yt_dlp/extractor/svt.py

   1 import json
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     determine_ext,
   8     dict_get,
   9     int_or_none,
  10     traverse_obj,
  11     try_get,
  12     unified_timestamp,
  13 )
  14
  15
  16 class SVTBaseIE(InfoExtractor):
  17     _GEO_COUNTRIES = ['SE']
  18
  19     def _extract_video(self, video_info, video_id):
  20         is_live = dict_get(video_info, ('live', 'simulcast'), default=False)
  21         m3u8_protocol = 'm3u8' if is_live else 'm3u8_native'
  22         formats = []
  23         subtitles = {}
  24         for vr in video_info['videoReferences']:
  25             player_type = vr.get('playerType') or vr.get('format')
  26             vurl = vr['url']
  27             ext = determine_ext(vurl)
  28             if ext == 'm3u8':
  29                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  30                     vurl, video_id,
  31                     ext='mp4', entry_protocol=m3u8_protocol,
  32                     m3u8_id=player_type, fatal=False)
  33                 formats.extend(fmts)
  34                 self._merge_subtitles(subs, target=subtitles)
  35             elif ext == 'f4m':
  36                 formats.extend(self._extract_f4m_formats(
  37                     vurl + '?hdcore=3.3.0', video_id,
  38                     f4m_id=player_type, fatal=False))
  39             elif ext == 'mpd':
  40                 fmts, subs = self._extract_mpd_formats_and_subtitles(
  41                     vurl, video_id, mpd_id=player_type, fatal=False)
  42                 formats.extend(fmts)
  43                 self._merge_subtitles(subs, target=subtitles)
  44             else:
  45                 formats.append({
  46                     'format_id': player_type,
  47                     'url': vurl,
  48                 })
  49         rights = try_get(video_info, lambda x: x['rights'], dict) or {}
  50         if not formats and rights.get('geoBlockedSweden'):
  51             self.raise_geo_restricted(
  52                 'This video is only available in Sweden',
  53                 countries=self._GEO_COUNTRIES, metadata_available=True)
  54
  55         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  56         if isinstance(subtitle_references, list):
  57             for sr in subtitle_references:
  58                 subtitle_url = sr.get('url')
  59                 subtitle_lang = sr.get('language', 'sv')
  60                 if subtitle_url:
  61                     sub = {
  62                         'url': subtitle_url,
  63                     }
  64                     if determine_ext(subtitle_url) == 'm3u8':
  65                         # XXX: no way of testing, is it ever hit?
  66                         sub['ext'] = 'vtt'
  67                     subtitles.setdefault(subtitle_lang, []).append(sub)
  68
  69         title = video_info.get('title')
  70
  71         series = video_info.get('programTitle')
  72         season_number = int_or_none(video_info.get('season'))
  73         episode = video_info.get('episodeTitle')
  74         episode_number = int_or_none(video_info.get('episodeNumber'))
  75
  76         timestamp = unified_timestamp(rights.get('validFrom'))
  77         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  78         age_limit = None
  79         adult = dict_get(
  80             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  81             skip_false_values=False)
  82         if adult is not None:
  83             age_limit = 18 if adult else 0
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'formats': formats,
  89             'subtitles': subtitles,
  90             'duration': duration,
  91             'timestamp': timestamp,
  92             'age_limit': age_limit,
  93             'series': series,
  94             'season_number': season_number,
  95             'episode': episode,
  96             'episode_number': episode_number,
  97             'is_live': is_live,
  98         }
  99
 100
 101 class SVTIE(SVTBaseIE):
 102     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 103     _EMBED_REGEX = [r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % _VALID_URL]
 104     _TEST = {
 105         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 106         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 107         'info_dict': {
 108             'id': '2900353',
 109             'ext': 'mp4',
 110             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 111             'duration': 27,
 112             'age_limit': 0,
 113         },
 114     }
 115
 116     def _real_extract(self, url):
 117         mobj = self._match_valid_url(url)
 118         widget_id = mobj.group('widget_id')
 119         article_id = mobj.group('id')
 120
 121         info = self._download_json(
 122             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 123             article_id)
 124
 125         info_dict = self._extract_video(info['video'], article_id)
 126         info_dict['title'] = info['context']['title']
 127         return info_dict
 128
 129
 130 class SVTPlayBaseIE(SVTBaseIE):
 131     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 132
 133
 134 class SVTPlayIE(SVTPlayBaseIE):
 135     IE_DESC = 'SVT Play and Öppet arkiv'
 136     _VALID_URL = r'''(?x)
 137                     (?:
 138                         (?:
 139                             svt:|
 140                             https?://(?:www\.)?svt\.se/barnkanalen/barnplay/[^/]+/
 141                         )
 142                         (?P<svt_id>[^/?#&]+)|
 143                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 144                         (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))?
