diff options
Diffstat (limited to 'youtube_dl')
222 files changed, 9763 insertions, 3808 deletions
diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 00af78e06..12977bf80 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -21,24 +21,25 @@ import subprocess import socket import sys import time +import tokenize import traceback if os.name == 'nt': import ctypes from .compat import ( - compat_basestring, compat_cookiejar, compat_expanduser, compat_get_terminal_size, compat_http_client, compat_kwargs, compat_str, + compat_tokenize_tokenize, compat_urllib_error, compat_urllib_request, + compat_urllib_request_DataHandler, ) from .utils import ( - escape_url, ContentTooShortError, date_from_str, DateRange, @@ -49,7 +50,6 @@ from .utils import ( ExtractorError, format_bytes, formatSeconds, - HEADRequest, locked_file, make_HTTPS_handler, MaxDownloadsReached, @@ -70,6 +70,7 @@ from .utils import ( version_tuple, write_json_file, write_string, + YoutubeDLCookieProcessor, YoutubeDLHandler, prepend_extension, replace_extension, @@ -285,7 +286,11 @@ class YoutubeDL(object): self._num_downloads = 0 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] self._err_file = sys.stderr - self.params = params + self.params = { + # Default parameters + 'nocheckcertificate': False, + } + self.params.update(params) self.cache = Cache(self) if params.get('bidi_workaround', False): @@ -853,8 +858,8 @@ class YoutubeDL(object): else: raise Exception('Invalid result type: %s' % result_type) - def _apply_format_filter(self, format_spec, available_formats): - " Returns a tuple of the remaining format_spec and filtered formats " + def _build_format_filter(self, filter_spec): + " Returns a function to filter the formats according to the filter_spec " OPERATORS = { '<': operator.lt, @@ -864,13 +869,13 @@ class YoutubeDL(object): '=': operator.eq, '!=': operator.ne, } - operator_rex = re.compile(r'''(?x)\s*\[ + operator_rex = re.compile(r'''(?x)\s* (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps) \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) - \]$ + $ ''' % '|'.join(map(re.escape, OPERATORS.keys()))) - m = operator_rex.search(format_spec) + m = operator_rex.search(filter_spec) if m: try: comparison_value = int(m.group('value')) @@ -881,7 +886,7 @@ class YoutubeDL(object): if comparison_value is None: raise ValueError( 'Invalid value %r in format specification %r' % ( - m.group('value'), format_spec)) + m.group('value'), filter_spec)) op = OPERATORS[m.group('op')] if not m: @@ -889,85 +894,283 @@ class YoutubeDL(object): '=': operator.eq, '!=': operator.ne, } - str_operator_rex = re.compile(r'''(?x)\s*\[ + str_operator_rex = re.compile(r'''(?x) \s*(?P<key>ext|acodec|vcodec|container|protocol) \s*(?P<op>%s)(?P<none_inclusive>\s*\?)? \s*(?P<value>[a-zA-Z0-9_-]+) - \s*\]$ + \s*$ ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) - m = str_operator_rex.search(format_spec) + m = str_operator_rex.search(filter_spec) if m: comparison_value = m.group('value') op = STR_OPERATORS[m.group('op')] if not m: - raise ValueError('Invalid format specification %r' % format_spec) + raise ValueError('Invalid filter specification %r' % filter_spec) def _filter(f): actual_value = f.get(m.group('key')) if actual_value is None: return m.group('none_inclusive') return op(actual_value, comparison_value) - new_formats = [f for f in available_formats if _filter(f)] + return _filter + + def build_format_selector(self, format_spec): + def syntax_error(note, start): + message = ( + 'Invalid format specification: ' + '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) + return SyntaxError(message) + + PICKFIRST = 'PICKFIRST' + MERGE = 'MERGE' + SINGLE = 'SINGLE' + GROUP = 'GROUP' + FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) + + def _parse_filter(tokens): + filter_parts = [] + for type, string, start, _, _ in tokens: + if type == tokenize.OP and string == ']': + return ''.join(filter_parts) + else: + filter_parts.append(string) + + def _remove_unused_ops(tokens): + # Remove operators that we don't use and join them with the sourrounding strings + # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' + ALLOWED_OPS = ('/', '+', ',', '(', ')') + last_string, last_start, last_end, last_line = None, None, None, None + for type, string, start, end, line in tokens: + if type == tokenize.OP and string == '[': + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + last_string = None + yield type, string, start, end, line + # everything inside brackets will be handled by _parse_filter + for type, string, start, end, line in tokens: + yield type, string, start, end, line + if type == tokenize.OP and string == ']': + break + elif type == tokenize.OP and string in ALLOWED_OPS: + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + last_string = None + yield type, string, start, end, line + elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: + if not last_string: + last_string = string + last_start = start + last_end = end + else: + last_string += string + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + + def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): + selectors = [] + current_selector = None + for type, string, start, _, _ in tokens: + # ENCODING is only defined in python 3.x + if type == getattr(tokenize, 'ENCODING', None): + continue + elif type in [tokenize.NAME, tokenize.NUMBER]: + current_selector = FormatSelector(SINGLE, string, []) + elif type == tokenize.OP: + if string == ')': + if not inside_group: + # ')' will be handled by the parentheses group + tokens.restore_last_token() + break + elif inside_merge and string in ['/', ',']: + tokens.restore_last_token() + break + elif inside_choice and string == ',': + tokens.restore_last_token() + break + elif string == ',': + if not current_selector: + raise syntax_error('"," must follow a format selector', start) + selectors.append(current_selector) + current_selector = None + elif string == '/': + if not current_selector: + raise syntax_error('"/" must follow a format selector', start) + first_choice = current_selector + second_choice = _parse_format_selection(tokens, inside_choice=True) + current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) + elif string == '[': + if not current_selector: + current_selector = FormatSelector(SINGLE, 'best', []) + format_filter = _parse_filter(tokens) + current_selector.filters.append(format_filter) + elif string == '(': + if current_selector: + raise syntax_error('Unexpected "("', start) + group = _parse_format_selection(tokens, inside_group=True) + current_selector = FormatSelector(GROUP, group, []) + elif string == '+': + video_selector = current_selector + audio_selector = _parse_format_selection(tokens, inside_merge=True) + if not video_selector or not audio_selector: + raise syntax_error('"+" must be between two format selectors', start) + current_selector = FormatSelector(MERGE, (video_selector, audio_selector), []) + else: + raise syntax_error('Operator not recognized: "{0}"'.format(string), start) + elif type == tokenize.ENDMARKER: + break + if current_selector: + selectors.append(current_selector) + return selectors + + def _build_selector_function(selector): + if isinstance(selector, list): + fs = [_build_selector_function(s) for s in selector] + + def selector_function(formats): + for f in fs: + for format in f(formats): + yield format + return selector_function + elif selector.type == GROUP: + selector_function = _build_selector_function(selector.selector) + elif selector.type == PICKFIRST: + fs = [_build_selector_function(s) for s in selector.selector] + + def selector_function(formats): + for f in fs: + picked_formats = list(f(formats)) + if picked_formats: + return picked_formats + return [] + elif selector.type == SINGLE: + format_spec = selector.selector + + def selector_function(formats): + formats = list(formats) + if not formats: + return + if format_spec == 'all': + for f in formats: + yield f + elif format_spec in ['best', 'worst', None]: + format_idx = 0 if format_spec == 'worst' else -1 + audiovideo_formats = [ + f for f in formats + if f.get('vcodec') != 'none' and f.get('acodec') != 'none'] + if audiovideo_formats: + yield audiovideo_formats[format_idx] + # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format + elif (all(f.get('acodec') != 'none' for f in formats) or + all(f.get('vcodec') != 'none' for f in formats)): + yield formats[format_idx] + elif format_spec == 'bestaudio': + audio_formats = [ + f for f in formats + if f.get('vcodec') == 'none'] + if audio_formats: + yield audio_formats[-1] + elif format_spec == 'worstaudio': + audio_formats = [ + f for f in formats + if f.get('vcodec') == 'none'] + if audio_formats: + yield audio_formats[0] + elif format_spec == 'bestvideo': + video_formats = [ + f for f in formats + if f.get('acodec') == 'none'] + if video_formats: + yield video_formats[-1] + elif format_spec == 'worstvideo': + video_formats = [ + f for f in formats + if f.get('acodec') == 'none'] + if video_formats: + yield video_formats[0] + else: + extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] + if format_spec in extensions: + filter_f = lambda f: f['ext'] == format_spec + else: + filter_f = lambda f: f['format_id'] == format_spec + matches = list(filter(filter_f, formats)) + if matches: + yield matches[-1] + elif selector.type == MERGE: + def _merge(formats_info): + format_1, format_2 = [f['format_id'] for f in formats_info] + # The first format must contain the video and the + # second the audio + if formats_info[0].get('vcodec') == 'none': + self.report_error('The first format must ' + 'contain the video, try using ' + '"-f %s+%s"' % (format_2, format_1)) + return + output_ext = ( + formats_info[0]['ext'] + if self.params.get('merge_output_format') is None + else self.params['merge_output_format']) + return { + 'requested_formats': formats_info, + 'format': '%s+%s' % (formats_info[0].get('format'), + formats_info[1].get('format')), + 'format_id': '%s+%s' % (formats_info[0].get('format_id'), + formats_info[1].get('format_id')), + 'width': formats_info[0].get('width'), + 'height': formats_info[0].get('height'), + 'resolution': formats_info[0].get('resolution'), + 'fps': formats_info[0].get('fps'), + 'vcodec': formats_info[0].get('vcodec'), + 'vbr': formats_info[0].get('vbr'), + 'stretched_ratio': formats_info[0].get('stretched_ratio'), + 'acodec': formats_info[1].get('acodec'), + 'abr': formats_info[1].get('abr'), + 'ext': output_ext, + } + video_selector, audio_selector = map(_build_selector_function, selector.selector) - new_format_spec = format_spec[:-len(m.group(0))] - if not new_format_spec: - new_format_spec = 'best' + def selector_function(formats): + formats = list(formats) + for pair in itertools.product(video_selector(formats), audio_selector(formats)): + yield _merge(pair) - return (new_format_spec, new_formats) + filters = [self._build_format_filter(f) for f in selector.filters] - def select_format(self, format_spec, available_formats): - while format_spec.endswith(']'): - format_spec, available_formats = self._apply_format_filter( - format_spec, available_formats) - if not available_formats: - return None + def final_selector(formats): + for _filter in filters: + formats = list(filter(_filter, formats)) + return selector_function(formats) + return final_selector - if format_spec in ['best', 'worst', None]: - format_idx = 0 if format_spec == 'worst' else -1 - audiovideo_formats = [ - f for f in available_formats - if f.get('vcodec') != 'none' and f.get('acodec') != 'none'] - if audiovideo_formats: - return audiovideo_formats[format_idx] - # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format - elif (all(f.get('acodec') != 'none' for f in available_formats) or - all(f.get('vcodec') != 'none' for f in available_formats)): - return available_formats[format_idx] - elif format_spec == 'bestaudio': - audio_formats = [ - f for f in available_formats - if f.get('vcodec') == 'none'] - if audio_formats: - return audio_formats[-1] - elif format_spec == 'worstaudio': - audio_formats = [ - f for f in available_formats - if f.get('vcodec') == 'none'] - if audio_formats: - return audio_formats[0] - elif format_spec == 'bestvideo': - video_formats = [ - f for f in available_formats - if f.get('acodec') == 'none'] - if video_formats: - return video_formats[-1] - elif format_spec == 'worstvideo': - video_formats = [ - f for f in available_formats - if f.get('acodec') == 'none'] - if video_formats: - return video_formats[0] - else: - extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] - if format_spec in extensions: - filter_f = lambda f: f['ext'] == format_spec - else: - filter_f = lambda f: f['format_id'] == format_spec - matches = list(filter(filter_f, available_formats)) - if matches: - return matches[-1] - return None + stream = io.BytesIO(format_spec.encode('utf-8')) + try: + tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) + except tokenize.TokenError: + raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) + + class TokenIterator(object): + def __init__(self, tokens): + self.tokens = tokens + self.counter = 0 + + def __iter__(self): + return self + + def __next__(self): + if self.counter >= len(self.tokens): + raise StopIteration() + value = self.tokens[self.counter] + self.counter += 1 + return value + + next = __next__ + + def restore_last_token(self): + self.counter -= 1 + + parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) + return _build_selector_function(parsed_selector) def _calc_headers(self, info_dict): res = std_headers.copy() @@ -1030,13 +1233,20 @@ class YoutubeDL(object): except (ValueError, OverflowError, OSError): pass + subtitles = info_dict.get('subtitles') + if subtitles: + for _, subtitle in subtitles.items(): + for subtitle_format in subtitle: + if 'ext' not in subtitle_format: + subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() + if self.params.get('listsubtitles', False): if 'automatic_captions' in info_dict: self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions') - self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles') + self.list_subtitles(info_dict['id'], subtitles, 'subtitles') return info_dict['requested_subtitles'] = self.process_subtitles( - info_dict['id'], info_dict.get('subtitles'), + info_dict['id'], subtitles, info_dict.get('automatic_captions')) # We now pick which formats have to be downloaded @@ -1104,62 +1314,15 @@ class YoutubeDL(object): if req_format is None: req_format_list = [] if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and - info_dict['extractor'] in ['youtube', 'ted']): + info_dict['extractor'] in ['youtube', 'ted'] and + not info_dict.get('is_live')): merger = FFmpegMergerPP(self) if merger.available and merger.can_merge(): req_format_list.append('bestvideo+bestaudio') req_format_list.append('best') req_format = '/'.join(req_format_list) - formats_to_download = [] - if req_format == 'all': - formats_to_download = formats - else: - for rfstr in req_format.split(','): - # We can accept formats requested in the format: 34/5/best, we pick - # the first that is available, starting from left - req_formats = rfstr.split('/') - for rf in req_formats: - if re.match(r'.+?\+.+?', rf) is not None: - # Two formats have been requested like '137+139' - format_1, format_2 = rf.split('+') - formats_info = (self.select_format(format_1, formats), - self.select_format(format_2, formats)) - if all(formats_info): - # The first format must contain the video and the - # second the audio - if formats_info[0].get('vcodec') == 'none': - self.report_error('The first format must ' - 'contain the video, try using ' - '"-f %s+%s"' % (format_2, format_1)) - return - output_ext = ( - formats_info[0]['ext'] - if self.params.get('merge_output_format') is None - else self.params['merge_output_format']) - selected_format = { - 'requested_formats': formats_info, - 'format': '%s+%s' % (formats_info[0].get('format'), - formats_info[1].get('format')), - 'format_id': '%s+%s' % (formats_info[0].get('format_id'), - formats_info[1].get('format_id')), - 'width': formats_info[0].get('width'), - 'height': formats_info[0].get('height'), - 'resolution': formats_info[0].get('resolution'), - 'fps': formats_info[0].get('fps'), - 'vcodec': formats_info[0].get('vcodec'), - 'vbr': formats_info[0].get('vbr'), - 'stretched_ratio': formats_info[0].get('stretched_ratio'), - 'acodec': formats_info[1].get('acodec'), - 'abr': formats_info[1].get('abr'), - 'ext': output_ext, - } - else: - selected_format = None - else: - selected_format = self.select_format(rf, formats) - if selected_format is not None: - formats_to_download.append(selected_format) - break + format_selector = self.build_format_selector(req_format) + formats_to_download = list(format_selector(formats)) if not formats_to_download: raise ExtractorError('requested format not available', expected=True) @@ -1707,27 +1870,6 @@ class YoutubeDL(object): def urlopen(self, req): """ Start an HTTP download """ - - # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not - # always respected by websites, some tend to give out URLs with non percent-encoded - # non-ASCII characters (see telemb.py, ard.py [#3412]) - # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) - # To work around aforementioned issue we will replace request's original URL with - # percent-encoded one - req_is_string = isinstance(req, compat_basestring) - url = req if req_is_string else req.get_full_url() - url_escaped = escape_url(url) - - # Substitute URL if any change after escaping - if url != url_escaped: - if req_is_string: - req = url_escaped - else: - req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request - req = req_type( - url_escaped, data=req.data, headers=req.headers, - origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) - return self._opener.open(req, timeout=self._socket_timeout) def print_debug_header(self): @@ -1810,8 +1952,7 @@ class YoutubeDL(object): if os.access(opts_cookiefile, os.R_OK): self.cookiejar.load() - cookie_processor = compat_urllib_request.HTTPCookieProcessor( - self.cookiejar) + cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) if opts_proxy is not None: if opts_proxy == '': proxies = {} @@ -1827,8 +1968,9 @@ class YoutubeDL(object): debuglevel = 1 if self.params.get('debug_printtraffic') else 0 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) + data_handler = compat_urllib_request_DataHandler() opener = compat_urllib_request.build_opener( - proxy_handler, https_handler, cookie_processor, ydlh) + proxy_handler, https_handler, cookie_processor, ydlh, data_handler) # Delete the default user-agent header, which would otherwise apply in # cases where our custom HTTP handler doesn't come into play @@ -1880,7 +2022,7 @@ class YoutubeDL(object): (info_dict['extractor'], info_dict['id'], thumb_display_id)) try: uf = self.urlopen(t['url']) - with open(thumb_filename, 'wb') as thumbf: + with open(encodeFilename(thumb_filename), 'wb') as thumbf: shutil.copyfileobj(uf, thumbf) self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 55b22c889..5e2ed4d4b 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -9,7 +9,6 @@ import codecs import io import os import random -import shlex import sys @@ -20,6 +19,7 @@ from .compat import ( compat_expanduser, compat_getpass, compat_print, + compat_shlex_split, workaround_optparse_bug9161, ) from .utils import ( @@ -262,10 +262,10 @@ def _real_main(argv=None): parser.error('setting filesize xattr requested but python-xattr is not available') external_downloader_args = None if opts.external_downloader_args: - external_downloader_args = shlex.split(opts.external_downloader_args) + external_downloader_args = compat_shlex_split(opts.external_downloader_args) postprocessor_args = None if opts.postprocessor_args: - postprocessor_args = shlex.split(opts.postprocessor_args) + postprocessor_args = compat_shlex_split(opts.postprocessor_args) match_filter = ( None if opts.match_filter is None else match_filter_func(opts.match_filter)) diff --git a/youtube_dl/__main__.py b/youtube_dl/__main__.py index 65a0f891c..42a0f8c6f 100755 --- a/youtube_dl/__main__.py +++ b/youtube_dl/__main__.py @@ -11,7 +11,7 @@ if __package__ is None and not hasattr(sys, "frozen"): # direct call of __main__.py import os.path path = os.path.realpath(os.path.abspath(__file__)) - sys.path.append(os.path.dirname(os.path.dirname(path))) + sys.path.insert(0, os.path.dirname(os.path.dirname(path))) import youtube_dl diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index c3783337a..d103ab9ad 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -1,10 +1,14 @@ from __future__ import unicode_literals +import binascii import collections +import email import getpass +import io import optparse import os import re +import shlex import shutil import socket import subprocess @@ -38,11 +42,21 @@ except ImportError: # Python 2 import urlparse as compat_urlparse try: + import urllib.response as compat_urllib_response +except ImportError: # Python 2 + import urllib as compat_urllib_response + +try: import http.cookiejar as compat_cookiejar except ImportError: # Python 2 import cookielib as compat_cookiejar try: + import http.cookies as compat_cookies +except ImportError: # Python 2 + import Cookie as compat_cookies + +try: import html.entities as compat_html_entities except ImportError: # Python 2 import htmlentitydefs as compat_html_entities @@ -75,47 +89,113 @@ except ImportError: import BaseHTTPServer as compat_http_server try: + compat_str = unicode # Python 2 +except NameError: + compat_str = str + +try: + from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes from urllib.parse import unquote as compat_urllib_parse_unquote -except ImportError: - def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'): - if string == '': + from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus +except ImportError: # Python 2 + _asciire = (compat_urllib_parse._asciire if hasattr(compat_urllib_parse, '_asciire') + else re.compile('([\x00-\x7f]+)')) + + # HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus + # implementations from cpython 3.4.3's stdlib. Python 2's version + # is apparently broken (see https://github.com/rg3/youtube-dl/pull/6244) + + def compat_urllib_parse_unquote_to_bytes(string): + """unquote_to_bytes('abc%20def') -> b'abc def'.""" + # Note: strings are encoded as UTF-8. This is only an issue if it contains + # unescaped non-ASCII characters, which URIs should not. + if not string: + # Is it a string-like object? + string.split + return b'' + if isinstance(string, compat_str): + string = string.encode('utf-8') + bits = string.split(b'%') + if len(bits) == 1: return string - res = string.split('%') - if len(res) == 1: + res = [bits[0]] + append = res.append + for item in bits[1:]: + try: + append(compat_urllib_parse._hextochr[item[:2]]) + append(item[2:]) + except KeyError: + append(b'%') + append(item) + return b''.join(res) + + def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'): + """Replace %xx escapes by their single-character equivalent. The optional + encoding and errors parameters specify how to decode percent-encoded + sequences into Unicode characters, as accepted by the bytes.decode() + method. + By default, percent-encoded sequences are decoded with UTF-8, and invalid + sequences are replaced by a placeholder character. + + unquote('abc%20def') -> 'abc def'. + """ + if '%' not in string: + string.split return string if encoding is None: encoding = 'utf-8' if errors is None: errors = 'replace' - # pct_sequence: contiguous sequence of percent-encoded bytes, decoded - pct_sequence = b'' - string = res[0] - for item in res[1:]: - try: - if not item: - raise ValueError - pct_sequence += item[:2].decode('hex') - rest = item[2:] - if not rest: - # This segment was just a single percent-encoded character. - # May be part of a sequence of code units, so delay decoding. - # (Stored in pct_sequence). - continue - except ValueError: - rest = '%' + item - # Encountered non-percent-encoded characters. Flush the current - # pct_sequence. - string += pct_sequence.decode(encoding, errors) + rest - pct_sequence = b'' - if pct_sequence: - # Flush the final pct_sequence - string += pct_sequence.decode(encoding, errors) - return string + bits = _asciire.split(string) + res = [bits[0]] + append = res.append + for i in range(1, len(bits), 2): + append(compat_urllib_parse_unquote_to_bytes(bits[i]).decode(encoding, errors)) + append(bits[i + 1]) + return ''.join(res) + + def compat_urllib_parse_unquote_plus(string, encoding='utf-8', errors='replace'): + """Like unquote(), but also replace plus signs by spaces, as required for + unquoting HTML form values. + + unquote_plus('%7e/abc+def') -> '~/abc def' + """ + string = string.replace('+', ' ') + return compat_urllib_parse_unquote(string, encoding, errors) try: - compat_str = unicode # Python 2 -except NameError: - compat_str = str + from urllib.request import DataHandler as compat_urllib_request_DataHandler +except ImportError: # Python < 3.4 + # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py + class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler): + def data_open(self, req): + # data URLs as specified in RFC 2397. + # + # ignores POSTed data + # + # syntax: + # dataurl := "data:" [ mediatype ] [ ";base64" ] "," data + # mediatype := [ type "/" subtype ] *( ";" parameter ) + # data := *urlchar + # parameter := attribute "=" value + url = req.get_full_url() + + scheme, data = url.split(":", 1) + mediatype, data = data.split(",", 1) + + # even base64 encoded data URLs might be quoted so unquote in any case: + data = compat_urllib_parse_unquote_to_bytes(data) + if mediatype.endswith(";base64"): + data = binascii.a2b_base64(data) + mediatype = mediatype[:-7] + + if not mediatype: + mediatype = "text/plain;charset=US-ASCII" + + headers = email.message_from_string( + "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data))) + + return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url) try: compat_basestring = basestring # Python 2 @@ -190,6 +270,17 @@ except ImportError: # Python < 3.3 return "'" + s.replace("'", "'\"'\"'") + "'" +if sys.version_info >= (2, 7, 3): + compat_shlex_split = shlex.split +else: + # 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) + + def compat_ord(c): if type(c) is int: return c @@ -367,26 +458,32 @@ if hasattr(shutil, 'get_terminal_size'): # Python >= 3.3 else: _terminal_size = collections.namedtuple('terminal_size', ['columns', 'lines']) - def compat_get_terminal_size(): - columns = compat_getenv('COLUMNS', None) + def compat_get_terminal_size(fallback=(80, 24)): + columns = compat_getenv('COLUMNS') if columns: columns = int(columns) else: columns = None - lines = compat_getenv('LINES', None) + lines = compat_getenv('LINES') if lines: lines = int(lines) else: lines = None - try: - sp = subprocess.Popen( - ['stty', 'size'], - stdout=subprocess.PIPE, stderr=subprocess.PIPE) - out, err = sp.communicate() - lines, columns = map(int, out.split()) - except Exception: - pass + if columns is None or lines is None or columns <= 0 or lines <= 0: + try: + sp = subprocess.Popen( + ['stty', 'size'], + stdout=subprocess.PIPE, stderr=subprocess.PIPE) + out, err = sp.communicate() + _lines, _columns = map(int, out.split()) + except Exception: + _columns, _lines = _terminal_size(*fallback) + + if columns is None or columns <= 0: + columns = _columns + if lines is None or lines <= 0: + lines = _lines return _terminal_size(columns, lines) try: @@ -399,11 +496,17 @@ except TypeError: # Python 2.6 yield n n += step +if sys.version_info >= (3, 0): + from tokenize import tokenize as compat_tokenize_tokenize +else: + from tokenize import generate_tokens as compat_tokenize_tokenize + __all__ = [ 'compat_HTTPError', 'compat_basestring', 'compat_chr', 'compat_cookiejar', + 'compat_cookies', 'compat_expanduser', 'compat_get_terminal_size', 'compat_getenv', @@ -416,14 +519,20 @@ __all__ = [ 'compat_ord', 'compat_parse_qs', 'compat_print', + 'compat_shlex_split', 'compat_socket_create_connection', 'compat_str', 'compat_subprocess_get_DEVNULL', + 'compat_tokenize_tokenize', 'compat_urllib_error', 'compat_urllib_parse', 'compat_urllib_parse_unquote', + 'compat_urllib_parse_unquote_plus', + 'compat_urllib_parse_unquote_to_bytes', 'compat_urllib_parse_urlparse', 'compat_urllib_request', + 'compat_urllib_request_DataHandler', + 'compat_urllib_response', 'compat_urlparse', 'compat_urlretrieve', 'compat_xml_parse_error', diff --git a/youtube_dl/downloader/__init__.py b/youtube_dl/downloader/__init__.py index f110830c4..dccc59212 100644 --- a/youtube_dl/downloader/__init__.py +++ b/youtube_dl/downloader/__init__.py @@ -8,6 +8,7 @@ from .hls import NativeHlsFD from .http import HttpFD from .rtsp import RtspFD from .rtmp import RtmpFD +from .dash import DashSegmentsFD from ..utils import ( determine_protocol, @@ -20,6 +21,7 @@ PROTOCOL_MAP = { 'mms': RtspFD, 'rtsp': RtspFD, 'f4m': F4mFD, + 'http_dash_segments': DashSegmentsFD, } diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py index 97e755d4b..29a4500d3 100644 --- a/youtube_dl/downloader/common.py +++ b/youtube_dl/downloader/common.py @@ -325,7 +325,7 @@ class FileDownloader(object): ) # Check file already present - if filename != '-' and nooverwrites_and_exists or continuedl_and_exists: + if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists): self.report_file_already_downloaded(filename) self._hook_progress({ 'filename': filename, diff --git a/youtube_dl/downloader/dash.py b/youtube_dl/downloader/dash.py new file mode 100644 index 000000000..8b6fa2753 --- /dev/null +++ b/youtube_dl/downloader/dash.py @@ -0,0 +1,66 @@ +from __future__ import unicode_literals + +import re + +from .common import FileDownloader +from ..compat import compat_urllib_request + + +class DashSegmentsFD(FileDownloader): + """ + Download segments in a DASH manifest + """ + def real_download(self, filename, info_dict): + self.report_destination(filename) + tmpfilename = self.temp_name(filename) + base_url = info_dict['url'] + segment_urls = info_dict['segment_urls'] + + is_test = self.params.get('test', False) + remaining_bytes = self._TEST_FILE_SIZE if is_test else None + byte_counter = 0 + + def append_url_to_file(outf, target_url, target_name, remaining_bytes=None): + self.to_screen('[DashSegments] %s: Downloading %s' % (info_dict['id'], target_name)) + req = compat_urllib_request.Request(target_url) + if remaining_bytes is not None: + req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1)) + + data = self.ydl.urlopen(req).read() + + if remaining_bytes is not None: + data = data[:remaining_bytes] + + outf.write(data) + return len(data) + + def combine_url(base_url, target_url): + if re.match(r'^https?://', target_url): + return target_url + return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url) + + with open(tmpfilename, 'wb') as outf: + append_url_to_file( + outf, combine_url(base_url, info_dict['initialization_url']), + 'initialization segment') + for i, segment_url in enumerate(segment_urls): + segment_len = append_url_to_file( + outf, combine_url(base_url, segment_url), + 'segment %d / %d' % (i + 1, len(segment_urls)), + remaining_bytes) + byte_counter += segment_len + if remaining_bytes is not None: + remaining_bytes -= segment_len + if remaining_bytes <= 0: + break + + self.try_rename(tmpfilename, filename) + + self._hook_progress({ + 'downloaded_bytes': byte_counter, + 'total_bytes': byte_counter, + 'filename': filename, + 'status': 'finished', + }) + + return True diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index 1d5cc9904..2bc011266 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -5,6 +5,10 @@ import subprocess from .common import FileDownloader from ..utils import ( + cli_option, + cli_valueless_option, + cli_bool_option, + cli_configuration_args, encodeFilename, encodeArgument, ) @@ -45,18 +49,17 @@ class ExternalFD(FileDownloader): def supports(cls, info_dict): return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps') - def _source_address(self, command_option): - source_address = self.params.get('source_address') - if source_address is None: - return [] - return [command_option, source_address] + def _option(self, command_option, param): + return cli_option(self.params, command_option, param) + + def _bool_option(self, command_option, param, true_value='true', false_value='false', separator=None): + return cli_bool_option(self.params, command_option, param, true_value, false_value, separator) + + def _valueless_option(self, command_option, param, expected_value=True): + return cli_valueless_option(self.params, command_option, param, expected_value) def _configuration_args(self, default=[]): - ex_args = self.params.get('external_downloader_args') - if ex_args is None: - return default - assert isinstance(ex_args, list) - return ex_args + return cli_configuration_args(self.params, 'external_downloader_args', default) def _call_downloader(self, tmpfilename, info_dict): """ Either overwrite this or implement _make_cmd """ @@ -77,7 +80,19 @@ class CurlFD(ExternalFD): cmd = [self.exe, '--location', '-o', tmpfilename] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--interface') + cmd += self._option('--interface', 'source_address') + cmd += self._option('--proxy', 'proxy') + cmd += self._valueless_option('--insecure', 'nocheckcertificate') + cmd += self._configuration_args() + cmd += ['--', info_dict['url']] + return cmd + + +class AxelFD(ExternalFD): + def _make_cmd(self, tmpfilename, info_dict): + cmd = [self.exe, '-o', tmpfilename] + for key, val in info_dict['http_headers'].items(): + cmd += ['-H', '%s: %s' % (key, val)] cmd += self._configuration_args() cmd += ['--', info_dict['url']] return cmd @@ -88,7 +103,9 @@ class WgetFD(ExternalFD): cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies'] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--bind-address') + cmd += self._option('--bind-address', 'source_address') + cmd += self._option('--proxy', 'proxy') + cmd += self._valueless_option('--no-check-certificate', 'nocheckcertificate') cmd += self._configuration_args() cmd += ['--', info_dict['url']] return cmd @@ -105,7 +122,9 @@ class Aria2cFD(ExternalFD): cmd += ['--out', os.path.basename(tmpfilename)] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--interface') + cmd += self._option('--interface', 'source_address') + cmd += self._option('--all-proxy', 'proxy') + cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=') cmd += ['--', info_dict['url']] return cmd diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index b1a858c45..7f6143954 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -7,17 +7,17 @@ import os import time import xml.etree.ElementTree as etree -from .common import FileDownloader -from .http import HttpFD +from .fragment import FragmentFD from ..compat import ( compat_urlparse, compat_urllib_error, + compat_urllib_parse_urlparse, ) from ..utils import ( - struct_pack, - struct_unpack, encodeFilename, sanitize_open, + struct_pack, + struct_unpack, xpath_text, ) @@ -226,16 +226,13 @@ def _add_ns(prop): return '{http://ns.adobe.com/f4m/1.0}%s' % prop -class HttpQuietDownloader(HttpFD): - def to_screen(self, *args, **kargs): - pass - - -class F4mFD(FileDownloader): +class F4mFD(FragmentFD): """ A downloader for f4m manifests or AdobeHDS. """ + FD_NAME = 'f4m' + def _get_unencrypted_media(self, doc): media = doc.findall(_add_ns('media')) if not media: @@ -288,8 +285,10 @@ class F4mFD(FileDownloader): def real_download(self, filename, info_dict): man_url = info_dict['url'] requested_bitrate = info_dict.get('tbr') - self.to_screen('[download] Downloading f4m manifest') - manifest = self.ydl.urlopen(man_url).read() + self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME) + urlh = self.ydl.urlopen(man_url) + man_url = urlh.geturl() + manifest = urlh.read() doc = etree.fromstring(manifest) formats = [(int(f.attrib.get('bitrate', -1)), f) @@ -320,94 +319,53 @@ class F4mFD(FileDownloader): # For some akamai manifests we'll need to add a query to the fragment url akamai_pv = xpath_text(doc, _add_ns('pv-2.0')) - self.report_destination(filename) - http_dl = HttpQuietDownloader( - self.ydl, - { - 'continuedl': True, - 'quiet': True, - 'noprogress': True, - 'ratelimit': self.params.get('ratelimit', None), - 'test': self.params.get('test', False), - } - ) - tmpfilename = self.temp_name(filename) - (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb') + ctx = { + 'filename': filename, + 'total_frags': total_frags, + } + + self._prepare_frag_download(ctx) + + dest_stream = ctx['dest_stream'] write_flv_header(dest_stream) if not live: write_metadata_tag(dest_stream, metadata) - # This dict stores the download progress, it's updated by the progress - # hook - state = { - 'status': 'downloading', - 'downloaded_bytes': 0, - 'frag_index': 0, - 'frag_count': total_frags, - 'filename': filename, - 'tmpfilename': tmpfilename, - } - start = time.time() - - def frag_progress_hook(s): - if s['status'] not in ('downloading', 'finished'): - return - - frag_total_bytes = s.get('total_bytes', 0) - if s['status'] == 'finished': - state['downloaded_bytes'] += frag_total_bytes - state['frag_index'] += 1 - - estimated_size = ( - (state['downloaded_bytes'] + frag_total_bytes) / - (state['frag_index'] + 1) * total_frags) - time_now = time.time() - state['total_bytes_estimate'] = estimated_size - state['elapsed'] = time_now - start - - if s['status'] == 'finished': - progress = self.calc_percent(state['frag_index'], total_frags) - else: - frag_downloaded_bytes = s['downloaded_bytes'] - frag_progress = self.calc_percent(frag_downloaded_bytes, - frag_total_bytes) - progress = self.calc_percent(state['frag_index'], total_frags) - progress += frag_progress / float(total_frags) - - state['eta'] = self.calc_eta( - start, time_now, estimated_size, state['downloaded_bytes'] + frag_downloaded_bytes) - state['speed'] = s.get('speed') - self._hook_progress(state) + base_url_parsed = compat_urllib_parse_urlparse(base_url) - http_dl.add_progress_hook(frag_progress_hook) + self._start_frag_download(ctx) frags_filenames = [] while fragments_list: seg_i, frag_i = fragments_list.pop(0) name = 'Seg%d-Frag%d' % (seg_i, frag_i) - url = base_url + name + query = [] + if base_url_parsed.query: + query.append(base_url_parsed.query) if akamai_pv: - url += '?' + akamai_pv.strip(';') + query.append(akamai_pv.strip(';')) if info_dict.get('extra_param_to_segment_url'): - url += info_dict.get('extra_param_to_segment_url') - frag_filename = '%s-%s' % (tmpfilename, name) + query.append(info_dict['extra_param_to_segment_url']) + url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query)) + frag_filename = '%s-%s' % (ctx['tmpfilename'], name) try: - success = http_dl.download(frag_filename, {'url': url}) + success = ctx['dl'].download(frag_filename, {'url': url_parsed.geturl()}) if not success: return False - with open(frag_filename, 'rb') as down: - down_data = down.read() - reader = FlvReader(down_data) - while True: - _, box_type, box_data = reader.read_box_info() - if box_type == b'mdat': - dest_stream.write(box_data) - break + (down, frag_sanitized) = sanitize_open(frag_filename, 'rb') + down_data = down.read() + down.close() + reader = FlvReader(down_data) + while True: + _, box_type, box_data = reader.read_box_info() + if box_type == b'mdat': + dest_stream.write(box_data) + break if live: - os.remove(frag_filename) + os.remove(encodeFilename(frag_sanitized)) else: - frags_filenames.append(frag_filename) + frags_filenames.append(frag_sanitized) except (compat_urllib_error.HTTPError, ) as err: if live and (err.code == 404 or err.code == 410): # We didn't keep up with the live window. Continue @@ -425,20 +383,9 @@ class F4mFD(FileDownloader): msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1)) self.report_warning(msg) - dest_stream.close() + self._finish_frag_download(ctx) - elapsed = time.time() - start - self.try_rename(tmpfilename, filename) for frag_file in frags_filenames: - os.remove(frag_file) - - fsize = os.path.getsize(encodeFilename(filename)) - self._hook_progress({ - 'downloaded_bytes': fsize, - 'total_bytes': fsize, - 'filename': filename, - 'status': 'finished', - 'elapsed': elapsed, - }) + os.remove(encodeFilename(frag_file)) return True diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py new file mode 100644 index 000000000..5a64b29ee --- /dev/null +++ b/youtube_dl/downloader/fragment.py @@ -0,0 +1,111 @@ +from __future__ import division, unicode_literals + +import os +import time + +from .common import FileDownloader +from .http import HttpFD +from ..utils import ( + encodeFilename, + sanitize_open, +) + + +class HttpQuietDownloader(HttpFD): + def to_screen(self, *args, **kargs): + pass + + +class FragmentFD(FileDownloader): + """ + A base file downloader class for fragmented media (e.g. f4m/m3u8 manifests). + """ + + def _prepare_and_start_frag_download(self, ctx): + self._prepare_frag_download(ctx) + self._start_frag_download(ctx) + + def _prepare_frag_download(self, ctx): + self.to_screen('[%s] Total fragments: %d' % (self.FD_NAME, ctx['total_frags'])) + self.report_destination(ctx['filename']) + dl = HttpQuietDownloader( + self.ydl, + { + 'continuedl': True, + 'quiet': True, + 'noprogress': True, + 'ratelimit': self.params.get('ratelimit', None), + 'retries': self.params.get('retries', 0), + 'test': self.params.get('test', False), + } + ) + tmpfilename = self.temp_name(ctx['filename']) + dest_stream, tmpfilename = sanitize_open(tmpfilename, 'wb') + ctx.update({ + 'dl': dl, + 'dest_stream': dest_stream, + 'tmpfilename': tmpfilename, + }) + + def _start_frag_download(self, ctx): + total_frags = ctx['total_frags'] + # This dict stores the download progress, it's updated by the progress + # hook + state = { + 'status': 'downloading', + 'downloaded_bytes': 0, + 'frag_index': 0, + 'frag_count': total_frags, + 'filename': ctx['filename'], + 'tmpfilename': ctx['tmpfilename'], + } + start = time.time() + ctx['started'] = start + + def frag_progress_hook(s): + if s['status'] not in ('downloading', 'finished'): + return + + frag_total_bytes = s.get('total_bytes', 0) + if s['status'] == 'finished': + state['downloaded_bytes'] += frag_total_bytes + state['frag_index'] += 1 + + estimated_size = ( + (state['downloaded_bytes'] + frag_total_bytes) / + (state['frag_index'] + 1) * total_frags) + time_now = time.time() + state['total_bytes_estimate'] = estimated_size + state['elapsed'] = time_now - start + + if s['status'] == 'finished': + progress = self.calc_percent(state['frag_index'], total_frags) + else: + frag_downloaded_bytes = s['downloaded_bytes'] + frag_progress = self.calc_percent(frag_downloaded_bytes, + frag_total_bytes) + progress = self.calc_percent(state['frag_index'], total_frags) + progress += frag_progress / float(total_frags) + + state['eta'] = self.calc_eta( + start, time_now, estimated_size, state['downloaded_bytes'] + frag_downloaded_bytes) + state['speed'] = s.get('speed') + self._hook_progress(state) + + ctx['dl'].add_progress_hook(frag_progress_hook) + + return start + + def _finish_frag_download(self, ctx): + ctx['dest_stream'].close() + elapsed = time.time() - ctx['started'] + self.try_rename(ctx['tmpfilename'], ctx['filename']) + fsize = os.path.getsize(encodeFilename(ctx['filename'])) + + self._hook_progress({ + 'downloaded_bytes': fsize, + 'total_bytes': fsize, + 'filename': ctx['filename'], + 'status': 'finished', + 'elapsed': elapsed, + }) diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index 8be4f4249..9a83a73dd 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -4,15 +4,15 @@ import os import re import subprocess -from ..postprocessor.ffmpeg import FFmpegPostProcessor from .common import FileDownloader -from ..compat import ( - compat_urlparse, - compat_urllib_request, -) +from .fragment import FragmentFD + +from ..compat import compat_urlparse +from ..postprocessor.ffmpeg import FFmpegPostProcessor from ..utils import ( encodeArgument, encodeFilename, + sanitize_open, ) @@ -28,10 +28,21 @@ class HlsFD(FileDownloader): return False ffpp.check_version() - args = [ - encodeArgument(opt) - for opt in (ffpp.executable, '-y', '-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc')] - args.append(encodeFilename(tmpfilename, True)) + args = [ffpp.executable, '-y'] + + if info_dict['http_headers'] and re.match(r'^https?://', url): + # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv: + # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header. + args += [ + '-headers', + ''.join('%s: %s\r\n' % (key, val) for key, val in info_dict['http_headers'].items())] + + args += ['-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc'] + + args = [encodeArgument(opt) for opt in args] + args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True)) + + self._debug_cmd(args) retval = subprocess.call(args) if retval == 0: @@ -51,54 +62,51 @@ class HlsFD(FileDownloader): return False -class NativeHlsFD(FileDownloader): +class NativeHlsFD(FragmentFD): """ A more limited implementation that does not require ffmpeg """ + FD_NAME = 'hlsnative' + def real_download(self, filename, info_dict): - url = info_dict['url'] - self.report_destination(filename) - tmpfilename = self.temp_name(filename) + man_url = info_dict['url'] + self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME) + manifest = self.ydl.urlopen(man_url).read() - self.to_screen( - '[hlsnative] %s: Downloading m3u8 manifest' % info_dict['id']) - data = self.ydl.urlopen(url).read() - s = data.decode('utf-8', 'ignore') - segment_urls = [] + s = manifest.decode('utf-8', 'ignore') + fragment_urls = [] 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(url, line)) - segment_urls.append(segment_url) - - is_test = self.params.get('test', False) - remaining_bytes = self._TEST_FILE_SIZE if is_test else None - byte_counter = 0 - with open(tmpfilename, 'wb') as outf: - for i, segurl in enumerate(segment_urls): - self.to_screen( - '[hlsnative] %s: Downloading segment %d / %d' % - (info_dict['id'], i + 1, len(segment_urls))) - seg_req = compat_urllib_request.Request(segurl) - if remaining_bytes is not None: - seg_req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1)) - - segment = self.ydl.urlopen(seg_req).read() - if remaining_bytes is not None: - segment = segment[:remaining_bytes] - remaining_bytes -= len(segment) - outf.write(segment) - byte_counter += len(segment) - if remaining_bytes is not None and remaining_bytes <= 0: + 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 - self._hook_progress({ - 'downloaded_bytes': byte_counter, - 'total_bytes': byte_counter, + ctx = { 'filename': filename, - 'status': 'finished', - }) - self.try_rename(tmpfilename, filename) + 'total_frags': len(fragment_urls), + } + + self._prepare_and_start_frag_download(ctx) + + 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) + + self._finish_frag_download(ctx) + + for frag_file in frags_filenames: + os.remove(encodeFilename(frag_file)) + return True diff --git a/youtube_dl/downloader/http.py b/youtube_dl/downloader/http.py index b7f144af9..a29f5cf31 100644 --- a/youtube_dl/downloader/http.py +++ b/youtube_dl/downloader/http.py @@ -4,6 +4,7 @@ import errno import os import socket import time +import re from .common import FileDownloader from ..compat import ( @@ -57,6 +58,24 @@ class HttpFD(FileDownloader): # Establish connection try: data = self.ydl.urlopen(request) + # When trying to resume, Content-Range HTTP header of response has to be checked + # to match the value of requested Range HTTP header. This is due to a webservers + # that don't support resuming and serve a whole file with no Content-Range + # set in response despite of requested Range (see + # https://github.com/rg3/youtube-dl/issues/6057#issuecomment-126129799) + if resume_len > 0: + content_range = data.headers.get('Content-Range') + if content_range: + content_range_m = re.search(r'bytes (\d+)-', content_range) + # Content-Range is present and matches requested Range, resume is possible + if content_range_m and resume_len == int(content_range_m.group(1)): + break + # Content-Range is either not present or invalid. Assuming remote webserver is + # trying to send the whole file, resume is not possible, so wiping the local file + # and performing entire redownload + self.report_unable_to_resume() + resume_len = 0 + open_mode = 'wb' break except (compat_urllib_error.HTTPError, ) as err: if (err.code < 500 or err.code >= 600) and err.code != 416: diff --git a/youtube_dl/downloader/rtmp.py b/youtube_dl/downloader/rtmp.py index 7d19bb808..f1d219ba9 100644 --- a/youtube_dl/downloader/rtmp.py +++ b/youtube_dl/downloader/rtmp.py @@ -105,7 +105,7 @@ class RtmpFD(FileDownloader): protocol = info_dict.get('rtmp_protocol', None) real_time = info_dict.get('rtmp_real_time', False) no_resume = info_dict.get('no_resume', False) - continue_dl = info_dict.get('continuedl', True) + continue_dl = self.params.get('continuedl', True) self.report_destination(filename) tmpfilename = self.temp_name(filename) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 5cc03b875..10286aa88 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -19,9 +19,14 @@ from .anysex import AnySexIE from .aol import AolIE from .allocine import AllocineIE from .aparat import AparatIE +from .appleconnect import AppleConnectIE from .appletrailers import AppleTrailersIE from .archiveorg import ArchiveOrgIE -from .ard import ARDIE, ARDMediathekIE +from .ard import ( + ARDIE, + ARDMediathekIE, + SportschauIE, +) from .arte import ( ArteTvIE, ArteTVPlus7IE, @@ -38,7 +43,11 @@ from .azubu import AzubuIE from .baidu import BaiduVideoIE from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE -from .bbccouk import BBCCoUkIE +from .bbc import ( + BBCCoUkIE, + BBCCoUkArticleIE, + BBCIE, +) from .beeg import BeegIE from .behindkink import BehindKinkIE from .beatportpro import BeatportProIE @@ -68,6 +77,7 @@ from .cbssports import CBSSportsIE from .ccc import CCCIE from .ceskatelevize import CeskaTelevizeIE from .channel9 import Channel9IE +from .chaturbate import ChaturbateIE from .chilloutzone import ChilloutzoneIE from .chirbit import ( ChirbitIE, @@ -80,6 +90,7 @@ from .cliphunter import CliphunterIE from .clipsyndicate import ClipsyndicateIE from .cloudy import CloudyIE from .clubic import ClubicIE +from .clyp import ClypIE from .cmt import CMTIE from .cnet import CNETIE from .cnn import ( @@ -110,6 +121,7 @@ from .dailymotion import ( ) from .daum import DaumIE from .dbtv import DBTVIE +from .dcn import DCNIE from .dctp import DctpTvIE from .deezer import DeezerPlaylistIE from .democracynow import DemocracynowIE @@ -130,7 +142,6 @@ from .dump import DumpIE from .dumpert import DumpertIE from .defense import DefenseGouvFrIE from .discovery import DiscoveryIE -from .divxstage import DivxStageIE from .dropbox import DropboxIE from .eagleplatform import EaglePlatformIE from .ebaumsworld import EbaumsWorldIE @@ -150,6 +161,8 @@ from .eporner import EpornerIE from .eroprofile import EroProfileIE from .escapist import EscapistIE from .espn import ESPNIE +from .esri import EsriVideoIE +from .europa import EuropaIE from .everyonesmixtape import EveryonesMixtapeIE from .exfm import ExfmIE from .expotv import ExpoTVIE @@ -157,14 +170,12 @@ from .extremetube import ExtremeTubeIE from .facebook import FacebookIE from .faz import FazIE from .fc2 import FC2IE +from .fczenit import FczenitIE from .firstpost import FirstpostIE from .firsttv import FirstTVIE from .fivemin import FiveMinIE from .fivetv import FiveTVIE -from .fktv import ( - FKTVIE, - FKTVPosteckeIE, -) +from .fktv import FKTVIE from .flickr import FlickrIE from .folketinget import FolketingetIE from .footyroom import FootyRoomIE @@ -220,7 +231,6 @@ from .historicfilms import HistoricFilmsIE from .history import HistoryIE from .hitbox import HitboxIE, HitboxLiveIE from .hornbunny import HornBunnyIE -from .hostingbulk import HostingBulkIE from .hotnewhiphop import HotNewHipHopIE from .howcast import HowcastIE from .howstuffworks import HowStuffWorksIE @@ -232,13 +242,21 @@ from .imdb import ( ImdbIE, ImdbListIE ) -from .imgur import ImgurIE +from .imgur import ( + ImgurIE, + ImgurAlbumIE, +) from .ina import InaIE +from .indavideo import ( + IndavideoIE, + IndavideoEmbedIE, +) from .infoq import InfoQIE from .instagram import InstagramIE, InstagramUserIE from .internetvideoarchive import InternetVideoArchiveIE from .iprima import IPrimaIE from .iqiyi import IqiyiIE +from .ir90tv import Ir90TvIE from .ivi import ( IviIE, IviCompilationIE @@ -271,6 +289,7 @@ from .kuwo import ( ) from .la7 import LA7IE from .laola1tv import Laola1TvIE +from .lecture2go import Lecture2GoIE from .letv import ( LetvIE, LetvTvIE, @@ -281,6 +300,11 @@ from .lifenews import ( LifeNewsIE, LifeEmbedIE, ) +from .limelight import ( + LimelightMediaIE, + LimelightChannelIE, + LimelightChannelListIE, +) from .liveleak import LiveLeakIE from .livestream import ( LivestreamIE, @@ -298,7 +322,6 @@ from .macgamestore import MacGameStoreIE from .mailru import MailRuIE from .malemotion import MalemotionIE from .mdr import MDRIE -from .megavideoz import MegaVideozIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mgoon import MgoonIE @@ -325,11 +348,12 @@ from .mtv import ( MTVIE, MTVServicesEmbeddedIE, MTVIggyIE, + MTVDEIE, ) from .muenchentv import MuenchenTVIE from .musicplayon import MusicPlayOnIE -from .musicvault import MusicVaultIE from .muzu import MuzuTVIE +from .mwave import MwaveIE from .myspace import MySpaceIE, MySpaceAlbumIE from .myspass import MySpassIE from .myvi import MyviIE @@ -343,10 +367,14 @@ from .nbc import ( NBCNewsIE, NBCSportsIE, NBCSportsVPlayerIE, + MSNBCIE, ) from .ndr import ( NDRIE, NJoyIE, + NDREmbedBaseIE, + NDREmbedIE, + NJoyEmbedIE, ) from .ndtv import NDTVIE from .netzkino import NetzkinoIE @@ -382,7 +410,11 @@ from .normalboots import NormalbootsIE from .nosvideo import NosVideoIE from .nova import NovaIE from .novamov import NovaMovIE -from .nowness import NownessIE +from .nowness import ( + NownessIE, + NownessPlaylistIE, + NownessSeriesIE, +) from .nowtv import NowTVIE from .nowvideo import NowVideoIE from .npo import ( @@ -390,7 +422,8 @@ from .npo import ( NPOLiveIE, NPORadioIE, NPORadioFragmentIE, - TegenlichtVproIE, + VPROIE, + WNLIE ) from .nrk import ( NRKIE, @@ -411,7 +444,6 @@ from .ooyala import ( OoyalaIE, OoyalaExternalIE, ) -from .openfilm import OpenFilmIE from .orf import ( ORFTVthekIE, ORFOE1IE, @@ -421,6 +453,10 @@ from .orf import ( from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE from .pbs import PBSIE +from .periscope import ( + PeriscopeIE, + QuickscopeIE, +) from .philharmoniedeparis import PhilharmonieDeParisIE from .phoenix import PhoenixIE from .photobucket import PhotobucketIE @@ -429,8 +465,13 @@ from .planetaplay import PlanetaPlayIE from .pladform import PladformIE from .played import PlayedIE from .playfm import PlayFMIE +from .playtvak import PlaytvakIE from .playvid import PlayvidIE from .playwire import PlaywireIE +from .pluralsight import ( + PluralsightIE, + PluralsightCourseIE, +) from .podomatic import PodomaticIE from .porn91 import Porn91IE from .pornhd import PornHdIE @@ -461,6 +502,7 @@ from .radiobremen import RadioBremenIE from .radiofrance import RadioFranceIE from .rai import RaiIE from .rbmaradio import RBMARadioIE +from .rds import RDSIE from .redtube import RedTubeIE from .restudy import RestudyIE from .reverbnation import ReverbNationIE @@ -475,6 +517,7 @@ from .rtl2 import RTL2IE from .rtp import RTPIE from .rts import RTSIE from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE +from .rtvnh import RTVNHIE from .ruhd import RUHDIE from .rutube import ( RutubeIE, @@ -501,6 +544,7 @@ from .senateisvp import SenateISVPIE 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 @@ -545,6 +589,7 @@ from .spankwire import SpankwireIE from .spiegel import SpiegelIE, SpiegelArticleIE from .spiegeltv import SpiegeltvIE from .spike import SpikeIE +from .stitcher import StitcherIE from .sport5 import Sport5IE from .sportbox import ( SportBoxIE, @@ -580,6 +625,7 @@ from .techtalks import TechTalksIE from .ted import TEDIE from .telebruxelles import TeleBruxellesIE from .telecinco import TelecincoIE +from .telegraaf import TelegraafIE from .telemb import TeleMBIE from .teletask import TeleTaskIE from .tenplay import TenPlayIE @@ -587,7 +633,10 @@ from .testurl import TestURLIE from .testtube import TestTubeIE from .tf1 import TF1IE from .theonion import TheOnionIE -from .theplatform import ThePlatformIE +from .theplatform import ( + ThePlatformIE, + ThePlatformFeedIE, +) from .thesixtyone import TheSixtyOneIE from .thisamericanlife import ThisAmericanLifeIE from .thisav import ThisAVIE @@ -645,7 +694,7 @@ from .twitch import ( TwitchBookmarksIE, TwitchStreamIE, ) -from .twitter import TwitterCardIE +from .twitter import TwitterCardIE, TwitterIE from .ubu import UbuIE from .udemy import ( UdemyIE, @@ -671,7 +720,6 @@ from .vgtv import ( from .vh1 import VH1IE from .vice import ViceIE from .viddler import ViddlerIE -from .videobam import VideoBamIE from .videodetective import VideoDetectiveIE from .videolecturesnet import VideoLecturesNetIE from .videofyme import VideofyMeIE @@ -706,6 +754,7 @@ from .vk import ( VKIE, VKUserVideosIE, ) +from .vlive import VLiveIE from .vodlocker import VodlockerIE from .voicerepublic import VoiceRepublicIE from .vporn import VpornIE diff --git a/youtube_dl/extractor/abc.py b/youtube_dl/extractor/abc.py index dc0fb85d6..c0e5d1abf 100644 --- a/youtube_dl/extractor/abc.py +++ b/youtube_dl/extractor/abc.py @@ -1,16 +1,20 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor +from ..utils import ( + ExtractorError, + js_to_json, + int_or_none, +) class ABCIE(InfoExtractor): IE_NAME = 'abc.net.au' - _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)' + _VALID_URL = r'http://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334', 'md5': 'cb3dd03b18455a661071ee1e28344d9f', 'info_dict': { @@ -19,23 +23,62 @@ class ABCIE(InfoExtractor): 'title': 'Australia to help staff Ebola treatment centre in Sierra Leone', 'description': 'md5:809ad29c67a05f54eb41f2a105693a67', }, - } + }, { + 'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326', + 'md5': 'db2a5369238b51f9811ad815b69dc086', + 'info_dict': { + 'id': 'NvqvPeNZsHU', + 'ext': 'mp4', + 'upload_date': '20150816', + 'uploader': 'ABC News (Australia)', + 'description': 'Government backbencher Warren Entsch introduces a cross-party sponsored bill to legalise same-sex marriage, saying the bill is designed to promote "an inclusive Australia, not a divided one.". Read more here: http://ab.co/1Mwc6ef', + 'uploader_id': 'NewsOnABC', + 'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill', + }, + 'add_ie': ['Youtube'], + }, { + 'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080', + 'md5': 'b96eee7c9edf4fc5a358a0252881cc1f', + 'info_dict': { + 'id': '6880080', + 'ext': 'mp3', + 'title': 'NAB lifts interest rates, following Westpac and CBA', + 'description': 'md5:f13d8edc81e462fce4a0437c7dc04728', + }, + }, { + 'url': 'http://www.abc.net.au/news/2015-10-19/6866214', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - urls_info_json = self._search_regex( - r'inlineVideoData\.push\((.*?)\);', webpage, 'video urls', - flags=re.DOTALL) - urls_info = json.loads(urls_info_json.replace('\'', '"')) + mobj = re.search( + r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);', + webpage) + if mobj is None: + raise ExtractorError('Unable to extract video urls') + + urls_info = self._parse_json( + mobj.group('json_data'), video_id, transform_source=js_to_json) + + if not isinstance(urls_info, list): + urls_info = [urls_info] + + if mobj.group('type') == 'YouTube': + return self.playlist_result([ + self.url_result(url_info['url']) for url_info in urls_info]) + formats = [{ 'url': url_info['url'], - 'width': int(url_info['width']), - 'height': int(url_info['height']), - 'tbr': int(url_info['bitrate']), - 'filesize': int(url_info['filesize']), + 'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none', + 'width': int_or_none(url_info.get('width')), + 'height': int_or_none(url_info.get('height')), + 'tbr': int_or_none(url_info.get('bitrate')), + 'filesize': int_or_none(url_info.get('filesize')), } for url_info in urls_info] + self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/academicearth.py b/youtube_dl/extractor/academicearth.py index 47313fba8..34095501c 100644 --- a/youtube_dl/extractor/academicearth.py +++ b/youtube_dl/extractor/academicearth.py @@ -15,7 +15,7 @@ class AcademicEarthCourseIE(InfoExtractor): 'title': 'Laws of Nature', 'description': 'Introduce yourself to the laws of nature with these free online college lectures from Yale, Harvard, and MIT.', }, - 'playlist_count': 4, + 'playlist_count': 3, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/adultswim.py b/youtube_dl/extractor/adultswim.py index 39335b827..3ae618e71 100644 --- a/youtube_dl/extractor/adultswim.py +++ b/youtube_dl/extractor/adultswim.py @@ -5,6 +5,7 @@ import re from .common import InfoExtractor from ..utils import ( + determine_ext, ExtractorError, float_or_none, xpath_text, @@ -40,7 +41,8 @@ class AdultSwimIE(InfoExtractor): 'id': 'rQxZvXQ4ROaSOqq-or2Mow', 'title': 'Rick and Morty - Pilot', 'description': "Rick moves in with his daughter's family and establishes himself as a bad influence on his grandson, Morty. " - } + }, + 'skip': 'This video is only available for registered users', }, { 'url': 'http://www.adultswim.com/videos/playlists/american-parenting/putting-francine-out-of-business/', 'playlist': [ @@ -123,7 +125,6 @@ class AdultSwimIE(InfoExtractor): else: collections = bootstrapped_data['show']['collections'] collection, video_info = self.find_collection_containing_video(collections, episode_path) - # Video wasn't found in the collections, let's try `slugged_video`. if video_info is None: if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path: @@ -133,7 +134,15 @@ class AdultSwimIE(InfoExtractor): show = bootstrapped_data['show'] show_title = show['title'] - segment_ids = [clip['videoPlaybackID'] for clip in video_info['clips']] + stream = video_info.get('stream') + clips = [stream] if stream else video_info.get('clips') + if not clips: + raise ExtractorError( + 'This video is only available via cable service provider subscription that' + ' is not currently supported. You may want to use --cookies.' + if video_info.get('auth') is True else 'Unable to find stream or clips', + expected=True) + segment_ids = [clip['videoPlaybackID'] for clip in clips] episode_id = video_info['id'] episode_title = video_info['title'] @@ -142,7 +151,7 @@ class AdultSwimIE(InfoExtractor): entries = [] for part_num, segment_id in enumerate(segment_ids): - segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=mobile' % segment_id + segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=desktop' % segment_id segment_title = '%s - %s' % (show_title, episode_title) if len(segment_ids) > 1: @@ -156,19 +165,32 @@ class AdultSwimIE(InfoExtractor): xpath_text(idoc, './/trt', 'segment duration').strip()) formats = [] - file_els = idoc.findall('.//files/file') + file_els = idoc.findall('.//files/file') or idoc.findall('./files/file') + unique_urls = [] + unique_file_els = [] for file_el in file_els: + media_url = file_el.text + if not media_url or determine_ext(media_url) == 'f4m': + continue + if file_el.text not in unique_urls: + unique_urls.append(file_el.text) + unique_file_els.append(file_el) + + for file_el in unique_file_els: bitrate = file_el.attrib.get('bitrate') ftype = file_el.attrib.get('type') - - formats.append({ - 'format_id': '%s_%s' % (bitrate, ftype), - 'url': file_el.text.strip(), - # The bitrate may not be a number (for example: 'iphone') - 'tbr': int(bitrate) if bitrate.isdigit() else None, - 'quality': 1 if ftype == 'hd' else -1 - }) + media_url = file_el.text + if determine_ext(media_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, segment_title, 'mp4', preference=0, m3u8_id='hls')) + else: + formats.append({ + 'format_id': '%s_%s' % (bitrate, ftype), + 'url': file_el.text.strip(), + # The bitrate may not be a number (for example: 'iphone') + 'tbr': int(bitrate) if bitrate.isdigit() else None, + }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/airmozilla.py b/youtube_dl/extractor/airmozilla.py index 611ad1e9d..f8e70f4e5 100644 --- a/youtube_dl/extractor/airmozilla.py +++ b/youtube_dl/extractor/airmozilla.py @@ -20,14 +20,14 @@ class AirMozillaIE(InfoExtractor): 'id': '6x4q2w', 'ext': 'mp4', 'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco', - 'thumbnail': 're:https://\w+\.cloudfront\.net/6x4q2w/poster\.jpg\?t=\d+', + 'thumbnail': 're:https?://vid\.ly/(?P<id>[0-9a-z-]+)/poster', 'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...', 'timestamp': 1422487800, 'upload_date': '20150128', 'location': 'SFO Commons', 'duration': 3780, 'view_count': int, - 'categories': ['Main'], + 'categories': ['Main', 'Privacy'], } } diff --git a/youtube_dl/extractor/aljazeera.py b/youtube_dl/extractor/aljazeera.py index 612708e25..184a14a4f 100644 --- a/youtube_dl/extractor/aljazeera.py +++ b/youtube_dl/extractor/aljazeera.py @@ -16,6 +16,7 @@ class AlJazeeraIE(InfoExtractor): 'uploader': 'Al Jazeera English', }, 'add_ie': ['Brightcove'], + 'skip': 'Not accessible from Travis CI server', } def _real_extract(self, url): diff --git a/youtube_dl/extractor/anitube.py b/youtube_dl/extractor/anitube.py index 31f0d417c..23f942ae2 100644 --- a/youtube_dl/extractor/anitube.py +++ b/youtube_dl/extractor/anitube.py @@ -26,8 +26,8 @@ class AnitubeIE(InfoExtractor): 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, 'key') + key = self._search_regex( + r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key') config_xml = self._download_xml( 'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key) diff --git a/youtube_dl/extractor/appleconnect.py b/youtube_dl/extractor/appleconnect.py new file mode 100644 index 000000000..ea7a70393 --- /dev/null +++ b/youtube_dl/extractor/appleconnect.py @@ -0,0 +1,50 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + str_to_int, + ExtractorError +) + + +class AppleConnectIE(InfoExtractor): + _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)' + _TEST = { + 'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3', + 'md5': '10d0f2799111df4cb1c924520ca78f98', + 'info_dict': { + 'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3', + 'ext': 'm4v', + 'title': 'Energy', + 'uploader': 'Drake', + 'thumbnail': 'http://is5.mzstatic.com/image/thumb/Video5/v4/78/61/c5/7861c5fa-ad6d-294b-1464-cf7605b911d6/source/1920x1080sr.jpg', + 'upload_date': '20150710', + 'timestamp': 1436545535, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + try: + video_json = self._html_search_regex( + r'class="auc-video-data">(\{.*?\})', webpage, 'json') + except ExtractorError: + raise ExtractorError('This post doesn\'t contain a video', expected=True) + + video_data = self._parse_json(video_json, video_id) + timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp')) + like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count')) + + return { + 'id': video_id, + 'url': video_data['sslSrc'], + 'title': video_data['title'], + 'description': video_data['description'], + 'uploader': video_data['artistName'], + 'thumbnail': video_data['artworkUrl'], + 'timestamp': timestamp, + 'like_count': like_count, + } diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py index 576f03b5b..f68dc3236 100644 --- a/youtube_dl/extractor/appletrailers.py +++ b/youtube_dl/extractor/appletrailers.py @@ -13,53 +13,53 @@ from ..utils import ( class AppleTrailersIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)' _TESTS = [{ - "url": "http://trailers.apple.com/trailers/wb/manofsteel/", + 'url': 'http://trailers.apple.com/trailers/wb/manofsteel/', 'info_dict': { 'id': 'manofsteel', }, - "playlist": [ + 'playlist': [ { - "md5": "d97a8e575432dbcb81b7c3acb741f8a8", - "info_dict": { - "id": "manofsteel-trailer4", - "ext": "mov", - "duration": 111, - "title": "Trailer 4", - "upload_date": "20130523", - "uploader_id": "wb", + 'md5': 'd97a8e575432dbcb81b7c3acb741f8a8', + 'info_dict': { + 'id': 'manofsteel-trailer4', + 'ext': 'mov', + 'duration': 111, + 'title': 'Trailer 4', + 'upload_date': '20130523', + 'uploader_id': 'wb', }, }, { - "md5": "b8017b7131b721fb4e8d6f49e1df908c", - "info_dict": { - "id": "manofsteel-trailer3", - "ext": "mov", - "duration": 182, - "title": "Trailer 3", - "upload_date": "20130417", - "uploader_id": "wb", + 'md5': 'b8017b7131b721fb4e8d6f49e1df908c', + 'info_dict': { + 'id': 'manofsteel-trailer3', + 'ext': 'mov', + 'duration': 182, + 'title': 'Trailer 3', + 'upload_date': '20130417', + 'uploader_id': 'wb', }, }, { - "md5": "d0f1e1150989b9924679b441f3404d48", - "info_dict": { - "id": "manofsteel-trailer", - "ext": "mov", - "duration": 148, - "title": "Trailer", - "upload_date": "20121212", - "uploader_id": "wb", + 'md5': 'd0f1e1150989b9924679b441f3404d48', + 'info_dict': { + 'id': 'manofsteel-trailer', + 'ext': 'mov', + 'duration': 148, + 'title': 'Trailer', + 'upload_date': '20121212', + 'uploader_id': 'wb', }, }, { - "md5": "5fe08795b943eb2e757fa95cb6def1cb", - "info_dict": { - "id": "manofsteel-teaser", - "ext": "mov", - "duration": 93, - "title": "Teaser", - "upload_date": "20120721", - "uploader_id": "wb", + 'md5': '5fe08795b943eb2e757fa95cb6def1cb', + 'info_dict': { + 'id': 'manofsteel-teaser', + 'ext': 'mov', + 'duration': 93, + 'title': 'Teaser', + 'upload_date': '20120721', + 'uploader_id': 'wb', }, }, ] diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 6a35ea463..6f465789b 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -8,6 +8,7 @@ from .generic import GenericIE from ..utils import ( determine_ext, ExtractorError, + get_element_by_attribute, qualities, int_or_none, parse_duration, @@ -22,19 +23,125 @@ class ARDMediathekIE(InfoExtractor): _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?' _TESTS = [{ - 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', - 'only_matching': True, + 'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114', + 'info_dict': { + 'id': '29582122', + 'ext': 'mp4', + 'title': 'Ich liebe das Leben trotzdem', + 'description': 'md5:45e4c225c72b27993314b31a84a5261c', + 'duration': 4557, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { - 'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916', + 'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916', + 'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e', 'info_dict': { - 'id': '22490580', + 'id': '29522730', 'ext': 'mp4', - 'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)', - 'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.', + 'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)', + 'description': 'md5:196392e79876d0ac94c94e8cdb2875f1', + 'duration': 5252, }, - 'skip': 'Blocked outside of Germany', + }, { + # 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', + 'md5': '219d94d8980b4f538c7fcb0865eb7f2c', + 'info_dict': { + 'id': '28488308', + 'ext': 'mp3', + 'title': 'Tod eines Fußballers', + 'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef', + 'duration': 3240, + }, + }, { + 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', + 'only_matching': True, }] + def _extract_media_info(self, media_info_url, webpage, video_id): + media_info = self._download_json( + media_info_url, video_id, 'Downloading media JSON') + + formats = self._extract_formats(media_info, video_id) + + if not formats: + if '"fsk"' in webpage: + raise ExtractorError( + 'This video is only available after 20:00', expected=True) + elif media_info.get('_geoblocked'): + raise ExtractorError('This video is not available due to geo restriction', expected=True) + + self._sort_formats(formats) + + duration = int_or_none(media_info.get('_duration')) + thumbnail = media_info.get('_previewImage') + + subtitles = {} + subtitle_url = media_info.get('_subtitleUrl') + if subtitle_url: + subtitles['de'] = [{ + 'ext': 'srt', + 'url': subtitle_url, + }] + + return { + 'id': video_id, + 'duration': duration, + 'thumbnail': thumbnail, + 'formats': formats, + 'subtitles': subtitles, + } + + def _extract_formats(self, media_info, video_id): + type_ = media_info.get('_type') + media_array = media_info.get('_mediaArray', []) + formats = [] + for num, media in enumerate(media_array): + for stream in media.get('_mediaStreamArray', []): + stream_urls = stream.get('_stream') + if not stream_urls: + continue + if not isinstance(stream_urls, list): + stream_urls = [stream_urls] + quality = stream.get('_quality') + server = stream.get('_server') + for stream_url in stream_urls: + ext = determine_ext(stream_url) + 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')) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + stream_url, video_id, 'mp4', preference=1, m3u8_id='hls')) + else: + if server and server.startswith('rtmp'): + f = { + 'url': server, + 'play_path': stream_url, + 'format_id': 'a%s-rtmp-%s' % (num, quality), + } + elif stream_url.startswith('http'): + f = { + 'url': stream_url, + 'format_id': 'a%s-%s-%s' % (num, ext, quality) + } + else: + continue + m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url) + if m: + f.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + if type_ == 'audio': + f['vcodec'] = 'none' + formats.append(f) + return formats + def _real_extract(self, url): # determine video id from url m = re.match(self._VALID_URL, url) @@ -92,46 +199,22 @@ class ARDMediathekIE(InfoExtractor): 'format_id': fid, 'url': furl, }) + self._sort_formats(formats) + info = { + 'formats': formats, + } else: # request JSON file - media_info = self._download_json( - 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id) - # The second element of the _mediaArray contains the standard http urls - streams = media_info['_mediaArray'][1]['_mediaStreamArray'] - if not streams: - if '"fsk"' in webpage: - raise ExtractorError('This video is only available after 20:00') - - formats = [] - for s in streams: - if type(s['_stream']) == list: - for index, url in enumerate(s['_stream'][::-1]): - quality = s['_quality'] + index - formats.append({ - 'quality': quality, - 'url': url, - 'format_id': '%s-%s' % (determine_ext(url), quality) - }) - continue - - format = { - 'quality': s['_quality'], - 'url': s['_stream'], - } - - format['format_id'] = '%s-%s' % ( - determine_ext(format['url']), format['quality']) + info = self._extract_media_info( + 'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id) - formats.append(format) - - self._sort_formats(formats) - - return { + info.update({ 'id': video_id, 'title': title, 'description': description, - 'formats': formats, 'thumbnail': thumbnail, - } + }) + + return info class ARDIE(InfoExtractor): @@ -189,3 +272,41 @@ 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 76de24477..2a00da3ee 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -4,6 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( find_xpath_attr, unified_strdate, @@ -77,7 +81,13 @@ class ArteTVPlus7IE(InfoExtractor): def _extract_from_webpage(self, webpage, video_id, lang): json_url = self._html_search_regex( [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'], - webpage, 'json vp url') + webpage, 'json vp url', default=None) + if not json_url: + iframe_url = self._html_search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1', + webpage, 'iframe url', group='url') + json_url = compat_parse_qs( + compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0] return self._extract_from_json_url(json_url, video_id, lang) def _extract_from_json_url(self, json_url, video_id, lang): diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 505877b77..c1ef8051d 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -10,6 +10,8 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + float_or_none, + int_or_none, ) @@ -52,11 +54,11 @@ class BandcampIE(InfoExtractor): ext, abr_str = format_id.split('-', 1) formats.append({ 'format_id': format_id, - 'url': format_url, + 'url': self._proto_relative_url(format_url, 'http:'), 'ext': ext, 'vcodec': 'none', 'acodec': ext, - 'abr': int(abr_str), + 'abr': int_or_none(abr_str), }) self._sort_formats(formats) @@ -65,7 +67,7 @@ class BandcampIE(InfoExtractor): 'id': compat_str(data['id']), 'title': data['title'], 'formats': formats, - 'duration': float(data['duration']), + 'duration': float_or_none(data.get('duration')), } else: raise ExtractorError('No free songs found') @@ -93,8 +95,8 @@ class BandcampIE(InfoExtractor): final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url') # If we could correctly generate the .rand field the url would be # in the "download_url" key - final_url = self._search_regex( - r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL') + final_url = self._proto_relative_url(self._search_regex( + r'"retry_url":"(.