diff options
Diffstat (limited to 'youtube_dl')
48 files changed, 824 insertions, 289 deletions
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index 088f59586..e5a542ed5 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -5,9 +5,6 @@ import subprocess import sys import time -if os.name == 'nt': - import ctypes - from .utils import ( compat_urllib_error, compat_urllib_request, @@ -151,16 +148,8 @@ class FileDownloader(object): def to_stderr(self, message): self.ydl.to_screen(message) - def to_cons_title(self, message): - """Set console/terminal window title to message.""" - if not self.params.get('consoletitle', False): - return - if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): - # c_wchar_p() might not be necessary if `message` is - # already of type unicode() - ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) - elif 'TERM' in os.environ: - self.to_screen('\033]0;%s\007' % message, skip_eol=True) + def to_console_title(self, message): + self.ydl.to_console_title(message) def trouble(self, *args, **kargs): self.ydl.trouble(*args, **kargs) @@ -249,7 +238,7 @@ class FileDownloader(object): else: self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' % (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True) - self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' % + self.to_console_title(u'youtube-dl - %s of %s at %s ETA %s' % (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip())) def report_resuming_byte(self, resume_len): diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 5253c39e1..2700051cf 100644 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -5,6 +5,7 @@ from __future__ import absolute_import import errno import io +import json import os import re import shutil @@ -13,7 +14,34 @@ import sys import time import traceback -from .utils import * +if os.name == 'nt': + import ctypes + +from .utils import ( + compat_http_client, + compat_print, + compat_str, + compat_urllib_error, + compat_urllib_request, + ContentTooShortError, + date_from_str, + DateRange, + determine_ext, + DownloadError, + encodeFilename, + ExtractorError, + locked_file, + MaxDownloadsReached, + PostProcessingError, + preferredencoding, + SameFileError, + sanitize_filename, + subtitles_filename, + takewhile_inclusive, + UnavailableVideoError, + write_json_file, + write_string, +) from .extractor import get_info_extractor, gen_extractors from .FileDownloader import FileDownloader @@ -57,6 +85,7 @@ class YoutubeDL(object): forcethumbnail: Force printing thumbnail URL. forcedescription: Force printing description. forcefilename: Force printing final filename. + forcejson: Force printing info_dict as JSON. simulate: Do not download the video files. format: Video format code. format_limit: Highest quality format to try. @@ -176,6 +205,37 @@ class YoutubeDL(object): output = output.encode(preferredencoding()) sys.stderr.write(output) + def to_console_title(self, message): + if not self.params.get('consoletitle', False): + return + if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): + # c_wchar_p() might not be necessary if `message` is + # already of type unicode() + ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) + elif 'TERM' in os.environ: + write_string(u'\033]0;%s\007' % message, self._screen_file) + + def save_console_title(self): + if not self.params.get('consoletitle', False): + return + if 'TERM' in os.environ: + # Save the title on stack + write_string(u'\033[22;0t', self._screen_file) + + def restore_console_title(self): + if not self.params.get('consoletitle', False): + return + if 'TERM' in os.environ: + # Restore the title from stack + write_string(u'\033[23;0t', self._screen_file) + + def __enter__(self): + self.save_console_title() + return self + + def __exit__(self, *args): + self.restore_console_title() + def fixed_template(self): """Checks if the output template is fixed.""" return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None) @@ -254,7 +314,7 @@ class YoutubeDL(object): """Report file has already been fully downloaded.""" try: self.to_screen(u'[download] %s has already been downloaded' % file_name) - except (UnicodeEncodeError) as err: + except UnicodeEncodeError: self.to_screen(u'[download] The file has already been downloaded') def increment_downloads(self): @@ -385,7 +445,7 @@ class YoutubeDL(object): result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system if result_type == 'video': self.add_extra_info(ie_result, extra_info) - return self.process_video_result(ie_result) + return self.process_video_result(ie_result, download=download) elif result_type == 'url': # We have to add extra_info to the results because it may be # contained in a playlist @@ -579,7 +639,7 @@ class YoutubeDL(object): # Forced printings if self.params.get('forcetitle', False): - compat_print(info_dict['title']) + compat_print(info_dict['fulltitle']) if self.params.get('forceid', False): compat_print(info_dict['id']) if self.params.get('forceurl', False): @@ -593,6 +653,8 @@ class YoutubeDL(object): compat_print(filename) if self.params.get('forceformat', False): compat_print(info_dict['format']) + if self.params.get('forcejson', False): + compat_print(json.dumps(info_dict)) # Do nothing else if in simulate mode if self.params.get('simulate', False): @@ -640,7 +702,7 @@ class YoutubeDL(object): # subtitles download errors are already managed as troubles in relevant IE # that way it will silently go on when used with unsupporting IE subtitles = info_dict['subtitles'] - sub_format = self.params.get('subtitlesformat') + sub_format = self.params.get('subtitlesformat', 'srt') for sub_lang in subtitles.keys(): sub = subtitles[sub_lang] if sub is None: @@ -655,7 +717,7 @@ class YoutubeDL(object): return if self.params.get('writeinfojson', False): - infofn = filename + u'.info.json' + infofn = os.path.splitext(filename)[0] + u'.info.json' self.report_writeinfojson(infofn) try: json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) @@ -781,20 +843,42 @@ class YoutubeDL(object): return res def list_formats(self, info_dict): + def format_note(fdict): + if fdict.get('format_note') is not None: + return fdict['format_note'] + res = u'' + if fdict.get('vcodec') is not None: + res += u'%-5s' % fdict['vcodec'] + elif fdict.get('vbr') is not None: + res += u'video' + if fdict.get('vbr') is not None: + res += u'@%4dk' % fdict['vbr'] + if fdict.get('acodec') is not None: + if res: + res += u', ' + res += u'%-5s' % fdict['acodec'] + elif fdict.get('abr') is not None: + if res: + res += u', ' + res += 'audio' + if fdict.get('abr') is not None: + res += u'@%3dk' % fdict['abr'] + return res + def line(format): return (u'%-20s%-10s%-12s%s' % ( format['format_id'], format['ext'], self.format_resolution(format), - format.get('format_note', ''), + format_note(format), ) ) formats = info_dict.get('formats', [info_dict]) formats_s = list(map(line, formats)) if len(formats) > 1: - formats_s[0] += (' ' if formats[0].get('format_note') else '') + '(worst)' - formats_s[-1] += (' ' if formats[-1].get('format_note') else '') + '(best)' + formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)' + formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)' header_line = line({ 'format_id': u'format code', 'ext': u'extension', diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 4dee487ab..64ebf4d48 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -34,6 +34,7 @@ __authors__ = ( 'Andras Elso', 'Jelle van der Waa', 'Marcin Cieślak', + 'Anton Larionov', ) __license__ = 'Public Domain' @@ -306,6 +307,9 @@ def parseOpts(overrideArguments=None): verbosity.add_option('--get-format', action='store_true', dest='getformat', help='simulate, quiet but print output format', default=False) + verbosity.add_option('-j', '--dump-json', + action='store_true', dest='dumpjson', + help='simulate, quiet but print JSON information', default=False) verbosity.add_option('--newline', action='store_true', dest='progress_with_newline', help='output progress bar as new lines', default=False) verbosity.add_option('--no-progress', @@ -603,13 +607,12 @@ def _real_main(argv=None): u' file! Use "%%(ext)s" instead of %r' % determine_ext(outtmpl, u'')) - # YoutubeDL - ydl = YoutubeDL({ + ydl_opts = { 'usenetrc': opts.usenetrc, 'username': opts.username, 'password': opts.password, 'videopassword': opts.videopassword, - 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson), 