yt_dlp/extractor/odnoklassniki.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_etree_fromstring,
   6     compat_parse_qs,
   7     compat_urllib_parse_unquote,
   8     compat_urllib_parse_urlparse,
   9 )
  10 from ..networking import HEADRequest
  11 from ..utils import (
  12     ExtractorError,
  13     float_or_none,
  14     int_or_none,
  15     qualities,
  16     smuggle_url,
  17     traverse_obj,
  18     unescapeHTML,
  19     unified_strdate,
  20     unsmuggle_url,
  21     url_or_none,
  22     urlencode_postdata,
  23 )
  24
  25
  26 class OdnoklassnikiIE(InfoExtractor):
  27     _VALID_URL = r'''(?x)
  28                 https?://
  29                     (?:(?:www|m|mobile)\.)?
  30                     (?:odnoklassniki|ok)\.ru/
  31                     (?:
  32                         video(?P<embed>embed)?/|
  33                         web-api/video/moviePlayer/|
  34                         live/|
  35                         dk\?.*?st\.mvId=
  36                     )
  37                     (?P<id>[\d-]+)
  38                 '''
  39     _EMBED_REGEX = [r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1']
  40     _TESTS = [{
  41         'note': 'Coub embedded',
  42         'url': 'http://ok.ru/video/1484130554189',
  43         'info_dict': {
  44             'id': '1keok9',
  45             'ext': 'mp4',
  46             'timestamp': 1545580896,
  47             'view_count': int,
  48             'thumbnail': r're:^https?://.*\.jpg$',
  49             'title': 'Народная забава',
  50             'uploader': 'Nevata',
  51             'upload_date': '20181223',
  52             'age_limit': 0,
  53             'uploader_id': 'nevata.s',
  54             'like_count': int,
  55             'duration': 8.08,
  56             'repost_count': int,
  57         },
  58     }, {
  59         'note': 'vk.com embedded',
  60         'url': 'https://ok.ru/video/3568183087575',
  61         'info_dict': {
  62             'id': '-165101755_456243749',
  63             'ext': 'mp4',
  64             'uploader_id': '-165101755',
  65             'duration': 132,
  66             'timestamp': 1642869935,
  67             'upload_date': '20220122',
  68             'thumbnail': str,
  69             'title': str,
  70             'uploader': str,
  71         },
  72         'skip': 'vk extractor error',
  73     }, {
  74         # metadata in JSON, webm_dash with Firefox UA
  75         'url': 'http://ok.ru/video/20079905452',
  76         'md5': '8f477d8931c531374a3e36daec617b2c',
  77         'info_dict': {
  78             'id': '20079905452',
  79             'ext': 'webm',
  80             'title': 'Культура меняет нас (прекрасный ролик!))',
  81             'thumbnail': str,
  82             'duration': 100,
  83             'upload_date': '20141207',
  84             'uploader_id': '330537914540',
  85             'uploader': 'Виталий Добровольский',
  86             'like_count': int,
  87             'age_limit': 0,
  88         },
  89         'params': {
  90             'format': 'bv[ext=webm]',
  91             'http_headers': {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; rv:102.0) Gecko/20100101 Firefox/102.0'},
  92         },
  93     }, {
  94         # metadataUrl
  95         'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
  96         'md5': '2bae2f58eefe1b3d26f3926c4a64d2f3',
  97         'info_dict': {
  98             'id': '63567059965189-0',
  99             'ext': 'mp4',
 100             'title': 'Девушка без комплексов ...',
 101             'thumbnail': str,
 102             'duration': 191,
 103             'upload_date': '20150518',
 104             'uploader_id': '534380003155',
 105             'uploader': '☭ Андрей Мещанинов ☭',
 106             'like_count': int,
 107             'age_limit': 0,
 108             'start_time': 5,
 109         },
 110         'params': {'skip_download': 'm3u8'},
 111     }, {
 112         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
 113         'url': 'https://ok.ru/video/3952212382174',
 114         'md5': '5fb5f83ce16cb212d6bf887282b5da53',
 115         'info_dict': {
 116             'id': '5axVgHHDBvU',
 117             'ext': 'mp4',
 118             'title': 'Youtube-dl 101: What is it and HOW to use it! Full Download Walkthrough and Guide',
 119             'description': 'md5:b57209eeb9d5c2f20c984dfb58862097',