+?)"', final_url_webpage, 'final video URL'), 'http:') return { 'id': video_id, diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py new file mode 100644 index 000000000..2cdce1eb9 --- /dev/null +++ b/youtube_dl/extractor/bbc.py @@ -0,0 +1,935 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + float_or_none, + int_or_none, + parse_duration, + parse_iso8601, + remove_end, + unescapeHTML, +) +from ..compat import compat_HTTPError + + +class BBCCoUkIE(InfoExtractor): + IE_NAME = 'bbc.co.uk' + IE_DESC = 'BBC iPlayer' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P<id>[\da-z]{8})' + + _MEDIASELECTOR_URLS = [ + # Provides HQ HLS streams with even better quality that pc mediaset but fails + # with geolocation in some cases when it's even not geo restricted at all (e.g. + # http://www.bbc.co.uk/programmes/b06bp7lf) + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s', + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s', + ] + + _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection' + _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist' + + _NAMESPACES = ( + _MEDIASELECTION_NS, + _EMP_PLAYLIST_NS, + ) + + _TESTS = [ + { + 'url': 'http://www.bbc.co.uk/programmes/b039g8p7', + 'info_dict': { + 'id': 'b039d07m', + 'ext': 'flv', + 'title': 'Kaleidoscope, Leonard Cohen', + 'description': 'The Canadian poet and songwriter reflects on his musical career.', + 'duration': 1740, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/', + 'info_dict': { + 'id': 'b00yng1d', + 'ext': 'flv', + 'title': 'The Man in Black: Series 3: The Printed Name', + 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.", + 'duration': 1800, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Episode is no longer available on BBC iPlayer Radio', + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/', + 'info_dict': { + 'id': 'b00yng1d', + 'ext': 'flv', + 'title': 'The Voice UK: Series 3: Blind Auditions 5', + 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.", + 'duration': 5100, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion', + 'info_dict': { + 'id': 'b03k3pb7', + 'ext': 'flv', + 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction", + 'description': '2. Invasion', + 'duration': 3600, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, { + 'url': 'http://www.bbc.co.uk/programmes/b04v20dw', + 'info_dict': { + 'id': 'b04v209v', + 'ext': 'flv', + 'title': 'Pete Tong, The Essential New Tune Special', + 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!", + 'duration': 10800, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3', + 'note': 'Audio', + 'info_dict': { + 'id': 'p02frcch', + 'ext': 'flv', + 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix', + 'description': 'French house superstar Madeon takes us out of the club and onto the after party.', + 'duration': 3507, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz', + 'note': 'Video', + 'info_dict': { + 'id': 'p025c103', + 'ext': 'flv', + 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)', + 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014', + 'duration': 226, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls', + 'info_dict': { + 'id': 'p02n76xf', + 'ext': 'flv', + 'title': 'Natural World, 2015-2016: 2. Super Powered Owls', + 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d', + 'duration': 3540, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'geolocation', + }, { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition', + 'info_dict': { + 'id': 'b05zmgw1', + 'ext': 'flv', + 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.', + 'title': 'Royal Academy Summer Exhibition', + 'duration': 3540, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'geolocation', + }, { + # iptv-all mediaset fails with geolocation however there is no geo restriction + # for this programme at all + 'url': 'http://www.bbc.co.uk/programmes/b06bp7lf', + 'info_dict': { + 'id': 'b06bp7kf', + 'ext': 'flv', + 'title': "Annie Mac's Friday Night, B.Traits sits in for Annie", + 'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.', + 'duration': 10800, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo', + 'only_matching': True, + } + ] + + class MediaSelectionError(Exception): + def __init__(self, id): + self.id = id + + def _extract_asx_playlist(self, connection, programme_id): + asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist') + return [ref.get('href') for ref in asx.findall('./Entry/ref')] + + def _extract_connection(self, connection, programme_id): + formats = [] + kind = connection.get('kind') + protocol = connection.get('protocol') + supplier = connection.get('supplier') + if protocol == 'http': + href = connection.get('href') + transfer_format = connection.get('transferFormat') + # ASX playlist + if supplier == 'asx': + for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)): + formats.append({ + 'url': ref, + 'format_id': 'ref%s_%s' % (i, supplier), + }) + # Skip DASH until supported + elif transfer_format == 'dash': + pass + elif transfer_format == 'hls': + m3u8_formats = self._extract_m3u8_formats( + href, programme_id, ext='mp4', entry_protocol='m3u8_native', + m3u8_id=supplier, fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + # Direct link + else: + formats.append({ + 'url': href, + 'format_id': supplier or kind or protocol, + }) + elif protocol == 'rtmp': + application = connection.get('application', 'ondemand') + auth_string = connection.get('authString') + identifier = connection.get('identifier') + server = connection.get('server') + formats.append({ + 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string), + 'play_path': identifier, + 'app': '%s?%s' % (application, auth_string), + 'page_url': 'http://www.bbc.co.uk', + 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf', + 'rtmp_live': False, + 'ext': 'flv', + 'format_id': supplier, + }) + return formats + + def _extract_items(self, playlist): + return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS) + + def _findall_ns(self, element, xpath): + elements = [] + for ns in self._NAMESPACES: + elements.extend(element.findall(xpath % ns)) + return elements + + def _extract_medias(self, media_selection): + error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS) + if error is None: + media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS) + if error is not None: + raise BBCCoUkIE.MediaSelectionError(error.get('id')) + return self._findall_ns(media_selection, './{%s}media') + + def _extract_connections(self, media): + return self._findall_ns(media, './{%s}connection') + + def _extract_video(self, media, programme_id): + formats = [] + vbr = int_or_none(media.get('bitrate')) + vcodec = media.get('encoding') + service = media.get('service') + width = int_or_none(media.get('width')) + height = int_or_none(media.get('height')) + file_size = int_or_none(media.get('media_file_size')) + for connection in self._extract_connections(media): + conn_formats = self._extract_connection(connection, programme_id) + for format in conn_formats: + format.update({ + 'width': width, + 'height': height, + 'vbr': vbr, + 'vcodec': vcodec, + 'filesize': file_size, + }) + if service: + format['format_id'] = '%s_%s' % (service, format['format_id']) + formats.extend(conn_formats) + return formats + + def _extract_audio(self, media, programme_id): + formats = [] + abr = int_or_none(media.get('bitrate')) + acodec = media.get('encoding') + service = media.get('service') + for connection in self._extract_connections(media): + conn_formats = self._extract_connection(connection, programme_id) + for format in conn_formats: + format.update({ + 'format_id': '%s_%s' % (service, format['format_id']), + 'abr': abr, + 'acodec': acodec, + }) + formats.extend(conn_formats) + return formats + + def _get_subtitles(self, media, programme_id): + subtitles = {} + for connection in self._extract_connections(media): + captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') + lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') + subtitles[lang] = [ + { + 'url': connection.get('href'), + 'ext': 'ttml', + }, + ] + return subtitles + + def _raise_extractor_error(self, media_selection_error): + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, media_selection_error.id), + expected=True) + + def _download_media_selector(self, programme_id): + last_exception = None + for mediaselector_url in self._MEDIASELECTOR_URLS: + try: + return self._download_media_selector_url( + mediaselector_url % programme_id, programme_id) + except BBCCoUkIE.MediaSelectionError as e: + if e.id in ('notukerror', 'geolocation'): + last_exception = e + continue + self._raise_extractor_error(e) + self._raise_extractor_error(last_exception) + + def _download_media_selector_url(self, url, programme_id=None): + try: + media_selection = self._download_xml( + url, programme_id, 'Downloading media selection XML') + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: + media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) + else: + raise + return self._process_media_selector(media_selection, programme_id) + + def _process_media_selector(self, media_selection, programme_id): + formats = [] + subtitles = None + + for media in self._extract_medias(media_selection): + kind = media.get('kind') + if kind == 'audio': + formats.extend(self._extract_audio(media, programme_id)) + elif kind == 'video': + formats.extend(self._extract_video(media, programme_id)) + elif kind == 'captions': + subtitles = self.extract_subtitles(media, programme_id) + return formats, subtitles + + def _download_playlist(self, playlist_id): + try: + playlist = self._download_json( + 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id, + playlist_id, 'Downloading playlist JSON') + + version = playlist.get('defaultAvailableVersion') + if version: + smp_config = version['smpConfig'] + title = smp_config['title'] + description = smp_config['summary'] + for item in smp_config['items']: + kind = item['kind'] + if kind != 'programme' and kind != 'radioProgramme': + continue + programme_id = item.get('vpid') + duration = int_or_none(item.get('duration')) + formats, subtitles = self._download_media_selector(programme_id) + return programme_id, title, description, duration, formats, subtitles + except ExtractorError as ee: + if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404): + raise + + # fallback to legacy playlist + return self._process_legacy_playlist(playlist_id) + + def _process_legacy_playlist_url(self, url, display_id): + playlist = self._download_legacy_playlist_url(url, display_id) + return self._extract_from_legacy_playlist(playlist, display_id) + + def _process_legacy_playlist(self, playlist_id): + return self._process_legacy_playlist_url( + 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id) + + def _download_legacy_playlist_url(self, url, playlist_id=None): + return self._download_xml( + url, playlist_id, 'Downloading legacy playlist XML') + + def _extract_from_legacy_playlist(self, playlist, playlist_id): + no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS) + if no_items is not None: + reason = no_items.get('reason') + if reason == 'preAvailability': + msg = 'Episode %s is not yet available' % playlist_id + elif reason == 'postAvailability': + msg = 'Episode %s is no longer available' % playlist_id + elif reason == 'noMedia': + msg = 'Episode %s is not currently available' % playlist_id + else: + msg = 'Episode %s is not available: %s' % (playlist_id, reason) + raise ExtractorError(msg, expected=True) + + for item in self._extract_items(playlist): + kind = item.get('kind') + if kind != 'programme' and kind != 'radioProgramme': + continue + title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text + description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS) + description = description_el.text if description_el is not None else None + + def get_programme_id(item): + def get_from_attributes(item): + for p in('identifier', 'group'): + value = item.get(p) + if value and re.match(r'^[pb][\da-z]{7}$', value): + return value + get_from_attributes(item) + mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS) + if mediator is not None: + return get_from_attributes(mediator) + + programme_id = get_programme_id(item) + duration = int_or_none(item.get('duration')) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + else: + formats, subtitles = self._process_media_selector(item, playlist_id) + programme_id = playlist_id + + return programme_id, title, description, duration, formats, subtitles + + def _real_extract(self, url): + group_id = self._match_id(url) + + webpage = self._download_webpage(url, group_id, 'Downloading video page') + + programme_id = None + + tviplayer = self._search_regex( + r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById', + webpage, 'player', default=None) + + if tviplayer: + player = self._parse_json(tviplayer, group_id).get('player', {}) + duration = int_or_none(player.get('duration')) + programme_id = player.get('vpid') + + if not programme_id: + programme_id = self._search_regex( + r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + title = self._og_search_title(webpage) + description = self._search_regex( + r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', + webpage, 'description', fatal=False) + else: + programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) + + self._sort_formats(formats) + + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + } + + +class BBCIE(BBCCoUkIE): + IE_NAME = 'bbc' + IE_DESC = 'BBC' + _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)' + + _MEDIASELECTOR_URLS = [ + # Provides HQ HLS streams but fails with geolocation in some cases when it's + # even not geo restricted at all + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s', + # Provides more formats, namely direct mp4 links, but fails on some videos with + # notukerror for non UK (?) users (e.g. + # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) + 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s', + # Provides fewer formats, but works everywhere for everybody (hopefully) + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s', + ] + + _TESTS = [{ + # article with multiple videos embedded with data-playable containing vpids + 'url': 'http://www.bbc.com/news/world-europe-32668511', + 'info_dict': { + 'id': 'world-europe-32668511', + 'title': 'Russia stages massive WW2 parade despite Western boycott', + 'description': 'md5:00ff61976f6081841f759a08bf78cc9c', + }, + 'playlist_count': 2, + }, { + # article with multiple videos embedded with data-playable (more videos) + 'url': 'http://www.bbc.com/news/business-28299555', + 'info_dict': { + 'id': 'business-28299555', + 'title': 'Farnborough Airshow: Video highlights', + 'description': 'BBC reports and video highlights at the Farnborough Airshow.', + }, + 'playlist_count': 9, + 'skip': 'Save time', + }, { + # article with multiple videos embedded with `new SMP()` + # broken + 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460', + 'info_dict': { + 'id': '3662a707-0af9-3149-963f-47bea720b460', + 'title': 'BBC Blogs - Adam Curtis - BUGGER', + }, + 'playlist_count': 18, + }, { + # single video embedded with data-playable containing vpid + 'url': 'http://www.bbc.com/news/world-europe-32041533', + 'info_dict': { + 'id': 'p02mprgb', + 'ext': 'mp4', + 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV', + 'description': 'md5:2868290467291b37feda7863f7a83f54', + 'duration': 47, + 'timestamp': 1427219242, + 'upload_date': '20150324', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # article with single video embedded with data-playable containing XML playlist + # with direct video links as progressiveDownloadUrl (for now these are extracted) + # and playlist with f4m and m3u8 as streamingUrl + 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu', + 'info_dict': { + 'id': '150615_telabyad_kentin_cogu', + 'ext': 'mp4', + 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde", + 'timestamp': 1434397334, + 'upload_date': '20150615', + }, + 'params': { + 'skip_download': True, + } + }, { + # single video embedded with data-playable containing XML playlists (regional section) + 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw', + 'info_dict': { + 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw', + 'ext': 'mp4', + 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción', + 'timestamp': 1434713142, + 'upload_date': '20150619', + }, + 'params': { + 'skip_download': True, + } + }, { + # single video from video playlist embedded with vxp-playlist-data JSON + 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376', + 'info_dict': { + 'id': 'p02w6qjc', + 'ext': 'mp4', + 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''', + 'duration': 56, + }, + 'params': { + 'skip_download': True, + } + }, { + # single video story with digitalData + 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret', + 'info_dict': { + 'id': 'p02q6gc4', + 'ext': 'flv', + 'title': 'Sri Lanka’s spicy secret', + 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.', + 'timestamp': 1437674293, + 'upload_date': '20150723', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # single video story without digitalData + 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star', + 'info_dict': { + 'id': 'p018zqqg', + 'ext': 'mp4', + 'title': 'Hyundai Santa Fe Sport: Rock star', + 'description': 'md5:b042a26142c4154a6e472933cf20793d', + 'timestamp': 1415867444, + 'upload_date': '20141113', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # single video with playlist.sxml URL in playlist param + 'url': 'http://www.bbc.com/sport/0/football/33653409', + 'info_dict': { + 'id': 'p02xycnp', + 'ext': 'mp4', + 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?', + 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.', + 'duration': 140, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # article with multiple videos embedded with playlist.sxml in playlist param + 'url': 'http://www.bbc.com/sport/0/football/34475836', + 'info_dict': { + 'id': '34475836', + 'title': 'What Liverpool can expect from Klopp', + }, + 'playlist_count': 3, + }, { + # single video with playlist URL from weather section + 'url': 'http://www.bbc.com/weather/features/33601775', + 'only_matching': True, + }, { + # custom redirection to www.bbc.com + 'url': 'http://www.bbc.co.uk/news/science-environment-33661876', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) 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. + # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu) + # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml + source_files = media_meta.get('sourceFiles') + if source_files: + return [{ + 'url': f['url'], + 'format_id': format_id, + 'ext': f.get('encoding'), + 'tbr': float_or_none(f.get('bitrate'), 1000), + 'filesize': int_or_none(f.get('filesize')), + } for format_id, f in source_files.items() if f.get('url')], [] + + programme_id = media_meta.get('externalId') + if programme_id: + return self._download_media_selector(programme_id) + + # Process playlist.sxml as legacy playlist + href = media_meta.get('href') + if href: + playlist = self._download_legacy_playlist_url(href) + _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id) + return formats, subtitles + + return [], [] + + def _extract_from_playlist_sxml(self, url, playlist_id, timestamp): + programme_id, title, description, duration, formats, subtitles = \ + self._process_legacy_playlist_url(url, playlist_id) + self._sort_formats(formats) + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + timestamp = None + playlist_title = None + playlist_description = None + + ld = self._parse_json( + self._search_regex( + r'(?s)<script type="application/ld\+json">(.+?)</script>', + webpage, 'ld json', default='{}'), + playlist_id, fatal=False) + if ld: + timestamp = parse_iso8601(ld.get('datePublished')) + playlist_title = ld.get('headline') + playlist_description = ld.get('articleBody') + + if not timestamp: + timestamp = parse_iso8601(self._search_regex( + [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"', + r'itemprop="datePublished"[^>]+datetime="([^"]+)"', + r'"datePublished":\s*"([^"]+)'], + webpage, 'date', default=None)) + + entries = [] + + # article with multiple videos embedded with playlist.sxml (e.g. + # http://www.bbc.com/sport/0/football/34475836) + playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage) + if playlists: + entries = [ + self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp) + for playlist_url in playlists] + + # news article with multiple videos embedded with data-playable + data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage) + if data_playables: + for _, data_playable_json in data_playables: + data_playable = self._parse_json( + unescapeHTML(data_playable_json), playlist_id, fatal=False) + if not data_playable: + continue + settings = data_playable.get('settings', {}) + if settings: + # data-playable with video vpid in settings.playlistObject.items (e.g. + # http://www.bbc.com/news/world-us-canada-34473351) + playlist_object = settings.get('playlistObject', {}) + if playlist_object: + items = playlist_object.get('items') + if items and isinstance(items, list): + title = playlist_object['title'] + description = playlist_object.get('summary') + duration = int_or_none(items[0].get('duration')) + programme_id = items[0].get('vpid') + formats, subtitles = self._download_media_selector(programme_id) + self._sort_formats(formats) + entries.append({ + 'id': programme_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + }) + else: + # data-playable without vpid but with a playlist.sxml URLs + # in otherSettings.playlist (e.g. + # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani) + playlist = data_playable.get('otherSettings', {}).get('playlist', {}) + if playlist: + entries.append(self._extract_from_playlist_sxml( + playlist.get('progressiveDownloadUrl'), playlist_id, timestamp)) + + if entries: + playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News') + playlist_description = playlist_description or self._og_search_description(webpage, default=None) + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) + + # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) + programme_id = self._search_regex( + [r'data-video-player-vpid="([\da-z]{8})"', + r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'], + webpage, 'vpid', default=None) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + self._sort_formats(formats) + # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star) + digital_data = self._parse_json( + self._search_regex( + r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'), + programme_id, fatal=False) + page_info = digital_data.get('page', {}).get('pageInfo', {}) + title = page_info.get('pageName') or self._og_search_title(webpage) + description = page_info.get('description') or self._og_search_description(webpage) + timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + } + + playlist_title = self._html_search_regex( + r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title') + playlist_description = self._og_search_description(webpage, default=None) + + def extract_all(pattern): + return list(filter(None, map( + lambda s: self._parse_json(s, playlist_id, fatal=False), + re.findall(pattern, webpage)))) + + # Multiple video article (e.g. + # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460) + EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?' + entries = [] + for match in extract_all(r'new\s+SMP\(({.+?})\)'): + embed_url = match.get('playerSettings', {}).get('externalEmbedUrl') + if embed_url and re.match(EMBED_URL, embed_url): + entries.append(embed_url) + entries.extend(re.findall( + r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage)) + if entries: + return self.playlist_result( + [self.url_result(entry, 'BBCCoUk') for entry in entries], + playlist_id, playlist_title, playlist_description) + + # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511) + medias = extract_all(r"data-media-meta='({[^']+})'") + + if not medias: + # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international) + media_asset = self._search_regex( + r'mediaAssetPage\.init\(\s*({.+?}), "/', + webpage, 'media asset', default=None) + if media_asset: + media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False) + medias = [] + for video in media_asset_page.get('videos', {}).values(): + medias.extend(video.values()) + + if not medias: + # Multiple video playlist with single `now playing` entry (e.g. + # http://www.bbc.com/news/video_and_audio/must_see/33767813) + vxp_playlist = self._parse_json( + self._search_regex( + r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>', + webpage, 'playlist data'), + playlist_id) + playlist_medias = [] + for item in vxp_playlist: + media = item.get('media') + if not media: + continue + playlist_medias.append(media) + # Download single video if found media with asset id matching the video id from URL + if item.get('advert', {}).get('assetId') == playlist_id: + medias = [media] + break + # Fallback to the whole playlist + if not medias: + medias = playlist_medias + + entries = [] + for num, media_meta in enumerate(medias, start=1): + formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id) + if not formats: + continue + self._sort_formats(formats) + + video_id = media_meta.get('externalId') + if not video_id: + video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num) + + title = media_meta.get('caption') + if not title: + title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num) + + duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration')) + + images = [] + for image in media_meta.get('images', {}).values(): + images.extend(image.values()) + if 'image' in media_meta: + images.append(media_meta['image']) + + thumbnails = [{ + 'url': image.get('href'), + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in images] + + entries.append({ + 'id': video_id, + 'title': title, + 'thumbnails': thumbnails, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + }) + + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) + + +class BBCCoUkArticleIE(InfoExtractor): + _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)' + IE_NAME = 'bbc.co.uk:article' + IE_DESC = 'BBC articles' + + _TEST = { + 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer', + 'info_dict': { + 'id': '3jNQLTMrPlYGTBn0WV6M2MS', + 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four', + 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.', + }, + 'playlist_count': 4, + 'add_ie': ['BBCCoUk'], + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage).strip() + + entries = [self.url_result(programme_url) for programme_url in re.findall( + r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)] + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/bbccouk.py b/youtube_dl/extractor/bbccouk.py deleted file mode 100644 index 5825d2867..000000000 --- a/youtube_dl/extractor/bbccouk.py +++ /dev/null @@ -1,379 +0,0 @@ -from __future__ import unicode_literals - -import xml.etree.ElementTree - -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - int_or_none, -) -from ..compat import compat_HTTPError - - -class BBCCoUkIE(InfoExtractor): - IE_NAME = 'bbc.co.uk' - IE_DESC = 'BBC iPlayer' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})' - - _TESTS = [ - { - 'url': 'http://www.bbc.co.uk/programmes/b039g8p7', - 'info_dict': { - 'id': 'b039d07m', - 'ext': 'flv', - 'title': 'Kaleidoscope, Leonard Cohen', - 'description': 'The Canadian poet and songwriter reflects on his musical career.', - 'duration': 1740, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/', - 'info_dict': { - 'id': 'b00yng1d', - 'ext': 'flv', - 'title': 'The Man in Black: Series 3: The Printed Name', - 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.", - 'duration': 1800, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Episode is no longer available on BBC iPlayer Radio', - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/', - 'info_dict': { - 'id': 'b00yng1d', - 'ext': 'flv', - 'title': 'The Voice UK: Series 3: Blind Auditions 5', - 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.", - 'duration': 5100, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion', - 'info_dict': { - 'id': 'b03k3pb7', - 'ext': 'flv', - 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction", - 'description': '2. Invasion', - 'duration': 3600, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', - }, { - 'url': 'http://www.bbc.co.uk/programmes/b04v20dw', - 'info_dict': { - 'id': 'b04v209v', - 'ext': 'flv', - 'title': 'Pete Tong, The Essential New Tune Special', - 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!", - 'duration': 10800, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3', - 'note': 'Audio', - 'info_dict': { - 'id': 'p02frcch', - 'ext': 'flv', - 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix', - 'description': 'French house superstar Madeon takes us out of the club and onto the after party.', - 'duration': 3507, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz', - 'note': 'Video', - 'info_dict': { - 'id': 'p025c103', - 'ext': 'flv', - 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)', - 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014', - 'duration': 226, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls', - 'info_dict': { - 'id': 'p02n76xf', - 'ext': 'flv', - 'title': 'Natural World, 2015-2016: 2. Super Powered Owls', - 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d', - 'duration': 3540, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'geolocation', - }, { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition', - 'info_dict': { - 'id': 'b05zmgw1', - 'ext': 'flv', - 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.', - 'title': 'Royal Academy Summer Exhibition', - 'duration': 3540, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'geolocation', - }, { - 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', - 'only_matching': True, - }, { - 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3', - 'only_matching': True, - }, { - 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo', - 'only_matching': True, - } - ] - - def _extract_asx_playlist(self, connection, programme_id): - asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist') - return [ref.get('href') for ref in asx.findall('./Entry/ref')] - - def _extract_connection(self, connection, programme_id): - formats = [] - protocol = connection.get('protocol') - supplier = connection.get('supplier') - if protocol == 'http': - href = connection.get('href') - # ASX playlist - if supplier == 'asx': - for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)): - formats.append({ - 'url': ref, - 'format_id': 'ref%s_%s' % (i, supplier), - }) - # Direct link - else: - formats.append({ - 'url': href, - 'format_id': supplier, - }) - elif protocol == 'rtmp': - application = connection.get('application', 'ondemand') - auth_string = connection.get('authString') - identifier = connection.get('identifier') - server = connection.get('server') - formats.append({ - 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string), - 'play_path': identifier, - 'app': '%s?%s' % (application, auth_string), - 'page_url': 'http://www.bbc.co.uk', - 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf', - 'rtmp_live': False, - 'ext': 'flv', - 'format_id': supplier, - }) - return formats - - def _extract_items(self, playlist): - return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item') - - def _extract_medias(self, media_selection): - error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error') - if error is not None: - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True) - return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media') - - def _extract_connections(self, media): - return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection') - - def _extract_video(self, media, programme_id): - formats = [] - vbr = int(media.get('bitrate')) - vcodec = media.get('encoding') - service = media.get('service') - width = int(media.get('width')) - height = int(media.get('height')) - file_size = int(media.get('media_file_size')) - for connection in self._extract_connections(media): - conn_formats = self._extract_connection(connection, programme_id) - for format in conn_formats: - format.update({ - 'format_id': '%s_%s' % (service, format['format_id']), - 'width': width, - 'height': height, - 'vbr': vbr, - 'vcodec': vcodec, - 'filesize': file_size, - }) - formats.extend(conn_formats) - return formats - - def _extract_audio(self, media, programme_id): - formats = [] - abr = int(media.get('bitrate')) - acodec = media.get('encoding') - service = media.get('service') - for connection in self._extract_connections(media): - conn_formats = self._extract_connection(connection, programme_id) - for format in conn_formats: - format.update({ - 'format_id': '%s_%s' % (service, format['format_id']), - 'abr': abr, - 'acodec': acodec, - }) - formats.extend(conn_formats) - return formats - - def _get_subtitles(self, media, programme_id): - subtitles = {} - for connection in self._extract_connections(media): - captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') - lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') - subtitles[lang] = [ - { - 'url': connection.get('href'), - 'ext': 'ttml', - }, - ] - return subtitles - - def _download_media_selector(self, programme_id): - try: - media_selection = self._download_xml( - 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id, - programme_id, 'Downloading media selection XML') - except ExtractorError as ee: - if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: - media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) - else: - raise - - formats = [] - subtitles = None - - for media in self._extract_medias(media_selection): - kind = media.get('kind') - if kind == 'audio': - formats.extend(self._extract_audio(media, programme_id)) - elif kind == 'video': - formats.extend(self._extract_video(media, programme_id)) - elif kind == 'captions': - subtitles = self.extract_subtitles(media, programme_id) - - return formats, subtitles - - def _download_playlist(self, playlist_id): - try: - playlist = self._download_json( - 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id, - playlist_id, 'Downloading playlist JSON') - - version = playlist.get('defaultAvailableVersion') - if version: - smp_config = version['smpConfig'] - title = smp_config['title'] - description = smp_config['summary'] - for item in smp_config['items']: - kind = item['kind'] - if kind != 'programme' and kind != 'radioProgramme': - continue - programme_id = item.get('vpid') - duration = int(item.get('duration')) - formats, subtitles = self._download_media_selector(programme_id) - return programme_id, title, description, duration, formats, subtitles - except ExtractorError as ee: - if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404): - raise - - # fallback to legacy playlist - playlist = self._download_xml( - 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, - playlist_id, 'Downloading legacy playlist XML') - - no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems') - if no_items is not None: - reason = no_items.get('reason') - if reason == 'preAvailability': - msg = 'Episode %s is not yet available' % playlist_id - elif reason == 'postAvailability': - msg = 'Episode %s is no longer available' % playlist_id - elif reason == 'noMedia': - msg = 'Episode %s is not currently available' % playlist_id - else: - msg = 'Episode %s is not available: %s' % (playlist_id, reason) - raise ExtractorError(msg, expected=True) - - for item in self._extract_items(playlist): - kind = item.get('kind') - if kind != 'programme' and kind != 'radioProgramme': - continue - title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text - description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text - programme_id = item.get('identifier') - duration = int(item.get('duration')) - formats, subtitles = self._download_media_selector(programme_id) - - return programme_id, title, description, duration, formats, subtitles - - def _real_extract(self, url): - group_id = self._match_id(url) - - webpage = self._download_webpage(url, group_id, 'Downloading video page') - - programme_id = None - - tviplayer = self._search_regex( - r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById', - webpage, 'player', default=None) - - if tviplayer: - player = self._parse_json(tviplayer, group_id).get('player', {}) - duration = int_or_none(player.get('duration')) - programme_id = player.get('vpid') - - if not programme_id: - programme_id = self._search_regex( - r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) - - if programme_id: - formats, subtitles = self._download_media_selector(programme_id) - title = self._og_search_title(webpage) - description = self._search_regex( - r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', - webpage, 'description', fatal=False) - else: - programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) - - self._sort_formats(formats) - - return { - 'id': programme_id, - 'title': title, - 'description': description, - 'thumbnail': self._og_search_thumbnail(webpage, default=None), - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - } diff --git a/youtube_dl/extractor/beeg.py b/youtube_dl/extractor/beeg.py index b38057f2f..e6c928699 100644 --- a/youtube_dl/extractor/beeg.py +++ b/youtube_dl/extractor/beeg.py @@ -1,65 +1,67 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_iso8601, +) class BeegIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)' _TEST = { 'url': 'http://beeg.com/5416503', - 'md5': '1bff67111adb785c51d1b42959ec10e5', + 'md5': '46c384def73b33dbc581262e5ee67cef', 'info_dict': { 'id': '5416503', 'ext': 'mp4', 'title': 'Sultry Striptease', - 'description': 'md5:6db3c6177972822aaba18652ff59c773', - 'categories': list, # NSFW - 'thumbnail': 're:https?://.*\.jpg$', + 'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2', + 'timestamp': 1391813355, + 'upload_date': '20140207', + 'duration': 383, + 'tags': list, 'age_limit': 18, } } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - webpage = self._download_webpage(url, video_id) - - quality_arr = self._search_regex( - r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats') + video_id = self._match_id(url) - formats = [{ - 'url': fmt[1], - 'format_id': fmt[0], - 'height': int(fmt[0][:-1]), - } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)] + video = self._download_json( + 'http://beeg.com/api/v1/video/%s' % video_id, video_id) + formats = [] + for format_id, video_url in video.items(): + height = self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None) + if not height: + continue + formats.append({ + 'url': self._proto_relative_url(video_url.replace('{DATA_MARKERS}', ''), 'http:'), + 'format_id': format_id, + 'height': int(height), + }) self._sort_formats(formats) - title = self._html_search_regex( - r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title') + title = video['title'] + video_id = video.get('id') or video_id + display_id = video.get('code') + description = video.get('desc') - description = self._html_search_regex( - r'<meta name="description" content="([^"]*)"', - webpage, 'description', fatal=False) - thumbnail = self._html_search_regex( - r'\'previewer.url\'\s*:\s*"([^"]*)"', - webpage, 'thumbnail', fatal=False) + timestamp = parse_iso8601(video.get('date'), ' ') + duration = int_or_none(video.get('duration')) - categories_str = self._html_search_regex( - r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False) - categories = ( - None if categories_str is None - else categories_str.split(',')) + tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None return { 'id': video_id, + 'display_id': display_id, 'title': title, 'description': description, - 'thumbnail': thumbnail, - 'categories': categories, + 'timestamp': timestamp, + 'duration': duration, + 'tags': tags, 'formats': formats, 'age_limit': 18, } diff --git a/youtube_dl/extractor/bet.py b/youtube_dl/extractor/bet.py index 26b934543..03dad4636 100644 --- a/youtube_dl/extractor/bet.py +++ b/youtube_dl/extractor/bet.py @@ -1,7 +1,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_unquote from ..utils import ( xpath_text, xpath_with_ns, @@ -57,7 +57,7 @@ class BetIE(InfoExtractor): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - media_url = compat_urllib_parse.unquote(self._search_regex( + media_url = compat_urllib_parse_unquote(self._search_regex( [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"], webpage, 'media URL')) diff --git a/youtube_dl/extractor/bild.py b/youtube_dl/extractor/bild.py index 4d8cce1ef..1a0184861 100644 --- a/youtube_dl/extractor/bild.py +++ b/youtube_dl/extractor/bild.py @@ -4,7 +4,7 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( int_or_none, - fix_xml_ampersands, + unescapeHTML, ) @@ -17,26 +17,24 @@ class BildIE(InfoExtractor): 'info_dict': { 'id': '38184146', 'ext': 'mp4', - 'title': 'BILD hat sie getestet', + 'title': 'Das können die neuen iPads', + 'description': 'md5:a4058c4fa2a804ab59c00d7244bbf62f', 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 196, - 'description': 'Mit dem iPad Air 2 und dem iPad Mini 3 hat Apple zwei neue Tablet-Modelle präsentiert. BILD-Reporter Sven Stein durfte die Geräte bereits testen. ', } } def _real_extract(self, url): video_id = self._match_id(url) - xml_url = url.split(".bild.html")[0] + ",view=xml.bild.xml" - doc = self._download_xml(xml_url, video_id, transform_source=fix_xml_ampersands) - - duration = int_or_none(doc.attrib.get('duration'), scale=1000) + video_data = self._download_json( + url.split('.bild.html')[0] + ',view=json.bild.html', video_id) return { 'id': video_id, - 'title': doc.attrib['ueberschrift'], - 'description': doc.attrib.get('text'), - 'url': doc.attrib['src'], - 'thumbnail': doc.attrib.get('img'), - 'duration': duration, + 'title': unescapeHTML(video_data['title']).strip(), + 'description': unescapeHTML(video_data.get('description')), + 'url': video_data['clipList'][0]['srces'][0]['src'], + 'thumbnail': video_data.get('poster'), + 'duration': int_or_none(video_data.get('durationSec')), } diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index bf60450c2..ecc17ebeb 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -41,8 +41,15 @@ class BiliBiliIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - if self._search_regex(r'(此视频不存在或被删除)', webpage, 'error message', default=None): - raise ExtractorError('The video does not exist or was deleted', expected=True) + if '(此视频不存在或被删除)' in webpage: + raise ExtractorError( + 'The video does not exist or was deleted', expected=True) + + if '>你没有权限浏览！ 由于版权相关问题 我们不对您所在的地区提供服务<' in webpage: + raise ExtractorError( + 'The video is not available in your region due to copyright reasons', + expected=True) + video_code = self._search_regex( r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code') diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py index fb56cd78d..c3296283d 100644 --- a/youtube_dl/extractor/bliptv.py +++ b/youtube_dl/extractor/bliptv.py @@ -5,7 +5,6 @@ import re from .common import InfoExtractor from ..compat import ( - compat_str, compat_urllib_request, compat_urlparse, ) @@ -14,6 +13,8 @@ from ..utils import ( int_or_none, parse_iso8601, unescapeHTML, + xpath_text, + xpath_with_ns, ) @@ -23,10 +24,10 @@ class BlipTVIE(InfoExtractor): _TESTS = [ { 'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352', - 'md5': 'c6934ad0b6acf2bd920720ec888eb812', + 'md5': '80baf1ec5c3d2019037c1c707d676b9f', 'info_dict': { 'id': '5779306', - 'ext': 'mov', + 'ext': 'm4v', 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3', 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596', 'timestamp': 1323138843, @@ -100,6 +101,20 @@ class BlipTVIE(InfoExtractor): 'vcodec': 'none', } }, + { + # missing duration + 'url': 'http://blip.tv/rss/flash/6700880', + 'info_dict': { + 'id': '6684191', + 'ext': 'm4v', + 'title': 'Cowboy Bebop: Gateway Shuffle Review', + 'description': 'md5:3acc480c0f9ae157f5fe88547ecaf3f8', + 'timestamp': 1386639757, + 'upload_date': '20131210', + 'uploader': 'sfdebris', + 'uploader_id': '706520', + } + } ] @staticmethod @@ -128,35 +143,34 @@ class BlipTVIE(InfoExtractor): rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS') - def blip(s): - return '{http://blip.tv/dtd/blip/1.0}%s' % s - - def media(s): - return '{http://search.yahoo.com/mrss/}%s' % s - - def itunes(s): - return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s + def _x(p): + return xpath_with_ns(p, { + 'blip': 'http://blip.tv/dtd/blip/1.0', + 'media': 'http://search.yahoo.com/mrss/', + 'itunes': 'http://www.itunes.com/dtds/podcast-1.0.dtd', + }) item = rss.find('channel/item') - video_id = item.find(blip('item_id')).text - title = item.find('./title').text - description = clean_html(compat_str(item.find(blip('puredescription')).text)) - timestamp = parse_iso8601(item.find(blip('datestamp')).text) - uploader = item.find(blip('user')).text - uploader_id = item.find(blip('userid')).text - duration = int(item.find(blip('runtime')).text) - media_thumbnail = item.find(media('thumbnail')) - thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text - categories = [category.text for category in item.findall('category')] + video_id = xpath_text(item, _x('blip:item_id'), 'video id') or lookup_id + title = xpath_text(item, 'title', 'title', fatal=True) + description = clean_html(xpath_text(item, _x('blip:puredescription'), 'description')) + timestamp = parse_iso8601(xpath_text(item, _x('blip:datestamp'), 'timestamp')) + uploader = xpath_text(item, _x('blip:user'), 'uploader') + uploader_id = xpath_text(item, _x('blip:userid'), 'uploader id') + duration = int_or_none(xpath_text(item, _x('blip:runtime'), 'duration')) + media_thumbnail = item.find(_x('media:thumbnail')) + thumbnail = (media_thumbnail.get('url') if media_thumbnail is not None + else xpath_text(item, 'image', 'thumbnail')) + categories = [category.text for category in item.findall('category') if category is not None] formats = [] subtitles_urls = {} - media_group = item.find(media('group')) - for media_content in media_group.findall(media('content')): + media_group = item.find(_x('media:group')) + for media_content in media_group.findall(_x('media:content')): url = media_content.get('url') - role = media_content.get(blip('role')) + role = media_content.get(_x('blip:role')) msg = self._download_webpage( url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url', video_id, 'Resolving URL for %s' % role) @@ -175,8 +189,8 @@ class BlipTVIE(InfoExtractor): 'url': real_url, 'format_id': role, 'format_note': media_type, - 'vcodec': media_content.get(blip('vcodec')) or 'none', - 'acodec': media_content.get(blip('acodec')), + 'vcodec': media_content.get(_x('blip:vcodec')) or 'none', + 'acodec': media_content.get(_x('blip:acodec')), 'filesize': media_content.get('filesize'), 'width': int_or_none(media_content.get('width')), 'height': int_or_none(media_content.get('height')), diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 809287d14..aa08051b1 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -18,6 +18,7 @@ class BreakIE(InfoExtractor): 'id': '2468056', 'ext': 'mp4', 'title': 'When Girls Act Like D-Bags', + 'age_limit': 13, } }, { 'url': 'http://www.break.com/video/ugc/baby-flex-2773063', diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py index c4fefefe4..f6a1ff381 100644 --- a/youtube_dl/extractor/canalc2.py +++ b/youtube_dl/extractor/canalc2.py @@ -4,38 +4,53 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import parse_duration class Canalc2IE(InfoExtractor): IE_NAME = 'canalc2.tv' - _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)' + _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)' _TEST = { - 'url': 'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui', + 'url': 'http://www.canalc2.tv/video/12163', 'md5': '060158428b650f896c542dfbb3d6487f', 'info_dict': { 'id': '12163', - 'ext': 'mp4', - 'title': 'Terrasses du Numérique' + 'ext': 'flv', + 'title': 'Terrasses du Numérique', + 'duration': 122, + }, + 'params': { + 'skip_download': True, # Requires rtmpdump } } def _real_extract(self, url): - video_id = re.match(self._VALID_URL, url).group('id') - # We need to set the voir field for getting the file name - url = 'http://www.canalc2.tv/video.asp?idVideo=%s&voir=oui' % video_id + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - file_name = self._search_regex( - r"so\.addVariable\('file','(.*?)'\);", - webpage, 'file name') - video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name + video_url = self._search_regex( + r'jwplayer\((["\'])Player\1\)\.setup\({[^}]*file\s*:\s*(["\'])(?P<file>.+?)\2', + webpage, 'video_url', group='file') + formats = [{'url': video_url}] + if video_url.startswith('rtmp://'): + rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url) + formats[0].update({ + 'url': rtmp.group('url'), + 'ext': 'flv', + 'app': rtmp.group('app'), + 'play_path': rtmp.group('play_path'), + 'page_url': url, + }) title = self._html_search_regex( - r'class="evenement8">(.*?)</a>', webpage, 'title') + r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.*?)</h3>', webpage, 'title') + duration = parse_duration(self._search_regex( + r'id=["\']video_duree["\'][^>]*>([^<]+)', + webpage, 'duration', fatal=False)) return { 'id': video_id, - 'ext': 'mp4', - 'url': video_url, 'title': title, + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 699b4f7d0..004372f8d 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -78,7 +78,8 @@ class CanalplusIE(InfoExtractor): if video_id is None: webpage = self._download_webpage(url, display_id) video_id = self._search_regex( - r'<canal:player[^>]+?videoId="(\d+)"', webpage, 'video id') + [r'<canal:player[^>]+?videoId=(["\'])(?P<id>\d+)', r'id=["\']canal_video_player(?P<id>\d+)'], + webpage, 'video id', group='id') info_url = self._VIDEO_INFO_TEMPLATE % (site_id, video_id) doc = self._download_xml(info_url, video_id, 'Downloading video XML') @@ -106,15 +107,11 @@ class CanalplusIE(InfoExtractor): continue format_id = fmt.tag if format_id == 'HLS': - hls_formats = self._extract_m3u8_formats(format_url, video_id, 'flv') - for fmt in hls_formats: - fmt['preference'] = preference(format_id) - formats.extend(hls_formats) + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', preference=preference(format_id))) elif format_id == 'HDS': - hds_formats = self._extract_f4m_formats(format_url + '?hdcore=2.11.3', video_id) - for fmt in hds_formats: - fmt['preference'] = preference(format_id) - formats.extend(hds_formats) + formats.extend(self._extract_f4m_formats( + format_url + '?hdcore=2.11.3', video_id, preference=preference(format_id))) else: formats.append({ 'url': format_url, diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py index 65f6be623..e857e66f4 100644 --- a/youtube_dl/extractor/ceskatelevize.py +++ b/youtube_dl/extractor/ceskatelevize.py @@ -7,6 +7,7 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_request, compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_parse_urlparse, ) from ..utils import ( @@ -16,55 +17,81 @@ from ..utils import ( class CeskaTelevizeIE(InfoExtractor): - _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)' - - _TESTS = [ - { - 'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220', + _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(?:[^/]+/)*(?P<id>[^/#?]+)/*(?:[#?].*)?$' + _TESTS = [{ + 'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220', + 'info_dict': { + 'id': '61924494876951776', + 'ext': 'mp4', + 'title': 'Hyde Park Civilizace', + 'description': 'md5:fe93f6eda372d150759d11644ebbfb4a', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 3350, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + 'info_dict': { + 'id': '61924494876844374', + 'ext': 'mp4', + 'title': 'První republika: Zpěvačka z Dupárny Bobina', + 'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 88.4, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + # video with 18+ caution trailer + 'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/', + 'info_dict': { + 'id': '215562210900007-bogotart', + 'title': 'Queer: Bogotart', + 'description': 'Alternativní průvodce současným queer světem', + }, + 'playlist': [{ 'info_dict': { - 'id': '214411058091220', + 'id': '61924494876844842', 'ext': 'mp4', - 'title': 'Hyde Park Civilizace', - 'description': 'Věda a současná civilizace. Interaktivní pořad - prostor pro vaše otázky a komentáře', - 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 3350, - }, - 'params': { - # m3u8 download - 'skip_download': True, + 'title': 'Queer: Bogotart (Varování 18+)', + 'duration': 10.2, }, - }, - { - 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina', + }, { 'info_dict': { - 'id': '14716', + 'id': '61924494877068022', 'ext': 'mp4', - 'title': 'První republika: Zpěvačka z Dupárny Bobina', - 'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.', + 'title': 'Queer: Bogotart (Queer)', 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 88.4, - }, - 'params': { - # m3u8 download - 'skip_download': True, + 'duration': 1558.3, }, + }], + 'params': { + # m3u8 download + 'skip_download': True, }, - ] + }] def _real_extract(self, url): url = url.replace('/porady/', '/ivysilani/').replace('/video/', '') mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + playlist_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, playlist_id) NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.' if '%s</p>' % NOT_AVAILABLE_STRING in webpage: raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type') - episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id') + typ = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type') + episode_id = self._html_search_regex( + r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id') data = { 'playlist[0][type]': typ, @@ -82,42 +109,52 @@ class CeskaTelevizeIE(InfoExtractor): req.add_header('X-Requested-With', 'XMLHttpRequest') req.add_header('Referer', url) - playlistpage = self._download_json(req, video_id) + playlistpage = self._download_json(req, playlist_id) playlist_url = playlistpage['url'] if playlist_url == 'error_region': raise ExtractorError(NOT_AVAILABLE_STRING, expected=True) - req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlist_url)) + req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url)) req.add_header('Referer', url) - playlist = self._download_json(req, video_id) - - item = playlist['playlist'][0] - formats = [] - for format_id, stream_url in item['streamUrls'].items(): - formats.extend(self._extract_m3u8_formats(stream_url, video_id, 'mp4')) - self._sort_formats(formats) - - title = self._og_search_title(webpage) - description = self._og_search_description(webpage) - duration = float_or_none(item.get('duration')) - thumbnail = item.get('previewImageUrl') - - subtitles = {} - subs = item.get('subtitles') - if subs: - subtitles = self.extract_subtitles(episode_id, subs) - - return { - 'id': episode_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - } + playlist_title = self._og_search_title(webpage) + playlist_description = self._og_search_description(webpage) + + playlist = self._download_json(req, playlist_id)['playlist'] + playlist_len = len(playlist) + + entries = [] + for item in playlist: + 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')) + self._sort_formats(formats) + + item_id = item.get('id') or item['assetId'] + title = item['title'] + + duration = float_or_none(item.get('duration')) + thumbnail = item.get('previewImageUrl') + + subtitles = {} + if item.get('type') == 'VOD': + subs = item.get('subtitles') + if subs: + subtitles = self.extract_subtitles(episode_id, subs) + + entries.append({ + 'id': item_id, + 'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title), + 'description': playlist_description if playlist_len == 1 else None, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + }) + + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) def _get_subtitles(self, episode_id, subs): original_subtitles = self._download_webpage( diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py index 3dfc24f5b..c74553dcf 100644 --- a/youtube_dl/extractor/channel9.py +++ b/youtube_dl/extractor/channel9.py @@ -3,7 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + parse_filesize, + qualities, +) class Channel9IE(InfoExtractor): @@ -28,7 +32,7 @@ class Channel9IE(InfoExtractor): 'title': 'Developer Kick-Off Session: Stuff We Love', 'description': 'md5:c08d72240b7c87fcecafe2692f80e35f', 'duration': 4576, - 'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg', + 'thumbnail': 're:http://.*\.jpg', 'session_code': 'KOS002', 'session_day': 'Day 1', 'session_room': 'Arena 1A', @@ -44,31 +48,29 @@ class Channel9IE(InfoExtractor): 'title': 'Self-service BI with Power BI - nuclear testing', 'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', 'duration': 1540, - 'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg', + '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, + }, } ] _RSS_URL = 'http://channel9.msdn.com/%s/RSS' - # Sorted by quality - _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4'] - - def _restore_bytes(self, formatted_size): - if not formatted_size: - return 0 - m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size) - if not m: - return 0 - units = m.group('units') - try: - exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper()) - except ValueError: - return 0 - size = float(m.group('size')) - return int(size * (1024 ** exponent)) - def _formats_from_html(self, html): FORMAT_REGEX = r''' (?x) @@ -78,16 +80,20 @@ class Channel9IE(InfoExtractor): <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s* </div>)? # File size part may be missing ''' - # Extract known formats + quality = qualities(( + 'MP3', 'MP4', + 'Low Quality WMV', 'Low Quality MP4', + 'Mid Quality WMV', 'Mid Quality MP4', + 'High Quality WMV', 'High Quality MP4')) formats = [{ 'url': x.group('url'), 'format_id': x.group('quality'), 'format_note': x.group('note'), 'format': '%s (%s)' % (x.group('quality'), x.group('note')), - 'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate - 'preference': self._known_formats.index(x.group('quality')), + 'filesize_approx': parse_filesize(x.group('filesize')), + 'quality': quality(x.group('quality')), 'vcodec': 'none' if x.group('note') == 'Audio only' else None, - } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats] + } for x in list(re.finditer(FORMAT_REGEX, html))] self._sort_formats(formats) @@ -158,7 +164,7 @@ class Channel9IE(InfoExtractor): def _extract_session_day(self, html): m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html) - return m.group('day') if m is not None else None + return m.group('day').strip() if m is not None else None def _extract_session_room(self, html): m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html) @@ -224,12 +230,12 @@ class Channel9IE(InfoExtractor): if contents is None: return contents - authors = self._extract_authors(html) + if len(contents) > 1: + raise ExtractorError('Got more than one entry') + result = contents[0] + result['authors'] = self._extract_authors(html) - for content in contents: - content['authors'] = authors - - return contents + return result def _extract_session(self, html, content_path): contents = self._extract_content(html, content_path) diff --git a/youtube_dl/extractor/chaturbate.py b/youtube_dl/extractor/chaturbate.py new file mode 100644 index 000000000..0b67ba67d --- /dev/null +++ b/youtube_dl/extractor/chaturbate.py @@ -0,0 +1,50 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ExtractorError + + +class ChaturbateIE(InfoExtractor): + _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?P<id>[^/?#]+)' + _TESTS = [{ + 'url': 'https://www.chaturbate.com/siswet19/', + 'info_dict': { + 'id': 'siswet19', + 'ext': 'mp4', + 'title': 're:^siswet19 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'age_limit': 18, + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + }, { + 'url': 'https://en.chaturbate.com/siswet19/', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + m3u8_url = self._search_regex( + r'src=(["\'])(?P<url>http.+?\.m3u8.*?)\1', webpage, + 'playlist', default=None, group='url') + + if not m3u8_url: + error = self._search_regex( + r'<span[^>]+class=(["\'])desc_span\1[^>]*>(?P<error>[^<]+)</span>', + webpage, 'error', group='error') + raise ExtractorError(error, expected=True) + + formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') + + return { + 'id': video_id, + 'title': self._live_title(video_id), + 'thumbnail': 'https://cdn-s.highwebmedia.com/uHK3McUtGCG3SMFcd4ZJsRv8/roomimage/%s.jpg' % video_id, + 'age_limit': self._rta_search(webpage), + 'is_live': True, + 'formats': formats, + } diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py index c949a4814..fd1770dac 100644 --- a/youtube_dl/extractor/cinemassacre.py +++ b/youtube_dl/extractor/cinemassacre.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..utils import ExtractorError from .bliptv import BlipTVIE +from .screenwavemedia import ScreenwaveMediaIE class CinemassacreIE(InfoExtractor): @@ -83,10 +84,10 @@ class CinemassacreIE(InfoExtractor): playerdata_url = self._search_regex( [ - r'src="(http://(?:player2\.screenwavemedia\.com|player\.screenwavemedia\.com/play)/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', - r'<iframe[^>]+src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"', + ScreenwaveMediaIE.EMBED_PATTERN, + r'<iframe[^>]+src="(?P<url>(?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"', ], - webpage, 'player data URL', default=None) + webpage, 'player data URL', default=None, group='url') if not playerdata_url: playerdata_url = BlipTVIE._extract_url(webpage) if not playerdata_url: diff --git a/youtube_dl/extractor/clipfish.py b/youtube_dl/extractor/clipfish.py index a5c3cb7c6..7af903571 100644 --- a/youtube_dl/extractor/clipfish.py +++ b/youtube_dl/extractor/clipfish.py @@ -1,53 +1,68 @@ from __future__ import unicode_literals import re -import time -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( - ExtractorError, - parse_duration, + determine_ext, + int_or_none, + js_to_json, + parse_iso8601, + remove_end, ) class ClipfishIE(InfoExtractor): - IE_NAME = 'clipfish' - - _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?clipfish\.de/(?:[^/]+/)+video/(?P<id>[0-9]+)' _TEST = { 'url': 'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/', - 'md5': '2521cd644e862936cf2e698206e47385', + 'md5': '79bc922f3e8a9097b3d68a93780fd475', 'info_dict': { 'id': '3966754', 'ext': 'mp4', 'title': 'FIFA 14 - E3 2013 Trailer', + 'timestamp': 1370938118, + 'upload_date': '20130611', 'duration': 82, - }, - 'skip': 'Blocked in the US' + } } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) - - info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' % - (video_id, int(time.time()))) - doc = self._download_xml( - info_url, video_id, note='Downloading info page') - title = doc.find('title').text - video_url = doc.find('filename').text - if video_url is None: - xml_bytes = xml.etree.ElementTree.tostring(doc) - raise ExtractorError('Cannot find video URL in document %r' % - xml_bytes) - thumbnail = doc.find('imageurl').text - duration = parse_duration(doc.find('duration').text) + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + video_info = self._parse_json( + js_to_json(self._html_search_regex( + '(?s)videoObject\s*=\s*({.+?});', webpage, 'video object')), + video_id) + + formats = [] + for video_url in re.findall(r'var\s+videourl\s*=\s*"([^"]+)"', webpage): + ext = determine_ext(video_url) + if ext == 'm3u8': + formats.append({ + 'url': video_url.replace('de.hls.fra.clipfish.de', 'hls.fra.clipfish.de'), + 'ext': 'mp4', + 'format_id': 'hls', + }) + else: + formats.append({ + 'url': video_url, + 'format_id': ext, + }) + self._sort_formats(formats) + + title = remove_end(self._og_search_title(webpage), ' - Video') + thumbnail = self._og_search_thumbnail(webpage) + duration = int_or_none(video_info.get('length')) + timestamp = parse_iso8601(self._html_search_meta('uploadDate', webpage, 'upload date')) return { 'id': video_id, 'title': title, - 'url': video_url, + 'formats': formats, 'thumbnail': thumbnail, 'duration': duration, + 'timestamp': timestamp, } diff --git a/youtube_dl/extractor/clubic.py b/youtube_dl/extractor/clubic.py index 14f215c5c..1dfa7c12e 100644 --- a/youtube_dl/extractor/clubic.py +++ b/youtube_dl/extractor/clubic.py @@ -12,9 +12,9 @@ from ..utils import ( class ClubicIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?clubic\.com/video/[^/]+/video.*-(?P<id>[0-9]+)\.html' + _VALID_URL = r'http://(?:www\.)?clubic\.com/video/(?:[^/]+/)*video.*-(?P<id>[0-9]+)\.html' - _TEST = { + _TESTS = [{ 'url': 'http://www.clubic.com/video/clubic-week/video-clubic-week-2-0-le-fbi-se-lance-dans-la-photo-d-identite-448474.html', 'md5': '1592b694ba586036efac1776b0b43cd3', 'info_dict': { @@ -24,7 +24,10 @@ class ClubicIE(InfoExtractor): 'description': 're:Gueule de bois chez Nokia. Le constructeur a indiqué cette.*', 'thumbnail': 're:^http://img\.clubic\.com/.*\.jpg$', } - } + }, { + 'url': 'http://www.clubic.com/video/video-clubic-week-2-0-apple-iphone-6s-et-plus-mais-surtout-le-pencil-469792.html', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) diff --git a/youtube_dl/extractor/clyp.py b/youtube_dl/extractor/clyp.py new file mode 100644 index 000000000..57e643799 --- /dev/null +++ b/youtube_dl/extractor/clyp.py @@ -0,0 +1,57 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + float_or_none, + parse_iso8601, +) + + +class ClypIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?clyp\.it/(?P<id>[a-z0-9]+)' + _TEST = { + 'url': 'https://clyp.it/ojz2wfah', + 'md5': '1d4961036c41247ecfdcc439c0cddcbb', + 'info_dict': { + 'id': 'ojz2wfah', + 'ext': 'mp3', + 'title': 'Krisson80 - bits wip wip', + 'description': '#Krisson80BitsWipWip #chiptune\n#wip', + 'duration': 263.21, + 'timestamp': 1443515251, + 'upload_date': '20150929', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + metadata = self._download_json( + 'https://api.clyp.it/%s' % audio_id, audio_id) + + formats = [] + for secure in ('', 'Secure'): + for ext in ('Ogg', 'Mp3'): + format_id = '%s%s' % (secure, ext) + format_url = metadata.get('%sUrl' % format_id) + if format_url: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'vcodec': 'none', + }) + self._sort_formats(formats) + + title = metadata['Title'] + description = metadata.get('Description') + duration = float_or_none(metadata.get('Duration')) + timestamp = parse_iso8601(metadata.get('DateCreated')) + + return { + 'id': audio_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/comcarcoff.py b/youtube_dl/extractor/comcarcoff.py index 9c25b2223..81f3d7697 100644 --- a/youtube_dl/extractor/comcarcoff.py +++ b/youtube_dl/extractor/comcarcoff.py @@ -36,7 +36,7 @@ class ComCarCoffIE(InfoExtractor): webpage, 'full data json')) video_id = full_data['activeVideo']['video'] - video_data = full_data['videos'][video_id] + video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id] thumbnails = [{ 'url': video_data['images']['thumb'], }, { diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index 91ebb0ce5..3e4bd10b6 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -151,12 +151,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor): mobj = re.match(self._VALID_URL, url) if mobj.group('shortname'): - if mobj.group('shortname') in ('tds', 'thedailyshow'): - url = 'http://thedailyshow.cc.com/full-episodes/' - else: - url = 'http://thecolbertreport.cc.com/full-episodes/' - mobj = re.match(self._VALID_URL, url, re.VERBOSE) - assert mobj is not None + return self.url_result('http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes') if mobj.group('clip'): if mobj.group('videotitle'): diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 82f5de2d8..10c0d5d1f 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -14,10 +14,13 @@ import xml.etree.ElementTree from ..compat import ( compat_cookiejar, - compat_HTTPError, + compat_cookies, + compat_getpass, compat_http_client, compat_urllib_error, + compat_urllib_parse, compat_urllib_parse_urlparse, + compat_urllib_request, compat_urlparse, compat_str, ) @@ -27,12 +30,18 @@ from ..utils import ( bug_reports_message, clean_html, compiled_regex_type, + determine_ext, ExtractorError, + fix_xml_ampersands, float_or_none, int_or_none, RegexNotFoundError, sanitize_filename, unescapeHTML, + unified_strdate, + url_basename, + xpath_text, + xpath_with_ns, ) @@ -63,7 +72,7 @@ class InfoExtractor(object): Potential fields: * url Mandatory. The URL of the video file - * ext Will be calculated from url if missing + * ext Will be calculated from URL if missing * format A human-readable description of the format ("mp4 container with h264/opus"). Calculated from the format_id, width, height. @@ -143,6 +152,7 @@ class InfoExtractor(object): description: Full video description. uploader: Full name of the video uploader. creator: The main artist who created 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). If not explicitly set, calculated from timestamp. @@ -153,13 +163,15 @@ class InfoExtractor(object): lower to higher preference, each element is a dictionary with the "ext" entry and one of: * "data": The subtitles file contents - * "url": A url pointing to the subtitles file + * "url": A URL pointing to the subtitles file + "ext" will be calculated from URL if missing automatic_captions: Like 'subtitles', used by the YoutubeIE for automatically generated captions duration: Length of the video in seconds, as an integer. view_count: How many users have watched the video on the platform. like_count: Number of positive ratings of the video dislike_count: Number of negative ratings of the video + repost_count: Number of reposts of the video average_rating: Average rating give by users, the scale used depends on the webpage comment_count: Number of comments on the video comments: A list of comments, each with one or more of the following @@ -174,13 +186,18 @@ class InfoExtractor(object): Set to "root" to indicate that this is a comment to the original video. age_limit: Age restriction for the video, as an integer (years) - webpage_url: The url to the video webpage, if given to youtube-dl it + 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) categories: A list of categories that the video falls in, for example ["Sports", "Berlin"] + tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] is_live: True, False, or None (=unknown). Whether this video is a live stream that goes on instead of a fixed-length video. + start_time: Time in seconds where the reproduction should start, as + specified in the URL. + end_time: Time in seconds where the reproduction should end, as + specified in the URL. Unless mentioned otherwise, the fields should be Unicode strings. @@ -191,8 +208,8 @@ class InfoExtractor(object): There must be a key "entries", which is a list, an iterable, or a PagedList object, each element of which is a valid dictionary by this specification. - Additionally, playlists can have "title" and "id" attributes with the same - semantics as videos (see above). + Additionally, playlists can have "title", "description" and "id" attributes + with the same semantics as videos (see above). _type "multi_video" indicates that there are multiple videos that @@ -496,10 +513,22 @@ class InfoExtractor(object): """Report attempt to log in.""" self.to_screen('Logging in') + @staticmethod + def raise_login_required(msg='This video is only available for registered users'): + raise ExtractorError( + '%s. Use --username and --password or --netrc to provide account credentials.' % msg, + expected=True) + + @staticmethod + def raise_geo_restricted(msg='This video is not available from your location due to geo restriction'): + raise ExtractorError( + '%s. You might want to use --proxy to workaround.' % msg, + expected=True) + # Methods for following #608 @staticmethod def url_result(url, ie=None, video_id=None, video_title=None): - """Returns a url that points to a page that should be processed""" + """Returns a URL that points to a page that should be processed""" # TODO: ie should be the class used for getting the info video_info = {'_type': 'url', 'url': url, @@ -597,7 +626,7 @@ class InfoExtractor(object): return (username, password) - def _get_tfa_info(self): + def _get_tfa_info(self, note='two-factor verification code'): """ Get the two-factor authentication info TODO - asking the user will be required for sms/phone verify @@ -611,19 +640,26 @@ class InfoExtractor(object): if downloader_params.get('twofactor', None) is not None: return downloader_params['twofactor'] - return None + return compat_getpass('Type %s and press [Return]: ' % note) # Helper functions for extracting OpenGraph info @staticmethod def _og_regexes(prop): - content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' - property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) + content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' + property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' + % {'prop': re.escape(prop)}) template = r'<meta[^>]+?%s[^>]+?%s' return [ template % (property_re, content_re), template % (content_re, property_re), ] + @staticmethod + def _meta_regex(prop): + return r'''(?isx)<meta + (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) + [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) + def _og_search_property(self, prop, html, name=None, **kargs): if name is None: name = 'OpenGraph %s' % prop @@ -633,7 +669,7 @@ class InfoExtractor(object): return unescapeHTML(escaped) def _og_search_thumbnail(self, html, **kargs): - return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs) + return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) def _og_search_description(self, html, **kargs): return self._og_search_property('description', html, fatal=False, **kargs) @@ -654,9 +690,7 @@ class InfoExtractor(object): if display_name is None: display_name = name return self._html_search_regex( - r'''(?isx)<meta - (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1) - [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(name), + self._meta_regex(name), html, display_name, fatal=fatal, group='content', **kwargs) def _dc_search_uploader(self, html): @@ -706,10 +740,26 @@ class InfoExtractor(object): 'twitter card player') @staticmethod - def _form_hidden_inputs(html): - return dict(re.findall( - r'<input\s+type="hidden"\s+name="([^"]+)"\s+(?:id="[^"]+"\s+)?value="([^"]*)"', - html)) + def _hidden_inputs(html): + html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) + hidden_inputs = {} + for input in re.findall(r'(?i)<input([^>]+)>', html): + if not re.search(r'type=(["\'])(?:hidden|submit)\1', input): + continue + name = re.search(r'name=(["\'])(?P<value>.+?)\1', input) + if not name: + continue + value = re.search(r'value=(["\'])(?P<value>.*?)