'forceurl': opts.geturl, 'forcetitle': opts.gettitle, 'forceid': opts.getid, @@ -617,8 +620,9 @@ def _real_main(argv=None): 'forcedescription': opts.getdescription, 'forcefilename': opts.getfilename, 'forceformat': opts.getformat, + 'forcejson': opts.dumpjson, 'simulate': opts.simulate, - 'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat), + 'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson), 'format': opts.format, 'format_limit': opts.format_limit, 'listformats': opts.listformats, @@ -667,61 +671,63 @@ def _real_main(argv=None): 'youtube_print_sig_code': opts.youtube_print_sig_code, 'age_limit': opts.age_limit, 'download_archive': opts.download_archive, - }) + } - if opts.verbose: - write_string(u'[debug] youtube-dl version ' + __version__ + u'\n') - try: - sp = subprocess.Popen( - ['git', 'rev-parse', '--short', 'HEAD'], - stdout=subprocess.PIPE, stderr=subprocess.PIPE, - cwd=os.path.dirname(os.path.abspath(__file__))) - out, err = sp.communicate() - out = out.decode().strip() - if re.match('[0-9a-f]+', out): - write_string(u'[debug] Git HEAD: ' + out + u'\n') - except: + with YoutubeDL(ydl_opts) as ydl: + if opts.verbose: + write_string(u'[debug] youtube-dl version ' + __version__ + u'\n') try: - sys.exc_clear() + sp = subprocess.Popen( + ['git', 'rev-parse', '--short', 'HEAD'], + stdout=subprocess.PIPE, stderr=subprocess.PIPE, + cwd=os.path.dirname(os.path.abspath(__file__))) + out, err = sp.communicate() + out = out.decode().strip() + if re.match('[0-9a-f]+', out): + write_string(u'[debug] Git HEAD: ' + out + u'\n') except: - pass - write_string(u'[debug] Python version %s - %s' %(platform.python_version(), platform_name()) + u'\n') - - proxy_map = {} - for handler in opener.handlers: - if hasattr(handler, 'proxies'): - proxy_map.update(handler.proxies) - write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n') - - ydl.add_default_info_extractors() - - # PostProcessors - # Add the metadata pp first, the other pps will copy it - if opts.addmetadata: - ydl.add_post_processor(FFmpegMetadataPP()) - if opts.extractaudio: - ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites)) - if opts.recodevideo: - ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo)) - if opts.embedsubtitles: - ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat)) - - # Update version - if opts.update_self: - update_self(ydl.to_screen, opts.verbose) - - # Maybe do nothing - if len(all_urls) < 1: - if not opts.update_self: - parser.error(u'you must provide at least one URL') - else: - sys.exit() + try: + sys.exc_clear() + except: + pass + write_string(u'[debug] Python version %s - %s' % + (platform.python_version(), platform_name()) + u'\n') + + proxy_map = {} + for handler in opener.handlers: + if hasattr(handler, 'proxies'): + proxy_map.update(handler.proxies) + write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n') + + ydl.add_default_info_extractors() + + # PostProcessors + # Add the metadata pp first, the other pps will copy it + if opts.addmetadata: + ydl.add_post_processor(FFmpegMetadataPP()) + if opts.extractaudio: + ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites)) + if opts.recodevideo: + ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo)) + if opts.embedsubtitles: + ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat)) + + # Update version + if opts.update_self: + update_self(ydl.to_screen, opts.verbose) + + # Maybe do nothing + if len(all_urls) < 1: + if not opts.update_self: + parser.error(u'you must provide at least one URL') + else: + sys.exit() - try: - retcode = ydl.download(all_urls) - except MaxDownloadsReached: - ydl.to_screen(u'--max-download limit reached, aborting.') - retcode = 101 + try: + retcode = ydl.download(all_urls) + except MaxDownloadsReached: + ydl.to_screen(u'--max-download limit reached, aborting.') + retcode = 101 # Dump cookie jar if requested if opts.cookiefile is not None: diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 0594a3666..02f9e2546 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -1,5 +1,6 @@ from .appletrailers import AppleTrailersIE from .addanime import AddAnimeIE +from .anitube import AnitubeIE from .archiveorg import ArchiveOrgIE from .ard import ARDIE from .arte import ( @@ -10,7 +11,7 @@ from .arte import ( ) from .auengine import AUEngineIE from .bambuser import BambuserIE, BambuserChannelIE -from .bandcamp import BandcampIE +from .bandcamp import BandcampIE, BandcampAlbumIE from .bliptv import BlipTVIE, BlipTVUserIE from .bloomberg import BloombergIE from .breakcom import BreakIE @@ -25,6 +26,7 @@ from .comedycentral import ComedyCentralIE from .condenast import CondeNastIE from .criterion import CriterionIE from .cspan import CSpanIE +from .d8 import D8IE from .dailymotion import ( DailymotionIE, DailymotionPlaylistIE, @@ -80,7 +82,7 @@ from .keezmovies import KeezMoviesIE from .kickstarter import KickStarterIE from .keek import KeekIE from .liveleak import LiveLeakIE -from .livestream import LivestreamIE +from .livestream import LivestreamIE, LivestreamOriginalIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mit import TechTVMITIE, MITIE @@ -116,19 +118,24 @@ from .slashdot import SlashdotIE from .slideshare import SlideshareIE from .sohu import SohuIE from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE -from .southparkstudios import SouthParkStudiosIE +from .southparkstudios import ( + SouthParkStudiosIE, + SouthparkDeIE, +) from .space import SpaceIE from .spankwire import SpankwireIE from .spiegel import SpiegelIE from .stanfordoc import StanfordOpenClassroomIE from .statigram import StatigramIE from .steam import SteamIE +from .streamcloud import StreamcloudIE from .sztvhu import SztvHuIE from .teamcoco import TeamcocoIE from .techtalks import TechTalksIE from .ted import TEDIE from .tf1 import TF1IE from .thisav import ThisAVIE +from .toutv import TouTvIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE from .tube8 import Tube8IE diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py new file mode 100644 index 000000000..691d5a844 --- /dev/null +++ b/youtube_dl/extractor/anitube.py @@ -0,0 +1,55 @@ +import re +import xml.etree.ElementTree + +from .common import InfoExtractor + + +class AnitubeIE(InfoExtractor): + IE_NAME = u'anitube.se' + _VALID_URL = r'https?://(?:www\.)?anitube\.se/video/(?P<id>\d+)' + + _TEST = { + u'url': u'http://www.anitube.se/video/36621', + u'md5': u'59d0eeae28ea0bc8c05e7af429998d43', + u'file': u'36621.mp4', + u'info_dict': { + u'id': u'36621', + u'ext': u'mp4', + u'title': u'Recorder to Randoseru 01', + }, + u'skip': u'Blocked in the US', + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + key = self._html_search_regex(r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', + webpage, u'key') + + webpage_config = self._download_webpage('http://www.anitube.se/nuevo/econfig.php?key=%s' % key, + key) + config_xml = xml.etree.ElementTree.fromstring(webpage_config.encode('utf-8')) + + video_title = config_xml.find('title').text + + formats = [] + video_url = config_xml.find('file') + if video_url is not None: + formats.append({ + 'format_id': 'sd', + 'url': video_url.text, + }) + video_url = config_xml.find('filehd') + if video_url is not None: + formats.append({ + 'format_id': 'hd', + 'url': video_url.text, + }) + + return { + 'id': video_id, + 'title': video_title, + 'formats': formats + } diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index b35a679e3..44d0b5d70 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -69,7 +69,7 @@ class ArteTvIE(InfoExtractor): lang = mobj.group('lang') return self._extract_liveweb(url, name, lang) - if re.search(self._LIVE_URL, video_id) is not None: + if re.search(self._LIVE_URL, url) is not None: raise ExtractorError(u'Arte live streams are not yet supported, sorry') # self.extractLiveStream(url) # return @@ -115,7 +115,7 @@ class ArteTvIE(InfoExtractor): event_doc = config_doc.find('event') url_node = event_doc.find('video').find('urlHd') if url_node is None: - url_node = video_doc.find('urlSd') + url_node = event_doc.find('urlSd') return {'id': video_id, 'title': event_doc.find('name%s' % lang.capitalize()).text, diff --git a/youtube_dl/extractor/auengine.py b/youtube_dl/extractor/auengine.py index 0febbff4f..95c038003 100644 --- a/youtube_dl/extractor/auengine.py +++ b/youtube_dl/extractor/auengine.py @@ -1,10 +1,10 @@ -import os.path import re from .common import InfoExtractor from ..utils import ( compat_urllib_parse, - compat_urllib_parse_urlparse, + determine_ext, + ExtractorError, ) class AUEngineIE(InfoExtractor): @@ -25,22 +25,25 @@ class AUEngineIE(InfoExtractor): title = self._html_search_regex(r'<title>(?P<title>.