X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=b2c714505232e8d93836a4464533746e6b3a1b88;hb=13a75688a55f32cde316b0f7d5992ff4a1f6d279;hp=8619f38381e42f46367f8044ba3552d1b1a0f7dd;hpb=027ffdca0d3174963a4269ce8de8519cfed7a12c;p=youtube-dl diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 8619f3838..b2c714505 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -27,6 +27,7 @@ from ..compat import ( compat_str, ) from ..utils import ( + bool_or_none, clean_html, dict_get, error_to_compat_str, @@ -116,6 +117,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 'f.req': json.dumps(f_req), 'flowName': 'GlifWebSignIn', 'flowEntry': 'ServiceLogin', + # TODO: reverse actual botguard identifier generation algo + 'bgRequest': '["identifier",""]', }) return self._download_json( url, None, note=note, errnote=errnote, @@ -368,10 +371,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): (?:www\.)?hooktube\.com/| (?:www\.)?yourepeat\.com/| tube\.majestyc\.net/| + # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances (?:(?:www|dev)\.)?invidio\.us/| - (?:www\.)?invidiou\.sh/| - (?:www\.)?invidious\.snopyta\.org/| + (?:(?:www|no)\.)?invidiou\.sh/| + (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/| (?:www\.)?invidious\.kabi\.tk/| + (?:www\.)?invidious\.enkirton\.net/| + (?:www\.)?invidious\.13ad\.de/| + (?:www\.)?invidious\.mastodon\.host/| + (?:www\.)?tube\.poal\.co/| (?:www\.)?vid\.wxzm\.sx/| youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls @@ -500,6 +508,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # RTMP (unnamed) '_rtmp': {'protocol': 'rtmp'}, + + # av01 video only formats sometimes served with "unknown" codecs + '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, + '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, + '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, + '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, } _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt') @@ -1306,11 +1320,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _parse_sig_js(self, jscode): funcname = self._search_regex( - (r'(["\'])signature\1\s*,\s*(?P[a-zA-Z0-9$]+)\(', + (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P[a-zA-Z0-9$]+)\(', + r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P[a-zA-Z0-9$]+)\(', + r'(?P[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)', + # Obsolete patterns + r'(["\'])signature\1\s*,\s*(?P[a-zA-Z0-9$]+)\(', r'\.sig\|\|(?P[a-zA-Z0-9$]+)\(', - r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P[a-zA-Z0-9$]+)\(', - r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P[a-zA-Z0-9$]+)\(', - r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P[a-zA-Z0-9$]+)\('), + r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P[a-zA-Z0-9$]+)\(', + r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P[a-zA-Z0-9$]+)\(', + r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P[a-zA-Z0-9$]+)\(', + r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P[a-zA-Z0-9$]+)\(', + r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P[a-zA-Z0-9$]+)\(', + r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P[a-zA-Z0-9$]+)\('), jscode, 'Initial JS player signature function name', group='sig') jsi = JSInterpreter(jscode) @@ -1575,8 +1596,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return video_id def _extract_annotations(self, video_id): - url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id - return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') + return self._download_webpage( + 'https://www.youtube.com/annotations_invideo', video_id, + note='Downloading annotations', + errnote='Unable to download video annotations', fatal=False, + query={ + 'features': 1, + 'legacy': 1, + 'video_id': video_id, + }) @staticmethod def _extract_chapters(description, duration): @@ -1789,9 +1817,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): raise ExtractorError( 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id) - if video_info.get('license_info'): - raise ExtractorError('This video is DRM protected.', expected=True) - video_details = try_get( player_response, lambda x: x['videoDetails'], dict) or {} @@ -1871,6 +1896,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if view_count is None and video_details: view_count = int_or_none(video_details.get('viewCount')) + if is_live is None: + is_live = bool_or_none(video_details.get('isLive')) + # Check for "rental" videos if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True) @@ -1927,7 +1955,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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: + if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'): continue stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0])) # Unsupported FORMAT_STREAM_TYPE_OTF @@ -2052,8 +2080,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): url_or_none(try_get( player_response, lambda x: x['streamingData']['hlsManifestUrl'], - compat_str)) or - url_or_none(try_get( + compat_str)) + or url_or_none(try_get( video_info, lambda x: x['hlsvp'][0], compat_str))) if manifest_url: formats = [] @@ -2102,10 +2130,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self._downloader.report_warning('unable to extract uploader nickname') channel_id = ( - str_or_none(video_details.get('channelId')) or - self._html_search_meta( - 'channelId', video_webpage, 'channel id', default=None) or - self._search_regex( + str_or_none(video_details.get('channelId')) + or self._html_search_meta( + 'channelId', video_webpage, 'channel id', default=None) + or self._search_regex( r'data-channel-external-id=(["\'])(?P(?:(?!\1).)