yt_dlp/extractor/svt.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_str
   5 from ..utils import (
   6     determine_ext,
   7     dict_get,
   8     int_or_none,
   9     unified_timestamp,
  10     str_or_none,
  11     strip_or_none,
  12     try_get,
  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 for now due to download test asserts that segment is > 10000 bytes and svt uses
 167             # init segments that are smaller
 168             # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B
 169             'skip_download': True,
 170         },
 171     }, {
 172         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 173         'only_matching': True,
 174     }, {
 175         'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
 176         'only_matching': True,
 177     }, {
 178         # geo restricted to Sweden
 179         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 180         'only_matching': True,
 181     }, {
 182         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 183         'only_matching': True,
 184     }, {
 185         'url': 'https://www.svtplay.se/kanaler/svt1',
 186         'only_matching': True,
 187     }, {
 188         'url': 'svt:1376446-003A',
 189         'only_matching': True,
 190     }, {
 191         'url': 'svt:14278044',
 192         'only_matching': True,
 193     }, {
 194         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 195         'only_matching': True,
 196     }, {
 197         'url': 'svt:eWv5MLX',
 198         'only_matching': True,
 199     }]
 200
 201     def _extract_by_video_id(self, video_id, webpage=None):
 202         data = self._download_json(
 203             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 204             video_id, headers=self.geo_verification_headers())
 205         info_dict = self._extract_video(data, video_id)
 206         if not info_dict.get('title'):
 207             title = dict_get(info_dict, ('episode', 'series'))
 208             if not title and webpage:
 209                 title = re.sub(
 210                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 211             if not title:
 212                 title = video_id
 213             info_dict['title'] = title
 214         return info_dict
 215
 216     def _real_extract(self, url):
 217         mobj = self._match_valid_url(url)
 218         video_id = mobj.group('id')
 219         svt_id = mobj.group('svt_id') or mobj.group('modal_id')
 220
 221         if svt_id:
 222             return self._extract_by_video_id(svt_id)
 223
 224         webpage = self._download_webpage(url, video_id)
 225
 226         data = self._parse_json(
 227             self._search_regex(
 228                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 229                 group='json'),
 230             video_id, fatal=False)
 231
 232         thumbnail = self._og_search_thumbnail(webpage)
 233
 234         if data:
 235             video_info = try_get(
 236                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 237                 dict)
 238             if video_info:
 239                 info_dict = self._extract_video(video_info, video_id)
 240                 info_dict.update({
 241                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 242                     'thumbnail': thumbnail,
 243                 })
 244                 return info_dict
 245
 246             svt_id = try_get(
 247                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 248                 compat_str)
 249
 250         if not svt_id:
 251             svt_id = self._search_regex(
 252                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 253                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id),
 254                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 255                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 256                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 257                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 258                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 259                 webpage, 'video id')
 260
 261         info_dict = self._extract_by_video_id(svt_id, webpage)
 262         info_dict['thumbnail'] = thumbnail
 263
 264         return info_dict
 265
 266
 267 class SVTSeriesIE(SVTPlayBaseIE):
 268     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 269     _TESTS = [{
 270         'url': 'https://www.svtplay.se/rederiet',
 271         'info_dict': {
 272             'id': '14445680',
 273             'title': 'Rederiet',
 274             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 275         },
 276         'playlist_mincount': 318,
 277     }, {
 278         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 279         'info_dict': {
 280             'id': 'season-2-14445680',
 281             'title': 'Rederiet - Säsong 2',
 282             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 283         },
 284         'playlist_mincount': 12,
 285     }]
 286
 287     @classmethod
 288     def suitable(cls, url):
 289         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 290
 291     def _real_extract(self, url):
 292         series_slug, season_id = self._match_valid_url(url).groups()
 293
 294         series = self._download_json(
 295             'https://api.svt.se/contento/graphql', series_slug,
 296             'Downloading series page', query={
 297                 'query': '''{
 298   listablesBySlug(slugs: ["%s"]) {
 299     associatedContent(include: [productionPeriod, season]) {
 300       items {
 301         item {
 302           ... on Episode {
 303             videoSvtId
 304           }
 305         }
 306       }
 307       id
 308       name
 309     }
 310     id
 311     longDescription
 312     name
 313     shortDescription
 314   }
 315 }''' % series_slug,
 316             })['data']['listablesBySlug'][0]
 317
 318         season_name = None
 319
 320         entries = []
 321         for season in series['associatedContent']:
 322             if not isinstance(season, dict):
 323                 continue
 324             if season_id:
 325                 if season.get('id') != season_id:
 326                     continue
 327                 season_name = season.get('name')
 328             items = season.get('items')
 329             if not isinstance(items, list):
 330                 continue
 331             for item in items:
 332                 video = item.get('item') or {}
 333                 content_id = video.get('videoSvtId')
 334                 if not content_id or not isinstance(content_id, compat_str):
 335                     continue
 336                 entries.append(self.url_result(
 337                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 338
 339         title = series.get('name')
 340         season_name = season_name or season_id
 341
 342         if title and season_name:
 343             title = '%s - %s' % (title, season_name)
 344         elif season_id:
 345             title = season_id
 346
 347         return self.playlist_result(
 348             entries, season_id or series.get('id'), title,
 349             dict_get(series, ('longDescription', 'shortDescription')))
 350
 351
 352 class SVTPageIE(InfoExtractor):
 353     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 354     _TESTS = [{
 355         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 356         'info_dict': {
 357             'id': '25298267',
 358             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 359         },
 360         'playlist_count': 4,
 361     }, {
 362         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 363         'info_dict': {
 364             'id': '24243746',
 365             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 366         },
 367         'playlist_count': 2,
 368     }, {
 369         # only programTitle
 370         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 371         'info_dict': {
 372             'id': '8439V2K',
 373             'ext': 'mp4',
 374             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 375             'duration': 27,
 376             'age_limit': 0,
 377         },
 378     }, {
 379         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 380         'only_matching': True,
 381     }, {
 382         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 383         'only_matching': True,
 384     }]
 385
 386     @classmethod
 387     def suitable(cls, url):
 388         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 389
 390     def _real_extract(self, url):
 391         path, display_id = self._match_valid_url(url).groups()
 392
 393         article = self._download_json(
 394             'https://api.svt.se/nss-api/page/' + path, display_id,
 395             query={'q': 'articles'})['articles']['content'][0]
 396
 397         entries = []
 398
 399         def _process_content(content):
 400             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 401                 video_id = compat_str(content['image']['svtId'])
 402                 entries.append(self.url_result(
 403                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 404
 405         for media in article.get('media', []):
 406             _process_content(media)
 407
 408         for obj in article.get('structuredBody', []):
 409             _process_content(obj.get('content') or {})
 410
 411         return self.playlist_result(
 412             entries, str_or_none(article.get('id')),
 413             strip_or_none(article.get('title')))