\1', input) + if not value: + continue + hidden_inputs[name.group('value')] = value.group('value') + return hidden_inputs + + def _form_hidden_inputs(self, form_id, html): + form = self._search_regex( + r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id, + html, '%s form' % form_id, group='form') + return self._hidden_inputs(form) def _sort_formats(self, formats, field_preference=None): if not formats: @@ -791,7 +841,7 @@ class InfoExtractor(object): self._request_webpage(url, video_id, 'Checking %s URL' % item) return True except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError): + if isinstance(e.cause, compat_urllib_error.URLError): self.to_screen( '%s: %s URL is invalid, skipping' % (video_id, item)) return False @@ -821,10 +871,19 @@ class InfoExtractor(object): self.to_screen(msg) time.sleep(timeout) - def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None): + 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): manifest = self._download_xml( manifest_url, video_id, 'Downloading f4m manifest', - 'Unable to download f4m manifest') + 'Unable to download f4m manifest', + # Some manifests may be malformed, e.g. prosiebensat1 generated manifests + # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) + transform_source=transform_source, + fatal=fatal) + + if manifest is False: + return manifest formats = [] manifest_version = '1.0' @@ -834,8 +893,22 @@ class InfoExtractor(object): media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') for i, media_el in enumerate(media_nodes): if manifest_version == '2.0': - manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/' + - (media_el.attrib.get('href') or media_el.attrib.get('url'))) + media_url = media_el.attrib.get('href') or media_el.attrib.get('url') + if not media_url: + continue + manifest_url = ( + media_url if media_url.startswith('http://') or media_url.startswith('https://') + else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) + # If media_url is itself a f4m manifest do the recursive extraction + # 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': + f4m_formats = self._extract_f4m_formats( + manifest_url, video_id, preference, f4m_id, fatal=fatal) + if f4m_formats: + formats.extend(f4m_formats) + 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)])), @@ -870,13 +943,14 @@ class InfoExtractor(object): if re.match(r'^https?://', u) else compat_urlparse.urljoin(m3u8_url, u)) - m3u8_doc = self._download_webpage( + m3u8_doc, urlh = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) if m3u8_doc is False: return m3u8_doc + m3u8_url = urlh.geturl() last_info = None last_media = None kv_rex = re.compile( @@ -937,69 +1011,237 @@ class InfoExtractor(object): self._sort_formats(formats) return formats - # TODO: improve extraction - def _extract_smil_formats(self, smil_url, video_id, fatal=True): - smil = self._download_xml( - smil_url, video_id, 'Downloading SMIL file', - 'Unable to download SMIL file', fatal=fatal) + @staticmethod + def _xpath_ns(path, namespace=None): + if not namespace: + return path + out = [] + for c in path.split('/'): + if not c or c == '.': + out.append(c) + else: + out.append('{%s}%s' % (namespace, c)) + return '/'.join(out) + + def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None): + smil = self._download_smil(smil_url, video_id, fatal=fatal) + if smil is False: assert not fatal return [] - base = smil.find('./head/meta').get('base') + namespace = self._parse_smil_namespace(smil) + + return self._parse_smil_formats( + smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) + + def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): + smil = self._download_smil(smil_url, video_id, fatal=fatal) + if smil is False: + return {} + return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) + + def _download_smil(self, smil_url, video_id, fatal=True): + return self._download_xml( + smil_url, video_id, 'Downloading SMIL file', + 'Unable to download SMIL file', fatal=fatal) + + def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): + namespace = self._parse_smil_namespace(smil) + + formats = self._parse_smil_formats( + smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) + subtitles = self._parse_smil_subtitles(smil, namespace=namespace) + + video_id = os.path.splitext(url_basename(smil_url))[0] + title = None + description = None + upload_date = None + for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): + name = meta.attrib.get('name') + content = meta.attrib.get('content') + if not name or not content: + continue + if not title and name == 'title': + title = content + elif not description and name in ('description', 'abstract'): + description = content + elif not upload_date and name == 'date': + upload_date = unified_strdate(content) + + thumbnails = [{ + 'id': image.get('type'), + 'url': image.get('src'), + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] + + return { + 'id': video_id, + 'title': title or video_id, + 'description': description, + 'upload_date': upload_date, + 'thumbnails': thumbnails, + 'formats': formats, + 'subtitles': subtitles, + } + + def _parse_smil_namespace(self, smil): + return self._search_regex( + r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) + + def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): + base = smil_url + for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): + b = meta.get('base') or meta.get('httpBase') + if b: + base = b + break formats = [] rtmp_count = 0 - if smil.findall('./body/seq/video'): - video = smil.findall('./body/seq/video')[0] - fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) - formats.extend(fmts) - else: - for video in smil.findall('./body/switch/video'): - fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) - formats.extend(fmts) + http_count = 0 + + videos = smil.findall(self._xpath_ns('.//video', namespace)) + for video in videos: + src = video.get('src') + if not src: + continue + + 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') + src_ext = determine_ext(src) + streamer = video.get('streamer') or base + + if proto == 'rtmp' or streamer.startswith('rtmp'): + rtmp_count += 1 + formats.append({ + 'url': streamer, + 'play_path': src, + 'ext': 'flv', + 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), + 'tbr': bitrate, + 'filesize': filesize, + 'width': width, + 'height': height, + }) + if transform_rtmp_url: + streamer, src = transform_rtmp_url(streamer, src) + formats[-1].update({ + 'url': streamer, + 'play_path': src, + }) + continue + + src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) + + if proto == 'm3u8' or src_ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + continue + + if src_ext == 'f4m': + f4m_url = src_url + if not f4m_params: + f4m_params = { + 'hdcore': '3.2.0', + 'plugin': 'flowplayer-3.2.0.1', + } + f4m_url += '&' if '?' in f4m_url else '?' + f4m_url += compat_urllib_parse.urlencode(f4m_params) + f4m_formats = self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + continue + + if src_url.startswith('http') and self._is_valid_url(src, video_id): + http_count += 1 + formats.append({ + 'url': src_url, + 'ext': ext or src_ext or 'flv', + 'format_id': 'http-%d' % (bitrate or http_count), + 'tbr': bitrate, + 'filesize': filesize, + 'width': width, + 'height': height, + }) + continue self._sort_formats(formats) return formats - def _parse_smil_video(self, video, video_id, base, rtmp_count): - src = video.get('src') - if not src: - return ([], rtmp_count) - bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - proto = video.get('proto') - if not proto: - if base: - if base.startswith('rtmp'): - proto = 'rtmp' - elif base.startswith('http'): - proto = 'http' - ext = video.get('ext') - if proto == 'm3u8': - return (self._extract_m3u8_formats(src, video_id, ext), rtmp_count) - elif proto == 'rtmp': - rtmp_count += 1 - streamer = video.get('streamer') or base - return ([{ - 'url': streamer, - 'play_path': src, - 'ext': 'flv', - 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), - 'tbr': bitrate, - 'width': width, - 'height': height, - }], rtmp_count) - elif proto.startswith('http'): - return ([{ - 'url': base + src, - 'ext': ext or 'flv', - 'tbr': bitrate, - 'width': width, - 'height': height, - }], rtmp_count) + def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): + subtitles = {} + for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): + src = textstream.get('src') + if not src: + continue + ext = textstream.get('ext') or determine_ext(src) + if not ext: + type_ = textstream.get('type') + SUBTITLES_TYPES = { + 'text/vtt': 'vtt', + 'text/srt': 'srt', + 'application/smptett+xml': 'tt', + } + if type_ in SUBTITLES_TYPES: + ext = SUBTITLES_TYPES[type_] + lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang + subtitles.setdefault(lang, []).append({ + 'url': src, + 'ext': ext, + }) + return subtitles + + def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): + xspf = self._download_xml( + playlist_url, playlist_id, 'Downloading xpsf playlist', + 'Unable to download xspf manifest', fatal=fatal) + if xspf is False: + return [] + return self._parse_xspf(xspf, playlist_id) + + def _parse_xspf(self, playlist, playlist_id): + NS_MAP = { + 'xspf': 'http://xspf.org/ns/0/', + 's1': 'http://static.streamone.nl/player/ns/0', + } + + entries = [] + for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): + title = xpath_text( + track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) + description = xpath_text( + track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') + thumbnail = xpath_text( + track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') + duration = float_or_none( + xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) + + formats = [{ + 'url': location.text, + 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), + 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), + 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), + } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] + self._sort_formats(formats) + + entries.append({ + 'id': playlist_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + }) + return entries def _live_title(self, name): """ Generate the title for a live video """ @@ -1035,6 +1277,12 @@ class InfoExtractor(object): None, '/', True, False, expire_time, '', None, None, None) self._downloader.cookiejar.set_cookie(cookie) + def _get_cookies(self, url): + """ Return a compat_cookies.SimpleCookie with the cookies for the url """ + req = compat_urllib_request.Request(url) + self._downloader.cookiejar.add_cookie_header(req) + return compat_cookies.SimpleCookie(req.get_header('Cookie')) + def get_testcases(self, include_onlymatching=False): t = getattr(self, '_TEST', None) if t: @@ -1073,6 +1321,23 @@ class InfoExtractor(object): def _get_subtitles(self, *args, **kwargs): raise NotImplementedError("This method must be implemented by subclasses") + @staticmethod + def _merge_subtitle_items(subtitle_list1, subtitle_list2): + """ Merge subtitle items for one language. Items with duplicated URLs + will be dropped. """ + list1_urls = set([item['url'] for item in subtitle_list1]) + ret = list(subtitle_list1) + ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls]) + return ret + + @classmethod + def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2): + """ Merge two subtitle dictionaries, language by language. """ + ret = dict(subtitle_dict1) + for lang in subtitle_dict2: + ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang]) + return ret + def extract_automatic_captions(self, *args, **kwargs): if (self._downloader.params.get('writeautomaticsub', False) or self._downloader.params.get('listsubtitles')): @@ -1086,7 +1351,7 @@ class InfoExtractor(object): class SearchInfoExtractor(InfoExtractor): """ Base class for paged search queries extractors. - They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} + They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} Instances should define _SEARCH_KEY and _MAX_RESULTS. """ diff --git a/youtube_dl/extractor/condenast.py b/youtube_dl/extractor/condenast.py index 3db4db4e4..6f92ae2ed 100644 --- a/youtube_dl/extractor/condenast.py +++ b/youtube_dl/extractor/condenast.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..compat import ( @@ -12,6 +11,7 @@ from ..compat import ( ) from ..utils import ( orderedSet, + remove_end, ) @@ -24,21 +24,33 @@ class CondeNastIE(InfoExtractor): # The keys are the supported sites and the values are the name to be shown # to the user and in the extractor description. _SITES = { - 'wired': 'WIRED', + 'allure': 'Allure', + 'architecturaldigest': 'Architectural Digest', + 'arstechnica': 'Ars Technica', + 'bonappetit': 'Bon Appétit', + 'brides': 'Brides', + 'cnevids': 'Condé Nast', + 'cntraveler': 'Condé Nast Traveler', + 'details': 'Details', + 'epicurious': 'Epicurious', + 'glamour': 'Glamour', + 'golfdigest': 'Golf Digest', 'gq': 'GQ', + 'newyorker': 'The New Yorker', + 'self': 'SELF', + 'teenvogue': 'Teen Vogue', + 'vanityfair': 'Vanity Fair', 'vogue': 'Vogue', - 'glamour': 'Glamour', + 'wired': 'WIRED', 'wmagazine': 'W Magazine', - 'vanityfair': 'Vanity Fair', - 'cnevids': 'Condé Nast', } - _VALID_URL = r'http://(video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) + _VALID_URL = r'http://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) - EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed)/.+?' % '|'.join(_SITES.keys()) + EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys()) - _TEST = { + _TESTS = [{ 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', 'md5': '1921f713ed48aabd715691f774c451f7', 'info_dict': { @@ -47,7 +59,16 @@ class CondeNastIE(InfoExtractor): 'title': '3D Printed Speakers Lit With LED', 'description': 'Check out these beautiful 3D printed LED speakers. You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.', } - } + }, { + # JS embed + 'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js', + 'md5': 'f1a6f9cafb7083bab74a710f65d08999', + 'info_dict': { + 'id': '55f9cf8b61646d1acf00000c', + 'ext': 'mp4', + 'title': '3D printed TSA Travel Sentry keys really do open TSA locks', + } + }] def _extract_series(self, url, webpage): title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>', @@ -86,8 +107,8 @@ class CondeNastIE(InfoExtractor): info_url = base_info_url + data info_page = self._download_webpage(info_url, video_id, 'Downloading video info') - video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info') - video_info = json.loads(video_info) + video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info') + video_info = self._parse_json(video_info, video_id) formats = [{ 'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']), @@ -111,6 +132,13 @@ class CondeNastIE(InfoExtractor): url_type = mobj.group('type') item_id = mobj.group('id') + # Convert JS embed to regular embed + if url_type == 'embedjs': + parsed_url = compat_urlparse.urlparse(url) + url = compat_urlparse.urlunparse(parsed_url._replace( + path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/'))) + url_type = 'embed' + self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site]) webpage = self._download_webpage(url, item_id) diff --git a/youtube_dl/extractor/criterion.py b/youtube_dl/extractor/criterion.py index 4fb178165..dedb810a0 100644 --- a/youtube_dl/extractor/criterion.py +++ b/youtube_dl/extractor/criterion.py @@ -27,9 +27,7 @@ class CriterionIE(InfoExtractor): final_url = self._search_regex( r'so.addVariable\("videoURL", "(.+?)"\)\;', webpage, 'video url') title = self._og_search_title(webpage) - description = self._html_search_regex( - r'<meta name="description" content="(.+?)" />', - webpage, 'video description') + description = self._html_search_meta('description', webpage) thumbnail = self._search_regex( r'so.addVariable\("thumbnailURL", "(.+?)"\)\;', webpage, 'thumbnail url') diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index 73f1e22ef..f8ce10111 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -12,23 +12,76 @@ from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_request, + compat_urlparse, ) from ..utils import ( ExtractorError, bytes_to_intlist, intlist_to_bytes, + int_or_none, + remove_end, unified_strdate, urlencode_postdata, + xpath_text, ) from ..aes import ( aes_cbc_decrypt, ) -class CrunchyrollIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)' +class CrunchyrollBaseIE(InfoExtractor): _NETRC_MACHINE = 'crunchyroll' + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + self.report_login() + login_url = 'https://www.crunchyroll.com/?a=formhandler' + data = urlencode_postdata({ + 'formname': 'RpcApiUser_Login', + 'name': username, + 'password': password, + }) + login_request = compat_urllib_request.Request(login_url, data) + login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + self._download_webpage(login_request, None, False, 'Wrong login info') + + def _real_initialize(self): + self._login() + + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): + request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request) + else compat_urllib_request.Request(url_or_request)) + # Accept-Language must be set explicitly to accept any language to avoid issues + # similar to https://github.com/rg3/youtube-dl/issues/6797. + # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction + # should be imposed or not (from what I can see it just takes the first language + # ignoring the priority and requires it to correspond the IP). By the way this causes + # Crunchyroll to not work in georestriction cases in some browsers that don't place + # the locale lang first in header. However allowing any language seems to workaround the issue. + request.add_header('Accept-Language', '*') + return super(CrunchyrollBaseIE, self)._download_webpage( + request, video_id, note, errnote, fatal, tries, timeout, encoding) + + @staticmethod + def _add_skip_wall(url): + parsed_url = compat_urlparse.urlparse(url) + qs = compat_urlparse.parse_qs(parsed_url.query) + # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message: + # > This content may be inappropriate for some people. + # > Are you sure you want to continue? + # since it's not disabled by default in crunchyroll account's settings. + # See https://github.com/rg3/youtube-dl/issues/7202. + qs['skip_wall'] = ['1'] + return compat_urlparse.urlunparse( + parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True))) + + +class CrunchyrollIE(CrunchyrollBaseIE): + _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)' _TESTS = [{ 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', 'info_dict': { @@ -60,10 +113,13 @@ class CrunchyrollIE(InfoExtractor): # rtmp 'skip_download': True, }, - }, { 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', 'only_matching': True, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617', + 'only_matching': True, }] _FORMAT_IDS = { @@ -73,24 +129,6 @@ class CrunchyrollIE(InfoExtractor): '1080': ('80', '108'), } - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - self.report_login() - login_url = 'https://www.crunchyroll.com/?a=formhandler' - data = urlencode_postdata({ - 'formname': 'RpcApiUser_Login', - 'name': username, - 'password': password, - }) - login_request = compat_urllib_request.Request(login_url, data) - login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - self._download_webpage(login_request, None, False, 'Wrong login info') - - def _real_initialize(self): - self._login() - def _decrypt_subtitles(self, data, iv, id): data = bytes_to_intlist(base64.b64decode(data.encode('utf-8'))) iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8'))) @@ -207,7 +245,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text def _get_subtitles(self, video_id, webpage): subtitles = {} - for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): + for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage): sub_page = self._download_webpage( 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id, video_id, note='Downloading subtitles for ' + sub_name) @@ -233,8 +271,10 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text else: webpage_url = 'http://www.' + mobj.group('url') - webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage') - note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='') + webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage') + note_m = self._html_search_regex( + r'<div class="showmedia-trailer-notice">(.+?)</div>', + webpage, 'trailer-notice', default='') if note_m: raise ExtractorError(note_m) @@ -244,17 +284,24 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text if msg.get('type') == 'error': raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True) + if 'To view this, please log in to verify you are 18 or older.' in webpage: + self.raise_login_required() + video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL) video_title = re.sub(r' {2,}', ' ', video_title) video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='') if not video_description: video_description = None - video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) + video_upload_date = self._html_search_regex( + [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'], + webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) if video_upload_date: video_upload_date = unified_strdate(video_upload_date) - video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL) + video_uploader = self._html_search_regex( + 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_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url')) playerdata_req = compat_urllib_request.Request(playerdata_url) playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url}) playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') @@ -278,13 +325,33 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text stream_info = streamdata.find('./{default}preload/stream_info') video_url = stream_info.find('./host').text video_play_path = stream_info.find('./file').text - formats.append({ + metadata = stream_info.find('./metadata') + format_info = { + 'format': video_format, + 'format_id': video_format, + 'height': int_or_none(xpath_text(metadata, './height')), + 'width': int_or_none(xpath_text(metadata, './width')), + } + + if '.fplive.net/' in video_url: + video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip()) + parsed_video_url = compat_urlparse.urlparse(video_url) + direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace( + netloc='v.lvlt.crcdn.net', + path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1]))) + if self._is_valid_url(direct_video_url, video_id, video_format): + format_info.update({ + 'url': direct_video_url, + }) + formats.append(format_info) + continue + + format_info.update({ 'url': video_url, 'play_path': video_play_path, 'ext': 'flv', - 'format': video_format, - 'format_id': video_format, }) + formats.append(format_info) subtitles = self.extract_subtitles(video_id, webpage) @@ -300,9 +367,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text } -class CrunchyrollShowPlaylistIE(InfoExtractor): +class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): IE_NAME = "crunchyroll:playlist" - _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?$' + _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)' _TESTS = [{ 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi', @@ -311,12 +378,25 @@ class CrunchyrollShowPlaylistIE(InfoExtractor): 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi' }, 'playlist_count': 13, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova', + 'info_dict': { + 'id': 'cosplay-complex-ova', + 'title': 'Cosplay Complex OVA' + }, + 'playlist_count': 3, + 'skip': 'Georestricted', + }, { + # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14 + 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1', + 'only_matching': True, }] def _real_extract(self, url): show_id = self._match_id(url) - webpage = self._download_webpage(url, show_id) + webpage = self._download_webpage(self._add_skip_wall(url), show_id) title = self._html_search_regex( r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>', webpage, 'title') diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 8852f0add..bc7823931 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -13,8 +13,9 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + determine_ext, int_or_none, - orderedSet, + parse_iso8601, str_to_int, unescapeHTML, ) @@ -28,10 +29,16 @@ class DailymotionBaseInfoExtractor(InfoExtractor): request.add_header('Cookie', 'family_filter=off; ff=off') return request + def _download_webpage_handle_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage_handle(request, *args, **kwargs) + + def _download_webpage_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage(request, *args, **kwargs) -class DailymotionIE(DailymotionBaseInfoExtractor): - """Information Extractor for Dailymotion""" +class DailymotionIE(DailymotionBaseInfoExtractor): _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)' IE_NAME = 'dailymotion' @@ -50,9 +57,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'info_dict': { 'id': 'x2iuewm', 'ext': 'mp4', - 'uploader': 'IGN', 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News', + 'description': 'Several come bundled with the Steam Controller.', + 'thumbnail': 're:^https?:.*\.(?:jpg|png)$', + 'duration': 74, + 'timestamp': 1425657362, 'upload_date': '20150306', + 'uploader': 'IGN', + 'uploader_id': 'xijv66', + 'age_limit': 0, + 'view_count': int, + 'comment_count': int, } }, # Vevo video @@ -81,46 +96,128 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'uploader': 'HotWaves1012', 'age_limit': 18, } + }, + # geo-restricted, player v5 + { + 'url': 'http://www.dailymotion.com/video/xhza0o', + 'only_matching': True, } ] def _real_extract(self, url): video_id = self._match_id(url) - url = 'https://www.dailymotion.com/video/%s' % video_id - # Retrieve video webpage to extract further information - request = self._build_request(url) - webpage = self._download_webpage(request, video_id) + webpage = self._download_webpage_no_ff( + 'https://www.dailymotion.com/video/%s' % video_id, video_id) + + age_limit = self._rta_search(webpage) - # Extract URL, uploader and title from webpage - self.report_extraction(video_id) + description = self._og_search_description(webpage) or self._html_search_meta( + 'description', webpage, 'description') + + view_count = str_to_int(self._search_regex( + [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"', + r'video_views_count[^>]+>\s+([\d\.,]+)'], + webpage, 'view count', fatal=False)) + comment_count = int_or_none(self._search_regex( + r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"', + webpage, 'comment count', fatal=False)) + + player_v5 = self._search_regex( + [r'buildPlayer\(({.+?})\);', r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);'], + webpage, 'player v5', default=None) + if player_v5: + player = self._parse_json(player_v5, video_id) + metadata = player['metadata'] + + self._check_error(metadata) + + formats = [] + for quality, media_list in metadata['qualities'].items(): + for media in media_list: + media_url = media.get('url') + if not media_url: + continue + type_ = media.get('type') + if type_ == 'application/vnd.lumberjack.manifest': + continue + ext = determine_ext(media_url) + if type_ == 'application/x-mpegURL' or ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + media_url, video_id, 'mp4', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif type_ == 'application/f4m' or ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + media_url, video_id, preference=-1, f4m_id='hds', fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + else: + f = { + 'url': media_url, + 'format_id': quality, + } + m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url) + if m: + f.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + formats.append(f) + self._sort_formats(formats) + + title = metadata['title'] + duration = int_or_none(metadata.get('duration')) + timestamp = int_or_none(metadata.get('created_time')) + thumbnail = metadata.get('poster_url') + uploader = metadata.get('owner', {}).get('screenname') + uploader_id = metadata.get('owner', {}).get('id') + + subtitles = {} + for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items(): + subtitles[subtitle_lang] = [{ + 'ext': determine_ext(subtitle_url), + 'url': subtitle_url, + } for subtitle_url in subtitle.get('urls', [])] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'age_limit': age_limit, + 'view_count': view_count, + 'comment_count': comment_count, + 'formats': formats, + 'subtitles': subtitles, + } - # It may just embed a vevo video: - m_vevo = re.search( + # vevo embed + vevo_id = self._search_regex( r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)', - webpage) - if m_vevo is not None: - vevo_id = m_vevo.group('id') - self.to_screen('Vevo video detected: %s' % vevo_id) - return self.url_result('vevo:%s' % vevo_id, ie='Vevo') + webpage, 'vevo embed', default=None) + if vevo_id: + return self.url_result('vevo:%s' % vevo_id, 'Vevo') - age_limit = self._rta_search(webpage) + # fallback old player + embed_page = self._download_webpage_no_ff( + 'https://www.dailymotion.com/embed/video/%s' % video_id, + video_id, 'Downloading embed page') - video_upload_date = None - mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage) - if mobj is not None: - video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3) - - embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id - embed_request = self._build_request(embed_url) - embed_page = self._download_webpage( - embed_request, video_id, 'Downloading embed page') - info = self._search_regex(r'var info = ({.*?}),$', embed_page, - 'video info', flags=re.MULTILINE) - info = json.loads(info) - if info.get('error') is not None: - msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] - raise ExtractorError(msg, expected=True) + timestamp = parse_iso8601(self._html_search_meta( + 'video:release_date', webpage, 'upload date')) + + info = self._parse_json( + self._search_regex( + r'var info = ({.*?}),$', embed_page, + 'video info', flags=re.MULTILINE), + video_id) + + self._check_error(info) formats = [] for (key, format_id) in self._FORMATS: @@ -138,16 +235,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'width': width, 'height': height, }) - if not formats: - raise ExtractorError('Unable to extract video URL') + self._sort_formats(formats) # subtitles video_subtitles = self.extract_subtitles(video_id, webpage) - view_count = str_to_int(self._search_regex( - r'video_views_count[^>]+>\s+([\d\.,]+)', - webpage, 'view count', fatal=False)) - title = self._og_search_title(webpage, default=None) if title is None: title = self._html_search_regex( @@ -158,14 +250,21 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'id': video_id, 'formats': formats, 'uploader': info['owner.screenname'], - 'upload_date': video_upload_date, + 'timestamp': timestamp, 'title': title, + 'description': description, 'subtitles': video_subtitles, 'thumbnail': info['thumbnail_url'], 'age_limit': age_limit, 'view_count': view_count, + 'duration': info['duration'] } + def _check_error(self, info): + if info.get('error') is not None: + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True) + def _get_subtitles(self, video_id, webpage): try: sub_list = self._download_webpage( @@ -197,18 +296,26 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): }] def _extract_entries(self, id): - video_ids = [] + video_ids = set() + processed_urls = set() for pagenum in itertools.count(1): - request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum)) - webpage = self._download_webpage(request, - id, 'Downloading page %s' % pagenum) + page_url = self._PAGE_TEMPLATE % (id, pagenum) + webpage, urlh = self._download_webpage_handle_no_ff( + page_url, id, 'Downloading page %s' % pagenum) + if urlh.geturl() in processed_urls: + self.report_warning('Stopped at duplicated page %s, which is the same as %s' % ( + page_url, urlh.geturl()), id) + break + + processed_urls.add(urlh.geturl()) - video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage)) + for video_id in re.findall(r'data-xid="(.+?)"', webpage): + if video_id not in video_ids: + yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') + video_ids.add(video_id) if re.search(self._MORE_PAGES_INDICATOR, webpage) is None: break - return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') - for video_id in orderedSet(video_ids)] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -225,7 +332,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): class DailymotionUserIE(DailymotionPlaylistIE): IE_NAME = 'dailymotion:user' - _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$' + _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)' _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s' _TESTS = [{ 'url': 'https://www.dailymotion.com/user/nqtv', @@ -234,6 +341,17 @@ class DailymotionUserIE(DailymotionPlaylistIE): 'title': 'Rémi Gaillard', }, 'playlist_mincount': 100, + }, { + 'url': 'http://www.dailymotion.com/user/UnderProject', + 'info_dict': { + 'id': 'UnderProject', + 'title': 'UnderProject', + }, + 'playlist_mincount': 1800, + 'expected_warnings': [ + 'Stopped at duplicated page', + ], + 'skip': 'Takes too long time', }] def _real_extract(self, url): @@ -284,8 +402,7 @@ class DailymotionCloudIE(DailymotionBaseInfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = self._build_request(url) - webpage = self._download_webpage(request, video_id) + webpage = self._download_webpage_no_ff(url, video_id) title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title') diff --git a/youtube_dl/extractor/dcn.py b/youtube_dl/extractor/dcn.py new file mode 100644 index 000000000..6f2fea5ff --- /dev/null +++ b/youtube_dl/extractor/dcn.py @@ -0,0 +1,84 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_request, +) +from ..utils import ( + int_or_none, + parse_iso8601, +) + + +class DCNIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/.+|show/\d+/.+?)/(?P<id>\d+)' + _TEST = { + 'url': 'http://www.dcndigital.ae/#/show/199074/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375/6887', + 'info_dict': + { + 'id': '17375', + 'ext': 'mp4', + 'title': 'رحلة العمر : الحلقة 1', + 'description': 'md5:0156e935d870acb8ef0a66d24070c6d6', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 2041, + 'timestamp': 1227504126, + 'upload_date': '20081124', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + request = compat_urllib_request.Request( + 'http://admin.mangomolo.com/analytics/index.php/plus/video?id=%s' % video_id, + headers={'Origin': 'http://www.dcndigital.ae'}) + + video = self._download_json(request, video_id) + title = video.get('title_en') or video['title_ar'] + + webpage = self._download_webpage( + 'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' + + compat_urllib_parse.urlencode({ + 'id': video['id'], + 'user_id': video['user_id'], + 'signature': video['signature'], + 'countries': 'Q0M=', + 'filter': 'DENY', + }), video_id) + + m3u8_url = self._html_search_regex(r'file:\s*"([^"]+)', webpage, 'm3u8 url') + formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') + + 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', + }) + + self._sort_formats(formats) + + img = video.get('img') + thumbnail = 'http://admin.mangomolo.com/analytics/%s' % img if img else None + duration = int_or_none(video.get('duration')) + description = video.get('description_en') or video.get('description_ar') + timestamp = parse_iso8601(video.get('create_time') or video.get('update_time'), ' ') + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/dfb.py b/youtube_dl/extractor/dfb.py index 8049779b0..263532cc6 100644 --- a/youtube_dl/extractor/dfb.py +++ b/youtube_dl/extractor/dfb.py @@ -3,42 +3,47 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import unified_strdate class DFBIE(InfoExtractor): IE_NAME = 'tv.dfb.de' - _VALID_URL = r'https?://tv\.dfb\.de/video/[^/]+/(?P<id>\d+)' + _VALID_URL = r'https?://tv\.dfb\.de/video/(?P<display_id>[^/]+)/(?P<id>\d+)' _TEST = { - 'url': 'http://tv.dfb.de/video/highlights-des-empfangs-in-berlin/9070/', + 'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/', # The md5 is different each time 'info_dict': { - 'id': '9070', + 'id': '11633', + 'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland', 'ext': 'flv', - 'title': 'Highlights des Empfangs in Berlin', - 'upload_date': '20140716', + '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') - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, display_id) player_info = self._download_xml( 'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id, - 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()), video_id) + 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) return { 'id': video_id, + 'display_id': display_id, 'title': video_info.find('title').text, - 'url': manifest_url, - 'ext': 'flv', 'thumbnail': self._og_search_thumbnail(webpage), - 'upload_date': ''.join(video_info.find('time_date').text.split('.')[::-1]), + 'upload_date': unified_strdate(video_info.find('time_date').text), + 'formats': formats, } diff --git a/youtube_dl/extractor/dhm.py b/youtube_dl/extractor/dhm.py index 3ed1f1663..44e0c5d4d 100644 --- a/youtube_dl/extractor/dhm.py +++ b/youtube_dl/extractor/dhm.py @@ -1,10 +1,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - xpath_text, - parse_duration, -) +from ..utils import parse_duration class DHMIE(InfoExtractor): @@ -34,24 +31,14 @@ class DHMIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + playlist_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, playlist_id) playlist_url = self._search_regex( r"file\s*:\s*'([^']+)'", webpage, 'playlist url') - playlist = self._download_xml(playlist_url, video_id) - - track = playlist.find( - './{http://xspf.org/ns/0/}trackList/{http://xspf.org/ns/0/}track') - - video_url = xpath_text( - track, './{http://xspf.org/ns/0/}location', - 'video url', fatal=True) - thumbnail = xpath_text( - track, './{http://xspf.org/ns/0/}image', - 'thumbnail') + entries = self._extract_xspf_playlist(playlist_url, playlist_id) title = self._search_regex( [r'dc:title="([^"]+)"', r'<title> »([^<]+)</title>'], @@ -63,11 +50,10 @@ class DHMIE(InfoExtractor): r'<em>Length\s*</em>\s*:\s*</strong>([^<]+)', webpage, 'duration', default=None)) - return { - 'id': video_id, - 'url': video_url, + entries[0].update({ 'title': title, 'description': description, 'duration': duration, - 'thumbnail': thumbnail, - } + }) + + return self.playlist_result(entries, playlist_id) diff --git a/youtube_dl/extractor/divxstage.py b/youtube_dl/extractor/divxstage.py deleted file mode 100644 index b88379e06..000000000 --- a/youtube_dl/extractor/divxstage.py +++ /dev/null @@ -1,27 +0,0 @@ -from __future__ import unicode_literals - -from .novamov import NovaMovIE - - -class DivxStageIE(NovaMovIE): - IE_NAME = 'divxstage' - IE_DESC = 'DivxStage' - - _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'divxstage\.(?:eu|net|ch|co|at|ag|to)'} - - _HOST = 'www.divxstage.eu' - - _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<' - _TITLE_REGEX = r'<div class="video_det">\s*<strong>([^<]+)</strong>' - _DESCRIPTION_REGEX = r'<div class="video_det">\s*<strong>[^<]+</strong>\s*<p>([^<]+)</p>' - - _TEST = { - 'url': 'http://www.divxstage.eu/video/57f238e2e5e01', - 'md5': '63969f6eb26533a1968c4d325be63e72', - 'info_dict': { - 'id': '57f238e2e5e01', - 'ext': 'flv', - 'title': 'youtubedl test video', - 'description': 'This is a test video for youtubedl.', - } - } diff --git a/youtube_dl/extractor/dramafever.py b/youtube_dl/extractor/dramafever.py index ca41a3abf..38e6597c8 100644 --- a/youtube_dl/extractor/dramafever.py +++ b/youtube_dl/extractor/dramafever.py @@ -23,8 +23,23 @@ class DramaFeverBaseIE(InfoExtractor): _LOGIN_URL = 'https://www.dramafever.com/accounts/login/' _NETRC_MACHINE = 'dramafever' + _CONSUMER_SECRET = 'DA59dtVXYLxajktV' + + _consumer_secret = None + + def _get_consumer_secret(self): + mainjs = self._download_webpage( + 'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js', + None, 'Downloading main.js', fatal=False) + if not mainjs: + return self._CONSUMER_SECRET + return self._search_regex( + r"var\s+cs\s*=\s*'([^']+)'", mainjs, + 'consumer secret', default=self._CONSUMER_SECRET) + def _real_initialize(self): self._login() + self._consumer_secret = self._get_consumer_secret() def _login(self): (username, password) = self._get_login_info() @@ -119,6 +134,23 @@ class DramaFeverIE(DramaFeverBaseIE): 'url': href, }] + series_id, episode_number = video_id.split('.') + episode_info = self._download_json( + # We only need a single episode info, so restricting page size to one episode + # and dealing with page number as with episode number + r'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_number=%s&page_size=1' + % (self._consumer_secret, series_id, episode_number), + video_id, 'Downloading episode info JSON', fatal=False) + if episode_info: + value = episode_info.get('value') + if value: + subfile = value[0].get('subfile') or value[0].get('new_subfile') + if subfile and subfile != 'http://www.dramafever.com/st/': + subtitles.setdefault('English', []).append({ + 'ext': 'srt', + 'url': subfile, + }) + return { 'id': video_id, 'title': title, @@ -152,27 +184,14 @@ class DramaFeverSeriesIE(DramaFeverBaseIE): 'playlist_count': 20, }] - _CONSUMER_SECRET = 'DA59dtVXYLxajktV' _PAGE_SIZE = 60 # max is 60 (see http://api.drama9.com/#get--api-4-episode-series-) - def _get_consumer_secret(self, video_id): - mainjs = self._download_webpage( - 'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js', - video_id, 'Downloading main.js', fatal=False) - if not mainjs: - return self._CONSUMER_SECRET - return self._search_regex( - r"var\s+cs\s*=\s*'([^']+)'", mainjs, - 'consumer secret', default=self._CONSUMER_SECRET) - def _real_extract(self, url): series_id = self._match_id(url) - consumer_secret = self._get_consumer_secret(series_id) - series = self._download_json( 'http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s' - % (consumer_secret, series_id), + % (self._consumer_secret, series_id), series_id, 'Downloading series JSON')['series'][series_id] title = clean_html(series['name']) @@ -182,7 +201,7 @@ class DramaFeverSeriesIE(DramaFeverBaseIE): for page_num in itertools.count(1): episodes = self._download_json( 'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d' - % (consumer_secret, series_id, self._PAGE_SIZE, page_num), + % (self._consumer_secret, series_id, self._PAGE_SIZE, page_num), series_id, 'Downloading episodes JSON page #%d' % page_num) for episode in episodes.get('value', []): episode_url = episode.get('episode_url') diff --git a/youtube_dl/extractor/dumpert.py b/youtube_dl/extractor/dumpert.py index 999fb5620..1f00386fe 100644 --- a/youtube_dl/extractor/dumpert.py +++ b/youtube_dl/extractor/dumpert.py @@ -9,8 +9,8 @@ from ..utils import qualities class DumpertIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/mediabase/(?P<id>[0-9]+/[0-9a-zA-Z]+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?dumpert\.nl/(?:mediabase|embed)/(?P<id>[0-9]+/[0-9a-zA-Z]+)' + _TESTS = [{ 'url': 'http://www.dumpert.nl/mediabase/6646981/951bc60f/', 'md5': '1b9318d7d5054e7dcb9dc7654f21d643', 'info_dict': { @@ -20,11 +20,15 @@ class DumpertIE(InfoExtractor): 'description': 'Niet schrikken hoor', 'thumbnail': 're:^https?://.*\.jpg$', } - } + }, { + 'url': 'http://www.dumpert.nl/embed/6675421/dc440fe7/', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) + url = 'https://www.dumpert.nl/mediabase/' + video_id req = compat_urllib_request.Request(url) req.add_header('Cookie', 'nsfw=1; cpc=10') webpage = self._download_webpage(req, video_id) diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py index 688dfc2f7..7bbf617d4 100644 --- a/youtube_dl/extractor/eagleplatform.py +++ b/youtube_dl/extractor/eagleplatform.py @@ -21,7 +21,7 @@ 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': '0b7994faa2bd5c0f69a3db6db28d078d', + 'md5': '70f5187fb620f2c1d503b3b22fd4efe3', 'info_dict': { 'id': '227304', 'ext': 'mp4', @@ -36,7 +36,7 @@ class EaglePlatformIE(InfoExtractor): # http://muz-tv.ru/play/7129/ # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true 'url': 'eagleplatform:media.clipyou.ru:12820', - 'md5': '6c2ebeab03b739597ce8d86339d5a905', + 'md5': '90b26344ba442c8e44aa4cf8f301164a', 'info_dict': { 'id': '12820', 'ext': 'mp4', @@ -48,7 +48,8 @@ class EaglePlatformIE(InfoExtractor): 'skip': 'Georestricted', }] - def _handle_error(self, response): + @staticmethod + def _handle_error(response): status = int_or_none(response.get('status', 200)) if status != 200: raise ExtractorError(' '.join(response['errors']), expected=True) @@ -58,6 +59,9 @@ class EaglePlatformIE(InfoExtractor): self._handle_error(response) return response + def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'): + return self._download_json(url_or_request, video_id, note)['data'][0] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id') @@ -69,7 +73,7 @@ class EaglePlatformIE(InfoExtractor): title = media['title'] description = media.get('description') - thumbnail = media.get('snapshot') + thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:') duration = int_or_none(media.get('duration')) view_count = int_or_none(media.get('views')) @@ -78,13 +82,20 @@ class EaglePlatformIE(InfoExtractor): if age_restriction: age_limit = 0 if age_restriction == 'allow_all' else 18 - m3u8_data = self._download_json( - media['sources']['secure_m3u8']['auto'], - video_id, 'Downloading m3u8 JSON') + secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:') + m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON') formats = self._extract_m3u8_formats( - m3u8_data['data'][0], video_id, - 'mp4', entry_protocol='m3u8_native') + m3u8_url, video_id, + 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') + + mp4_url = self._get_video_url( + # Secure mp4 URL is constructed according to Player.prototype.mp4 from + # http://lentaru.media.eagleplatform.com/player/player.js + re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4', secure_m3u8), + video_id, 'Downloading mp4 JSON') + formats.append({'url': mp4_url, 'format_id': 'mp4'}) + self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/ehow.py b/youtube_dl/extractor/ehow.py index 9cb1bf301..b1cd4f5d4 100644 --- a/youtube_dl/extractor/ehow.py +++ b/youtube_dl/extractor/ehow.py @@ -1,9 +1,7 @@ from __future__ import unicode_literals -from ..compat import ( - compat_urllib_parse, -) from .common import InfoExtractor +from ..compat import compat_urllib_parse_unquote class EHowIE(InfoExtractor): @@ -26,7 +24,7 @@ class EHowIE(InfoExtractor): webpage = self._download_webpage(url, video_id) video_url = self._search_regex( r'(?:file|source)=(http[^\'"&]*)', webpage, 'video URL') - final_url = compat_urllib_parse.unquote(video_url) + final_url = compat_urllib_parse_unquote(video_url) uploader = self._html_search_meta('uploader', webpage) title = self._og_search_title(webpage).replace(' | eHow', '') diff --git a/youtube_dl/extractor/engadget.py b/youtube_dl/extractor/engadget.py index 4ea37ebd9..e4180701d 100644 --- a/youtube_dl/extractor/engadget.py +++ b/youtube_dl/extractor/engadget.py @@ -10,7 +10,7 @@ from ..utils import ( class EngadgetIE(InfoExtractor): _VALID_URL = r'''(?x)https?://www.engadget.com/ - (?:video/5min/(?P<id>\d+)| + (?:video(?:/5min)?/(?P<id>\d+)| [\d/]+/.*?) ''' diff --git a/youtube_dl/extractor/eroprofile.py b/youtube_dl/extractor/eroprofile.py index 316033cf1..7fcd0151d 100644 --- a/youtube_dl/extractor/eroprofile.py +++ b/youtube_dl/extractor/eroprofile.py @@ -71,8 +71,7 @@ class EroProfileIE(InfoExtractor): m = re.search(r'You must be logged in to view this video\.', webpage) if m: - raise ExtractorError( - 'This video requires login. Please specify a username and password and try again.', expected=True) + self.raise_login_required('This video requires login') video_id = self._search_regex( [r"glbUpdViews\s*\('\d*','(\d+)'", r'p/report/video/(\d+)'], diff --git a/youtube_dl/extractor/esri.py b/youtube_dl/extractor/esri.py new file mode 100644 index 000000000..bf5d2019f --- /dev/null +++ b/youtube_dl/extractor/esri.py @@ -0,0 +1,74 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + int_or_none, + parse_filesize, + unified_strdate, +) + + +class EsriVideoIE(InfoExtractor): + _VALID_URL = r'https?://video\.esri\.com/watch/(?P<id>[0-9]+)' + _TEST = { + 'url': 'https://video.esri.com/watch/1124/arcgis-online-_dash_-developing-applications', + 'md5': 'd4aaf1408b221f1b38227a9bbaeb95bc', + 'info_dict': { + 'id': '1124', + 'ext': 'mp4', + 'title': 'ArcGIS Online - Developing Applications', + 'description': 'Jeremy Bartley demonstrates how to develop applications with ArcGIS Online.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 185, + 'upload_date': '20120419', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + formats = [] + for width, height, content in re.findall( + r'(?s)<li><strong>(\d+)x(\d+):</strong>(.+?)</li>', webpage): + for video_url, ext, filesize in re.findall( + r'<a[^>]+href="([^"]+)">([^<]+) \(([^<]+)\)</a>', content): + formats.append({ + 'url': compat_urlparse.urljoin(url, video_url), + 'ext': ext.lower(), + 'format_id': '%s-%s' % (ext.lower(), height), + 'width': int(width), + 'height': int(height), + 'filesize_approx': parse_filesize(filesize), + }) + self._sort_formats(formats) + + title = self._html_search_meta('title', webpage, 'title') + description = self._html_search_meta( + 'description', webpage, 'description', fatal=False) + + thumbnail = self._html_search_meta('thumbnail', webpage, 'thumbnail', fatal=False) + if thumbnail: + thumbnail = re.sub(r'_[st]\.jpg$', '_x.jpg', thumbnail) + + duration = int_or_none(self._search_regex( + [r'var\s+videoSeconds\s*=\s*(\d+)', r"'duration'\s*:\s*(\d+)"], + webpage, 'duration', fatal=False)) + + upload_date = unified_strdate(self._html_search_meta( + 'last-modified', webpage, 'upload date', fatal=None)) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'upload_date': upload_date, + 'formats': formats + } diff --git a/youtube_dl/extractor/europa.py b/youtube_dl/extractor/europa.py new file mode 100644 index 000000000..adc43919e --- /dev/null +++ b/youtube_dl/extractor/europa.py @@ -0,0 +1,93 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + int_or_none, + orderedSet, + parse_duration, + qualities, + unified_strdate, + xpath_text +) + + +class EuropaIE(InfoExtractor): + _VALID_URL = r'https?://ec\.europa\.eu/avservices/(?:video/player|audio/audioDetails)\.cfm\?.*?\bref=(?P<id>[A-Za-z0-9-]+)' + _TESTS = [{ + 'url': 'http://ec.europa.eu/avservices/video/player.cfm?ref=I107758', + 'md5': '574f080699ddd1e19a675b0ddf010371', + 'info_dict': { + 'id': 'I107758', + 'ext': 'mp4', + 'title': 'TRADE - Wikileaks on TTIP', + 'description': 'NEW LIVE EC Midday press briefing of 11/08/2015', + 'thumbnail': 're:^https?://.*\.jpg$', + 'upload_date': '20150811', + 'duration': 34, + 'view_count': int, + 'formats': 'mincount:3', + } + }, { + 'url': 'http://ec.europa.eu/avservices/video/player.cfm?sitelang=en&ref=I107786', + 'only_matching': True, + }, { + 'url': 'http://ec.europa.eu/avservices/audio/audioDetails.cfm?ref=I-109295&sitelang=en', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + playlist = self._download_xml( + 'http://ec.europa.eu/avservices/video/player/playlist.cfm?ID=%s' % video_id, video_id) + + def get_item(type_, preference): + items = {} + for item in playlist.findall('./info/%s/item' % type_): + lang, label = xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None) + if lang and label: + items[lang] = label.strip() + for p in preference: + if items.get(p): + return items[p] + + query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + preferred_lang = query.get('sitelang', ('en', ))[0] + + preferred_langs = orderedSet((preferred_lang, 'en', 'int')) + + title = get_item('title', preferred_langs) or video_id + description = get_item('description', preferred_langs) + thumbnmail = xpath_text(playlist, './info/thumburl', 'thumbnail') + upload_date = unified_strdate(xpath_text(playlist, './info/date', 'upload date')) + duration = parse_duration(xpath_text(playlist, './info/duration', 'duration')) + view_count = int_or_none(xpath_text(playlist, './info/views', 'views')) + + language_preference = qualities(preferred_langs[::-1]) + + formats = [] + for file_ in playlist.findall('./files/file'): + video_url = xpath_text(file_, './url') + if not video_url: + continue + lang = xpath_text(file_, './lg') + formats.append({ + 'url': video_url, + 'format_id': lang, + 'format_note': xpath_text(file_, './lglabel'), + 'language_preference': language_preference(lang) + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnmail, + 'upload_date': upload_date, + 'duration': duration, + 'view_count': view_count, + 'formats': formats + } diff --git a/youtube_dl/extractor/expotv.py b/youtube_dl/extractor/expotv.py index a38b773e8..1585a03bb 100644 --- a/youtube_dl/extractor/expotv.py +++ b/youtube_dl/extractor/expotv.py @@ -33,20 +33,27 @@ class ExpoTVIE(InfoExtractor): webpage = self._download_webpage(url, video_id) player_key = self._search_regex( r'<param name="playerKey" value="([^"]+)"', webpage, 'player key') - config_url = 'http://client.expotv.com/video/config/%s/%s' % ( - video_id, player_key) config = self._download_json( - config_url, video_id, - note='Downloading video configuration') + 'http://client.expotv.com/video/config/%s/%s' % (video_id, player_key), + video_id, 'Downloading video configuration') - formats = [{ - 'url': fcfg['file'], - 'height': int_or_none(fcfg.get('height')), - 'format_note': fcfg.get('label'), - 'ext': self._search_regex( - r'filename=.*\.([a-z0-9_A-Z]+)&', fcfg['file'], - 'file extension', default=None), - } for fcfg in config['sources']] + formats = [] + for fcfg in config['sources']: + media_url = fcfg.get('file') + if not media_url: + continue + if fcfg.get('type') == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')) + else: + formats.append({ + 'url': media_url, + 'height': int_or_none(fcfg.get('height')), + 'format_id': fcfg.get('label'), + 'ext': self._search_regex( + r'filename=.*\.([a-z0-9_A-Z]+)&', media_url, + 'file extension', default=None) or fcfg.get('type'), + }) self._sort_formats(formats) title = self._og_search_title(webpage) diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index 82dc27bc6..f53c51615 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -9,14 +9,15 @@ from ..compat import ( compat_http_client, compat_str, compat_urllib_error, - compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_request, ) from ..utils import ( ExtractorError, - int_or_none, limit_length, urlencode_postdata, + get_element_by_id, + clean_html, ) @@ -42,6 +43,7 @@ class FacebookIE(InfoExtractor): 'id': '637842556329505', 'ext': 'mp4', 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam', + 'uploader': 'Tennis on Facebook', } }, { 'note': 'Video without discernible title', @@ -50,6 +52,7 @@ class FacebookIE(InfoExtractor): 'id': '274175099429670', 'ext': 'mp4', 'title': 'Facebook video #274175099429670', + 'uploader': 'Asif Nawab Butt', }, 'expected_warnings': [ 'title' @@ -136,18 +139,22 @@ class FacebookIE(InfoExtractor): else: raise ExtractorError('Cannot parse data') data = dict(json.loads(m.group(1))) - params_raw = compat_urllib_parse.unquote(data['params']) + params_raw = compat_urllib_parse_unquote(data['params']) params = json.loads(params_raw) - video_data = params['video_data'][0] formats = [] - for quality in ['sd', 'hd']: - src = video_data.get('%s_src' % quality) - if src is not None: - formats.append({ - 'format_id': quality, - 'url': src, - }) + for format_id, f in params['video_data'].items(): + if not f or not isinstance(f, list): + continue + for quality in ('sd', 'hd'): + for src_type in ('src', 'src_no_ratelimit'): + src = f[0].get('%s_%s' % (quality, src_type)) + if src: + formats.append({ + 'format_id': '%s_%s_%s' % (format_id, quality, src_type), + 'url': src, + 'preference': -10 if format_id == 'progressive' else 0, + }) if not formats: raise ExtractorError('Cannot find video formats') @@ -161,11 +168,11 @@ class FacebookIE(InfoExtractor): video_title = limit_length(video_title, 80) if not video_title: video_title = 'Facebook video #%s' % video_id + uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage)) return { 'id': video_id, 'title': video_title, 'formats': formats, - 'duration': int_or_none(video_data.get('video_duration')), - 'thumbnail': video_data.get('thumbnail_src'), + 'uploader': uploader, } diff --git a/youtube_dl/extractor/fc2.py b/youtube_dl/extractor/fc2.py index 1ccc1a964..a406945e8 100644 --- a/youtube_dl/extractor/fc2.py +++ b/youtube_dl/extractor/fc2.py @@ -10,12 +10,13 @@ from ..compat import ( compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, ) class FC2IE(InfoExtractor): - _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)?content/(?P<id>[^/]+)' + _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)*content/(?P<id>[^/]+)' IE_NAME = 'fc2' _NETRC_MACHINE = 'fc2' _TESTS = [{ @@ -37,6 +38,9 @@ class FC2IE(InfoExtractor): 'password': '(snip)', 'skip': 'requires actual password' } + }, { + 'url': 'http://video.fc2.com/en/a/content/20130926eZpARwsF', + 'only_matching': True, }] def _login(self): @@ -52,10 +56,7 @@ class FC2IE(InfoExtractor): 'Submit': ' Login ', } - # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode - # chokes on unicode - login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items()) - login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8') + login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') request = compat_urllib_request.Request( 'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data) @@ -80,13 +81,13 @@ class FC2IE(InfoExtractor): title = self._og_search_title(webpage) thumbnail = self._og_search_thumbnail(webpage) - refer = url.replace('/content/', '/a/content/') + refer = url.replace('/content/', '/a/content/') if '/a/content/' not in url else url mimi = hashlib.md5((video_id + '_gGddgPfeaf_gzyr').encode('utf-8')).hexdigest() info_url = ( "http://video.fc2.com/ginfo.php?mimi={1:s}&href={2:s}&v={0:s}&fversion=WIN%2011%2C6%2C602%2C180&from=2&otag=0&upid={0:s}&tk=null&". - format(video_id, mimi, compat_urllib_request.quote(refer, safe='').replace('.', '%2E'))) + format(video_id, mimi, compat_urllib_request.quote(refer, safe=b'').replace('.', '%2E'))) info_webpage = self._download_webpage( info_url, video_id, note='Downloading info page') diff --git a/youtube_dl/extractor/fczenit.py b/youtube_dl/extractor/fczenit.py new file mode 100644 index 000000000..f1f150ef2 --- /dev/null +++ b/youtube_dl/extractor/fczenit.py @@ -0,0 +1,41 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class FczenitIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?fc-zenit\.ru/video/gl(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://fc-zenit.ru/video/gl6785/', + 'md5': '458bacc24549173fe5a5aa29174a5606', + 'info_dict': { + 'id': '6785', + 'ext': 'mp4', + 'title': '«Зенит-ТВ»: как Олег Шатов играл против «Урала»', + }, + } + + def _real_extract(self, url): + 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') + + bitrates_raw = self._html_search_regex(r'bitrates:.*\n(.*)\]', webpage, 'video URL') + bitrates = re.findall(r'url:.?\'(.+?)\'.*?bitrate:.?([0-9]{3}?)', bitrates_raw) + + formats = [{ + 'url': furl, + 'tbr': tbr, + } for furl, tbr in bitrates] + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video_title, + 'formats': formats, + } diff --git a/youtube_dl/extractor/fivemin.py b/youtube_dl/extractor/fivemin.py index 157094e8c..2955965d9 100644 --- a/youtube_dl/extractor/fivemin.py +++ b/youtube_dl/extractor/fivemin.py @@ -2,11 +2,15 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import ( - compat_str, compat_urllib_parse, + compat_parse_qs, + compat_urllib_parse_urlparse, + compat_urlparse, ) from ..utils import ( ExtractorError, + parse_duration, + replace_extension, ) @@ -28,6 +32,7 @@ class FiveMinIE(InfoExtractor): 'id': '518013791', 'ext': 'mp4', 'title': 'iPad Mini with Retina Display Review', + 'duration': 177, }, }, { @@ -38,9 +43,52 @@ class FiveMinIE(InfoExtractor): 'id': '518086247', 'ext': 'mp4', 'title': 'How to Make a Next-Level Fruit Salad', + 'duration': 184, }, }, ] + _ERRORS = { + 'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.', + 'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.', + 'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.', + 'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.', + 'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.', + 'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.', + } + _QUALITIES = { + 1: { + 'width': 640, + 'height': 360, + }, + 2: { + 'width': 854, + 'height': 480, + }, + 4: { + 'width': 1280, + 'height': 720, + }, + 8: { + 'width': 1920, + 'height': 1080, + }, + 16: { + 'width': 640, + 'height': 360, + }, + 32: { + 'width': 854, + 'height': 480, + }, + 64: { + 'width': 1280, + 'height': 720, + }, + 128: { + 'width': 640, + 'height': 360, + }, + } def _real_extract(self, url): video_id = self._match_id(url) @@ -59,26 +107,36 @@ class FiveMinIE(InfoExtractor): 'https://syn.5min.com/handlers/SenseHandler.ashx?' + query, video_id) if not response['success']: - err_msg = response['errorMessage'] - if err_msg == 'ErrorVideoUserNotGeo': - msg = 'Video not available from your location' - else: - msg = 'Aol said: %s' % err_msg - raise ExtractorError(msg, expected=True, video_id=video_id) + raise ExtractorError( + '%s said: %s' % ( + self.IE_NAME, + self._ERRORS.get(response['errorMessage'], response['errorMessage'])), + expected=True) info = response['binding'][0] - second_id = compat_str(int(video_id[:-2]) + 1) formats = [] - for quality, height in [(1, 320), (2, 480), (4, 720), (8, 1080)]: - if any(r['ID'] == quality for r in info['Renditions']): + parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs( + compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0]) + for rendition in info['Renditions']: + if rendition['RenditionType'] == 'm3u8': + formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls')) + elif rendition['RenditionType'] == 'aac': + continue + else: + rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType']))) + quality = self._QUALITIES.get(rendition['ID'], {}) formats.append({ - 'format_id': compat_str(quality), - 'url': 'http://avideos.5min.com/%s/%s/%s_%s.mp4' % (second_id[-3:], second_id, video_id, quality), - 'height': height, + 'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']), + 'url': rendition_url, + 'width': quality.get('width'), + 'height': quality.get('height'), }) + self._sort_formats(formats) return { 'id': video_id, 'title': info['Title'], + 'thumbnail': info.get('ThumbURL'), + 'duration': parse_duration(info.get('Duration')), 'formats': formats, } diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py index 190d9f9ad..40ea27895 100644 --- a/youtube_dl/extractor/fktv.py +++ b/youtube_dl/extractor/fktv.py @@ -1,13 +1,12 @@ from __future__ import unicode_literals import re -import random -import json from .common import InfoExtractor from ..utils import ( - get_element_by_id, clean_html, + determine_ext, + ExtractorError, ) @@ -17,66 +16,40 @@ class FKTVIE(InfoExtractor): _TEST = { 'url': 'http://fernsehkritik.tv/folge-1', + 'md5': '21f0b0c99bce7d5b524eb1b17b1c6d79', 'info_dict': { - 'id': '00011', - 'ext': 'flv', + 'id': '1', + 'ext': 'mp4', 'title': 'Folge 1 vom 10. April 2007', - 'description': 'md5:fb4818139c7cfe6907d4b83412a6864f', + 'thumbnail': 're:^https?://.*\.jpg$', }, } def _real_extract(self, url): - episode = int(self._match_id(url)) - - video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%s.jpg' % episode - start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/Start' % episode, - episode) - playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage, - 'playlist', flags=re.DOTALL) - files = json.loads(re.sub('{[^{}]*?}', '{}', playlist)) - - videos = [] - for i, _ in enumerate(files, 1): - video_id = '%04d%d' % (episode, i) - video_url = 'http://fernsehkritik.tv/js/directme.php?file=%s%s.flv' % (episode, '' if i == 1 else '-%d' % i) - videos.append({ - 'ext': 'flv', - 'id': video_id, - 'url': video_url, - 'title': clean_html(get_element_by_id('eptitle', start_webpage)), - 'description': clean_html(get_element_by_id('contentlist', start_webpage)), - 'thumbnail': video_thumbnail - }) - return { - '_type': 'multi_video', - 'entries': videos, - 'id': 'folge-%s' % episode, - } - - -class FKTVPosteckeIE(InfoExtractor): - IE_NAME = 'fernsehkritik.tv:postecke' - _VALID_URL = r'http://(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)' - _TEST = { - 'url': 'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120', - 'md5': '262f0adbac80317412f7e57b4808e5c4', - 'info_dict': { - 'id': '0120', - 'ext': 'flv', - 'title': 'Postecke 120', - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - episode = int(mobj.group('ep')) - - server = random.randint(2, 4) - video_id = '%04d' % episode - video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode) - video_title = 'Postecke %d' % episode + episode = self._match_id(url) + + webpage = self._download_webpage( + 'http://fernsehkritik.tv/folge-%s/play' % episode, episode) + title = clean_html(self._html_search_regex( + '<h3>([^<]+)</h3>', webpage, 'title')) + matches = re.search( + r'(?s)<video(?:(?!poster)[^>])+(?:poster="([^"]+)")?[^>]*>(.*)</video>', + webpage) + if matches is None: + raise ExtractorError('Unable to extract the video') + + poster, sources = matches.groups() + if poster is None: + self.report_warning('unable to extract thumbnail') + + urls = re.findall(r'<source[^>]+src="([^"]+)"', sources) + formats = [{ + 'url': furl, + 'format_id': determine_ext(furl), + } for furl in urls] return { - 'id': video_id, - 'url': video_url, - 'title': video_title, + 'id': episode, + 'title': title, + 'formats': formats, + 'thumbnail': poster, } diff --git a/youtube_dl/extractor/folketinget.py b/youtube_dl/extractor/folketinget.py index 0fb29de75..75399fa7d 100644 --- a/youtube_dl/extractor/folketinget.py +++ b/youtube_dl/extractor/folketinget.py @@ -30,6 +30,10 @@ class FolketingetIE(InfoExtractor): 'upload_date': '20141120', 'duration': 3960, }, + 'params': { + # rtmp download + 'skip_download': True, + }, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index b2284ab01..fb6d108c0 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -32,6 +32,7 @@ class FourTubeIE(InfoExtractor): 'view_count': int, 'like_count': int, 'categories': list, + 'age_limit': 18, } } @@ -45,10 +46,10 @@ class FourTubeIE(InfoExtractor): thumbnail = self._html_search_meta('thumbnailUrl', webpage) uploader_id = self._html_search_regex( r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">', - webpage, 'uploader id') + webpage, 'uploader id', fatal=False) uploader = self._html_search_regex( r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">', - webpage, 'uploader') + webpage, 'uploader', fatal=False) categories_html = self._search_regex( r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>', @@ -67,13 +68,24 @@ class FourTubeIE(InfoExtractor): webpage, 'like count', fatal=False)) duration = parse_duration(self._html_search_meta('duration', webpage)) - params_js = self._search_regex( - r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', - webpage, 'initialization parameters' - ) - params = self._parse_json('[%s]' % params_js, video_id) - media_id = params[0] - sources = ['%s' % p for p in params[2]] + media_id = self._search_regex( + r'<button[^>]+data-id=(["\'])(?P<id>\d+)\1[^>]+data-quality=', webpage, + 'media id', default=None, group='id') + sources = [ + quality + for _, quality in re.findall(r'<button[^>]+data-quality=(["\'])(.+?)\1', webpage)] + if not (media_id and sources): + player_js = self._download_webpage( + self._search_regex( + r'<script[^>]id=(["\'])playerembed\1[^>]+src=(["\'])(?P<url>.+?)\2', + webpage, 'player JS', group='url'), + video_id, 'Downloading player JS') + params_js = self._search_regex( + r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', + player_js, 'initialization parameters') + params = self._parse_json('[%s]' % params_js, video_id) + media_id = params[0] + sources = ['%s' % p for p in params[2]] token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format( media_id, '+'.join(sources)) diff --git a/youtube_dl/extractor/foxnews.py b/youtube_dl/extractor/foxnews.py index 917f76b1e..3a4a59135 100644 --- a/youtube_dl/extractor/foxnews.py +++ b/youtube_dl/extractor/foxnews.py @@ -1,5 +1,7 @@ from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..utils import ( parse_iso8601, @@ -8,7 +10,8 @@ from ..utils import ( class FoxNewsIE(InfoExtractor): - _VALID_URL = r'https?://video\.foxnews\.com/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)' + IE_DESC = 'Fox News and Fox Business Video' + _VALID_URL = r'https?://(?P<host>video\.fox(?:news|business)\.com)/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)' _TESTS = [ { 'url': 'http://video.foxnews.com/v/3937480/frozen-in-time/#sp=show-clips', @@ -42,13 +45,19 @@ class FoxNewsIE(InfoExtractor): 'url': 'http://video.foxnews.com/v/video-embed.html?video_id=3937480&d=video.foxnews.com', 'only_matching': True, }, + { + 'url': 'http://video.foxbusiness.com/v/4442309889001', + 'only_matching': True, + }, ] def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + host = mobj.group('host') video = self._download_json( - 'http://video.foxnews.com/v/feed/video/%s.js?template=fox' % video_id, video_id) + 'http://%s/v/feed/video/%s.js?template=fox' % (host, video_id), video_id) item = video['channel']['item'] title = item['title'] diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index b2c984bf2..8e60cf60f 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -6,15 +6,11 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( clean_html, ExtractorError, int_or_none, - float_or_none, parse_duration, determine_ext, ) @@ -59,12 +55,12 @@ class FranceTVBaseInfoExtractor(InfoExtractor): # See https://github.com/rg3/youtube-dl/issues/3963 # m3u8 urls work fine continue - video_url_parsed = compat_urllib_parse_urlparse(video_url) f4m_url = self._download_webpage( - 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url_parsed.path, + 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url, video_id, 'Downloading f4m manifest token', fatal=False) if f4m_url: - formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id)) + formats.extend(self._extract_f4m_formats( + f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, 1, format_id)) elif ext == 'm3u8': formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id)) elif video_url.startswith('rtmp'): @@ -82,28 +78,48 @@ class FranceTVBaseInfoExtractor(InfoExtractor): }) self._sort_formats(formats) + title = info['titre'] + subtitle = info.get('sous_titre') + if subtitle: + title += ' - %s' % subtitle + + subtitles = {} + subtitles_list = [{ + 'url': subformat['url'], + 'ext': subformat.get('format'), + } for subformat in info.get('subtitles', []) if subformat.get('url')] + if subtitles_list: + subtitles['fr'] = subtitles_list + return { 'id': video_id, - 'title': info['titre'], + 'title': title, 'description': clean_html(info['synopsis']), 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']), - 'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']), + 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, + 'subtitles': subtitles, } class PluzzIE(FranceTVBaseInfoExtractor): IE_NAME = 'pluzz.francetv.fr' - _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html' + _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P<id>.+?)\.html' # Can't use tests, videos expire in 7 days def _real_extract(self, url): - title = re.match(self._VALID_URL, url).group(1) - webpage = self._download_webpage(url, title) - video_id = self._search_regex( - r'data-diffusion="(\d+)"', webpage, 'ID') + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + video_id = self._html_search_meta( + 'id_video', webpage, 'video id', default=None) + if not video_id: + video_id = self._search_regex( + r'data-diffusion=["\'](\d+)', webpage, 'video id') + return self._extract_video(video_id, 'Pluzz') @@ -119,6 +135,9 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): 'title': 'Soir 3', 'upload_date': '20130826', 'timestamp': 1377548400, + 'subtitles': { + 'fr': 'mincount:2', + }, }, }, { 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html', @@ -160,11 +179,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): class FranceTVIE(FranceTVBaseInfoExtractor): IE_NAME = 'francetv' IE_DESC = 'France 2, 3, 4, 5 and Ô' - _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/ - (?: - emissions/.*?/(videos|emissions)/(?P<id>[^/?]+) - | (emissions?|jt)/(?P<key>[^/?]+) - )''' + _VALID_URL = r'''(?x) + https?:// + (?: + (?:www\.)?france[2345o]\.fr/ + (?: + emissions/[^/]+/(?:videos|diffusions)| + emission/[^/]+| + videos| + jt + ) + /| + embed\.francetv\.fr/\?ue= + ) + (?P<id>[^/?]+) + ''' _TESTS = [ # france2 @@ -208,37 +237,59 @@ class FranceTVIE(FranceTVBaseInfoExtractor): }, # france5 { - 'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968', - 'md5': '78f0f4064f9074438e660785bbf2c5d9', + 'url': 'http://www.france5.fr/emissions/c-a-dire/videos/quels_sont_les_enjeux_de_cette_rentree_politique__31-08-2015_908948?onglet=tous&page=1', + 'md5': 'f6c577df3806e26471b3d21631241fd0', 'info_dict': { - 'id': '108961659', + 'id': '123327454', 'ext': 'flv', - 'title': 'C à dire ?!', - 'description': 'md5:1a4aeab476eb657bf57c4ff122129f81', - 'upload_date': '20140915', - 'timestamp': 1410795000, + 'title': 'C à dire ?! - Quels sont les enjeux de cette rentrée politique ?', + 'description': 'md5:4a0d5cb5dce89d353522a84462bae5a4', + 'upload_date': '20150831', + 'timestamp': 1441035120, }, }, # franceo { - 'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013', - 'md5': '52f0bfe202848b15915a2f39aaa8981b', + 'url': 'http://www.franceo.fr/jt/info-soir/18-07-2015', + 'md5': '47d5816d3b24351cdce512ad7ab31da8', + 'info_dict': { + 'id': '125377621', + 'ext': 'flv', + 'title': 'Infô soir', + 'description': 'md5:01b8c6915a3d93d8bbbd692651714309', + 'upload_date': '20150718', + 'timestamp': 1437241200, + 'duration': 414, + }, + }, + { + # francetv embed + 'url': 'http://embed.francetv.fr/?ue=8d7d3da1e3047c42ade5a5d7dfd3fc87', 'info_dict': { - 'id': '108634970', + 'id': 'EV_30231', 'ext': 'flv', - 'title': 'Infô Afrique', - 'description': 'md5:ebf346da789428841bee0fd2a935ea55', - 'upload_date': '20140915', - 'timestamp': 1410822000, + 'title': 'Alcaline, le concert avec Calogero', + 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff', + 'upload_date': '20150226', + 'timestamp': 1424989860, + 'duration': 5400, }, }, + { + 'url': 'http://www.france4.fr/emission/highlander/diffusion-du-17-07-2015-04h05', + 'only_matching': True, + }, + { + 'url': 'http://www.franceo.fr/videos/125377617', + 'only_matching': True, + } ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id')) + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) video_id, catalogue = self._html_search_regex( - r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', + r'href="http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video ID').split('@') return self._extract_video(video_id, catalogue) diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index dd87257c4..f5f13689c 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -53,7 +53,7 @@ class FunnyOrDieIE(InfoExtractor): for bitrate in bitrates: for link in links: formats.append({ - 'url': '%s%d.%s' % (link[0], bitrate, link[1]), + 'url': self._proto_relative_url('%s%d.