+?)</title>', webpage, u'title') title = title.strip() - links = re.findall(r'[^A-Za-z0-9]?(?:file|url):\s*["\'](http[^\'"&]*)', webpage) - links = [compat_urllib_parse.unquote(l) for l in links] + links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage) + links = map(compat_urllib_parse.unquote, links) + + thumbnail = None + video_url = None for link in links: - root, pathext = os.path.splitext(compat_urllib_parse_urlparse(link).path) - if pathext == '.png': + if link.endswith('.png'): thumbnail = link - elif pathext == '.mp4': - url = link - ext = pathext + elif '/videos/' in link: + video_url = link + if not video_url: + raise ExtractorError(u'Could not find video URL') + ext = u'.' + determine_ext(video_url) if ext == title[-len(ext):]: title = title[:-len(ext)] - ext = ext[1:] - return [{ + + return { 'id': video_id, - 'url': url, - 'ext': ext, + 'url': video_url, 'title': title, 'thumbnail': thumbnail, - }] + } diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py index f3b36f473..967568c4a 100644 --- a/youtube_dl/extractor/bambuser.py +++ b/youtube_dl/extractor/bambuser.py @@ -15,7 +15,8 @@ class BambuserIE(InfoExtractor): _TEST = { u'url': u'http://bambuser.com/v/4050584', - u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', + # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388 + #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', u'info_dict': { u'id': u'4050584', u'ext': u'flv', diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 129a20f44..359d4174b 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -3,13 +3,16 @@ import re from .common import InfoExtractor from ..utils import ( + compat_str, + compat_urlparse, ExtractorError, ) class BandcampIE(InfoExtractor): + IE_NAME = u'Bandcamp' _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)' - _TEST = { + _TESTS = [{ u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song', u'file': u'1812978515.mp3', u'md5': u'cdeb30cdae1921719a3cbcab696ef53c', @@ -17,7 +20,29 @@ class BandcampIE(InfoExtractor): u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad" }, u'skip': u'There is a limit of 200 free downloads / month for the test song' - } + }, { + u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', + u'playlist': [ + { + u'file': u'1353101989.mp3', + u'md5': u'39bc1eded3476e927c724321ddf116cf', + u'info_dict': { + u'title': u'Intro', + } + }, + { + u'file': u'38097443.mp3', + u'md5': u'1a2c32e2691474643e912cc6cd4bffaa', + u'info_dict': { + u'title': u'Kero One - Keep It Alive (Blazo remix)', + } + }, + ], + u'params': { + u'playlistend': 2 + }, + u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test' + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -26,6 +51,26 @@ class BandcampIE(InfoExtractor): # We get the link to the free download page m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) if m_download is None: + m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage) + if m_trackinfo: + json_code = m_trackinfo.group(1) + data = json.loads(json_code) + + entries = [] + for d in data: + formats = [{ + 'format_id': 'format_id', + 'url': format_url, + 'ext': format_id.partition('-')[0] + } for format_id, format_url in sorted(d['file'].items())] + entries.append({ + 'id': compat_str(d['id']), + 'title': d['title'], + 'formats': formats, + }) + + return self.playlist_result(entries, title, title) + else: raise ExtractorError(u'No free songs found') download_link = m_download.group(1) @@ -61,3 +106,25 @@ class BandcampIE(InfoExtractor): } return [track_info] + + +class BandcampAlbumIE(InfoExtractor): + IE_NAME = u'Bandcamp:album' + _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + title = mobj.group('title') + webpage = self._download_webpage(url, title) + tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage) + if not tracks_paths: + raise ExtractorError(u'The page doesn\'t contain any track') + entries = [ + self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key()) + for t_path in tracks_paths] + title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title') + return { + '_type': 'playlist', + 'title': title, + 'entries': entries, + } diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 1db9b24cf..bfa2a8b40 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -5,6 +5,7 @@ import xml.etree.ElementTree from .common import InfoExtractor from ..utils import unified_strdate + class CanalplusIE(InfoExtractor): _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))' _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s' @@ -25,7 +26,7 @@ class CanalplusIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = mobj.groupdict().get('id') if video_id is None: webpage = self._download_webpage(url, mobj.group('path')) video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id') diff --git a/youtube_dl/extractor/collegehumor.py b/youtube_dl/extractor/collegehumor.py index 8d4c93d6d..0c29acfb1 100644 --- a/youtube_dl/extractor/collegehumor.py +++ b/youtube_dl/extractor/collegehumor.py @@ -71,10 +71,8 @@ class CollegeHumorIE(InfoExtractor): adoc = xml.etree.ElementTree.fromstring(manifestXml) try: - media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0] - node_id = media_node.attrib['url'] video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text - except IndexError as err: + except IndexError: raise ExtractorError(u'Invalid manifest file') url_pr = compat_urllib_parse_urlparse(info['thumbnail']) info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','') diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 9c20d30b4..eb3435c77 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -71,6 +71,10 @@ class InfoExtractor(object): ("3D" or "DASH video") * width Width of the video, if known * height Height of the video, if known + * abr Average audio bitrate in KBit/s + * acodec Name of the audio codec in use + * vbr Average video bitrate in KBit/s + * vcodec Name of the video codec in use webpage_url: The url to the video webpage, if given to youtube-dl it should allow to get the same result again. (It will be set by YoutubeDL if it's missing) @@ -315,13 +319,19 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod - def _og_regex(prop): - return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop) + def _og_regexes(prop): + content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')' + property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop) + template = r'<meta[^>]+?%s[^>]+?%s' + return [ + template % (property_re, content_re), + template % (content_re, property_re), + ] def _og_search_property(self, prop, html, name=None, **kargs): if name is None: name = 'OpenGraph %s' % prop - escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs) + escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) if escaped is None: return None return unescapeHTML(escaped) @@ -336,10 +346,21 @@ class InfoExtractor(object): return self._og_search_property('title', html, **kargs) def _og_search_video_url(self, html, name='video url', secure=True, **kargs): - regexes = [self._og_regex('video')] - if secure: regexes.insert(0, self._og_regex('video:secure_url')) + regexes = self._og_regexes('video') + if secure: regexes = self._og_regexes('video:secure_url') + regexes return self._html_search_regex(regexes, html, name, **kargs) + def _html_search_meta(self, name, html, display_name=None): + if display_name is None: + display_name = name + return self._html_search_regex( + r'''(?ix)<meta(?