 120             'uploader': 'Lod Mer',
 121             'uploader_id': '575186401502',
 122             'duration': 1529,
 123             'age_limit': 0,
 124             'upload_date': '20210405',
 125             'comment_count': int,
 126             'live_status': 'not_live',
 127             'view_count': int,
 128             'thumbnail': 'https://i.mycdn.me/i?r=AEHujHvw2RjEbemUCNEorZbxYpb_p_9AcN2FmGik64Krkcmz37YtlY093oAM5-HIEAt7Zi9s0CiBOSDmbngC-I-k&fn=external_8',
 129             'uploader_url': 'https://www.youtube.com/@MrKewlkid94',
 130             'channel_follower_count': int,
 131             'tags': ['youtube-dl', 'youtube playlists', 'download videos', 'download audio'],
 132             'channel_id': 'UCVGtvURtEURYHtJFUegdSug',
 133             'like_count': int,
 134             'availability': 'public',
 135             'channel_url': 'https://www.youtube.com/channel/UCVGtvURtEURYHtJFUegdSug',
 136             'categories': ['Education'],
 137             'playable_in_embed': True,
 138             'channel': 'BornToReact',
 139         },
 140     }, {
 141         # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
 142         'url': 'http://ok.ru/video/62036049272859-0',
 143         'info_dict': {
 144             'id': '62036049272859-0',
 145             'ext': 'mp4',
 146             'title': 'МУЗЫКА     ДОЖДЯ .',
 147             'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
 148             'upload_date': '20120106',
 149             'uploader_id': '473534735899',
 150             'uploader': 'МARINA D',
 151             'age_limit': 0,
 152         },
 153         'params': {
 154             'skip_download': True,
 155         },
 156         'skip': 'Video has not been found',
 157     }, {
 158         'note': 'Only available in mobile webpage',
 159         'url': 'https://m.ok.ru/video/2361249957145',
 160         'info_dict': {
 161             'id': '2361249957145',
 162             'ext': 'mp4',
 163             'title': 'Быковское крещение',
 164             'duration': 3038.181,
 165             'thumbnail': r're:^https?://i\.mycdn\.me/videoPreview\?.+',
 166         },
 167     }, {
 168         'note': 'subtitles',
 169         'url': 'https://ok.ru/video/4249587550747',
 170         'info_dict': {
 171             'id': '4249587550747',
 172             'ext': 'mp4',
 173             'title': 'Small Country An African Childhood (2020) (1080p) +subtitle',
 174             'uploader': 'Sunflower Movies',
 175             'uploader_id': '595802161179',
 176             'upload_date': '20220816',
 177             'duration': 6728,
 178             'age_limit': 0,
 179             'thumbnail': r're:^https?://i\.mycdn\.me/videoPreview\?.+',
 180             'like_count': int,
 181             'subtitles': dict,
 182         },
 183         'params': {
 184             'skip_download': True,
 185         },
 186     }, {
 187         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
 188         'only_matching': True,
 189     }, {
 190         'url': 'http://www.ok.ru/video/20648036891',
 191         'only_matching': True,
 192     }, {
 193         'url': 'http://www.ok.ru/videoembed/20648036891',
 194         'only_matching': True,
 195     }, {
 196         'url': 'http://m.ok.ru/video/20079905452',
 197         'only_matching': True,
 198     }, {
 199         'url': 'http://mobile.ok.ru/video/20079905452',
 200         'only_matching': True,
 201     }, {
 202         'url': 'https://www.ok.ru/live/484531969818',
 203         'only_matching': True,
 204     }, {
 205         'url': 'https://m.ok.ru/dk?st.cmd=movieLayer&st.discId=863789452017&st.retLoc=friend&st.rtu=%2Fdk%3Fst.cmd%3DfriendMovies%26st.mode%3Down%26st.mrkId%3D%257B%2522uploadedMovieMarker%2522%253A%257B%2522marker%2522%253A%25221519410114503%2522%252C%2522hasMore%2522%253Atrue%257D%252C%2522sharedMovieMarker%2522%253A%257B%2522marker%2522%253Anull%252C%2522hasMore%2522%253Afalse%257D%257D%26st.friendId%3D561722190321%26st.frwd%3Don%26_prevCmd%3DfriendMovies%26tkn%3D7257&st.discType=MOVIE&st.mvId=863789452017&_prevCmd=friendMovies&tkn=3648#lst#',
 206         'only_matching': True,
 207     }, {
 208         # Paid video
 209         'url': 'https://ok.ru/video/954886983203',
 210         'only_matching': True,
 211     }, {
 212         'url': 'https://ok.ru/videoembed/2932705602075',
 213         'info_dict': {
 214             'id': '2932705602075',