 145                     )
 146                     '''
 147     _TESTS = [{
 148         'url': 'https://www.svtplay.se/video/30479064',
 149         'md5': '2382036fd6f8c994856c323fe51c426e',
 150         'info_dict': {
 151             'id': '8zVbDPA',
 152             'ext': 'mp4',
 153             'title': 'Designdrömmar i Stenungsund',
 154             'timestamp': 1615770000,
 155             'upload_date': '20210315',
 156             'duration': 3519,
 157             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 158             'age_limit': 0,
 159             'subtitles': {
 160                 'sv': [{
 161                     'ext': 'vtt',
 162                 }]
 163             },
 164         },
 165         'params': {
 166             'skip_download': 'm3u8',
 167         },
 168         'skip': 'Episode is no longer available',
 169     }, {
 170         'url': 'https://www.svtplay.se/video/emBxBQj',
 171         'md5': '2382036fd6f8c994856c323fe51c426e',
 172         'info_dict': {
 173             'id': 'eyBd9aj',
 174             'ext': 'mp4',
 175             'title': '1. Farlig kryssning',
 176             'timestamp': 1491019200,
 177             'upload_date': '20170401',
 178             'duration': 2566,
 179             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 180             'age_limit': 0,
 181             'episode': '1. Farlig kryssning',
 182             'series': 'Rederiet',
 183             'subtitles': {
 184                 'sv': 'count:3'
 185             },
 186         },
 187         'params': {
 188             'skip_download': 'm3u8',
 189         },
 190     }, {
 191         'url': 'https://www.svtplay.se/video/jz2rYz7/anders-hansen-moter/james-fallon?info=visa',
 192         'info_dict': {
 193             'id': 'jvXAGVb',
 194             'ext': 'mp4',
 195             'title': 'James Fallon',
 196             'timestamp': 1673917200,
 197             'upload_date': '20230117',
 198             'duration': 1081,
 199             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 200             'age_limit': 0,
 201             'episode': 'James Fallon',
 202             'series': 'Anders Hansen möter...',
 203         },
 204         'params': {
 205             'skip_download': 'dash',
 206         },
 207     }, {
 208         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 209         'only_matching': True,
 210     }, {
 211         'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
 212         'only_matching': True,
 213     }, {
 214         # geo restricted to Sweden
 215         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 216         'only_matching': True,
 217     }, {
 218         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 219         'only_matching': True,
 220     }, {
 221         'url': 'https://www.svtplay.se/kanaler/svt1',
 222         'only_matching': True,
 223     }, {
 224         'url': 'svt:1376446-003A',
 225         'only_matching': True,
 226     }, {
 227         'url': 'svt:14278044',
 228         'only_matching': True,
 229     }, {
 230         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 231         'only_matching': True,
 232     }, {
 233         'url': 'svt:eWv5MLX',
 234         'only_matching': True,
 235     }]
 236
 237     def _extract_by_video_id(self, video_id, webpage=None):
 238         data = self._download_json(
 239             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 240             video_id, headers=self.geo_verification_headers())
 241         info_dict = self._extract_video(data, video_id)
 242         if not info_dict.get('title'):
 243             title = dict_get(info_dict, ('episode', 'series'))
 244             if not title and webpage:
 245                 title = re.sub(
 246                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 247             if not title:
 248                 title = video_id
 249             info_dict['title'] = title
 250         return info_dict
 251
 252     def _real_extract(self, url):
 253         mobj = self._match_valid_url(url)
 254         video_id = mobj.group('id')
 255         svt_id = mobj.group('svt_id') or mobj.group('modal_id')
 256
 257         if svt_id:
 258             return self._extract_by_video_id(svt_id)
 259
 260         webpage = self._download_webpage(url, video_id)
 261
 262         data = self._parse_json(
 263             self._search_regex(