=[^>]+(?:name|property)=["\']%s["\']) + [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), + html, display_name, fatal=False) + + def _dc_search_uploader(self, html): + return self._html_search_meta('dc.creator', html, 'uploader') + def _rta_search(self, html): # See http://www.rtalabel.org/index.php?content=howtofaq#single if re.search(r'(?ix)<meta\s+name="rating"\s+' @@ -348,6 +369,23 @@ class InfoExtractor(object): return 18 return 0 + def _media_rating_search(self, html): + # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ + rating = self._html_search_meta('rating', html) + + if not rating: + return None + + RATING_TABLE = { + 'safe for kids': 0, + 'general': 8, + '14 years': 14, + 'mature': 17, + 'restricted': 19, + } + return RATING_TABLE.get(rating.lower(), None) + + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/d8.py b/youtube_dl/extractor/d8.py new file mode 100644 index 000000000..a56842b16 --- /dev/null +++ b/youtube_dl/extractor/d8.py @@ -0,0 +1,22 @@ +# encoding: utf-8 +from .canalplus import CanalplusIE + + +class D8IE(CanalplusIE): + _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)' + _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s' + IE_NAME = u'd8.tv' + + _TEST = { + u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html', + u'file': u'966289.flv', + u'info_dict': { + u'title': u'Campagne intime - Documentaire exceptionnel', + u'description': u'md5:d2643b799fb190846ae09c61e59a859f', + u'upload_date': u'20131108', + }, + u'params': { + # rtmp + u'skip_download': True, + }, + } diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index e87690f9d..71f5e03ee 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -186,7 +186,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): webpage = self._download_webpage(request, id, u'Downloading page %s' % pagenum) - playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) + playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage) video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: diff --git a/youtube_dl/extractor/eighttracks.py b/youtube_dl/extractor/eighttracks.py index 2cfbcd363..f21ef8853 100644 --- a/youtube_dl/extractor/eighttracks.py +++ b/youtube_dl/extractor/eighttracks.py @@ -1,4 +1,3 @@ -import itertools import json import random import re diff --git a/youtube_dl/extractor/escapist.py b/youtube_dl/extractor/escapist.py index 3aa2da52c..b1242f6bc 100644 --- a/youtube_dl/extractor/escapist.py +++ b/youtube_dl/extractor/escapist.py @@ -11,11 +11,11 @@ from ..utils import ( class EscapistIE(InfoExtractor): - _VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$' + _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$' _TEST = { u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate', u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4', - u'md5': u'c6793dbda81388f4264c1ba18684a74d', + u'md5': u'ab3a706c681efca53f0a35f1415cf0d1', u'info_dict': { u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", u"uploader": u"the-escapist-presents", @@ -25,50 +25,60 @@ class EscapistIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) showName = mobj.group('showname') videoId = mobj.group('episode') self.report_extraction(videoId) webpage = self._download_webpage(url, videoId) - videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"', + videoDesc = self._html_search_regex( + r'<meta name="description" content="([^"]*)"', webpage, u'description', fatal=False) - playerUrl = self._og_search_video_url(webpage, name='player url') + playerUrl = self._og_search_video_url(webpage, name=u'player URL') - title = self._html_search_regex('<meta name="title" content="([^"]*)"', - webpage, u'player url').split(' : ')[-1] + title = self._html_search_regex( + r'<meta name="title" content="([^"]*)"', + webpage, u'title').split(' : ')[-1] - configUrl = self._search_regex('config=(.*)$', playerUrl, u'config url') + configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL') configUrl = compat_urllib_parse.unquote(configUrl) - configJSON = self._download_webpage(configUrl, videoId, - u'Downloading configuration', - u'unable to download configuration') - - # Technically, it's JavaScript, not JSON - configJSON = configJSON.replace("'", '"') - + formats = [] + + def _add_format(name, cfgurl): + configJSON = self._download_webpage( + cfgurl, videoId, + u'Downloading ' + name + ' configuration', + u'Unable to download ' + name + ' configuration') + + # Technically, it's JavaScript, not JSON + configJSON = configJSON.replace("'", '"') + + try: + config = json.loads(configJSON) + except (ValueError,) as err: + raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err)) + playlist = config['playlist'] + formats.append({ + 'url': playlist[1]['url'], + 'format_id': name, + }) + + _add_format(u'normal', configUrl) + hq_url = (configUrl + + ('&hq=1' if '?' in configUrl else configUrl + '?hq=1')) try: - config = json.loads(configJSON) - except (ValueError,) as err: - raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err)) + _add_format(u'hq', hq_url) + except ExtractorError: + pass # That's fine, we'll just use normal quality - playlist = config['playlist'] - videoUrl = playlist[1]['url'] - - info = { + return { 'id': videoId, - 'url': videoUrl, + 'formats': formats, 'uploader': showName, - 'upload_date': None, 'title': title, - 'ext': 'mp4', 'thumbnail': self._og_search_thumbnail(webpage), 'description': videoDesc, 'player_url': playerUrl, } - - return [info] diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index f8bdfc2d3..3b210710e 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -1,5 +1,4 @@ import json -import netrc import re import socket diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py index 9c89362ef..dba1a8dc2 100644 --- a/youtube_dl/extractor/fktv.py +++ b/youtube_dl/extractor/fktv.py @@ -39,7 +39,6 @@ class FKTVIE(InfoExtractor): for i, _ in enumerate(files, 1): video_id = '%04d%d' % (episode, i) video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i) - video_title = 'Fernsehkritik %d.%d' % (episode, i) videos.append({ 'id': video_id, 'url': video_url, diff --git a/youtube_dl/extractor/gamekings.py b/youtube_dl/extractor/gamekings.py index 4b4259447..c91669b0e 100644 --- a/youtube_dl/extractor/gamekings.py +++ b/youtube_dl/extractor/gamekings.py @@ -1,9 +1,6 @@ import re from .common import InfoExtractor -from ..utils import ( - determine_ext, -) class GamekingsIE(InfoExtractor): @@ -11,7 +8,8 @@ class GamekingsIE(InfoExtractor): _TEST = { u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/", u'file': u'20130811.mp4', - u'md5': u'17f6088f7d0149ff2b46f2714bdb1954', + # MD5 is flaky, seems to change regularly + #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3', u'info_dict': { u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review", u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.", diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 098768361..9645b00c3 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -24,7 +24,7 @@ class GameSpotIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - page_id = video_id = mobj.group('page_id') + page_id = mobj.group('page_id') webpage = self._download_webpage(url, page_id) data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video') data_video = json.loads(unescapeHTML(data_video_json)) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index c7552fddb..0b5f2b2bb 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -162,6 +162,16 @@ class GenericIE(InfoExtractor): raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # 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(r'<title>(.*)</title>', + webpage, u'video title', default=u'video', flags=re.DOTALL) + # Look for BrightCove: bc_url = BrightcoveIE._extract_brightcove_url(webpage) if bc_url is not None: @@ -177,17 +187,20 @@ class GenericIE(InfoExtractor): return self.url_result(surl, 'Vimeo') # Look for embedded YouTube player - mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage) - if mobj: - surl = unescapeHTML(mobj.group(u'url')) - return self.url_result(surl, 'Youtube') + matches = re.findall( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\1', webpage) + if matches: + urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') + for tuppl in matches] + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) # Look for Bandcamp pages with custom domain mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) if mobj is not None: burl = unescapeHTML(mobj.group(1)) - return self.url_result(burl, 'Bandcamp') + # Don't set the extractor because it can be a track url or an album + return self.url_result(burl) # Start with something easy: JW Player in SWFObject mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) @@ -226,15 +239,6 @@ class GenericIE(InfoExtractor): video_extension = os.path.splitext(video_id)[1][1:] video_id = os.path.splitext(video_id)[0] - # it's tempting to parse this further, but you would - # have to take into account all the variations like - # Video Title - Site Name - # 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(r'<title>(.