, default, use_remote_dns): + try: + return socket.inet_aton(destaddr) + except socket.error: + if use_remote_dns and self._proxy.remote_dns: + return default + else: + return socket.inet_aton(socket.gethostbyname(destaddr)) + + def _setup_socks4(self, address, is_4a=False): + destaddr, port = address + + ipaddr = self._resolve_address(destaddr, SOCKS4_DEFAULT_DSTIP, use_remote_dns=is_4a) + + packet = compat_struct_pack('!BBH', SOCKS4_VERSION, Socks4Command.CMD_CONNECT, port) + ipaddr + + username = (self._proxy.username or '').encode('utf-8') + packet += username + b'\x00' + + if is_4a and self._proxy.remote_dns: + packet += destaddr.encode('utf-8') + b'\x00' + + self.sendall(packet) + + version, resp_code, dstport, dsthost = compat_struct_unpack('!BBHI', self.recvall(8)) + + self._check_response_version(SOCKS4_REPLY_VERSION, version) + + if resp_code != Socks4Error.ERR_SUCCESS: + self.close() + raise Socks4Error(resp_code) + + return (dsthost, dstport) + + def _setup_socks4a(self, address): + self._setup_socks4(address, is_4a=True) + + def _socks5_auth(self): + packet = compat_struct_pack('!B', SOCKS5_VERSION) + + auth_methods = [Socks5Auth.AUTH_NONE] + if self._proxy.username and self._proxy.password: + auth_methods.append(Socks5Auth.AUTH_USER_PASS) + + packet += compat_struct_pack('!B', len(auth_methods)) + packet += compat_struct_pack('!{0}B'.format(len(auth_methods)), *auth_methods) + + self.sendall(packet) + + version, method = self._recv_bytes(2) + + self._check_response_version(SOCKS5_VERSION, version) + + if method == Socks5Auth.AUTH_NO_ACCEPTABLE: + self.close() + raise Socks5Error(method) + + if method == Socks5Auth.AUTH_USER_PASS: + username = self._proxy.username.encode('utf-8') + password = self._proxy.password.encode('utf-8') + packet = compat_struct_pack('!B', SOCKS5_USER_AUTH_VERSION) + packet += self._len_and_data(username) + self._len_and_data(password) + self.sendall(packet) + + version, status = self._recv_bytes(2) + + self._check_response_version(SOCKS5_USER_AUTH_VERSION, version) + + if status != SOCKS5_USER_AUTH_SUCCESS: + self.close() + raise Socks5Error(Socks5Error.ERR_GENERAL_FAILURE) + + def _setup_socks5(self, address): + destaddr, port = address + + ipaddr = self._resolve_address(destaddr, None, use_remote_dns=True) + + self._socks5_auth() + + reserved = 0 + packet = compat_struct_pack('!BBB', SOCKS5_VERSION, Socks5Command.CMD_CONNECT, reserved) + if ipaddr is None: + destaddr = destaddr.encode('utf-8') + packet += compat_struct_pack('!B', Socks5AddressType.ATYP_DOMAINNAME) + packet += self._len_and_data(destaddr) + else: + packet += compat_struct_pack('!B', Socks5AddressType.ATYP_IPV4) + ipaddr + packet += compat_struct_pack('!H', port) + + self.sendall(packet) + + version, status, reserved, atype = self._recv_bytes(4) + + self._check_response_version(SOCKS5_VERSION, version) + + if status != Socks5Error.ERR_SUCCESS: + self.close() + raise Socks5Error(status) + + if atype == Socks5AddressType.ATYP_IPV4: + destaddr = self.recvall(4) + elif atype == Socks5AddressType.ATYP_DOMAINNAME: + alen = compat_ord(self.recv(1)) + destaddr = self.recvall(alen) + elif atype == Socks5AddressType.ATYP_IPV6: + destaddr = self.recvall(16) + destport = compat_struct_unpack('!H', self.recvall(2))[0] + + return (destaddr, destport) + + def _make_proxy(self, connect_func, address): + if not self._proxy: + return connect_func(self, address) + + result = connect_func(self, (self._proxy.host, self._proxy.port)) + if result != 0 and result is not None: + return result + setup_funcs = { + ProxyType.SOCKS4: self._setup_socks4, + ProxyType.SOCKS4A: self._setup_socks4a, + ProxyType.SOCKS5: self._setup_socks5, + } + setup_funcs[self._proxy.type](address) + return result + + def connect(self, address): + self._make_proxy(socket.socket.connect, address) + + def connect_ex(self, address): + return self._make_proxy(socket.socket.connect_ex, address) diff --git a/youtube_dl/swfinterp.py b/youtube_dl/swfinterp.py index 06c1d6cc1..7cf490aa4 100644 --- a/youtube_dl/swfinterp.py +++ b/youtube_dl/swfinterp.py @@ -4,10 +4,12 @@ import collections import io import zlib -from .compat import compat_str +from .compat import ( + compat_str, + compat_struct_unpack, +) from .utils import ( ExtractorError, - struct_unpack, ) @@ -23,17 +25,17 @@ def _extract_tags(file_contents): file_contents[:1]) # Determine number of bits in framesize rectangle - framesize_nbits = struct_unpack('!B', content[:1])[0] >> 3 + framesize_nbits = compat_struct_unpack('!B', content[:1])[0] >> 3 framesize_len = (5 + 4 * framesize_nbits + 7) // 8 pos = framesize_len + 2 + 2 while pos < len(content): - header16 = struct_unpack('<H', content[pos:pos + 2])[0] + header16 = compat_struct_unpack('<H', content[pos:pos + 2])[0] pos += 2 tag_code = header16 >> 6 tag_len = header16 & 0x3f if tag_len == 0x3f: - tag_len = struct_unpack('<I', content[pos:pos + 4])[0] + tag_len = compat_struct_unpack('<I', content[pos:pos + 4])[0] pos += 4 assert pos + tag_len <= len(content), \ ('Tag %d ends at %d+%d - that\'s longer than the file (%d)' @@ -101,7 +103,7 @@ def _read_int(reader): for _ in range(5): buf = reader.read(1) assert len(buf) == 1 - b = struct_unpack('<B', buf)[0] + b = compat_struct_unpack('<B', buf)[0] res = res | ((b & 0x7f) << shift) if b & 0x80 == 0: break @@ -127,7 +129,7 @@ def _s24(reader): bs = reader.read(3) assert len(bs) == 3 last_byte = b'\xff' if (ord(bs[2:3]) >= 0x80) else b'\x00' - return struct_unpack('<i', bs + last_byte)[0] + return compat_struct_unpack('<i', bs + last_byte)[0] def _read_string(reader): @@ -146,7 +148,7 @@ def _read_bytes(count, reader): def _read_byte(reader): resb = _read_bytes(1, reader=reader) - res = struct_unpack('<B', resb)[0] + res = compat_struct_unpack('<B', resb)[0] return res diff --git a/youtube_dl/update.py b/youtube_dl/update.py index 676ebe1c4..ebce9666a 100644 --- a/youtube_dl/update.py +++ b/youtube_dl/update.py @@ -83,11 +83,8 @@ def update_self(to_screen, verbose, opener): print_notes(to_screen, versions_info['versions']) - filename = sys.argv[0] - # Py2EXE: Filename could be different - if hasattr(sys, 'frozen') and not os.path.isfile(filename): - if os.path.isfile(filename + '.exe'): - filename += '.exe' + # sys.executable is set to the full pathname of the exe-file for py2exe + filename = sys.executable if hasattr(sys, 'frozen') else sys.argv[0] if not os.access(filename, os.W_OK): to_screen('ERROR: no write permissions on %s' % filename) @@ -95,7 +92,7 @@ def update_self(to_screen, verbose, opener): # Py2EXE if hasattr(sys, 'frozen'): - exe = os.path.abspath(filename) + exe = filename directory = os.path.dirname(exe) if not os.access(directory, os.W_OK): to_screen('ERROR: no write permissions on %s' % directory) diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index f333e4712..3498697b6 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -14,8 +14,8 @@ import email.utils import errno import functools import gzip -import itertools import io +import itertools import json import locale import math @@ -24,9 +24,8 @@ import os import pipes import platform import re -import ssl import socket -import struct +import ssl import subprocess import sys import tempfile @@ -40,27 +39,44 @@ from .compat import ( compat_chr, compat_etree_fromstring, compat_html_entities, + compat_html_entities_html5, compat_http_client, compat_kwargs, compat_parse_qs, + compat_shlex_quote, compat_socket_create_connection, compat_str, + compat_struct_pack, compat_urllib_error, compat_urllib_parse, compat_urllib_parse_urlencode, compat_urllib_parse_urlparse, + compat_urllib_parse_unquote_plus, compat_urllib_request, compat_urlparse, compat_xpath, - shlex_quote, +) + +from .socks import ( + ProxyType, + sockssocket, ) +def register_socks_protocols(): + # "Register" SOCKS protocols + # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 + # URLs with protocols not in urlparse.uses_netloc are not handled correctly + for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): + if scheme not in compat_urlparse.uses_netloc: + compat_urlparse.uses_netloc.append(scheme) + + # This is not clearly defined otherwise compiled_regex_type = type(re.compile('')) std_headers = { - 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/44.0 (Chrome)', + 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/47.0 (Chrome)', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', 'Accept-Encoding': 'gzip, deflate', @@ -89,6 +105,54 @@ KNOWN_EXTENSIONS = ( 'wav', 'f4f', 'f4m', 'm3u8', 'smil') +# needed for sanitizing filenames in restricted mode +ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', + itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'], + 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy'))) + +DATE_FORMATS = ( + '%d %B %Y', + '%d %b %Y', + '%B %d %Y', + '%b %d %Y', + '%b %dst %Y %I:%M', + '%b %dnd %Y %I:%M', + '%b %dth %Y %I:%M', + '%Y %m %d', + '%Y-%m-%d', + '%Y/%m/%d', + '%Y/%m/%d %H:%M:%S', + '%Y-%m-%d %H:%M:%S', + '%Y-%m-%d %H:%M:%S.%f', + '%d.%m.%Y %H:%M', + '%d.%m.%Y %H.%M', + '%Y-%m-%dT%H:%M:%SZ', + '%Y-%m-%dT%H:%M:%S.%fZ', + '%Y-%m-%dT%H:%M:%S.%f0Z', + '%Y-%m-%dT%H:%M:%S', + '%Y-%m-%dT%H:%M:%S.%f', + '%Y-%m-%dT%H:%M', +) + +DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) +DATE_FORMATS_DAY_FIRST.extend([ + '%d-%m-%Y', + '%d.%m.%Y', + '%d.%m.%y', + '%d/%m/%Y', + '%d/%m/%y', + '%d/%m/%Y %H:%M:%S', +]) + +DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) +DATE_FORMATS_MONTH_FIRST.extend([ + '%m-%d-%Y', + '%m.%d.%Y', + '%m/%d/%Y', + '%m/%d/%y', + '%m/%d/%Y %H:%M:%S', +]) + def preferredencoding(): """Get preferred encoding. @@ -246,18 +310,26 @@ def get_element_by_id(id, html): return get_element_by_attribute('id', id, html) -def get_element_by_attribute(attribute, value, html): +def get_element_by_class(class_name, html): + return get_element_by_attribute( + 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name), + html, escape_value=False) + + +def get_element_by_attribute(attribute, value, html, escape_value=True): """Return the content of the tag with the specified attribute in the passed HTML document""" + value = re.escape(value) if escape_value else value + m = re.search(r'''(?xs) <([a-zA-Z0-9:._-]+) - (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*? + (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? \s+%s=['"]?%s['"]? - (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*? + (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? \s*> (?P<content>.*?) </\1> - ''' % (re.escape(attribute), re.escape(value)), html) + ''' % (re.escape(attribute), value), html) if not m: return None @@ -365,6 +437,8 @@ def sanitize_filename(s, restricted=False, is_id=False): Set is_id if this is not an arbitrary string, but an ID that should be kept if possible """ def replace_insane(char): + if restricted and char in ACCENT_CHARS: + return ACCENT_CHARS[char] if char == '?' or ord(char) < 32 or ord(char) == 127: return '' elif char == '"': @@ -434,12 +508,19 @@ def orderedSet(iterable): return res -def _htmlentity_transform(entity): +def _htmlentity_transform(entity_with_semicolon): """Transforms an HTML entity to a character.""" + entity = entity_with_semicolon[:-1] + # Known