 215             'ext': 'mp4',
 216             'thumbnail': 'https://i.mycdn.me/videoPreview?id=1369902483995&type=37&idx=2&tkn=fqlnoQD_xwq5ovIlKfgNyU08qmM&fn=external_8',
 217             'title': 'Boosty для тебя!',
 218             'uploader_id': '597811038747',
 219             'like_count': 0,
 220             'duration': 35,
 221         },
 222     }]
 223
 224     _WEBPAGE_TESTS = [{
 225         'url': 'https://boosty.to/ikakprosto/posts/56cedaca-b56a-4dfd-b3ed-98c79cfa0167',
 226         'info_dict': {
 227             'id': '3950343629563',
 228             'ext': 'mp4',
 229             'thumbnail': 'https://i.mycdn.me/videoPreview?id=2776238394107&type=37&idx=11&tkn=F3ejkUFcpuI4DnMRxrDGcH5YcmM&fn=external_8',
 230             'title': 'Заяц Бусти.mp4',
 231             'uploader_id': '571368965883',
 232             'like_count': 0,
 233             'duration': 10444,
 234         },
 235         'skip': 'Site no longer embeds',
 236     }]
 237
 238     def _clear_cookies(self, cdn_url):
 239         # Direct http downloads will fail if CDN cookies are set
 240         # so we need to reset them after each format extraction
 241         self.cookiejar.clear(domain='.mycdn.me')
 242         self.cookiejar.clear(domain=urllib.parse.urlparse(cdn_url).hostname)
 243
 244     @classmethod
 245     def _extract_embed_urls(cls, url, webpage):
 246         for x in super()._extract_embed_urls(url, webpage):
 247             yield smuggle_url(x, {'referrer': url})
 248
 249     def _real_extract(self, url):
 250         try:
 251             return self._extract_desktop(url)
 252         except ExtractorError as e:
 253             try:
 254                 return self._extract_mobile(url)
 255             except ExtractorError:
 256                 # error message of desktop webpage is in English
 257                 raise e
 258
 259     def _extract_desktop(self, url):
 260         start_time = int_or_none(compat_parse_qs(
 261             compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
 262
 263         url, smuggled = unsmuggle_url(url, {})
 264         video_id, is_embed = self._match_valid_url(url).group('id', 'embed')
 265         mode = 'videoembed' if is_embed else 'video'
 266
 267         webpage = self._download_webpage(
 268             f'https://ok.ru/{mode}/{video_id}', video_id,
 269             note='Downloading desktop webpage',
 270             headers={'Referer': smuggled['referrer']} if smuggled.get('referrer') else {})
 271
 272         error = self._search_regex(
 273             r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<',
 274             webpage, 'error', default=None)
 275         # Direct link from boosty
 276         if (error == 'The author of this video has not been found or is blocked'
 277                 and not smuggled.get('referrer') and mode == 'videoembed'):
 278             return self._extract_desktop(smuggle_url(url, {'referrer': 'https://boosty.to'}))
 279         elif error:
 280             raise ExtractorError(error, expected=True)
 281
 282         player = self._parse_json(
 283             unescapeHTML(self._search_regex(
 284                 r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
 285                 webpage, 'player', group='player')),
 286             video_id)
 287
 288         # embedded external player
 289         if player.get('isExternalPlayer') and player.get('url'):
 290             return self.url_result(player['url'])
 291
 292         flashvars = player['flashvars']
 293
 294         metadata = flashvars.get('metadata')
 295         if metadata:
 296             metadata = self._parse_json(metadata, video_id)
 297         else:
 298             data = {}
 299             st_location = flashvars.get('location')
 300             if st_location:
 301                 data['st.location'] = st_location
 302             metadata = self._download_json(
 303                 compat_urllib_parse_unquote(flashvars['metadataUrl']),
 304                 video_id, 'Downloading metadata JSON',
 305                 data=urlencode_postdata(data))
 306
 307         movie = metadata['movie']
 308
 309         # Some embedded videos may not contain title in movie dict (e.g.
 310         # http://ok.ru/video/62036049272859-0) thus we allow missing title
 311         # here and it's going to be extracted later by an extractor that
 312         # will process the actual embed.