*)</title>', - webpage, u'video title', default=u'video', flags=re.DOTALL) - # video uploader is domain name video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*', url, u'video uploader') diff --git a/youtube_dl/extractor/jeuxvideo.py b/youtube_dl/extractor/jeuxvideo.py index 6bb54b932..0020c47cf 100644 --- a/youtube_dl/extractor/jeuxvideo.py +++ b/youtube_dl/extractor/jeuxvideo.py @@ -22,7 +22,7 @@ class JeuxVideoIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - title = re.match(self._VALID_URL, url).group(1) + title = mobj.group(1) webpage = self._download_webpage(url, title) xml_link = self._html_search_regex( r'<param name="flashvars" value="config=(.*?)" />', diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index 4531fd6ab..5f548437c 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -1,16 +1,17 @@ import re import json +import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( compat_urllib_parse_urlparse, compat_urlparse, - get_meta_content, - ExtractorError, + xpath_with_ns, ) class LivestreamIE(InfoExtractor): + IE_NAME = u'livestream' _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$' _TEST = { u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370', @@ -54,3 +55,44 @@ class LivestreamIE(InfoExtractor): info = json.loads(self._download_webpage(api_url, video_id, u'Downloading video info')) return self._extract_video_info(info) + + +# The original version of Livestream uses a different system +class LivestreamOriginalIE(InfoExtractor): + IE_NAME = u'livestream:original' + _VALID_URL = r'https?://www\.livestream\.com/(?P<user>[^/]+)/video\?.*?clipId=(?P<id>.*?)(&|$)' + _TEST = { + u'url': u'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', + u'info_dict': { + u'id': u'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', + u'ext': u'flv', + u'title': u'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital', + }, + u'params': { + # rtmp + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + user = mobj.group('user') + api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id) + + api_response = self._download_webpage(api_url, video_id) + info = xml.etree.ElementTree.fromstring(api_response.encode('utf-8')) + item = info.find('channel').find('item') + ns = {'media': 'http://search.yahoo.com/mrss'} + thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url'] + # Remove the extension and number from the path (like 1.jpg) + path = self._search_regex(r'(user-files/.+)_.*?\.jpg$', thumbnail_url, u'path') + + return { + 'id': video_id, + 'title': item.find('title').text, + 'url': 'rtmp://extondemand.livestream.com/ondemand', + 'play_path': 'mp4:trans/dv15/mogulus-{0}.mp4'.format(path), + 'ext': 'flv', + 'thumbnail': thumbnail_url, + } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 24a79ae13..04afd6c4c 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -48,7 +48,7 @@ class MTVIE(InfoExtractor): def _transform_rtmp_url(rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url) if not m: - raise ExtractorError(u'Cannot transform RTMP url') + return rtmp_video_url base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' return base + m.group('finalid') @@ -59,7 +59,6 @@ class MTVIE(InfoExtractor): if '/error_country_block.swf' in metadataXml: raise ExtractorError(u'This video is not available from your country.', expected=True) mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8')) - renditions = mdoc.findall('.//rendition') formats = [] for rendition in mdoc.findall('.//rendition'): diff --git a/youtube_dl/extractor/nhl.py b/youtube_dl/extractor/nhl.py index 224f56ac8..458fe4063 100644 --- a/youtube_dl/extractor/nhl.py +++ b/youtube_dl/extractor/nhl.py @@ -72,7 +72,7 @@ class NHLIE(NHLBaseInfoExtractor): class NHLVideocenterIE(NHLBaseInfoExtractor): IE_NAME = u'nhl.com:videocenter' - IE_DESC = u'Download the first 12 videos from a videocenter category' + IE_DESC = u'NHL videocenter category' _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?' @classmethod diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 75cf4bb9f..8b3471919 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -6,7 +6,6 @@ from ..utils import ( compat_urllib_parse_urlparse, compat_urllib_request, compat_urllib_parse, - unescapeHTML, ) from ..aes import ( aes_decrypt_text diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py index 994778e16..3bbda128e 100644 --- a/youtube_dl/extractor/redtube.py +++ b/youtube_dl/extractor/redtube.py @@ -8,7 +8,9 @@ class RedTubeIE(InfoExtractor): _TEST = { u'url': u'http://www.redtube.com/66418', u'file': u'66418.mp4', - u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d', + # md5 varies from time to time, as in + # https://travis-ci.org/rg3/youtube-dl/jobs/14052463#L295 + #u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d', u'info_dict': { u"title": u"Sucked on a toilet", u"age_limit": 18, diff --git a/youtube_dl/extractor/rtlnow.py b/youtube_dl/extractor/rtlnow.py index 9ac7c3be8..2f238de35 100644 --- a/youtube_dl/extractor/rtlnow.py +++ b/youtube_dl/extractor/rtlnow.py @@ -63,18 +63,6 @@ class RTLnowIE(InfoExtractor): }, }, { - u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1', - u'file': u'129679.flv', - u'info_dict': { - u'upload_date': u'20131016', - u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...', - u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig', - }, - u'params': { - u'skip_download': True, - }, - }, - { u'url': u'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10', u'file': u'124903.flv', u'info_dict': { diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 4717fbb77..67b2dff9c 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -59,6 +59,7 @@ class SoundcloudIE(InfoExtractor): ] _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28' + _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf' @classmethod def suitable(cls, url): @@ -83,28 +84,37 @@ class SoundcloudIE(InfoExtractor): thumbnail = thumbnail.replace('-large', '-t500x500') result = { 'id': track_id, - 'url': info['stream_url'] + '?client_id=' + self._CLIENT_ID, 'uploader': info['user']['username'], 'upload_date': unified_strdate(info['created_at']), 'title': info['title'], - 'ext': u'mp3', + 'ext': info.get('original_format', u'mp3'), 'description': info['description'], 'thumbnail': thumbnail, } if info.get('downloadable', False): + # We can build a direct link to the song result['url'] = 'https://api.soundcloud.com/tracks/{0}/download?client_id={1}'.format(track_id, self._CLIENT_ID) - if not info.get('streamable', False): - # We have to get the rtmp url + else: + # We have to retrieve the url stream_json = self._download_webpage( - 'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._CLIENT_ID), + 'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID), track_id, u'Downloading track url') - rtmp_url = json.loads(stream_json)['rtmp_mp3_128_url'] - # The url doesn't have an rtmp app, we have to extract the playpath - url, path = rtmp_url.split('mp3:', 1) - result.update({ - 'url': url, - 'play_path': 'mp3:' + path, - }) + # There should be only one entry in the dictionary + key, stream_url = list(json.loads(stream_json).items())[0] + if key.startswith(u'http'): + result['url'] = stream_url + elif key.startswith(u'rtmp'): + # The url doesn't have an rtmp app, we have to extract the playpath + url, path = stream_url.split('mp3:', 1) + result.update({ + 'url': url, + 'play_path': 'mp3:' + path, + }) + else: + # We fallback to the stream_url in the original info, this + # cannot be always used, sometimes it can give an HTTP 404 error + result['url'] = info['stream_url'] + '?client_id=' + self._CLIENT_ID, + return result def _real_extract(self, url): @@ -158,7 +168,6 @@ class SoundcloudSetIE(SoundcloudIE): resolv_url = self._resolv_url(url) info_json = self._download_webpage(resolv_url, full_title) - videos = [] info = json.loads(info_json) if 'errors' in info: for err in info['errors']: diff --git a/youtube_dl/extractor/southparkstudios.py b/youtube_dl/extractor/southparkstudios.py index b1e96b679..a711531e6 100644 --- a/youtube_dl/extractor/southparkstudios.py +++ b/youtube_dl/extractor/southparkstudios.py @@ -5,21 +5,19 @@ from .mtv import MTVIE, _media_xml_tag class SouthParkStudiosIE(MTVIE): IE_NAME = u'southparkstudios.com' - _VALID_URL = r'https?://www\.southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$)' + _VALID_URL = r'(https?://)?