+)\1', video_webpage, 'channel id', default=None, group='id')) channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None @@ -2227,6 +2255,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage, 'view count', default=None)) + average_rating = ( + float_or_none(video_details.get('averageRating')) + or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0]))) + # subtitles video_subtitles = self.extract_subtitles(video_id, video_webpage) automatic_captions = self.extract_automatic_captions(video_id, video_webpage) @@ -2323,6 +2355,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '"token" parameter not in video info for unknown reason', video_id=video_id) + if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])): + raise ExtractorError('This video is DRM protected.', expected=True) + self._sort_formats(formats) self.mark_watched(video_id, video_info, player_response) @@ -2353,7 +2388,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'view_count': view_count, 'like_count': like_count, 'dislike_count': dislike_count, - 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]), + 'average_rating': average_rating, 'formats': formats, 'is_live': is_live, 'start_time': start_time, @@ -2397,7 +2432,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): (%(playlist_id)s) )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE} _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' - _VIDEO_RE = r'href="\s*/watch\?v=(?P[0-9A-Za-z_-]{11})&[^"]*?index=(?P\d+)(?:[^>]+>(?P[^<]+))?' + _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', @@ -2420,6 +2455,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'title': '29C3: Not my department', 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', + 'uploader': 'Christiaan008', + 'uploader_id': 'ChRiStIaAn008', }, 'playlist_count': 95, }, { @@ -2428,6 +2465,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'title': '[OLD]Team Fortress 2 (Class-based LP)', 'id': 'PLBB231211A4F62143', + 'uploader': 'Wickydoo', + 'uploader_id': 'Wickydoo', }, 'playlist_mincount': 26, }, { @@ -2436,6 +2475,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'title': 'Uploads from Cauchemar', 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q', + 'uploader': 'Cauchemar', + 'uploader_id': 'Cauchemar89', }, 'playlist_mincount': 799, }, { @@ -2453,13 +2494,17 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'title': 'JODA15', 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + 'uploader': 'milan', + 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw', } }, { 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', 'playlist_mincount': 485, 'info_dict': { - 'title': '2017 華語最新單曲 (2/24更新)', + 'title': '2018 Chinese New Singles (11/6 updated)', 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', + 'uploader': 'LBK', + 'uploader_id': 'sdragonfang', } }, { 'note': 'Embedded SWF player', @@ -2468,13 +2513,16 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'title': 'JODA7', 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ', - } + }, + 'skip': 'This playlist does not exist', }, { 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos', 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA', 'info_dict': { 'title': 'Uploads from Interstellar Movie', 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', + 'uploader': 'Interstellar Movie', + 'uploader_id': 'InterstellarMovie1', }, 'playlist_mincount': 21, }, { @@ -2499,6 +2547,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'This video is not available.', 'add_ie': [YoutubeIE.ie_key()], }, { 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5', @@ -2510,7 +2559,6 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'uploader_id': 'backuspagemuseum', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum', 'upload_date': '20161008', - 'license': 'Standard YouTube License', 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a', 'categories': ['Nonprofits & Activism'], 'tags': list, @@ -2521,6 +2569,16 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'noplaylist': True, 'skip_download': True, }, + }, { + # https://github.com/ytdl-org/youtube-dl/issues/21844 + 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba', + 'info_dict': { + 'title': 'Data Analysis with Dr Mike Pound', + 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba', + 'uploader_id': 'Computerphile', + 'uploader': 'Computerphile', + }, + 'playlist_mincount': 11, }, { 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21', 'only_matching': True, @@ -2564,9 +2622,9 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) title_span = ( - search_title('playlist-title') or - search_title('title long-title') or - search_title('title')) + search_title('playlist-title') + or search_title('title long-title') + or search_title('title')) title = clean_html(title_span) return self.playlist_result(url_results, playlist_id, title) @@ -2687,6 +2745,8 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w', 'title': 'Uploads from lex will', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', } }, { 'note': 'Age restricted channel', @@ -2696,6 +2756,8 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): 'info_dict': { 'id': 'UUs0ifCMCm1icqRbqhUINa0w', 'title': 'Uploads from Deus Ex', + 'uploader': 'Deus Ex', + 'uploader_id': 'DeusExOfficial', }, }, { 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA', @@ -2780,6 +2842,8 @@ class YoutubeUserIE(YoutubeChannelIE): 'info_dict': { 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ', 'title': 'Uploads from The Linux Foundation', + 'uploader': 'The Linux Foundation', + 'uploader_id': 'TheLinuxFoundation', } }, { # Only available via https://www.youtube.com/c/12minuteathlete/videos @@ -2789,6 +2853,8 @@ class YoutubeUserIE(YoutubeChannelIE): 'info_dict': { 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ', 'title': 'Uploads from 12 Minute Athlete', + 'uploader': '12 Minute Athlete', + 'uploader_id': 'the12minuteathlete', } }, { 'url': 'ytuser:phihag', @@ -2882,7 +2948,7 @@ class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor): 'playlist_mincount': 4, 'info_dict': { 'id': 'ThirstForScience', - 'title': 'Thirst for Science', + 'title': 'ThirstForScience', }, }, { # with "Load more" button @@ -2899,6 +2965,7 @@ class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor): 'id': 'UCiU1dHvZObB2iP6xkJ__Icw', 'title': 'Chem Player', }, + 'skip': 'Blocked', }]