non-numeric HTML entity if entity in compat_html_entities.name2codepoint: return compat_chr(compat_html_entities.name2codepoint[entity]) + # TODO: HTML5 allows entities without a semicolon. For example, + # 'Éric' should be decoded as 'Éric'. + if entity_with_semicolon in compat_html_entities_html5: + return compat_html_entities_html5[entity_with_semicolon] + mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) if mobj is not None: numstr = mobj.group(1) @@ -464,7 +545,7 @@ def unescapeHTML(s): assert type(s) == compat_str return re.sub( - r'&([^;]+);', lambda m: _htmlentity_transform(m.group(1)), s) + r'&([^;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) def get_subprocess_encoding(): @@ -745,8 +826,15 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): self._params = params def http_open(self, req): + conn_class = compat_http_client.HTTPConnection + + socks_proxy = req.headers.get('Ytdl-socks-proxy') + if socks_proxy: + conn_class = make_socks_conn_class(conn_class, socks_proxy) + del req.headers['Ytdl-socks-proxy'] + return self.do_open(functools.partial( - _create_http_connection, self, compat_http_client.HTTPConnection, False), + _create_http_connection, self, conn_class, False), req) @staticmethod @@ -832,9 +920,13 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 if sys.version_info >= (3, 0): location = location.encode('iso-8859-1').decode('utf-8') + else: + location = location.decode('utf-8') location_escaped = escape_url(location) if location != location_escaped: del resp.headers['Location'] + if sys.version_info < (3, 0): + location_escaped = location_escaped.encode('utf-8') resp.headers['Location'] = location_escaped return resp @@ -842,6 +934,49 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): https_response = http_response +def make_socks_conn_class(base_class, socks_proxy): + assert issubclass(base_class, ( + compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection)) + + url_components = compat_urlparse.urlparse(socks_proxy) + if url_components.scheme.lower() == 'socks5': + socks_type = ProxyType.SOCKS5 + elif url_components.scheme.lower() in ('socks', 'socks4'): + socks_type = ProxyType.SOCKS4 + elif url_components.scheme.lower() == 'socks4a': + socks_type = ProxyType.SOCKS4A + + def unquote_if_non_empty(s): + if not s: + return s + return compat_urllib_parse_unquote_plus(s) + + proxy_args = ( + socks_type, + url_components.hostname, url_components.port or 1080, + True, # Remote DNS + unquote_if_non_empty(url_components.username), + unquote_if_non_empty(url_components.password), + ) + + class SocksConnection(base_class): + def connect(self): + self.sock = sockssocket() + self.sock.setproxy(*proxy_args) + if type(self.timeout) in (int, float): + self.sock.settimeout(self.timeout) + self.sock.connect((self.host, self.port)) + + if isinstance(self, compat_http_client.HTTPSConnection): + if hasattr(self, '_context'): # Python > 2.6 + self.sock = self._context.wrap_socket( + self.sock, server_hostname=self.host) + else: + self.sock = ssl.wrap_socket(self.sock) + + return SocksConnection + + class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): def __init__(self, params, https_conn_class=None, *args, **kwargs): compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs) @@ -850,12 +985,20 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): def https_open(self, req): kwargs = {} + conn_class = self._https_conn_class + if hasattr(self, '_context'): # python > 2.6 kwargs['context'] = self._context if hasattr(self, '_check_hostname'): # python 3.x kwargs['check_hostname'] = self._check_hostname + + socks_proxy = req.headers.get('Ytdl-socks-proxy') + if socks_proxy: + conn_class = make_socks_conn_class(conn_class, socks_proxy) + del