%s' % (link[0], bitrate, link[1])), 'format_id': '%s-%d' % (link[1], bitrate), 'vbr': bitrate, }) diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 2d33fa7f5..b3f1bafcc 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -5,7 +5,7 @@ import json from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, compat_urlparse, ) from ..utils import ( @@ -75,7 +75,7 @@ class GameSpotIE(InfoExtractor): return { 'id': data_video['guid'], 'display_id': page_id, - 'title': compat_urllib_parse.unquote(data_video['title']), + 'title': compat_urllib_parse_unquote(data_video['title']), 'formats': formats, 'description': self._html_search_meta('description', webpage), 'thumbnail': self._og_search_thumbnail(webpage), diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index 43f916412..a6834db43 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -7,7 +7,10 @@ from ..compat import ( compat_urllib_parse, compat_urllib_request, ) -from ..utils import remove_end +from ..utils import ( + remove_end, + HEADRequest, +) class GDCVaultIE(InfoExtractor): @@ -73,10 +76,20 @@ class GDCVaultIE(InfoExtractor): return video_formats def _parse_flv(self, xml_description): - video_formats = [] + formats = [] akamai_url = xml_description.find('./metadata/akamaiHost').text + audios = xml_description.find('./metadata/audios') + if audios is not None: + for audio in audios: + formats.append({ + 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, + 'play_path': remove_end(audio.get('url'), '.flv'), + 'ext': 'flv', + 'vcodec': 'none', + 'format_id': audio.get('code'), + }) slide_video_path = xml_description.find('./metadata/slideVideo').text - video_formats.append({ + formats.append({ 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, 'play_path': remove_end(slide_video_path, '.flv'), 'ext': 'flv', @@ -86,7 +99,7 @@ class GDCVaultIE(InfoExtractor): 'format_id': 'slides', }) speaker_video_path = xml_description.find('./metadata/speakerVideo').text - video_formats.append({ + formats.append({ 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, 'play_path': remove_end(speaker_video_path, '.flv'), 'ext': 'flv', @@ -95,7 +108,7 @@ class GDCVaultIE(InfoExtractor): 'preference': -1, 'format_id': 'speaker', }) - return video_formats + return formats def _login(self, webpage_url, display_id): (username, password) = self._get_login_info() @@ -133,16 +146,18 @@ class GDCVaultIE(InfoExtractor): r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);', start_page, 'url', default=None) if direct_url: - video_url = 'http://www.gdcvault.com/' + direct_url title = self._html_search_regex( r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>', start_page, 'title') + video_url = 'http://www.gdcvault.com' + direct_url + # resolve the url so that we can detect the correct extension + head = self._request_webpage(HEADRequest(video_url), video_id) + video_url = head.geturl() return { 'id': video_id, 'display_id': display_id, 'url': video_url, - 'ext': 'flv', 'title': title, } @@ -168,8 +183,8 @@ class GDCVaultIE(InfoExtractor): # Fallback to the older format xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename') - xml_decription_url = xml_root + 'xml/' + xml_name - xml_description = self._download_xml(xml_decription_url, display_id) + xml_description_url = xml_root + 'xml/' + xml_name + xml_description = self._download_xml(xml_description_url, display_id) video_title = xml_description.find('./metadata/title').text video_formats = self._parse_mp4(xml_description) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 392ad3648..ca5fbafb2 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -4,11 +4,11 @@ from __future__ import unicode_literals import os import re +import sys from .common import InfoExtractor from .youtube import YoutubeIE from ..compat import ( - compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_request, compat_urlparse, @@ -49,6 +49,8 @@ from .vimeo import VimeoIE from .dailymotion import DailymotionCloudIE from .onionstudios import OnionStudiosIE from .snagfilms import SnagFilmsEmbedIE +from .screenwavemedia import ScreenwaveMediaIE +from .mtv import MTVServicesEmbeddedIE class GenericIE(InfoExtractor): @@ -131,6 +133,89 @@ class GenericIE(InfoExtractor): 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624', } }, + # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng + { + 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml', + 'info_dict': { + 'id': 'smil', + 'ext': 'mp4', + 'title': 'Automatics, robotics and biocybernetics', + 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'formats': 'mincount:16', + 'subtitles': 'mincount:1', + }, + 'params': { + 'force_generic_extractor': True, + 'skip_download': True, + }, + }, + # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html + { + 'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil', + 'info_dict': { + 'id': 'hds', + 'ext': 'flv', + 'title': 'hds', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from https://www.restudy.dk/video/play/id/1637 + { + 'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml', + 'info_dict': { + 'id': 'video_1637', + 'ext': 'flv', + 'title': 'video_1637', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm + { + 'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil', + 'info_dict': { + 'id': 'smil-service', + 'ext': 'flv', + 'title': 'smil-service', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370 + { + 'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil', + 'info_dict': { + 'id': '4719370', + 'ext': 'mp4', + 'title': '571de1fd-47bc-48db-abf9-238872a58d1f', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html + { + 'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf', + 'info_dict': { + 'id': 'mZlp2ctYIUEB', + 'ext': 'mp4', + 'title': 'Tikibad ontruimd wegens brand', + 'description': 'md5:05ca046ff47b931f9b04855015e163a4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 33, + }, + 'params': { + 'skip_download': True, + }, + }, # google redirect { 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE', @@ -148,6 +233,22 @@ class GenericIE(InfoExtractor): } }, { + # redirect in Refresh HTTP header + 'url': 'https://www.facebook.com/l.php?u=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpO8h3EaFRdo&h=TAQHsoToz&enc=AZN16h-b6o4Zq9pZkCCdOLNKMN96BbGMNtcFwHSaazus4JHT_MFYkAA-WARTX2kvsCIdlAIyHZjl6d33ILIJU7Jzwk_K3mcenAXoAzBNoZDI_Q7EXGDJnIhrGkLXo_LJ_pAa2Jzbx17UHMd3jAs--6j2zaeto5w9RTn8T_1kKg3fdC5WPX9Dbb18vzH7YFX0eSJmoa6SP114rvlkw6pkS1-T&s=1', + 'info_dict': { + 'id': 'pO8h3EaFRdo', + 'ext': 'mp4', + 'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set', + 'description': 'md5:6294cc1af09c4049e0652b51a2df10d5', + 'upload_date': '20150917', + 'uploader_id': 'brtvofficial', + 'uploader': 'Boiler Room', + }, + 'params': { + 'skip_download': False, + }, + }, + { 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', 'md5': '85b90ccc9d73b4acd9138d3af4c27f89', 'info_dict': { @@ -237,6 +338,19 @@ class GenericIE(InfoExtractor): }, 'add_ie': ['Ooyala'], }, + { + # ooyala video embedded with http://player.ooyala.com/iframe.js + 'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/', + 'info_dict': { + 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB', + 'ext': 'mp4', + 'title': '"Steve Jobs: Man in the Machine" trailer', + 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."', + }, + 'params': { + 'skip_download': True, + }, + }, # multiple ooyala embeds on SBN network websites { 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', @@ -277,14 +391,6 @@ class GenericIE(InfoExtractor): 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', }, }, - # BBC iPlayer embeds - { - 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER', - 'info_dict': { - 'title': 'BBC - Blogs - Adam Curtis - BUGGER', - }, - 'playlist_mincount': 18, - }, # RUTV embed { 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', @@ -408,6 +514,26 @@ class GenericIE(InfoExtractor): 'skip_download': 'Requires rtmpdump' } }, + # francetv embed + { + 'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero', + 'info_dict': { + 'id': 'EV_30231', + 'ext': 'mp4', + 'title': 'Alcaline, le concert avec Calogero', + 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff', + 'upload_date': '20150226', + 'timestamp': 1424989860, + 'duration': 5400, + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + }, + 'expected_warnings': [ + 'Forbidden' + ] + }, # Condé Nast embed { 'url': 'http://www.wired.com/2014/04/honda-asimo/', @@ -894,6 +1020,16 @@ class GenericIE(InfoExtractor): 'description': 'New experience with Acrobat DC', 'duration': 248.667, }, + }, + # ScreenwaveMedia embed + { + 'url': 'http://www.thecinemasnob.com/the-cinema-snob/a-nightmare-on-elm-street-2-freddys-revenge1', + 'md5': '24ace5baba0d35d55c6810b51f34e9e0', + 'info_dict': { + 'id': 'cinemasnob-55d26273809dd', + 'ext': 'mp4', + 'title': 'cinemasnob', + }, } ] @@ -1099,11 +1235,15 @@ class GenericIE(InfoExtractor): self.report_extraction(video_id) - # Is it an RSS feed? + # Is it an RSS feed, a SMIL file or a XSPF playlist? try: doc = parse_xml(webpage) if doc.tag == 'rss': return self._extract_rss(url, video_id, doc) + elif re.match(r'^(?:{[^}]+})?smil$', doc.tag): + return self._parse_smil(doc, url, video_id) + elif doc.tag == '{http://xspf.org/ns/0/}playlist': + return self.playlist_result(self._parse_xspf(doc, video_id), video_id) except compat_xml_parse_error: pass @@ -1115,7 +1255,7 @@ class GenericIE(InfoExtractor): # Sometimes embedded video player is hidden behind percent encoding # (e.g. https://github.com/rg3/youtube-dl/issues/2448) # Unescaping the whole page allows to handle those cases in a generic way - webpage = compat_urllib_parse.unquote(webpage) + webpage = compat_urllib_parse_unquote(webpage) # it's tempting to parse this further, but you would # have to take into account all the variations like @@ -1177,6 +1317,12 @@ class GenericIE(InfoExtractor): if vimeo_url is not None: return self.url_result(vimeo_url) + vid_me_embed_url = self._search_regex( + r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]', + webpage, 'vid.me embed', default=None) + if vid_me_embed_url is not None: + return self.url_result(vid_me_embed_url, 'Vidme') + # Look for embedded YouTube player matches = re.findall(r'''(?x) (?: @@ -1303,7 +1449,7 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) # Look for Ooyala videos - mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or + mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage)) @@ -1369,7 +1515,7 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage) if mobj is not None: - return self.url_result(compat_urllib_parse.unquote(mobj.group('url'))) + return self.url_result(compat_urllib_parse_unquote(mobj.group('url'))) # Look for funnyordie embed matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage) @@ -1432,6 +1578,13 @@ class GenericIE(InfoExtractor): if mobj is not None: return self.url_result(mobj.group('url'), 'ArteTVEmbed') + # Look for embedded francetv player + mobj = re.search( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1', + webpage) + if mobj is not None: + return self.url_result(mobj.group('url')) + # Look for embedded smotri.com player smotri_url = SmotriIE._extract_url(webpage) if smotri_url: @@ -1459,12 +1612,9 @@ class GenericIE(InfoExtractor): return self.url_result(url, ie='Vulture') # Look for embedded mtvservices player - mobj = re.search( - r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"', - webpage) - if mobj is not None: - url = unescapeHTML(mobj.group('url')) - return self.url_result(url, ie='MTVServicesEmbedded') + mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage) + if mtvservices_url: + return self.url_result(mtvservices_url, ie='MTVServicesEmbedded') # Look for embedded yahoo player mobj = re.search( @@ -1503,7 +1653,7 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url'), 'MLB') mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL, + r'<(?:iframe|script)[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL, webpage) if mobj is not None: return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast') @@ -1594,6 +1744,11 @@ class GenericIE(InfoExtractor): if snagfilms_url: return self.url_result(snagfilms_url) + # Look for ScreenwaveMedia embeds + mobj = re.search(ScreenwaveMediaIE.EMBED_PATTERN, webpage) + if mobj is not None: + return self.url_result(unescapeHTML(mobj.group('url')), 'ScreenwaveMedia') + # Look for AdobeTVVideo embeds mobj = re.search( r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]', @@ -1631,7 +1786,7 @@ class GenericIE(InfoExtractor): if not found: # Broaden the findall a little bit: JWPlayer JS loader found = filter_video(re.findall( - r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)) + r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)) if not found: # Flow player found = filter_video(re.findall(r'''(?xs) @@ -1657,7 +1812,7 @@ class GenericIE(InfoExtractor): found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)) if not found: # HTML5 video - found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage) + found = re.findall(r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage) if not found: REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)' found = re.search( @@ -1668,9 +1823,12 @@ class GenericIE(InfoExtractor): # Look also in Refresh HTTP header refresh_header = head_response.headers.get('Refresh') if refresh_header: + # In python 2 response HTTP headers are bytestrings + if sys.version_info < (3, 0) and isinstance(refresh_header, str): + refresh_header = refresh_header.decode('iso-8859-1') found = re.search(REDIRECT_REGEX, refresh_header) if found: - new_url = compat_urlparse.urljoin(url, found.group(1)) + new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1))) self.report_following_redirect(new_url) return { '_type': 'url', @@ -1682,7 +1840,7 @@ class GenericIE(InfoExtractor): entries = [] for video_url in found: video_url = compat_urlparse.urljoin(url, video_url) - video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) + video_id = compat_urllib_parse_unquote(os.path.basename(video_url)) # Sometimes, jwplayer extraction will result in a YouTube URL if YoutubeIE.suitable(video_url): @@ -1692,7 +1850,8 @@ class GenericIE(InfoExtractor): # here's a fun little line of code for you: video_id = os.path.splitext(video_id)[0] - if determine_ext(video_url) == 'smil': + ext = determine_ext(video_url) + if ext == 'smil': entries.append({ 'id': video_id, 'formats': self._extract_smil_formats(video_url, video_id), @@ -1700,6 +1859,8 @@ class GenericIE(InfoExtractor): 'title': video_title, 'age_limit': age_limit, }) + elif ext == 'xspf': + return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id) else: entries.append({ 'id': video_id, diff --git a/youtube_dl/extractor/globo.py b/youtube_dl/extractor/globo.py index 8a95793ca..33d6432a6 100644 --- a/youtube_dl/extractor/globo.py +++ b/youtube_dl/extractor/globo.py @@ -13,6 +13,7 @@ from ..compat import ( from ..utils import ( ExtractorError, float_or_none, + int_or_none, ) @@ -359,13 +360,8 @@ class GloboIE(InfoExtractor): self._API_URL_TEMPLATE % video_id, video_id)['videos'][0] title = video['title'] - duration = float_or_none(video['duration'], 1000) - like_count = video['likes'] - uploader = video['channel'] - uploader_id = video['channel_id'] formats = [] - for resource in video['resources']: resource_id = resource.get('_id') if not resource_id: @@ -407,6 +403,11 @@ class GloboIE(InfoExtractor): self._sort_formats(formats) + duration = float_or_none(video.get('duration'), 1000) + like_count = int_or_none(video.get('likes')) + uploader = video.get('channel') + uploader_id = video.get('channel_id') + return { 'id': video_id, 'title': title, diff --git a/youtube_dl/extractor/gorillavid.py b/youtube_dl/extractor/gorillavid.py index aabf07a20..d23e3eac1 100644 --- a/youtube_dl/extractor/gorillavid.py +++ b/youtube_dl/extractor/gorillavid.py @@ -10,15 +10,16 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + encode_dict, int_or_none, ) class GorillaVidIE(InfoExtractor): - IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in and realvid.net' + IE_DESC = 'GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net and filehoot.com' _VALID_URL = r'''(?x) https?://(?P<host>(?:www\.)? - (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net))/ + (?:daclips\.in|gorillavid\.in|movpod\.in|fastvideo\.in|realvid\.net|filehoot\.com))/ (?:embed-)?(?P<id>[0-9a-zA-Z]+)(?:-[0-9]+x[0-9]+\.html)? ''' @@ -67,18 +68,27 @@ class GorillaVidIE(InfoExtractor): }, { 'url': 'http://movpod.in/0wguyyxi1yca', 'only_matching': True, + }, { + 'url': 'http://filehoot.com/3ivfabn7573c.html', + 'info_dict': { + 'id': '3ivfabn7573c', + 'ext': 'mp4', + 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', + 'thumbnail': 're:http://.*\.jpg', + } }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - webpage = self._download_webpage('http://%s/%s' % (mobj.group('host'), video_id), video_id) + 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: raise ExtractorError('Video %s does not exist' % video_id, expected=True) - fields = self._form_hidden_inputs(webpage) + fields = self._hidden_inputs(webpage) if fields['op'] == 'download1': countdown = int_or_none(self._search_regex( @@ -87,7 +97,7 @@ class GorillaVidIE(InfoExtractor): if countdown: self._sleep(countdown, video_id) - post = compat_urllib_parse.urlencode(fields) + post = compat_urllib_parse.urlencode(encode_dict(fields)) req = compat_urllib_request.Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') @@ -95,7 +105,7 @@ class GorillaVidIE(InfoExtractor): webpage = self._download_webpage(req, video_id, 'Downloading video page') title = self._search_regex( - [r'style="z-index: [0-9]+;">([^<]+)</span>', r'>Watch (.+) '], + [r'style="z-index: [0-9]+;">([^<]+)</span>', r'<td nowrap>([^<]+)</td>', r'>Watch (.+) '], webpage, 'title', default=None) or self._og_search_title(webpage) video_url = self._search_regex( r'file\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'file url') diff --git a/youtube_dl/extractor/hostingbulk.py b/youtube_dl/extractor/hostingbulk.py deleted file mode 100644 index 63f579592..000000000 --- a/youtube_dl/extractor/hostingbulk.py +++ /dev/null @@ -1,80 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) -from ..utils import ( - ExtractorError, - int_or_none, - urlencode_postdata, -) - - -class HostingBulkIE(InfoExtractor): - _VALID_URL = r'''(?x) - https?://(?:www\.)?hostingbulk\.com/ - (?:embed-)?(?P<id>[A-Za-z0-9]{12})(?:-\d+x\d+)?\.html''' - _FILE_DELETED_REGEX = r'<b>File Not Found</b>' - _TEST = { - 'url': 'http://hostingbulk.com/n0ulw1hv20fm.html', - 'md5': '6c8653c8ecf7ebfa83b76e24b7b2fe3f', - 'info_dict': { - 'id': 'n0ulw1hv20fm', - 'ext': 'mp4', - 'title': 'md5:5afeba33f48ec87219c269e054afd622', - 'filesize': 6816081, - 'thumbnail': 're:^http://.*\.jpg$', - } - } - - def _real_extract(self, url): - video_id = self._match_id(url) - url = 'http://hostingbulk.com/{0:}.html'.format(video_id) - - # Custom request with cookie to set language to English, so our file - # deleted regex would work. - request = compat_urllib_request.Request( - url, headers={'Cookie': 'lang=english'}) - webpage = self._download_webpage(request, video_id) - - if re.search(self._FILE_DELETED_REGEX, webpage) is not None: - raise ExtractorError('Video %s does not exist' % video_id, - expected=True) - - title = self._html_search_regex(r'<h3>(.*?)</h3>', webpage, 'title') - filesize = int_or_none( - self._search_regex( - r'<small>\((\d+)\sbytes?\)</small>', - webpage, - 'filesize', - fatal=False - ) - ) - thumbnail = self._search_regex( - r'<img src="([^"]+)".+?class="pic"', - webpage, 'thumbnail', fatal=False) - - fields = self._form_hidden_inputs(webpage) - - request = compat_urllib_request.Request(url, urlencode_postdata(fields)) - request.add_header('Content-type', 'application/x-www-form-urlencoded') - response = self._request_webpage(request, video_id, - 'Submiting download request') - video_url = response.geturl() - - formats = [{ - 'format_id': 'sd', - 'filesize': filesize, - 'url': video_url, - }] - - return { - 'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'formats': formats, - } diff --git a/youtube_dl/extractor/howstuffworks.py b/youtube_dl/extractor/howstuffworks.py index e97339121..663e6632a 100644 --- a/youtube_dl/extractor/howstuffworks.py +++ b/youtube_dl/extractor/howstuffworks.py @@ -10,7 +10,7 @@ from ..utils import ( class HowStuffWorksIE(InfoExtractor): - _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm' + _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*(?:\d+-)?(?P<id>.+?)-video\.htm' _TESTS = [ { 'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm', @@ -46,6 +46,10 @@ class HowStuffWorksIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg$', }, }, + { + 'url': 'http://shows.howstuffworks.com/stuff-to-blow-your-mind/optical-illusions-video.htm', + 'only_matching': True, + } ] def _real_extract(self, url): diff --git a/youtube_dl/extractor/iconosquare.py b/youtube_dl/extractor/iconosquare.py index 70e4c0d41..a39f422e9 100644 --- a/youtube_dl/extractor/iconosquare.py +++ b/youtube_dl/extractor/iconosquare.py @@ -1,7 +1,11 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import int_or_none +from ..utils import ( + int_or_none, + get_element_by_id, + remove_end, +) class IconosquareIE(InfoExtractor): @@ -12,7 +16,7 @@ class IconosquareIE(InfoExtractor): 'info_dict': { 'id': '522207370455279102_24101272', 'ext': 'mp4', - 'title': 'Instagram media by @aguynamedpatrick (Patrick Janelle)', + 'title': 'Instagram photo by @aguynamedpatrick (Patrick Janelle)', 'description': 'md5:644406a9ec27457ed7aa7a9ebcd4ce3d', 'timestamp': 1376471991, 'upload_date': '20130814', @@ -29,8 +33,7 @@ class IconosquareIE(InfoExtractor): webpage = self._download_webpage(url, video_id) media = self._parse_json( - self._search_regex( - r'window\.media\s*=\s*({.+?});\n', webpage, 'media'), + get_element_by_id('mediaJson', webpage), video_id) formats = [{ @@ -41,9 +44,7 @@ class IconosquareIE(InfoExtractor): } for format_id, f in media['videos'].items()] self._sort_formats(formats) - title = self._html_search_regex( - r'<title>(.+?)(?: *\(Videos?\))? \| (?:Iconosquare|Statigram)</title>', - webpage, 'title') + title = remove_end(self._og_search_title(webpage), ' - via Iconosquare') timestamp = int_or_none(media.get('created_time') or media.get('caption', {}).get('created_time')) description = media.get('caption', {}).get('text') @@ -61,6 +62,14 @@ class IconosquareIE(InfoExtractor): 'height': int_or_none(t.get('height')) } for thumbnail_id, t in media.get('images', {}).items()] + comments = [{ + 'id': comment.get('id'), + 'text': comment['text'], + 'timestamp': int_or_none(comment.get('created_time')), + 'author': comment.get('from', {}).get('full_name'), + 'author_id': comment.get('from', {}).get('username'), + } for comment in media.get('comments', {}).get('data', []) if 'text' in comment] + return { 'id': video_id, 'title': title, @@ -72,4 +81,5 @@ class IconosquareIE(InfoExtractor): 'comment_count': comment_count, 'like_count': like_count, 'formats': formats, + 'comments': comments, } diff --git a/youtube_dl/extractor/imdb.py b/youtube_dl/extractor/imdb.py index 4bb574cf3..02e1e428e 100644 --- a/youtube_dl/extractor/imdb.py +++ b/youtube_dl/extractor/imdb.py @@ -4,8 +4,8 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urlparse, +from ..utils import ( + qualities, ) @@ -30,24 +30,33 @@ class ImdbIE(InfoExtractor): descr = self._html_search_regex( r'(?s)<span itemprop="description">(.*?)</span>', webpage, 'description', fatal=False) - available_formats = re.findall( - r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage, - flags=re.MULTILINE) + player_url = 'http://www.imdb.com/video/imdb/vi%s/imdb/single' % video_id + player_page = self._download_webpage( + player_url, video_id, 'Downloading player page') + # the player page contains the info for the default format, we have to + # fetch other pages for the rest of the formats + extra_formats = re.findall(r'href="(?P<url>%s.*?)".*?>(?P<name>.*?)<' % re.escape(player_url), player_page) + format_pages = [ + self._download_webpage( + f_url, video_id, 'Downloading info for %s format' % f_name) + for f_url, f_name in extra_formats] + format_pages.append(player_page) + + quality = qualities(['SD', '480p', '720p']) formats = [] - for f_id, f_path in available_formats: - f_path = f_path.strip() - format_page = self._download_webpage( - compat_urlparse.urljoin(url, f_path), - 'Downloading info for %s format' % f_id) + for format_page in format_pages: 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'] formats.append({ 'format_id': f_id, 'url': format_info['videoInfoList'][0]['videoUrl'], + 'quality': quality(f_id), }) + self._sort_formats(formats) return { 'id': video_id, diff --git a/youtube_dl/extractor/imgur.py b/youtube_dl/extractor/imgur.py index d692ea79a..70c8ca64e 100644 --- a/youtube_dl/extractor/imgur.py +++ b/youtube_dl/extractor/imgur.py @@ -13,7 +13,7 @@ from ..utils import ( class ImgurIE(InfoExtractor): - _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?P<id>[a-zA-Z0-9]+)' + _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?!gallery)(?P<id>[a-zA-Z0-9]+)' _TESTS = [{ 'url': 'https://i.imgur.com/A61SaA1.gifv', @@ -97,3 +97,28 @@ class ImgurIE(InfoExtractor): 'description': self._og_search_description(webpage), 'title': self._og_search_title(webpage), } + + +class ImgurAlbumIE(InfoExtractor): + _VALID_URL = r'https?://(?:i\.)?imgur\.com/gallery/(?P<id>[a-zA-Z0-9]+)' + + _TEST = { + 'url': 'http://imgur.com/gallery/Q95ko', + 'info_dict': { + 'id': 'Q95ko', + }, + 'playlist_count': 25, + } + + def _real_extract(self, url): + album_id = self._match_id(url) + + album_images = self._download_json( + 'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id, + album_id)['data']['images'] + + entries = [ + self.url_result('http://imgur.com/%s' % image['hash']) + for image in album_images if image.get('hash')] + + return self.playlist_result(entries, album_id) diff --git a/youtube_dl/extractor/indavideo.py b/youtube_dl/extractor/indavideo.py new file mode 100644 index 000000000..12fb5e8e1 --- /dev/null +++ b/youtube_dl/extractor/indavideo.py @@ -0,0 +1,142 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_age_limit, + parse_iso8601, +) + + +class IndavideoEmbedIE(InfoExtractor): + _VALID_URL = r'https?://(?:(?:embed\.)?indavideo\.hu/player/video/|assets\.indavideo\.hu/swf/player\.swf\?.*\b(?:v(?:ID|id))=)(?P<id>[\da-f]+)' + _TESTS = [{ + 'url': 'http://indavideo.hu/player/video/1bdc3c6d80/', + 'md5': 'f79b009c66194acacd40712a6778acfa', + 'info_dict': { + 'id': '1837039', + 'ext': 'mp4', + 'title': 'Cicatánc', + 'description': '', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'cukiajanlo', + 'uploader_id': '83729', + 'timestamp': 1439193826, + 'upload_date': '20150810', + 'duration': 72, + 'age_limit': 0, + 'tags': ['tánc', 'cica', 'cuki', 'cukiajanlo', 'newsroom'], + }, + }, { + 'url': 'http://embed.indavideo.hu/player/video/1bdc3c6d80?autostart=1&hide=1', + 'only_matching': True, + }, { + 'url': 'http://assets.indavideo.hu/swf/player.swf?v=fe25e500&vID=1bdc3c6d80&autostart=1&hide=1&i=1', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'http://amfphp.indavideo.hu/SYm0json.php/player.playerHandler.getVideoData/%s' % video_id, + video_id)['data'] + + title = video['title'] + + video_urls = video.get('video_files', []) + video_file = video.get('video_file') + if video: + video_urls.append(video_file) + video_urls = list(set(video_urls)) + + video_prefix = video_urls[0].rsplit('/', 1)[0] + + for flv_file in video.get('flv_files', []): + flv_url = '%s/%s' % (video_prefix, flv_file) + if flv_url not in video_urls: + video_urls.append(flv_url) + + formats = [{ + 'url': video_url, + 'height': self._search_regex(r'\.(\d{3,4})\.mp4$', video_url, 'height', default=None), + } for video_url in video_urls] + self._sort_formats(formats) + + timestamp = video.get('date') + if timestamp: + # upload date is in CEST + timestamp = parse_iso8601(timestamp + ' +0200', ' ') + + thumbnails = [{ + 'url': self._proto_relative_url(thumbnail) + } for thumbnail in video.get('thumbnails', [])] + + tags = [tag['title'] for tag in video.get('tags', [])] + + return { + 'id': video.get('id') or video_id, + 'title': title, + 'description': video.get('description'), + 'thumbnails': thumbnails, + 'uploader': video.get('user_name'), + 'uploader_id': video.get('user_id'), + 'timestamp': timestamp, + 'duration': int_or_none(video.get('length')), + 'age_limit': parse_age_limit(video.get('age_limit')), + 'tags': tags, + 'formats': formats, + } + + +class IndavideoIE(InfoExtractor): + _VALID_URL = r'https?://(?:.+?\.)?indavideo\.hu/video/(?P<id>[^/#?]+)' + _TESTS = [{ + 'url': 'http://indavideo.hu/video/Vicces_cica_1', + 'md5': '8c82244ba85d2a2310275b318eb51eac', + 'info_dict': { + 'id': '1335611', + 'display_id': 'Vicces_cica_1', + 'ext': 'mp4', + 'title': 'Vicces cica', + 'description': 'Játszik a tablettel. :D', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Jet_Pack', + 'uploader_id': '491217', + 'timestamp': 1390821212, + 'upload_date': '20140127', + 'duration': 7, + 'age_limit': 0, + 'tags': ['vicces', 'macska', 'cica', 'ügyes', 'nevetés', 'játszik', 'Cukiság', 'Jet_Pack'], + }, + }, { + 'url': 'http://index.indavideo.hu/video/2015_0728_beregszasz', + 'only_matching': True, + }, { + 'url': 'http://auto.indavideo.hu/video/Sajat_utanfutoban_a_kis_tacsko', + 'only_matching': True, + }, { + 'url': 'http://erotika.indavideo.hu/video/Amator_tini_punci', + 'only_matching': True, + }, { + 'url': 'http://film.indavideo.hu/video/f_hrom_nagymamm_volt', + 'only_matching': True, + }, { + 'url': 'http://palyazat.indavideo.hu/video/Embertelen_dal_Dodgem_egyuttes', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + embed_url = self._search_regex( + r'<link[^>]+rel="video_src"[^>]+href="(.+?)"', webpage, 'embed url') + + return { + '_type': 'url_transparent', + 'ie_key': 'IndavideoEmbed', + 'url': embed_url, + 'display_id': display_id, + } diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py index 91a1b3ccb..71cfd12c5 100644 --- a/youtube_dl/extractor/infoq.py +++ b/youtube_dl/extractor/infoq.py @@ -4,7 +4,7 @@ import base64 from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, compat_urlparse, ) @@ -39,7 +39,7 @@ class InfoQIE(InfoExtractor): # Extract video URL encoded_id = self._search_regex( r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id') - real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8')) + real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8')) playpath = 'mp4:' + real_id video_filename = playpath.split('/')[-1] diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 0f6707d7c..2df1da3f0 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -3,19 +3,13 @@ from __future__ import unicode_literals import hashlib import math -import os.path import random -import re import time import uuid -import zlib from .common import InfoExtractor from ..compat import compat_urllib_parse -from ..utils import ( - ExtractorError, - url_basename, -) +from ..utils import ExtractorError class IqiyiIE(InfoExtractor): @@ -39,62 +33,57 @@ class IqiyiIE(InfoExtractor): 'title': '名侦探柯南第752集', }, 'playlist': [{ - 'md5': '7e49376fecaffa115d951634917fe105', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '41b75ba13bb7ac0e411131f92bc4f6ca', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '0cee1dd0a3d46a83e71e2badeae2aab0', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '4f8ad72373b0c491b582e7c196b0b1f9', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': 'd89ad028bcfad282918e8098e811711d', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '9cb1e5c95da25dff0660c32ae50903b7', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '155116e0ff1867bbc9b98df294faabc9', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '53f5db77622ae14fa493ed2a278a082b', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }], + 'params': { + 'skip_download': True, + }, }] _FORMATS_MAP = [ @@ -106,6 +95,10 @@ class IqiyiIE(InfoExtractor): ('10', 'h1'), ] + @staticmethod + def md5_text(text): + return hashlib.md5(text.encode('utf-8')).hexdigest() + def construct_video_urls(self, data, video_id, _uuid): def do_xor(x, y): a = y % 3 @@ -132,7 +125,7 @@ class IqiyiIE(InfoExtractor): 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 hashlib.md5((t + mg + x).encode('utf8')).hexdigest() + return self.md5_text(t + mg + x) video_urls_dict = {} for format_item in data['vp']['tkl'][0]['vs']: @@ -190,20 +183,19 @@ class IqiyiIE(InfoExtractor): def get_raw_data(self, tvid, video_id, enc_key, _uuid): tm = str(int(time.time())) + tail = tm + tvid param = { 'key': 'fvip', - 'src': hashlib.md5(b'youtube-dl').hexdigest(), + 'src': self.md5_text('youtube-dl'), 'tvId': tvid, 'vid': video_id, 'vinfo': 1, 'tm': tm, - 'enc': hashlib.md5( - (enc_key + tm + tvid).encode('utf8')).hexdigest(), + 'enc': self.md5_text(enc_key + tail), 'qyid': _uuid, 'tn': random.random(), 'um': 0, - 'authkey': hashlib.md5( - (tm + tvid).encode('utf8')).hexdigest() + 'authkey': self.md5_text(self.md5_text('') + tail), } api_url = 'http://cache.video.qiyi.com/vms' + '?' + \ @@ -212,20 +204,10 @@ class IqiyiIE(InfoExtractor): return raw_data def get_enc_key(self, swf_url, video_id): - filename, _ = os.path.splitext(url_basename(swf_url)) - enc_key_json = self._downloader.cache.load('iqiyi-enc-key', filename) - if enc_key_json is not None: - return enc_key_json[0] - - req = self._request_webpage( - swf_url, video_id, note='download swf content') - cn = req.read() - cn = zlib.decompress(cn[8:]) - pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv') - enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8') - - self._downloader.cache.store('iqiyi-enc-key', filename, [enc_key]) - + # TODO: automatic key extraction + # last update at 2015-10-22 for Zombie::bite + # '7223c67061dbea1259d0ceb44f44b6d62288f4f80c972170de5201d2321060270e05'[2:66][0::2] + enc_key = '2c76de15dcb44bd28ff0927d50d31620' return enc_key def _real_extract(self, url): diff --git a/youtube_dl/extractor/ir90tv.py b/youtube_dl/extractor/ir90tv.py new file mode 100644 index 000000000..214bcd5b5 --- /dev/null +++ b/youtube_dl/extractor/ir90tv.py @@ -0,0 +1,42 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import remove_start + + +class Ir90TvIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?90tv\.ir/video/(?P<id>[0-9]+)/.*' + _TESTS = [{ + 'url': 'http://90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218', + 'md5': '411dbd94891381960cb9e13daa47a869', + 'info_dict': { + 'id': '95719', + 'ext': 'mp4', + 'title': 'شایعات نقل و انتقالات مهم فوتبال اروپا 94/02/18', + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://www.90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + title = remove_start(self._html_search_regex( + r'<title>([^<]+)</title>', webpage, 'title'), '90tv.ir :: ') + + video_url = self._search_regex( + r'<source[^>]+src="([^"]+)"', webpage, 'video url') + + thumbnail = self._search_regex(r'poster="([^"]+)"', webpage, 'thumbnail url', fatal=False) + + return { + 'url': video_url, + 'id': video_id, + 'title': title, + 'video_url': video_url, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/jeuxvideo.py b/youtube_dl/extractor/jeuxvideo.py index d0720ff56..eef7daa29 100644 --- a/youtube_dl/extractor/jeuxvideo.py +++ b/youtube_dl/extractor/jeuxvideo.py @@ -8,9 +8,9 @@ from .common import InfoExtractor class JeuxVideoIE(InfoExtractor): - _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm' + _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)\.htm' - _TEST = { + _TESTS = [{ 'url': 'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm', 'md5': '046e491afb32a8aaac1f44dd4ddd54ee', 'info_dict': { @@ -19,13 +19,16 @@ class JeuxVideoIE(InfoExtractor): 'title': 'Tearaway : GC 2013 : Tearaway nous présente ses papiers d\'identité', 'description': 'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.', }, - } + }, { + 'url': 'http://www.jeuxvideo.com/videos/chroniques/434220/l-histoire-du-jeu-video-la-saturn.htm', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) title = mobj.group(1) webpage = self._download_webpage(url, title) - title = self._html_search_meta('name', webpage) + title = self._html_search_meta('name', webpage) or self._og_search_title(webpage) config_url = self._html_search_regex( r'data-src="(/contenu/medias/video.php.*?)"', webpage, 'config URL') diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index d28730492..3dca0e566 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -13,12 +13,24 @@ from ..utils import ( class KalturaIE(InfoExtractor): _VALID_URL = r'''(?x) - (?:kaltura:| - https?://(:?(?:www|cdnapisec)\.)?kaltura\.com/index\.php/kwidget/(?:[^/]+/)*?wid/_ - )(?P<partner_id>\d+) - (?::| - /(?:[^/]+/)*?entry_id/ - )(?P<id>[0-9a-z_]+)''' + (?: + kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)| + https?:// + (:?(?:www|cdnapisec)\.)?kaltura\.com/ + (?: + (?: + # flash player + index\.php/kwidget/ + (?:[^/]+/)*?wid/_(?P<partner_id>\d+)/ + (?:[^/]+/)*?entry_id/(?P<id>[0-9a-z_]+)| + # html5 player + html5/html5lib/ + (?:[^/]+/)*?entry_id/(?P<id_html5>[0-9a-z_]+) + .*\?.*\bwid=_(?P<partner_id_html5>\d+) + ) + ) + ) + ''' _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?' _TESTS = [ { @@ -43,6 +55,10 @@ class KalturaIE(InfoExtractor): 'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3', 'only_matching': True, }, + { + 'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342', + 'only_matching': True, + } ] def _kaltura_api_call(self, video_id, actions, *args, **kwargs): @@ -105,9 +121,9 @@ class KalturaIE(InfoExtractor): video_id, actions, note='Downloading video info JSON') def _real_extract(self, url): - video_id = self._match_id(url) mobj = re.match(self._VALID_URL, url) - partner_id, entry_id = mobj.group('partner_id'), mobj.group('id') + partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5') + entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5') info, source_data = self._get_video_info(entry_id, partner_id) @@ -126,7 +142,7 @@ class KalturaIE(InfoExtractor): self._sort_formats(formats) return { - 'id': video_id, + 'id': entry_id, 'title': info['name'], 'formats': formats, 'description': info.get('description'), diff --git a/youtube_dl/extractor/karaoketv.py b/youtube_dl/extractor/karaoketv.py index e3b43ff8d..06daf5a89 100644 --- a/youtube_dl/extractor/karaoketv.py +++ b/youtube_dl/extractor/karaoketv.py @@ -2,7 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_unquote_plus from ..utils import ( js_to_json, ) @@ -24,7 +24,7 @@ class KaraoketvIE(InfoExtractor): webpage = self._download_webpage(url, video_id) page_video_url = self._og_search_video_url(webpage, video_id) - config_json = compat_urllib_parse.unquote_plus(self._search_regex( + config_json = compat_urllib_parse_unquote_plus(self._search_regex( r'config=(.*)', page_video_url, 'configuration')) urls_info_json = self._download_json( diff --git a/youtube_dl/extractor/keek.py b/youtube_dl/extractor/keek.py index c0956ba09..94a03d277 100644 --- a/youtube_dl/extractor/keek.py +++ b/youtube_dl/extractor/keek.py @@ -1,46 +1,39 @@ +# coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor class KeekIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<id>\w+)' + _VALID_URL = r'https?://(?:www\.)?keek\.com/keek/(?P<id>\w+)' IE_NAME = 'keek' _TEST = { - 'url': 'https://www.keek.com/ytdl/keeks/NODfbab', - 'md5': '09c5c109067536c1cec8bac8c21fea05', + 'url': 'https://www.keek.com/keek/NODfbab', + 'md5': '9b0636f8c0f7614afa4ea5e4c6e57e83', 'info_dict': { 'id': 'NODfbab', 'ext': 'mp4', - 'uploader': 'youtube-dl project', - 'uploader_id': 'ytdl', - 'title': 'test chars: "\'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de .', + 'title': 'md5:35d42050a3ece241d5ddd7fdcc6fd896', + 'uploader': 'ytdl', + 'uploader_id': 'eGT5bab', }, } def _real_extract(self, url): video_id = self._match_id(url) - video_url = 'http://cdn.keek.com/keek/video/%s' % video_id - thumbnail = 'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id webpage = self._download_webpage(url, video_id) - raw_desc = self._html_search_meta('description', webpage) - if raw_desc: - uploader = self._html_search_regex( - r'Watch (.*?)\s+\(', raw_desc, 'uploader', fatal=False) - uploader_id = self._html_search_regex( - r'Watch .*?\(@(.+?)\)', raw_desc, 'uploader_id', fatal=False) - else: - uploader = None - uploader_id = None - return { 'id': video_id, - 'url': video_url, + 'url': self._og_search_video_url(webpage), 'ext': 'mp4', - 'title': self._og_search_title(webpage), - 'thumbnail': thumbnail, - 'uploader': uploader, - 'uploader_id': uploader_id, + 'title': self._og_search_description(webpage).strip(), + 'thumbnail': self._og_search_thumbnail(webpage), + 'uploader': self._search_regex( + r'data-username=(["\'])(?P<uploader>.+?)\1', webpage, + 'uploader', fatal=False, group='uploader'), + 'uploader_id': self._search_regex( + r'data-user-id=(["\'])(?P<uploader_id>.+?)\1', webpage, + 'uploader id', fatal=False, group='uploader_id'), } diff --git a/youtube_dl/extractor/kontrtube.py b/youtube_dl/extractor/kontrtube.py index 720bc939b..a59c529f4 100644 --- a/youtube_dl/extractor/kontrtube.py +++ b/youtube_dl/extractor/kontrtube.py @@ -4,7 +4,10 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import int_or_none +from ..utils import ( + int_or_none, + parse_duration, +) class KontrTubeIE(InfoExtractor): @@ -34,33 +37,28 @@ class KontrTubeIE(InfoExtractor): webpage = self._download_webpage( url, display_id, 'Downloading page') - video_url = self._html_search_regex( + video_url = self._search_regex( r"video_url\s*:\s*'(.+?)/?',", webpage, 'video URL') - thumbnail = self._html_search_regex( - r"preview_url\s*:\s*'(.+?)/?',", webpage, 'video thumbnail', fatal=False) + thumbnail = self._search_regex( + r"preview_url\s*:\s*'(.+?)/?',", webpage, 'thumbnail', fatal=False) title = self._html_search_regex( - r'<title>(.+?)</title>', webpage, 'video title') + r'(?s)<h2>(.+?)</h2>', webpage, 'title') description = self._html_search_meta( - 'description', webpage, 'video description') + 'description', webpage, 'description') - mobj = re.search( - r'<div class="col_2">Длительность: <span>(?P<minutes>\d+)м:(?P<seconds>\d+)с</span></div>', - webpage) - duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None + duration = self._search_regex( + r'Длительность: <em>([^<]+)</em>', webpage, 'duration', fatal=False) + if duration: + duration = parse_duration(duration.replace('мин', 'min').replace('сек', 'sec')) - view_count = self._html_search_regex( - r'<div class="col_2">Просмотров: <span>(\d+)</span></div>', + view_count = self._search_regex( + r'Просмотров: <em>([^<]+)</em>', webpage, 'view count', fatal=False) + if view_count: + view_count = int_or_none(view_count.replace(' ', '')) - comment_count = None - comment_str = self._html_search_regex( - r'Комментарии: <span>([^<]+)</span>', webpage, 'comment count', fatal=False) - if comment_str.startswith('комментариев нет'): - comment_count = 0 - else: - mobj = re.search(r'\d+ из (?P<total>\d+) комментариев', comment_str) - if mobj: - comment_count = mobj.group('total') + comment_count = int_or_none(self._search_regex( + r'Комментарии \((\d+)\)<', webpage, ' comment count', fatal=False)) return { 'id': video_id, diff --git a/youtube_dl/extractor/krasview.py b/youtube_dl/extractor/krasview.py index 96f95979a..0ae8ebd68 100644 --- a/youtube_dl/extractor/krasview.py +++ b/youtube_dl/extractor/krasview.py @@ -25,6 +25,9 @@ class KrasViewIE(InfoExtractor): 'duration': 27, 'thumbnail': 're:^https?://.*\.jpg', }, + 'params': { + 'skip_download': 'Not accessible from Travis CI server', + }, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/kuwo.py b/youtube_dl/extractor/kuwo.py index 1077846f2..0c8ed5d07 100644 --- a/youtube_dl/extractor/kuwo.py +++ b/youtube_dl/extractor/kuwo.py @@ -57,6 +57,7 @@ class KuwoIE(KuwoBaseIE): 'upload_date': '20080122', 'description': 'md5:ed13f58e3c3bf3f7fd9fbc4e5a7aa75c' }, + 'skip': 'this song has been offline because of copyright issues', }, { 'url': 'http://www.kuwo.cn/yinyue/6446136/', 'info_dict': { @@ -76,9 +77,11 @@ class KuwoIE(KuwoBaseIE): webpage = self._download_webpage( url, song_id, note='Download song detail info', errnote='Unable to get song detail info') + if '对不起，该歌曲由于版权问题已被下线，将返回网站首页' in webpage: + raise ExtractorError('this song has been offline because of copyright issues', expected=True) song_name = self._html_search_regex( - r'<h1[^>]+title="([^"]+)">', webpage, 'song name') + 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) @@ -202,6 +205,7 @@ class KuwoSingerIE(InfoExtractor): 'title': 'Ali', }, 'playlist_mincount': 95, + 'skip': 'Regularly stalls travis build', # See https://travis-ci.org/rg3/youtube-dl/jobs/78878540 }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/lecture2go.py b/youtube_dl/extractor/lecture2go.py new file mode 100644 index 000000000..40a3d2346 --- /dev/null +++ b/youtube_dl/extractor/lecture2go.py @@ -0,0 +1,62 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + parse_duration, + int_or_none, +) + + +class Lecture2GoIE(InfoExtractor): + _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)' + _TEST = { + 'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473', + 'md5': 'ac02b570883020d208d405d5a3fd2f7f', + 'info_dict': { + 'id': '17473', + 'ext': 'flv', + 'title': '2 - Endliche Automaten und reguläre Sprachen', + 'creator': 'Frank Heitmann', + 'duration': 5220, + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title') + + formats = [] + for url in set(re.findall(r'"src","([^"]+)"', webpage)): + ext = determine_ext(url) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats(url, video_id)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats(url, video_id)) + else: + formats.append({ + 'url': url, + }) + + self._sort_formats(formats) + + creator = self._html_search_regex( + r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False) + duration = parse_duration(self._html_search_regex( + r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False)) + view_count = int_or_none(self._html_search_regex( + r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False)) + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'creator': creator, + 'duration': duration, + 'view_count': view_count, + } diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index ba2ae8085..effd9eb92 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -9,12 +9,14 @@ from .common import InfoExtractor from ..compat import ( compat_urllib_parse, compat_urllib_request, - compat_urlparse, + compat_ord, ) from ..utils import ( determine_ext, ExtractorError, parse_iso8601, + int_or_none, + encode_data_uri, ) @@ -24,15 +26,16 @@ class LetvIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.letv.com/ptv/vplay/22005890.html', - 'md5': 'cab23bd68d5a8db9be31c9a222c1e8df', + 'md5': 'edadcfe5406976f42f9f266057ee5e40', 'info_dict': { 'id': '22005890', 'ext': 'mp4', 'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家', - 'timestamp': 1424747397, - 'upload_date': '20150224', 'description': 'md5:a9cb175fd753e2962176b7beca21a47c', - } + }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'url': 'http://www.letv.com/ptv/vplay/1415246.html', 'info_dict': { @@ -41,16 +44,22 @@ class LetvIE(InfoExtractor): 'title': '美人天下01', 'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda', }, + 'params': { + 'hls_prefer_native': True, + }, }, { 'note': 'This video is available only in Mainland China, thus a proxy is needed', 'url': 'http://www.letv.com/ptv/vplay/1118082.html', - 'md5': 'f80936fbe20fb2f58648e81386ff7927', + 'md5': '2424c74948a62e5f31988438979c5ad1', 'info_dict': { 'id': '1118082', 'ext': 'mp4', 'title': '与龙共舞 完整版', 'description': 'md5:7506a5eeb1722bb9d4068f85024e3986', }, + 'params': { + 'hls_prefer_native': True, + }, 'skip': 'Only available in China', }] @@ -73,6 +82,27 @@ class LetvIE(InfoExtractor): _loc3_ = self.ror(_loc3_, _loc2_ % 17) return _loc3_ + # see M3U8Encryption class in KLetvPlayer.swf + @staticmethod + def decrypt_m3u8(encrypted_data): + if encrypted_data[:5].decode('utf-8').lower() != 'vc_01': + return encrypted_data + encrypted_data = encrypted_data[5:] + + _loc4_ = bytearray() + while encrypted_data: + b = compat_ord(encrypted_data[0]) + _loc4_.extend([b // 16, b & 0x0f]) + encrypted_data = encrypted_data[1:] + idx = len(_loc4_) - 11 + _loc4_ = _loc4_[idx:] + _loc4_[:idx] + _loc7_ = bytearray() + while _loc4_: + _loc7_.append(_loc4_[0] * 16 + _loc4_[1]) + _loc4_ = _loc4_[2:] + + return bytes(_loc7_) + def _real_extract(self, url): media_id = self._match_id(url) page = self._download_webpage(url, media_id) @@ -114,27 +144,32 @@ class LetvIE(InfoExtractor): for format_id in formats: if format_id in dispatch: media_url = playurl['domain'][0] + dispatch[format_id][0] - - # Mimic what flvxz.com do - url_parts = list(compat_urlparse.urlparse(media_url)) - qs = dict(compat_urlparse.parse_qs(url_parts[4])) - qs.update({ - 'platid': '14', - 'splatid': '1401', - 'tss': 'no', - 'retry': 1 + media_url += '&' + compat_urllib_parse.urlencode({ + 'm3v': 1, + 'format': 1, + 'expect': 3, + 'rateid': format_id, }) - url_parts[4] = compat_urllib_parse.urlencode(qs) - media_url = compat_urlparse.urlunparse(url_parts) + + nodes_data = self._download_json( + media_url, media_id, + 'Download JSON metadata for format %s' % format_id) + + req = self._request_webpage( + nodes_data['nodelist'][0]['location'], media_id, + note='Downloading m3u8 information for format %s' % format_id) + + m3u8_data = self.decrypt_m3u8(req.read()) url_info_dict = { - 'url': media_url, + 'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'), 'ext': determine_ext(dispatch[format_id][1]), 'format_id': format_id, + 'protocol': 'm3u8', } if format_id[-1:] == 'p': - url_info_dict['height'] = format_id[:-1] + url_info_dict['height'] = int_or_none(format_id[:-1]) urls.append(url_info_dict) diff --git a/youtube_dl/extractor/libsyn.py b/youtube_dl/extractor/libsyn.py index 9ab1416f5..d375695f5 100644 --- a/youtube_dl/extractor/libsyn.py +++ b/youtube_dl/extractor/libsyn.py @@ -8,9 +8,9 @@ from ..utils import unified_strdate class LibsynIE(InfoExtractor): - _VALID_URL = r'https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+)' + _VALID_URL = r'(?P<mainurl>https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+))' - _TEST = { + _TESTS = [{ 'url': 'http://html5-player.libsyn.com/embed/episode/id/3377616/', 'md5': '443360ee1b58007bc3dcf09b41d093bb', 'info_dict': { @@ -19,12 +19,24 @@ class LibsynIE(InfoExtractor): 'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart", 'description': 'md5:601cb790edd05908957dae8aaa866465', 'upload_date': '20150220', + 'thumbnail': 're:^https?://.*', }, - } + }, { + 'url': 'https://html5-player.libsyn.com/embed/episode/id/3727166/height/75/width/200/theme/standard/direction/no/autoplay/no/autonext/no/thumbnail/no/preload/no/no_addthis/no/', + 'md5': '6c5cb21acd622d754d3b1a92b582ce42', + 'info_dict': { + 'id': '3727166', + 'ext': 'mp3', + 'title': 'Clients From Hell Podcast - How a Sex Toy Company Kickstarted my Freelance Career', + 'upload_date': '20150818', + 'thumbnail': 're:^https?://.*', + } + }] def _real_extract(self, url): - video_id = self._match_id(url) - + m = re.match(self._VALID_URL, url) + video_id = m.group('id') + url = m.group('mainurl') webpage = self._download_webpage(url, video_id) formats = [{ @@ -32,20 +44,18 @@ class LibsynIE(InfoExtractor): } for media_url in set(re.findall('var\s+mediaURL(?:Libsyn)?\s*=\s*"([^"]+)"', webpage))] podcast_title = self._search_regex( - r'<h2>([^<]+)</h2>', webpage, 'title') + r'<h2>([^<]+)</h2>', webpage, 'podcast title', default=None) episode_title = self._search_regex( - r'<h3>([^<]+)</h3>', webpage, 'title', default=None) + r'(?:<div class="episode-title">|<h3>)([^<]+)</', webpage, 'episode title') title = '%s - %s' % (podcast_title, episode_title) if podcast_title else episode_title description = self._html_search_regex( r'<div id="info_text_body">(.+?)</div>', webpage, - 'description', fatal=False) - + 'description', default=None) thumbnail = self._search_regex( r'<img[^>]+class="info-show-icon"[^>]+src="([^"]+)"', webpage, 'thumbnail', fatal=False) - release_date = unified_strdate(self._search_regex( r'<div class="release_date">Released: ([^<]+)<', webpage, 'release date', fatal=False)) diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py new file mode 100644 index 000000000..fb03dd527 --- /dev/null +++ b/youtube_dl/extractor/limelight.py @@ -0,0 +1,229 @@ +# 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, +) + + +class LimelightBaseIE(InfoExtractor): + _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s' + _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json' + + def _call_playlist_service(self, item_id, method, fatal=True): + return self._download_json( + self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method), + item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal) + + def _call_api(self, organization_id, item_id, method): + return self._download_json( + self._API_URL % (organization_id, self._API_PATH, item_id, method), + item_id, 'Downloading API %s JSON' % method) + + def _extract(self, item_id, pc_method, mobile_method, meta_method): + pc = self._call_playlist_service(item_id, pc_method) + metadata = self._call_api(pc['orgId'], item_id, meta_method) + mobile = self._call_playlist_service(item_id, mobile_method, fatal=False) + return pc, mobile, metadata + + def _extract_info(self, streams, mobile_urls, properties): + video_id = properties['media_id'] + formats = [] + + for stream in streams: + stream_url = stream.get('url') + if not stream_url: + continue + if '.f4m' in stream_url: + formats.extend(self._extract_f4m_formats(stream_url, video_id)) + else: + fmt = { + 'url': stream_url, + 'abr': float_or_none(stream.get('audioBitRate')), + 'vbr': float_or_none(stream.get('videoBitRate')), + 'fps': float_or_none(stream.get('videoFrameRate')), + 'width': int_or_none(stream.get('videoWidthInPixels')), + 'height': int_or_none(stream.get('videoHeightInPixels')), + 'ext': determine_ext(stream_url) + } + rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url) + if rtmp: + format_id = 'rtmp' + if stream.get('videoBitRate'): + format_id += '-%d' % int_or_none(stream['videoBitRate']) + fmt.update({ + 'url': rtmp.group('url'), + 'play_path': rtmp.group('playpath'), + 'app': rtmp.group('app'), + 'ext': 'flv', + 'format_id': format_id, + }) + formats.append(fmt) + + for mobile_url in mobile_urls: + media_url = mobile_url.get('mobileUrl') + if not media_url: + continue + format_id = mobile_url.get('targetMediaPlatform') + if determine_ext(media_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', entry_protocol='m3u8_native', + preference=-1, m3u8_id=format_id)) + else: + formats.append({ + 'url': media_url, + 'format_id': format_id, + 'preference': -1, + }) + + self._sort_formats(formats) + + title = properties['title'] + description = properties.get('description') + timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date')) + duration = float_or_none(properties.get('duration_in_milliseconds'), 1000) + filesize = int_or_none(properties.get('total_storage_in_bytes')) + categories = [properties.get('category')] + tags = properties.get('tags', []) + thumbnails = [{ + 'url': thumbnail['url'], + 'width': int_or_none(thumbnail.get('width')), + 'height': int_or_none(thumbnail.get('height')), + } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')] + + subtitles = {} + for caption in properties.get('captions', {}): + lang = caption.get('language_code') + subtitles_url = caption.get('url') + if lang and subtitles_url: + subtitles[lang] = [{ + 'url': subtitles_url, + }] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'formats': formats, + 'timestamp': timestamp, + 'duration': duration, + 'filesize': filesize, + 'categories': categories, + 'tags': tags, + 'thumbnails': thumbnails, + 'subtitles': subtitles, + } + + +class LimelightMediaIE(LimelightBaseIE): + IE_NAME = 'limelight' + _VALID_URL = r'(?:limelight:media:|http://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})' + _TESTS = [{ + 'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86', + 'info_dict': { + 'id': '3ffd040b522b4485b6d84effc750cd86', + 'ext': 'flv', + 'title': 'HaP and the HB Prince Trailer', + 'description': 'md5:8005b944181778e313d95c1237ddb640', + 'thumbnail': 're:^https?://.*\.jpeg$', + 'duration': 144.23, + 'timestamp': 1244136834, + 'upload_date': '20090604', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # video with subtitles + 'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335', + 'info_dict': { + 'id': 'a3e00274d4564ec4a9b29b9466432335', + 'ext': 'flv', + 'title': '3Play Media Overview Video', + 'description': '', + 'thumbnail': 're:^https?://.*\.jpeg$', + 'duration': 78.101, + 'timestamp': 1338929955, + 'upload_date': '20120605', + 'subtitles': 'mincount:9', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }] + _PLAYLIST_SERVICE_PATH = 'media' + _API_PATH = 'media' + + def _real_extract(self, url): + video_id = self._match_id(url) + + pc, mobile, metadata = self._extract( + video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties') + + return self._extract_info( + pc['playlistItems'][0].get('streams', []), + mobile['mediaList'][0].get('mobileUrls', []) if mobile else [], + metadata) + + +class LimelightChannelIE(LimelightBaseIE): + IE_NAME = 'limelight:channel' + _VALID_URL = r'(?:limelight:channel:|http://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})' + _TEST = { + 'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082', + 'info_dict': { + 'id': 'ab6a524c379342f9b23642917020c082', + 'title': 'Javascript Sample Code', + }, + 'playlist_mincount': 3, + } + _PLAYLIST_SERVICE_PATH = 'channel' + _API_PATH = 'channels' + + def _real_extract(self, url): + channel_id = self._match_id(url) + + pc, mobile, medias = self._extract( + channel_id, 'getPlaylistByChannelId', + 'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media') + + entries = [ + self._extract_info( + pc['playlistItems'][i].get('streams', []), + mobile['mediaList'][i].get('mobileUrls', []) if mobile else [], + medias['media_list'][i]) + for i in range(len(medias['media_list']))] + + return self.playlist_result(entries, channel_id, pc['title']) + + +class LimelightChannelListIE(LimelightBaseIE): + IE_NAME = 'limelight:channel_list' + _VALID_URL = r'(?:limelight:channel_list:|http://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})' + _TEST = { + 'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b', + 'info_dict': { + 'id': '301b117890c4465c8179ede21fd92e2b', + 'title': 'Website - Hero Player', + }, + 'playlist_mincount': 2, + } + _PLAYLIST_SERVICE_PATH = 'channel_list' + + def _real_extract(self, url): + channel_list_id = self._match_id(url) + + channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById') + + entries = [ + self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel') + for channel in channel_list['channelList']] + + return self.playlist_result(entries, channel_list_id, channel_list['title']) diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index a00f6e5e5..5c973e75c 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -11,13 +11,13 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + clean_html, int_or_none, ) class LyndaBaseIE(InfoExtractor): _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' - _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true' _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' _NETRC_MACHINE = 'lynda' @@ -41,7 +41,7 @@ class LyndaBaseIE(InfoExtractor): request, None, 'Logging in as %s' % username) # Not (yet) logged in - m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page) + 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) @@ -70,7 +70,16 @@ class LyndaBaseIE(InfoExtractor): request, None, 'Confirming log in and log out from another device') - if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None: + 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') @@ -109,9 +118,7 @@ class LyndaIE(LyndaBaseIE): 'lynda returned error: %s' % video_json['Message'], expected=True) if video_json['HasAccess'] is False: - raise ExtractorError( - 'Video %s is only available for members. ' - % video_id + self._ACCOUNT_CREDENTIALS_HINT, expected=True) + self.raise_login_required('Video %s is only available for members' % video_id) video_id = compat_str(video_json['ID']) duration = video_json['DurationInSeconds'] @@ -133,13 +140,14 @@ class LyndaIE(LyndaBaseIE): prioritized_streams = video_json.get('PrioritizedStreams') if prioritized_streams: - formats.extend([ - { - 'url': video_url, - 'width': int_or_none(format_id), - 'format_id': format_id, - } for format_id, video_url in prioritized_streams['0'].items() - ]) + for prioritized_stream_id, prioritized_stream in prioritized_streams.items(): + formats.extend([ + { + 'url': video_url, + 'width': int_or_none(format_id), + 'format_id': '%s-%s' % (prioritized_stream_id, format_id), + } for format_id, video_url in prioritized_stream.items() + ]) self._check_formats(formats, video_id) self._sort_formats(formats) diff --git a/youtube_dl/extractor/mailru.py b/youtube_dl/extractor/mailru.py index 54a14cb94..ab1300185 100644 --- a/youtube_dl/extractor/mailru.py +++ b/youtube_dl/extractor/mailru.py @@ -25,6 +25,7 @@ class MailRuIE(InfoExtractor): 'uploader_id': 'sonypicturesrus@mail.ru', 'duration': 184, }, + 'skip': 'Not accessible from Travis CI server', }, { 'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html', @@ -39,6 +40,7 @@ class MailRuIE(InfoExtractor): 'uploader_id': 'hitech@corp.mail.ru', 'duration': 245, }, + 'skip': 'Not accessible from Travis CI server', }, ] diff --git a/youtube_dl/extractor/malemotion.py b/youtube_dl/extractor/malemotion.py index 0b85a59d1..92511a671 100644 --- a/youtube_dl/extractor/malemotion.py +++ b/youtube_dl/extractor/malemotion.py @@ -2,9 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, -) +from ..compat import compat_urllib_parse_unquote class MalemotionIE(InfoExtractor): @@ -24,7 +22,7 @@ class MalemotionIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_url = compat_urllib_parse.unquote(self._search_regex( + 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') diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py index 5fdd19027..fc7499958 100644 --- a/youtube_dl/extractor/mdr.py +++ b/youtube_dl/extractor/mdr.py @@ -29,7 +29,7 @@ class MDRIE(InfoExtractor): doc = self._download_xml(domain + xmlurl, video_id) formats = [] for a in doc.findall('./assets/asset'): - url_el = a.find('.//progressiveDownloadUrl') + url_el = a.find('./progressiveDownloadUrl') if url_el is None: continue abr = int(a.find('bitrateAudio').text) // 1000 diff --git a/youtube_dl/extractor/megavideoz.py b/youtube_dl/extractor/megavideoz.py deleted file mode 100644 index af7ff07ea..000000000 --- a/youtube_dl/extractor/megavideoz.py +++ /dev/null @@ -1,56 +0,0 @@ -# encoding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - float_or_none, - xpath_text, -) - - -class MegaVideozIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?megavideoz\.eu/video/(?P<id>[^/]+)(?:/(?P<display_id>[^/]+))?' - _TEST = { - 'url': 'http://megavideoz.eu/video/WM6UB919XMXH/SMPTE-Universal-Film-Leader', - 'info_dict': { - 'id': '48723', - 'display_id': 'SMPTE-Universal-Film-Leader', - 'ext': 'mp4', - 'title': 'SMPTE Universal Film Leader', - 'thumbnail': 're:https?://.*?\.jpg', - 'duration': 10.93, - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') or video_id - - webpage = self._download_webpage(url, display_id) - - if any(p in webpage for p in ('>Video Not Found<', '>404 Error<')): - raise ExtractorError('Video %s does not exist' % video_id, expected=True) - - config = self._download_xml( - self._search_regex( - r"var\s+cnf\s*=\s*'([^']+)'", webpage, 'cnf url'), - display_id) - - video_url = xpath_text(config, './file', 'video url', fatal=True) - title = xpath_text(config, './title', 'title', fatal=True) - thumbnail = xpath_text(config, './image', 'thumbnail') - duration = float_or_none(xpath_text(config, './duration', 'duration')) - video_id = xpath_text(config, './mediaid', 'video id') or video_id - - return { - 'id': video_id, - 'display_id': display_id, - 'url': video_url, - 'title': title, - 'thumbnail': thumbnail, - 'duration': duration - } diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py index 8bc333b02..6e2e73a51 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -6,6 +6,7 @@ from .common import InfoExtractor from ..compat import ( compat_parse_qs, compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_request, ) from ..utils import ( @@ -155,7 +156,7 @@ class MetacafeIE(InfoExtractor): video_url = None mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage) if mobj is not None: - mediaURL = compat_urllib_parse.unquote(mobj.group(1)) + mediaURL = compat_urllib_parse_unquote(mobj.group(1)) video_ext = mediaURL[-3:] # Extract gdaKey if available diff --git a/youtube_dl/extractor/mit.py b/youtube_dl/extractor/mit.py index d7ab6a9ae..f088ab9e2 100644 --- a/youtube_dl/extractor/mit.py +++ b/youtube_dl/extractor/mit.py @@ -18,12 +18,12 @@ class TechTVMITIE(InfoExtractor): _TEST = { 'url': 'http://techtv.mit.edu/videos/25418-mit-dna-learning-center-set', - 'md5': '1f8cb3e170d41fd74add04d3c9330e5f', + 'md5': '00a3a27ee20d44bcaa0933ccec4a2cf7', 'info_dict': { 'id': '25418', 'ext': 'mp4', - 'title': 'MIT DNA Learning Center Set', - 'description': 'md5:82313335e8a8a3f243351ba55bc1b474', + 'title': 'MIT DNA and Protein Sets', + 'description': 'md5:46f5c69ce434f0a97e7c628cc142802d', }, } @@ -33,8 +33,8 @@ class TechTVMITIE(InfoExtractor): 'http://techtv.mit.edu/videos/%s' % video_id, video_id) clean_page = re.compile(r'<!--.*?-->', re.S).sub('', raw_page) - base_url = self._search_regex( - r'ipadUrl: \'(.+?cloudfront.net/)', raw_page, 'base url') + base_url = self._proto_relative_url(self._search_regex( + r'ipadUrl: \'(.+?cloudfront.net/)', raw_page, 'base url'), 'http:') formats_json = self._search_regex( r'bitrates: (\[.+?\])', raw_page, 'video formats') formats_mit = json.loads(formats_json) diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py index 7091f3335..3142fcde2 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -1,73 +1,85 @@ from __future__ import unicode_literals -import json - from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, - compat_urlparse, -) +from ..compat import compat_urllib_parse from ..utils import ( + encode_dict, get_element_by_attribute, - parse_duration, - strip_jsonp, + int_or_none, ) class MiTeleIE(InfoExtractor): - IE_NAME = 'mitele.es' + IE_DESC = 'mitele.es' _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/' _TESTS = [{ 'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/', + 'md5': '757b0b66cbd7e0a97226d7d3156cb3e9', 'info_dict': { - 'id': '0fce117d', - 'ext': 'mp4', - 'title': 'Programa 144 - Tor, la web invisible', - 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', + 'id': '0NF1jJnxS1Wu3pHrmvFyw2', 'display_id': 'programa-144', + 'ext': 'flv', + 'title': 'Tor, la web invisible', + 'description': 'md5:3b6fce7eaa41b2d97358726378d9369f', + 'thumbnail': 're:(?i)^https?://.*\.jpg$', 'duration': 2913, }, - 'params': { - # m3u8 download - 'skip_download': True, - }, }] 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) + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) - 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') + config_url = self._search_regex( + r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url') - 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') + 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(encode_dict(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)) + + title = self._search_regex( + r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title') + + 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')) return { - 'id': embed_data['videoId'], - 'display_id': episode, - 'title': info_el.find('title').text, - 'formats': formats, + 'id': video_id, + 'display_id': display_id, + 'title': title, 'description': get_element_by_attribute('class', 'text', webpage), - 'thumbnail': info_el.find('thumb').text, - 'duration': parse_duration(info_el.find('duration').text), + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 425a4ccf1..d47aeceda 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -3,9 +3,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, -) +from ..compat import compat_urllib_parse_unquote from ..utils import ( ExtractorError, HEADRequest, @@ -60,7 +58,7 @@ class MixcloudIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) uploader = mobj.group(1) cloudcast_name = mobj.group(2) - track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name))) + track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name))) webpage = self._download_webpage(url, track_id) diff --git a/youtube_dl/extractor/mofosex.py b/youtube_dl/extractor/mofosex.py index 2cec12d35..9bf99a54a 100644 --- a/youtube_dl/extractor/mofosex.py +++ b/youtube_dl/extractor/mofosex.py @@ -5,9 +5,9 @@ import re from .common import InfoExtractor from ..compat import ( + compat_urllib_parse_unquote, compat_urllib_parse_urlparse, compat_urllib_request, - compat_urllib_parse, ) @@ -34,7 +34,7 @@ class MofosexIE(InfoExtractor): webpage = self._download_webpage(req, video_id) video_title = self._html_search_regex(r'<h1>(.+?)<', webpage, 'title') - video_url = compat_urllib_parse.unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, 'video_url')) + video_url = compat_urllib_parse_unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, 'video_url')) path = compat_urllib_parse_urlparse(video_url).path extension = os.path.splitext(path)[1][1:] format = path.split('/')[5].split('_')[:2] diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 88dcd4f73..7c0c4e50e 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -9,12 +9,15 @@ from ..compat import ( compat_urllib_parse, compat_urllib_request, ) -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + remove_start, +) class MonikerIE(InfoExtractor): IE_DESC = 'allmyvideos.net and vidspot.net' - _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?P<id>[a-zA-Z0-9_-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:allmyvideos|vidspot)\.net/(?:(?:2|v)/v-)?(?P<id>[a-zA-Z0-9_-]+)' _TESTS = [{ 'url': 'http://allmyvideos.net/jih3nce3x6wn', @@ -25,6 +28,14 @@ class MonikerIE(InfoExtractor): 'title': 'youtube-dl test video', }, }, { + 'url': 'http://allmyvideos.net/embed-jih3nce3x6wn', + 'md5': '710883dee1bfc370ecf9fa6a89307c88', + 'info_dict': { + 'id': 'jih3nce3x6wn', + 'ext': 'mp4', + 'title': 'youtube-dl test video', + }, + }, { 'url': 'http://vidspot.net/l2ngsmhs8ci5', 'md5': '710883dee1bfc370ecf9fa6a89307c88', 'info_dict': { @@ -35,10 +46,25 @@ class MonikerIE(InfoExtractor): }, { 'url': 'https://www.vidspot.net/l2ngsmhs8ci5', 'only_matching': True, + }, { + 'url': 'http://vidspot.net/2/v-ywDf99', + 'md5': '5f8254ce12df30479428b0152fb8e7ba', + 'info_dict': { + 'id': 'ywDf99', + 'ext': 'mp4', + 'title': 'IL FAIT LE MALIN EN PORSHE CAYENNE ( mais pas pour longtemps)', + 'description': 'IL FAIT LE MALIN EN PORSHE CAYENNE.', + }, + }, { + 'url': 'http://allmyvideos.net/v/v-HXZm5t', + 'only_matching': True, }] def _real_extract(self, url): - video_id = self._match_id(url) + orig_video_id = self._match_id(url) + video_id = remove_start(orig_video_id, 'embed-') + url = url.replace(orig_video_id, video_id) + assert re.match(self._VALID_URL, url) is not None orig_webpage = self._download_webpage(url, video_id) if '>File Not Found<' in orig_webpage: @@ -50,18 +76,30 @@ class MonikerIE(InfoExtractor): raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error), expected=True) - fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) - data = dict(fields) + builtin_url = self._search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+?/builtin-.+?)\1', + orig_webpage, 'builtin URL', default=None, group='url') - post = compat_urllib_parse.urlencode(data) - 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='Downloading video page ...') + if builtin_url: + req = compat_urllib_request.Request(builtin_url) + req.add_header('Referer', url) + webpage = self._download_webpage(req, video_id, 'Downloading builtin page') + title = self._og_search_title(orig_webpage).strip() + description = self._og_search_description(orig_webpage).strip() + else: + fields = re.findall(r'type="hidden" name="(.+?)"\s* value="?(.+?)">', orig_webpage) + data = dict(fields) + + post = compat_urllib_parse.urlencode(data) + 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='Downloading video page ...') - title = os.path.splitext(data['fname'])[0] + title = os.path.splitext(data['fname'])[0] + description = None # Could be several links with different quality links = re.findall(r'"file" : "?(.+?)",', webpage) @@ -75,5 +113,6 @@ class MonikerIE(InfoExtractor): return { 'id': video_id, 'title': title, + 'description': description, 'formats': formats, } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index b48fac5e3..302c9bf35 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -67,7 +67,7 @@ class MTVServicesInfoExtractor(InfoExtractor): return [{'url': url, 'ext': 'mp4'}] def _extract_video_formats(self, mdoc, mtvn_id): - if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None: + if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None: if mtvn_id is not None and self._MOBILE_TEMPLATE is not None: self.to_screen('The normal version is not available from your ' 'country, trying with the mobile version') @@ -114,7 +114,8 @@ class MTVServicesInfoExtractor(InfoExtractor): # Remove the templates, like &device={device} mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url) if 'acceptMethods' not in mediagen_url: - mediagen_url += '&acceptMethods=fms' + mediagen_url += '&' if '?' in mediagen_url else '?' + mediagen_url += 'acceptMethods=fms' mediagen_doc = self._download_xml(mediagen_url, video_id, 'Downloading video urls') @@ -141,7 +142,7 @@ class MTVServicesInfoExtractor(InfoExtractor): if title_el is None: title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title') if title_el is None: - title_el = itemdoc.find('.//title') + title_el = itemdoc.find('.//title') or itemdoc.find('./title') if title_el.text is None: title_el = None @@ -174,8 +175,11 @@ class MTVServicesInfoExtractor(InfoExtractor): if self._LANG: info_url += 'lang=%s&' % self._LANG info_url += data + return self._get_videos_info_from_url(info_url, video_id) + + def _get_videos_info_from_url(self, url, video_id): idoc = self._download_xml( - info_url, video_id, + url, video_id, 'Downloading info', transform_source=fix_xml_ampersands) return self.playlist_result( [self._get_video_info(item) for item in idoc.findall('.//item')]) @@ -196,7 +200,13 @@ class MTVServicesInfoExtractor(InfoExtractor): if mgid is None or ':' not in mgid: mgid = self._search_regex( [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'], - webpage, 'mgid') + webpage, 'mgid', default=None) + + if not mgid: + sm4_embed = self._html_search_meta( + 'sm4:video:embed', webpage, 'sm4 embed', default='') + mgid = self._search_regex( + r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid') videos_info = self._get_videos_info(mgid) return videos_info @@ -218,6 +228,13 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): }, } + @staticmethod + def _extract_url(webpage): + mobj = re.search( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) + if mobj: + return mobj.group('url') + def _get_feed_url(self, uri): video_id = self._id_from_uri(uri) site_id = uri.replace(video_id, '') @@ -288,3 +305,65 @@ class MTVIggyIE(MTVServicesInfoExtractor): } } _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/' + + +class MTVDEIE(MTVServicesInfoExtractor): + IE_NAME = 'mtv.de' + _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$' + _TESTS = [{ + 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum', + 'info_dict': { + 'id': 'music_video-a50bc5f0b3aa4b3190aa', + 'ext': 'mp4', + 'title': 'MusicVideo_cro-traum', + 'description': 'Cro - Traum', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97) + 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen', + 'info_dict': { + 'id': 'local_playlist-f5ae778b9832cc837189', + 'ext': 'mp4', + 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + # single video in pagePlaylist with different id + 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3', + 'info_dict': { + 'id': 'local_playlist-4e760566473c4c8c5344', + 'ext': 'mp4', + 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1', + 'description': 'MTV Movies Supercut', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + playlist = self._parse_json( + self._search_regex( + r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'), + video_id) + + # news pages contain single video in playlist with different id + if len(playlist) == 1: + return self._get_videos_info_from_url(playlist[0]['mrss'], video_id) + + for item in playlist: + item_id = item.get('id') + if item_id and compat_str(item_id) == video_id: + return self._get_videos_info_from_url(item['mrss'], video_id) diff --git a/youtube_dl/extractor/musicvault.py b/youtube_dl/extractor/musicvault.py deleted file mode 100644 index 0e46ac7c1..000000000 --- a/youtube_dl/extractor/musicvault.py +++ /dev/null @@ -1,63 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor - - -class MusicVaultIE(InfoExtractor): - _VALID_URL = r'https?://www\.musicvault\.com/(?P<uploader_id>[^/?#]*)/video/(?P<display_id>[^/?#]*)_(?P<id>[0-9]+)\.html' - _TEST = { - 'url': 'http://www.musicvault.com/the-allman-brothers-band/video/straight-from-the-heart_1010863.html', - 'md5': '3adcbdb3dcc02d647539e53f284ba171', - 'info_dict': { - 'id': '1010863', - 'ext': 'mp4', - 'uploader_id': 'the-allman-brothers-band', - 'title': 'Straight from the Heart', - 'duration': 244, - 'uploader': 'The Allman Brothers Band', - 'thumbnail': 're:^https?://.*/thumbnail/.*', - 'upload_date': '20131219', - 'location': 'Capitol Theatre (Passaic, NJ)', - 'description': 'Listen to The Allman Brothers Band perform Straight from the Heart at Capitol Theatre (Passaic, NJ) on Dec 16, 1981', - 'timestamp': int, - } - } - - 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) - - thumbnail = self._search_regex( - r'<meta itemprop="thumbnail" content="([^"]+)"', - webpage, 'thumbnail', fatal=False) - - data_div = self._search_regex( - r'(?s)<div class="data">(.*?)</div>', webpage, 'data fields') - uploader = self._html_search_regex( - r'<h1.