(www\.)?(?P<url>southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$))' _FEED_URL = 'http://www.southparkstudios.com/feeds/video-player/mrss' - _TEST = { + # Overwrite MTVIE properties we don't want + _TESTS = [{ u'url': u'http://www.southparkstudios.com/clips/104437/bat-daded#tab=featured', u'file': u'a7bff6c2-ed00-11e0-aca6-0026b9414f30.mp4', u'info_dict': { u'title': u'Bat Daded', u'description': u'Randy disqualifies South Park by getting into a fight with Bat Dad.', }, - } - - # Overwrite MTVIE properties we don't want - _TESTS = [] + }] def _get_thumbnail_url(self, uri, itemdoc): search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) @@ -31,8 +29,23 @@ class SouthParkStudiosIE(MTVIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) + url = u'http://www.' + mobj.group(u'url') video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) mgid = self._search_regex(r'swfobject.embedSWF\(".*?(mgid:.*?)"', webpage, u'mgid') return self._get_videos_info(mgid) + +class SouthparkDeIE(SouthParkStudiosIE): + IE_NAME = u'southpark.de' + _VALID_URL = r'(https?://)?(www\.)?(?P<url>southpark\.de/(clips|alle-episoden)/(?P<id>.+?)(\?|#|$))' + _FEED_URL = 'http://www.southpark.de/feeds/video-player/mrss/' + + _TESTS = [{ + u'url': u'http://www.southpark.de/clips/uygssh/the-government-wont-respect-my-privacy#tab=featured', + u'file': u'85487c96-b3b9-4e39-9127-ad88583d9bf2.mp4', + u'info_dict': { + u'title': u'The Government Won\'t Respect My Privacy', + u'description': u'Cartman explains the benefits of "Shitter" to Stan, Kyle and Craig.', + }, + }] diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py index 97f9c268a..9e2ad0d99 100644 --- a/youtube_dl/extractor/spankwire.py +++ b/youtube_dl/extractor/spankwire.py @@ -6,7 +6,6 @@ from ..utils import ( compat_urllib_parse_urlparse, compat_urllib_request, compat_urllib_parse, - unescapeHTML, ) from ..aes import ( aes_decrypt_text @@ -36,11 +35,12 @@ class SpankwireIE(InfoExtractor): webpage = self._download_webpage(req, video_id) video_title = self._html_search_regex(r'<h1>([^<]+)', webpage, u'title') - video_uploader = self._html_search_regex(r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False) - thumbnail = self._html_search_regex(r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False) - description = self._html_search_regex(r'>\s*Description:</div>\s*<[^>]*>([^<]+)', webpage, u'description', fatal=False) - if len(description) == 0: - description = None + video_uploader = self._html_search_regex( + r'by:\s*<a [^>]*>(.+?)</a>', webpage, u'uploader', fatal=False) + thumbnail = self._html_search_regex( + r'flashvars\.image_url = "([^"]+)', webpage, u'thumbnail', fatal=False) + description = self._html_search_regex( + r'<div\s+id="descriptionContent">([^<]+)<', webpage, u'description', fatal=False) video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'flashvars\.quality_[0-9]{3}p = "([^"]+)', webpage))) if webpage.find('flashvars\.encrypted = "true"') != -1: diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 13c86401c..19ce585cf 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -6,14 +6,22 @@ from .common import InfoExtractor class SpiegelIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$' - _TEST = { + _TESTS = [{ u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', u'file': u'1259285.mp4', u'md5': u'2c2754212136f35fb4b19767d242f66e', u'info_dict': { u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv" } - } + }, + { + u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html', + u'file': u'1309159.mp4', + u'md5': u'f2cdf638d7aa47654e251e1aee360af1', + u'info_dict': { + u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers' + } + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) @@ -21,25 +29,38 @@ class SpiegelIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>', - webpage, u'title') + video_title = self._html_search_regex( + r'<div class="module-title">(.*?)</div>', webpage, u'title') xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' - xml_code = self._download_webpage(xml_url, video_id, - note=u'Downloading XML', errnote=u'Failed to download XML') + xml_code = self._download_webpage( + xml_url, video_id, + note=u'Downloading XML', errnote=u'Failed to download XML') idoc = xml.etree.ElementTree.fromstring(xml_code) - last_type = idoc[-1] - filename = last_type.findall('./filename')[0].text - duration = float(last_type.findall('./duration')[0].text) - video_url = 'http://video2.spiegel.de/flash/' + filename - video_ext = filename.rpartition('.')[2] + formats = [ + { + 'format_id': n.tag.rpartition('type')[2], + 'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text, + 'width': int(n.find('./width').text), + 'height': int(n.find('./height').text), + 'abr': int(n.find('./audiobitrate').text), + 'vbr': int(n.find('./videobitrate').text), + 'vcodec': n.find('./codec').text, + 'acodec': 'MP4A', + } + for n in list(idoc) + # Blacklist type 6, it's extremely LQ and not available on the same server + if n.tag.startswith('type') and n.tag != 'type6' + ] + formats.sort(key=lambda f: f['vbr']) + duration = float(idoc[0].findall('./duration')[0].text) + info = { 'id': video_id, - 'url': video_url, - 'ext': video_ext, 'title': video_title, 'duration': duration, + 'formats': formats, } - return [info] + return info diff --git a/youtube_dl/extractor/streamcloud.py b/youtube_dl/extractor/streamcloud.py new file mode 100644 index 000000000..d476693ec --- /dev/null +++ b/youtube_dl/extractor/streamcloud.py @@ -0,0 +1,65 @@ +# coding: utf-8 +import re +import time + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse, + compat_urllib_request, +) + + +class StreamcloudIE(InfoExtractor): + IE_NAME = u'streamcloud.eu' + _VALID_URL = r'https?://streamcloud\.eu/(?P<id>[a-zA-Z0-9_-]+)/(?P<fname>[^#?]*)\.html' + + _TEST = { + u'url': u'http://streamcloud.eu/skp9j99s4bpz/youtube-dl_test_video_____________-BaW_jenozKc.mp4.html', + u'file': u'skp9j99s4bpz.mp4', + u'md5': u'6bea4c7fa5daaacc2a946b7146286686', + u'info_dict': { + u'title': u'youtube-dl test video \'/\\ ä ↭', + u'duration': 9, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + orig_webpage = self._download_webpage(url, video_id) + + fields = re.findall(r'''(?x)<input\s+ + type="(?:hidden|submit)"\s+ + name="([^"]+)"\s+ + (?:id="[^"]+"\s+)? + value="([^"]*)" + ''', orig_webpage) + post = compat_urllib_parse.urlencode(fields) + + self.to_screen('%s: Waiting for timeout' % video_id) + time.sleep(12) + headers = { + b'Content-Type': b'application/x-www-form-urlencoded', + } + req = compat_urllib_request.Request(url, post, headers) + + webpage = self._download_webpage( + req, video_id, note=u'Downloading video page ...') + title = self._html_search_regex( + r'<h1[^>]*>([^<]+)<', webpage, u'title') + video_url = self._search_regex( + r'file:\s*"([^"]+)"', webpage, u'video URL') + duration_str = self._search_regex( + r'duration:\s*"?([0-9]+)"?', webpage, u'duration', fatal=False) + duration = None if duration_str is None else int(duration_str) + thumbnail = self._search_regex( + r'image:\s*"([^"]+)"', webpage, u'thumbnail URL', fatal=False) + + return { + 'id': video_id, + 'title': title, + 'url': video_url, + 'duration': duration, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/sztvhu.py b/youtube_dl/extractor/sztvhu.py index 81fa35c4b..c9359fafb 100644 --- a/youtube_dl/extractor/sztvhu.py +++ b/youtube_dl/extractor/sztvhu.py @@ -15,7 +15,8 @@ class SztvHuIE(InfoExtractor): u'info_dict': { u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren", u"description": u'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...', - } + }, + u'skip': u'Service temporarily disabled as of 2013-11-20' } def _real_extract(self, url): diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py index bc48620f0..165d9f88b 100644 --- a/youtube_dl/extractor/teamcoco.py +++ b/youtube_dl/extractor/teamcoco.py @@ -60,7 +60,7 @@ class TeamcocoIE(InfoExtractor): return -1 formats.sort(key=sort_key) if not formats: - raise RegexNotFoundError(u'Unable to extract video URL') + raise ExtractorError(u'Unable to extract video URL') return { 'id': video_id, diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 76cfdfb90..4bca62ba0 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -4,7 +4,6 @@ import re from .subtitles import SubtitlesInfoExtractor from ..utils import ( - compat_str, RegexNotFoundError, ) @@ -43,26 +42,25 @@ class TEDIE(SubtitlesInfoExtractor): self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) return [self._playlist_videos_info(url,name,playlist_id)] - def _playlist_videos_info(self,url,name,playlist_id=0): + + def _playlist_videos_info(self, url, name, playlist_id): '''Returns the videos of the playlist''' - video_RE=r''' - <li\ id="talk_(\d+)"([.