 264                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 265                 group='json'),
 266             video_id, fatal=False)
 267
 268         thumbnail = self._og_search_thumbnail(webpage)
 269
 270         if data:
 271             video_info = try_get(
 272                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 273                 dict)
 274             if video_info:
 275                 info_dict = self._extract_video(video_info, video_id)
 276                 info_dict.update({
 277                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 278                     'thumbnail': thumbnail,
 279                 })
 280                 return info_dict
 281
 282             svt_id = try_get(
 283                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 284                 compat_str)
 285
 286         if not svt_id:
 287             nextjs_data = self._search_nextjs_data(webpage, video_id, fatal=False)
 288             svt_id = traverse_obj(nextjs_data, (
 289                 'props', 'urqlState', ..., 'data', {json.loads}, 'detailsPageByPath',
 290                 'video', 'svtId', {str}), get_all=False)
 291
 292         if not svt_id:
 293             svt_id = self._search_regex(
 294                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 295                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/[\w-]+/[^"\']*\b(?:modalId|id)=([\w-]+)'),
 296                 webpage, 'video id')
 297
 298         info_dict = self._extract_by_video_id(svt_id, webpage)
 299         info_dict['thumbnail'] = thumbnail
 300
 301         return info_dict
 302
 303
 304 class SVTSeriesIE(SVTPlayBaseIE):
 305     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 306     _TESTS = [{
 307         'url': 'https://www.svtplay.se/rederiet',
 308         'info_dict': {
 309             'id': '14445680',
 310             'title': 'Rederiet',
 311             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 312         },
 313         'playlist_mincount': 318,
 314     }, {
 315         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 316         'info_dict': {
 317             'id': 'season-2-14445680',
 318             'title': 'Rederiet - Säsong 2',
 319             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 320         },
 321         'playlist_mincount': 12,
 322     }]
 323
 324     @classmethod
 325     def suitable(cls, url):
 326         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 327
 328     def _real_extract(self, url):
 329         series_slug, season_id = self._match_valid_url(url).groups()
 330
 331         series = self._download_json(
 332             'https://api.svt.se/contento/graphql', series_slug,
 333             'Downloading series page', query={
 334                 'query': '''{
 335   listablesBySlug(slugs: ["%s"]) {
 336     associatedContent(include: [productionPeriod, season]) {
 337       items {
 338         item {
 339           ... on Episode {
 340             videoSvtId
 341           }
 342         }
 343       }
 344       id
 345       name
 346     }
 347     id
 348     longDescription
 349     name
 350     shortDescription
 351   }
 352 }''' % series_slug,
 353             })['data']['listablesBySlug'][0]
 354
 355         season_name = None
 356
 357         entries = []
 358         for season in series['associatedContent']:
 359             if not isinstance(season, dict):
 360                 continue
 361             if season_id:
 362                 if season.get('id') != season_id:
 363                     continue
 364                 season_name = season.get('name')
 365             items = season.get('items')
 366             if not isinstance(items, list):
 367                 continue
 368             for item in items:
 369                 video = item.get('item') or {}
 370                 content_id = video.get('videoSvtId')
 371                 if not content_id or not isinstance(content_id, compat_str):
 372                     continue
 373                 entries.append(self.url_result(
 374                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 375
 376         title = series.get('name')
 377         season_name = season_name or season_id
 378
 379         if title and season_name:
 380             title = '%s - %s' % (title, season_name)
 381         elif season_id:
 382             title = season_id