*?>(.*?)</h1>', data_div, 'uploader', fatal=False) - title = self._html_search_regex( - r'<h2.*?>(.*?)</h2>', data_div, 'title') - location = self._html_search_regex( - r'<h4.*?>(.*?)</h4>', data_div, 'location', fatal=False) - - kaltura_id = self._search_regex( - r'<div id="video-detail-player" data-kaltura-id="([^"]+)"', - webpage, 'kaltura ID') - wid = self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid') - - return { - 'id': mobj.group('id'), - '_type': 'url_transparent', - 'url': 'kaltura:%s:%s' % (wid, kaltura_id), - 'ie_key': 'Kaltura', - 'display_id': display_id, - 'uploader_id': mobj.group('uploader_id'), - 'thumbnail': thumbnail, - 'description': self._html_search_meta('description', webpage), - 'location': location, - 'title': title, - 'uploader': uploader, - } diff --git a/youtube_dl/extractor/mwave.py b/youtube_dl/extractor/mwave.py new file mode 100644 index 000000000..66b523197 --- /dev/null +++ b/youtube_dl/extractor/mwave.py @@ -0,0 +1,58 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + int_or_none, + parse_duration, +) + + +class MwaveIE(InfoExtractor): + _VALID_URL = r'https?://mwave\.interest\.me/mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859', + 'md5': 'c930e27b7720aaa3c9d0018dfc8ff6cc', + 'info_dict': { + 'id': '168859', + 'ext': 'flv', + 'title': '[M COUNTDOWN] SISTAR - SHAKE IT', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'M COUNTDOWN', + 'duration': 206, + 'view_count': int, + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + vod_info = self._download_json( + 'http://mwave.interest.me/onair/vod_info.m?vodtype=CL§orid=&endinfo=Y&id=%s' % video_id, + video_id, 'Download vod JSON') + + formats = [] + for num, cdn_info in enumerate(vod_info['cdn']): + stream_url = cdn_info.get('url') + if not stream_url: + continue + stream_name = cdn_info.get('name') or compat_str(num) + f4m_stream = self._download_json( + stream_url, video_id, + 'Download %s stream JSON' % stream_name) + f4m_url = f4m_stream.get('fileurl') + if not f4m_url: + continue + formats.extend( + self._extract_f4m_formats(f4m_url + '&hdcore=3.0.3', video_id, f4m_id=stream_name)) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': vod_info['title'], + 'thumbnail': vod_info.get('cover'), + 'uploader': vod_info.get('program_title'), + 'duration': parse_duration(vod_info.get('time')), + 'view_count': int_or_none(vod_info.get('hit')), + 'formats': formats, + } diff --git a/youtube_dl/extractor/myspass.py b/youtube_dl/extractor/myspass.py index 5b9b9fbcd..4557a2b13 100644 --- a/youtube_dl/extractor/myspass.py +++ b/youtube_dl/extractor/myspass.py @@ -35,7 +35,8 @@ class MySpassIE(InfoExtractor): # get metadata metadata_url = META_DATA_URL_TEMPLATE % video_id - metadata = self._download_xml(metadata_url, video_id) + metadata = self._download_xml( + metadata_url, video_id, transform_source=lambda s: s.strip()) # extract values from metadata url_flv_el = metadata.find('url_flv') diff --git a/youtube_dl/extractor/myvideo.py b/youtube_dl/extractor/myvideo.py index 5e754fcff..c96f472a3 100644 --- a/youtube_dl/extractor/myvideo.py +++ b/youtube_dl/extractor/myvideo.py @@ -10,6 +10,7 @@ from .common import InfoExtractor from ..compat import ( compat_ord, compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_request, ) from ..utils import ( @@ -107,7 +108,7 @@ class MyVideoIE(InfoExtractor): if not a == '_encxml': params[a] = b else: - encxml = compat_urllib_parse.unquote(b) + encxml = compat_urllib_parse_unquote(b) if not params.get('domain'): params['domain'] = 'www.myvideo.de' xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params)) @@ -135,7 +136,7 @@ class MyVideoIE(InfoExtractor): video_url = None mobj = re.search('connectionurl=\'(.*?)\'', dec_data) if mobj: - video_url = compat_urllib_parse.unquote(mobj.group(1)) + video_url = compat_urllib_parse_unquote(mobj.group(1)) if 'myvideo2flash' in video_url: self.report_warning( 'Rewriting URL to use unencrypted rtmp:// ...', @@ -147,10 +148,10 @@ class MyVideoIE(InfoExtractor): mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data) if mobj is None: raise ExtractorError('unable to extract url') - video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2)) + video_url = compat_urllib_parse_unquote(mobj.group(1)) + compat_urllib_parse_unquote(mobj.group(2)) video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file') - video_file = compat_urllib_parse.unquote(video_file) + video_file = compat_urllib_parse_unquote(video_file) if not video_file.endswith('f4m'): ppath, prefix = video_file.split('.') @@ -159,7 +160,7 @@ class MyVideoIE(InfoExtractor): video_playpath = '' video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj') - video_swfobj = compat_urllib_parse.unquote(video_swfobj) + video_swfobj = compat_urllib_parse_unquote(video_swfobj) video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>", webpage, 'title') diff --git a/youtube_dl/extractor/nationalgeographic.py b/youtube_dl/extractor/nationalgeographic.py index c18640c5a..6fc9e7b05 100644 --- a/youtube_dl/extractor/nationalgeographic.py +++ b/youtube_dl/extractor/nationalgeographic.py @@ -8,25 +8,40 @@ from ..utils import ( class NationalGeographicIE(InfoExtractor): - _VALID_URL = r'http://video\.nationalgeographic\.com/video/.*?' - - _TEST = { - 'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo', - 'info_dict': { - 'id': '4DmDACA6Qtk_', - 'ext': 'flv', - 'title': 'Mating Crabs Busted by Sharks', - 'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3', + _VALID_URL = r'http://video\.nationalgeographic\.com/.*?' + + _TESTS = [ + { + 'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo', + 'info_dict': { + 'id': '4DmDACA6Qtk_', + 'ext': 'flv', + 'title': 'Mating Crabs Busted by Sharks', + 'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3', + }, + 'add_ie': ['ThePlatform'], }, - 'add_ie': ['ThePlatform'], - } + { + 'url': 'http://video.nationalgeographic.com/wild/when-sharks-attack/the-real-jaws', + 'info_dict': { + 'id': '_JeBD_D7PlS5', + 'ext': 'flv', + 'title': 'The Real Jaws', + 'description': 'md5:8d3e09d9d53a85cd397b4b21b2c77be6', + }, + 'add_ie': ['ThePlatform'], + }, + ] def _real_extract(self, url): name = url_basename(url) webpage = self._download_webpage(url, name) - feed_url = self._search_regex(r'data-feed-url="([^"]+)"', webpage, 'feed url') - guid = self._search_regex(r'data-video-guid="([^"]+)"', webpage, 'guid') + feed_url = self._search_regex( + r'data-feed-url="([^"]+)"', webpage, 'feed url') + guid = self._search_regex( + r'id="(?:videoPlayer|player-container)"[^>]+data-guid="([^"]+)"', + webpage, 'guid') feed = self._download_xml('%s?byGuid=%s' % (feed_url, guid), name) content = feed.find('.//{http://search.yahoo.com/mrss/}content') @@ -34,5 +49,6 @@ class NationalGeographicIE(InfoExtractor): return self.url_result(smuggle_url( 'http://link.theplatform.com/s/ngs/%s?format=SMIL&formats=MPEG4&manifest=f4m' % theplatform_id, - # For some reason, the normal links don't work and we must force the use of f4m + # For some reason, the normal links don't work and we must force + # the use of f4m {'force_smil_url': True})) diff --git a/youtube_dl/extractor/naver.py b/youtube_dl/extractor/naver.py index 925967753..1f5fc2145 100644 --- a/youtube_dl/extractor/naver.py +++ b/youtube_dl/extractor/naver.py @@ -10,7 +10,6 @@ from ..compat import ( ) from ..utils import ( ExtractorError, - clean_html, ) @@ -46,11 +45,11 @@ class NaverIE(InfoExtractor): m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"', webpage) if m_id is None: - m_error = re.search( - r'(?s)<div class="(?:nation_error|nation_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>', - webpage) - if m_error: - raise ExtractorError(clean_html(m_error.group('msg')), expected=True) + error = self._html_search_regex( + r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>', + webpage, 'error', default=None) + if error: + raise ExtractorError(error, expected=True) raise ExtractorError('couldn\'t extract vid and key') vid = m_id.group(1) key = m_id.group(2) diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index dc2091be0..e683d24c4 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -124,7 +124,7 @@ class NBCSportsIE(InfoExtractor): class NBCNewsIE(InfoExtractor): _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/ (?:video/.+?/(?P<id>\d+)| - (?:feature|nightly-news)/[^/]+/(?P<title>.+)) + (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+)) ''' _TESTS = [ @@ -169,6 +169,10 @@ class NBCNewsIE(InfoExtractor): 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5', }, }, + { + 'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952', + 'only_matching': True, + }, ] def _real_extract(self, url): @@ -232,3 +236,28 @@ class NBCNewsIE(InfoExtractor): 'url': info['videoAssets'][-1]['publicUrl'], 'ie_key': 'ThePlatform', } + + +class MSNBCIE(InfoExtractor): + # https URLs redirect to corresponding http ones + _VALID_URL = r'http://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', + '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/ndr.py b/youtube_dl/extractor/ndr.py index 79a13958b..ba06d8a98 100644 --- a/youtube_dl/extractor/ndr.py +++ b/youtube_dl/extractor/ndr.py @@ -1,130 +1,384 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor from ..utils import ( - ExtractorError, + determine_ext, int_or_none, + parse_iso8601, qualities, - parse_duration, ) class NDRBaseIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + display_id = next(group for group in mobj.groups() if group) + webpage = self._download_webpage(url, display_id) + return self._extract_embed(webpage, display_id) - page = self._download_webpage(url, video_id, 'Downloading page') - title = self._og_search_title(page).strip() - description = self._og_search_description(page) - if description: - description = description.strip() +class NDRIE(NDRBaseIE): + IE_NAME = 'ndr' + IE_DESC = 'NDR.de - Norddeutscher Rundfunk' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[^/?#]+),[\da-z]+\.html' + _TESTS = [{ + # httpVideo, same content id + 'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html', + 'md5': '6515bc255dc5c5f8c85bbc38e035a659', + 'info_dict': { + 'id': 'hafengeburtstag988', + 'display_id': 'Party-Poette-und-Parade', + 'ext': 'mp4', + 'title': 'Party, Pötte und Parade', + 'description': 'md5:ad14f9d2f91d3040b6930c697e5f6b4c', + 'uploader': 'ndrtv', + 'timestamp': 1431108900, + 'upload_date': '20150510', + 'duration': 3498, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideo, different content id + 'url': 'http://www.ndr.de/sport/fussball/40-Osnabrueck-spielt-sich-in-einen-Rausch,osna270.html', + 'md5': '1043ff203eab307f0c51702ec49e9a71', + 'info_dict': { + 'id': 'osna272', + 'display_id': '40-Osnabrueck-spielt-sich-in-einen-Rausch', + 'ext': 'mp4', + 'title': 'Osnabrück - Wehen Wiesbaden: Die Highlights', + 'description': 'md5:32e9b800b3d2d4008103752682d5dc01', + 'uploader': 'ndrtv', + 'timestamp': 1442059200, + 'upload_date': '20150912', + 'duration': 510, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpAudio, same content id + 'url': 'http://www.ndr.de/info/La-Valette-entgeht-der-Hinrichtung,audio51535.html', + 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', + 'info_dict': { + 'id': 'audio51535', + 'display_id': 'La-Valette-entgeht-der-Hinrichtung', + 'ext': 'mp3', + 'title': 'La Valette entgeht der Hinrichtung', + 'description': 'md5:22f9541913a40fe50091d5cdd7c9f536', + 'uploader': 'ndrinfo', + 'timestamp': 1290626100, + 'upload_date': '20140729', + 'duration': 884, + }, + 'params': { + 'skip_download': True, + }, + }] + + def _extract_embed(self, webpage, display_id): + embed_url = self._html_search_meta( + 'embedURL', webpage, 'embed URL', fatal=True) + description = self._search_regex( + r'<p[^>]+itemprop="description">([^<]+)</p>', + webpage, 'description', fatal=False) + timestamp = parse_iso8601( + self._search_regex( + r'<span itemprop="datePublished" content="([^"]+)">', + webpage, 'upload date', fatal=False)) + return { + '_type': 'url_transparent', + 'url': embed_url, + 'display_id': display_id, + 'description': description, + 'timestamp': timestamp, + } - duration = int_or_none(self._html_search_regex(r'duration: (\d+),\n', page, 'duration', default=None)) - if not duration: - duration = parse_duration(self._html_search_regex( - r'(<span class="min">\d+</span>:<span class="sec">\d+</span>)', - page, 'duration', default=None)) - formats = [] +class NJoyIE(NDRBaseIE): + IE_NAME = 'njoy' + IE_DESC = 'N-JOY' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?:(?P<display_id>[^/?#]+),)?(?P<id>[\da-z]+)\.html' + _TESTS = [{ + # httpVideo, same content id + 'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html', + 'md5': 'cb63be60cd6f9dd75218803146d8dc67', + 'info_dict': { + 'id': 'comedycontest2480', + 'display_id': 'Benaissa-beim-NDR-Comedy-Contest', + 'ext': 'mp4', + 'title': 'Benaissa beim NDR Comedy Contest', + 'description': 'md5:f057a6c4e1c728b10d33b5ffd36ddc39', + 'uploader': 'ndrtv', + 'upload_date': '20141129', + 'duration': 654, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideo, different content id + 'url': 'http://www.n-joy.de/musik/Das-frueheste-DJ-Set-des-Nordens-live-mit-Felix-Jaehn-,felixjaehn168.html', + 'md5': '417660fffa90e6df2fda19f1b40a64d8', + 'info_dict': { + 'id': 'dockville882', + 'display_id': 'Das-frueheste-DJ-Set-des-Nordens-live-mit-Felix-Jaehn-', + 'ext': 'mp4', + 'title': '"Ich hab noch nie" mit Felix Jaehn', + 'description': 'md5:85dd312d53be1b99e1f998a16452a2f3', + 'uploader': 'njoy', + 'upload_date': '20150822', + 'duration': 211, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.n-joy.de/radio/webradio/morningshow209.html', + 'only_matching': True, + }] + + def _extract_embed(self, webpage, display_id): + video_id = self._search_regex( + r'<iframe[^>]+id="pp_([\da-z]+)"', webpage, 'embed id') + description = self._search_regex( + r'<div[^>]+class="subline"[^>]*>[^<]+</div>\s*<p>([^<]+)</p>', + webpage, 'description', fatal=False) + return { + '_type': 'url_transparent', + 'ie_key': 'NDREmbedBase', + 'url': 'ndr:%s' % video_id, + 'display_id': display_id, + 'description': description, + } + + +class NDREmbedBaseIE(InfoExtractor): + IE_NAME = 'ndr:embed:base' + _VALID_URL = r'(?:ndr:(?P<id_s>[\da-z]+)|https?://www\.ndr\.de/(?P<id>[\da-z]+)-ppjson\.json)' + _TESTS = [{ + 'url': 'ndr:soundcheck3366', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/soundcheck3366-ppjson.json', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') or mobj.group('id_s') + + ppjson = self._download_json( + 'http://www.ndr.de/%s-ppjson.json' % video_id, video_id) - mp3_url = re.search(r'''\{src:'(?P<audio>[^']+)', type:"audio/mp3"},''', page) - if mp3_url: - formats.append({ - 'url': mp3_url.group('audio'), - 'format_id': 'mp3', - }) + playlist = ppjson['playlist'] - thumbnail = None + formats = [] + quality_key = qualities(('xs', 's', 'm', 'l', 'xl')) - video_url = re.search(r'''3: \{src:'(?P<video>.+?)\.(lo|hi|hq)\.mp4', type:"video/mp4"},''', page) - if video_url: - thumbnails = re.findall(r'''\d+: \{src: "([^"]+)"(?: \|\| '[^']+')?, quality: '([^']+)'}''', page) - if thumbnails: - quality_key = qualities(['xs', 's', 'm', 'l', 'xl']) - largest = max(thumbnails, key=lambda thumb: quality_key(thumb[1])) - thumbnail = 'http://www.ndr.de' + largest[0] + for format_id, f in playlist.items(): + src = f.get('src') + if not src: + continue + ext = determine_ext(src, None) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats( + src + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, f4m_id='hds')) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + src, video_id, m3u8_id='hls', entry_protocol='m3u8_native')) + else: + quality = f.get('quality') + ff = { + 'url': src, + 'format_id': quality or format_id, + 'quality': quality_key(quality), + } + type_ = f.get('type') + if type_ and type_.split('/')[0] == 'audio': + ff['vcodec'] = 'none' + ff['ext'] = ext or 'mp3' + formats.append(ff) + self._sort_formats(formats) - for format_id in 'lo', 'hi', 'hq': - formats.append({ - 'url': '%s.%s.mp4' % (video_url.group('video'), format_id), - 'format_id': format_id, - }) + config = playlist['config'] - if not formats: - raise ExtractorError('No media links available for %s' % video_id) + live = playlist.get('config', {}).get('streamType') in ['httpVideoLive', 'httpAudioLive'] + title = config['title'] + if live: + title = self._live_title(title) + uploader = ppjson.get('config', {}).get('branding') + upload_date = ppjson.get('config', {}).get('publicationDate') + duration = int_or_none(config.get('duration')) + + thumbnails = [{ + 'id': thumbnail.get('quality') or thumbnail_id, + 'url': thumbnail['src'], + 'preference': quality_key(thumbnail.get('quality')), + } for thumbnail_id, thumbnail in config.get('poster', {}).items() if thumbnail.get('src')] return { 'id': video_id, 'title': title, - 'description': description, - 'thumbnail': thumbnail, + 'is_live': live, + 'uploader': uploader if uploader != '-' else None, + 'upload_date': upload_date[0:8] if upload_date else None, 'duration': duration, + 'thumbnails': thumbnails, 'formats': formats, } -class NDRIE(NDRBaseIE): - IE_NAME = 'ndr' - IE_DESC = 'NDR.de - Mediathek' - _VALID_URL = r'https?://www\.ndr\.de/.+?(?P<id>\d+)\.html' - - _TESTS = [ - { - 'url': 'http://www.ndr.de/fernsehen/sendungen/nordmagazin/Kartoffeltage-in-der-Lewitz,nordmagazin25866.html', - 'md5': '5bc5f5b92c82c0f8b26cddca34f8bb2c', - 'note': 'Video file', - 'info_dict': { - 'id': '25866', - 'ext': 'mp4', - 'title': 'Kartoffeltage in der Lewitz', - 'description': 'md5:48c4c04dde604c8a9971b3d4e3b9eaa8', - 'duration': 166, - }, - 'skip': '404 Not found', - }, - { - 'url': 'http://www.ndr.de/fernsehen/Party-Poette-und-Parade,hafengeburtstag988.html', - 'md5': 'dadc003c55ae12a5d2f6bd436cd73f59', - 'info_dict': { - 'id': '988', - 'ext': 'mp4', - 'title': 'Party, Pötte und Parade', - 'description': 'Hunderttausende feiern zwischen Speicherstadt und St. Pauli den 826. Hafengeburtstag. Die NDR Sondersendung zeigt die schönsten und spektakulärsten Bilder vom Auftakt.', - 'duration': 3498, - }, - }, - { - 'url': 'http://www.ndr.de/info/audio51535.html', - 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', - 'note': 'Audio file', - 'info_dict': { - 'id': '51535', - 'ext': 'mp3', - 'title': 'La Valette entgeht der Hinrichtung', - 'description': 'md5:22f9541913a40fe50091d5cdd7c9f536', - 'duration': 884, - } - } - ] - +class NDREmbedIE(NDREmbedBaseIE): + IE_NAME = 'ndr:embed' + _VALID_URL = r'https?://www\.ndr\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)\.html' + _TESTS = [{ + 'url': 'http://www.ndr.de/fernsehen/sendungen/ndr_aktuell/ndraktuell28488-player.html', + 'md5': '8b9306142fe65bbdefb5ce24edb6b0a9', + 'info_dict': { + 'id': 'ndraktuell28488', + 'ext': 'mp4', + 'title': 'Norddeutschland begrüßt Flüchtlinge', + 'is_live': False, + 'uploader': 'ndrtv', + 'upload_date': '20150907', + 'duration': 132, + }, + }, { + 'url': 'http://www.ndr.de/ndr2/events/soundcheck/soundcheck3366-player.html', + 'md5': '002085c44bae38802d94ae5802a36e78', + 'info_dict': { + 'id': 'soundcheck3366', + 'ext': 'mp4', + 'title': 'Ella Henderson braucht Vergleiche nicht zu scheuen', + 'is_live': False, + 'uploader': 'ndr2', + 'upload_date': '20150912', + 'duration': 3554, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/info/audio51535-player.html', + 'md5': 'bb3cd38e24fbcc866d13b50ca59307b8', + 'info_dict': { + 'id': 'audio51535', + 'ext': 'mp3', + 'title': 'La Valette entgeht der Hinrichtung', + 'is_live': False, + 'uploader': 'ndrinfo', + 'upload_date': '20140729', + 'duration': 884, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/visite/visite11010-externalPlayer.html', + 'md5': 'ae57f80511c1e1f2fd0d0d3d31aeae7c', + 'info_dict': { + 'id': 'visite11010', + 'ext': 'mp4', + 'title': 'Visite - die ganze Sendung', + 'is_live': False, + 'uploader': 'ndrtv', + 'upload_date': '20150902', + 'duration': 3525, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpVideoLive + 'url': 'http://www.ndr.de/fernsehen/livestream/livestream217-externalPlayer.html', + 'info_dict': { + 'id': 'livestream217', + 'ext': 'flv', + 'title': 're:^NDR Fernsehen Niedersachsen \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + 'upload_date': '20150910', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.ndr.de/ndrkultur/audio255020-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/nordtour/nordtour7124-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/kultur/film/videos/videoimport10424-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/hamburg_journal/hamj43006-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/sendungen/weltbilder/weltbilder4518-player.html', + 'only_matching': True, + }, { + 'url': 'http://www.ndr.de/fernsehen/doku952-player.html', + 'only_matching': True, + }] -class NJoyIE(NDRBaseIE): - IE_NAME = 'N-JOY' - _VALID_URL = r'https?://www\.n-joy\.de/.+?(?P<id>\d+)\.html' - _TEST = { - 'url': 'http://www.n-joy.de/entertainment/comedy/comedy_contest/Benaissa-beim-NDR-Comedy-Contest,comedycontest2480.html', - 'md5': 'cb63be60cd6f9dd75218803146d8dc67', +class NJoyEmbedIE(NDREmbedBaseIE): + IE_NAME = 'njoy:embed' + _VALID_URL = r'https?://www\.n-joy\.de/(?:[^/]+/)+(?P<id>[\da-z]+)-(?:player|externalPlayer)_[^/]+\.html' + _TESTS = [{ + # httpVideo + 'url': 'http://www.n-joy.de/events/reeperbahnfestival/doku948-player_image-bc168e87-5263-4d6d-bd27-bb643005a6de_theme-n-joy.html', + 'md5': '8483cbfe2320bd4d28a349d62d88bd74', 'info_dict': { - 'id': '2480', + 'id': 'doku948', 'ext': 'mp4', - 'title': 'Benaissa beim NDR Comedy Contest', - 'description': 'Von seinem sehr "behaarten" Leben lässt sich Benaissa trotz aller Schwierigkeiten nicht unterkriegen.', - 'duration': 654, - } - } + 'title': 'Zehn Jahre Reeperbahn Festival - die Doku', + 'is_live': False, + 'upload_date': '20150807', + 'duration': 1011, + }, + }, { + # httpAudio + 'url': 'http://www.n-joy.de/news_wissen/stefanrichter100-player_image-d5e938b1-f21a-4b9a-86b8-aaba8bca3a13_theme-n-joy.html', + 'md5': 'd989f80f28ac954430f7b8a48197188a', + 'info_dict': { + 'id': 'stefanrichter100', + 'ext': 'mp3', + 'title': 'Interview mit einem Augenzeugen', + 'is_live': False, + 'uploader': 'njoy', + 'upload_date': '20150909', + 'duration': 140, + }, + 'params': { + 'skip_download': True, + }, + }, { + # httpAudioLive, no explicit ext + 'url': 'http://www.n-joy.de/news_wissen/webradioweltweit100-player_image-3fec0484-2244-4565-8fb8-ed25fd28b173_theme-n-joy.html', + 'info_dict': { + 'id': 'webradioweltweit100', + 'ext': 'mp3', + 'title': 're:^N-JOY Weltweit \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + 'uploader': 'njoy', + 'upload_date': '20150810', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.n-joy.de/musik/dockville882-player_image-3905259e-0803-4764-ac72-8b7de077d80a_theme-n-joy.html', + 'only_matching': True, + }, { + 'url': 'http://www.n-joy.de/radio/sendungen/morningshow/urlaubsfotos190-player_image-066a5df1-5c95-49ec-a323-941d848718db_theme-n-joy.html', + 'only_matching': True, + }, { + 'url': 'http://www.n-joy.de/entertainment/comedy/krudetv290-player_image-ab261bfe-51bf-4bf3-87ba-c5122ee35b3d_theme-n-joy.html', + 'only_matching': True, + }] diff --git a/youtube_dl/extractor/nextmedia.py b/youtube_dl/extractor/nextmedia.py index c10784f6b..d1688457f 100644 --- a/youtube_dl/extractor/nextmedia.py +++ b/youtube_dl/extractor/nextmedia.py @@ -126,7 +126,8 @@ class AppleDailyIE(NextMediaIE): 'thumbnail': 're:^https?://.*\.jpg$', 'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd', 'upload_date': '20150128', - } + }, + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { # No thumbnail 'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/', @@ -140,10 +141,19 @@ class AppleDailyIE(NextMediaIE): }, 'expected_warnings': [ 'video thumbnail', - ] + ], + 'skip': 'redirect to http://www.appledaily.com.tw/animation/', }, { 'url': 'http://www.appledaily.com.tw/appledaily/article/supplement/20140417/35770334/', - 'only_matching': True, + 'md5': 'eaa20e6b9df418c912d7f5dec2ba734d', + 'info_dict': { + 'id': '35770334', + 'ext': 'mp4', + 'title': '咖啡占卜測 XU裝熟指數', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:7b859991a6a4fedbdf3dd3b66545c748', + 'upload_date': '20140417', + }, }] _URL_PATTERN = r'\{url: \'(.+)\'\}' diff --git a/youtube_dl/extractor/nfl.py b/youtube_dl/extractor/nfl.py index dc54634a5..200874d68 100644 --- a/youtube_dl/extractor/nfl.py +++ b/youtube_dl/extractor/nfl.py @@ -16,53 +16,118 @@ from ..utils import ( class NFLIE(InfoExtractor): IE_NAME = 'nfl.com' - _VALID_URL = r'''(?x)https?:// - (?P<host>(?:www\.)?(?:nfl\.com|.*?\.clubs\.nfl\.com))/ - (?:.+?/)* - (?P<id>(?:[a-z0-9]{16}|\w{8}\-(?:\w{4}\-){3}\w{12}))''' - _TESTS = [ - { - 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights', - 'md5': '394ef771ddcd1354f665b471d78ec4c6', - 'info_dict': { - 'id': '0ap3000000398478', - 'ext': 'mp4', - 'title': 'Week 3: Redskins vs. Eagles highlights', - 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478', - 'upload_date': '20140921', - 'timestamp': 1411337580, - 'thumbnail': 're:^https?://.*\.jpg$', - } + _VALID_URL = r'''(?x) + https?:// + (?P<host> + (?:www\.)? + (?: + (?: + nfl| + buffalobills| + miamidolphins| + patriots| + newyorkjets| + baltimoreravens| + bengals| + clevelandbrowns| + steelers| + houstontexans| + colts| + jaguars| + titansonline| + denverbroncos| + kcchiefs| + raiders| + chargers| + dallascowboys| + giants| + philadelphiaeagles| + redskins| + chicagobears| + detroitlions| + packers| + vikings| + atlantafalcons| + panthers| + neworleanssaints| + buccaneers| + azcardinals| + stlouisrams| + 49ers| + seahawks + )\.com| + .+?\.clubs\.nfl\.com + ) + )/ + (?:.+?/)* + (?P<id>[^/#?&]+) + ''' + _TESTS = [{ + 'url': 'http://www.nfl.com/videos/nfl-game-highlights/0ap3000000398478/Week-3-Redskins-vs-Eagles-highlights', + 'md5': '394ef771ddcd1354f665b471d78ec4c6', + 'info_dict': { + 'id': '0ap3000000398478', + 'ext': 'mp4', + 'title': 'Week 3: Redskins vs. Eagles highlights', + 'description': 'md5:56323bfb0ac4ee5ab24bd05fdf3bf478', + 'upload_date': '20140921', + 'timestamp': 1411337580, + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266', + 'md5': 'cf85bdb4bc49f6e9d3816d130c78279c', + 'info_dict': { + 'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266', + 'ext': 'mp4', + 'title': 'LIVE: Post Game vs. Browns', + 'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8', + 'upload_date': '20131229', + 'timestamp': 1388354455, + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish', + 'info_dict': { + 'id': '0ap3000000467607', + 'ext': 'mp4', + 'title': 'Frustrations flare on the field', + 'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.', + 'timestamp': 1422850320, + 'upload_date': '20150202', }, - { - 'url': 'http://prod.www.steelers.clubs.nfl.com/video-and-audio/videos/LIVE_Post_Game_vs_Browns/9d72f26a-9e2b-4718-84d3-09fb4046c266', - 'md5': 'cf85bdb4bc49f6e9d3816d130c78279c', - 'info_dict': { - 'id': '9d72f26a-9e2b-4718-84d3-09fb4046c266', - 'ext': 'mp4', - 'title': 'LIVE: Post Game vs. Browns', - 'description': 'md5:6a97f7e5ebeb4c0e69a418a89e0636e8', - 'upload_date': '20131229', - 'timestamp': 1388354455, - 'thumbnail': 're:^https?://.*\.jpg$', - } + }, { + 'url': 'http://www.patriots.com/video/2015/09/18/10-days-gillette', + 'md5': '4c319e2f625ffd0b481b4382c6fc124c', + 'info_dict': { + 'id': 'n-238346', + 'ext': 'mp4', + 'title': '10 Days at Gillette', + 'description': 'md5:8cd9cd48fac16de596eadc0b24add951', + 'timestamp': 1442618809, + 'upload_date': '20150918', }, - { - 'url': 'http://www.nfl.com/news/story/0ap3000000467586/article/patriots-seahawks-involved-in-lategame-skirmish', - 'info_dict': { - 'id': '0ap3000000467607', - 'ext': 'mp4', - 'title': 'Frustrations flare on the field', - 'description': 'Emotions ran high at the end of the Super Bowl on both sides of the ball after a dramatic finish.', - 'timestamp': 1422850320, - 'upload_date': '20150202', - }, + }, { + # lowercase data-contentid + 'url': 'http://www.steelers.com/news/article-1/Tomlin-on-Ben-getting-Vick-ready/56399c96-4160-48cf-a7ad-1d17d4a3aef7', + 'info_dict': { + 'id': '12693586-6ea9-4743-9c1c-02c59e4a5ef2', + 'ext': 'mp4', + 'title': 'Tomlin looks ahead to Ravens on a short week', + 'description': 'md5:32f3f7b139f43913181d5cbb24ecad75', + 'timestamp': 1443459651, + 'upload_date': '20150928', }, - { - 'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood', - 'only_matching': True, - } - ] + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.nfl.com/videos/nfl-network-top-ten/09000d5d810a6bd4/Top-10-Gutsiest-Performances-Jack-Youngblood', + 'only_matching': True, + }, { + 'url': 'http://www.buffalobills.com/video/videos/Rex_Ryan_Show_World_Wide_Rex/b1dcfab2-3190-4bb1-bfc0-d6e603d6601a', + 'only_matching': True, + }] @staticmethod def prepend_host(host, url): @@ -95,13 +160,14 @@ class NFLIE(InfoExtractor): webpage = self._download_webpage(url, video_id) config_url = NFLIE.prepend_host(host, self._search_regex( - r'(?:config|configURL)\s*:\s*"([^"]+)"', webpage, 'config URL', - default='static/content/static/config/video/config.json')) + r'(?:(?:config|configURL)\s*:\s*|<nflcs:avplayer[^>]+data-config\s*=\s*)(["\'])(?P<config>.+?)\1', + webpage, 'config URL', default='static/content/static/config/video/config.json', + group='config')) # For articles, the id in the url is not the video id video_id = self._search_regex( - r'contentId\s*:\s*"([^"]+)"', webpage, 'video id', default=video_id) - config = self._download_json(config_url, video_id, - note='Downloading player config') + r'(?:<nflcs:avplayer[^>]+data-content[Ii]d\s*=\s*|content[Ii]d\s*:\s*)(["\'])(?P<id>.+?)\1', + webpage, 'video id', default=video_id, group='id') + config = self._download_json(config_url, video_id, 'Downloading player config') url_template = NFLIE.prepend_host( host, '{contentURLTemplate:}'.format(**config)) video_data = self._download_json( diff --git a/youtube_dl/extractor/nhl.py b/youtube_dl/extractor/nhl.py index 279b18386..e98a5ef89 100644 --- a/youtube_dl/extractor/nhl.py +++ b/youtube_dl/extractor/nhl.py @@ -72,7 +72,7 @@ class NHLBaseInfoExtractor(InfoExtractor): class NHLIE(NHLBaseInfoExtractor): IE_NAME = 'nhl.com' - _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console)?(?:\?(?:.*?[?&])?)(?:id|hlg)=(?P<id>[-0-9a-zA-Z,]+)' + _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console|embed)?(?:\?(?:.*?[?&])?)(?:id|hlg|playlist)=(?P<id>[-0-9a-zA-Z,]+)' _TESTS = [{ 'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614', @@ -136,6 +136,9 @@ class NHLIE(NHLBaseInfoExtractor): 'params': { 'skip_download': True, # Requires rtmpdump } + }, { + 'url': 'http://video.nhl.com/videocenter/embed?playlist=836127', + 'only_matching': True, }] def _real_extract(self, url): @@ -146,9 +149,9 @@ class NHLIE(NHLBaseInfoExtractor): class NHLNewsIE(NHLBaseInfoExtractor): IE_NAME = 'nhl.com:news' IE_DESC = 'NHL news' - _VALID_URL = r'https?://(?:www\.)?nhl\.com/ice/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)' + _VALID_URL = r'https?://(?:.+?\.)?nhl\.com/(?:ice|club)/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.nhl.com/ice/news.htm?id=750727', 'md5': '4b3d1262e177687a3009937bd9ec0be8', 'info_dict': { @@ -159,13 +162,26 @@ class NHLNewsIE(NHLBaseInfoExtractor): 'duration': 37, 'upload_date': '20150128', }, - } + }, { + # iframe embed + 'url': 'http://sabres.nhl.com/club/news.htm?id=780189', + 'md5': '9f663d1c006c90ac9fb82777d4294e12', + 'info_dict': { + 'id': '836127', + 'ext': 'mp4', + 'title': 'Morning Skate: OTT vs. BUF (9/23/15)', + 'description': "Brian Duff chats with Tyler Ennis prior to Buffalo's first preseason home game.", + 'duration': 93, + 'upload_date': '20150923', + }, + }] def _real_extract(self, url): news_id = self._match_id(url) webpage = self._download_webpage(url, news_id) video_id = self._search_regex( - [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'"], + [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'", + r'<iframe[^>]+src=["\']https?://video.*?\.nhl\.com/videocenter/embed\?.*\bplaylist=(\d+)'], webpage, 'video id') return self._real_extract_video(video_id) diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index 0f8aa5ada..bda1cff05 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -12,6 +12,7 @@ from ..compat import ( compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, parse_duration, @@ -100,10 +101,7 @@ class NiconicoIE(InfoExtractor): 'mail': username, 'password': password, } - # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode - # chokes on unicode - login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items()) - login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8') + login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') request = compat_urllib_request.Request( 'https://secure.nicovideo.jp/secure/login', login_data) login_results = self._download_webpage( diff --git a/youtube_dl/extractor/ninegag.py b/youtube_dl/extractor/ninegag.py index 7f842b5c2..a06d38afd 100644 --- a/youtube_dl/extractor/ninegag.py +++ b/youtube_dl/extractor/ninegag.py @@ -1,7 +1,6 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..utils import str_to_int @@ -9,61 +8,93 @@ from ..utils import str_to_int class NineGagIE(InfoExtractor): IE_NAME = '9gag' - _VALID_URL = r'''(?x)^https?://(?:www\.)?9gag\.tv/ - (?: - v/(?P<numid>[0-9]+)| - p/(?P<id>[a-zA-Z0-9]+)/(?P<display_id>[^?#/]+) - ) - ''' + _VALID_URL = r'https?://(?:www\.)?9gag(?:\.com/tv|\.tv)/(?:p|embed)/(?P<id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^?#/]+))?' _TESTS = [{ - "url": "http://9gag.tv/v/1912", - "info_dict": { - "id": "1912", - "ext": "mp4", - "description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)", - "title": "\"People Are Awesome 2013\" Is Absolutely Awesome", + 'url': 'http://9gag.com/tv/p/Kk2X5/people-are-awesome-2013-is-absolutely-awesome', + 'info_dict': { + 'id': 'Kk2X5', + 'ext': 'mp4', + 'description': 'This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)', + 'title': '\"People Are Awesome 2013\" Is Absolutely Awesome', 'uploader_id': 'UCdEH6EjDKwtTe-sO2f0_1XA', 'uploader': 'CompilationChannel', 'upload_date': '20131110', - "view_count": int, - "thumbnail": "re:^https?://", + 'view_count': int, }, - 'add_ie': ['Youtube'] + 'add_ie': ['Youtube'], }, { - 'url': 'http://9gag.tv/p/KklwM/alternate-banned-opening-scene-of-gravity?ref=fsidebar', + 'url': 'http://9gag.com/tv/p/aKolP3', 'info_dict': { - 'id': 'KklwM', + 'id': 'aKolP3', 'ext': 'mp4', - 'display_id': 'alternate-banned-opening-scene-of-gravity', - "description": "While Gravity was a pretty awesome movie already, YouTuber Krishna Shenoi came up with a way to improve upon it, introducing a much better solution to Sandra Bullock's seemingly endless tumble in space. The ending is priceless.", - 'title': "Banned Opening Scene Of \"Gravity\" That Changes The Whole Movie", - 'uploader': 'Krishna Shenoi', - 'upload_date': '20140401', - 'uploader_id': 'krishnashenoi93', + 'title': 'This Guy Travelled 11 countries In 44 days Just To Make This Amazing Video', + 'description': "I just saw more in 1 minute than I've seen in 1 year. This guy's video is epic!!", + 'uploader_id': 'rickmereki', + 'uploader': 'Rick Mereki', + 'upload_date': '20110803', + 'view_count': int, }, + 'add_ie': ['Vimeo'], + }, { + 'url': 'http://9gag.com/tv/p/KklwM', + 'only_matching': True, + }, { + 'url': 'http://9gag.tv/p/Kk2X5', + 'only_matching': True, + }, { + 'url': 'http://9gag.com/tv/embed/a5Dmvl', + 'only_matching': True, }] + _EXTERNAL_VIDEO_PROVIDER = { + '1': { + 'url': '%s', + 'ie_key': 'Youtube', + }, + '2': { + 'url': 'http://player.vimeo.com/video/%s', + 'ie_key': 'Vimeo', + }, + '3': { + 'url': 'http://instagram.com/p/%s', + 'ie_key': 'Instagram', + }, + '4': { + 'url': 'http://vine.co/v/%s', + 'ie_key': 'Vine', + }, + } + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('numid') or mobj.group('id') + video_id = mobj.group('id') display_id = mobj.group('display_id') or video_id webpage = self._download_webpage(url, display_id) - post_view = json.loads(self._html_search_regex( - r'var postView = new app\.PostView\({\s*post:\s*({.+?}),\s*posts:\s*prefetchedCurrentPost', webpage, 'post view')) + post_view = self._parse_json( + self._search_regex( + r'var\s+postView\s*=\s*new\s+app\.PostView\({\s*post:\s*({.+?})\s*,\s*posts:\s*prefetchedCurrentPost', + webpage, 'post view'), + display_id) - youtube_id = post_view['videoExternalId'] + ie_key = None + source_url = post_view.get('sourceUrl') + if not source_url: + external_video_id = post_view['videoExternalId'] + external_video_provider = post_view['videoExternalProvider'] + source_url = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['url'] % external_video_id + ie_key = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['ie_key'] title = post_view['title'] - description = post_view['description'] - view_count = str_to_int(post_view['externalView']) + description = post_view.get('description') + view_count = str_to_int(post_view.get('externalView')) thumbnail = post_view.get('thumbnail_700w') or post_view.get('ogImageUrl') or post_view.get('thumbnail_300w') return { '_type': 'url_transparent', - 'url': youtube_id, - 'ie_key': 'Youtube', + 'url': source_url, + 'ie_key': ie_key, 'id': video_id, 'display_id': display_id, 'title': title, diff --git a/youtube_dl/extractor/nowness.py b/youtube_dl/extractor/nowness.py index 6b2f3f55a..b97f62fdb 100644 --- a/youtube_dl/extractor/nowness.py +++ b/youtube_dl/extractor/nowness.py @@ -1,64 +1,134 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .brightcove import BrightcoveIE from .common import InfoExtractor from ..utils import ExtractorError +from ..compat import ( + compat_str, + compat_urllib_request, +) + + +class NownessBaseIE(InfoExtractor): + def _extract_url_result(self, post): + if post['type'] == 'video': + for media in post['media']: + if media['type'] == 'video': + video_id = media['content'] + source = media['source'] + if source == 'brightcove': + player_code = self._download_webpage( + 'http://www.nowness.com/iframe?id=%s' % video_id, video_id, + note='Downloading player JavaScript', + errnote='Unable to download player JavaScript') + bc_url = BrightcoveIE._extract_brightcove_url(player_code) + if bc_url is None: + raise ExtractorError('Could not find player definition') + return self.url_result(bc_url, 'Brightcove') + elif source == 'vimeo': + return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo') + elif source == 'youtube': + return self.url_result(video_id, 'Youtube') + elif source == 'cinematique': + # youtube-dl currently doesn't support cinematique + # return self.url_result('http://cinematique.com/embed/%s' % video_id, 'Cinematique') + pass + def _api_request(self, url, request_path): + display_id = self._match_id(url) + request = compat_urllib_request.Request( + 'http://api.nowness.com/api/' + request_path % display_id, + headers={ + 'X-Nowness-Language': 'zh-cn' if 'cn.nowness.com' in url else 'en-us', + }) + return display_id, self._download_json(request, display_id) -class NownessIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])' - _TESTS = [ - { - 'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation', - 'md5': '068bc0202558c2e391924cb8cc470676', - 'info_dict': { - 'id': '2520295746001', - 'ext': 'mp4', - 'title': 'Candor: The Art of Gesticulation', - 'description': 'Candor: The Art of Gesticulation', - 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'Nowness', - } +class NownessIE(NownessBaseIE): + IE_NAME = 'nowness' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/(?:story|(?:series|category)/[^/]+)/(?P<id>[^/]+?)(?:$|[?#])' + _TESTS = [{ + 'url': 'https://www.nowness.com/story/candor-the-art-of-gesticulation', + 'md5': '068bc0202558c2e391924cb8cc470676', + 'info_dict': { + 'id': '2520295746001', + 'ext': 'mp4', + 'title': 'Candor: The Art of Gesticulation', + 'description': 'Candor: The Art of Gesticulation', + 'thumbnail': 're:^https?://.*\.jpg', + 'uploader': 'Nowness', }, - { - 'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr', - 'md5': 'e79cf125e387216f86b2e0a5b5c63aa3', - 'info_dict': { - 'id': '3716354522001', - 'ext': 'mp4', - 'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', - 'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', - 'thumbnail': 're:^https?://.*\.jpg', - 'uploader': 'Nowness', - } + }, { + 'url': 'https://cn.nowness.com/story/kasper-bjorke-ft-jaakko-eino-kalevi-tnr', + 'md5': 'e79cf125e387216f86b2e0a5b5c63aa3', + 'info_dict': { + 'id': '3716354522001', + 'ext': 'mp4', + 'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', + 'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR', + 'thumbnail': 're:^https?://.*\.jpg', + 'uploader': 'Nowness', }, - ] + }, { + # vimeo + 'url': 'https://www.nowness.com/series/nowness-picks/jean-luc-godard-supercut', + 'md5': '9a5a6a8edf806407e411296ab6bc2a49', + 'info_dict': { + 'id': '130020913', + 'ext': 'mp4', + 'title': 'Bleu, Blanc, Rouge - A Godard Supercut', + 'description': 'md5:f0ea5f1857dffca02dbd37875d742cec', + 'thumbnail': 're:^https?://.*\.jpg', + 'upload_date': '20150607', + 'uploader': 'Cinema Sem Lei', + 'uploader_id': 'cinemasemlei', + }, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('slug') + _, post = self._api_request(url, 'post/getBySlug/%s') + return self._extract_url_result(post) - webpage = self._download_webpage(url, video_id) - player_url = self._search_regex( - r'"([^"]+/content/issue-[0-9.]+.js)"', webpage, 'player URL') - real_id = self._search_regex( - r'\sdata-videoId="([0-9]+)"', webpage, 'internal video ID') - player_code = self._download_webpage( - player_url, video_id, - note='Downloading player JavaScript', - errnote='Player download failed') - player_code = player_code.replace("'+d+'", real_id) +class NownessPlaylistIE(NownessBaseIE): + IE_NAME = 'nowness:playlist' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/playlist/(?P<id>\d+)' + _TEST = { + 'url': 'https://www.nowness.com/playlist/3286/i-guess-thats-why-they-call-it-the-blues', + 'info_dict': { + 'id': '3286', + }, + 'playlist_mincount': 8, + } - bc_url = BrightcoveIE._extract_brightcove_url(player_code) - if bc_url is None: - raise ExtractorError('Could not find player definition') - return { - '_type': 'url', - 'url': bc_url, - 'ie_key': 'Brightcove', - } + def _real_extract(self, url): + playlist_id, playlist = self._api_request(url, 'post?PlaylistId=%s') + entries = [self._extract_url_result(item) for item in playlist['items']] + return self.playlist_result(entries, playlist_id) + + +class NownessSeriesIE(NownessBaseIE): + IE_NAME = 'nowness:series' + _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/series/(?P<id>[^/]+?)(?:$|[?#])' + _TEST = { + 'url': 'https://www.nowness.com/series/60-seconds', + 'info_dict': { + 'id': '60', + 'title': '60 Seconds', + 'description': 'One-minute wisdom in a new NOWNESS series', + }, + 'playlist_mincount': 4, + } + + def _real_extract(self, url): + display_id, series = self._api_request(url, 'series/getBySlug/%s') + entries = [self._extract_url_result(post) for post in series['posts']] + series_title = None + series_description = None + translations = series.get('translations', []) + if translations: + series_title = translations[0].get('title') or translations[0]['seoTitle'] + series_description = translations[0].get('seoDescription') + return self.playlist_result( + entries, compat_str(series['id']), series_title, series_description) diff --git a/youtube_dl/extractor/nowtv.py b/youtube_dl/extractor/nowtv.py index 173e46cd8..b0bdffc4e 100644 --- a/youtube_dl/extractor/nowtv.py +++ b/youtube_dl/extractor/nowtv.py @@ -1,12 +1,11 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import compat_str from ..utils import ( ExtractorError, + determine_ext, int_or_none, parse_iso8601, parse_duration, @@ -15,7 +14,7 @@ from ..utils import ( class NowTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?nowtv\.de/(?P<station>rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<id>.+?)/player' + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<id>.+?)/(?:player|preview)' _TESTS = [{ # rtl @@ -23,7 +22,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203519', 'display_id': 'bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Die neuen Bauern und eine Hochzeit', 'description': 'md5:e234e1ed6d63cf06be5c070442612e7e', 'thumbnail': 're:^https?://.*\.jpg$', @@ -32,7 +31,7 @@ class NowTVIE(InfoExtractor): 'duration': 2786, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -41,7 +40,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203481', 'display_id': 'berlin-tag-nacht/berlin-tag-nacht-folge-934', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Berlin - Tag & Nacht (Folge 934)', 'description': 'md5:c85e88c2e36c552dfe63433bc9506dd0', 'thumbnail': 're:^https?://.*\.jpg$', @@ -50,7 +49,7 @@ class NowTVIE(InfoExtractor): 'duration': 2641, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -59,7 +58,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '165780', 'display_id': 'alarm-fuer-cobra-11-die-autobahnpolizei/hals-und-beinbruch-2014-08-23-21-10-00', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Hals- und Beinbruch', 'description': 'md5:b50d248efffe244e6f56737f0911ca57', 'thumbnail': 're:^https?://.*\.jpg$', @@ -68,7 +67,7 @@ class NowTVIE(InfoExtractor): 'duration': 2742, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -77,7 +76,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '99205', 'display_id': 'medicopter-117/angst', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Angst!', 'description': 'md5:30cbc4c0b73ec98bcd73c9f2a8c17c4e', 'thumbnail': 're:^https?://.*\.jpg$', @@ -86,7 +85,7 @@ class NowTVIE(InfoExtractor): 'duration': 3025, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -95,7 +94,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203521', 'display_id': 'ratgeber-geld/thema-ua-der-erste-blick-die-apple-watch', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Thema u.a.: Der erste Blick: Die Apple Watch', 'description': 'md5:4312b6c9d839ffe7d8caf03865a531af', 'thumbnail': 're:^https?://.*\.jpg$', @@ -104,7 +103,7 @@ class NowTVIE(InfoExtractor): 'duration': 1083, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -113,7 +112,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '128953', 'display_id': 'der-hundeprofi/buero-fall-chihuahua-joel', - 'ext': 'mp4', + 'ext': 'flv', 'title': "Büro-Fall / Chihuahua 'Joel'", 'description': 'md5:e62cb6bf7c3cc669179d4f1eb279ad8d', 'thumbnail': 're:^https?://.*\.jpg$', @@ -122,18 +121,28 @@ class NowTVIE(InfoExtractor): 'duration': 3092, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, + }, { + 'url': 'http://www.nowtv.de/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit/preview', + 'only_matching': True, + }, { + 'url': 'http://www.nowtv.at/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit/preview?return=/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', + 'only_matching': True, + }, { + 'url': 'http://www.nowtv.de/rtl2/echtzeit/list/aktuell/schnelles-geld-am-ende-der-welt/player', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('id') - station = mobj.group('station') + display_id = self._match_id(url) + display_id_split = display_id.split('/') + if len(display_id) > 2: + display_id = '/'.join((display_id_split[0], display_id_split[-1])) info = self._download_json( - 'https://api.nowtv.de/v3/movies/%s?fields=*,format,files' % display_id, + 'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id, display_id) video_id = compat_str(info['id']) @@ -148,29 +157,19 @@ class NowTVIE(InfoExtractor): raise ExtractorError( 'Video %s is not available for free' % video_id, expected=True) - f = info.get('format', {}) - station = f.get('station') or station - - STATIONS = { - 'rtl': 'rtlnow', - 'rtl2': 'rtl2now', - 'vox': 'voxnow', - 'nitro': 'rtlnitronow', - 'ntv': 'n-tvnow', - 'superrtl': 'superrtlnow' - } - formats = [] for item in files['items']: - item_path = remove_start(item['path'], '/') - tbr = int_or_none(item['bitrate']) - m3u8_url = 'http://hls.fra.%s.de/hls-vod-enc/%s.m3u8' % (STATIONS[station], item_path) - m3u8_url = m3u8_url.replace('now/', 'now/videos/') + if determine_ext(item['path']) != 'f4v': + continue + app, play_path = remove_start(item['path'], '/').split('/', 1) formats.append({ - 'url': m3u8_url, - 'format_id': '%s-%sk' % (item['id'], tbr), - 'ext': 'mp4', - 'tbr': tbr, + 'url': 'rtmpe://fms.rtl.de', + 'app': app, + 'play_path': 'mp4:%s' % play_path, + 'ext': 'flv', + 'page_url': 'http://rtlnow.rtl.de', + 'player_url': 'http://cdn.static-fra.de/now/vodplayer.swf', + 'tbr': int_or_none(item.get('bitrate')), }) self._sort_formats(formats) @@ -178,6 +177,8 @@ class NowTVIE(InfoExtractor): description = info.get('articleLong') or info.get('articleShort') timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') duration = parse_duration(info.get('duration')) + + f = info.get('format', {}) thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') return { diff --git a/youtube_dl/extractor/nowvideo.py b/youtube_dl/extractor/nowvideo.py index dec09cdfe..17baa9679 100644 --- a/youtube_dl/extractor/nowvideo.py +++ b/youtube_dl/extractor/nowvideo.py @@ -7,7 +7,7 @@ class NowVideoIE(NovaMovIE): IE_NAME = 'nowvideo' IE_DESC = 'NowVideo' - _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|sx|eu|at|ag|co|li)'} + _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'nowvideo\.(?:ch|ec|sx|eu|at|ag|co|li)'} _HOST = 'www.nowvideo.ch' diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index 62d12b7a6..eb12fb810 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -1,5 +1,7 @@ from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..utils import ( fix_xml_ampersands, @@ -7,7 +9,6 @@ from ..utils import ( qualities, strip_jsonp, unified_strdate, - url_basename, ) @@ -37,8 +38,21 @@ class NPOBaseIE(InfoExtractor): class NPOIE(NPOBaseIE): - IE_NAME = 'npo.nl' - _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?P<id>[^/?]+)' + IE_NAME = 'npo' + IE_DESC = 'npo.nl and ntr.nl' + _VALID_URL = r'''(?x) + (?: + npo:| + https?:// + (?:www\.)? + (?: + npo\.nl/(?!live|radio)(?:[^/]+/){2}| + ntr\.nl/(?:[^/]+/){2,}| + omroepwnl\.nl/video/fragment/[^/]+__ + ) + ) + (?P<id>[^/?#]+) + ''' _TESTS = [ { @@ -58,7 +72,7 @@ class NPOIE(NPOBaseIE): 'info_dict': { 'id': 'VARA_101191800', 'ext': 'm4v', - 'title': 'De Mega Mike & Mega Thomas show', + 'title': 'De Mega Mike & Mega Thomas show: The best of.', 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4', 'upload_date': '20090227', 'duration': 2400, @@ -70,8 +84,8 @@ class NPOIE(NPOBaseIE): 'info_dict': { 'id': 'VPWON_1169289', 'ext': 'm4v', - 'title': 'Tegenlicht', - 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1', + 'title': 'Tegenlicht: De toekomst komt uit Afrika', + 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea', 'upload_date': '20130225', 'duration': 3000, }, @@ -100,6 +114,30 @@ class NPOIE(NPOBaseIE): 'title': 'Hoe gaat Europa verder na Parijs?', }, }, + { + 'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content', + 'md5': '01c6a2841675995da1f0cf776f03a9c3', + 'info_dict': { + 'id': 'VPWON_1233944', + 'ext': 'm4v', + 'title': 'Aap, poot, pies', + 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde', + 'upload_date': '20150508', + 'duration': 599, + }, + }, + { + 'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698', + 'md5': 'd30cd8417b8b9bca1fdff27428860d08', + 'info_dict': { + 'id': 'POW_00996502', + 'ext': 'm4v', + 'title': '''"Dit is wel een 'landslide'..."''', + 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8', + 'upload_date': '20150508', + 'duration': 462, + }, + } ] def _real_extract(self, url): @@ -114,6 +152,18 @@ class NPOIE(NPOBaseIE): transform_source=strip_jsonp, ) + # For some videos actual video id (prid) is different (e.g. for + # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698 + # video id is POMS_WNL_853698 but prid is POW_00996502) + video_id = metadata.get('prid') or video_id + + # titel is too generic in some cases so utilize aflevering_titel as well + # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html) + title = metadata['titel'] + sub_title = metadata.get('aflevering_titel') + if sub_title and sub_title != title: + title += ': %s' % sub_title + token = self._get_token(video_id) formats = [] @@ -186,8 +236,8 @@ class NPOIE(NPOBaseIE): return { 'id': video_id, - 'title': metadata['titel'], - 'description': metadata['info'], + 'title': title, + 'description': metadata.get('info'), 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'], 'upload_date': unified_strdate(metadata.get('gidsdatum')), 'duration': parse_duration(metadata.get('tijdsduur')), @@ -356,9 +406,9 @@ class NPORadioFragmentIE(InfoExtractor): } -class TegenlichtVproIE(NPOIE): - IE_NAME = 'tegenlicht.vpro.nl' - _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?' +class VPROIE(NPOIE): + IE_NAME = 'vpro' + _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html' _TESTS = [ { @@ -367,17 +417,72 @@ class TegenlichtVproIE(NPOIE): 'info_dict': { 'id': 'VPWON_1169289', 'ext': 'm4v', - 'title': 'Tegenlicht', - 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1', + 'title': 'De toekomst komt uit Afrika', + 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea', 'upload_date': '20130225', }, }, + { + 'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html', + 'info_dict': { + 'id': 'sergio-herman', + 'title': 'Sergio Herman: Fucking perfect', + }, + 'playlist_count': 2, + }, + { + # playlist with youtube embed + 'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html', + 'info_dict': { + 'id': 'education-education', + 'title': '2Doc', + }, + 'playlist_count': 2, + } ] def _real_extract(self, url): - name = url_basename(url) - webpage = self._download_webpage(url, name) - urn = self._html_search_meta('mediaurn', webpage) - info_page = self._download_json( - 'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name) - return self._get_info(info_page['mid']) + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id) + for video_id in re.findall(r'data-media-id="([^"]+)"', webpage) + ] + + playlist_title = self._search_regex( + r'<title>\s*([^>]+?)\s*-\s*Teledoc\s*-\s*VPRO\s*</title>', + webpage, 'playlist title', default=None) or self._og_search_title(webpage) + + return self.playlist_result(entries, playlist_id, playlist_title) + + +class WNLIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+' + + _TEST = { + 'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515', + 'info_dict': { + 'id': 'vandaag-de-dag-6-mei', + 'title': 'Vandaag de Dag 6 mei', + }, + 'playlist_count': 4, + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result('npo:%s' % video_id, 'NPO') + for video_id, part in re.findall( + r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>(Deel \d+)', webpage) + ] + + playlist_title = self._html_search_regex( + r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>', + webpage, 'playlist title') + + return self.playlist_result(entries, playlist_id, playlist_title) diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 9e4581cf9..8ac38a174 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_urlparse from ..utils import ( ExtractorError, float_or_none, @@ -49,7 +50,7 @@ class NRKIE(InfoExtractor): if data['usageRights']['isGeoBlocked']: raise ExtractorError( - 'NRK har ikke rettig-heter til å vise dette programmet utenfor Norge', + 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge', expected=True) video_url = data['mediaUrl'] + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81' @@ -116,7 +117,8 @@ class NRKPlaylistIE(InfoExtractor): class NRKTVIE(InfoExtractor): - _VALID_URL = r'(?P<baseurl>https?://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?' + 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 = [ { @@ -188,31 +190,22 @@ class NRKTVIE(InfoExtractor): 'duration': 6947.5199999999995, }, 'skip': 'Only works from Norway', + }, + { + 'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#', + 'only_matching': True, } ] - def _debug_print(self, txt): - if self._downloader.params.get('verbose', False): - self.to_screen('[debug] %s' % txt) - - def _get_subtitles(self, subtitlesurl, video_id, baseurl): - url = "%s%s" % (baseurl, subtitlesurl) - self._debug_print('%s: Subtitle url: %s' % (video_id, url)) - captions = self._download_xml( - url, video_id, 'Downloading subtitles') - lang = captions.get('lang', 'no') - return {lang: [ - {'ext': 'ttml', 'url': 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) + return self._extract_f4m_formats( + manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds') def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') part_id = mobj.group('part_id') - baseurl = mobj.group('baseurl') + base_url = mobj.group('baseurl') webpage = self._download_webpage(url, video_id) @@ -268,15 +261,18 @@ class NRKTVIE(InfoExtractor): 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')) + 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[ ]*=[ ]*"([^"]+)"', - webpage, 'subtitle URL', default=None) - subtitles = None + r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1', + webpage, 'subtitle URL', default=None, group='url') + subtitles = {} if subtitles_url: - subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl) + subtitles['no'] = [{ + 'ext': 'ttml', + 'url': compat_urlparse.urljoin(base_url, subtitles_url), + }] return { 'id': video_id, diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index 6c7149fe3..184c7a323 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -2,8 +2,9 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_unquote from ..utils import ( + ExtractorError, unified_strdate, int_or_none, qualities, @@ -12,20 +13,23 @@ from ..utils import ( class OdnoklassnikiIE(InfoExtractor): - _VALID_URL = r'https?://(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>[\d-]+)' + _VALID_URL = r'https?://(?:www\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)' _TESTS = [{ # metadata in JSON 'url': 'http://ok.ru/video/20079905452', - 'md5': '8e24ad2da6f387948e7a7d44eb8668fe', + 'md5': '6ba728d85d60aa2e6dd37c9e70fdc6bc', 'info_dict': { 'id': '20079905452', 'ext': 'mp4', 'title': 'Культура меняет нас (прекрасный ролик!))', 'duration': 100, + 'upload_date': '20141207', 'uploader_id': '330537914540', 'uploader': 'Виталий Добровольский', 'like_count': int, + 'age_limit': 0, }, + 'skip': 'Video has been blocked', }, { # metadataUrl 'url': 'http://ok.ru/video/63567059965189-0', @@ -35,13 +39,36 @@ class OdnoklassnikiIE(InfoExtractor): 'ext': 'mp4', 'title': 'Девушка без комплексов ...', 'duration': 191, + 'upload_date': '20150518', 'uploader_id': '534380003155', - 'uploader': 'Андрей Мещанинов', + 'uploader': '☭ Андрей Мещанинов ☭', 'like_count': int, + 'age_limit': 0, + }, + }, { + # YouTube embed (metadataUrl, provider == USER_YOUTUBE) + 'url': 'http://ok.ru/video/64211978996595-1', + 'md5': '5d7475d428845cd2e13bae6f1a992278', + 'info_dict': { + 'id': '64211978996595-1', + 'ext': 'mp4', + 'title': 'Космическая среда от 26 августа 2015', + 'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0', + 'duration': 440, + 'upload_date': '20150826', + 'uploader_id': '750099571', + 'uploader': 'Алина П', + 'age_limit': 0, }, }, { 'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452', 'only_matching': True, + }, { + 'url': 'http://www.ok.ru/video/20648036891', + 'only_matching': True, + }, { + 'url': 'http://www.ok.ru/videoembed/20648036891', + 'only_matching': True, }] def _real_extract(self, url): @@ -50,9 +77,16 @@ class OdnoklassnikiIE(InfoExtractor): webpage = self._download_webpage( 'http://ok.ru/video/%s' % video_id, video_id) + error = self._search_regex( + r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<', + webpage, 'error', default=None) + if error: + raise ExtractorError(error, expected=True) + player = self._parse_json( unescapeHTML(self._search_regex( - r'data-attributes="([^"]+)"', webpage, 'player')), + r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id, + webpage, 'player', group='player')), video_id) flashvars = player['flashvars'] @@ -62,7 +96,7 @@ class OdnoklassnikiIE(InfoExtractor): metadata = self._parse_json(metadata, video_id) else: metadata = self._download_json( - compat_urllib_parse.unquote(flashvars['metadataUrl']), + compat_urllib_parse_unquote(flashvars['metadataUrl']), video_id, 'Downloading metadata JSON') movie = metadata['movie'] @@ -85,16 +119,7 @@ class OdnoklassnikiIE(InfoExtractor): like_count = int_or_none(metadata.get('likeCount')) - quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd')) - - formats = [{ - 'url': f['url'], - 'ext': 'mp4', - 'format_id': f['name'], - 'quality': quality(f['name']), - } for f in metadata['videos']] - - return { + info = { 'id': video_id, 'title': title, 'thumbnail': thumbnail, @@ -104,5 +129,24 @@ class OdnoklassnikiIE(InfoExtractor): 'uploader_id': uploader_id, 'like_count': like_count, 'age_limit': age_limit, - 'formats': formats, } + + if metadata.get('provider') == 'USER_YOUTUBE': + info.update({ + '_type': 'url_transparent', + 'url': movie['contentId'], + }) + return info + + quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd')) + + formats = [{ + 'url': f['url'], + 'ext': 'mp4', + 'format_id': f['name'], + 'quality': quality(f['name']), + } for f in metadata['videos']] + self._sort_formats(formats) + + info['formats'] = formats + return info diff --git a/youtube_dl/extractor/onionstudios.py b/youtube_dl/extractor/onionstudios.py index 8fa507dec..0f1f448fe 100644 --- a/youtube_dl/extractor/onionstudios.py +++ b/youtube_dl/extractor/onionstudios.py @@ -49,19 +49,21 @@ class OnionStudiosIE(InfoExtractor): self._sort_formats(formats) title = self._search_regex( - r'share_title\s*=\s*"([^"]+)"', webpage, 'title') + r'share_title\s*=\s*(["\'])(?P<title>[^\1]+?)\1', + webpage, 'title', group='title') description = self._search_regex( - r'share_description\s*=\s*"([^"]+)"', webpage, - 'description', default=None) + r'share_description\s*=\s*(["\'])(?P<description>[^\1]+?)\1', + webpage, 'description', default=None, group='description') thumbnail = self._search_regex( - r'poster="([^"]+)"', webpage, 'thumbnail', default=False) + r'poster\s*=\s*(["\'])(?P<thumbnail>[^\1]+?)\1', + webpage, 'thumbnail', default=False, group='thumbnail') uploader_id = self._search_regex( - r'twitter_handle\s*=\s*"([^"]+)"', - webpage, 'uploader id', fatal=False) + 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:([^"]+)"', - webpage, 'uploader', default=False) + r'window\.channelName\s*=\s*(["\'])Embedded:(?P<uploader>[^\1]+?)\1', + webpage, 'uploader', default=False, group='uploader') return { 'id': video_id, diff --git a/youtube_dl/extractor/openfilm.py b/youtube_dl/extractor/openfilm.py deleted file mode 100644 index 2249657eb..000000000 --- a/youtube_dl/extractor/openfilm.py +++ /dev/null @@ -1,70 +0,0 @@ -from __future__ import unicode_literals - -import json - -from .common import InfoExtractor -from ..utils import ( - parse_iso8601, - compat_urllib_parse, - parse_age_limit, - int_or_none, -) - - -class OpenFilmIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)openfilm\.com/videos/(?P<id>.+)' - _TEST = { - 'url': 'http://www.openfilm.com/videos/human-resources-remastered', - 'md5': '42bcd88c2f3ec13b65edf0f8ad1cac37', - 'info_dict': { - 'id': '32736', - 'display_id': 'human-resources-remastered', - 'ext': 'mp4', - 'title': 'Human Resources (Remastered)', - 'description': 'Social Engineering in the 20th Century.', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 7164, - 'timestamp': 1334756988, - 'upload_date': '20120418', - 'uploader_id': '41117', - 'view_count': int, - 'age_limit': 0, - }, - } - - def _real_extract(self, url): - display_id = self._match_id(url) - - webpage = self._download_webpage(url, display_id) - - player = compat_urllib_parse.unquote_plus( - self._og_search_video_url(webpage)) - - video = json.loads(self._search_regex( - r'\bp=({.+?})(?:&|$)', player, 'video JSON')) - - video_url = '%s1.mp4' % video['location'] - video_id = video.get('video_id') - display_id = video.get('alias') or display_id - title = video.get('title') - description = video.get('description') - thumbnail = video.get('main_thumb') - duration = int_or_none(video.get('duration')) - timestamp = parse_iso8601(video.get('dt_published'), ' ') - uploader_id = video.get('user_id') - view_count = int_or_none(video.get('views_count')) - age_limit = parse_age_limit(video.get('age_limit')) - - return { - 'id': video_id, - 'display_id': display_id, - 'url': video_url, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'timestamp': timestamp, - 'uploader_id': uploader_id, - 'view_count': view_count, - 'age_limit': age_limit, - } diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index fec5d65ad..3448736a2 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -32,7 +32,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2365006249', 'ext': 'mp4', - 'title': 'A More Perfect Union', + 'title': 'Constitution USA with Peter Sagal - A More Perfect Union', 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a', 'duration': 3190, }, @@ -46,7 +46,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2365297690', 'ext': 'mp4', - 'title': 'Losing Iraq', + 'title': 'FRONTLINE - Losing Iraq', 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e', 'duration': 5050, }, @@ -60,7 +60,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2201174722', 'ext': 'mp4', - 'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist', + 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist', 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28', 'duration': 801, }, @@ -72,7 +72,7 @@ class PBSIE(InfoExtractor): 'id': '2365297708', 'ext': 'mp4', 'description': 'md5:68d87ef760660eb564455eb30ca464fe', - 'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full', + 'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full', 'duration': 6559, 'thumbnail': 're:^https?://.*\.jpg$', }, @@ -88,10 +88,11 @@ class PBSIE(InfoExtractor): 'display_id': 'killer-typhoon', 'ext': 'mp4', 'description': 'md5:c741d14e979fc53228c575894094f157', - 'title': 'Killer Typhoon', + 'title': 'NOVA - Killer Typhoon', 'duration': 3172, 'thumbnail': 're:^https?://.*\.jpg$', 'upload_date': '20140122', + 'age_limit': 10, }, 'params': { 'skip_download': True, # requires ffmpeg @@ -107,12 +108,46 @@ class PBSIE(InfoExtractor): { 'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/', 'info_dict': { - 'id': '2280706814', + 'id': '2276541483', 'display_id': 'player', 'ext': 'mp4', - 'title': 'Death and the Civil War', + 'title': 'American Experience - Death and the Civil War, Chapter 1', 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.', - 'duration': 6705, + 'duration': 682, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + }, + { + 'url': 'http://video.pbs.org/video/2365367186/', + 'info_dict': { + 'id': '2365367186', + 'display_id': '2365367186', + 'ext': 'mp4', + 'title': 'To Catch A Comet - Full Episode', + 'description': 'On November 12, 2014, billions of kilometers from Earth, spacecraft orbiter Rosetta and lander Philae did what no other had dared to attempt \u2014 land on the volatile surface of a comet as it zooms around the sun at 67,000 km/hr. The European Space Agency hopes this mission can help peer into our past and unlock secrets of our origins.', + 'duration': 3342, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, + 'skip': 'Expired', + }, + { + # Video embedded in iframe containing angle brackets as attribute's value (e.g. + # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see + # https://github.com/rg3/youtube-dl/issues/7059) + 'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/', + 'info_dict': { + 'id': '2365546844', + 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business', + 'ext': 'mp4', + 'title': "A Chef's Life - Season 3, Ep. 5: Prickly Business", + 'description': 'md5:61db2ddf27c9912f09c241014b118ed1', + 'duration': 1480, 'thumbnail': 're:^https?://.*\.jpg$', }, 'params': { @@ -120,6 +155,12 @@ class PBSIE(InfoExtractor): }, } ] + _ERRORS = { + 101: 'We\'re sorry, but this video is not yet available.', + 403: 'We\'re sorry, but this video is not available in your region due to right restrictions.', + 404: 'We are experiencing technical difficulties that are preventing us from playing the video at this time. Please check back again soon.', + 410: 'This video has expired and is no longer available for online streaming.', + } def _extract_webpage(self, url): mobj = re.match(self._VALID_URL, url) @@ -151,7 +192,7 @@ class PBSIE(InfoExtractor): return media_id, presumptive_id, upload_date url = self._search_regex( - r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']', + r'(?s)<iframe[^>]+?(?:[a-z-]+?=["\'].*?["\'][^>]+?)*?\bsrc=["\']([^\'"]+partnerplayer[^\'"]+)["\']', webpage, 'player URL') mobj = re.match(self._VALID_URL, url) @@ -197,13 +238,11 @@ class PBSIE(InfoExtractor): 'Downloading %s video url info' % encoding_name) if redirect_info['status'] == 'error': - if redirect_info['http_code'] == 403: - message = ( - 'The video is not available in your region due to ' - 'right restrictions') - else: - message = redirect_info['message'] - raise ExtractorError(message, expected=True) + raise ExtractorError( + '%s said: %s' % ( + self.IE_NAME, + self._ERRORS.get(redirect_info['http_code'], redirect_info['message'])), + expected=True) format_url = redirect_info.get('url') if not format_url: @@ -232,6 +271,12 @@ class PBSIE(InfoExtractor): 'url': closed_captions_url, }] + # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc) + # Try turning it to 'program - title' naming scheme if possible + alt_title = info.get('program', {}).get('title') + if alt_title: + info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title']) + return { 'id': video_id, 'display_id': display_id, diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py new file mode 100644 index 000000000..8ad936758 --- /dev/null +++ b/youtube_dl/extractor/periscope.py @@ -0,0 +1,99 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_request, +) +from ..utils import parse_iso8601 + + +class PeriscopeIE(InfoExtractor): + IE_DESC = 'Periscope' + _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P<id>[^/?#]+)' + _TEST = { + 'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==', + 'md5': '65b57957972e503fcbbaeed8f4fa04ca', + 'info_dict': { + 'id': '56102209', + 'ext': 'mp4', + 'title': 'Bec Boop - 🚠✈️🇬🇧 Fly above #London in Emirates Air Line cable car at night 🇬🇧✈️🚠 #BoopScope 🎀💗', + 'timestamp': 1438978559, + 'upload_date': '20150807', + 'uploader': 'Bec Boop', + 'uploader_id': '1465763', + }, + 'skip': 'Expires in 24 hours', + } + + def _call_api(self, method, token): + return self._download_json( + 'https://api.periscope.tv/api/v2/%s?token=%s' % (method, token), token) + + def _real_extract(self, url): + token = self._match_id(url) + + broadcast_data = self._call_api('getBroadcastPublic', token) + 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') + + title = '%s - %s' % (uploader, status) if uploader else status + state = broadcast.get('state').lower() + if state == 'running': + title = self._live_title(title) + timestamp = parse_iso8601(broadcast.get('created_at')) + + thumbnails = [{ + 'url': broadcast[image], + } for image in ('image_url', 'image_url_small') if broadcast.get(image)] + + stream = self._call_api('getAccessPublic', token) + + formats = [] + for format_id in ('replay', 'rtmp', 'hls', 'https_hls'): + video_url = stream.get(format_id + '_url') + if not video_url: + continue + f = { + 'url': video_url, + 'ext': 'flv' if format_id == 'rtmp' else 'mp4', + } + if format_id != 'rtmp': + f['protocol'] = 'm3u8_native' if state == 'ended' else 'm3u8' + formats.append(f) + self._sort_formats(formats) + + return { + 'id': broadcast.get('id') or token, + 'title': title, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'thumbnails': thumbnails, + 'formats': formats, + } + + +class QuickscopeIE(InfoExtractor): + IE_DESC = 'Quick Scope' + _VALID_URL = r'https?://watchonperiscope\.com/broadcast/(?P<id>\d+)' + _TEST = { + 'url': 'https://watchonperiscope.com/broadcast/56180087', + 'only_matching': True, + } + + def _real_extract(self, url): + broadcast_id = self._match_id(url) + request = compat_urllib_request.Request( + 'https://watchonperiscope.com/api/accessChannel', compat_urllib_parse.urlencode({ + 'broadcast_id': broadcast_id, + 'entry_ticket': '', + 'from_push': 'false', + 'uses_sessions': 'true', + }).encode('utf-8')) + return self.url_result( + self._download_json(request, broadcast_id)['share_url'], 'Periscope') diff --git a/youtube_dl/extractor/photobucket.py b/youtube_dl/extractor/photobucket.py index c66db3cdc..788411ccc 100644 --- a/youtube_dl/extractor/photobucket.py +++ b/youtube_dl/extractor/photobucket.py @@ -4,7 +4,7 @@ import json import re from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_unquote class PhotobucketIE(InfoExtractor): @@ -34,7 +34,7 @@ class PhotobucketIE(InfoExtractor): info_json = self._search_regex(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (.*?)\);', webpage, 'info json') info = json.loads(info_json) - url = compat_urllib_parse.unquote(self._html_search_regex(r'file=(.+\.mp4)', info['linkcodes']['html'], 'url')) + url = compat_urllib_parse_unquote(self._html_search_regex(r'file=(.+\.mp4)', info['linkcodes']['html'], 'url')) return { 'id': video_id, 'url': url, diff --git a/youtube_dl/extractor/played.py b/youtube_dl/extractor/played.py index 9fe1524f2..8a1c296dd 100644 --- a/youtube_dl/extractor/played.py +++ b/youtube_dl/extractor/played.py @@ -38,7 +38,7 @@ class PlayedIE(InfoExtractor): if m_error: raise ExtractorError(m_error.group('msg'), expected=True) - data = self._form_hidden_inputs(orig_webpage) + data = self._hidden_inputs(orig_webpage) self._sleep(2, video_id) diff --git a/youtube_dl/extractor/playtvak.py b/youtube_dl/extractor/playtvak.py new file mode 100644 index 000000000..e360404f7 --- /dev/null +++ b/youtube_dl/extractor/playtvak.py @@ -0,0 +1,181 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_urlparse, + compat_urllib_parse, +) +from ..utils import ( + ExtractorError, + int_or_none, + parse_iso8601, + qualities, +) + + +class PlaytvakIE(InfoExtractor): + IE_DESC = 'Playtvak.cz, iDNES.cz and Lidovky.cz' + _VALID_URL = r'https?://(?:.+?\.)?(?:playtvak|idnes|lidovky|metro)\.cz/.*\?(?:c|idvideo)=(?P<id>[^&]+)' + _TESTS = [{ + 'url': 'http://www.playtvak.cz/vyzente-vosy-a-srsne-ze-zahrady-dn5-/hodinovy-manzel.aspx?c=A150730_150323_hodinovy-manzel_kuko', + 'md5': '4525ae312c324b4be2f4603cc78ceb4a', + 'info_dict': { + 'id': 'A150730_150323_hodinovy-manzel_kuko', + 'ext': 'mp4', + 'title': 'Vyžeňte vosy a sršně ze zahrady', + 'description': 'md5:f93d398691044d303bc4a3de62f3e976', + 'thumbnail': 're:(?i)^https?://.*\.(?:jpg|png)$', + 'duration': 279, + 'timestamp': 1438732860, + 'upload_date': '20150805', + 'is_live': False, + } + }, { # live video test + 'url': 'http://slowtv.playtvak.cz/planespotting-0pr-/planespotting.aspx?c=A150624_164934_planespotting_cat', + 'info_dict': { + 'id': 'A150624_164934_planespotting_cat', + 'ext': 'flv', + 'title': 're:^Přímý přenos iDNES.cz [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': 'Sledujte provoz na ranveji Letiště Václava Havla v Praze', + 'thumbnail': 're:(?i)^https?://.*\.(?:jpg|png)$', + 'is_live': True, + }, + 'params': { + 'skip_download': True, # requires rtmpdump + }, + }, { # idnes.cz + 'url': 'http://zpravy.idnes.cz/pes-zavreny-v-aute-rozbijeni-okynek-v-aute-fj5-/domaci.aspx?c=A150809_104116_domaci_pku', + 'md5': '819832ba33cd7016e58a6658577fe289', + 'info_dict': { + 'id': 'A150809_104116_domaci_pku', + 'ext': 'mp4', + 'title': 'Zavřeli jsme mraženou pizzu do auta. Upekla se', + 'description': 'md5:01e73f02329e2e5760bd5eed4d42e3c2', + 'thumbnail': 're:(?i)^https?://.*\.(?:jpg|png)$', + 'duration': 39, + 'timestamp': 1438969140, + 'upload_date': '20150807', + 'is_live': False, + } + }, { # lidovky.cz + 'url': 'http://www.lidovky.cz/dalsi-demonstrace-v-praze-o-migraci-duq-/video.aspx?c=A150808_214044_ln-video_ELE', + 'md5': 'c7209ac4ba9d234d4ad5bab7485bcee8', + 'info_dict': { + 'id': 'A150808_214044_ln-video_ELE', + 'ext': 'mp4', + 'title': 'Táhni! Demonstrace proti imigrantům budila emoce', + 'description': 'md5:97c81d589a9491fbfa323c9fa3cca72c', + 'thumbnail': 're:(?i)^https?://.*\.