\s]*?)data-id="(?P<video_id>\d+)" - ([.\s]*?)data-playlist_item_id="(\d+)" - ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)" - ''' - video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>' - webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage') - m_videos=re.finditer(video_RE,webpage,re.VERBOSE) - m_names=re.finditer(video_name_RE,webpage) + + webpage = self._download_webpage( + url, playlist_id, u'Downloading playlist webpage') + matches = re.finditer( + r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>', + webpage) playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>', webpage, 'playlist title') - playlist_entries = [] - for m_video, m_name in zip(m_videos,m_names): - talk_url='http://www.ted.com%s' % m_name.group('talk_url') - playlist_entries.append(self.url_result(talk_url, 'TED')) - return self.playlist_result(playlist_entries, playlist_id = playlist_id, playlist_title = playlist_title) + playlist_entries = [ + self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED') + for m in matches + ] + return self.playlist_result( + playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title) def _talk_info(self, url, video_id=0): """Return the video for the talk in the url""" @@ -85,7 +83,7 @@ class TEDIE(SubtitlesInfoExtractor): 'ext': 'mp4', 'url': stream['file'], 'format': stream['id'] - } for stream in info['htmlStreams']] + } for stream in info['htmlStreams']] video_id = info['id'] @@ -95,7 +93,7 @@ class TEDIE(SubtitlesInfoExtractor): self._list_available_subtitles(video_id, webpage) return - info = { + return { 'id': video_id, 'title': title, 'thumbnail': thumbnail, @@ -104,11 +102,6 @@ class TEDIE(SubtitlesInfoExtractor): 'formats': formats, } - # TODO: Remove when #980 has been merged - info.update(info['formats'][-1]) - - return info - def _get_available_subtitles(self, video_id, webpage): try: options = self._search_regex(r'(?:<select name="subtitles_language_select" id="subtitles_language_select">)(.*?)(?:</select>)', webpage, 'subtitles_language_select', flags=re.DOTALL) @@ -119,6 +112,6 @@ class TEDIE(SubtitlesInfoExtractor): url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l) sub_lang_list[l] = url return sub_lang_list - except RegexNotFoundError as err: + except RegexNotFoundError: self._downloader.report_warning(u'video doesn\'t have subtitles') return {} diff --git a/youtube_dl/extractor/toutv.py b/youtube_dl/extractor/toutv.py new file mode 100644 index 000000000..2f728d3dc --- /dev/null +++ b/youtube_dl/extractor/toutv.py @@ -0,0 +1,74 @@ +# coding: utf-8 +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + unified_strdate, +) + + +class TouTvIE(InfoExtractor): + IE_NAME = u'tou.tv' + _VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))' + + _TEST = { + u'url': u'http://www.tou.tv/30-vies/S04E41', + u'file': u'30-vies_S04E41.mp4', + u'info_dict': { + u'title': u'30 vies Saison 4 / Épisode 41', + u'description': u'md5:da363002db82ccbe4dafeb9cab039b09', + u'age_limit': 8, + u'uploader': u'Groupe des Nouveaux Médias', + u'duration': 1296, + u'upload_date': u'20131118', + u'thumbnail': u'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg', + }, + u'params': { + u'skip_download': True, # Requires rtmpdump + }, + u'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, u'media ID') + + streams_url = u'http://release.theplatform.com/content.select?pid=' + mediaId + streams_webpage = self._download_webpage( + streams_url, video_id, note=u'Downloading stream list') + + streams_doc = xml.etree.ElementTree.fromstring( + streams_webpage.encode('utf-8')) + video_url = next(n.text + for n in streams_doc.findall('.//choice/url') + if u'//ad.doubleclick' not in n.text) + if video_url.endswith('/Unavailable.flv'): + raise ExtractorError( + u'Access to this video is blocked from outside of Canada', + expected=True) + + duration_str = self._html_search_meta( + 'video:duration', webpage, u'duration') + duration = int(duration_str) if duration_str else None + upload_date_str = self._html_search_meta( + 'video:release_date', webpage, u'upload date') + upload_date = unified_strdate(upload_date_str) if upload_date_str else None + + return { + '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', + } diff --git a/youtube_dl/extractor/tube8.py b/youtube_dl/extractor/tube8.py index d4b7603c7..4d9d41db3 100644 --- a/youtube_dl/extractor/tube8.py +++ b/youtube_dl/extractor/tube8.py @@ -5,8 +5,6 @@ from .common import InfoExtractor from ..utils import ( compat_urllib_parse_urlparse, compat_urllib_request, - compat_urllib_parse, - unescapeHTML, ) from ..aes import ( aes_decrypt_text diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index 32e0f5037..bfed9dd04 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -13,9 +13,10 @@ class TvpIE(InfoExtractor): u'md5': u'148408967a6a468953c0a75cbdaf0d7a', u'file': u'12878238.wmv', u'info_dict': { - u'title': u'31.10.2013', - u'description': u'31.10.2013', + u'title': u'31.10.2013 - Odcinek 2', + u'description': u'31.10.2013 - Odcinek 2', }, + u'skip': u'Download has to use same server IP as extraction. Therefore, a good (load-balancing) DNS resolver will make the download fail.' } def _real_extract(self, url): diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 3f6020f74..4378b1780 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -78,12 +78,13 @@ class VevoIE(InfoExtractor): continue format_url = self._SMIL_BASE_URL + m.group('path') - format_note = ('%(vcodec)s@%(vbr)4sk, %(acodec)s@%(abr)3sk' % - m.groupdict()) formats.append({ 'url': format_url, 'format_id': u'SMIL_' + m.group('cbr'), - 'format_note': format_note, + 'vcodec': m.group('vcodec'), + 'acodec': m.group('acodec'), + 'vbr': int(m.group('vbr')), + 'abr': int(m.group('abr')), 'ext': m.group('ext'), 'width': int(m.group('width')), 'height': int(m.group('height')), diff --git a/youtube_dl/extractor/videopremium.py b/youtube_dl/extractor/videopremium.py index 65f39b982..4800415bd 100644 --- a/youtube_dl/extractor/videopremium.py +++ b/youtube_dl/extractor/videopremium.py @@ -24,12 +24,16 @@ class VideoPremiumIE(InfoExtractor): webpage_url = 'http://videopremium.tv/' + video_id webpage = self._download_webpage(webpage_url, video_id) - self.report_extraction(video_id) + if re.match(r"^<html><head><script[^>]*>window.location\s*=", webpage): + # Download again, we need a cookie + webpage = self._download_webpage( + webpage_url, video_id, + note=u'Downloading webpage again (with cookie)') - video_title = self._html_search_regex(r'<h2(?:.*?)>\s*(.+?)\s*<', - webpage, u'video title') + video_title = self._html_search_regex( + r'<h2(?:.*?)>\s*(.+?)\s*<', webpage, u'video title') - return [{ + return { 'id': video_id, 'url': "rtmp://e%d.md.iplay.md/play" % random.randint(1, 16), 'play_path': "mp4:%s.f4v" % video_id, @@ -37,4 +41,4 @@ class VideoPremiumIE(InfoExtractor): 'player_url': "http://videopremium.tv/uplayer/uppod.swf", 'ext': 'f4v', 'title': video_title, - }] + }