req.headers['Ytdl-socks-proxy'] + return self.do_open(functools.partial( - _create_http_connection, self, self._https_conn_class, True), + _create_http_connection, self, conn_class, True), req, **kwargs) @@ -883,6 +1026,24 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): https_response = http_response +def extract_timezone(date_str): + m = re.search( + r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', + date_str) + if not m: + timezone = datetime.timedelta() + else: + date_str = date_str[:-len(m.group('tz'))] + if not m.group('sign'): + timezone = datetime.timedelta() + else: + sign = 1 if m.group('sign') == '+' else -1 + timezone = datetime.timedelta( + hours=sign * int(m.group('hours')), + minutes=sign * int(m.group('minutes'))) + return timezone, date_str + + def parse_iso8601(date_str, delimiter='T', timezone=None): """ Return a UNIX timestamp from the given date """ @@ -892,20 +1053,8 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): date_str = re.sub(r'\.[0-9]+', '', date_str) if timezone is None: - m = re.search( - r'(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', - date_str) - if not m: - timezone = datetime.timedelta() - else: - date_str = date_str[:-len(m.group(0))] - if not m.group('sign'): - timezone = datetime.timedelta() - else: - sign = 1 if m.group('sign') == '+' else -1 - timezone = datetime.timedelta( - hours=sign * int(m.group('hours')), - minutes=sign * int(m.group('minutes'))) + timezone, date_str = extract_timezone(date_str) + try: date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) dt = datetime.datetime.strptime(date_str, date_format) - timezone @@ -914,6 +1063,10 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): pass +def date_formats(day_first=True): + return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST + + def unified_strdate(date_str, day_first=True): """Return a string with the date in the format YYYYMMDD""" @@ -922,52 +1075,11 @@ def unified_strdate(date_str, day_first=True): upload_date = None # Replace commas date_str = date_str.replace(',', ' ') - # %z (UTC offset) is only supported in python>=3.2 - if not re.match(r'^[0-9]{1,2}-[0-9]{1,2}-[0-9]{4}$', date_str): - date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str) # Remove AM/PM + timezone date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) + _, date_str = extract_timezone(date_str) - format_expressions = [ - '%d %B %Y', - '%d %b %Y', - '%B %d %Y', - '%b %d %Y', - '%b %dst %Y %I:%M', - '%b %dnd %Y %I:%M', - '%b %dth %Y %I:%M', - '%Y %m %d', - '%Y-%m-%d', - '%Y/%m/%d', - '%Y/%m/%d %H:%M:%S', - '%Y-%m-%d %H:%M:%S', - '%Y-%m-%d %H:%M:%S.%f', - '%d.%m.%Y %H:%M', - '%d.%m.%Y %H.%M', - '%Y-%m-%dT%H:%M:%SZ', - '%Y-%m-%dT%H:%M:%S.%fZ', - '%Y-%m-%dT%H:%M:%S.%f0Z', - '%Y-%m-%dT%H:%M:%S', - '%Y-%m-%dT%H:%M:%S.%f', - '%Y-%m-%dT%H:%M', - ] - if day_first: - format_expressions.extend([ - '%d-%m-%Y', - '%d.%m.%Y', - '%d/%m/%Y', - '%d/%m/%y', - '%d/%m/%Y %H:%M:%S', - ]) - else: - format_expressions.extend([ - '%m-%d-%Y', - '%m.%d.%Y', - '%m/%d/%Y', - '%m/%d/%y', - '%m/%d/%Y %H:%M:%S', - ]) - for expression in format_expressions: + for expression in date_formats(day_first): try: upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') except ValueError: @@ -975,11 +1087,37 @@ def unified_strdate(date_str, day_first=True): if upload_date is None: timetuple = email.utils.parsedate_tz(date_str) if timetuple: - upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') + try: + upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') + except ValueError: + pass if upload_date is not None: return compat_str(upload_date) +def unified_timestamp(date_str, day_first=True): + if date_str is None: + return None + + date_str = date_str.replace(',', ' ') + + pm_delta = datetime.timedelta(hours=12 if re.search(r'(?i)PM', date_str) else 0) + timezone, date_str = extract_timezone(date_str) + + # Remove AM/PM + timezone + date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) + + for expression in date_formats(day_first): + try: + dt = datetime.datetime.strptime(date_str, expression) - timezone + pm_delta + return calendar.timegm(dt.timetuple()) + except ValueError: + pass + timetuple = email.utils.parsedate_tz(date_str) + if timetuple: + return calendar.timegm(timetuple.timetuple()) + + def determine_ext(url, default_ext='unknown_video'): if url is None: return default_ext @@ -1186,7 +1324,7 @@ def bytes_to_intlist(bs): def intlist_to_bytes(xs): if not xs: return b'' - return struct_pack('%dB' % len(xs), *xs) + return compat_struct_pack('%dB' % len(xs), *xs) # Cross-platform file locking @@ -1314,6 +1452,8 @@ def shell_quote(args): def smuggle_url(url, data): """ Pass additional data in a URL for internal use. """ + url, idata = unsmuggle_url(url, {}) + data.update(idata) sdata = compat_urllib_parse_urlencode( {'__youtubedl_smuggle': json.dumps(data)}) return url + '#' + sdata @@ -1469,15 +1609,11 @@ def setproctitle(title): def remove_start(s, start): - if s.startswith(start): - return s[len(start):] - return s + return s[len(start):] if s is not None and s.startswith(start) else s def remove_end(s, end): - if s.endswith(end): - return s[:-len(end)] - return s + return s[:-len(end)] if s is not None and s.endswith(end) else s def remove_quotes(s): @@ -1499,6 +1635,11 @@ class HEADRequest(compat_urllib_request.Request): return 'HEAD' +class PUTRequest(compat_urllib_request.Request): + def get_method(self): + return 'PUT' + + def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): if get_attr: if v is not None: @@ -1534,6 +1675,10 @@ def float_or_none(v, scale=1, invscale=1, default=None): return default +def strip_or_none(v): + return None if v is None else v.strip() + + def parse_duration(s): if not isinstance(s, compat_basestring): return None @@ -1754,24 +1899,6 @@ def escape_url(url): fragment=escape_rfc3986(url_parsed.fragment) ).geturl() -try: - struct.pack('!I', 0) -except TypeError: - # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument - # See https://bugs.python.org/issue19099 - def struct_pack(spec, *args): - if isinstance(spec, compat_str): - spec = spec.encode('ascii') - return struct.pack(spec, *args) - - def struct_unpack(spec, *args): - if isinstance(spec, compat_str): - spec = spec.encode('ascii') - return struct.unpack(spec, *args) -else: - struct_pack = struct.pack - struct_unpack = struct.unpack - def read_batch_urls(batch_fd): def fixup(url): @@ -1808,7 +1935,13 @@ def update_Request(req, url=None, data=None, headers={}, query={}): req_headers.update(headers) req_data = data or req.data req_url = update_url_query(url or req.get_full_url(), query) - req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request + req_get_method = req.get_method() + if req_get_method == 'HEAD': + req_type = HEADRequest + elif req_get_method == 'PUT': + req_type = PUTRequest + else: + req_type = compat_urllib_request.Request new_req = req_type( req_url, data=req_data, headers=req_headers, origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) @@ -1827,6 +1960,16 @@ def dict_get(d, key_or_keys, default=None, skip_false_values=True): return d.get(key_or_keys, default) +def try_get(src, getter, expected_type=None): + try: + v = getter(src) + except (AttributeError, KeyError, TypeError, IndexError): + pass + else: + if expected_type is None or isinstance(v, expected_type): + return v + + def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): return string if isinstance(string, compat_str) else compat_str(string, encoding, errors) @@ -1849,7 +1992,7 @@ def parse_age_limit(s): def strip_jsonp(code): return re.sub( - r'(?s)^[a-zA-Z0-9_.]