(?:jpg|png)$', + 'timestamp': 1439052180, + 'upload_date': '20150808', + 'is_live': False, + } + }, { # metro.cz + 'url': 'http://www.metro.cz/video-pod-billboardem-se-na-vltavske-roztocil-kolotoc-deti-vozil-jen-par-hodin-1hx-/metro-extra.aspx?c=A141111_173251_metro-extra_row', + 'md5': '84fc1deedcac37b7d4a6ccae7c716668', + 'info_dict': { + 'id': 'A141111_173251_metro-extra_row', + 'ext': 'mp4', + 'title': 'Recesisté udělali z billboardu kolotoč', + 'description': 'md5:7369926049588c3989a66c9c1a043c4c', + 'thumbnail': 're:(?i)^https?://.*\.(?:jpg|png)$', + 'timestamp': 1415725500, + 'upload_date': '20141111', + 'is_live': False, + } + }, { + 'url': 'http://www.playtvak.cz/embed.aspx?idvideo=V150729_141549_play-porad_kuko', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + info_url = self._html_search_regex( + r'Misc\.videoFLV\(\s*{\s*data\s*:\s*"([^"]+)"', webpage, 'info url') + + parsed_url = compat_urlparse.urlparse(info_url) + + qs = compat_urlparse.parse_qs(parsed_url.query) + qs.update({ + 'reklama': ['0'], + 'type': ['js'], + }) + + info_url = compat_urlparse.urlunparse( + parsed_url._replace(query=compat_urllib_parse.urlencode(qs, True))) + + json_info = self._download_json( + info_url, video_id, + transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1]) + + item = None + for i in json_info['items']: + if i.get('type') == 'video' or i.get('type') == 'stream': + item = i + break + if not item: + raise ExtractorError('No suitable stream found') + + quality = qualities(('low', 'middle', 'high')) + + formats = [] + for fmt in item['video']: + video_url = fmt.get('file') + if not video_url: + continue + + format_ = fmt['format'] + format_id = '%s_%s' % (format_, fmt['quality']) + preference = None + + if format_ in ('mp4', 'webm'): + ext = format_ + elif format_ == 'rtmp': + ext = 'flv' + elif format_ == 'apple': + ext = 'mp4' + # Some streams have mp3 audio which does not play + # well with ffmpeg filter aac_adtstoasc + preference = -1 + elif format_ == 'adobe': # f4m manifest fails with 404 in 80% of requests + continue + else: # Other formats not supported yet + continue + + formats.append({ + 'url': video_url, + 'ext': ext, + 'format_id': format_id, + 'quality': quality(fmt.get('quality')), + 'preference': preference, + }) + self._sort_formats(formats) + + title = item['title'] + is_live = item['type'] == 'stream' + if is_live: + title = self._live_title(title) + description = self._og_search_description(webpage, default=None) or self._html_search_meta( + 'description', webpage, 'description') + timestamp = None + duration = None + if not is_live: + duration = int_or_none(item.get('length')) + timestamp = item.get('published') + if timestamp: + timestamp = parse_iso8601(timestamp[:-5]) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': item.get('image'), + 'duration': duration, + 'timestamp': timestamp, + 'is_live': is_live, + 'formats': formats, + } diff --git a/youtube_dl/extractor/playvid.py b/youtube_dl/extractor/playvid.py index c3e667e9e..2eb4fd96d 100644 --- a/youtube_dl/extractor/playvid.py +++ b/youtube_dl/extractor/playvid.py @@ -4,7 +4,8 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urllib_parse_unquote_plus, ) from ..utils import ( clean_html, @@ -44,7 +45,7 @@ class PlayvidIE(InfoExtractor): flashvars = self._html_search_regex( r'flashvars="(.+?)"', webpage, 'flashvars') - infos = compat_urllib_parse.unquote(flashvars).split(r'&') + infos = compat_urllib_parse_unquote(flashvars).split(r'&') for info in infos: videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$', info) if videovars_match: @@ -52,7 +53,7 @@ class PlayvidIE(InfoExtractor): val = videovars_match.group(2) if key == 'title': - video_title = compat_urllib_parse.unquote_plus(val) + video_title = compat_urllib_parse_unquote_plus(val) if key == 'duration': try: duration = int(val) diff --git a/youtube_dl/extractor/playwire.py b/youtube_dl/extractor/playwire.py index bdc71017b..6d138ef25 100644 --- a/youtube_dl/extractor/playwire.py +++ b/youtube_dl/extractor/playwire.py @@ -19,7 +19,7 @@ class PlaywireIE(InfoExtractor): 'id': '3353705', 'ext': 'mp4', 'title': 'S04_RM_UCL_Rus', - 'thumbnail': 're:^http://.*\.png$', + 'thumbnail': 're:^https?://.*\.png$', 'duration': 145.94, }, }, { diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py new file mode 100644 index 000000000..fd32836cc --- /dev/null +++ b/youtube_dl/extractor/pluralsight.py @@ -0,0 +1,207 @@ +from __future__ import unicode_literals + +import re +import json + +from .common import InfoExtractor +from ..compat import ( + compat_str, + compat_urllib_parse, + compat_urllib_request, + compat_urlparse, +) +from ..utils import ( + ExtractorError, + int_or_none, + parse_duration, +) + + +class PluralsightIE(InfoExtractor): + IE_NAME = 'pluralsight' + _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P<author>[^&]+)&name=(?P<name>[^&]+)(?:&mode=live)?&clip=(?P<clip>\d+)&course=(?P<course>[^&]+)' + _LOGIN_URL = 'https://www.pluralsight.com/id/' + _NETRC_MACHINE = 'pluralsight' + + _TEST = { + 'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas', + 'md5': '4d458cf5cf4c593788672419a8dd4cf8', + 'info_dict': { + 'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04', + 'ext': 'mp4', + 'title': 'Management of SQL Server - Demo Monitoring', + 'duration': 338, + }, + 'skip': 'Requires pluralsight account credentials', + } + + def _real_initialize(self): + self._login() + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + self.raise_login_required('Pluralsight account is required') + + login_page = self._download_webpage( + self._LOGIN_URL, None, 'Downloading login page') + + login_form = self._hidden_inputs(login_page) + + login_form.update({ + 'Username': username.encode('utf-8'), + 'Password': password.encode('utf-8'), + }) + + post_url = self._search_regex( + r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, + 'post url', default=self._LOGIN_URL, group='url') + + if not post_url.startswith('http'): + post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) + + request = compat_urllib_request.Request( + post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8')) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') + + response = self._download_webpage( + request, None, 'Logging in as %s' % username) + + error = self._search_regex( + r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>', + response, 'error message', default=None) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + author = mobj.group('author') + name = mobj.group('name') + clip_id = mobj.group('clip') + course = mobj.group('course') + + display_id = '%s-%s' % (name, clip_id) + + webpage = self._download_webpage(url, display_id) + + collection = self._parse_json( + self._search_regex( + r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', + webpage, 'modules'), + display_id) + + module, clip = None, None + + for module_ in collection: + if module_.get('moduleName') == name: + module = module_ + for clip_ in module_.get('clips', []): + clip_index = clip_.get('clipIndex') + if clip_index is None: + continue + if compat_str(clip_index) == clip_id: + clip = clip_ + break + + if not clip: + raise ExtractorError('Unable to resolve clip') + + QUALITIES = { + 'low': {'width': 640, 'height': 480}, + 'medium': {'width': 848, 'height': 640}, + 'high': {'width': 1024, 'height': 768}, + } + + ALLOWED_QUALITIES = ( + ('webm', ('high',)), + ('mp4', ('low', 'medium', 'high',)), + ) + + formats = [] + for ext, qualities in ALLOWED_QUALITIES: + for quality in qualities: + f = QUALITIES[quality].copy() + clip_post = { + 'a': author, + 'cap': 'false', + 'cn': clip_id, + 'course': course, + 'lc': 'en', + 'm': name, + 'mt': ext, + 'q': '%dx%d' % (f['width'], f['height']), + } + request = compat_urllib_request.Request( + 'http://www.pluralsight.com/training/Player/ViewClip', + json.dumps(clip_post).encode('utf-8')) + request.add_header('Content-Type', 'application/json;charset=utf-8') + format_id = '%s-%s' % (ext, quality) + clip_url = self._download_webpage( + request, display_id, 'Downloading %s URL' % format_id, fatal=False) + if not clip_url: + continue + f.update({ + 'url': clip_url, + 'ext': ext, + 'format_id': format_id, + }) + formats.append(f) + self._sort_formats(formats) + + # TODO: captions + # http://www.pluralsight.com/training/Player/ViewClip + cap = true + # or + # http://www.pluralsight.com/training/Player/Captions + # { a = author, cn = clip_id, lc = end, m = name } + + return { + 'id': clip['clipName'], + 'title': '%s - %s' % (module['title'], clip['title']), + 'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')), + 'creator': author, + 'formats': formats + } + + +class PluralsightCourseIE(InfoExtractor): + IE_NAME = 'pluralsight:course' + _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P<id>[^/]+)' + _TEST = { + # Free course from Pluralsight Starter Subscription for Microsoft TechNet + # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz + 'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas', + 'info_dict': { + 'id': 'hosting-sql-server-windows-azure-iaas', + 'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals', + 'description': 'md5:61b37e60f21c4b2f91dc621a977d0986', + }, + 'playlist_count': 31, + } + + def _real_extract(self, url): + course_id = self._match_id(url) + + # TODO: PSM cookie + + course = self._download_json( + 'http://www.pluralsight.com/data/course/%s' % course_id, + course_id, 'Downloading course JSON') + + title = course['title'] + description = course.get('description') or course.get('shortDescription') + + course_data = self._download_json( + 'http://www.pluralsight.com/data/course/content/%s' % course_id, + course_id, 'Downloading course data JSON') + + entries = [] + for module in course_data: + for clip in module.get('clips', []): + player_parameters = clip.get('playerParameters') + if not player_parameters: + continue + entries.append(self.url_result( + 'http://www.pluralsight.com/training/player?%s' % player_parameters, + 'Pluralsight')) + + return self.playlist_result(entries, course_id, title, description) diff --git a/youtube_dl/extractor/porn91.py b/youtube_dl/extractor/porn91.py index 72d1b2718..3e15533e9 100644 --- a/youtube_dl/extractor/porn91.py +++ b/youtube_dl/extractor/porn91.py @@ -22,6 +22,7 @@ class Porn91IE(InfoExtractor): 'title': '18岁大一漂亮学妹，水嫩性感，再爽一次！', 'ext': 'mp4', 'duration': 431, + 'age_limit': 18, } } @@ -68,4 +69,5 @@ class Porn91IE(InfoExtractor): 'url': video_url, 'duration': duration, 'comment_count': comment_count, + 'age_limit': self._rta_search(webpage), } diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 8172bc997..a656ad85a 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -5,7 +5,8 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urllib_parse_unquote_plus, compat_urllib_parse_urlparse, compat_urllib_request, ) @@ -19,7 +20,7 @@ from ..aes import ( class PornHubIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)' + _VALID_URL = r'https?://(?:[a-z]+\.)?pornhub\.com/(?:view_video\.php\?viewkey=|embed/)(?P<id>[0-9a-z]+)' _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': '882f488fa1f0026f023f33576004a2ed', @@ -33,6 +34,9 @@ class PornHubIE(InfoExtractor): }, { 'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d', 'only_matching': True, + }, { + 'url': 'http://fr.pornhub.com/view_video.php?viewkey=ph55ca2f9760862', + 'only_matching': True, }] @classmethod @@ -69,7 +73,7 @@ class PornHubIE(InfoExtractor): webpage, 'uploader', fatal=False) thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False) if thumbnail: - thumbnail = compat_urllib_parse.unquote(thumbnail) + thumbnail = compat_urllib_parse_unquote(thumbnail) view_count = self._extract_count( r'<span class="count">([\d,\.]+)</span> views', webpage, 'view') @@ -80,9 +84,9 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment') - video_urls = list(map(compat_urllib_parse.unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage))) + video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage))) if webpage.find('"encrypted":true') != -1: - password = compat_urllib_parse.unquote_plus( + password = compat_urllib_parse_unquote_plus( self._search_regex(r'"video_title":"([^"]+)', webpage, 'password')) video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) @@ -93,7 +97,7 @@ class PornHubIE(InfoExtractor): format = path.split('/')[5].split('_')[:2] format = "-".join(format) - m = re.match(r'^(?P<height>[0-9]+)P-(?P<tbr>[0-9]+)K$', format) + m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format) if m is None: height = None tbr = None diff --git a/youtube_dl/extractor/primesharetv.py b/youtube_dl/extractor/primesharetv.py index 9aa0c862a..304359dc5 100644 --- a/youtube_dl/extractor/primesharetv.py +++ b/youtube_dl/extractor/primesharetv.py @@ -29,7 +29,7 @@ class PrimeShareTVIE(InfoExtractor): if '>File not exist<' in webpage: raise ExtractorError('Video %s does not exist' % video_id, expected=True) - fields = self._form_hidden_inputs(webpage) + fields = self._hidden_inputs(webpage) headers = { 'Referer': url, diff --git a/youtube_dl/extractor/promptfile.py b/youtube_dl/extractor/promptfile.py index 81a63c7fc..8190ed676 100644 --- a/youtube_dl/extractor/promptfile.py +++ b/youtube_dl/extractor/promptfile.py @@ -35,7 +35,7 @@ class PromptFileIE(InfoExtractor): raise ExtractorError('Video %s does not exist' % video_id, expected=True) - fields = self._form_hidden_inputs(webpage) + fields = self._hidden_inputs(webpage) post = compat_urllib_parse.urlencode(fields) req = compat_urllib_request.Request(url, post) req.add_header('Content-type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py index 536a42dc8..effcf1db3 100644 --- a/youtube_dl/extractor/prosiebensat1.py +++ b/youtube_dl/extractor/prosiebensat1.py @@ -9,8 +9,11 @@ from ..compat import ( compat_urllib_parse, ) from ..utils import ( - unified_strdate, + ExtractorError, + determine_ext, + float_or_none, int_or_none, + unified_strdate, ) @@ -21,6 +24,11 @@ class ProSiebenSat1IE(InfoExtractor): _TESTS = [ { + # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242 + # in response to fixing https://github.com/rg3/youtube-dl/issues/6215: + # - malformed f4m manifest support + # - proper handling of URLs starting with `https?://` in 2.0 manifests + # - recursive child f4m manifests extraction 'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge', 'info_dict': { 'id': '2104602', @@ -208,7 +216,7 @@ class ProSiebenSat1IE(InfoExtractor): clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id') access_token = 'prosieben' - client_name = 'kolibri-1.12.6' + 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({ @@ -218,10 +226,13 @@ class ProSiebenSat1IE(InfoExtractor): 'ids': clip_id, }) - videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON') + video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0] + + if video.get('is_protected') is True: + raise ExtractorError('This video is DRM protected.', expected=True) - duration = float(videos[0]['duration']) - source_ids = [source['id'] for source in videos[0]['sources']] + duration = float_or_none(video.get('duration')) + source_ids = [source['id'] for source in video['sources']] source_ids_str = ','.join(map(str, source_ids)) g = '01!8d8F_)r9]4s[qeuXfP%' @@ -275,8 +286,9 @@ class ProSiebenSat1IE(InfoExtractor): 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>.+)$', source['url']) + mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url) if not mobj: continue path = mobj.group('path') @@ -293,9 +305,11 @@ class ProSiebenSat1IE(InfoExtractor): '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'], + 'url': source_url, 'vbr': fix_bitrate(source['bitrate']), }) diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py index 1654a641f..c98539f6a 100644 --- a/youtube_dl/extractor/qqmusic.py +++ b/youtube_dl/extractor/qqmusic.py @@ -25,7 +25,7 @@ class QQMusicIE(InfoExtractor): 'id': '004295Et37taLD', 'ext': 'mp3', 'title': '可惜没如果', - 'upload_date': '20141227', + 'release_date': '20141227', 'creator': '林俊杰', 'description': 'md5:d327722d0361576fde558f1ac68a7065', 'thumbnail': 're:^https?://.*\.jpg$', @@ -38,11 +38,26 @@ class QQMusicIE(InfoExtractor): 'id': '004MsGEo3DdNxV', 'ext': 'mp3', 'title': '如果', - 'upload_date': '20050626', + 'release_date': '20050626', 'creator': '李季美', 'description': 'md5:46857d5ed62bc4ba84607a805dccf437', 'thumbnail': 're:^https?://.*\.jpg$', } + }, { + 'note': 'lyrics not in .lrc format', + 'url': 'http://y.qq.com/#type=song&mid=001JyApY11tIp6', + 'info_dict': { + 'id': '001JyApY11tIp6', + 'ext': 'mp3', + 'title': 'Shadows Over Transylvania', + 'release_date': '19970225', + 'creator': 'Dark Funeral', + 'description': 'md5:ed14d5bd7ecec19609108052c25b2c11', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, + }, }] _FORMATS = { @@ -112,15 +127,27 @@ class QQMusicIE(InfoExtractor): self._check_formats(formats, mid) self._sort_formats(formats) - return { + actual_lrc_lyrics = ''.join( + line + '\n' for line in re.findall( + r'(?m)^(\[[0-9]{2}:[0-9]{2}(?:\.[0-9]{2,})?\][^\n]*|\[[^\]]*\])', lrc_content)) + + info_dict = { 'id': mid, 'formats': formats, 'title': song_name, - 'upload_date': publish_time, + 'release_date': publish_time, 'creator': singer, 'description': lrc_content, - 'thumbnail': thumbnail_url, + 'thumbnail': thumbnail_url } + if actual_lrc_lyrics: + info_dict['subtitles'] = { + 'origin': [{ + 'ext': 'lrc', + 'data': actual_lrc_lyrics, + }] + } + return info_dict class QQPlaylistBaseIE(InfoExtractor): diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py index 1631faf29..7ff1d06c4 100644 --- a/youtube_dl/extractor/rai.py +++ b/youtube_dl/extractor/rai.py @@ -5,6 +5,7 @@ import re from .common import InfoExtractor from ..compat import ( compat_urllib_parse, + compat_urlparse, ) from ..utils import ( parse_duration, @@ -72,6 +73,18 @@ class RaiIE(InfoExtractor): 'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!', 'uploader': 'RaiTre', } + }, + { + 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', + 'md5': '037104d2c14132887e5e4cf114569214', + 'info_dict': { + 'id': '0c7a664b-d0f4-4b2c-8835-3f82e46f433e', + 'ext': 'flv', + 'title': 'Il pacco', + 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', + 'uploader': 'RaiTre', + 'upload_date': '20141221', + }, } ] @@ -90,11 +103,14 @@ class RaiIE(InfoExtractor): relinker_url = self._extract_relinker_url(webpage) if not relinker_url: - iframe_path = self._search_regex( - r'<iframe[^>]+src="/?(dl/[^"]+\?iframe\b[^"]*)"', + 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) webpage = self._download_webpage( - '%s/%s' % (host, iframe_path), video_id) + iframe_url, video_id) relinker_url = self._extract_relinker_url(webpage) relinker = self._download_json( diff --git a/youtube_dl/extractor/rds.py b/youtube_dl/extractor/rds.py new file mode 100644 index 000000000..796adfdf9 --- /dev/null +++ b/youtube_dl/extractor/rds.py @@ -0,0 +1,73 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + parse_duration, + parse_iso8601, +) + + +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+)' + + _TESTS = [{ + 'url': 'http://www.rds.ca/videos/football/nfl/fowler-jr-prend-la-direction-de-jacksonville-3.1132799', + 'info_dict': { + 'id': '3.1132799', + 'display_id': 'fowler-jr-prend-la-direction-de-jacksonville', + 'ext': 'mp4', + 'title': 'Fowler Jr. prend la direction de Jacksonville', + 'description': 'Dante Fowler Jr. est le troisième choix du repêchage 2015 de la NFL. ', + 'timestamp': 1430397346, + 'upload_date': '20150430', + 'duration': 154.354, + 'age_limit': 0, + } + }, { + 'url': 'http://www.rds.ca/vid%C3%A9os/un-voyage-positif-3.877934', + 'only_matching': True, + }] + + 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) + + # 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( + '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( + [r'<link[^>]+itemprop="thumbnailUrl"[^>]+href="([^"]+)"', + r'<span[^>]+itemprop="thumbnailUrl"[^>]+content="([^"]+)"'], + webpage, 'thumbnail', fatal=False) + timestamp = parse_iso8601(self._search_regex( + r'<span[^>]+itemprop="uploadDate"[^>]+content="([^"]+)"', + webpage, 'upload date', fatal=False)) + duration = parse_duration(self._search_regex( + r'<span[^>]+itemprop="duration"[^>]+content="([^"]+)"', + webpage, 'duration', fatal=False)) + age_limit = self._family_friendly_search(webpage) + + return { + 'id': video_id, + 'display_id': display_id, + 'url': video_url, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + 'duration': duration, + 'age_limit': age_limit, + } diff --git a/youtube_dl/extractor/rtbf.py b/youtube_dl/extractor/rtbf.py index e4215d546..04a66df90 100644 --- a/youtube_dl/extractor/rtbf.py +++ b/youtube_dl/extractor/rtbf.py @@ -36,7 +36,7 @@ class RTBFIE(InfoExtractor): data = self._parse_json( unescapeHTML(self._search_regex( - r'data-video="([^"]+)"', webpage, 'data video')), + r'data-media="([^"]+)"', webpage, 'data video')), video_id) if data.get('provider').lower() == 'youtube': diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py index 04158b993..d9cfbf180 100644 --- a/youtube_dl/extractor/rte.py +++ b/youtube_dl/extractor/rte.py @@ -9,16 +9,16 @@ from ..utils import ( class RteIE(InfoExtractor): - _VALID_URL = r'http?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/(?P<id>[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)' _TEST = { - 'url': 'http://www.rte.ie/player/de/show/10363114/', + 'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/', 'info_dict': { - 'id': '10363114', + 'id': '10478715', 'ext': 'mp4', - 'title': 'One News', + 'title': 'Watch iWitness online', 'thumbnail': 're:^https?://.*\.jpg$', - 'description': 'The One O\'Clock News followed by Weather.', - 'duration': 436.844, + 'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.', + 'duration': 60.046, }, 'params': { 'skip_download': 'f4m fails with --test atm' diff --git a/youtube_dl/extractor/rtl2.py b/youtube_dl/extractor/rtl2.py index 72cd80498..25f7faf76 100644 --- a/youtube_dl/extractor/rtl2.py +++ b/youtube_dl/extractor/rtl2.py @@ -1,6 +1,7 @@ # encoding: utf-8 from __future__ import unicode_literals +import re from .common import InfoExtractor @@ -8,22 +9,28 @@ class RTL2IE(InfoExtractor): _VALID_URL = r'http?://(?:www\.)?rtl2\.de/[^?#]*?/(?P<id>[^?#/]*?)(?:$|/(?:$|[?#]))' _TESTS = [{ 'url': 'http://www.rtl2.de/sendung/grip-das-motormagazin/folge/folge-203-0', - 'md5': 'bfcc179030535b08dc2b36b469b5adc7', 'info_dict': { 'id': 'folge-203-0', 'ext': 'f4v', 'title': 'GRIP sucht den Sommerkönig', 'description': 'Matthias, Det und Helge treten gegeneinander an.' }, + 'params': { + # rtmp download + 'skip_download': True, + }, }, { 'url': 'http://www.rtl2.de/sendung/koeln-50667/video/5512-anna/21040-anna-erwischt-alex/', - 'md5': 'ffcd517d2805b57ce11a58a2980c2b02', 'info_dict': { 'id': '21040-anna-erwischt-alex', 'ext': 'mp4', 'title': 'Anna erwischt Alex!', 'description': 'Anna ist Alex\' Tochter bei Köln 50667.' }, + 'params': { + # rtmp download + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -34,12 +41,18 @@ class RTL2IE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - vico_id = self._html_search_regex( - r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id') - vivi_id = self._html_search_regex( - r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id') + mobj = re.search( + r'<div[^>]+data-collection="(?P<vico_id>\d+)"[^>]+data-video="(?P<vivi_id>\d+)"', + webpage) + if mobj: + vico_id = mobj.group('vico_id') + vivi_id = mobj.group('vivi_id') + else: + vico_id = self._html_search_regex( + r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id') + vivi_id = self._html_search_regex( + r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id') info_url = 'http://www.rtl2.de/video/php/get_video.php?vico_id=' + vico_id + '&vivi_id=' + vivi_id - webpage = self._download_webpage(info_url, '') info = self._download_json(info_url, video_id) video_info = info['video'] diff --git a/youtube_dl/extractor/rtlnl.py b/youtube_dl/extractor/rtlnl.py index a4d3d73ff..543d94417 100644 --- a/youtube_dl/extractor/rtlnl.py +++ b/youtube_dl/extractor/rtlnl.py @@ -44,6 +44,21 @@ class RtlNlIE(InfoExtractor): '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) + '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$', + 'timestamp': 1437233400, + 'upload_date': '20150718', + 'duration': 30.474, + }, + 'params': { + 'skip_download': True, + }, + }, { # encrypted m3u8 streams, georestricted 'url': 'http://www.rtlxl.nl/#!/afl-2-257632/52a74543-c504-4cde-8aa8-ec66fe8d68a7', 'only_matching': True, @@ -59,22 +74,29 @@ class RtlNlIE(InfoExtractor): uuid) material = info['material'][0] - progname = info['abstracts'][0]['name'] - subtitle = material['title'] or info['episodes'][0]['name'] - description = material.get('synopsis') or info['episodes'][0]['synopsis'] + title = info['abstracts'][0]['name'] + subtitle = material.get('title') + if subtitle: + title += ' - %s' % subtitle + description = material.get('synopsis') meta = info.get('meta', {}) - # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118) - # NB: nowadays, recent ffmpeg and avconv can handle these encrypted streams, so - # this adaptive -> flash workaround is not required in general, but it also - # allows bypassing georestriction therefore is retained for now. - videopath = material['videopath'].replace('/adaptive/', '/flash/') + # m3u8 streams are encrypted and may not be handled properly by older ffmpeg/avconv. + # To workaround this previously adaptive -> flash trick was used to obtain + # unencrypted m3u8 streams (see https://github.com/rg3/youtube-dl/issues/4118) + # and bypass georestrictions as well. + # Currently, unencrypted m3u8 playlists are (intentionally?) invalid and therefore + # unusable albeit can be fixed by simple string replacement (see + # https://github.com/rg3/youtube-dl/pull/6337) + # Since recent ffmpeg and avconv handle encrypted streams just fine encrypted + # streams are used now. + videopath = material['videopath'] m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4') - video_urlpart = videopath.split('/flash/')[1][:-5] + video_urlpart = videopath.split('/adaptive/')[1][:-5] PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4' formats.extend([ @@ -107,7 +129,7 @@ class RtlNlIE(InfoExtractor): return { 'id': uuid, - 'title': '%s - %s' % (progname, subtitle), + 'title': title, 'formats': formats, 'timestamp': material['original_date'], 'description': description, diff --git a/youtube_dl/extractor/rtp.py b/youtube_dl/extractor/rtp.py index ecf4939cd..82b323cdd 100644 --- a/youtube_dl/extractor/rtp.py +++ b/youtube_dl/extractor/rtp.py @@ -18,6 +18,10 @@ class RTPIE(InfoExtractor): 'description': 'As paixões musicais de António Cartaxo e António Macedo', 'thumbnail': 're:^https?://.*\.jpg', }, + 'params': { + # rtmp download + 'skip_download': True, + }, }, { 'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas', 'only_matching': True, diff --git a/youtube_dl/extractor/rts.py b/youtube_dl/extractor/rts.py index 9fbe239d8..12639f08b 100644 --- a/youtube_dl/extractor/rts.py +++ b/youtube_dl/extractor/rts.py @@ -19,7 +19,16 @@ from ..utils import ( class RTSIE(InfoExtractor): IE_DESC = 'RTS.ch' - _VALID_URL = r'https?://(?:www\.)?rts\.ch/(?:(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html|play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+))' + _VALID_URL = r'''(?x) + (?: + rts:(?P<rts_id>\d+)| + https?:// + (?:www\.)?rts\.ch/ + (?: + (?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html| + play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+) + ) + )''' _TESTS = [ { @@ -123,6 +132,15 @@ class RTSIE(InfoExtractor): }, }, { + # article with videos on rhs + 'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html', + 'info_dict': { + 'id': '6693917', + 'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse', + }, + 'playlist_mincount': 5, + }, + { 'url': 'http://www.rts.ch/play/tv/le-19h30/video/le-chantier-du-nouveau-parlement-vaudois-a-permis-une-trouvaille-historique?id=6348280', 'only_matching': True, } @@ -130,7 +148,7 @@ class RTSIE(InfoExtractor): def _real_extract(self, url): m = re.match(self._VALID_URL, url) - video_id = m.group('id') or m.group('id_new') + video_id = m.group('rts_id') or m.group('id') or m.group('id_new') display_id = m.group('display_id') or m.group('display_id_new') def download_json(internal_id): @@ -143,6 +161,15 @@ class RTSIE(InfoExtractor): # video_id extracted out of URL is not always a real id if 'video' not in all_info and 'audio' not in all_info: page = self._download_webpage(url, display_id) + + # article with videos on rhs + videos = re.findall( + r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:rts:video:(\d+)"', + page) + if videos: + entries = [self.url_result('rts:%s' % video_urn, 'RTS') for video_urn in videos] + return self.playlist_result(entries, video_id, self._og_search_title(page)) + internal_id = self._html_search_regex( r'<(?:video|audio) data-id="([0-9]+)"', page, 'internal video id') diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py index 82cd98ac7..5b97d33ca 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -6,7 +6,7 @@ import re import time from .common import InfoExtractor -from ..compat import compat_urlparse +from ..compat import compat_urllib_request, compat_urlparse from ..utils import ( ExtractorError, float_or_none, @@ -102,7 +102,9 @@ class RTVEALaCartaIE(InfoExtractor): if info['state'] == 'DESPU': raise ExtractorError('The video is no longer available', expected=True) png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id) - png = self._download_webpage(png_url, video_id, 'Downloading url information') + png_request = compat_urllib_request.Request(png_url) + png_request.add_header('Referer', url) + png = self._download_webpage(png_request, video_id, 'Downloading url information') video_url = _decrypt_url(png) if not video_url.endswith('.f4m'): auth_url = video_url.replace( diff --git a/youtube_dl/extractor/rtvnh.py b/youtube_dl/extractor/rtvnh.py new file mode 100644 index 000000000..7c9d4b0cd --- /dev/null +++ b/youtube_dl/extractor/rtvnh.py @@ -0,0 +1,47 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ExtractorError + + +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', + 'info_dict': { + 'id': '131946', + 'ext': 'mp4', + 'title': 'Grote zoektocht in zee bij Zandvoort naar vermiste vrouw', + 'thumbnail': 're:^https?:.*\.jpg$' + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + meta = self._parse_json(self._download_webpage( + 'http://www.rtvnh.nl/video/json?m=' + video_id, video_id), video_id) + + status = meta.get('status') + if status != 200: + 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']}) + + return { + 'id': video_id, + 'title': meta['title'].strip(), + 'thumbnail': meta.get('image'), + 'formats': formats + } diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py index 5b1c3577a..d94dc7399 100644 --- a/youtube_dl/extractor/rutube.py +++ b/youtube_dl/extractor/rutube.py @@ -30,6 +30,7 @@ class RutubeIE(InfoExtractor): 'uploader': 'NTDRussian', 'uploader_id': '29790', 'upload_date': '20131016', + 'age_limit': 0, }, 'params': { # It requires ffmpeg (m3u8 download) diff --git a/youtube_dl/extractor/ruutu.py b/youtube_dl/extractor/ruutu.py index 4e22628d0..a16b73ff4 100644 --- a/youtube_dl/extractor/ruutu.py +++ b/youtube_dl/extractor/ruutu.py @@ -6,19 +6,19 @@ from ..compat import compat_urllib_parse_urlparse from ..utils import ( determine_ext, int_or_none, + xpath_attr, xpath_text, ) class RuutuIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?ruutu\.fi/ohjelmat/(?:[^/?#]+/)*(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?ruutu\.fi/video/(?P<id>\d+)' _TESTS = [ { - 'url': 'http://www.ruutu.fi/ohjelmat/oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi', + 'url': 'http://www.ruutu.fi/video/2058907', 'md5': 'ab2093f39be1ca8581963451b3c0234f', 'info_dict': { 'id': '2058907', - 'display_id': 'oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi', 'ext': 'mp4', 'title': 'Oletko aina halunnut tietää mitä tapahtuu vain hetki ennen lähetystä? - Nyt se selvisi!', 'description': 'md5:cfc6ccf0e57a814360df464a91ff67d6', @@ -28,14 +28,13 @@ class RuutuIE(InfoExtractor): }, }, { - 'url': 'http://www.ruutu.fi/ohjelmat/superpesis/superpesis-katso-koko-kausi-ruudussa', + 'url': 'http://www.ruutu.fi/video/2057306', 'md5': '065a10ae4d5b8cfd9d0c3d332465e3d9', 'info_dict': { 'id': '2057306', - 'display_id': 'superpesis-katso-koko-kausi-ruudussa', 'ext': 'mp4', 'title': 'Superpesis: katso koko kausi Ruudussa', - 'description': 'md5:44c44a99fdbe5b380ab74ebd75f0af77', + 'description': 'md5:da2736052fef3b2bd5e0005e63c25eac', 'thumbnail': 're:^https?://.*\.jpg$', 'duration': 40, 'age_limit': 0, @@ -44,29 +43,10 @@ class RuutuIE(InfoExtractor): ] def _real_extract(self, url): - display_id = self._match_id(url) + video_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - - video_id = self._search_regex( - r'data-media-id="(\d+)"', webpage, 'media id') - - video_xml_url = None - - media_data = self._search_regex( - r'jQuery\.extend\([^,]+,\s*(.+?)\);', webpage, - 'media data', default=None) - if media_data: - media_json = self._parse_json(media_data, display_id, fatal=False) - if media_json: - xml_url = media_json.get('ruutuplayer', {}).get('xmlUrl') - if xml_url: - video_xml_url = xml_url.replace('{ID}', video_id) - - if not video_xml_url: - video_xml_url = 'http://gatling.ruutu.fi/media-xml-cache?id=%s' % video_id - - video_xml = self._download_xml(video_xml_url, video_id) + video_xml = self._download_xml( + 'http://gatling.ruutu.fi/media-xml-cache?id=%s' % video_id, video_id) formats = [] processed_urls = [] @@ -94,7 +74,7 @@ class RuutuIE(InfoExtractor): preference = -1 if proto == 'rtmp' else 1 label = child.get('label') tbr = int_or_none(child.get('bitrate')) - width, height = [int_or_none(x) for x in child.get('resolution', '').split('x')] + width, height = [int_or_none(x) for x in child.get('resolution', 'x').split('x')[:2]] formats.append({ 'format_id': '%s-%s' % (proto, label if label else tbr), 'url': video_url, @@ -109,10 +89,9 @@ class RuutuIE(InfoExtractor): return { 'id': video_id, - 'display_id': display_id, - 'title': self._og_search_title(webpage), - 'description': self._og_search_description(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), + 'title': xpath_attr(video_xml, './/Behavior/Program', 'program_name', 'title', fatal=True), + 'description': xpath_attr(video_xml, './/Behavior/Program', 'description', 'description'), + 'thumbnail': xpath_attr(video_xml, './/Behavior/Startpicture', 'href', 'thumbnail'), 'duration': int_or_none(xpath_text(video_xml, './/Runtime', 'duration')), 'age_limit': int_or_none(xpath_text(video_xml, './/AgeLimit', 'age limit')), 'formats': formats, diff --git a/youtube_dl/extractor/safari.py b/youtube_dl/extractor/safari.py index f3c80708c..a602af692 100644 --- a/youtube_dl/extractor/safari.py +++ b/youtube_dl/extractor/safari.py @@ -20,7 +20,6 @@ from ..utils import ( class SafariBaseIE(InfoExtractor): _LOGIN_URL = 'https://www.safaribooksonline.com/accounts/login/' _SUCCESSFUL_LOGIN_REGEX = r'<a href="/accounts/logout/"[^>]*>Sign Out</a>' - _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to supply credentials for safaribooksonline.com' _NETRC_MACHINE = 'safari' _API_BASE = 'https://www.safaribooksonline.com/api/v1/book' @@ -37,9 +36,7 @@ class SafariBaseIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - raise ExtractorError( - self._ACCOUNT_CREDENTIALS_HINT, - expected=True) + self.raise_login_required('safaribooksonline.com account is required') headers = std_headers if 'Referer' not in headers: diff --git a/youtube_dl/extractor/sbs.py b/youtube_dl/extractor/sbs.py index d4bd1a0d7..d6ee2d9e2 100644 --- a/youtube_dl/extractor/sbs.py +++ b/youtube_dl/extractor/sbs.py @@ -1,17 +1,12 @@ # -*- coding: utf-8 -*- from __future__ import unicode_literals -import re from .common import InfoExtractor -from ..utils import ( - js_to_json, - remove_end, -) class SBSIE(InfoExtractor): IE_DESC = 'sbs.com.au' - _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/ondemand/video/(?:single/)?(?P<id>[0-9]+)' + _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/(?:ondemand|news)/video/(?:single/)?(?P<id>[0-9]+)' _TESTS = [{ # Original URL is handled by the generic IE which finds the iframe: @@ -21,39 +16,36 @@ class SBSIE(InfoExtractor): 'info_dict': { 'id': '320403011771', 'ext': 'mp4', - 'title': 'Dingo Conservation', - 'description': 'Dingoes are on the brink of extinction; most of the animals we think are dingoes are in fact crossbred with wild dogs. This family run a dingo conservation park to prevent their extinction', + 'title': 'Dingo Conservation (The Feed)', + 'description': 'md5:f250a9856fca50d22dec0b5b8015f8a5', 'thumbnail': 're:http://.*\.jpg', + 'duration': 308, }, - 'add_ies': ['generic'], }, { 'url': 'http://www.sbs.com.au/ondemand/video/320403011771/Dingo-Conservation-The-Feed', 'only_matching': True, + }, { + 'url': 'http://www.sbs.com.au/news/video/471395907773/The-Feed-July-9', + 'only_matching': True, }] def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - player = self._search_regex( - r'(?s)playerParams\.releaseUrls\s*=\s*(\{.*?\n\});\n', - webpage, 'player') - player = re.sub(r"'\s*\+\s*[\da-zA-Z_]+\s*\+\s*'", '', player) - - release_urls = self._parse_json(js_to_json(player), video_id) + webpage = self._download_webpage( + 'http://www.sbs.com.au/ondemand/video/single/%s?context=web' % video_id, video_id) - theplatform_url = release_urls.get('progressive') or release_urls['standard'] + player_params = self._parse_json( + self._search_regex( + r'(?s)var\s+playerParams\s*=\s*({.+?});', webpage, 'playerParams'), + video_id) - title = remove_end(self._og_search_title(webpage), ' (The Feed)') - description = self._html_search_meta('description', webpage) - thumbnail = self._og_search_thumbnail(webpage) + urls = player_params['releaseUrls'] + theplatform_url = (urls.get('progressive') or urls.get('standard') or + urls.get('html') or player_params['relatedItemsURL']) return { '_type': 'url_transparent', 'id': video_id, 'url': theplatform_url, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, } diff --git a/youtube_dl/extractor/screenwavemedia.py b/youtube_dl/extractor/screenwavemedia.py index d1ab66b32..05f93904c 100644 --- a/youtube_dl/extractor/screenwavemedia.py +++ b/youtube_dl/extractor/screenwavemedia.py @@ -7,12 +7,13 @@ from .common import InfoExtractor from ..utils import ( int_or_none, unified_strdate, + js_to_json, ) class ScreenwaveMediaIE(InfoExtractor): - _VALID_URL = r'http://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?[^"]*\bid=(?P<id>.+)' - + _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', 'only_matching': True, @@ -22,59 +23,71 @@ class ScreenwaveMediaIE(InfoExtractor): video_id = self._match_id(url) playerdata = self._download_webpage( - 'http://player.screenwavemedia.com/play/player.php?id=%s' % video_id, + 'http://player.screenwavemedia.com/player.php?id=%s' % video_id, video_id, 'Downloading player webpage') vidtitle = self._search_regex( r'\'vidtitle\'\s*:\s*"([^"]+)"', playerdata, 'vidtitle').replace('\\/', '/') - vidurl = self._search_regex( - r'\'vidurl\'\s*:\s*"([^"]+)"', playerdata, 'vidurl').replace('\\/', '/') - - videolist_url = None - - mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata) - if mobj: - videoserver = mobj.group('videoserver') - mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata) - vidid = mobj.group('vidid') if mobj else video_id - videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid) - else: - mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata) - if mobj: - videolist_url = mobj.group('smil') - - if videolist_url: - videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML') - formats = [] - baseurl = vidurl[:vidurl.rfind('/') + 1] - for video in videolist.findall('.//video'): - src = video.get('src') - if not src: + + playerconfig = self._download_webpage( + 'http://player.screenwavemedia.com/player.js', + video_id, 'Downloading playerconfig webpage') + + videoserver = self._search_regex(r'SWMServer\s*=\s*"([\d\.]+)"', playerdata, 'videoserver') + + sources = self._parse_json( + js_to_json( + re.sub( + r'(?s)/\*.*?\*/', '', + self._search_regex( + r"sources\s*:\s*(\[[^\]]+?\])", playerconfig, + 'sources', + ).replace( + "' + thisObj.options.videoserver + '", + videoserver + ).replace( + "' + playerVidId + '", + video_id + ) + ) + ), + video_id, fatal=False + ) + + # Fallback to hardcoded sources if JS changes again + if not sources: + self.report_warning('Falling back to a hardcoded list of streams') + sources = [{ + 'file': 'http://%s/vod/%s_%s.mp4' % (videoserver, video_id, format_id), + 'type': 'mp4', + 'label': format_label, + } for format_id, format_label in ( + ('low', '144p Low'), ('med', '160p Med'), ('high', '360p High'), ('hd1', '720p HD1'))] + sources.append({ + 'file': 'http://%s/vod/smil:%s.smil/playlist.m3u8' % (videoserver, video_id), + 'type': 'hls', + }) + + formats = [] + for source in sources: + if source['type'] == 'hls': + formats.extend(self._extract_m3u8_formats(source['file'], video_id)) + else: + file_ = source.get('file') + if not file_: continue - file_ = src.partition(':')[-1] - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - bitrate = int_or_none(video.get('system-bitrate'), scale=1000) - format = { - 'url': baseurl + file_, - 'format_id': src.rpartition('.')[0].rpartition('_')[-1], - } - if width or height: - format.update({ - 'tbr': bitrate, - 'width': width, - 'height': height, - }) - else: - format.update({ - 'abr': bitrate, - 'vcodec': 'none', - }) - formats.append(format) - else: - formats = [{ - 'url': vidurl, - }] + format_label = source.get('label') + format_id = self._search_regex( + r'_(.+?)\.[^.]+$', file_, 'format id', default=None) + height = int_or_none(self._search_regex( + r'^(\d+)[pP]', format_label, 'height', default=None)) + formats.append({ + 'url': source['file'], + 'format_id': format_id, + 'format': format_label, + 'ext': source.get('type'), + 'height': height, + }) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/senateisvp.py b/youtube_dl/extractor/senateisvp.py index 9c53704ea..474ebb49b 100644 --- a/youtube_dl/extractor/senateisvp.py +++ b/youtube_dl/extractor/senateisvp.py @@ -121,9 +121,9 @@ class SenateISVPIE(InfoExtractor): 'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=', }] else: - hdcore_sign = '?hdcore=3.1.0' + hdcore_sign = 'hdcore=3.1.0' url_params = (domain, video_id, stream_num) - f4m_url = '%s/z/%s_1@%s/manifest.f4m' % url_params + hdcore_sign + f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'): # URLs without the extra param induce an 404 error diff --git a/youtube_dl/extractor/sexykarma.py b/youtube_dl/extractor/sexykarma.py index 6446d26dc..e33483674 100644 --- a/youtube_dl/extractor/sexykarma.py +++ b/youtube_dl/extractor/sexykarma.py @@ -29,6 +29,7 @@ class SexyKarmaIE(InfoExtractor): 'view_count': int, 'comment_count': int, 'categories': list, + 'age_limit': 18, } }, { 'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html', diff --git a/youtube_dl/extractor/shahid.py b/youtube_dl/extractor/shahid.py new file mode 100644 index 000000000..f76fb12c0 --- /dev/null +++ b/youtube_dl/extractor/shahid.py @@ -0,0 +1,107 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_urllib_parse +from ..utils import ( + ExtractorError, + int_or_none, + parse_iso8601, +) + + +class ShahidIE(InfoExtractor): + _VALID_URL = r'https?://shahid\.mbc\.net/ar/episode/(?P<id>\d+)/?' + _TESTS = [{ + 'url': 'https://shahid.mbc.net/ar/episode/90574/%D8%A7%D9%84%D9%85%D9%84%D9%83-%D8%B9%D8%A8%D8%AF%D8%A7%D9%84%D9%84%D9%87-%D8%A7%D9%84%D8%A5%D9%86%D8%B3%D8%A7%D9%86-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-3.html', + 'info_dict': { + 'id': '90574', + 'ext': 'mp4', + 'title': 'الملك عبدالله الإنسان الموسم 1 كليب 3', + 'description': 'الفيلم الوثائقي - الملك عبد الله الإنسان', + 'duration': 2972, + 'timestamp': 1422057420, + 'upload_date': '20150123', + }, + 'params': { + # m3u8 download + 'skip_download': True, + } + }, { + # shahid plus subscriber only + 'url': 'https://shahid.mbc.net/ar/episode/90511/%D9%85%D8%B1%D8%A7%D9%8A%D8%A7-2011-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1.html', + 'only_matching': True + }] + + def _handle_error(self, response): + if not isinstance(response, dict): + return + error = response.get('error') + if error: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, '\n'.join(error.values())), + expected=True) + + def _download_json(self, url, video_id, note='Downloading JSON metadata'): + response = super(ShahidIE, self)._download_json(url, video_id, note)['data'] + self._handle_error(response) + return response + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + api_vars = { + 'id': video_id, + 'type': 'player', + 'url': 'http://api.shahid.net/api/v1_1', + 'playerType': 'episode', + } + + flashvars = self._search_regex( + r'var\s+flashvars\s*=\s*({[^}]+})', webpage, 'flashvars', default=None) + if flashvars: + for key in api_vars.keys(): + value = self._search_regex( + r'\b%s\s*:\s*(?P<q>["\'])(?P<value>.+?)(?P=q)' % key, + flashvars, 'type', default=None, group='value') + if value: + api_vars[key] = value + + player = self._download_json( + 'https://shahid.mbc.net/arContent/getPlayerContent-param-.id-%s.type-%s.html' + % (video_id, api_vars['type']), video_id, 'Downloading player JSON') + + formats = self._extract_m3u8_formats(player['url'], video_id, 'mp4') + + video = self._download_json( + '%s/%s/%s?%s' % ( + api_vars['url'], api_vars['playerType'], api_vars['id'], + compat_urllib_parse.urlencode({ + 'apiKey': 'sh@hid0nlin3', + 'hash': 'b2wMCTHpSmyxGqQjJFOycRmLSex+BpTK/ooxy6vHaqs=', + })), + video_id, 'Downloading video JSON') + + video = video[api_vars['playerType']] + + title = video['title'] + description = video.get('description') + thumbnail = video.get('thumbnailUrl') + duration = int_or_none(video.get('duration')) + timestamp = parse_iso8601(video.get('referenceDate')) + categories = [ + category['name'] + for category in video.get('genres', []) if 'name' in category] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'categories': categories, + 'formats': formats, + } diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py index 6e2b94e7d..c5636e8e9 100644 --- a/youtube_dl/extractor/shared.py +++ b/youtube_dl/extractor/shared.py @@ -14,17 +14,28 @@ from ..utils import ( class SharedIE(InfoExtractor): - _VALID_URL = r'http://shared\.sx/(?P<id>[\da-z]{10})' + IE_DESC = 'shared.sx and vivo.sx' + _VALID_URL = r'http://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})' - _TEST = { + _TESTS = [{ 'url': 'http://shared.sx/0060718775', 'md5': '106fefed92a8a2adb8c98e6a0652f49b', 'info_dict': { 'id': '0060718775', 'ext': 'mp4', 'title': 'Bmp4', + 'filesize': 1720110, }, - } + }, { + 'url': 'http://vivo.sx/d7ddda0e78', + 'md5': '15b3af41be0b4fe01f4df075c2678b2c', + 'info_dict': { + 'id': 'd7ddda0e78', + 'ext': 'mp4', + 'title': 'Chicken', + 'filesize': 528031, + }, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -34,7 +45,7 @@ class SharedIE(InfoExtractor): raise ExtractorError( 'Video %s does not exist' % video_id, expected=True) - download_form = self._form_hidden_inputs(webpage) + download_form = self._hidden_inputs(webpage) request = compat_urllib_request.Request( url, compat_urllib_parse.urlencode(download_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py index 93a7cfe15..35a81ee87 100644 --- a/youtube_dl/extractor/smotri.py +++ b/youtube_dl/extractor/smotri.py @@ -330,10 +330,7 @@ class SmotriBroadcastIE(InfoExtractor): (username, password) = self._get_login_info() if username is None: - raise ExtractorError( - 'Erotic broadcasts allowed only for registered users, ' - 'use --username and --password options to provide account credentials.', - expected=True) + self.raise_login_required('Erotic broadcasts allowed only for registered users') login_form = { 'login-hint53': '1', diff --git a/youtube_dl/extractor/snagfilms.py b/youtube_dl/extractor/snagfilms.py index cf495f310..6977afb27 100644 --- a/youtube_dl/extractor/snagfilms.py +++ b/youtube_dl/extractor/snagfilms.py @@ -24,6 +24,15 @@ class SnagFilmsEmbedIE(InfoExtractor): 'title': '#whilewewatch', } }, { + # invalid labels, 360p is better that 480p + 'url': 'http://www.snagfilms.com/embed/player?filmId=17ca0950-a74a-11e0-a92a-0026bb61d036', + 'md5': '882fca19b9eb27ef865efeeaed376a48', + 'info_dict': { + 'id': '17ca0950-a74a-11e0-a92a-0026bb61d036', + 'ext': 'mp4', + 'title': 'Life in Limbo', + } + }, { 'url': 'http://www.snagfilms.com/embed/player?filmId=0000014c-de2f-d5d6-abcf-ffef58af0017', 'only_matching': True, }] @@ -52,14 +61,15 @@ class SnagFilmsEmbedIE(InfoExtractor): if not file_: continue type_ = source.get('type') - format_id = source.get('label') ext = determine_ext(file_) - if any(_ == 'm3u8' for _ in (type_, ext)): + format_id = source.get('label') or ext + if all(v == 'm3u8' 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', file_, 'bitrate', default=None)) + [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext], + file_, 'bitrate', default=None)) height = int_or_none(self._search_regex( r'^(\d+)[pP]$', format_id, 'height', default=None)) formats.append({ diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 118ca4832..2b60d354a 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -29,7 +29,7 @@ class SoundcloudIE(InfoExtractor): _VALID_URL = r'''(?x)^(?:https?://)? (?:(?:(?:www\.|m\.)?soundcloud\.com/ (?P<uploader>[\w\d-]+)/ - (?!sets/|(?:likes|tracks)/?(?:$|[?#])) + (?!(?:tracks|sets(?:/[^/?#]+)?|reposts|likes|spotlight)/?(?:$|[?#])) (?P<title>[\w\d-]+)/? (?P<token>[^?]+?)?(?:[?].*)?$) |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+) @@ -113,7 +113,7 @@ class SoundcloudIE(InfoExtractor): }, ] - _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28' + _CLIENT_ID = '02gUJC0hH2ct1EGOcYXQIzRFU91c72Ea' _IPHONE_CLIENT_ID = '376f225bf427445fc4bfb6b99b72e0bf' def report_resolve(self, video_id): @@ -282,69 +282,150 @@ class SoundcloudSetIE(SoundcloudIE): msgs = (compat_str(err['error_message']) for err in info['errors']) raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs)) + entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in info['tracks']] + return { '_type': 'playlist', - 'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']], + 'entries': entries, 'id': '%s' % info['id'], 'title': info['title'], } class SoundcloudUserIE(SoundcloudIE): - _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$' + _VALID_URL = r'''(?x) + https?:// + (?:(?:www|m)\.)?soundcloud\.com/ + (?P<user>[^/]+) + (?:/ + (?P<rsrc>tracks|sets|reposts|likes|spotlight) + )? + /?(?:[?#].*)?$ + ''' IE_NAME = 'soundcloud:user' _TESTS = [{ - 'url': 'https://soundcloud.com/the-concept-band', + 'url': 'https://soundcloud.com/the-akashic-chronicler', 'info_dict': { - 'id': '9615865', - 'title': 'The Royal Concept', + 'id': '114582580', + 'title': 'The Akashic Chronicler (All)', }, - 'playlist_mincount': 12 + 'playlist_mincount': 111, }, { - 'url': 'https://soundcloud.com/the-concept-band/likes', + 'url': 'https://soundcloud.com/the-akashic-chronicler/tracks', 'info_dict': { - 'id': '9615865', - 'title': 'The Royal Concept', + 'id': '114582580', + 'title': 'The Akashic Chronicler (Tracks)', }, - 'playlist_mincount': 1, + 'playlist_mincount': 50, }, { - 'url': 'https://soundcloud.com/the-akashic-chronicler/tracks', - 'only_matching': True, + 'url': 'https://soundcloud.com/the-akashic-chronicler/sets', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Playlists)', + }, + 'playlist_mincount': 3, + }, { + 'url': 'https://soundcloud.com/the-akashic-chronicler/reposts', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Reposts)', + }, + 'playlist_mincount': 7, + }, { + 'url': 'https://soundcloud.com/the-akashic-chronicler/likes', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Likes)', + }, + 'playlist_mincount': 321, + }, { + 'url': 'https://soundcloud.com/grynpyret/spotlight', + 'info_dict': { + 'id': '7098329', + 'title': 'Grynpyret (Spotlight)', + }, + 'playlist_mincount': 1, }] + _API_BASE = 'https://api.soundcloud.com' + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + _BASE_URL_MAP = { + 'all': '%s/profile/soundcloud:users:%%s' % _API_V2_BASE, + 'tracks': '%s/users/%%s/tracks' % _API_BASE, + 'sets': '%s/users/%%s/playlists' % _API_V2_BASE, + 'reposts': '%s/profile/soundcloud:users:%%s/reposts' % _API_V2_BASE, + 'likes': '%s/users/%%s/likes' % _API_V2_BASE, + 'spotlight': '%s/users/%%s/spotlight' % _API_V2_BASE, + } + + _TITLE_MAP = { + 'all': 'All', + 'tracks': 'Tracks', + 'sets': 'Playlists', + 'reposts': 'Reposts', + 'likes': 'Likes', + 'spotlight': 'Spotlight', + } + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) uploader = mobj.group('user') - resource = mobj.group('rsrc') - if resource is None: - resource = 'tracks' - elif resource == 'likes': - resource = 'favorites' url = 'http://soundcloud.com/%s/' % uploader resolv_url = self._resolv_url(url) user = self._download_json( resolv_url, uploader, 'Downloading user info') - base_url = 'http://api.soundcloud.com/users/%s/%s.json?' % (uploader, resource) + + resource = mobj.group('rsrc') or 'all' + base_url = self._BASE_URL_MAP[resource] % user['id'] + + next_href = None entries = [] for i in itertools.count(): - data = compat_urllib_parse.urlencode({ - 'offset': i * 50, - 'limit': 50, - 'client_id': self._CLIENT_ID, - }) - new_entries = self._download_json( - base_url + data, uploader, 'Downloading track page %s' % (i + 1)) - if len(new_entries) == 0: + if not next_href: + data = compat_urllib_parse.urlencode({ + 'offset': i * 50, + 'limit': 50, + 'client_id': self._CLIENT_ID, + 'linked_partitioning': '1', + 'representation': 'speedy', + }) + next_href = base_url + '?' + data + + response = self._download_json( + next_href, uploader, 'Downloading track page %s' % (i + 1)) + + collection = response['collection'] + + if not collection: self.to_screen('%s: End page received' % uploader) break - entries.extend(self.url_result(e['permalink_url'], 'Soundcloud') for e in new_entries) + + def resolve_permalink_url(candidates): + for cand in candidates: + if isinstance(cand, dict): + permalink_url = cand.get('permalink_url') + if permalink_url and permalink_url.startswith('http'): + return permalink_url + + for e in collection: + permalink_url = resolve_permalink_url((e, e.get('track'), e.get('playlist'))) + if permalink_url: + entries.append(self.url_result(permalink_url)) + + if 'next_href' in response: + next_href = response['next_href'] + if not next_href: + break + else: + next_href = None return { '_type': 'playlist', 'id': compat_str(user['id']), - 'title': user['username'], + 'title': '%s (%s)' % (user['username'], self._TITLE_MAP[resource]), 'entries': entries, } @@ -379,9 +460,7 @@ class SoundcloudPlaylistIE(SoundcloudIE): data = self._download_json( base_url + data, playlist_id, 'Downloading playlist') - entries = [ - self._extract_info_dict(t, quiet=True, secret_token=token) - for t in data['tracks']] + entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in data['tracks']] return { '_type': 'playlist', diff --git a/youtube_dl/extractor/southpark.py b/youtube_dl/extractor/southpark.py index 7fb165a87..87b650468 100644 --- a/youtube_dl/extractor/southpark.py +++ b/youtube_dl/extractor/southpark.py @@ -45,6 +45,14 @@ class SouthParkDeIE(SouthParkIE): 'title': 'The Government Won\'t Respect My Privacy', 'description': 'Cartman explains the benefits of "Shitter" to Stan, Kyle and Craig.', }, + }, { + # non-ASCII characters in initial URL + 'url': 'http://www.southpark.de/alle-episoden/s18e09-hashtag-aufwärmen', + 'playlist_count': 4, + }, { + # non-ASCII characters in redirect URL + 'url': 'http://www.southpark.de/alle-episoden/s18e09', + 'playlist_count': 4, }] diff --git a/youtube_dl/extractor/spankwire.py b/youtube_dl/extractor/spankwire.py index bff75d6b2..9e8fb35b2 100644 --- a/youtube_dl/extractor/spankwire.py +++ b/youtube_dl/extractor/spankwire.py @@ -4,7 +4,7 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_parse_urlparse, compat_urllib_request, ) @@ -16,8 +16,9 @@ from ..aes import aes_decrypt_text class SpankwireIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<videoid>[0-9]+)/?)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?P<url>spankwire\.com/[^/]*/video(?P<id>[0-9]+)/?)' + _TESTS = [{ + # download URL pattern: */<height>P_<tbr>K_<video_id>.mp4 'url': 'http://www.spankwire.com/Buckcherry-s-X-Rated-Music-Video-Crazy-Bitch/video103545/', 'md5': '8bbfde12b101204b39e4b9fe7eb67095', 'info_dict': { @@ -30,14 +31,27 @@ class SpankwireIE(InfoExtractor): 'upload_date': '20070507', 'age_limit': 18, } - } + }, { + # download URL pattern: */mp4_<format_id>_<video_id>.mp4 + 'url': 'http://www.spankwire.com/Titcums-Compiloation-I/video1921551/', + 'md5': '09b3c20833308b736ae8902db2f8d7e6', + 'info_dict': { + 'id': '1921551', + 'ext': 'mp4', + 'title': 'Titcums Compiloation I', + 'description': 'cum on tits', + 'uploader': 'dannyh78999', + 'uploader_id': '3056053', + 'upload_date': '20150822', + 'age_limit': 18, + }, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - url = 'http://www.' + mobj.group('url') + video_id = mobj.group('id') - req = compat_urllib_request.Request(url) + req = compat_urllib_request.Request('http://www.' + mobj.group('url')) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) @@ -54,7 +68,7 @@ class SpankwireIE(InfoExtractor): r'by:\s*<a [^>]*>(.+?)</a>', webpage, 'uploader', fatal=False) uploader_id = self._html_search_regex( - r'by:\s*<a href="/Profile\.aspx\?.*?UserId=(\d+).*?"', + r'by:\s*<a href="/(?:user/viewProfile|Profile\.aspx)\?.*?UserId=(\d+).*?"', webpage, 'uploader id', fatal=False) upload_date = unified_strdate(self._html_search_regex( r'</a> on (.+?) at \d+:\d+', @@ -67,9 +81,10 @@ class SpankwireIE(InfoExtractor): r'<span\s+id="spCommentCount"[^>]*>([\d,\.]+)</span>', webpage, 'comment count', fatal=False)) - video_urls = list(map( - compat_urllib_parse.unquote, - re.findall(r'playerData\.cdnPath[0-9]{3,}\s*=\s*(?:encodeURIComponent\()?["\']([^"\']+)["\']', webpage))) + videos = re.findall( + r'playerData\.cdnPath([0-9]{3,})\s*=\s*(?:encodeURIComponent\()?["\']([^"\']+)["\']', webpage) + heights = [int(video[0]) for video in videos] + video_urls = list(map(compat_urllib_parse_unquote, [video[1] for video in videos])) if webpage.find('flashvars\.encrypted = "true"') != -1: password = self._search_regex( r'flashvars\.video_title = "([^"]+)', @@ -79,21 +94,22 @@ class SpankwireIE(InfoExtractor): video_urls)) formats = [] - for video_url in video_urls: + for height, video_url in zip(heights, video_urls): path = compat_urllib_parse_urlparse(video_url).path - format = path.split('/')[4].split('_')[:2] - resolution, bitrate_str = format - format = "-".join(format) - height = int(resolution.rstrip('Pp')) - tbr = int(bitrate_str.rstrip('Kk')) - formats.append({ + _, quality = path.split('/')[4].split('_')[:2] + f = { 'url': video_url, - 'resolution': resolution, - 'format': format, - 'tbr': tbr, 'height': height, - 'format_id': format, - }) + } + tbr = self._search_regex(r'^(\d+)[Kk]$', quality, 'tbr', default=None) + if tbr: + f.update({ + 'tbr': int(tbr), + 'format_id': '%dp' % height, + }) + else: + f['format_id'] = quality + formats.append(f) self._sort_formats(formats) age_limit = self._rta_search(webpage) diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index b868241d5..5bd3c0087 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -9,7 +9,7 @@ from .spiegeltv import SpiegeltvIE class SpiegelIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$' + _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed|-iframe)?(?:\.html)?(?:#.*)?$' _TESTS = [{ 'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', 'md5': '2c2754212136f35fb4b19767d242f66e', @@ -39,6 +39,9 @@ class SpiegelIE(InfoExtractor): '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?"', } + }, { + 'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/spiegeltv.py b/youtube_dl/extractor/spiegeltv.py index 27f4033c5..034bd47ff 100644 --- a/youtube_dl/extractor/spiegeltv.py +++ b/youtube_dl/extractor/spiegeltv.py @@ -77,17 +77,21 @@ class SpiegeltvIE(InfoExtractor): 'rtmp_live': True, }) elif determine_ext(endpoint) == 'm3u8': - m3u8_formats = self._extract_m3u8_formats( - endpoint.replace('[video]', play_path), - video_id, 'm4v', - preference=1, # Prefer hls since it allows to workaround georestriction - m3u8_id='hls', fatal=False) - if m3u8_formats is not False: - formats.extend(m3u8_formats) + formats.append({ + 'url': endpoint.replace('[video]', play_path), + 'ext': 'm4v', + 'format_id': 'hls', # Prefer hls since it allows to workaround georestriction + 'protocol': 'm3u8', + 'preference': 1, + 'http_headers': { + 'Accept-Encoding': 'deflate', # gzip causes trouble on the server side + }, + }) else: formats.append({ 'url': endpoint, }) + self._check_formats(formats, video_id) thumbnails = [] for image in media_json['images']: diff --git a/youtube_dl/extractor/sportdeutschland.py b/youtube_dl/extractor/sportdeutschland.py index 1a57aebf1..7ec6c613f 100644 --- a/youtube_dl/extractor/sportdeutschland.py +++ b/youtube_dl/extractor/sportdeutschland.py @@ -38,10 +38,12 @@ class SportDeutschlandIE(InfoExtractor): 'upload_date': '20140825', 'description': 'md5:60a20536b57cee7d9a4ec005e8687504', 'timestamp': 1408976060, + 'duration': 2732, 'title': 'Li-Ning Badminton Weltmeisterschaft 2014 Kopenhagen: Herren Einzel, Wei Lee vs. Keun Lee', 'thumbnail': 're:^https?://.*\.jpg$', 'view_count': int, 'categories': ['Li-Ning Badminton WM 2014'], + } }] @@ -50,7 +52,7 @@ class SportDeutschlandIE(InfoExtractor): video_id = mobj.group('id') sport_id = mobj.group('sport') - api_url = 'http://splink.tv/api/permalinks/%s/%s' % ( + api_url = 'http://proxy.vidibusdynamic.net/sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % ( sport_id, video_id) req = compat_urllib_request.Request(api_url, headers={ 'Accept': 'application/vnd.vidibus.v2.html+json', @@ -58,12 +60,11 @@ class SportDeutschlandIE(InfoExtractor): }) data = self._download_json(req, video_id) - categories = list(data.get('section', {}).get('tags', {}).values()) asset = data['asset'] - assets_info = self._download_json(asset['url'], video_id) + categories = [data['section']['title']] formats = [] - smil_url = assets_info['video'] + smil_url = asset['video'] if '.smil' in smil_url: m3u8_url = smil_url.replace('.smil', '.m3u8') formats.extend( @@ -91,6 +92,7 @@ class SportDeutschlandIE(InfoExtractor): 'title': asset['title'], 'thumbnail': asset.get('image'), 'description': asset.get('teaser'), + 'duration': asset.get('duration'), 'categories': categories, 'view_count': asset.get('views'), 'rtmp_live': asset.get('live'), diff --git a/youtube_dl/extractor/stitcher.py b/youtube_dl/extractor/stitcher.py new file mode 100644 index 000000000..d5c852f52 --- /dev/null +++ b/youtube_dl/extractor/stitcher.py @@ -0,0 +1,81 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + int_or_none, + js_to_json, + unescapeHTML, +) + + +class StitcherIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?stitcher\.com/podcast/(?:[^/]+/)+e/(?:(?P<display_id>[^/#?&]+?)-)?(?P<id>\d+)(?:[/#?&]|$)' + _TESTS = [{ + 'url': 'http://www.stitcher.com/podcast/the-talking-machines/e/40789481?autoplay=true', + 'md5': '391dd4e021e6edeb7b8e68fbf2e9e940', + 'info_dict': { + 'id': '40789481', + 'ext': 'mp3', + 'title': 'Machine Learning Mastery and Cancer Clusters', + 'description': 'md5:55163197a44e915a14a1ac3a1de0f2d3', + 'duration': 1604, + 'thumbnail': 're:^https?://.*\.jpg', + }, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/vulture-tv/e/the-rare-hourlong-comedy-plus-40846275?autoplay=true', + 'info_dict': { + 'id': '40846275', + 'display_id': 'the-rare-hourlong-comedy-plus', + 'ext': 'mp3', + 'title': "The CW's 'Crazy Ex-Girlfriend'", + 'description': 'md5:04f1e2f98eb3f5cbb094cea0f9e19b17', + 'duration': 2235, + 'thumbnail': 're:^https?://.*\.jpg', + }, + 'params': { + 'skip_download': True, + }, + }, { + # escaped title + 'url': 'http://www.stitcher.com/podcast/marketplace-on-stitcher/e/40910226?autoplay=true', + 'only_matching': True, + }, { + 'url': 'http://www.stitcher.com/podcast/panoply/getting-in/e/episode-2a-how-many-extracurriculars-should-i-have-40876278?autoplay=true', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + audio_id = mobj.group('id') + display_id = mobj.group('display_id') or audio_id + + webpage = self._download_webpage(url, display_id) + + episode = self._parse_json( + js_to_json(self._search_regex( + r'(?s)var\s+stitcher\s*=\s*({.+?});\n', webpage, 'episode config')), + display_id)['config']['episode'] + + title = unescapeHTML(episode['title']) + formats = [{ + 'url': episode[episode_key], + 'ext': determine_ext(episode[episode_key]) or 'mp3', + 'vcodec': 'none', + } for episode_key in ('episodeURL',) if episode.get(episode_key)] + description = self._search_regex( + r'Episode Info:\s*</span>([^<]+)<', webpage, 'description', fatal=False) + duration = int_or_none(episode.get('duration')) + thumbnail = episode.get('episodeImage') + + return { + 'id': audio_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'duration': duration, + 'thumbnail': thumbnail, + 'formats': formats, + } diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py index bfe07b024..73e7657d4 100644 --- a/youtube_dl/extractor/tagesschau.py +++ b/youtube_dl/extractor/tagesschau.py @@ -8,17 +8,17 @@ from ..utils import parse_filesize class TagesschauIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:sendung/ts|video/video)(?P<id>-?[0-9]+)\.html' + _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_[^/#?]+?)?\.html' _TESTS = [{ - 'url': 'http://www.tagesschau.de/multimedia/video/video1399128.html', - 'md5': 'bcdeac2194fb296d599ce7929dfa4009', + 'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html', + 'md5': '917a228bc7df7850783bc47979673a09', 'info_dict': { - 'id': '1399128', + 'id': '102143', 'ext': 'mp4', - 'title': 'Harald Range, Generalbundesanwalt, zu den Ermittlungen', - 'description': 'md5:69da3c61275b426426d711bde96463ab', - 'thumbnail': 're:^http:.*\.jpg$', + 'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt', + 'description': 'md5:171feccd9d9b3dd54d05d501568f6359', + 'thumbnail': 're:^https?:.*\.jpg$', }, }, { 'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html', @@ -28,8 +28,39 @@ class TagesschauIE(InfoExtractor): 'ext': 'mp4', 'description': 'md5:695c01bfd98b7e313c501386327aea59', 'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr', - 'thumbnail': 're:^http:.*\.jpg$', - } + 'thumbnail': 're:^https?:.*\.jpg$', + }, + }, { + 'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html', + 'md5': 'aef45de271c4bf0a5db834aa40bf774c', + 'info_dict': { + 'id': '18407', + 'ext': 'mp3', + 'title': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', + 'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', + 'thumbnail': 're:^https?:.*\.jpg$', + }, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/weltspiegel-3167.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/tsvorzwanzig-959.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/bab/bab-3299~_bab-sendung-209.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html', + 'only_matching': True, }] _FORMATS = { @@ -49,19 +80,26 @@ class TagesschauIE(InfoExtractor): playerpage = self._download_webpage( player_url, display_id, 'Downloading player page') - medias = re.findall( - r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"', - playerpage) formats = [] - for url, ext, res in medias: + 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': res + '_' + ext, + '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_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1] + thumbnail = self._og_search_thumbnail(playerpage) title = self._og_search_title(webpage).strip() description = self._og_search_description(webpage).strip() else: @@ -99,17 +137,14 @@ class TagesschauIE(InfoExtractor): 'filesize_approx': parse_filesize(m.group('filesize_approx')), }) formats.append(format) - thumbnail_fn = self._search_regex( - r'(?s)<img alt="Sendungsbild".*?src="([^"]+)"', - webpage, 'thumbnail', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) description = self._html_search_regex( r'(?s)<p class="teasertext">(.*?)</p>', - webpage, 'description', fatal=False) + webpage, 'description', default=None) title = self._html_search_regex( r'<span class="headline".*?>(.*?)</span>', webpage, 'title') self._sort_formats(formats) - thumbnail = 'http://www.tagesschau.de' + thumbnail_fn return { 'id': display_id, diff --git a/youtube_dl/extractor/tapely.py b/youtube_dl/extractor/tapely.py index f1f43d0a7..744f9db38 100644 --- a/youtube_dl/extractor/tapely.py +++ b/youtube_dl/extractor/tapely.py @@ -16,7 +16,7 @@ from ..utils import ( class TapelyIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tape\.ly/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?' + _VALID_URL = r'https?://(?:www\.)?(?:tape\.ly|tapely\.com)/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?' _API_URL = 'http://tape.ly/showtape?id={0:}' _S3_SONG_URL = 'http://mytape.s3.amazonaws.com/{0:}' _SOUNDCLOUD_SONG_URL = 'http://api.soundcloud.com{0:}' @@ -42,6 +42,10 @@ class TapelyIE(InfoExtractor): 'ext': 'm4a', }, }, + { + 'url': 'https://tapely.com/my-grief-as-told-by-water', + 'only_matching': True, + }, ] def _real_extract(self, url): diff --git a/youtube_dl/extractor/telecinco.py b/youtube_dl/extractor/telecinco.py index a0c744fd1..2c8e9b941 100644 --- a/youtube_dl/extractor/telecinco.py +++ b/youtube_dl/extractor/telecinco.py @@ -1,26 +1,94 @@ # coding: utf-8 from __future__ import unicode_literals -from .mitele import MiTeleIE +import json +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urlparse, +) +from ..utils import ( + get_element_by_attribute, + parse_duration, + strip_jsonp, +) -class TelecincoIE(MiTeleIE): - IE_NAME = 'telecinco.es' - _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/(?:[^/]+/)?(?P<id>.*?)\.html' + +class TelecincoIE(InfoExtractor): + 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', 'info_dict': { 'id': 'MDSVID20141015_0058', 'ext': 'mp4', 'title': 'Con Martín Berasategui, hacer un bacalao al ...', 'duration': 662, }, - 'params': { - # m3u8 download - 'skip_download': True, + }, { + 'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html', + 'md5': '0a5b9f3cc8b074f50a0578f823a12694', + 'info_dict': { + 'id': 'MDSVID20150916_0128', + 'ext': 'mp4', + 'title': '¿Quién es este ex futbolista con el que hablan ...', + 'duration': 79, + }, + }, { + 'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html', + 'md5': 'ad1bfaaba922dd4a295724b05b68f86a', + 'info_dict': { + 'id': 'MDSVID20150513_0220', + 'ext': 'mp4', + 'title': '#DOYLACARA. Con la trata no hay trato', + 'duration': 50, }, }, { 'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html', 'only_matching': True, + }, { + 'url': 'http://www.telecinco.es/espanasinirmaslejos/Espana-gran-destino-turistico_2_1240605043.html', + 'only_matching': True, }] + + 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') + + 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), + } diff --git a/youtube_dl/extractor/telegraaf.py b/youtube_dl/extractor/telegraaf.py new file mode 100644 index 000000000..6f8333cfc --- /dev/null +++ b/youtube_dl/extractor/telegraaf.py @@ -0,0 +1,35 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import 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', + 'title': 'Tikibad ontruimd wegens brand', + 'description': 'md5:05ca046ff47b931f9b04855015e163a4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 33, + }, + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + playlist_url = self._search_regex( + r"iframe\.loadPlayer\('([^']+)'", webpage, 'player') + + entries = self._extract_xspf_playlist(playlist_url, playlist_id) + title = remove_end(self._og_search_title(webpage), ' - VIDEO') + description = self._og_search_description(webpage) + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 83d833e30..25edc3100 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -1,7 +1,7 @@ +# -*- coding: utf-8 -*- from __future__ import unicode_literals import re -import json import time import hmac import binascii @@ -10,7 +10,8 @@ import hashlib from .common import InfoExtractor from ..compat import ( - compat_str, + compat_parse_qs, + compat_urllib_parse_urlparse, ) from ..utils import ( determine_ext, @@ -18,12 +19,69 @@ from ..utils import ( xpath_with_ns, unsmuggle_url, int_or_none, + url_basename, + float_or_none, ) -_x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'}) +default_ns = 'http://www.w3.org/2005/SMIL21/Language' +_x = lambda p: xpath_with_ns(p, {'smil': default_ns}) -class ThePlatformIE(InfoExtractor): +class ThePlatformBaseIE(InfoExtractor): + def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'): + meta = self._download_xml(smil_url, video_id, note=note) + try: + error_msg = next( + n.attrib['abstract'] + for n in meta.findall(_x('.//smil:ref')) + if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired') + except StopIteration: + pass + else: + raise ExtractorError(error_msg, expected=True) + + formats = self._parse_smil_formats( + meta, smil_url, video_id, namespace=default_ns, + # the parameters are from syfy.com, other sites may use others, + # they also work for nbc.com + f4m_params={'g': 'UXWGVKRWHFSP', 'hdcore': '3.0.3'}, + transform_rtmp_url=lambda streamer, src: (streamer, 'mp4:' + src)) + + for _format in formats: + ext = determine_ext(_format['url']) + if ext == 'once': + _format['ext'] = 'mp4' + + self._sort_formats(formats) + + subtitles = self._parse_smil_subtitles(meta, default_ns) + + return formats, subtitles + + def get_metadata(self, path, video_id): + info_url = 'http://link.theplatform.com/s/%s?format=preview' % path + info = self._download_json(info_url, video_id) + + subtitles = {} + captions = info.get('captions') + if isinstance(captions, list): + for caption in captions: + lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type') + subtitles[lang] = [{ + 'ext': 'srt' if mime == 'text/srt' else 'ttml', + 'url': src, + }] + + return { + 'title': info['title'], + 'subtitles': subtitles, + 'description': info['description'], + 'thumbnail': info['defaultThumbnailUrl'], + 'duration': int_or_none(info.get('duration'), 1000), + } + + +class ThePlatformIE(ThePlatformBaseIE): _VALID_URL = r'''(?x) (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/ (?:(?P<media>(?:[^/]+/)+select/media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))? @@ -67,6 +125,20 @@ class ThePlatformIE(InfoExtractor): }, { 'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7', 'only_matching': True, + }, { + 'url': 'http://player.theplatform.com/p/2E2eJC/nbcNewsOffsite?guid=tdy_or_siri_150701', + 'md5': '734f3790fb5fc4903da391beeebc4836', + 'info_dict': { + 'id': 'tdy_or_siri_150701', + 'ext': 'mp4', + 'title': 'iPhone Siri’s sassy response to a math question has people talking', + 'description': 'md5:a565d1deadd5086f3331d57298ec6333', + 'duration': 83.0, + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1435752600, + 'upload_date': '20150701', + 'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"], + }, }] @staticmethod @@ -101,6 +173,24 @@ class ThePlatformIE(InfoExtractor): path += '/media' path += '/' + video_id + qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + if 'guid' in qs_dict: + webpage = self._download_webpage(url, video_id) + scripts = re.findall(r'<script[^>]+src="([^"]+)"', webpage) + feed_id = None + # feed id usually locates in the last script. + # Seems there's no pattern for the interested script filename, so + # I try one by one + for script in reversed(scripts): + feed_script = self._download_webpage(script, video_id, 'Downloading feed script') + feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None) + if feed_id is not None: + break + if feed_id is None: + raise ExtractorError('Unable to find feed id') + return self.url_result('http://feed.theplatform.com/f/%s/%s?byGuid=%s' % ( + provider_id, feed_id, qs_dict['guid'][0])) + if smuggled_data.get('force_smil_url', False): smil_url = url elif mobj.group('config'): @@ -108,7 +198,11 @@ class ThePlatformIE(InfoExtractor): config_url = config_url.replace('swf/', 'config/') config_url = config_url.replace('onsite/', 'onsite/config/') config = self._download_json(config_url, video_id, 'Downloading config') - smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m' + if 'releaseUrl' in config: + release_url = config['releaseUrl'] + else: + release_url = 'http://link.theplatform.com/s/%s?mbr=true' % path + smil_url = release_url + '&format=SMIL&formats=MPEG4&manifest=f4m' else: smil_url = 'http://link.theplatform.com/s/%s/meta.smil?format=smil&mbr=true' % path @@ -116,95 +210,85 @@ class ThePlatformIE(InfoExtractor): if sig: smil_url = self._sign_url(smil_url, sig['key'], sig['secret']) - meta = self._download_xml(smil_url, video_id) - try: - error_msg = next( - n.attrib['abstract'] - for n in meta.findall(_x('.