X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=91b9d59c68a55ec01a2ebeb7491601815abd56c1;hb=dc879c5a37dae588a5bb35d416635678356ad1b7;hp=25d056b3c21ea3f98c5c36384e86d243dfd4a913;hpb=494d664e679c5b0f85e3c899579e7eb8a1cc8246;p=youtube-dl diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 25d056b3c..91b9d59c6 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -41,7 +41,6 @@ from ..utils import ( orderedSet, parse_codecs, parse_duration, - qualities, remove_quotes, remove_start, smuggle_url, @@ -70,7 +69,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False - _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}' + _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}' def _set_language(self): self._set_cookie( @@ -373,7 +372,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): _VALID_URL = r"""(?x)^ ( (?:https?://|//) # http(s):// or protocol-independent URL - (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| + (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/| (?:www\.)?deturl\.com/www\.youtube\.com/| (?:www\.)?pwnyoutube\.com/| (?:www\.)?hooktube\.com/| @@ -384,13 +383,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): (?:(?: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\.)?invidious\.nixnet\.xyz/| + (?:www\.)?invidious\.drycat\.fr/| (?:www\.)?tube\.poal\.co/| (?:www\.)?vid\.wxzm\.sx/| (?:www\.)?yt\.elukerio\.org/| + (?:www\.)?yt\.lelux\.fi/| + (?:www\.)?kgg2m7yk5aybusll\.onion/| + (?:www\.)?qklhadlycap4cnod\.onion/| + (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/| + (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/| + (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/| + (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/| + (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/| youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls (?: # the various things that can precede the ID: @@ -1217,6 +1224,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'skip_download': True, }, }, + { + 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q', + 'only_matching': True, + }, ] def __init__(self, *args, **kwargs): @@ -1245,7 +1256,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _extract_signature_function(self, video_id, player_url, example_sig): id_m = re.match( - r'.*?-(?P[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P[a-z]+)$', + r'.*?[-.](?P[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P[a-z]+)$', player_url) if not id_m: raise ExtractorError('Cannot identify player %r' % player_url) @@ -1332,6 +1343,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): funcname = self._search_regex( (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'\b(?P[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)', 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$]+)\(', @@ -1909,6 +1921,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return int_or_none(self._search_regex( r'\bclen[=/](\d+)', media_url, 'filesize', default=None)) + streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or [] + streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or []) + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): self.report_rtmp_download() formats = [{ @@ -1917,10 +1932,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'url': video_info['conn'][0], 'player_url': player_url, }] - elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1): + elif not is_live and (streaming_formats or len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1): 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/ytdl-org/youtube-dl/issues/343 for more information.', expected=True) + formats = [] formats_spec = {} fmt_list = video_info.get('fmt_list', [''])[0] if fmt_list: @@ -1934,91 +1950,104 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'width': int_or_none(width_height[0]), 'height': int_or_none(width_height[1]), } - q = qualities(['small', 'medium', 'hd720']) - streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) - if streaming_formats: - for fmt in streaming_formats: - itag = str_or_none(fmt.get('itag')) - if not itag: - continue - quality = fmt.get('quality') - quality_label = fmt.get('qualityLabel') or quality - formats_spec[itag] = { - 'asr': int_or_none(fmt.get('audioSampleRate')), - 'filesize': int_or_none(fmt.get('contentLength')), - 'format_note': quality_label, - 'fps': int_or_none(fmt.get('fps')), - 'height': int_or_none(fmt.get('height')), - 'quality': q(quality), - # bitrate for itag 43 is always 2147483647 - 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None, - 'width': int_or_none(fmt.get('width')), - } - 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 or url_data.get('drm_families'): + for fmt in streaming_formats: + itag = str_or_none(fmt.get('itag')) + if not itag: + continue + quality = fmt.get('quality') + quality_label = fmt.get('qualityLabel') or quality + formats_spec[itag] = { + 'asr': int_or_none(fmt.get('audioSampleRate')), + 'filesize': int_or_none(fmt.get('contentLength')), + 'format_note': quality_label, + 'fps': int_or_none(fmt.get('fps')), + 'height': int_or_none(fmt.get('height')), + # bitrate for itag 43 is always 2147483647 + 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None, + 'width': int_or_none(fmt.get('width')), + } + + for fmt in streaming_formats: + if fmt.get('drm_families'): continue + url = url_or_none(fmt.get('url')) + + if not url: + cipher = fmt.get('cipher') + if not cipher: + continue + url_data = compat_parse_qs(cipher) + url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str)) + if not url: + continue + else: + cipher = None + url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0])) # Unsupported FORMAT_STREAM_TYPE_OTF if stream_type == 3: continue - format_id = url_data['itag'][0] - url = url_data['url'][0] - - if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True): - ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")' - jsplayer_url_json = self._search_regex( - ASSETS_RE, - embed_webpage if age_gate else video_webpage, - 'JS player URL (1)', default=None) - if not jsplayer_url_json and not age_gate: - # We need the embed website after all - if embed_webpage is None: - embed_url = proto + '://www.youtube.com/embed/%s' % video_id - embed_webpage = self._download_webpage( - embed_url, video_id, 'Downloading embed webpage') - jsplayer_url_json = self._search_regex( - ASSETS_RE, embed_webpage, 'JS player URL') - - player_url = json.loads(jsplayer_url_json) - if player_url is None: - player_url_json = self._search_regex( - r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', - video_webpage, 'age gate player URL') - player_url = json.loads(player_url_json) - if 'sig' in url_data: - url += '&signature=' + url_data['sig'][0] - elif 's' in url_data: - encrypted_sig = url_data['s'][0] + format_id = fmt.get('itag') or url_data['itag'][0] + if not format_id: + continue + format_id = compat_str(format_id) - if self._downloader.params.get('verbose'): + if cipher: + if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True): + ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")' + jsplayer_url_json = self._search_regex( + ASSETS_RE, + embed_webpage if age_gate else video_webpage, + 'JS player URL (1)', default=None) + if not jsplayer_url_json and not age_gate: + # We need the embed website after all + if embed_webpage is None: + embed_url = proto + '://www.youtube.com/embed/%s' % video_id + embed_webpage = self._download_webpage( + embed_url, video_id, 'Downloading embed webpage') + jsplayer_url_json = self._search_regex( + ASSETS_RE, embed_webpage, 'JS player URL') + + player_url = json.loads(jsplayer_url_json) if player_url is None: - player_version = 'unknown' - player_desc = 'unknown' - else: - if player_url.endswith('swf'): - player_version = self._search_regex( - r'-(.+?)(?:/watch_as3)?\.swf$', player_url, - 'flash player', fatal=False) - player_desc = 'flash player %s' % player_version + player_url_json = self._search_regex( + r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', + video_webpage, 'age gate player URL') + player_url = json.loads(player_url_json) + + if 'sig' in url_data: + url += '&signature=' + url_data['sig'][0] + elif 's' in url_data: + encrypted_sig = url_data['s'][0] + + if self._downloader.params.get('verbose'): + if player_url is None: + player_version = 'unknown' + player_desc = 'unknown' else: - player_version = self._search_regex( - [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', - r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'], - player_url, - 'html5 player', fatal=False) - player_desc = 'html5 player %s' % player_version - - parts_sizes = self._signature_cache_id(encrypted_sig) - self.to_screen('{%s} signature length %s, %s' % - (format_id, parts_sizes, player_desc)) - - signature = self._decrypt_signature( - encrypted_sig, video_id, player_url, age_gate) - sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature' - url += '&%s=%s' % (sp, signature) + if player_url.endswith('swf'): + player_version = self._search_regex( + r'-(.+?)(?:/watch_as3)?\.swf$', player_url, + 'flash player', fatal=False) + player_desc = 'flash player %s' % player_version + else: + player_version = self._search_regex( + [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', + r'(?:www|player(?:_ias)?)[-.]