\ No newline at end of file diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index d465bf20b..7d82c2cfa 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -151,7 +151,7 @@ class VimeoIE(InfoExtractor): config = json.loads(config_json) except RegexNotFoundError: # For pro videos or player.vimeo.com urls - config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'], + config = self._search_regex([r' = {config:({.+?}),assets:', r'(?:c|b)=({.+?});'], webpage, u'info section', flags=re.DOTALL) config = json.loads(config) except Exception as e: diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 03ad88bed..e3458d2bd 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -5,7 +5,6 @@ from .common import InfoExtractor from ..utils import ( compat_urllib_parse_urlparse, compat_urllib_request, - compat_urllib_parse, ) class XTubeIE(InfoExtractor): diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index c48c0e24f..9b09793eb 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -139,10 +139,10 @@ class YoutubeBaseInfoExtractor(InfoExtractor): class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): IE_DESC = u'YouTube.com' - _VALID_URL = r"""^ + _VALID_URL = r"""(?x)^ ( - (?:https?://)? # http(s):// (optional) - (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/| + (?:https?://|//)? # http(s):// or protocol-independent URL (optional) + (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| tube\.majestyc\.net/| youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls @@ -363,6 +363,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): u"uploader_id": u"justintimberlakeVEVO" } }, + { + u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", + u"file": u"yZIXLfi8CZQ.mp4", + u"note": u"Embed-only video (#1746)", + u"info_dict": { + u"upload_date": u"20120608", + u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", + u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", + u"uploader": u"SET India", + u"uploader_id": u"setindia" + } + }, ] @@ -370,7 +382,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" if YoutubePlaylistIE.suitable(url): return False - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + return re.match(cls._VALID_URL, url) is not None def __init__(self, *args, **kwargs): super(YoutubeIE, self).__init__(*args, **kwargs) @@ -1019,6 +1031,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): """Turn the encrypted s field into a working signature""" if player_url is not None: + if player_url.startswith(u'//'): + player_url = u'https:' + player_url try: player_id = (player_url, len(s)) if player_id not in self._player_cache: @@ -1098,7 +1112,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): params = compat_urllib_parse.urlencode({ 'lang': lang, 'v': video_id, - 'fmt': self._downloader.params.get('subtitlesformat'), + 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), 'name': l[0].encode('utf-8'), }) url = u'http://www.youtube.com/api/timedtext?' + params @@ -1111,7 +1125,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def _get_available_automatic_caption(self, video_id, webpage): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" - sub_format = self._downloader.params.get('subtitlesformat') + sub_format = self._downloader.params.get('subtitlesformat', 'srt') self.to_screen(u'%s: Looking for automatic captions' % video_id) mobj = re.search(r';ytplayer.config = ({.*?});', webpage) err_msg = u'Couldn\'t find automatic captions for %s' % video_id @@ -1270,7 +1284,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # We simulate the access to the video from www.youtube.com/v/{video_id} # this can be viewed without login into Youtube data = compat_urllib_parse.urlencode({'video_id': video_id, - 'el': 'embedded', + 'el': 'player_embedded', 'gl': 'US', 'hl': 'en', 'eurl': 'https://youtube.googleapis.com/v/' + video_id, @@ -1299,6 +1313,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: raise ExtractorError(u'"token" parameter not in video info for unknown reason') + if 'view_count' in video_info: + view_count = int(video_info['view_count'][0]) + else: + view_count = None + # Check for "rental" videos if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: raise ExtractorError(u'"rental" videos not supported') @@ -1487,6 +1506,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'age_limit': 18 if age_gate else 0, 'annotations': video_annotations, 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, + 'view_count': view_count, }) return results @@ -1578,20 +1598,31 @@ class YoutubeChannelIE(InfoExtractor): # Download channel page channel_id = mobj.group(1) video_ids = [] + url = 'https://www.youtube.com/channel/%s/videos' % channel_id + channel_page = self._download_webpage(url, channel_id) + if re.search(r'channel-header-autogenerated-label', channel_page) is not None: + autogenerated = True + else: + autogenerated = False - # Download all channel pages using the json-based channel_ajax query - for pagenum in itertools.count(1): - url = self._MORE_PAGES_URL % (pagenum, channel_id) - page = self._download_webpage(url, channel_id, - u'Downloading page #%s' % pagenum) - - page = json.loads(page) - - ids_in_page = self.extract_videos_from_page(page['content_html']) - video_ids.extend(ids_in_page) - - if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: - break + if autogenerated: + # The videos are contained in a single page + # the ajax pages can't be used, they are empty + video_ids = self.extract_videos_from_page(channel_page) + else: + # Download all channel pages using the json-based channel_ajax query + for pagenum in itertools.count(1): + url = self._MORE_PAGES_URL % (pagenum, channel_id) + page = self._download_webpage(url, channel_id, + u'Downloading page #%s' % pagenum) + + page = json.loads(page) + + ids_in_page = self.extract_videos_from_page(page['content_html']) + video_ids.extend(ids_in_page) + + if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: + break self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index faed7ff7f..c6a9d06f2 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -53,7 +53,7 @@ class ZDFIE(InfoExtractor): video_id, u'Get stream URL') - MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"' + #MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"' RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)' mobj = re.search(self._MEDIA_STREAM, media_link) diff --git a/youtube_dl/update.py b/youtube_dl/update.py index 0689a4891..e5f441707 100644 --- a/youtube_dl/update.py +++ b/youtube_dl/update.py @@ -2,11 +2,15 @@ import io import json import traceback import hashlib +import os import subprocess import sys from zipimport import zipimporter -from .utils import * +from .utils import ( + compat_str, + compat_urllib_request, +) from .version import __version__ def rsa_verify(message, signature, key): @@ -105,7 +109,7 @@ def update_self(to_screen, verbose): urlh = compat_urllib_request.urlopen(version['exe'][0]) newcontent = urlh.read() urlh.close() - except (IOError, OSError) as err: + except (IOError, OSError): if verbose: to_screen(compat_str(traceback.format_exc())) to_screen(u'ERROR: unable to download latest version') return @@ -118,7 +122,7 @@ def update_self(to_screen, verbose): try: with open(exe + '.new', 'wb') as outf: outf.write(newcontent) - except (IOError, OSError) as err: + except (IOError, OSError): if verbose: to_screen(compat_str(traceback.format_exc())) to_screen(u'ERROR: unable to write the new version') return @@ -137,7 +141,7 @@ start /b "" cmd /c del "%%~f0"&exit /b" subprocess.Popen([bat]) # Continues to run in the background return # Do not show premature success messages - except (IOError, OSError) as err: + except (IOError, OSError): if verbose: to_screen(compat_str(traceback.format_exc())) to_screen(u'ERROR: unable to overwrite current version') return @@ -148,7 +152,7 @@ start /b "" cmd /c del "%%~f0"&exit /b" urlh = compat_urllib_request.urlopen(version['bin'][0]) newcontent = urlh.read() urlh.close() - except (IOError, OSError) as err: + except (IOError, OSError): if verbose: to_screen(compat_str(traceback.format_exc())) to_screen(u'ERROR: unable to download latest version') return @@ -161,7 +165,7 @@ start /b "" cmd /c del "%%~f0"&exit /b" try: with open(filename, 'wb') as outf: outf.write(newcontent) - except (IOError, OSError) as err: + except (IOError, OSError): if verbose: to_screen(compat_str(traceback.format_exc())) to_screen(u'ERROR: unable to overwrite current version') return diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 1d9785341..0720fe9eb 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -734,6 +734,8 @@ def unified_strdate(date_str): '%Y/%m/%d %H:%M:%S', '%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', ] for expression in format_expressions: @@ -949,7 +951,16 @@ class locked_file(object): def shell_quote(args): - return ' '.join(map(pipes.quote, args)) + quoted_args = [] + encoding = sys.getfilesystemencoding() + if encoding is None: + encoding = 'utf-8' + for a in args: + if isinstance(a, bytes): + # We may get a filename encoded with 'encodeFilename' + a = a.decode(encoding) + quoted_args.append(pipes.quote(a)) + return u' '.join(quoted_args) def takewhile_inclusive(pred, seq): diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 26b91105f..c1f581cd6 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,2 +1,2 @@ -__version__ = '2013.11.13' +__version__ = '2013.11.22' |