//smil:ref')) - if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired') - except StopIteration: - pass - else: - raise ExtractorError(error_msg, expected=True) + formats, subtitles = self._extract_theplatform_smil(smil_url, video_id) - info_url = 'http://link.theplatform.com/s/%s?format=preview' % path - info_json = self._download_webpage(info_url, video_id) - info = json.loads(info_json) + ret = self.get_metadata(path, video_id) + combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles) + ret.update({ + 'id': video_id, + 'formats': formats, + 'subtitles': combined_subtitles, + }) + + return ret + + +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 = { + # 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': '22d2b84f058d3586efcd99e57d59d314', + 'info_dict': { + 'id': 'n_hardball_5biden_140207', + 'ext': 'mp4', + 'title': 'The Biden factor: will Joe run in 2016?', + 'description': 'Could Vice President Joe Biden be preparing a 2016 campaign? Mark Halperin and Sam Stein weigh in.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'upload_date': '20140208', + 'timestamp': 1391824260, + 'duration': 467.0, + 'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'], + }, + } + + 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] + formats = [] subtitles = {} - captions = info.get('captions') - if isinstance(captions, list): - for caption in captions: - lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type') - subtitles[lang] = [{ - 'ext': 'srt' if mime == 'text/srt' else 'ttml', - 'url': src, - }] + first_video_id = None + duration = None + for item in entry['media$content']: + smil_url = item['plfile$url'] + '&format=SMIL&Tracking=true&Embedded=true&formats=MPEG4,F4M' + cur_video_id = url_basename(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) - head = meta.find(_x('smil:head')) - body = meta.find(_x('smil:body')) + self._sort_formats(formats) - f4m_node = body.find(_x('smil:seq//smil:video')) - if f4m_node is None: - f4m_node = body.find(_x('smil:seq/smil:video')) - if f4m_node is not None and '.f4m' in f4m_node.attrib['src']: - f4m_url = f4m_node.attrib['src'] - if 'manifest.f4m?' not in f4m_url: - f4m_url += '?' - # the parameters are from syfy.com, other sites may use others, - # they also work for nbc.com - f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3' - formats = self._extract_f4m_formats(f4m_url, video_id) - else: - formats = [] - switch = body.find(_x('smil:switch')) - if switch is None: - switch = body.find(_x('smil:par//smil:switch')) - if switch is None: - switch = body.find(_x('smil:par/smil:switch')) - if switch is None: - switch = body.find(_x('smil:par')) - if switch is not None: - base_url = head.find(_x('smil:meta')).attrib['base'] - for f in switch.findall(_x('smil:video')): - attr = f.attrib - width = int_or_none(attr.get('width')) - height = int_or_none(attr.get('height')) - vbr = int_or_none(attr.get('system-bitrate'), 1000) - format_id = '%dx%d_%dk' % (width, height, vbr) - formats.append({ - 'format_id': format_id, - 'url': base_url, - 'play_path': 'mp4:' + attr['src'], - 'ext': 'flv', - 'width': width, - 'height': height, - 'vbr': vbr, - }) - else: - switch = body.find(_x('smil:seq//smil:switch')) - if switch is None: - switch = body.find(_x('smil:seq/smil:switch')) - for f in switch.findall(_x('smil:video')): - attr = f.attrib - vbr = int_or_none(attr.get('system-bitrate'), 1000) - ext = determine_ext(attr['src']) - if ext == 'once': - ext = 'mp4' - formats.append({ - 'format_id': compat_str(vbr), - 'url': attr['src'], - 'vbr': vbr, - 'ext': ext, - }) - self._sort_formats(formats) + thumbnails = [{ + 'url': thumbnail['plfile$url'], + 'width': int_or_none(thumbnail.get('plfile$width')), + 'height': int_or_none(thumbnail.get('plfile$height')), + } for thumbnail in entry.get('media$thumbnails', [])] - return { + timestamp = int_or_none(entry.get('media$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) + subtitles = self._merge_subtitles(subtitles, ret['subtitles']) + ret.update({ 'id': video_id, - 'title': info['title'], - 'subtitles': subtitles, 'formats': formats, - 'description': info['description'], - 'thumbnail': info['defaultThumbnailUrl'], - 'duration': int_or_none(info.get('duration'), 1000), - } + 'subtitles': subtitles, + 'thumbnails': thumbnails, + 'duration': duration, + 'timestamp': timestamp, + 'categories': categories, + }) + + return ret diff --git a/youtube_dl/extractor/tubitv.py b/youtube_dl/extractor/tubitv.py index 2c4b21807..4f86b3ee9 100644 --- a/youtube_dl/extractor/tubitv.py +++ b/youtube_dl/extractor/tubitv.py @@ -60,9 +60,7 @@ class TubiTvIE(InfoExtractor): webpage = self._download_webpage(url, video_id) if re.search(r"<(?:DIV|div) class='login-required-screen'>", webpage): - raise ExtractorError( - 'This video requires login, use --username and --password ' - 'options to provide account credentials.', expected=True) + self.raise_login_required('This video requires login') title = self._og_search_title(webpage) description = self._og_search_description(webpage) diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py index c89de5ba4..5f7ac4b35 100644 --- a/youtube_dl/extractor/tudou.py +++ b/youtube_dl/extractor/tudou.py @@ -2,14 +2,12 @@ from __future__ import unicode_literals -import re -import json - from .common import InfoExtractor +from ..compat import compat_str class TudouIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/.*?/(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])' + _VALID_URL = r'https?://(?:www\.)?tudou\.com/(?:listplay|programs(?:/view)?|albumplay)/([^/]+/)*(?P<id>[^/?#]+?)(?:\.html)?/?(?:$|[?#])' _TESTS = [{ 'url': 'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html', 'md5': '140a49ed444bd22f93330985d8475fcb', @@ -27,35 +25,41 @@ class TudouIE(InfoExtractor): 'title': 'La Sylphide-Bolshoi-Ekaterina Krysanova & Vyacheslav Lopatin 2012', 'thumbnail': 're:^https?://.*\.jpg$', } + }, { + 'url': 'http://www.tudou.com/albumplay/cJAHGih4yYg.html', + 'only_matching': True, }] - def _url_for_id(self, id, quality=None): - info_url = "http://v2.tudou.com/f?id=" + str(id) + _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf' + + def _url_for_id(self, video_id, quality=None): + info_url = 'http://v2.tudou.com/f?id=' + compat_str(video_id) if quality: info_url += '&hd' + quality - webpage = self._download_webpage(info_url, id, "Opening the info webpage") - final_url = self._html_search_regex('>(.+?)</f>', webpage, 'video url') + xml_data = self._download_xml(info_url, video_id, "Opening the info XML page") + final_url = xml_data.text return final_url def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - m = re.search(r'vcode:\s*[\'"](.+?)[\'"]', webpage) - if m and m.group(1): - return { - '_type': 'url', - 'url': 'youku:' + m.group(1), - 'ie_key': 'Youku' - } + youku_vcode = self._search_regex( + r'vcode\s*:\s*[\'"]([^\'"]*)[\'"]', webpage, 'youku vcode', default=None) + if youku_vcode: + return self.url_result('youku:' + youku_vcode, ie='Youku') title = self._search_regex( - r",kw:\s*['\"](.+?)[\"']", webpage, 'title') + r',kw\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'title') thumbnail_url = self._search_regex( - r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False) + r',pic\s*:\s*[\'"]([^\'"]+)[\'"]', webpage, 'thumbnail URL', fatal=False) + + player_url = self._search_regex( + r'playerUrl\s*:\s*[\'"]([^\'"]+\.swf)[\'"]', + webpage, 'player URL', default=self._PLAYER_URL) - segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments') - segments = json.loads(segs_json) + segments = self._parse_json(self._search_regex( + r'segs: \'([^\']+)\'', webpage, 'segments'), video_id) # It looks like the keys are the arguments that have to be passed as # the hd field in the request url, we pick the higher # Also, filter non-number qualities (see issue #3643). @@ -76,6 +80,9 @@ class TudouIE(InfoExtractor): 'ext': ext, 'title': title, 'thumbnail': thumbnail_url, + 'http_headers': { + 'Referer': player_url, + }, } result.append(part_info) diff --git a/youtube_dl/extractor/tumblr.py b/youtube_dl/extractor/tumblr.py index 9ead13a91..4f844706d 100644 --- a/youtube_dl/extractor/tumblr.py +++ b/youtube_dl/extractor/tumblr.py @@ -4,8 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .pornhub import PornHubIE -from .vimeo import VimeoIE +from ..utils import int_or_none class TumblrIE(InfoExtractor): @@ -31,6 +30,19 @@ class TumblrIE(InfoExtractor): 'thumbnail': 're:http://.*\.jpg', } }, { + 'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video', + 'md5': '7ae503065ad150122dc3089f8cf1546c', + 'info_dict': { + 'id': '130323439814', + 'ext': 'mp4', + 'title': 'HD Video Testing \u2014 Test description for my HD video', + 'description': 'md5:97cc3ab5fcd27ee4af6356701541319c', + 'thumbnail': 're:http://.*\.jpg', + }, + 'params': { + 'format': 'hd', + }, + }, { 'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching', 'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab', 'info_dict': { @@ -39,6 +51,9 @@ class TumblrIE(InfoExtractor): 'title': 'naked smoking & stretching', 'upload_date': '20150506', 'timestamp': 1430931613, + 'age_limit': 18, + 'uploader_id': '1638622', + 'uploader': 'naked-yogi', }, 'add_ie': ['Vidme'], }, { @@ -60,28 +75,46 @@ class TumblrIE(InfoExtractor): blog = m_url.group('blog_name') url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id) - webpage = self._download_webpage(url, video_id) + webpage, urlh = self._download_webpage_handle(url, video_id) - vid_me_embed_url = self._search_regex( - r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]', - webpage, 'vid.me embed', default=None) - if vid_me_embed_url is not None: - return self.url_result(vid_me_embed_url, 'Vidme') + iframe_url = self._search_regex( + r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'', + webpage, 'iframe url', default=None) + if iframe_url is None: + return self.url_result(urlh.geturl(), 'Generic') - pornhub_url = PornHubIE._extract_url(webpage) - if pornhub_url: - return self.url_result(pornhub_url, 'PornHub') + iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page') - vimeo_url = VimeoIE._extract_vimeo_url(url, webpage) - if vimeo_url: - return self.url_result(vimeo_url, 'Vimeo') + duration = None + sources = [] - iframe_url = self._search_regex( - r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'', - webpage, 'iframe url') - iframe = self._download_webpage(iframe_url, video_id) - video_url = self._search_regex(r'<source src="([^"]+)"', - iframe, 'video url') + sd_url = self._search_regex( + r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe, + 'sd video url', default=None, group='url') + if sd_url: + sources.append((sd_url, 'sd')) + + options = self._parse_json( + self._search_regex( + r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe, + 'hd video url', default='', group='options'), + video_id, fatal=False) + if options: + duration = int_or_none(options.get('duration')) + hd_url = options.get('hdUrl') + if hd_url: + sources.append((hd_url, 'hd')) + + formats = [{ + 'url': video_url, + 'ext': 'mp4', + 'format_id': format_id, + 'height': int_or_none(self._search_regex( + r'/(\d{3,4})$', video_url, 'height', default=None)), + 'quality': quality, + } for quality, (video_url, format_id) in enumerate(sources)] + + self._sort_formats(formats) # The only place where you can get a title, it's not complete, # but searching in other places doesn't work for all videos @@ -91,9 +124,9 @@ class TumblrIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, - 'ext': 'mp4', 'title': video_title, 'description': self._og_search_description(webpage, default=None), 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'formats': formats, } diff --git a/youtube_dl/extractor/tutv.py b/youtube_dl/extractor/tutv.py index fad720b68..822372ea1 100644 --- a/youtube_dl/extractor/tutv.py +++ b/youtube_dl/extractor/tutv.py @@ -10,10 +10,10 @@ class TutvIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?tu\.tv/videos/(?P<id>[^/?]+)' _TEST = { 'url': 'http://tu.tv/videos/robots-futbolistas', - 'md5': '627c7c124ac2a9b5ab6addb94e0e65f7', + 'md5': '0cd9e28ad270488911b0d2a72323395d', 'info_dict': { 'id': '2973058', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Robots futbolistas', }, } diff --git a/youtube_dl/extractor/tvplay.py b/youtube_dl/extractor/tvplay.py index 79863e781..b4683de54 100644 --- a/youtube_dl/extractor/tvplay.py +++ b/youtube_dl/extractor/tvplay.py @@ -104,6 +104,7 @@ class TVPlayIE(InfoExtractor): 'duration': 1492, 'timestamp': 1330522854, 'upload_date': '20120229', + 'age_limit': 18, }, 'params': { # rtmp download diff --git a/youtube_dl/extractor/tweakers.py b/youtube_dl/extractor/tweakers.py index c80ec15cf..f3198fb85 100644 --- a/youtube_dl/extractor/tweakers.py +++ b/youtube_dl/extractor/tweakers.py @@ -1,19 +1,13 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - xpath_text, - xpath_with_ns, - int_or_none, - float_or_none, -) 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': '1b5afa817403bb5baa08359dca31e6df', + 'md5': '3147e4ddad366f97476a93863e4557c8', 'info_dict': { 'id': '9926', 'ext': 'mp4', @@ -25,41 +19,7 @@ class TweakersIE(InfoExtractor): } def _real_extract(self, url): - video_id = self._match_id(url) - - playlist = self._download_xml( - 'https://tweakers.net/video/s1playlist/%s/playlist.xspf' % video_id, - video_id) - - NS_MAP = { - 'xspf': 'http://xspf.org/ns/0/', - 's1': 'http://static.streamone.nl/player/ns/0', - } - - track = playlist.find(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)) - - title = xpath_text( - track, xpath_with_ns('./xspf:title', NS_MAP), 'title') - description = xpath_text( - track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') - thumbnail = xpath_text( - track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') - duration = float_or_none( - xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), - 1000) - - formats = [{ - 'url': location.text, - 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), - 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), - 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), - } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - } + 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) diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index af2b798fb..3ec08b674 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -7,12 +7,18 @@ import random from .common import InfoExtractor from ..compat import ( + compat_parse_qs, compat_str, compat_urllib_parse, + compat_urllib_parse_urlparse, compat_urllib_request, + compat_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, + int_or_none, + parse_duration, parse_iso8601, ) @@ -22,8 +28,7 @@ class TwitchBaseIE(InfoExtractor): _API_BASE = 'https://api.twitch.tv' _USHER_BASE = 'http://usher.twitch.tv' - _LOGIN_URL = 'https://secure.twitch.tv/login' - _LOGIN_POST_URL = 'https://passport.twitch.tv/authorize' + _LOGIN_URL = 'http://www.twitch.tv/login' _NETRC_MACHINE = 'twitch' def _handle_error(self, response): @@ -56,19 +61,28 @@ class TwitchBaseIE(InfoExtractor): if username is None: return - login_page = self._download_webpage( + login_page, handle = self._download_webpage_handle( self._LOGIN_URL, None, 'Downloading login page') - login_form = self._form_hidden_inputs(login_page) + login_form = self._hidden_inputs(login_page) login_form.update({ - 'login': username.encode('utf-8'), - 'password': password.encode('utf-8'), + 'username': username, + 'password': password, }) + redirect_url = handle.geturl() + + post_url = self._search_regex( + r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page, + 'post url', default=redirect_url, group='url') + + if not post_url.startswith('http'): + post_url = compat_urlparse.urljoin(redirect_url, post_url) + request = compat_urllib_request.Request( - self._LOGIN_POST_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) - request.add_header('Referer', self._LOGIN_URL) + post_url, compat_urllib_parse.urlencode(encode_dict(login_form)).encode('utf-8')) + request.add_header('Referer', redirect_url) response = self._download_webpage( request, None, 'Logging in as %s' % username) @@ -129,14 +143,14 @@ class TwitchItemBaseIE(TwitchBaseIE): def _extract_info(self, info): return { 'id': info['_id'], - 'title': info['title'], - 'description': info['description'], - 'duration': info['length'], - 'thumbnail': info['preview'], - 'uploader': info['channel']['display_name'], - 'uploader_id': info['channel']['name'], - 'timestamp': parse_iso8601(info['recorded_at']), - 'view_count': info['views'], + 'title': info.get('title') or 'Untitled Broadcast', + 'description': info.get('description'), + 'duration': int_or_none(info.get('length')), + 'thumbnail': info.get('preview'), + 'uploader': info.get('channel', {}).get('display_name'), + 'uploader_id': info.get('channel', {}).get('name'), + 'timestamp': parse_iso8601(info.get('recorded_at')), + 'view_count': int_or_none(info.get('views')), } def _real_extract(self, url): @@ -184,8 +198,8 @@ class TwitchVodIE(TwitchItemBaseIE): _ITEM_TYPE = 'vod' _ITEM_SHORTCUT = 'v' - _TEST = { - 'url': 'http://www.twitch.tv/riotgames/v/6528877', + _TESTS = [{ + 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s', 'info_dict': { 'id': 'v6528877', 'ext': 'mp4', @@ -197,25 +211,61 @@ class TwitchVodIE(TwitchItemBaseIE): 'uploader': 'Riot Games', 'uploader_id': 'riotgames', 'view_count': int, + 'start_time': 310, }, 'params': { # m3u8 download 'skip_download': True, }, - } + }, { + # Untitled broadcast (title is None) + 'url': 'http://www.twitch.tv/belkao_o/v/11230755', + 'info_dict': { + 'id': 'v11230755', + 'ext': 'mp4', + 'title': 'Untitled Broadcast', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 1638, + 'timestamp': 1439746708, + 'upload_date': '20150816', + 'uploader': 'BelkAO_o', + 'uploader_id': 'belkao_o', + 'view_count': int, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }] def _real_extract(self, url): item_id = self._match_id(url) + info = self._download_info(self._ITEM_SHORTCUT, item_id) access_token = self._download_json( '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id, 'Downloading %s access token' % self._ITEM_TYPE) + formats = self._extract_m3u8_formats( - '%s/vod/%s?nauth=%s&nauthsig=%s&allow_source=true' - % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']), + '%s/vod/%s?%s' % ( + self._USHER_BASE, item_id, + compat_urllib_parse.urlencode({ + 'allow_source': 'true', + 'allow_spectre': 'true', + 'player': 'twitchweb', + 'nauth': access_token['token'], + 'nauthsig': access_token['sig'], + })), item_id, 'mp4') + self._prefer_source(formats) info['formats'] = formats + + parsed_url = compat_urllib_parse_urlparse(url) + query = compat_parse_qs(parsed_url.query) + if 't' in query: + info['start_time'] = parse_duration(query['t'][0]) + return info @@ -310,9 +360,9 @@ class TwitchBookmarksIE(TwitchPlaylistBaseIE): class TwitchStreamIE(TwitchBaseIE): IE_NAME = 'twitch:stream' - _VALID_URL = r'%s/(?P<id>[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + _VALID_URL = r'%s/(?P<id>[^/#?]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE - _TEST = { + _TESTS = [{ 'url': 'http://www.twitch.tv/shroomztv', 'info_dict': { 'id': '12772022048', @@ -331,7 +381,10 @@ class TwitchStreamIE(TwitchBaseIE): # m3u8 download 'skip_download': True, }, - } + }, { + 'url': 'http://www.twitch.tv/miracle_doto#profile-0', + 'only_matching': True, + }] def _real_extract(self, url): channel_id = self._match_id(url) @@ -346,6 +399,12 @@ class TwitchStreamIE(TwitchBaseIE): 'http://www.twitch.tv/%s/profile' % channel_id, 'TwitchProfile', channel_id) + # Channel name may be typed if different case than the original channel name + # (e.g. http://www.twitch.tv/TWITCHPLAYSPOKEMON) that will lead to constructing + # an invalid m3u8 URL. Working around by use of original channel name from stream + # JSON and fallback to lowercase if it's not available. + channel_id = stream.get('channel', {}).get('name') or channel_id.lower() + access_token = self._download_json( '%s/api/channels/%s/access_token' % (self._API_BASE, channel_id), channel_id, 'Downloading channel access token') diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1aaa06305..9d3e46b94 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -1,3 +1,4 @@ +# coding: utf-8 from __future__ import unicode_literals import re @@ -6,23 +7,51 @@ from .common import InfoExtractor from ..compat import compat_urllib_request from ..utils import ( float_or_none, - unescapeHTML, + xpath_text, + remove_end, ) class TwitterCardIE(InfoExtractor): + IE_NAME = 'twitter:card' _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)' - _TEST = { - 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', - 'md5': 'a74f50b310c83170319ba16de6955192', - 'info_dict': { - 'id': '560070183650213889', - 'ext': 'mp4', - 'title': 'TwitterCard', - 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 30.033, + _TESTS = [ + { + 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889', + 'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4', + 'info_dict': { + 'id': '560070183650213889', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 30.033, + } }, - } + { + 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768', + 'md5': '7ee2a553b63d1bccba97fbed97d9e1c8', + 'info_dict': { + 'id': '623160978427936768', + 'ext': 'mp4', + 'title': 'TwitterCard', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 80.155, + }, + }, + { + 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977', + 'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814', + 'info_dict': { + '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/', + 'upload_date': '20111013', + 'uploader': 'OMG! Ubuntu!', + 'uploader_id': 'omgubuntu', + }, + } + ] def _real_extract(self, url): video_id = self._match_id(url) @@ -40,10 +69,24 @@ class TwitterCardIE(InfoExtractor): request.add_header('User-Agent', user_agent) webpage = self._download_webpage(request, video_id) - config = self._parse_json( - unescapeHTML(self._search_regex( - r'data-player-config="([^"]+)"', webpage, 'data player config')), + youtube_url = self._html_search_regex( + r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"', + webpage, 'youtube iframe', default=None) + if youtube_url: + return self.url_result(youtube_url, 'Youtube') + + config = self._parse_json(self._html_search_regex( + r'data-player-config="([^"]+)"', webpage, 'data player config'), video_id) + if 'playlist' not in config: + if 'vmapUrl' in config: + vmap_data = self._download_xml(config['vmapUrl'], video_id) + video_url = xpath_text(vmap_data, './/MediaFile').strip() + formats.append({ + 'url': video_url, + }) + break # same video regardless of UA + continue video_url = config['playlist'][0]['source'] @@ -70,3 +113,54 @@ class TwitterCardIE(InfoExtractor): 'duration': duration, 'formats': formats, } + + +class TwitterIE(InfoExtractor): + IE_NAME = 'twitter' + _VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)' + _TEMPLATE_URL = 'https://twitter.com/%s/status/%s' + + _TEST = { + 'url': 'https://twitter.com/freethenipple/status/643211948184596480', + 'md5': '31cd83a116fc41f99ae3d909d4caf6a0', + 'info_dict': { + 'id': '643211948184596480', + 'ext': 'mp4', + 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!', + 'thumbnail': 're:^https?://.*\.jpg', + 'duration': 12.922, + 'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"', + 'uploader': 'FREE THE NIPPLE', + 'uploader_id': 'freethenipple', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user_id = mobj.group('user_id') + twid = mobj.group('id') + + webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid) + + username = remove_end(self._og_search_title(webpage), ' on Twitter') + + title = self._og_search_description(webpage).strip('').replace('\n', ' ') + + # strip 'https -_t.co_BJYgOjSeGA' junk from filenames + mobj = re.match(r'“(.*)\s+(https?://[^ ]+)”', title) + title, short_url = mobj.groups() + + card_id = self._search_regex( + r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url') + card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id + + return { + '_type': 'url_transparent', + 'ie_key': 'TwitterCard', + 'uploader_id': user_id, + 'uploader': username, + 'url': card_url, + 'webpage_url': url, + 'description': '%s on Twitter: "%s %s"' % (username, title, short_url), + 'title': username + ' - ' + title, + } diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index 4667ed83b..365d8b4bf 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -15,7 +15,8 @@ from ..utils import ( class UdemyIE(InfoExtractor): IE_NAME = 'udemy' _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)' - _LOGIN_URL = 'https://www.udemy.com/join/login-submit/' + _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1' + _ORIGIN_URL = 'https://www.udemy.com' _NETRC_MACHINE = 'udemy' _TESTS = [{ @@ -69,34 +70,39 @@ class UdemyIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - raise ExtractorError( - 'Udemy account is required, use --username and --password options to provide account credentials.', - expected=True) + self.raise_login_required('Udemy account is required') login_popup = self._download_webpage( - 'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None, - 'Downloading login popup') + 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<']) - if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>': + # already logged in + if is_logged(login_popup): return - csrf = self._html_search_regex( - r'<input type="hidden" name="csrf" value="(.+?)"', - login_popup, 'csrf token') + login_form = self._form_hidden_inputs('login-form', login_popup) + + login_form.update({ + 'email': username.encode('utf-8'), + 'password': password.encode('utf-8'), + }) - login_form = { - 'email': username, - 'password': password, - 'csrf': csrf, - 'displayType': 'json', - 'isSubmitted': '1', - } request = compat_urllib_request.Request( self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8')) - response = self._download_json( + 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) - if 'returnUrl' not in response: + if not is_logged(response): + error = self._html_search_regex( + r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>', + response, 'error message', default=None) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) raise ExtractorError('Unable to log in') def _real_extract(self, url): diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py index c39c278ab..73b05ecab 100644 --- a/youtube_dl/extractor/ustream.py +++ b/youtube_dl/extractor/ustream.py @@ -1,17 +1,20 @@ from __future__ import unicode_literals -import json import re from .common import InfoExtractor from ..compat import ( compat_urlparse, ) -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + int_or_none, + float_or_none, +) class UstreamIE(InfoExtractor): - _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)' + _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)' IE_NAME = 'ustream' _TESTS = [{ 'url': 'http://www.ustream.tv/recorded/20274954', @@ -19,8 +22,12 @@ class UstreamIE(InfoExtractor): 'info_dict': { 'id': '20274954', 'ext': 'flv', - 'uploader': 'Young Americans for Liberty', 'title': 'Young Americans for Liberty February 7, 2012 2:28 AM', + 'description': 'Young Americans for Liberty February 7, 2012 2:28 AM', + 'timestamp': 1328577035, + 'upload_date': '20120207', + 'uploader': 'yaliberty', + 'uploader_id': '6780869', }, }, { # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444 @@ -32,20 +39,21 @@ class UstreamIE(InfoExtractor): 'ext': 'flv', 'title': '-CG11- Canada Games Figure Skating', 'uploader': 'sportscanadatv', - } + }, + 'skip': 'This Pro Broadcaster has chosen to remove this video from the ustream.tv site.', }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') + video_id = m.group('id') # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990) if m.group('type') == 'embed/recorded': - video_id = m.group('videoID') + video_id = m.group('id') desktop_url = 'http://www.ustream.tv/recorded/' + video_id return self.url_result(desktop_url, 'Ustream') if m.group('type') == 'embed': - video_id = m.group('videoID') + video_id = m.group('id') webpage = self._download_webpage(url, video_id) desktop_video_id = self._html_search_regex( r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id') @@ -53,52 +61,50 @@ class UstreamIE(InfoExtractor): return self.url_result(desktop_url, 'Ustream') params = self._download_json( - 'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({ - 'brandId': 1, - 'videoId': int(video_id), - 'autoplay': False, - }), video_id) - - if 'error' in params: - raise ExtractorError(params['error']['message'], expected=True) - - video_url = params['flv'] + 'https://api.ustream.tv/videos/%s.json' % video_id, video_id) - webpage = self._download_webpage(url, video_id) + error = params.get('error') + if error: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error), expected=True) - self.report_extraction(video_id) + video = params['video'] - video_title = self._html_search_regex(r'data-title="(?P<title>.+)"', - webpage, 'title', default=None) + title = video['title'] + filesize = float_or_none(video.get('file_size')) - if not video_title: - try: - video_title = params['moduleConfig']['meta']['title'] - except KeyError: - pass - - if not video_title: - video_title = 'Ustream video ' + video_id + formats = [{ + 'id': video_id, + 'url': video_url, + 'ext': format_id, + 'filesize': filesize, + } for format_id, video_url in video['media_urls'].items()] + self._sort_formats(formats) - uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>', - webpage, 'uploader', fatal=False, flags=re.DOTALL, default=None) + description = video.get('description') + timestamp = int_or_none(video.get('created_at')) + duration = float_or_none(video.get('length')) + view_count = int_or_none(video.get('views')) - if not uploader: - try: - uploader = params['moduleConfig']['meta']['userName'] - except KeyError: - uploader = None + uploader = video.get('owner', {}).get('username') + uploader_id = video.get('owner', {}).get('id') - thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"', - webpage, 'thumbnail', fatal=False) + thumbnails = [{ + 'id': thumbnail_id, + 'url': thumbnail_url, + } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()] return { 'id': video_id, - 'url': video_url, - 'ext': 'flv', - 'title': video_title, + 'title': title, + 'description': description, + 'thumbnails': thumbnails, + 'timestamp': timestamp, + 'duration': duration, + 'view_count': view_count, 'uploader': uploader, - 'thumbnail': thumbnail, + 'uploader_id': uploader_id, + 'formats': formats, } diff --git a/youtube_dl/extractor/veehd.py b/youtube_dl/extractor/veehd.py index 346edf485..0d8d832cc 100644 --- a/youtube_dl/extractor/veehd.py +++ b/youtube_dl/extractor/veehd.py @@ -5,6 +5,7 @@ import json from .common import InfoExtractor from ..compat import ( + compat_urllib_parse_unquote, compat_urlparse, ) from ..utils import ( @@ -76,7 +77,7 @@ class VeeHDIE(InfoExtractor): if config_json: config = json.loads(config_json) - video_url = compat_urlparse.unquote(config['clip']['url']) + video_url = compat_urllib_parse_unquote(config['clip']['url']) if not video_url: video_url = self._html_search_regex( diff --git a/youtube_dl/extractor/vice.py b/youtube_dl/extractor/vice.py index 04e2b0ba7..01af7a995 100644 --- a/youtube_dl/extractor/vice.py +++ b/youtube_dl/extractor/vice.py @@ -1,5 +1,4 @@ from __future__ import unicode_literals -import re from .common import InfoExtractor from .ooyala import OoyalaIE @@ -7,25 +6,29 @@ from ..utils import ExtractorError class ViceIE(InfoExtractor): - _VALID_URL = r'http://www\.vice\.com/.*?/(?P<name>.+)' + _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)+(?P<id>.+)' - _TEST = { - 'url': 'http://www.vice.com/Fringes/cowboy-capitalists-part-1', - 'info_dict': { - 'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp', - 'ext': 'mp4', - 'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', - }, - 'params': { - # Requires ffmpeg (m3u8 manifest) - 'skip_download': True, - }, - } + _TESTS = [ + { + 'url': 'http://www.vice.com/Fringes/cowboy-capitalists-part-1', + 'info_dict': { + 'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp', + 'ext': 'mp4', + 'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', + }, + 'params': { + # Requires ffmpeg (m3u8 manifest) + 'skip_download': True, + }, + }, { + 'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab', + 'only_matching': True, + } + ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - webpage = self._download_webpage(url, name) + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) try: embed_code = self._search_regex( r'embedCode=([^&\'"]+)', webpage, diff --git a/youtube_dl/extractor/videobam.py b/youtube_dl/extractor/videobam.py deleted file mode 100644 index 0eb3d9414..000000000 --- a/youtube_dl/extractor/videobam.py +++ /dev/null @@ -1,81 +0,0 @@ -from __future__ import unicode_literals - -import re -import json - -from .common import InfoExtractor -from ..utils import int_or_none - - -class VideoBamIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videobam\.com/(?:videos/download/)?(?P<id>[a-zA-Z]+)' - - _TESTS = [ - { - 'url': 'http://videobam.com/OiJQM', - 'md5': 'db471f27763a531f10416a0c58b5a1e0', - 'info_dict': { - 'id': 'OiJQM', - 'ext': 'mp4', - 'title': 'Is Alcohol Worse Than Ecstasy?', - 'description': 'md5:d25b96151515c91debc42bfbb3eb2683', - 'uploader': 'frihetsvinge', - }, - }, - { - 'url': 'http://videobam.com/pqLvq', - 'md5': 'd9a565b5379a99126ef94e1d7f9a383e', - 'note': 'HD video', - 'info_dict': { - 'id': 'pqLvq', - 'ext': 'mp4', - 'title': '_', - } - }, - ] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - - page = self._download_webpage('http://videobam.com/%s' % video_id, video_id, 'Downloading page') - - formats = [] - - for preference, format_id in enumerate(['low', 'high']): - mobj = re.search(r"%s: '(?P<url>[^']+)'" % format_id, page) - if not mobj: - continue - formats.append({ - 'url': mobj.group('url'), - 'ext': 'mp4', - 'format_id': format_id, - 'preference': preference, - }) - - if not formats: - player_config = json.loads(self._html_search_regex(r'var player_config = ({.+?});', page, 'player config')) - formats = [{ - 'url': item['url'], - 'ext': 'mp4', - } for item in player_config['playlist'] if 'autoPlay' in item] - - self._sort_formats(formats) - - title = self._og_search_title(page, default='_', fatal=False) - description = self._og_search_description(page, default=None) - thumbnail = self._og_search_thumbnail(page) - uploader = self._html_search_regex(r'Upload by ([^<]+)</a>', page, 'uploader', fatal=False, default=None) - view_count = int_or_none( - self._html_search_regex(r'<strong>Views:</strong> (\d+) ', page, 'view count', fatal=False)) - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'uploader': uploader, - 'view_count': view_count, - 'formats': formats, - 'age_limit': 18, - } diff --git a/youtube_dl/extractor/videolecturesnet.py b/youtube_dl/extractor/videolecturesnet.py index d6a7eb203..649ac9433 100644 --- a/youtube_dl/extractor/videolecturesnet.py +++ b/youtube_dl/extractor/videolecturesnet.py @@ -3,19 +3,21 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_HTTPError, + compat_urlparse, +) from ..utils import ( - find_xpath_attr, - int_or_none, + ExtractorError, parse_duration, - unified_strdate, ) class VideoLecturesNetIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/' + _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$' IE_NAME = 'videolectures.net' - _TEST = { + _TESTS = [{ 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/', 'info_dict': { 'id': 'promogram_igor_mekjavic_eng', @@ -26,61 +28,55 @@ class VideoLecturesNetIE(InfoExtractor): 'duration': 565, 'thumbnail': 're:http://.*\.jpg', }, - } + }, { + # video with invalid direct format links (HTTP 403) + 'url': 'http://videolectures.net/russir2010_filippova_nlp/', + 'info_dict': { + 'id': 'russir2010_filippova_nlp', + 'ext': 'flv', + 'title': 'NLP at Google', + 'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3', + 'duration': 5352, + 'thumbnail': 're:http://.*\.jpg', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { + 'url': 'http://videolectures.net/deeplearning2015_montreal/', + 'info_dict': { + 'id': 'deeplearning2015_montreal', + 'title': 'Deep Learning Summer School, Montreal 2015', + 'description': 'md5:90121a40cc6926df1bf04dcd8563ed3b', + }, + 'playlist_count': 30, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) smil_url = 'http://videolectures.net/%s/video/1/smil.xml' % video_id - smil = self._download_xml(smil_url, video_id) - title = find_xpath_attr(smil, './/meta', 'name', 'title').attrib['content'] - description_el = find_xpath_attr(smil, './/meta', 'name', 'abstract') - description = ( - None if description_el is None - else description_el.attrib['content']) - upload_date = unified_strdate( - find_xpath_attr(smil, './/meta', 'name', 'date').attrib['content']) + try: + smil = self._download_smil(smil_url, video_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + # Probably a playlist + webpage = self._download_webpage(url, video_id) + entries = [ + self.url_result(compat_urlparse.urljoin(url, video_url), 'VideoLecturesNet') + for _, video_url in re.findall(r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', webpage)] + playlist_title = self._html_search_meta('title', webpage, 'title', fatal=True) + playlist_description = self._html_search_meta('description', webpage, 'description') + return self.playlist_result(entries, video_id, playlist_title, playlist_description) - switch = smil.find('.//switch') - duration = parse_duration(switch.attrib.get('dur')) - thumbnail_el = find_xpath_attr(switch, './image', 'type', 'thumbnail') - thumbnail = ( - None if thumbnail_el is None else thumbnail_el.attrib.get('src')) + info = self._parse_smil(smil, smil_url, video_id) - formats = [] - for v in switch.findall('./video'): - proto = v.attrib.get('proto') - if proto not in ['http', 'rtmp']: - continue - f = { - 'width': int_or_none(v.attrib.get('width')), - 'height': int_or_none(v.attrib.get('height')), - 'filesize': int_or_none(v.attrib.get('size')), - 'tbr': int_or_none(v.attrib.get('systemBitrate')) / 1000.0, - 'ext': v.attrib.get('ext'), - } - src = v.attrib['src'] - if proto == 'http': - if self._is_valid_url(src, video_id): - f['url'] = src - formats.append(f) - elif proto == 'rtmp': - f.update({ - 'url': v.attrib['streamer'], - 'play_path': src, - 'rtmp_real_time': True, - }) - formats.append(f) - self._sort_formats(formats) + info['id'] = video_id + + switch = smil.find('.//switch') + if switch is not None: + info['duration'] = parse_duration(switch.attrib.get('dur')) - return { - 'id': video_id, - 'title': title, - 'description': description, - 'upload_date': upload_date, - 'duration': duration, - 'thumbnail': thumbnail, - 'formats': formats, - } + return info diff --git a/youtube_dl/extractor/videomega.py b/youtube_dl/extractor/videomega.py index eb309a7cd..78ff6310a 100644 --- a/youtube_dl/extractor/videomega.py +++ b/youtube_dl/extractor/videomega.py @@ -8,20 +8,23 @@ from ..compat import compat_urllib_request class VideoMegaIE(InfoExtractor): - _VALID_URL = r'''(?x)https?:// - (?:www\.)?videomega\.tv/ - (?:iframe\.php|cdn\.php)?\?ref=(?P<id>[A-Za-z0-9]+) - ''' - _TEST = { - 'url': 'http://videomega.tv/?ref=4GNA688SU99US886ANG4', - 'md5': 'bf5c2f95c4c917536e80936af7bc51e1', + _VALID_URL = r'(?:videomega:|https?://(?:www\.)?videomega\.tv/(?:(?:view|iframe|cdn)\.php)?\?ref=)(?P<id>[A-Za-z0-9]+)' + _TESTS = [{ + 'url': 'http://videomega.tv/cdn.php?ref=AOSQBJYKIDDIKYJBQSOA', + 'md5': 'cc1920a58add3f05c6a93285b84fb3aa', 'info_dict': { - 'id': '4GNA688SU99US886ANG4', + 'id': 'AOSQBJYKIDDIKYJBQSOA', 'ext': 'mp4', - 'title': 'BigBuckBunny_320x180', + 'title': '1254207', 'thumbnail': 're:^https?://.*\.jpg$', } - } + }, { + 'url': 'http://videomega.tv/cdn.php?ref=AOSQBJYKIDDIKYJBQSOA&width=1070&height=600', + 'only_matching': True, + }, { + 'url': 'http://videomega.tv/view.php?ref=090051111052065112106089103052052103089106112065052111051090', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -29,12 +32,13 @@ class VideoMegaIE(InfoExtractor): iframe_url = 'http://videomega.tv/cdn.php?ref=%s' % video_id req = compat_urllib_request.Request(iframe_url) req.add_header('Referer', url) + req.add_header('Cookie', 'noadvtday=0') webpage = self._download_webpage(req, video_id) title = self._html_search_regex( - r'<title>(.*?)</title>', webpage, 'title') + r'<title>(.+?)</title>', webpage, 'title') title = re.sub( - r'(?:^[Vv]ideo[Mm]ega\.tv\s-\s?|\s?-\svideomega\.tv$)', '', title) + r'(?:^[Vv]ideo[Mm]ega\.tv\s-\s*|\s*-\svideomega\.tv$)', '', title) thumbnail = self._search_regex( r'<video[^>]+?poster="([^"]+)"', webpage, 'thumbnail', fatal=False) video_url = self._search_regex( diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index e0b55078b..3d63ed4f0 100644 --- a/youtube_dl/extractor/vidme.py +++ b/youtube_dl/extractor/vidme.py @@ -1,10 +1,12 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..compat import compat_HTTPError from ..utils import ( + ExtractorError, int_or_none, float_or_none, - str_to_int, + parse_iso8601, ) @@ -18,49 +20,185 @@ class VidmeIE(InfoExtractor): 'ext': 'mp4', 'title': 'Fishing for piranha - the easy way', 'description': 'source: https://www.facebook.com/photo.php?v=312276045600871', - 'duration': 119.92, + 'thumbnail': 're:^https?://.*\.jpg', 'timestamp': 1406313244, 'upload_date': '20140725', + 'age_limit': 0, + 'duration': 119.92, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + }, { + 'url': 'https://vid.me/Gc6M', + 'md5': 'f42d05e7149aeaec5c037b17e5d3dc82', + 'info_dict': { + 'id': 'Gc6M', + 'ext': 'mp4', + 'title': 'O Mere Dil ke chain - Arnav and Khushi VM', 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1441211642, + 'upload_date': '20150902', + 'uploader': 'SunshineM', + 'uploader_id': '3552827', + 'age_limit': 0, + 'duration': 223.72, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, }, }, { - # From http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching + # tests uploader field + 'url': 'https://vid.me/4Iib', + 'info_dict': { + 'id': '4Iib', + 'ext': 'mp4', + 'title': 'The Carver', + 'description': 'md5:e9c24870018ae8113be936645b93ba3c', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1433203629, + 'upload_date': '20150602', + 'uploader': 'Thomas', + 'uploader_id': '109747', + 'age_limit': 0, + 'duration': 97.859999999999999, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, { + # nsfw test from http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching 'url': 'https://vid.me/e/Wmur', + 'info_dict': { + 'id': 'Wmur', + 'ext': 'mp4', + 'title': 'naked smoking & stretching', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1430931613, + 'upload_date': '20150506', + 'uploader': 'naked-yogi', + 'uploader_id': '1638622', + 'age_limit': 18, + 'duration': 653.26999999999998, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, { + # nsfw, user-disabled + 'url': 'https://vid.me/dzGJ', + 'only_matching': True, + }, { + # suspended + 'url': 'https://vid.me/Ox3G', + 'only_matching': True, + }, { + # deleted + 'url': 'https://vid.me/KTPm', 'only_matching': True, + }, { + # no formats in the API response + 'url': 'https://vid.me/e5g', + 'info_dict': { + 'id': 'e5g', + 'ext': 'mp4', + 'title': 'Video upload (e5g)', + 'thumbnail': 're:^https?://.*\.jpg', + 'timestamp': 1401480195, + 'upload_date': '20140530', + 'uploader': None, + 'uploader_id': None, + 'age_limit': 0, + 'duration': 483, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): - url = url.replace('vid.me/e/', 'vid.me/') video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - video_url = self._html_search_regex( - r'<source src="([^"]+)"', webpage, 'video URL') + try: + response = self._download_json( + 'https://api.vid.me/videoByUrl/%s' % video_id, video_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400: + response = self._parse_json(e.cause.read(), video_id) + else: + raise + + error = response.get('error') + if error: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error), expected=True) + + video = response['video'] + + if video.get('state') == 'deleted': + raise ExtractorError( + 'Vidme said: Sorry, this video has been deleted.', + expected=True) + + if video.get('state') in ('user-disabled', 'suspended'): + raise ExtractorError( + 'Vidme said: This video has been suspended either due to a copyright claim, ' + 'or for violating the terms of use.', + expected=True) + + formats = [{ + 'format_id': f.get('type'), + 'url': f['uri'], + 'width': int_or_none(f.get('width')), + 'height': int_or_none(f.get('height')), + 'preference': 0 if f.get('type', '').endswith('clip') else 1, + } for f in video.get('formats', []) if f.get('uri')] + + if not formats and video.get('complete_url'): + formats.append({ + 'url': video.get('complete_url'), + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), + }) + + self._sort_formats(formats) - title = self._og_search_title(webpage) - description = self._og_search_description(webpage, default='') - thumbnail = self._og_search_thumbnail(webpage) - timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False)) - width = int_or_none(self._og_search_property('video:width', webpage, fatal=False)) - height = int_or_none(self._og_search_property('video:height', webpage, fatal=False)) - duration = float_or_none(self._html_search_regex( - r'data-duration="([^"]+)"', webpage, 'duration', fatal=False)) - view_count = str_to_int(self._html_search_regex( - r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False)) - like_count = str_to_int(self._html_search_regex( - r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">', - webpage, 'like count', fatal=False)) + title = video['title'] + description = video.get('description') + thumbnail = video.get('thumbnail_url') + timestamp = parse_iso8601(video.get('date_created'), ' ') + uploader = video.get('user', {}).get('username') + uploader_id = video.get('user', {}).get('user_id') + age_limit = 18 if video.get('nsfw') is True else 0 + duration = float_or_none(video.get('duration')) + view_count = int_or_none(video.get('view_count')) + like_count = int_or_none(video.get('likes_count')) + comment_count = int_or_none(video.get('comment_count')) return { 'id': video_id, - 'url': video_url, - 'title': title, + 'title': title or 'Video upload (%s)' % video_id, 'description': description, 'thumbnail': thumbnail, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'age_limit': age_limit, 'timestamp': timestamp, - 'width': width, - 'height': height, 'duration': duration, 'view_count': view_count, 'like_count': like_count, + 'comment_count': comment_count, + 'formats': formats, } diff --git a/youtube_dl/extractor/vier.py b/youtube_dl/extractor/vier.py index 15377097e..c76c20614 100644 --- a/youtube_dl/extractor/vier.py +++ b/youtube_dl/extractor/vier.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import re +import itertools from .common import InfoExtractor @@ -91,31 +92,27 @@ class VierVideosIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) program = mobj.group('program') - webpage = self._download_webpage(url, program) - page_id = mobj.group('page') if page_id: page_id = int(page_id) start_page = page_id - last_page = start_page + 1 playlist_id = '%s-page%d' % (program, page_id) else: start_page = 0 - last_page = int(self._search_regex( - r'videos\?page=(\d+)">laatste</a>', - webpage, 'last page', default=0)) + 1 playlist_id = program entries = [] - for current_page_id in range(start_page, last_page): + for current_page_id in itertools.count(start_page): current_page = self._download_webpage( 'http://www.vier.be/%s/videos?page=%d' % (program, current_page_id), program, - 'Downloading page %d' % (current_page_id + 1)) if current_page_id != page_id else webpage + 'Downloading page %d' % (current_page_id + 1)) page_entries = [ self.url_result('http://www.vier.be' + video_url, 'Vier') for video_url in re.findall( r'<h3><a href="(/[^/]+/videos/[^/]+(?:/\d+)?)">', current_page)] entries.extend(page_entries) + if page_id or '>Meer<' not in current_page: + break return self.playlist_result(entries, playlist_id) diff --git a/youtube_dl/extractor/viewster.py b/youtube_dl/extractor/viewster.py index 1742e66f4..7cf930d69 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -1,129 +1,172 @@ +# coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_request +from ..compat import ( + compat_HTTPError, + compat_urllib_request, + compat_urllib_parse, + compat_urllib_parse_unquote, +) +from ..utils import ( + determine_ext, + ExtractorError, + int_or_none, + parse_iso8601, + HEADRequest, +) class ViewsterIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?viewster\.com/movie/(?P<id>\d+-\d+-\d+)' + _VALID_URL = r'https?://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)' _TESTS = [{ - # movielink, paymethod=fre - 'url': 'http://www.viewster.com/movie/1293-19341-000/hout-wood/', - 'playlist': [{ - 'md5': '8f9d94b282d80c42b378dffdbb11caf3', - 'info_dict': { - 'id': '1293-19341-000-movie', - 'ext': 'flv', - 'title': "'Hout' (Wood) - Movie", - }, - }], - 'info_dict': { - 'id': '1293-19341-000', - 'title': "'Hout' (Wood)", - 'description': 'md5:925733185a9242ef96f436937683f33b', - } - }, { - # movielink, paymethod=adv + # movie, Type=Movie 'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/', - 'playlist': [{ - 'md5': '77a005453ca7396cbe3d35c9bea30aef', - 'info_dict': { - 'id': '1140-11855-000-movie', - 'ext': 'flv', - 'title': "THE LISTENING PROJECT - Movie", - }, - }], + 'md5': 'e642d1b27fcf3a4ffa79f194f5adde36', 'info_dict': { 'id': '1140-11855-000', - 'title': "THE LISTENING PROJECT", - 'description': 'md5:714421ae9957e112e672551094bf3b08', - } + 'ext': 'mp4', + 'title': 'The listening Project', + 'description': 'md5:bac720244afd1a8ea279864e67baa071', + 'timestamp': 1214870400, + 'upload_date': '20080701', + 'duration': 4680, + }, }, { - # direct links, no movielink - 'url': 'http://www.viewster.com/movie/1198-56411-000/sinister/', - 'playlist': [{ - 'md5': '0307b7eac6bfb21ab0577a71f6eebd8f', - 'info_dict': { - 'id': '1198-56411-000-trailer', - 'ext': 'mp4', - 'title': "Sinister - Trailer", - }, - }, { - 'md5': '80b9ee3ad69fb368f104cb5d9732ae95', - 'info_dict': { - 'id': '1198-56411-000-behind-scenes', - 'ext': 'mp4', - 'title': "Sinister - Behind Scenes", - }, - }, { - 'md5': '3b3ea897ecaa91fca57a8a94ac1b15c5', - 'info_dict': { - 'id': '1198-56411-000-scene-from-movie', - 'ext': 'mp4', - 'title': "Sinister - Scene from movie", - }, - }], + # series episode, Type=Episode + 'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/', + 'md5': '9243079a8531809efe1b089db102c069', 'info_dict': { - 'id': '1198-56411-000', - 'title': "Sinister", - 'description': 'md5:014c40b0488848de9683566a42e33372', - } + 'id': '1284-19427-001', + 'ext': 'mp4', + 'title': 'The World and a Wall', + 'description': 'md5:24814cf74d3453fdf5bfef9716d073e3', + 'timestamp': 1428192000, + 'upload_date': '20150405', + 'duration': 1500, + }, + }, { + # serie, Type=Serie + 'url': 'http://www.viewster.com/serie/1303-19426-000/', + 'info_dict': { + 'id': '1303-19426-000', + 'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?', + 'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11', + }, + 'playlist_count': 13, + }, { + # unfinished serie, no Type + 'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/', + 'info_dict': { + 'id': '1284-19427-000', + 'title': 'Baby Steps—Season 2', + 'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1', + }, + 'playlist_mincount': 16, + }, { + # geo restricted series + 'url': 'https://www.viewster.com/serie/1280-18794-002/', + 'only_matching': True, + }, { + # geo restricted video + 'url': 'https://www.viewster.com/serie/1280-18794-002/what-is-extraterritoriality-lawo/', + 'only_matching': True, }] _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' - def _real_extract(self, url): - video_id = self._match_id(url) - - request = compat_urllib_request.Request( - 'http://api.live.viewster.com/api/v1/movie/%s' % video_id) + def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True): + request = compat_urllib_request.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) - movie = self._download_json( - request, video_id, 'Downloading movie metadata JSON') + def _real_extract(self, url): + video_id = self._match_id(url) + # Get 'api_token' cookie + self._request_webpage(HEADRequest('http://www.viewster.com/'), video_id) + cookies = self._get_cookies('http://www.viewster.com/') + self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value) - title = movie.get('title') or movie['original_title'] - description = movie.get('synopsis') - thumbnail = movie.get('large_artwork') or movie.get('artwork') + info = self._download_json( + 'https://public-api.viewster.com/search/%s' % video_id, + video_id, 'Downloading entry JSON') - entries = [] - for clip in movie['play_list']: - entry = None + entry_id = info.get('Id') or info['id'] - # movielink api - link_request = clip.get('link_request') - if link_request: - request = compat_urllib_request.Request( - 'http://api.live.viewster.com/api/v1/movielink?movieid=%(movieid)s&action=%(action)s&paymethod=%(paymethod)s&price=%(price)s¤cy=%(currency)s&language=%(language)s&subtitlelanguage=%(subtitlelanguage)s&ischromecast=%(ischromecast)s' - % link_request) - request.add_header('Accept', self._ACCEPT_HEADER) + # unfinished serie has no Type + if info.get('Type') in ('Serie', None): + try: + episodes = self._download_json( + 'https://public-api.viewster.com/series/%s/episodes' % entry_id, + video_id, 'Downloading series JSON') + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + self.raise_geo_restricted() + else: + raise + entries = [ + self.url_result( + 'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster') + for episode in episodes] + title = (info.get('Title') or info['Synopsis']['Title']).strip() + description = info.get('Synopsis', {}).get('Detailed') + return self.playlist_result(entries, video_id, title, description) - movie_link = self._download_json( - request, video_id, 'Downloading movie link JSON', fatal=False) + 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 movie_link: - formats = self._extract_f4m_formats( - movie_link['url'] + '&hdcore=3.2.0&plugin=flowplayer-3.2.0.1', video_id) - self._sort_formats(formats) - entry = { - 'formats': formats, - } + if not formats and not info.get('LanguageSets') and not info.get('VODSettings'): + self.raise_geo_restricted() - # direct link - clip_url = clip.get('clip_data', {}).get('url') - if clip_url: - entry = { - 'url': clip_url, - 'ext': 'mp4', - } + self._sort_formats(formats) - if entry: - entry.update({ - 'id': '%s-%s' % (video_id, clip['canonical_title']), - 'title': '%s - %s' % (title, clip['title']), - }) - entries.append(entry) + synopsis = info.get('Synopsis', {}) + # Prefer title outside synopsis since it's less messy + title = (info.get('Title') or synopsis['Title']).strip() + description = synopsis.get('Detailed') or info.get('Synopsis', {}).get('Short') + duration = int_or_none(info.get('Duration')) + timestamp = parse_iso8601(info.get('ReleaseDate')) - playlist = self.playlist_result(entries, video_id, title, description) - playlist['thumbnail'] = thumbnail - return playlist + return { + 'id': video_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + } diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py index 51cdc6b65..ddbd395c8 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -88,6 +88,14 @@ class VikiBaseIE(InfoExtractor): if not self._token: self.report_warning('Unable to get session token, login has probably failed') + @staticmethod + def dict_selection(dict_obj, preferred_key): + if preferred_key in dict_obj: + return dict_obj.get(preferred_key) + + filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()])) + return filtered_dict[0] if filtered_dict else None + class VikiIE(VikiBaseIE): IE_NAME = 'viki' @@ -173,6 +181,19 @@ class VikiIE(VikiBaseIE): }, { 'url': 'http://www.viki.com/player/44699v', 'only_matching': True, + }, { + # non-English description + 'url': 'http://www.viki.com/videos/158036v-love-in-magic', + 'md5': '1713ae35df5a521b31f6dc40730e7c9c', + 'info_dict': { + 'id': '158036v', + 'ext': 'mp4', + 'uploader': 'I Planet Entertainment', + 'upload_date': '20111122', + 'timestamp': 1321985454, + 'description': 'md5:44b1e46619df3a072294645c770cef36', + 'title': 'Love In Magic', + }, }] def _real_extract(self, url): @@ -181,19 +202,14 @@ class VikiIE(VikiBaseIE): video = self._call_api( 'videos/%s.json' % video_id, video_id, 'Downloading video JSON') - title = None - titles = video.get('titles') - if titles: - title = titles.get('en') or titles[titles.keys()[0]] + title = self.dict_selection(video.get('titles', {}), 'en') 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') - if container_titles: - container_title = container_titles.get('en') or container_titles[container_titles.keys()[0]] - title = '%s - %s' % (container_title, title) + container_titles = video.get('container', {}).get('titles', {}) + container_title = self.dict_selection(container_titles, 'en') + title = '%s - %s' % (container_title, title) - descriptions = video.get('descriptions') - description = descriptions.get('en') or descriptions[titles.keys()[0]] if descriptions else None + description = self.dict_selection(video.get('descriptions', {}), 'en') duration = int_or_none(video.get('duration')) timestamp = parse_iso8601(video.get('created_at')) @@ -242,8 +258,8 @@ class VikiIE(VikiBaseIE): formats = [] for format_id, stream_dict in streams.items(): - height = self._search_regex( - r'^(\d+)[pP]$', format_id, 'height', default=None) + height = int_or_none(self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None)) for protocol, format_dict in stream_dict.items(): if format_id == 'm3u8': formats = self._extract_m3u8_formats( @@ -299,11 +315,9 @@ class VikiChannelIE(VikiBaseIE): 'containers/%s.json' % channel_id, channel_id, 'Downloading channel JSON') - titles = channel['titles'] - title = titles.get('en') or titles[titles.keys()[0]] + title = self.dict_selection(channel['titles'], 'en') - descriptions = channel['descriptions'] - description = descriptions.get('en') or descriptions[descriptions.keys()[0]] + description = self.dict_selection(channel['descriptions'], 'en') entries = [] for video_type in ('episodes', 'clips', 'movies'): diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index d63c03183..2437ae1eb 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -29,6 +29,7 @@ from ..utils import ( class VimeoBaseInfoExtractor(InfoExtractor): _NETRC_MACHINE = 'vimeo' _LOGIN_REQUIRED = False + _LOGIN_URL = 'https://vimeo.com/log_in' def _login(self): (username, password) = self._get_login_info() @@ -37,21 +38,30 @@ class VimeoBaseInfoExtractor(InfoExtractor): raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) return self.report_login() - login_url = 'https://vimeo.com/log_in' - webpage = self._download_webpage(login_url, None, False) - token = self._search_regex(r'xsrft":"(.*?)"', webpage, 'login token') + webpage = self._download_webpage(self._LOGIN_URL, None, False) + token, vuid = self._extract_xsrft_and_vuid(webpage) data = urlencode_postdata({ + 'action': 'login', 'email': username, 'password': password, - 'action': 'login', 'service': 'vimeo', 'token': token, }) - login_request = compat_urllib_request.Request(login_url, data) + login_request = compat_urllib_request.Request(self._LOGIN_URL, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - login_request.add_header('Cookie', 'xsrft=%s' % token) + login_request.add_header('Cookie', 'vuid=%s' % vuid) + login_request.add_header('Referer', self._LOGIN_URL) self._download_webpage(login_request, None, False, 'Wrong login info') + def _extract_xsrft_and_vuid(self, webpage): + xsrft = self._search_regex( + r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)', + webpage, 'login token', group='xsrft') + vuid = self._search_regex( + r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1', + webpage, 'vuid', group='vuid') + return xsrft, vuid + class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" @@ -75,12 +85,12 @@ class VimeoIE(VimeoBaseInfoExtractor): 'info_dict': { 'id': '56015672', 'ext': 'mp4', - "upload_date": "20121220", - "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - "uploader_id": "user7108434", - "uploader": "Filippo Valsorda", - "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - "duration": 10, + 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", + 'description': 'md5:2d3305bad981a06ff79f027f19865021', + 'upload_date': '20121220', + 'uploader_id': 'user7108434', + 'uploader': 'Filippo Valsorda', + 'duration': 10, }, }, { @@ -93,7 +103,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'openstreetmapus', 'uploader': 'OpenStreetMap US', 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', - 'description': 'md5:380943ec71b89736ff4bf27183233d09', + 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30', 'duration': 1595, }, }, @@ -123,7 +133,7 @@ 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 who love them.', + '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', }, 'params': { 'videopassword': 'youtube-dl', @@ -147,7 +157,6 @@ class VimeoIE(VimeoBaseInfoExtractor): }, { 'url': 'http://vimeo.com/76979871', - 'md5': '3363dd6ffebe3784d56f4132317fd446', 'note': 'Video with subtitles', 'info_dict': { 'id': '76979871', @@ -172,6 +181,11 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'user28849593', }, }, + { + 'url': 'https://vimeo.com/109815029', + 'note': 'Video not completely processed, "failed" seed status', + 'only_matching': True, + }, ] @staticmethod @@ -193,7 +207,7 @@ class VimeoIE(VimeoBaseInfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token') + token, vuid = self._extract_xsrft_and_vuid(webpage) data = urlencode_postdata({ 'password': password, 'token': token, @@ -203,7 +217,8 @@ class VimeoIE(VimeoBaseInfoExtractor): url = url.replace('http://', 'https://') password_request = compat_urllib_request.Request(url + '/password', data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Cookie', 'xsrft=%s' % token) + password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid) + password_request.add_header('Referer', url) return self._download_webpage( password_request, video_id, 'Verifying the password', 'Wrong password') @@ -263,20 +278,30 @@ class VimeoIE(VimeoBaseInfoExtractor): self.report_extraction(video_id) vimeo_config = self._search_regex( - r'vimeo\.config\s*=\s*({.+?});', webpage, + r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage, 'vimeo config', default=None) if vimeo_config: seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) if seed_status.get('state') == 'failed': raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, seed_status['title']), + '%s said: %s' % (self.IE_NAME, seed_status['title']), expected=True) # Extract the config JSON try: try: config_url = self._html_search_regex( - r' data-config-url="(.+?)"', webpage, 'config URL') + r' data-config-url="(.+?)"', webpage, + 'config URL', default=None) + if not config_url: + # Sometimes new react-based page is served instead of old one that require + # different config URL extraction approach (see + # https://github.com/rg3/youtube-dl/pull/7209) + vimeo_clip_page_config = self._search_regex( + r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage, + 'vimeo clip page config') + config_url = self._parse_json( + vimeo_clip_page_config, video_id)['player']['config_url'] config_json = self._download_webpage(config_url, video_id) config = json.loads(config_json) except RegexNotFoundError: @@ -386,14 +411,20 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': codec_extension, 'url': video_url, 'format_id': format_id, - 'width': file_info.get('width'), - 'height': file_info.get('height'), + 'width': int_or_none(file_info.get('width')), + 'height': int_or_none(file_info.get('height')), + 'tbr': int_or_none(file_info.get('bitrate')), }) formats = [] + m3u8_url = config_files.get('hls', {}).get('all') + if m3u8_url: + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) for key in ('other', 'sd', 'hd'): formats += files[key] - if len(formats) == 0: - raise ExtractorError('No known codec found') + self._sort_formats(formats) subtitles = {} text_tracks = config['request'].get('text_tracks') @@ -422,10 +453,11 @@ class VimeoIE(VimeoBaseInfoExtractor): } -class VimeoChannelIE(InfoExtractor): +class VimeoChannelIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:channel' _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])' _MORE_PAGES_INDICATOR = r'<a.+?rel="next"' + _TITLE = None _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"' _TESTS = [{ 'url': 'https://vimeo.com/channels/tributes', @@ -440,7 +472,7 @@ class VimeoChannelIE(InfoExtractor): return '%s/videos/page:%d/' % (base_url, pagenum) def _extract_list_title(self, webpage): - return self._html_search_regex(self._TITLE_RE, webpage, 'list title') + return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title') def _login_list_password(self, page_url, list_id, webpage): login_form = self._search_regex( @@ -452,8 +484,8 @@ class VimeoChannelIE(InfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True) - fields = self._form_hidden_inputs(login_form) - token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token') + fields = self._hidden_inputs(login_form) + token, vuid = self._extract_xsrft_and_vuid(webpage) fields['token'] = token fields['password'] = password post = urlencode_postdata(fields) @@ -462,6 +494,7 @@ class VimeoChannelIE(InfoExtractor): password_url = compat_urlparse.urljoin(page_url, password_path) password_request = compat_urllib_request.Request(password_url, post) password_request.add_header('Content-type', 'application/x-www-form-urlencoded') + password_request.add_header('Cookie', 'vuid=%s' % vuid) self._set_cookie('vimeo.com', 'xsrft', token) return self._download_webpage( @@ -499,7 +532,7 @@ class VimeoChannelIE(InfoExtractor): class VimeoUserIE(VimeoChannelIE): IE_NAME = 'vimeo:user' - _VALID_URL = r'https://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>' _TESTS = [{ 'url': 'https://vimeo.com/nkistudio/videos', @@ -603,14 +636,14 @@ class VimeoReviewIE(InfoExtractor): return self.url_result(player_url, 'Vimeo', video_id) -class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): +class VimeoWatchLaterIE(VimeoChannelIE): IE_NAME = 'vimeo:watchlater' IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)' - _VALID_URL = r'https://vimeo\.com/home/watchlater|:vimeowatchlater' + _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater' + _TITLE = 'Watch Later' _LOGIN_REQUIRED = True - _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<' _TESTS = [{ - 'url': 'https://vimeo.com/home/watchlater', + 'url': 'https://vimeo.com/watchlater', 'only_matching': True, }] @@ -626,7 +659,7 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): return request def _real_extract(self, url): - return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater') + return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') class VimeoLikesIE(InfoExtractor): diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py index c733a48fa..cb2a4b0b5 100644 --- a/youtube_dl/extractor/vine.py +++ b/youtube_dl/extractor/vine.py @@ -1,10 +1,14 @@ +# coding: utf-8 from __future__ import unicode_literals import re import itertools from .common import InfoExtractor -from ..utils import unified_strdate +from ..utils import ( + int_or_none, + unified_strdate, +) class VineIE(InfoExtractor): @@ -17,10 +21,12 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': 'Chicken.', 'alt_title': 'Vine by Jack Dorsey', - 'description': 'Chicken.', 'upload_date': '20130519', 'uploader': 'Jack Dorsey', 'uploader_id': '76', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/MYxVapFvz2z', @@ -29,11 +35,13 @@ class VineIE(InfoExtractor): 'id': 'MYxVapFvz2z', 'ext': 'mp4', 'title': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', - 'alt_title': 'Vine by Luna', - 'description': 'Fuck Da Police #Mikebrown #justice #ferguson #prayforferguson #protesting #NMOS14', + 'alt_title': 'Vine by Mars Ruiz', 'upload_date': '20140815', - 'uploader': 'Luna', + 'uploader': 'Mars Ruiz', 'uploader_id': '1102363502380728320', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/v/bxVjBbZlPUH', @@ -43,14 +51,33 @@ class VineIE(InfoExtractor): 'ext': 'mp4', 'title': '#mw3 #ac130 #killcam #angelofdeath', 'alt_title': 'Vine by Z3k3', - 'description': '#mw3 #ac130 #killcam #angelofdeath', 'upload_date': '20130430', 'uploader': 'Z3k3', 'uploader_id': '936470460173008896', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, }, }, { 'url': 'https://vine.co/oembed/MYxVapFvz2z.json', 'only_matching': True, + }, { + 'url': 'https://vine.co/v/e192BnZnZ9V', + 'info_dict': { + 'id': 'e192BnZnZ9V', + 'ext': 'mp4', + 'title': 'ยิ้ม~ เขิน~ อาย~ น่าร้ากอ้ะ >//< @n_whitewo @orlameena #lovesicktheseries #lovesickseason2', + 'alt_title': 'Vine by Pimry_zaa', + 'upload_date': '20150705', + 'uploader': 'Pimry_zaa', + 'uploader_id': '1135760698325307392', + 'like_count': int, + 'comment_count': int, + 'repost_count': int, + }, + 'params': { + 'skip_download': True, + }, }] def _real_extract(self, url): @@ -58,32 +85,33 @@ class VineIE(InfoExtractor): webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id) data = self._parse_json( - self._html_search_regex( - r'window\.POST_DATA = { %s: ({.+?}) };\s*</script>' % video_id, + self._search_regex( + r'window\.POST_DATA\s*=\s*{\s*%s\s*:\s*({.+?})\s*};\s*</script>' % video_id, webpage, 'vine data'), video_id) formats = [{ 'format_id': '%(format)s-%(rate)s' % f, - 'vcodec': f['format'], - 'quality': f['rate'], + 'vcodec': f.get('format'), + 'quality': f.get('rate'), 'url': f['videoUrl'], - } for f in data['videoUrls']] + } for f in data['videoUrls'] if f.get('videoUrl')] self._sort_formats(formats) + username = data.get('username') + return { 'id': video_id, - 'title': self._og_search_title(webpage), - 'alt_title': self._og_search_description(webpage, default=None), - 'description': data['description'], - 'thumbnail': data['thumbnailUrl'], - 'upload_date': unified_strdate(data['created']), - 'uploader': data['username'], - 'uploader_id': data['userIdStr'], - 'like_count': data['likes']['count'], - 'comment_count': data['comments']['count'], - 'repost_count': data['reposts']['count'], + 'title': data.get('description') or self._og_search_title(webpage), + 'alt_title': 'Vine by %s' % username if username else self._og_search_description(webpage, default=None), + 'thumbnail': data.get('thumbnailUrl'), + 'upload_date': unified_strdate(data.get('created')), + 'uploader': username, + 'uploader_id': data.get('userIdStr'), + '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')), 'formats': formats, } diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 8ac3aeac0..765e9e6fd 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -17,10 +17,12 @@ from ..utils import ( unescapeHTML, unified_strdate, ) +from .vimeo import VimeoIE class VKIE(InfoExtractor): - IE_NAME = 'vk.com' + IE_NAME = 'vk' + IE_DESC = 'VK' _VALID_URL = r'''(?x) https?:// (?: @@ -154,6 +156,11 @@ class VKIE(InfoExtractor): 'only_matching': True, }, { + # age restricted video, requires vk account credentials + 'url': 'https://vk.com/video205387401_164765225', + 'only_matching': True, + }, + { # vk wrapper 'url': 'http://www.biqle.ru/watch/847655_160197695', 'only_matching': True, @@ -168,7 +175,7 @@ class VKIE(InfoExtractor): login_page = self._download_webpage( 'https://vk.com', None, 'Downloading login page') - login_form = self._form_hidden_inputs(login_page) + login_form = self._hidden_inputs(login_page) login_form.update({ 'email': username.encode('cp1251'), @@ -204,6 +211,12 @@ class VKIE(InfoExtractor): info_page = self._download_webpage(info_url, video_id) + error_message = self._html_search_regex( + r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>', + info_page, 'error message', default=None) + if error_message: + raise ExtractorError(error_message, expected=True) + if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page): raise ExtractorError( 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', @@ -237,6 +250,10 @@ class VKIE(InfoExtractor): if youtube_url: return self.url_result(youtube_url, 'Youtube') + vimeo_url = VimeoIE._extract_vimeo_url(url, info_page) + if vimeo_url is not None: + return self.url_result(vimeo_url) + m_rutube = re.search( r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page) if m_rutube is not None: @@ -289,25 +306,34 @@ class VKIE(InfoExtractor): class VKUserVideosIE(InfoExtractor): - IE_NAME = 'vk.com:user-videos' - IE_DESC = 'vk.com:All of a user\'s videos' - _VALID_URL = r'https?://vk\.com/videos(?P<id>[0-9]+)(?:m\?.*)?' + IE_NAME = 'vk:uservideos' + IE_DESC = "VK - User's Videos" + _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)$' _TEMPLATE_URL = 'https://vk.com/videos' - _TEST = { + _TESTS = [{ 'url': 'http://vk.com/videos205387401', 'info_dict': { 'id': '205387401', + 'title': "Tom Cruise's Videos", }, 'playlist_mincount': 4, - } + }, { + 'url': 'http://vk.com/videos-77521', + 'only_matching': True, + }] def _real_extract(self, url): page_id = self._match_id(url) - page = self._download_webpage(url, page_id) - video_ids = orderedSet( - m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page)) - url_entries = [ + + webpage = self._download_webpage(url, page_id) + + entries = [ self.url_result( 'http://vk.com/video' + video_id, 'VK', video_id=video_id) - for video_id in video_ids] - return self.playlist_result(url_entries, page_id) + for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))] + + title = unescapeHTML(self._search_regex( + r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos', + webpage, 'title', default=page_id)) + + return self.playlist_result(entries, page_id, title) diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py new file mode 100644 index 000000000..86c1cb5ef --- /dev/null +++ b/youtube_dl/extractor/vlive.py @@ -0,0 +1,86 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import hmac +from hashlib import sha1 +from base64 import b64encode +from time import time + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + determine_ext +) +from ..compat import compat_urllib_parse + + +class VLiveIE(InfoExtractor): + IE_NAME = 'vlive' + # www.vlive.tv/video/ links redirect to m.vlive.tv/video/ for mobile devices + _VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://m.vlive.tv/video/1326', + 'md5': 'cc7314812855ce56de70a06a27314983', + 'info_dict': { + 'id': '1326', + 'ext': 'mp4', + 'title': '[V] Girl\'s Day\'s Broadcast', + 'creator': 'Girl\'s Day', + }, + } + _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH' + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage( + 'http://m.vlive.tv/video/%s' % video_id, + video_id, note='Download video page') + + title = self._og_search_title(webpage) + thumbnail = self._og_search_thumbnail(webpage) + creator = self._html_search_regex( + r'<span[^>]+class="name">([^<>]+)</span>', webpage, 'creator') + + url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id + msgpad = '%.0f' % (time() * 1000) + md = b64encode( + hmac.new(self._SECRET.encode('ascii'), + (url[:255] + msgpad).encode('ascii'), sha1).digest() + ) + url += '&' + compat_urllib_parse.urlencode({'msgpad': msgpad, 'md': md}) + playinfo = self._download_json(url, video_id, 'Downloading video json') + + if playinfo.get('message', '') != 'success': + raise ExtractorError(playinfo.get('message', 'JSON request unsuccessful')) + + if not playinfo.get('result'): + raise ExtractorError('No videos found.') + + formats = [] + for vid in playinfo['result'].get('videos', {}).get('list', []): + formats.append({ + 'url': vid['source'], + 'ext': 'mp4', + 'abr': vid.get('bitrate', {}).get('audio'), + 'vbr': vid.get('bitrate', {}).get('video'), + 'format_id': vid['encodingOption']['name'], + 'height': vid.get('height'), + 'width': vid.get('width'), + }) + self._sort_formats(formats) + + subtitles = {} + for caption in playinfo['result'].get('captions', {}).get('list', []): + subtitles[caption['language']] = [ + {'ext': determine_ext(caption['source'], default_ext='vtt'), + 'url': caption['source']}] + + return { + 'id': video_id, + 'title': title, + 'creator': creator, + 'thumbnail': thumbnail, + 'formats': formats, + 'subtitles': subtitles, + } diff --git a/youtube_dl/extractor/vodlocker.py b/youtube_dl/extractor/vodlocker.py index 4804692bf..ccf1928b5 100644 --- a/youtube_dl/extractor/vodlocker.py +++ b/youtube_dl/extractor/vodlocker.py @@ -26,7 +26,7 @@ class VodlockerIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - fields = self._form_hidden_inputs(webpage) + fields = self._hidden_inputs(webpage) if fields['op'] == 'download1': self._sleep(3, video_id) # they do detect when requests happen too fast! diff --git a/youtube_dl/extractor/washingtonpost.py b/youtube_dl/extractor/washingtonpost.py index 72eb010f8..ec8b99998 100644 --- a/youtube_dl/extractor/washingtonpost.py +++ b/youtube_dl/extractor/washingtonpost.py @@ -19,25 +19,25 @@ class WashingtonPostIE(InfoExtractor): 'title': 'Sinkhole of bureaucracy', }, 'playlist': [{ - 'md5': '79132cc09ec5309fa590ae46e4cc31bc', + 'md5': 'b9be794ceb56c7267d410a13f99d801a', 'info_dict': { 'id': 'fc433c38-b146-11e3-b8b3-44b1d1cd4c1f', 'ext': 'mp4', 'title': 'Breaking Points: The Paper Mine', - 'duration': 1287, + 'duration': 1290, 'description': 'Overly complicated paper pushing is nothing new to government bureaucracy. But the way federal retirement applications are filed may be the most outdated. David Fahrenthold explains.', 'uploader': 'The Washington Post', 'timestamp': 1395527908, 'upload_date': '20140322', }, }, { - 'md5': 'e1d5734c06865cc504ad99dc2de0d443', + 'md5': '1fff6a689d8770966df78c8cb6c8c17c', 'info_dict': { 'id': '41255e28-b14a-11e3-b8b3-44b1d1cd4c1f', 'ext': 'mp4', 'title': 'The town bureaucracy sustains', 'description': 'Underneath the friendly town of Boyers is a sea of government paperwork. In a disused limestone mine, hundreds of locals now track, file and process retirement applications for the federal government. We set out to find out what it\'s like to do paperwork 230 feet underground.', - 'duration': 2217, + 'duration': 2220, 'timestamp': 1395528005, 'upload_date': '20140322', 'uploader': 'The Washington Post', diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py index f69d46a28..e4f50e64c 100644 --- a/youtube_dl/extractor/wimp.py +++ b/youtube_dl/extractor/wimp.py @@ -1,40 +1,33 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from .youtube import YoutubeIE class WimpIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?wimp\.com/([^/]+)/' + _VALID_URL = r'http://(?:www\.)