([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'], + player_url, + 'html5 player', fatal=False) + player_desc = 'html5 player %s' % player_version + + parts_sizes = self._signature_cache_id(encrypted_sig) + self.to_screen('{%s} signature length %s, %s' % + (format_id, parts_sizes, player_desc)) + + signature = self._decrypt_signature( + encrypted_sig, video_id, player_url, age_gate) + sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature' + url += '&%s=%s' % (sp, signature) if 'ratebypass' not in url: url += '&ratebypass=yes' @@ -2038,24 +2067,33 @@ class YoutubeIE(YoutubeBaseInfoExtractor): mobj = re.search(r'^(?P\d+)[xX](?P\d+)$', url_data.get('size', [''])[0]) width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) + if width is None: + width = int_or_none(fmt.get('width')) + if height is None: + height = int_or_none(fmt.get('height')) + filesize = int_or_none(url_data.get( 'clen', [None])[0]) or _extract_filesize(url) - quality = url_data.get('quality', [None])[0] + quality = url_data.get('quality', [None])[0] or fmt.get('quality') + quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel') + + tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000) + or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None + fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps')) more_fields = { 'filesize': filesize, - 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000), + 'tbr': tbr, 'width': width, 'height': height, - 'fps': int_or_none(url_data.get('fps', [None])[0]), - 'format_note': url_data.get('quality_label', [None])[0] or quality, - 'quality': q(quality), + 'fps': fps, + 'format_note': quality_label or quality, } for key, value in more_fields.items(): if value: dct[key] = value - type_ = url_data.get('type', [None])[0] + type_ = url_data.get('type', [None])[0] or fmt.get('mimeType') if type_: type_split = type_.split(';') kind_ext = type_split[0].split('/') @@ -2432,7 +2470,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): (?:\w+\.)? (?: (?: - youtube\.com| + youtube(?:kids)?\.com| invidio\.us ) / @@ -2444,7 +2482,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist= ) ( - (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,} + (?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)?[0-9A-Za-z-_]{10,} # Top tracks, they can also include dots |(?:MC)[\w\.]* ) @@ -2457,20 +2495,23 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): _VIDEO_RE = _VIDEO_RE_TPL % r'(?P[0-9A-Za-z_-]{11})' IE_NAME = 'youtube:playlist' _TESTS = [{ - 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', + 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', 'info_dict': { - 'title': 'ytdl test PL', - 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', + 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA', + 'uploader': 'Sergey M.', + 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', + 'title': 'youtube-dl public playlist', }, - 'playlist_count': 3, + 'playlist_count': 1, }, { - 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', + 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf', 'info_dict': { - 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', - 'title': 'YDL_Empty_List', + 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA', + 'uploader': 'Sergey M.', + 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf', + 'title': 'youtube-dl empty playlist', }, 'playlist_count': 0, - 'skip': 'This playlist is private', }, { 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', @@ -2480,7 +2521,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'uploader': 'Christiaan008', 'uploader_id': 'ChRiStIaAn008', }, - 'playlist_count': 95, + 'playlist_count': 96, }, { 'note': 'issue #673', 'url': 'PLBB231211A4F62143', @@ -2614,6 +2655,9 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): }, { 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU', 'only_matching': True, + }, { + 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g', + 'only_matching': True, }] def _real_initialize(self): @@ -2709,7 +2753,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): page, 'title', default=None) _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*
  • \s*]+\bhref=' - uploader = self._search_regex( + uploader = self._html_search_regex( r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE, page, 'uploader', default=None) mobj = re.search( @@ -2784,7 +2828,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com channels' - _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P[0-9A-Za-z_-]+)' + _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie|kids)?\.com|(?:www\.)?invidio\.us)/channel/(?P[0-9A-Za-z_-]+)' _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' _VIDEO_RE = r'(?:title="(?P[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?' IE_NAME = 'youtube:channel' @@ -2812,6 +2856,9 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): }, { 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA', 'only_matching': True, + }, { + 'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA', + 'only_matching': True, }] @classmethod