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code) + r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code) def js_to_json(code): @@ -1857,24 +2000,38 @@ def js_to_json(code): v = m.group(0) if v in ('true', 'false', 'null'): return v - if v.startswith('"'): - v = re.sub(r"\\'", "'", v[1:-1]) - elif v.startswith("'"): - v = v[1:-1] - v = re.sub(r"\\\\|\\'|\"", lambda m: { - '\\\\': '\\\\', - "\\'": "'", + elif v.startswith('/*') or v == ',': + return "" + + if v[0] in ("'", '"'): + v = re.sub(r'(?s)\\.|"', lambda m: { '"': '\\"', - }[m.group(0)], v) + "\\'": "'", + '\\\n': '', + '\\x': '\\u00', + }.get(m.group(0), m.group(0)), v[1:-1]) + + INTEGER_TABLE = ( + (r'^0[xX][0-9a-fA-F]+', 16), + (r'^0+[0-7]+', 8), + ) + + for regex, base in INTEGER_TABLE: + im = re.match(regex, v) + if im: + i = int(im.group(0), base) + return '"%d":' % i if v.endswith(':') else '%d' % i + return '"%s"' % v - res = re.sub(r'''(?x) - "(?:[^"\\]*(?:\\\\|\\['"nu]))*[^"\\]*"| - '(?:[^'\\]*(?:\\\\|\\['"nu]))*[^'\\]*'| - [a-zA-Z_][.a-zA-Z_0-9]* + return re.sub(r'''(?sx) + "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| + '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| + /\*.*?\*/|,(?=\s*[\]}])| + [a-zA-Z_][.a-zA-Z_0-9]*| + \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?| + [0-9]+(?=\s*:) ''', fix_kv, code) - res = re.sub(r',(\s*[\]}])', lambda m: m.group(1), res) - return res def qualities(quality_ids): @@ -1922,7 +2079,7 @@ def ytdl_is_updateable(): def args_to_str(args): # Get a short string representation for a subprocess command - return ' '.join(shlex_quote(a) for a in args) + return ' '.join(compat_shlex_quote(a) for a in args) def error_to_compat_str(err): @@ -1935,13 +2092,20 @@ def error_to_compat_str(err): def mimetype2ext(mt): + if mt is None: + return None + ext = { 'audio/mp4': 'm4a', + # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as + # it's the most popular one + 'audio/mpeg': 'mp3', }.get(mt) if ext is not None: return ext _, _, res = mt.rpartition('/') + res = res.lower() return { '3gpp': '3gp', @@ -1953,15 +2117,17 @@ def mimetype2ext(mt): 'x-flv': 'flv', 'x-mp4-fragmented': 'mp4', 'x-ms-wmv': 'wmv', + 'mpegurl': 'm3u8', + 'x-mpegurl': 'm3u8', + 'vnd.apple.mpegurl': 'm3u8', + 'dash+xml': 'mpd', + 'f4m': 'f4m', + 'f4m+xml': 'f4m', }.get(res, res) def urlhandle_detect_ext(url_handle): - try: - url_handle.headers - getheader = lambda h: url_handle.headers[h] - except AttributeError: # Python < 3 - getheader = url_handle.info().getheader + getheader = url_handle.headers.get cd = getheader('Content-Disposition') if cd: @@ -2691,6 +2857,10 @@ class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): if proxy == '__noproxy__': return None # No Proxy + if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): + req.add_header('Ytdl-socks-proxy', proxy) + # youtube-dl's http/https handlers do wrapping the socket with socks + return None return compat_urllib_request.ProxyHandler.proxy_open( self, req, proxy, type) @@ -2748,3 +2918,16 @@ def decode_packed_codes(code): return re.sub( r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], obfucasted_code) + + +def parse_m3u8_attributes(attrib): + info = {} + for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): + if val.startswith('"'): + val = val[1:-1] + info[key] = val + return info + + +def urshift(val, n): + return val >> n if val >= 0 else (val + 0x100000000) >> n diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 140a67847..3e45977c9 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2016.04.19' +__version__ = '2016.07.09.1' |