 313         provider = metadata.get('provider')
 314         title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
 315
 316         thumbnail = movie.get('poster')
 317         duration = int_or_none(movie.get('duration'))
 318
 319         author = metadata.get('author', {})
 320         uploader_id = author.get('id')
 321         uploader = author.get('name')
 322
 323         upload_date = unified_strdate(self._html_search_meta(
 324             'ya:ovs:upload_date', webpage, 'upload date', default=None))
 325
 326         age_limit = None
 327         adult = self._html_search_meta(
 328             'ya:ovs:adult', webpage, 'age limit', default=None)
 329         if adult:
 330             age_limit = 18 if adult == 'true' else 0
 331
 332         like_count = int_or_none(metadata.get('likeCount'))
 333
 334         subtitles = {}
 335         for sub in traverse_obj(metadata, ('movie', 'subtitleTracks', ...), expected_type=dict):
 336             sub_url = sub.get('url')
 337             if not sub_url:
 338                 continue
 339             subtitles.setdefault(sub.get('language') or 'en', []).append({
 340                 'url': sub_url,
 341                 'ext': 'vtt',
 342             })
 343
 344         info = {
 345             'id': video_id,
 346             'title': title,
 347             'thumbnail': thumbnail,
 348             'duration': duration,
 349             'upload_date': upload_date,
 350             'uploader': uploader,
 351             'uploader_id': uploader_id,
 352             'like_count': like_count,
 353             'age_limit': age_limit,
 354             'start_time': start_time,
 355             'subtitles': subtitles,
 356         }
 357
 358         # pladform
 359         if provider == 'OPEN_GRAPH':
 360             info.update({
 361                 '_type': 'url_transparent',
 362                 'url': movie['contentId'],
 363             })
 364             return info
 365
 366         if provider == 'USER_YOUTUBE':
 367             info.update({
 368                 '_type': 'url_transparent',
 369                 'url': movie['contentId'],
 370             })
 371             return info
 372
 373         assert title
 374         if provider == 'LIVE_TV_APP':
 375             info['title'] = title
 376
 377         quality = qualities(('4', '0', '1', '2', '3', '5', '6', '7'))
 378
 379         formats = [{
 380             'url': f['url'],
 381             'ext': 'mp4',
 382             'format_id': f.get('name'),
 383         } for f in traverse_obj(metadata, ('videos', lambda _, v: url_or_none(v['url'])))]
 384
 385         m3u8_url = traverse_obj(metadata, 'hlsManifestUrl', 'ondemandHls')
 386         if m3u8_url:
 387             formats.extend(self._extract_m3u8_formats(
 388                 m3u8_url, video_id, 'mp4', 'm3u8_native',
 389                 m3u8_id='hls', fatal=False))
 390             self._clear_cookies(m3u8_url)
 391
 392         for mpd_id, mpd_key in [('dash', 'ondemandDash'), ('webm', 'metadataWebmUrl')]:
 393             mpd_url = metadata.get(mpd_key)
 394             if mpd_url:
 395                 formats.extend(self._extract_mpd_formats(
 396                     mpd_url, video_id, mpd_id=mpd_id, fatal=False))
 397                 self._clear_cookies(mpd_url)
 398
 399         dash_manifest = metadata.get('metadataEmbedded')
 400         if dash_manifest:
 401             formats.extend(self._parse_mpd_formats(
 402                 compat_etree_fromstring(dash_manifest), 'mpd'))
 403
 404         for fmt in formats:
 405             fmt_type = self._search_regex(
 406                 r'\btype[/=](\d)', fmt['url'],
 407                 'format type', default=None)
 408             if fmt_type:
 409                 fmt['quality'] = quality(fmt_type)
 410
 411         # Live formats
 412         m3u8_url = metadata.get('hlsMasterPlaylistUrl')
 413         if m3u8_url:
 414             formats.extend(self._extract_m3u8_formats(
 415                 m3u8_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 416             self._clear_cookies(m3u8_url)
 417         rtmp_url = metadata.get('rtmpUrl')
 418         if rtmp_url:
 419             formats.append({
 420                 'url': rtmp_url,
 421                 'format_id': 'rtmp',
 422                 'ext': 'flv',
 423             })
 424
 425         if not formats:
 426             payment_info = metadata.get('paymentInfo')
 427             if payment_info:
 428                 self.raise_no_formats('This video is paid, subscribe to download it', expected=True)
 429
 430         info['formats'] = formats
 431         return info
 432
 433     def _extract_mobile(self, url):
 434         video_id = self._match_id(url)
 435
 436         webpage = self._download_webpage(
 437             'http://m.ok.ru/video/%s' % video_id, video_id,
 438             note='Downloading mobile webpage')
 439
 440         error = self._search_regex(
 441             r'видео</a>\s*<div\s+class="empty">(.+?)</div>',
 442             webpage, 'error', default=None)
 443         if error:
 444             raise ExtractorError(error, expected=True)
 445
 446         json_data = self._search_regex(
 447             r'data-video="(.+?)"', webpage, 'json data')
 448         json_data = self._parse_json(unescapeHTML(json_data), video_id) or {}
 449
 450         redirect_url = self._request_webpage(HEADRequest(
 451             json_data['videoSrc']), video_id, 'Requesting download URL').url
 452         self._clear_cookies(redirect_url)
 453
 454         return {
 455             'id': video_id,
 456             'title': json_data.get('videoName'),
 457             'duration': float_or_none(json_data.get('videoDuration'), scale=1000),
 458             'thumbnail': json_data.get('videoPosterSrc'),
 459             'formats': [{
 460                 'format_id': 'mobile',
 461                 'url': redirect_url,
 462                 'ext': 'mp4',
 463             }]
 464         }