?wimp\.com/(?P<id>[^/]+)/' _TESTS = [{ 'url': 'http://www.wimp.com/maruexhausted/', - 'md5': 'f1acced123ecb28d9bb79f2479f2b6a1', + 'md5': 'ee21217ffd66d058e8b16be340b74883', 'info_dict': { 'id': 'maruexhausted', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Maru is exhausted.', 'description': 'md5:57e099e857c0a4ea312542b684a869b8', } }, { - # youtube video 'url': 'http://www.wimp.com/clowncar/', + 'md5': '4e2986c793694b55b37cf92521d12bb4', 'info_dict': { - 'id': 'cG4CEr2aiSg', + 'id': 'clowncar', 'ext': 'mp4', - 'title': 'Basset hound clown car...incredible!', - 'description': 'md5:8d228485e0719898c017203f900b3a35', - 'uploader': 'Gretchen Hoey', - 'uploader_id': 'gretchenandjeff1', - 'upload_date': '20140303', + 'title': 'It\'s like a clown car.', + 'description': 'md5:0e56db1370a6e49c5c1d19124c0d2fb2', }, - 'add_ie': ['Youtube'], }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) video_url = self._search_regex( [r"[\"']file[\"']\s*[:,]\s*[\"'](.+?)[\"']", r"videoId\s*:\s*[\"']([^\"']+)[\"']"], diff --git a/youtube_dl/extractor/xbef.py b/youtube_dl/extractor/xbef.py index 80c48c37d..4ff99e5ca 100644 --- a/youtube_dl/extractor/xbef.py +++ b/youtube_dl/extractor/xbef.py @@ -1,9 +1,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, -) +from ..compat import compat_urllib_parse_unquote class XBefIE(InfoExtractor): @@ -30,7 +28,7 @@ class XBefIE(InfoExtractor): config_url_enc = self._download_webpage( 'http://xbef.com/Main/GetVideoURLEncoded/%s' % video_id, video_id, note='Retrieving config URL') - config_url = compat_urllib_parse.unquote(config_url_enc) + config_url = compat_urllib_parse_unquote(config_url_enc) config = self._download_xml( config_url, video_id, note='Retrieving config') diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index b4ad513a0..8938c0e45 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -4,7 +4,6 @@ import re from .common import InfoExtractor from ..utils import ( - ExtractorError, unified_strdate, str_to_int, int_or_none, @@ -22,7 +21,7 @@ class XHamsterIE(InfoExtractor): 'ext': 'mp4', 'title': 'FemaleAgent Shy beauty takes the bait', 'upload_date': '20121014', - 'uploader_id': 'Ruseful2011', + 'uploader': 'Ruseful2011', 'duration': 893, 'age_limit': 18, } @@ -34,7 +33,7 @@ class XHamsterIE(InfoExtractor): 'ext': 'mp4', 'title': 'Britney Spears Sexy Booty', 'upload_date': '20130914', - 'uploader_id': 'jojo747400', + 'uploader': 'jojo747400', 'duration': 200, 'age_limit': 18, } @@ -46,12 +45,12 @@ class XHamsterIE(InfoExtractor): ] def _real_extract(self, url): - def extract_video_url(webpage): - mp4 = re.search(r'<video\s+.*?file="([^"]+)".*?>', webpage) - if mp4 is None: - raise ExtractorError('Unable to extract media URL') - else: - return mp4.group(1) + def extract_video_url(webpage, name): + return self._search_regex( + [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''', + r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''', + r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''], + webpage, name, group='mp4') def is_hd(webpage): return '<div class=\'icon iconHD\'' in webpage @@ -64,7 +63,9 @@ class XHamsterIE(InfoExtractor): mrss_url = '%s://xhamster.com/movies/%s/%s.html' % (proto, video_id, seo) webpage = self._download_webpage(mrss_url, video_id) - title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage, 'title') + title = self._html_search_regex( + [r'<title>(?P<title>.+?)(?:, (?:[^,]+? )?Porn: xHamster| - xHamster\.com)</title>', + r'<h1>([^<]+)</h1>'], webpage, 'title') # Only a few videos have an description mobj = re.search(r'<span>Description: </span>([^<]+)', webpage) @@ -75,10 +76,14 @@ class XHamsterIE(InfoExtractor): if upload_date: upload_date = unified_strdate(upload_date) - uploader_id = self._html_search_regex(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^<]+)', - webpage, 'uploader id', default='anonymous') + uploader = self._html_search_regex( + r"<a href='[^']+xhamster\.com/user/[^>]+>(?P<uploader>[^<]+)", + webpage, 'uploader', default='anonymous') - thumbnail = self._html_search_regex(r'<video\s+.*?poster="([^"]+)".*?>', webpage, 'thumbnail', fatal=False) + thumbnail = self._search_regex( + [r'''thumb\s*:\s*(?P<q>["'])(?P<thumbnail>.+?)(?P=q)''', + r'''<video[^>]+poster=(?P<q>["'])(?P<thumbnail>.+?)(?P=q)[^>]*>'''], + webpage, 'thumbnail', fatal=False, group='thumbnail') duration = parse_duration(self._html_search_regex(r'<span>Runtime:</span> (\d+:\d+)</div>', webpage, 'duration', fatal=False)) @@ -97,7 +102,9 @@ class XHamsterIE(InfoExtractor): hd = is_hd(webpage) - video_url = extract_video_url(webpage) + format_id = 'hd' if hd else 'sd' + + video_url = extract_video_url(webpage, format_id) formats = [{ 'url': video_url, 'format_id': 'hd' if hd else 'sd', @@ -108,7 +115,7 @@ class XHamsterIE(InfoExtractor): mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url') webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage') if is_hd(webpage): - video_url = extract_video_url(webpage) + video_url = extract_video_url(webpage, 'hd') formats.append({ 'url': video_url, 'format_id': 'hd', @@ -122,7 +129,7 @@ class XHamsterIE(InfoExtractor): 'title': title, 'description': description, 'upload_date': upload_date, - 'uploader_id': uploader_id, + 'uploader': uploader, 'thumbnail': thumbnail, 'duration': duration, 'view_count': view_count, diff --git a/youtube_dl/extractor/xnxx.py b/youtube_dl/extractor/xnxx.py index 79ed6c744..5a41f8ffa 100644 --- a/youtube_dl/extractor/xnxx.py +++ b/youtube_dl/extractor/xnxx.py @@ -2,9 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse, -) +from ..compat import compat_urllib_parse_unquote class XNXXIE(InfoExtractor): @@ -26,7 +24,7 @@ class XNXXIE(InfoExtractor): video_url = self._search_regex(r'flv_url=(.*?)&', webpage, 'video URL') - video_url = compat_urllib_parse.unquote(video_url) + video_url = compat_urllib_parse_unquote(video_url) video_title = self._html_search_regex(r'<title>(.*?)\s+-\s+XNXX.COM', webpage, 'title') diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 1644f53c8..779e4f46a 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -5,7 +5,7 @@ import re from .common import InfoExtractor from ..compat import ( compat_urllib_request, - compat_urllib_parse, + compat_urllib_parse_unquote, ) from ..utils import ( parse_duration, @@ -59,7 +59,7 @@ class XTubeIE(InfoExtractor): for format_id, video_url in re.findall( r'flashvars\.quality_(.+?)\s*=\s*"([^"]+)"', webpage): fmt = { - 'url': compat_urllib_parse.unquote(video_url), + 'url': compat_urllib_parse_unquote(video_url), 'format_id': format_id, } m = re.search(r'^(?P<height>\d+)[pP]', format_id) @@ -68,7 +68,7 @@ class XTubeIE(InfoExtractor): formats.append(fmt) if not formats: - video_url = compat_urllib_parse.unquote(self._search_regex( + video_url = compat_urllib_parse_unquote(self._search_regex( r'flashvars\.video_url\s*=\s*"([^"]+)"', webpage, 'video URL')) formats.append({'url': video_url}) diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index 5aac8adb3..8bbac54e2 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -19,7 +19,7 @@ class XuiteIE(InfoExtractor): _TESTS = [{ # Audio 'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2', - 'md5': '63a42c705772aa53fd4c1a0027f86adf', + 'md5': 'e79284c87b371424885448d11f6398c8', 'info_dict': { 'id': '3860914', 'ext': 'mp3', diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py index d8415bed4..5dcf2fdd1 100644 --- a/youtube_dl/extractor/xvideos.py +++ b/youtube_dl/extractor/xvideos.py @@ -4,7 +4,7 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, + compat_urllib_parse_unquote, compat_urllib_request, ) from ..utils import ( @@ -37,7 +37,7 @@ 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( + 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') diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index f9afbdbab..fca5ddc69 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -101,7 +101,7 @@ class YahooIE(InfoExtractor): } }, { 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', - 'md5': '67010fdf3a08d290e060a4dd96baa07b', + 'md5': '88e209b417f173d86186bef6e4d1f160', 'info_dict': { 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', 'ext': 'mp4', @@ -144,6 +144,17 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://tw.news.yahoo.com/-100120367.html', 'only_matching': True, + }, { + # 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', + 'info_dict': { + 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', + 'ext': 'mp4', + 'title': 'Communitary - Community Episode 1: Ladders', + 'description': 'md5:8fc39608213295748e1e289807838c97', + 'duration': 1646, + }, } ] @@ -171,6 +182,19 @@ class YahooIE(InfoExtractor): if nbc_sports_url: return self.url_result(nbc_sports_url, 'NBCSportsVPlayer') + # 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 + # embedded query result when present sounds reasonable. + config_json = self._search_regex( + r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)', + webpage, 'videoplayer applet', default=None) + if config_json: + config = self._parse_json(config_json, display_id, fatal=False) + if config: + sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') + if sapi: + return self._extract_info(display_id, sapi, webpage) + items_json = self._search_regex( r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, default=None) @@ -190,22 +214,10 @@ class YahooIE(InfoExtractor): video_id = info['id'] return self._get_info(video_id, display_id, webpage) - def _get_info(self, video_id, display_id, webpage): - region = self._search_regex( - r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', - webpage, 'region', fatal=False, default='US') - data = compat_urllib_parse.urlencode({ - 'protocol': 'http', - 'region': region, - }) - query_url = ( - 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' - '{id}?{data}'.format(id=video_id, data=data)) - query_result = self._download_json( - query_url, display_id, 'Downloading video info') - - info = query_result['query']['results']['mediaObj'][0] + def _extract_info(self, display_id, query, webpage): + info = query['query']['results']['mediaObj'][0] meta = info.get('meta') + video_id = info.get('id') if not meta: msg = info['status'].get('msg') @@ -231,6 +243,9 @@ class YahooIE(InfoExtractor): 'ext': 'flv', }) else: + if s.get('format') == 'm3u8_playlist': + format_info['protocol'] = 'm3u8_native' + format_info['ext'] = 'mp4' format_url = compat_urlparse.urljoin(host, path) format_info['url'] = format_url formats.append(format_info) @@ -264,6 +279,21 @@ class YahooIE(InfoExtractor): 'subtitles': subtitles, } + def _get_info(self, video_id, display_id, webpage): + region = self._search_regex( + r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', + webpage, 'region', fatal=False, default='US') + data = compat_urllib_parse.urlencode({ + 'protocol': 'http', + 'region': region, + }) + query_url = ( + 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + '{id}?{data}'.format(id=video_id, data=data)) + query_result = self._download_json( + query_url, display_id, 'Downloading video info') + return self._extract_info(display_id, query_result, webpage) + class YahooSearchIE(SearchInfoExtractor): IE_DESC = 'Yahoo screen search' diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index f4c0f5702..08dc81f3a 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -1,18 +1,38 @@ -# coding=utf-8 +# coding: utf-8 from __future__ import unicode_literals import re import hashlib from .common import InfoExtractor -from ..compat import compat_str +from ..compat import ( + compat_str, + compat_urllib_parse, + compat_urllib_request, +) from ..utils import ( int_or_none, float_or_none, ) -class YandexMusicBaseIE(InfoExtractor): +class YandexMusicTrackIE(InfoExtractor): + IE_NAME = 'yandexmusic:track' + IE_DESC = 'Яндекс.Музыка - Трек' + _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)' + + _TEST = { + 'url': 'http://music.yandex.ru/album/540508/track/4878838', + 'md5': 'f496818aa2f60b6c0062980d2e00dc20', + 'info_dict': { + 'id': '4878838', + 'ext': 'mp3', + 'title': 'Carlo Ambrosio - Gypsy Eyes 1', + 'filesize': 4628061, + 'duration': 193.04, + } + } + def _get_track_url(self, storage_dir, track_id): data = self._download_json( 'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s' @@ -26,6 +46,12 @@ class YandexMusicBaseIE(InfoExtractor): % (data['host'], key, data['ts'] + data['path'], storage[1])) def _get_track_info(self, track): + thumbnail = None + cover_uri = track.get('albums', [{}])[0].get('coverUri') + if cover_uri: + thumbnail = cover_uri.replace('%%', 'orig') + if not thumbnail.startswith('http'): + thumbnail = 'http://' + thumbnail return { 'id': track['id'], 'ext': 'mp3', @@ -33,26 +59,9 @@ class YandexMusicBaseIE(InfoExtractor): 'title': '%s - %s' % (track['artists'][0]['name'], track['title']), 'filesize': int_or_none(track.get('fileSize')), 'duration': float_or_none(track.get('durationMs'), 1000), + 'thumbnail': thumbnail, } - -class YandexMusicTrackIE(YandexMusicBaseIE): - IE_NAME = 'yandexmusic:track' - IE_DESC = 'Яндекс.Музыка - Трек' - _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)' - - _TEST = { - 'url': 'http://music.yandex.ru/album/540508/track/4878838', - 'md5': 'f496818aa2f60b6c0062980d2e00dc20', - 'info_dict': { - 'id': '4878838', - 'ext': 'mp3', - 'title': 'Carlo Ambrosio - Gypsy Eyes 1', - 'filesize': 4628061, - 'duration': 193.04, - } - } - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) album_id, track_id = mobj.group('album_id'), mobj.group('id') @@ -64,7 +73,15 @@ class YandexMusicTrackIE(YandexMusicBaseIE): return self._get_track_info(track) -class YandexMusicAlbumIE(YandexMusicBaseIE): +class YandexMusicPlaylistBaseIE(InfoExtractor): + def _build_playlist(self, tracks): + return [ + self.url_result( + 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id'])) + for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)] + + +class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE): IE_NAME = 'yandexmusic:album' IE_DESC = 'Яндекс.Музыка - Альбом' _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)' @@ -85,7 +102,7 @@ class YandexMusicAlbumIE(YandexMusicBaseIE): 'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id, album_id, 'Downloading album JSON') - entries = [self._get_track_info(track) for track in album['volumes'][0]] + entries = self._build_playlist(album['volumes'][0]) title = '%s - %s' % (album['artists'][0]['name'], album['title']) year = album.get('year') @@ -95,12 +112,12 @@ class YandexMusicAlbumIE(YandexMusicBaseIE): return self.playlist_result(entries, compat_str(album['id']), title) -class YandexMusicPlaylistIE(YandexMusicBaseIE): +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+)' - _TEST = { + _TESTS = [{ 'url': 'http://music.yandex.ru/users/music.partners/playlists/1245', 'info_dict': { 'id': '1245', @@ -108,20 +125,54 @@ class YandexMusicPlaylistIE(YandexMusicBaseIE): 'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9', }, 'playlist_count': 6, - } + }, { + # playlist exceeding the limit of 150 tracks shipped with webpage (see + # https://github.com/rg3/youtube-dl/issues/6666) + 'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036', + 'info_dict': { + 'id': '1036', + 'title': 'Музыка 90-х', + }, + 'playlist_count': 310, + }] def _real_extract(self, url): playlist_id = self._match_id(url) webpage = self._download_webpage(url, playlist_id) - playlist = self._parse_json( + mu = self._parse_json( self._search_regex( r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'), - playlist_id)['pageData']['playlist'] - - entries = [self._get_track_info(track) for track in playlist['tracks']] + playlist_id) + + playlist = mu['pageData']['playlist'] + tracks, track_ids = playlist['tracks'], playlist['trackIds'] + + # tracks dictionary shipped with webpage 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 = compat_urllib_request.Request( + 'https://music.yandex.ru/handlers/track-entries.jsx', + compat_urllib_parse.urlencode({ + 'entries': ','.join(missing_track_ids), + 'lang': mu.get('settings', {}).get('lang', 'en'), + 'external-domain': 'music.yandex.ru', + 'overembed': 'false', + 'sign': mu.get('authData', {}).get('user', {}).get('sign'), + 'strict': 'true', + }).encode('utf-8')) + 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( - entries, compat_str(playlist_id), + self._build_playlist(tracks), + compat_str(playlist_id), playlist['title'], playlist.get('description')) diff --git a/youtube_dl/extractor/ynet.py b/youtube_dl/extractor/ynet.py index 894678a23..869f3e819 100644 --- a/youtube_dl/extractor/ynet.py +++ b/youtube_dl/extractor/ynet.py @@ -5,7 +5,7 @@ import re import json from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import compat_urllib_parse_unquote_plus class YnetIE(InfoExtractor): @@ -34,7 +34,7 @@ class YnetIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - content = compat_urllib_parse.unquote_plus(self._og_search_video_url(webpage)) + content = compat_urllib_parse_unquote_plus(self._og_search_video_url(webpage)) config = json.loads(self._search_regex(r'config=({.+?})$', content, 'video config')) f4m_url = config['clip']['url'] title = self._og_search_title(webpage) diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 78caeb8b3..2e81d9223 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -49,6 +49,17 @@ class YoukuIE(InfoExtractor): }, 'playlist_count': 13, 'skip': 'Available in China only', + }, { + 'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html', + 'note': 'Video protected with password', + 'info_dict': { + 'id': 'XNjA1NzA2Njgw', + 'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起', + }, + 'playlist_count': 19, + 'params': { + 'videopassword': '100600', + }, }] def construct_video_urls(self, data1, data2): @@ -185,9 +196,15 @@ class YoukuIE(InfoExtractor): raw_data = self._download_json(req, video_id, note=note) return raw_data['data'][0] + video_password = self._downloader.params.get('videopassword', None) + # request basic data + basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id + if video_password: + basic_data_url += '?password=%s' % video_password + data1 = retrieve_data( - 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id, + basic_data_url, 'Downloading JSON metadata 1') data2 = retrieve_data( 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id, diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 4ba7c36db..9bf8d1eeb 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -1,121 +1,171 @@ from __future__ import unicode_literals - -import json import re -import sys from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urllib_request, -) +from ..compat import compat_urllib_request from ..utils import ( - ExtractorError, + int_or_none, + str_to_int, unescapeHTML, unified_strdate, ) -from ..aes import ( - aes_decrypt_text -) +from ..aes import aes_decrypt_text class YouPornIE(InfoExtractor): - _VALID_URL = r'^(?P<proto>https?://)(?:www\.)?(?P<url>youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+))' - _TEST = { + _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', 'info_dict': { 'id': '505835', + 'display_id': 'sex-ed-is-it-safe-to-masturbate-daily', 'ext': 'mp4', - 'upload_date': '20101221', + 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', 'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?', + 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Ask Dan And Jennifer', - 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', + 'upload_date': '20101221', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, 'age_limit': 18, - } - } + }, + }, { + # Anonymous User uploader + 'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4', + 'info_dict': { + 'id': '561726', + 'display_id': 'big-tits-awesome-brunette-on-amazing-webcam-show', + 'ext': 'mp4', + 'title': 'Big Tits Awesome Brunette On amazing webcam show', + 'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Anonymous User', + 'upload_date': '20111125', + 'average_rating': int, + 'view_count': int, + 'comment_count': int, + 'categories': list, + 'tags': list, + 'age_limit': 18, + }, + 'params': { + 'skip_download': True, + }, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - url = mobj.group('proto') + 'www.' + mobj.group('url') + video_id = mobj.group('id') + display_id = mobj.group('display_id') - req = compat_urllib_request.Request(url) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) - age_limit = self._rta_search(webpage) + request = compat_urllib_request.Request(url) + request.add_header('Cookie', 'age_verified=1') + webpage = self._download_webpage(request, display_id) + + title = self._search_regex( + [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>.+?)\1', + r'<h1[^>]+class=["\']heading\d?["\'][^>]*>([^<])<'], + webpage, 'title', group='title') - # Get JSON parameters - json_params = self._search_regex( - [r'videoJa?son\s*=\s*({.+})', - r'var\s+currentVideo\s*=\s*new\s+Video\((.+?)\)[,;]'], - webpage, 'JSON parameters') - try: - params = json.loads(json_params) - except ValueError: - raise ExtractorError('Invalid JSON') - - self.report_extraction(video_id) - try: - video_title = params['title'] - upload_date = unified_strdate(params['release_date_f']) - video_description = params['description'] - video_uploader = params['submitted_by'] - thumbnail = params['thumbnails'][0]['image'] - except KeyError: - raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1]) - - # Get all of the links from the page - DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' - download_list_html = self._search_regex(DOWNLOAD_LIST_RE, - webpage, 'download list').strip() - LINK_RE = r'<a href="([^"]+)">' - links = re.findall(LINK_RE, download_list_html) - - # Get all encrypted links - encrypted_links = re.findall(r'var encryptedQuality[0-9]{3}URL = \'([a-zA-Z0-9+/]+={0,2})\';', webpage) - for encrypted_link in encrypted_links: - link = aes_decrypt_text(encrypted_link, video_title, 32).decode('utf-8') + links = [] + + sources = self._search_regex( + r'sources\s*:\s*({.+?})', webpage, 'sources', default=None) + if sources: + for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources): + links.append(link) + + # Fallback #1 + for _, link in re.findall( + r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): + links.append(link) + + # Fallback #2, this also contains extra low quality 180p format + for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage): links.append(link) + # Fallback #3, encrypted links + for _, encrypted_link in re.findall( + r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage): + links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8')) + formats = [] - for link in links: - # A link looks like this: - # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0 - # A path looks like this: - # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4 - video_url = unescapeHTML(link) - path = compat_urllib_parse_urlparse(video_url).path - format_parts = path.split('/')[4].split('_')[:2] - - dn = compat_urllib_parse_urlparse(video_url).netloc.partition('.')[0] - - resolution = format_parts[0] - height = int(resolution[:-len('p')]) - bitrate = int(format_parts[1][:-len('k')]) - format = '-'.join(format_parts) + '-' + dn - - formats.append({ + for video_url in set(unescapeHTML(link) for link in links): + f = { 'url': video_url, - 'format': format, - 'format_id': format, - 'height': height, - 'tbr': bitrate, - 'resolution': resolution, - }) - + } + # Video URL's path looks like this: + # /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4 + # We will benefit from it by extracting some metadata + mobj = re.search(r'/(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url) + if mobj: + height = int(mobj.group('height')) + bitrate = int(mobj.group('bitrate')) + f.update({ + 'format_id': '%dp-%dk' % (height, bitrate), + 'height': height, + 'tbr': bitrate, + }) + formats.append(f) self._sort_formats(formats) - if not formats: - raise ExtractorError('ERROR: no known formats available for video') + description = self._html_search_regex( + r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>', + webpage, 'description', default=None) + thumbnail = self._search_regex( + r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1', + webpage, 'thumbnail', fatal=False, group='thumbnail') + + uploader = self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', + webpage, 'uploader', fatal=False) + upload_date = unified_strdate(self._html_search_regex( + r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</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>', + 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)) + comment_count = str_to_int(self._search_regex( + r'>All [Cc]omments? \(([\d,.]+)\)', + webpage, 'comment count', fatal=False)) + + def extract_tag_box(title): + tag_box = self._search_regex( + (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*' + '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title), + webpage, '%s tag box' % title, default=None) + if not tag_box: + return [] + return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box) + + categories = extract_tag_box('Category') + tags = extract_tag_box('Tags') return { 'id': video_id, - 'uploader': video_uploader, - 'upload_date': upload_date, - 'title': video_title, + 'display_id': display_id, + 'title': title, + 'description': description, 'thumbnail': thumbnail, - 'description': video_description, + 'uploader': uploader, + 'upload_date': upload_date, + 'average_rating': average_rating, + 'view_count': view_count, + 'comment_count': comment_count, + 'categories': categories, + 'tags': tags, 'age_limit': age_limit, 'formats': formats, } diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 3c629d38a..d7eda7aa7 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -17,21 +17,29 @@ from ..compat import ( compat_chr, compat_parse_qs, compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urllib_parse_unquote_plus, + compat_urllib_parse_urlparse, compat_urllib_request, compat_urlparse, compat_str, ) from ..utils import ( clean_html, + encode_dict, ExtractorError, float_or_none, get_element_by_attribute, get_element_by_id, int_or_none, orderedSet, + parse_duration, + remove_start, + smuggle_url, str_to_int, unescapeHTML, unified_strdate, + unsmuggle_url, uppercase_escape, ISO3166Utils, ) @@ -40,7 +48,7 @@ from ..utils import ( class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' - _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' + _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge' _NETRC_MACHINE = 'youtube' # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False @@ -104,10 +112,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 'hl': 'en_US', } - # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode - # chokes on unicode - login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items()) - login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') + login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii') req = compat_urllib_request.Request(self._LOGIN_URL, login_data) login_results = self._download_webpage( @@ -122,42 +127,25 @@ class YoutubeBaseInfoExtractor(InfoExtractor): # Two-Factor # TODO add SMS and phone call support - these require making a request and then prompting the user - if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None: - tfa_code = self._get_tfa_info() + if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None: + tfa_code = self._get_tfa_info('2-step verification code') - if tfa_code is None: - self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>') - self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)') + if not tfa_code: + self._downloader.report_warning( + 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>' + '(Note that only TOTP (Google Authenticator App) codes work at this time.)') return False - # Unlike the first login form, secTok and timeStmp are both required for the TFA form - - match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) - if match is None: - self._downloader.report_warning('Failed to get secTok - did the page structure change?') - secTok = match.group(1) - match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) - if match is None: - self._downloader.report_warning('Failed to get timeStmp - did the page structure change?') - timeStmp = match.group(1) - - tfa_form_strs = { - 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', - 'smsToken': '', - 'smsUserPin': tfa_code, - 'smsVerifyPin': 'Verify', - - 'PersistentCookie': 'yes', - 'checkConnection': '', - 'checkedDomains': 'youtube', - 'pstMsg': '1', - 'secTok': secTok, - 'timeStmp': timeStmp, - 'service': 'youtube', - 'hl': 'en_US', - } - tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items()) - tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii') + tfa_code = remove_start(tfa_code, 'G-') + + tfa_form_strs = self._form_hidden_inputs('challenge', login_results) + + tfa_form_strs.update({ + 'Pin': tfa_code, + 'TrustDevice': 'on', + }) + + tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii') tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) tfa_results = self._download_webpage( @@ -167,8 +155,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor): if tfa_results is False: return False - if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None: - self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.') + if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None: + self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.') return False if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None: self._downloader.report_warning('unable to log in - did the page structure change?') @@ -190,6 +178,52 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return +class YoutubePlaylistBaseInfoExtractor(InfoExtractor): + # Extract the video ids from the playlist pages + def _entries(self, page, playlist_id): + more_widget_html = content_html = page + for page_num in itertools.count(1): + for video_id, video_title in self.extract_videos_from_page(content_html): + yield self.url_result( + video_id, 'Youtube', video_id=video_id, + video_title=video_title) + + mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) + if not mobj: + break + + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), playlist_id, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + if not content_html.strip(): + # Some webpages show a "Load more" button but they don't + # have more videos + break + more_widget_html = more['load_more_widget_html'] + + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + for mobj in re.finditer(self._VIDEO_RE, page): + # The link with index 0 is not the first video of the playlist (not sure if still actual) + if 'index' in mobj.groupdict() and mobj.group('id') == '0': + continue + video_id = mobj.group('id') + video_title = unescapeHTML(mobj.group('title')) + if video_title: + video_title = video_title.strip() + try: + idx = ids_in_page.index(video_id) + if video_title and not titles_in_page[idx]: + titles_in_page[idx] = video_title + except ValueError: + ids_in_page.append(video_id) + titles_in_page.append(video_title) + return zip(ids_in_page, titles_in_page) + + class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ @@ -207,11 +241,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): |(?: # or the v= param in all its forms (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! - (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) + (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx) v= ) )) - |youtu\.be/ # just youtu.be/xxxx + |(?: + youtu\.be| # just youtu.be/xxxx + vid\.plus # or vid.plus/xxxx + )/ |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= ) )? # all until now is optional -> you can pass the naked ID @@ -277,13 +314,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, # Dash webm - '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'}, + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'}, '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, @@ -293,11 +330,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, - '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'}, - '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, # Dash webm audio '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, @@ -315,7 +352,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): IE_NAME = 'youtube' _TESTS = [ { - 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc', + 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9', 'info_dict': { 'id': 'BaW_jenozKc', 'ext': 'mp4', @@ -325,8 +362,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20121002', 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], 'like_count': int, 'dislike_count': int, + 'start_time': 1, + 'end_time': 9, } }, { @@ -337,7 +377,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'ext': 'mp4', 'upload_date': '20120506', 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', - 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f', + 'description': 'md5:782e8651347686cba06e58f71ab51773', + 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli', + 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop', + 'iconic ep', 'iconic', 'love', 'it'], 'uploader': 'Icona Pop', 'uploader_id': 'IconaPop', } @@ -353,6 +396,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'md5:64249768eec3bc4276236606ea996373', 'uploader': 'justintimberlakeVEVO', 'uploader_id': 'justintimberlakeVEVO', + 'age_limit': 18, } }, { @@ -369,6 +413,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } }, { + 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY', + 'note': 'Use the first video ID in the URL', + 'info_dict': { + 'id': 'BaW_jenozKc', + 'ext': 'mp4', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', + 'uploader': 'Philipp Hagemeister', + 'uploader_id': 'phihag', + 'upload_date': '20121002', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, + { 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I', 'note': '256k DASH audio (format 141) via DASH manifest', 'info_dict': { @@ -409,7 +473,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'nfWlot6h_JM', 'ext': 'm4a', 'title': 'Taylor Swift - Shake It Off', - 'description': 'md5:2acfda1b285bdd478ccec22f9918199d', + 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3', 'uploader': 'TaylorSwiftVEVO', 'uploader_id': 'TaylorSwiftVEVO', 'upload_date': '20140818', @@ -443,6 +507,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'The Witcher', 'uploader_id': 'WitcherGame', 'upload_date': '20140605', + 'age_limit': 18, }, }, # Age-gate video with encrypted signature @@ -456,6 +521,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'LloydVEVO', 'uploader_id': 'LloydVEVO', 'upload_date': '20110629', + 'age_limit': 18, }, }, # video_info is None (https://github.com/rg3/youtube-dl/issues/4421) @@ -480,7 +546,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'lqQg6PlCWgI', 'ext': 'mp4', - 'upload_date': '20120731', + 'upload_date': '20120724', 'uploader_id': 'olympic', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', 'uploader': 'Olympics', @@ -509,7 +575,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'url': 'qEJwOuvDf7I', 'info_dict': { 'id': 'qEJwOuvDf7I', - 'ext': 'mp4', + 'ext': 'webm', 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге', 'description': '', 'upload_date': '20150404', @@ -533,6 +599,81 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'dorappi2000', 'formats': 'mincount:33', }, + }, + # DASH manifest with segment_list + { + 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8', + 'md5': '8ce563a1d667b599d21064e982ab9e31', + 'info_dict': { + 'id': 'CsmdDsKjzN8', + 'ext': 'mp4', + 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510 + 'uploader': 'Airtek', + 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.', + 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ', + 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '135', # bestvideo + } + }, + { + # Multifeed videos (multiple cameras), URL is for Main Camera + 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs', + 'info_dict': { + 'id': 'jqWvoWXjCVs', + 'title': 'teamPGP: Rocket League Noob Stream', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + }, + 'playlist': [{ + 'info_dict': { + 'id': 'jqWvoWXjCVs', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': '6h8e8xoXJzg', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': 'PUOgX5z9xZw', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (grizzle)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': 'teuwxikvS5k', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (zim)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }], + 'params': { + 'skip_download': True, + }, + }, + { + 'url': 'http://vid.plus/FlRa-iH7PGw', + 'only_matching': True, } ] @@ -562,7 +703,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _extract_signature_function(self, video_id, player_url, example_sig): id_m = re.match( - r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$', + r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$', player_url) if not id_m: raise ExtractorError('Cannot identify player %r' % player_url) @@ -824,6 +965,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # TODO implement WebVTT downloading pass elif mime_type.startswith('audio/') or mime_type.startswith('video/'): + segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList') format_id = r.attrib['id'] video_url = url_el.text filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) @@ -837,6 +979,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'filesize': filesize, 'fps': int_or_none(r.attrib.get('frameRate')), } + if segment_list is not None: + f.update({ + 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'], + 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')], + 'protocol': 'http_dash_segments', + }) try: existing_format = next( fo for fo in formats @@ -858,14 +1006,28 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return formats def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + proto = ( 'http' if self._downloader.params.get('prefer_insecure', False) else 'https') + start_time = None + end_time = None + parsed_url = compat_urllib_parse_urlparse(url) + for component in [parsed_url.fragment, parsed_url.query]: + query = compat_parse_qs(component) + if start_time is None and 't' in query: + start_time = parse_duration(query['t'][0]) + if start_time is None and 'start' in query: + start_time = parse_duration(query['start'][0]) + if end_time is None and 'end' in query: + end_time = parse_duration(query['end'][0]) + # Extract original video URL from URL with redirection, like age verification, using next_url parameter mobj = re.search(self._NEXT_URL_RE, url) if mobj: - url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') + url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/') video_id = self.extract_id(url) # Get video webpage @@ -888,6 +1050,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Get video info embed_webpage = None + is_live = None if re.search(r'player-age-gate-content">', video_webpage) is not None: age_gate = True # We simulate the access to the video from www.youtube.com/v/{video_id} @@ -920,6 +1083,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Convert to the same format returned by compat_parse_qs video_info = dict((k, [v]) for k, v in args.items()) add_dash_mpd(video_info) + if args.get('livestream') == '1' or args.get('live_playback') == 1: + is_live = True if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): # We also try looking in get_video_info since it may contain different dashmpd # URL that points to a DASH manifest with possibly different itag set (some itags @@ -937,7 +1102,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): video_id, note=False, errnote='unable to download video info webpage') get_video_info = compat_parse_qs(video_info_webpage) - add_dash_mpd(get_video_info) + if get_video_info.get('use_cipher_signature') != ['True']: + add_dash_mpd(get_video_info) if not video_info: video_info = get_video_info if 'token' in get_video_info: @@ -946,7 +1112,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if 'reason' in video_info: if 'The uploader has not made this video available in your country.' in video_info['reason']: regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None) - if regions_allowed is not None: + if regions_allowed: raise ExtractorError('YouTube said: This video is available in %s only' % ( ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))), expected=True) @@ -958,6 +1124,55 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '"token" parameter not in video info for unknown reason', video_id=video_id) + # title + if 'title' in video_info: + video_title = video_info['title'][0] + else: + self._downloader.report_warning('Unable to extract video title') + video_title = '_' + + # description + video_description = get_element_by_id("eow-description", video_webpage) + if video_description: + video_description = re.sub(r'''(?x) + <a\s+ + (?:[a-zA-Z-]+="[^"]+"\s+)*? + title="([^"]+)"\s+ + (?:[a-zA-Z-]+="[^"]+"\s+)*? + class="yt-uix-redirect-link"\s*> + [^<]+ + </a> + ''', r'\1', video_description) + video_description = clean_html(video_description) + else: + fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) + if fd_mobj: + video_description = unescapeHTML(fd_mobj.group(1)) + else: + video_description = '' + + if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False): + if not self._downloader.params.get('noplaylist'): + entries = [] + feed_ids = [] + multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0]) + for feed in multifeed_metadata_list.split(','): + feed_data = compat_parse_qs(feed) + entries.append({ + '_type': 'url_transparent', + 'ie_key': 'Youtube', + 'url': smuggle_url( + '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]), + {'force_singlefeed': True}), + 'title': '%s (%s)' % (video_title, feed_data['title'][0]), + }) + feed_ids.append(feed_data['id'][0]) + self.to_screen( + 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' + % (', '.join(feed_ids), video_id)) + return self.playlist_result(entries, video_id, video_title, video_description) + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + if 'view_count' in video_info: view_count = int(video_info['view_count'][0]) else: @@ -973,7 +1188,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # uploader if 'author' not in video_info: raise ExtractorError('Unable to extract uploader name') - video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) + video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0]) # uploader_id video_uploader_id = None @@ -983,13 +1198,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): else: self._downloader.report_warning('unable to extract uploader nickname') - # title - if 'title' in video_info: - video_title = video_info['title'][0] - else: - self._downloader.report_warning('Unable to extract video title') - video_title = '_' - # thumbnail image # We try first to get a high quality image: m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', @@ -1000,7 +1208,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self._downloader.report_warning('unable to extract video thumbnail') video_thumbnail = None else: # don't panic if we can't find it - video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) + video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0]) # upload date upload_date = self._html_search_meta( @@ -1025,25 +1233,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): else: video_categories = None - # description - video_description = get_element_by_id("eow-description", video_webpage) - if video_description: - video_description = re.sub(r'''(?x) - <a\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? - title="([^"]+)"\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? - class="yt-uix-redirect-link"\s*> - [^<]+ - </a> - ''', r'\1', video_description) - video_description = clean_html(video_description) - else: - fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) - if fd_mobj: - video_description = unescapeHTML(fd_mobj.group(1)) - else: - video_description = '' + video_tags = [ + unescapeHTML(m.group('content')) + for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)] def _extract_count(count_name): return str_to_int(self._search_regex( @@ -1062,7 +1254,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self._downloader.report_warning('unable to extract video duration') video_duration = None else: - video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) + video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0])) # annotations video_annotations = None @@ -1094,7 +1286,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0] if 'rtmpe%3Dyes' in encoded_url_map: raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) - url_map = {} + formats = [] for url_data_str in encoded_url_map.split(','): url_data = compat_parse_qs(url_data_str) if 'itag' not in url_data or 'url' not in url_data: @@ -1140,7 +1332,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): player_desc = 'flash player %s' % player_version else: player_version = self._search_regex( - r'html5player-([^/]+?)(?:/html5player)?\.js', + r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', player_url, 'html5 player', fatal=False) player_desc = 'html5 player %s' % player_version @@ -1154,8 +1346,50 @@ class YoutubeIE(YoutubeBaseInfoExtractor): url += '&signature=' + signature if 'ratebypass' not in url: url += '&ratebypass=yes' - url_map[format_id] = url - formats = _map_to_format_list(url_map) + + # Some itags are not included in DASH manifest thus corresponding formats will + # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993). + # Trying to extract metadata from url_encoded_fmt_stream_map entry. + mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0]) + width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) + dct = { + 'format_id': format_id, + 'url': url, + 'player_url': player_url, + 'filesize': int_or_none(url_data.get('clen', [None])[0]), + 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000), + 'width': width, + 'height': height, + 'fps': int_or_none(url_data.get('fps', [None])[0]), + 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0], + } + type_ = url_data.get('type', [None])[0] + if type_: + type_split = type_.split(';') + kind_ext = type_split[0].split('/') + if len(kind_ext) == 2: + kind, ext = kind_ext + dct['ext'] = ext + if kind in ('audio', 'video'): + codecs = None + for mobj in re.finditer( + r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_): + if mobj.group('key') == 'codecs': + codecs = mobj.group('val') + break + if codecs: + codecs = codecs.split(',') + if len(codecs) == 2: + acodec, vcodec = codecs[0], codecs[1] + else: + acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0]) + dct.update({ + 'acodec': acodec, + 'vcodec': vcodec, + }) + if format_id in self._formats: + dct.update(self._formats[format_id]) + formats.append(dct) elif video_info.get('hlsvp'): manifest_url = video_info['hlsvp'][0] url_map = self._extract_from_m3u8(manifest_url, video_id) @@ -1213,6 +1447,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'thumbnail': video_thumbnail, 'description': video_description, 'categories': video_categories, + 'tags': video_tags, 'subtitles': video_subtitles, 'automatic_captions': automatic_captions, 'duration': video_duration, @@ -1224,10 +1459,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'dislike_count': dislike_count, 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]), 'formats': formats, + 'is_live': is_live, + 'start_time': start_time, + 'end_time': end_time, } -class YoutubePlaylistIE(YoutubeBaseInfoExtractor): +class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: (?:https?://)? @@ -1248,7 +1486,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,}) )""" _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' - _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?' IE_NAME = 'youtube:playlist' _TESTS = [{ 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', @@ -1365,37 +1603,11 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): else: self.report_warning('Youtube gives an alert message: ' + match) - # Extract the video ids from the playlist pages - def _entries(): - more_widget_html = content_html = page - for page_num in itertools.count(1): - matches = re.finditer(self._VIDEO_RE, content_html) - # We remove the duplicates and the link with index 0 - # (it's not the first video of the playlist) - new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') - for vid_id in new_ids: - yield self.url_result(vid_id, 'Youtube', video_id=vid_id) - - mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), playlist_id, - 'Downloading page #%s' % page_num, - transform_source=uppercase_escape) - content_html = more['content_html'] - if not content_html.strip(): - # Some webpages show a "Load more" button but they don't - # have more videos - break - more_widget_html = more['load_more_widget_html'] - playlist_title = self._html_search_regex( r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', page, 'title') - return self.playlist_result(_entries(), playlist_id, playlist_title) + return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title) def _real_extract(self, url): # Extract playlist id @@ -1421,36 +1633,31 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): return self._extract_playlist(playlist_id) -class YoutubeChannelIE(InfoExtractor): +class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com channels' _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' + _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?' IE_NAME = 'youtube:channel' _TESTS = [{ 'note': 'paginated channel', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', 'playlist_mincount': 91, 'info_dict': { - 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w', + 'title': 'Uploads from lex will', } + }, { + 'note': 'Age restricted channel', + # from https://www.youtube.com/user/DeusExOfficial + 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w', + 'playlist_mincount': 64, + 'info_dict': { + 'id': 'UUs0ifCMCm1icqRbqhUINa0w', + 'title': 'Uploads from Deus Ex', + }, }] - @staticmethod - def extract_videos_from_page(page): - ids_in_page = [] - titles_in_page = [] - for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page): - video_id = mobj.group('id') - video_title = unescapeHTML(mobj.group('title')) - try: - idx = ids_in_page.index(video_id) - if video_title and not titles_in_page[idx]: - titles_in_page[idx] = video_title - except ValueError: - ids_in_page.append(video_id) - titles_in_page.append(video_title) - return zip(ids_in_page, titles_in_page) - def _real_extract(self, url): channel_id = self._match_id(url) @@ -1462,12 +1669,15 @@ class YoutubeChannelIE(InfoExtractor): channel_page = self._download_webpage( url + '?view=57', channel_id, 'Downloading channel page', fatal=False) - 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-id="([^"]+)"', - channel_page, 'channel id', default=None) + if channel_page is False: + channel_playlist_id = False + else: + 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) if channel_playlist_id and channel_playlist_id.startswith('UC'): playlist_id = 'UU' + channel_playlist_id[2:] return self.url_result( @@ -1490,29 +1700,7 @@ class YoutubeChannelIE(InfoExtractor): for video_id, video_title in self.extract_videos_from_page(channel_page)] return self.playlist_result(entries, channel_id) - def _entries(): - more_widget_html = content_html = channel_page - for pagenum in itertools.count(1): - - for video_id, video_title in self.extract_videos_from_page(content_html): - yield self.url_result( - video_id, 'Youtube', video_id=video_id, - video_title=video_title) - - mobj = re.search( - r'data-uix-load-more-href="/?(?P<more>[^"]+)"', - more_widget_html) - if not mobj: - break - - more = self._download_json( - 'https://youtube.com/%s' % mobj.group('more'), channel_id, - 'Downloading page #%s' % (pagenum + 1), - transform_source=uppercase_escape) - content_html = more['content_html'] - more_widget_html = more['load_more_widget_html'] - - return self.playlist_result(_entries(), channel_id) + return self.playlist_result(self._entries(channel_page, channel_id), channel_id) class YoutubeUserIE(YoutubeChannelIE): @@ -1609,14 +1797,14 @@ class YoutubeSearchURLIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - query = compat_urllib_parse.unquote_plus(mobj.group('query')) + 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) + r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code) entries = [] for part_code in part_codes: part_title = self._html_search_regex( @@ -1643,8 +1831,8 @@ class YoutubeShowIE(InfoExtractor): _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' IE_NAME = 'youtube:show' _TESTS = [{ - 'url': 'http://www.youtube.com/show/airdisasters', - 'playlist_mincount': 3, + 'url': 'https://www.youtube.com/show/airdisasters', + 'playlist_mincount': 5, 'info_dict': { 'id': 'airdisasters', 'title': 'Air Disasters', @@ -1655,7 +1843,7 @@ class YoutubeShowIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) playlist_id = mobj.group('id') webpage = self._download_webpage( - url, playlist_id, 'Downloading show webpage') + 'https://www.youtube.com/show/%s/playlists' % playlist_id, playlist_id, 'Downloading show webpage') # There's one playlist for each season of the show m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) @@ -1778,6 +1966,7 @@ class YoutubeTruncatedURLIE(InfoExtractor): annotation_id=annotation_[^&]+| x-yt-cl=[0-9]+| hl=[^&]*| + t=[0-9]+ )? | attribution_link\?a=[^&]+ @@ -1800,6 +1989,9 @@ class YoutubeTruncatedURLIE(InfoExtractor): }, { 'url': 'https://www.youtube.com/watch?hl=en-GB', 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?t=2372', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 98f15177b..a795f56b3 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -9,6 +9,7 @@ from ..utils import ( int_or_none, unified_strdate, OnDemandPagedList, + xpath_text, ) @@ -19,13 +20,11 @@ def extract_from_xml_url(ie, video_id, xml_url): errnote='Failed to download video info') title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - duration = int(doc.find('.//details/lengthSec').text) - uploader_node = doc.find('.//details/originChannelTitle') - uploader = None if uploader_node is None else uploader_node.text - uploader_id_node = doc.find('.//details/originChannelId') - uploader_id = None if uploader_id_node is None else uploader_id_node.text - upload_date = unified_strdate(doc.find('.//details/airtime').text) + description = xpath_text(doc, './/information/detail', 'description') + duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration')) + uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader') + uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id') + upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date')) def xml_to_format(fnode): video_url = fnode.find('url').text @@ -40,15 +39,14 @@ def extract_from_xml_url(ie, video_id, xml_url): ext = format_m.group('container') proto = format_m.group('proto').lower() - quality = fnode.find('./quality').text - abr = int(fnode.find('./audioBitrate').text) // 1000 - vbr_node = fnode.find('./videoBitrate') - vbr = None if vbr_node is None else int(vbr_node.text) // 1000 + quality = xpath_text(fnode, './quality', 'quality') + abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000) + vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000) - width_node = fnode.find('./width') - width = None if width_node is None else int_or_none(width_node.text) - height_node = fnode.find('./height') - height = None if height_node is None else int_or_none(height_node.text) + width = int_or_none(xpath_text(fnode, './width', 'width')) + height = int_or_none(xpath_text(fnode, './height', 'height')) + + filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize')) format_note = '' if not format_note: @@ -64,12 +62,31 @@ def extract_from_xml_url(ie, video_id, xml_url): 'vbr': vbr, 'width': width, 'height': height, - 'filesize': int_or_none(fnode.find('./filesize').text), + 'filesize': filesize, 'format_note': format_note, 'protocol': proto, '_available': is_available, } + def xml_to_thumbnails(fnode): + thumbnails = [] + for node in fnode: + thumbnail_url = node.text + if not thumbnail_url: + continue + thumbnail = { + 'url': thumbnail_url, + } + if 'key' in node.attrib: + m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key']) + if m: + thumbnail['width'] = int(m.group(1)) + thumbnail['height'] = int(m.group(2)) + thumbnails.append(thumbnail) + return thumbnails + + thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage')) + format_nodes = doc.findall('.//formitaeten/formitaet') formats = list(filter( lambda f: f['_available'], @@ -81,6 +98,7 @@ def extract_from_xml_url(ie, video_id, xml_url): 'title': title, 'description': description, 'duration': duration, + 'thumbnails': thumbnails, 'uploader': uploader, 'uploader_id': uploader_id, 'upload_date': upload_date, diff --git a/youtube_dl/extractor/zingmp3.py b/youtube_dl/extractor/zingmp3.py index 7dc1e2f2b..437eecb67 100644 --- a/youtube_dl/extractor/zingmp3.py +++ b/youtube_dl/extractor/zingmp3.py @@ -9,9 +9,11 @@ from ..utils import ExtractorError class ZingMp3BaseInfoExtractor(InfoExtractor): - def _extract_item(self, item): + def _extract_item(self, item, fatal=True): error_message = item.find('./errormessage').text if error_message: + if not fatal: + return raise ExtractorError( '%s returned error: %s' % (self.IE_NAME, error_message), expected=True) @@ -43,7 +45,9 @@ class ZingMp3BaseInfoExtractor(InfoExtractor): entries = [] for i, item in enumerate(items, 1): - entry = self._extract_item(item) + entry = self._extract_item(item, fatal=False) + if not entry: + continue entry['id'] = '%s-%d' % (id, i) entries.append(entry) @@ -85,7 +89,7 @@ class ZingMp3SongIE(ZingMp3BaseInfoExtractor): class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor): - _VALID_URL = r'https?://mp3\.zing\.vn/album/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html' + _VALID_URL = r'https?://mp3\.zing\.vn/(?:album|playlist)/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html' _TESTS = [{ 'url': 'http://mp3.zing.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html', 'info_dict': { @@ -94,6 +98,9 @@ class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor): 'title': 'Lâu Đài Tình Ái - Bằng Kiều ft. Minh Tuyết | Album 320 lossless', }, 'playlist_count': 10, + }, { + 'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html', + 'only_matching': True, }] IE_NAME = 'zingmp3:album' IE_DESC = 'mp3.zing.vn albums' diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 85365d769..3dd6d290b 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import os.path import optparse -import shlex import sys from .downloader.external import list_external_downloaders @@ -11,6 +10,7 @@ from .compat import ( compat_get_terminal_size, compat_getenv, compat_kwargs, + compat_shlex_split, ) from .utils import ( preferredencoding, @@ -28,7 +28,7 @@ def parseOpts(overrideArguments=None): try: res = [] for l in optionf: - res += shlex.split(l, comments=True) + res += compat_shlex_split(l, comments=True) finally: optionf.close() return res @@ -219,7 +219,7 @@ def parseOpts(overrideArguments=None): selection.add_option( '--playlist-items', dest='playlist_items', metavar='ITEM_SPEC', default=None, - help='Playlist video items to download. Specify indices of the videos in the playlist seperated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.') + help='Playlist video items to download. Specify indices of the videos in the playlist separated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.') selection.add_option( '--match-title', dest='matchtitle', metavar='REGEX', @@ -276,7 +276,7 @@ def parseOpts(overrideArguments=None): 'For example, to only match videos that have been liked more than ' '100 times and disliked less than 50 times (or the dislike ' 'functionality is not available at the given service), but who ' - 'also have a description, use --match-filter ' + 'also have a description, use --match-filter ' '"like_count > 100 & dislike_count <? 50 & description" .' )) selection.add_option( @@ -320,7 +320,7 @@ def parseOpts(overrideArguments=None): authentication.add_option( '--video-password', dest='videopassword', metavar='PASSWORD', - help='Video password (vimeo, smotri)') + help='Video password (vimeo, smotri, youku)') video_format = optparse.OptionGroup(parser, 'Video Format Options') video_format.add_option( @@ -602,7 +602,7 @@ def parseOpts(overrideArguments=None): filesystem.add_option( '-A', '--auto-number', action='store_true', dest='autonumber', default=False, - help='[deprecated; use -o "%(autonumber)s-%(title)s.%(ext)s" ] Number downloaded files starting from 00000') + help='[deprecated; use -o "%(autonumber)s-%(title)s.%(ext)s" ] Number downloaded files starting from 00000') filesystem.add_option( '-t', '--title', action='store_true', dest='usetitle', default=False, diff --git a/youtube_dl/postprocessor/common.py b/youtube_dl/postprocessor/common.py index 4191d040b..599dd1df2 100644 --- a/youtube_dl/postprocessor/common.py +++ b/youtube_dl/postprocessor/common.py @@ -4,6 +4,7 @@ import os from ..utils import ( PostProcessingError, + cli_configuration_args, encodeFilename, ) @@ -61,11 +62,7 @@ class PostProcessor(object): self._downloader.report_warning(errnote) def _configuration_args(self, default=[]): - pp_args = self._downloader.params.get('postprocessor_args') - if pp_args is None: - return default - assert isinstance(pp_args, list) - return pp_args + return cli_configuration_args(self._downloader.params, 'postprocessor_args', default) class AudioConversionError(PostProcessingError): diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 1ecce22e7..4f320e124 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -135,7 +135,10 @@ class FFmpegPostProcessor(PostProcessor): files_cmd = [] for path in input_paths: - files_cmd.extend([encodeArgument('-i'), encodeFilename(path, True)]) + files_cmd.extend([ + encodeArgument('-i'), + encodeFilename(self._ffmpeg_filename_argument(path), True) + ]) cmd = ([encodeFilename(self.executable, True), encodeArgument('-y')] + files_cmd + [encodeArgument(o) for o in opts] + @@ -155,10 +158,10 @@ class FFmpegPostProcessor(PostProcessor): self.run_ffmpeg_multiple_files([path], out_path, opts) def _ffmpeg_filename_argument(self, fn): - # ffmpeg broke --, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details - if fn.startswith('-'): - return './' + fn - return fn + # Always use 'file:' because the filename may contain ':' (ffmpeg + # interprets that as a protocol) or can start with '-' (-- is broken in + # ffmpeg, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details) + return 'file:' + fn class FFmpegExtractAudioPP(FFmpegPostProcessor): @@ -265,7 +268,7 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor): # If we download foo.mp3 and convert it to... foo.mp3, then don't delete foo.mp3, silly. if (new_path == path or (self._nopostoverwrites and os.path.exists(encodeFilename(new_path)))): - self._downloader.to_screen('[youtube] Post-process file %s exists, skipping' % new_path) + self._downloader.to_screen('[ffmpeg] Post-process file %s exists, skipping' % new_path) return [], information try: diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 942f76d24..558c9c7d5 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -3,6 +3,7 @@ from __future__ import unicode_literals +import base64 import calendar import codecs import contextlib @@ -139,21 +140,24 @@ def write_json_file(obj, fn): if sys.version_info >= (2, 7): - def find_xpath_attr(node, xpath, key, val): + def find_xpath_attr(node, xpath, key, val=None): """ Find the xpath xpath[@key=val] """ - assert re.match(r'^[a-zA-Z-]+$', key) - assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) - expr = xpath + "[@%s='%s']" % (key, val) + assert re.match(r'^[a-zA-Z_-]+$', key) + if val: + assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) + expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val)) return node.find(expr) else: - def find_xpath_attr(node, xpath, key, val): + def find_xpath_attr(node, xpath, key, val=None): # Here comes the crazy part: In 2.6, if the xpath is a unicode, # .//node does not match if a node is a direct child of . ! if isinstance(xpath, compat_str): xpath = xpath.encode('ascii') for f in node.findall(xpath): - if f.attrib.get(key) == val: + if key not in f.attrib: + continue + if val is None or f.attrib.get(key) == val: return f return None @@ -173,12 +177,12 @@ def xpath_with_ns(path, ns_map): return '/'.join(replaced) -def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): +def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): if sys.version_info < (2, 7): # Crazy 2.6 xpath = xpath.encode('ascii') n = node.find(xpath) - if n is None or n.text is None: + if n is None: if default is not NO_DEFAULT: return default elif fatal: @@ -186,9 +190,37 @@ def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): raise ExtractorError('Could not find XML element %s' % name) else: return None + return n + + +def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): + n = xpath_element(node, xpath, name, fatal=fatal, default=default) + if n is None or n == default: + return n + if n.text is None: + if default is not NO_DEFAULT: + return default + elif fatal: + name = xpath if name is None else name + raise ExtractorError('Could not find XML element\'s text %s' % name) + else: + return None return n.text +def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): + n = find_xpath_attr(node, xpath, key) + if n is None: + if default is not NO_DEFAULT: + return default + elif fatal: + name = '%s[@%s]' % (xpath, key) if name is None else name + raise ExtractorError('Could not find XML attribute %s' % name) + else: + return None + return n.attrib[key] + + def get_element_by_id(id, html): """Return the content of the tag with the specified ID in the passed HTML document""" return get_element_by_attribute("id", id, html) @@ -576,16 +608,19 @@ class ContentTooShortError(Exception): download is too small for what the server announced first, indicating the connection was probably interrupted. """ - # Both in bytes - downloaded = None - expected = None def __init__(self, downloaded, expected): + # Both in bytes self.downloaded = downloaded self.expected = expected def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): + # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting + # expected HTTP responses to meet HTTP/1.0 or later (see also + # https://github.com/rg3/youtube-dl/issues/6727) + if sys.version_info < (3, 0): + kwargs[b'strict'] = True hc = http_class(*args, **kwargs) source_address = ydl_handler._params.get('source_address') if source_address is not None: @@ -650,6 +685,26 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): return ret def http_request(self, req): + # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not + # always respected by websites, some tend to give out URLs with non percent-encoded + # non-ASCII characters (see telemb.py, ard.py [#3412]) + # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) + # To work around aforementioned issue we will replace request's original URL with + # percent-encoded one + # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) + # the code of this workaround has been moved here from YoutubeDL.urlopen() + url = req.get_full_url() + url_escaped = escape_url(url) + + # Substitute URL if any change after escaping + if url != url_escaped: + req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request + new_req = req_type( + url_escaped, data=req.data, headers=req.headers, + origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) + new_req.timeout = req.timeout + req = new_req + for h, v in std_headers.items(): # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 # The dict keys are capitalized because of this bug by urllib @@ -694,6 +749,18 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): gz = io.BytesIO(self.deflate(resp.read())) resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) resp.msg = old_resp.msg + # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see + # https://github.com/rg3/youtube-dl/issues/6457). + if 300 <= resp.code < 400: + location = resp.headers.get('Location') + if location: + # 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') + location_escaped = escape_url(location) + if location != location_escaped: + del resp.headers['Location'] + resp.headers['Location'] = location_escaped return resp https_request = http_request @@ -717,15 +784,41 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): req, **kwargs) +class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): + def __init__(self, cookiejar=None): + compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar) + + def http_response(self, request, response): + # Python 2 will choke on next HTTP request in row if there are non-ASCII + # characters in Set-Cookie HTTP header of last response (see + # https://github.com/rg3/youtube-dl/issues/6769). + # In order to at least prevent crashing we will percent encode Set-Cookie + # header before HTTPCookieProcessor starts processing it. + # if sys.version_info < (3, 0) and response.headers: + # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'): + # set_cookie = response.headers.get(set_cookie_header) + # if set_cookie: + # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ") + # if set_cookie != set_cookie_escaped: + # del response.headers[set_cookie_header] + # response.headers[set_cookie_header] = set_cookie_escaped + return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response) + + https_request = compat_urllib_request.HTTPCookieProcessor.http_request + https_response = http_response + + def parse_iso8601(date_str, delimiter='T', timezone=None): """ Return a UNIX timestamp from the given date """ if date_str is None: return None + date_str = re.sub(r'\.[0-9]+', '', date_str) + if timezone is None: m = re.search( - r'(\.[0-9]+)?(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', + r'(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', date_str) if not m: timezone = datetime.timedelta() @@ -738,9 +831,12 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): timezone = datetime.timedelta( hours=sign * int(m.group('hours')), minutes=sign * int(m.group('minutes'))) - date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) - dt = datetime.datetime.strptime(date_str, date_format) - timezone - return calendar.timegm(dt.timetuple()) + try: + date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) + dt = datetime.datetime.strptime(date_str, date_format) - timezone + return calendar.timegm(dt.timetuple()) + except ValueError: + pass def unified_strdate(date_str, day_first=True): @@ -1281,7 +1377,12 @@ def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): v = getattr(v, get_attr, None) if v == '': v = None - return default if v is None else (int(v) * invscale // scale) + if v is None: + return default + try: + return int(v) * invscale // scale + except ValueError: + return default def str_or_none(v, default=None): @@ -1297,7 +1398,12 @@ def str_to_int(int_str): def float_or_none(v, scale=1, invscale=1, default=None): - return default if v is None else (float(v) * invscale / scale) + if v is None: + return default + try: + return float(v) * invscale / scale + except ValueError: + return default def parse_duration(s): @@ -1309,10 +1415,10 @@ def parse_duration(s): m = re.match( r'''(?ix)(?:P?T)? (?: - (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*| + (?P<only_mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*| (?P<only_hours>[0-9.]+)\s*(?:hours?)| - \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*| + \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?\.?|minutes?)\s*| (?: (?: (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)? @@ -1546,6 +1652,10 @@ def urlencode_postdata(*args, **kargs): return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii') +def encode_dict(d, encoding='utf-8'): + return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items()) + + try: etree_iter = xml.etree.ElementTree.Element.iter except AttributeError: # Python <=2.6 @@ -1596,8 +1706,8 @@ def js_to_json(code): if v in ('true', 'false', 'null'): return v if v.startswith('"'): - 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: { '\\\\': '\\\\', @@ -1691,6 +1801,10 @@ def urlhandle_detect_ext(url_handle): return mimetype2ext(getheader('Content-Type')) +def encode_data_uri(data, mime_type): + return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) + + def age_restricted(content_limit, age_limit): """ Returns True iff the content should be blocked """ @@ -1886,6 +2000,32 @@ def dfxp2srt(dfxp_data): return ''.join(out) +def cli_option(params, command_option, param): + param = params.get(param) + return [command_option, param] if param is not None else [] + + +def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): + param = params.get(param) + assert isinstance(param, bool) + if separator: + return [command_option + separator + (true_value if param else false_value)] + return [command_option, true_value if param else false_value] + + +def cli_valueless_option(params, command_option, param, expected_value=True): + param = params.get(param) + return [command_option] if param == expected_value else [] + + +def cli_configuration_args(params, param, default=[]): + ex_args = params.get(param) + if ex_args is None: + return default + assert isinstance(ex_args, list) + return ex_args + + class ISO639Utils(object): # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt _lang_map = { diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 3364647ed..125e8ccf5 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2015.07.07' +__version__ = '2015.10.24' |