 383
 384         return self.playlist_result(
 385             entries, season_id or series.get('id'), title,
 386             dict_get(series, ('longDescription', 'shortDescription')))
 387
 388
 389 class SVTPageIE(SVTBaseIE):
 390     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?:[^/?#]+/)*(?P<id>[^/?&#]+)'
 391     _TESTS = [{
 392         'url': 'https://www.svt.se/nyheter/lokalt/skane/viktor-18-forlorade-armar-och-ben-i-sepsis-vill-ateruppta-karaten-och-bli-svetsare',
 393         'info_dict': {
 394             'title': 'Viktor, 18, förlorade armar och ben i sepsis – vill återuppta karaten och bli svetsare',
 395             'id': 'viktor-18-forlorade-armar-och-ben-i-sepsis-vill-ateruppta-karaten-och-bli-svetsare',
 396         },
 397         'playlist_count': 2,
 398     }, {
 399         'url': 'https://www.svt.se/nyheter/lokalt/skane/forsvarsmakten-om-trafikkaoset-pa-e22-kunde-inte-varit-dar-snabbare',
 400         'info_dict': {
 401             'id': 'jXvk42E',
 402             'title': 'Försvarsmakten om trafikkaoset på E22: Kunde inte varit där snabbare',
 403             'ext': 'mp4',
 404             "duration": 80,
 405             'age_limit': 0,
 406             'timestamp': 1704370009,
 407             'episode': 'Försvarsmakten om trafikkaoset på E22: Kunde inte varit där snabbare',
 408             'series': 'Lokala Nyheter Skåne',
 409             'upload_date': '20240104'
 410         },
 411         'params': {
 412             'skip_download': True,
 413         }
 414     }, {
 415         'url': 'https://www.svt.se/nyheter/svtforum/2023-tungt-ar-for-svensk-media',
 416         'info_dict': {
 417             'title': '2023 tungt år för svensk media',
 418             'id': 'ewqAZv4',
 419             'ext': 'mp4',
 420             "duration": 3074,
 421             'age_limit': 0,
 422             'series': '',
 423             'timestamp': 1702980479,
 424             'upload_date': '20231219',
 425             'episode': 'Mediestudier'
 426         },
 427         'params': {
 428             'skip_download': True,
 429         }
 430     }, {
 431         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 432         'info_dict': {
 433             'id': '25298267',
 434             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 435         },
 436         'playlist_count': 4,
 437         'skip': 'Video is gone'
 438     }, {
 439         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 440         'info_dict': {
 441             'id': '24243746',
 442             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 443         },
 444         'playlist_count': 2,
 445         'skip': 'Video is gone'
 446     }, {
 447         # only programTitle
 448         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 449         'info_dict': {
 450             'id': '8439V2K',
 451             'ext': 'mp4',
 452             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 453             'duration': 27,
 454             'age_limit': 0,
 455         },
 456         'skip': 'Video is gone'
 457     }, {
 458         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 459         'only_matching': True,
 460     }, {
 461         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 462         'only_matching': True,
 463     }]
 464
 465     @classmethod
 466     def suitable(cls, url):
 467         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 468
 469     def _real_extract(self, url):
 470         display_id = self._match_id(url)
 471
 472         webpage = self._download_webpage(url, display_id)
 473         title = self._og_search_title(webpage)
 474
 475         urql_state = self._search_json(
 476             r'window\.svt\.nyh\.urqlState\s*=', webpage, 'json data', display_id)
 477
 478         data = traverse_obj(urql_state, (..., 'data', {str}, {json.loads}), get_all=False) or {}
 479
 480         def entries():
 481             for video_id in set(traverse_obj(data, (
 482                 'page', (('topMedia', 'svtId'), ('body', ..., 'video', 'svtId')), {str}
 483             ))):
 484                 info = self._extract_video(
 485                     self._download_json(f'https://api.svt.se/video/{video_id}', video_id), video_id)
 486                 info['title'] = title
 487                 yield info
 488
 489         return self.playlist_result(entries(), display_id, title)