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         self._sort_formats(formats)
  55
  56         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  57         if isinstance(subtitle_references, list):
  58             for sr in subtitle_references:
  59                 subtitle_url = sr.get('url')
  60                 subtitle_lang = sr.get('language', 'sv')
  61                 if subtitle_url:
  62                     sub = {
  63                         'url': subtitle_url,
  64                     }
  65                     if determine_ext(subtitle_url) == 'm3u8':
  66                         # XXX: no way of testing, is it ever hit?
  67                         sub['ext'] = 'vtt'
  68                     subtitles.setdefault(subtitle_lang, []).append(sub)
  69
  70         title = video_info.get('title')
  71
  72         series = video_info.get('programTitle')
  73         season_number = int_or_none(video_info.get('season'))
  74         episode = video_info.get('episodeTitle')
  75         episode_number = int_or_none(video_info.get('episodeNumber'))
  76
  77         timestamp = unified_timestamp(rights.get('validFrom'))
  78         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  79         age_limit = None
  80         adult = dict_get(
  81             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  82             skip_false_values=False)
  83         if adult is not None:
  84             age_limit = 18 if adult else 0
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'formats': formats,
  90             'subtitles': subtitles,
  91             'duration': duration,
  92             'timestamp': timestamp,
  93             'age_limit': age_limit,
  94             'series': series,
  95             'season_number': season_number,
  96             'episode': episode,
  97             'episode_number': episode_number,
  98             'is_live': is_live,
  99         }
 100
 101
 102 class SVTIE(SVTBaseIE):
 103     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 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     @staticmethod
 117     def _extract_url(webpage):
 118         mobj = re.search(
 119             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 120         if mobj:
 121             return mobj.group('url')
 122
 123     def _real_extract(self, url):
 124         mobj = self._match_valid_url(url)
 125         widget_id = mobj.group('widget_id')
 126         article_id = mobj.group('id')
 127
 128         info = self._download_json(
 129             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 130             article_id)
 131
 132         info_dict = self._extract_video(info['video'], article_id)
 133         info_dict['title'] = info['context']['title']
 134         return info_dict
 135
 136
 137 class SVTPlayBaseIE(SVTBaseIE):
 138     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 139
 140
 141 class SVTPlayIE(SVTPlayBaseIE):
 142     IE_DESC = 'SVT Play and Öppet arkiv'
 143     _VALID_URL = r'''(?x)
 144                     (?:
 145                         (?:
 146                             svt:|
 147                             https?://(?:www\.)?svt\.se/barnkanalen/barnplay/[^/]+/
 148                         )
 149                         (?P<svt_id>[^/?#&]+)|
 150                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 151                         (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))?
 152                     )
 153                     '''
 154     _TESTS = [{
 155         'url': 'https://www.svtplay.se/video/30479064',
 156         'md5': '2382036fd6f8c994856c323fe51c426e',
 157         'info_dict': {
 158             'id': '8zVbDPA',
 159             'ext': 'mp4',
 160             'title': 'Designdrömmar i Stenungsund',
 161             'timestamp': 1615770000,
 162             'upload_date': '20210315',
 163             'duration': 3519,
 164             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 165             'age_limit': 0,
 166             'subtitles': {
 167                 'sv': [{
 168                     'ext': 'vtt',
 169                 }]
 170             },
 171         },
 172         'params': {
 173             # skip for now due to download test asserts that segment is > 10000 bytes and svt uses
 174             # init segments that are smaller
 175             # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B
 176             'skip_download': True,
 177         },
 178     }, {
 179         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 180         'only_matching': True,
 181     }, {
 182         'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
 183         'only_matching': True,
 184     }, {
 185         # geo restricted to Sweden
 186         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 187         'only_matching': True,
 188     }, {
 189         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 190         'only_matching': True,
 191     }, {
 192         'url': 'https://www.svtplay.se/kanaler/svt1',
 193         'only_matching': True,
 194     }, {
 195         'url': 'svt:1376446-003A',
 196         'only_matching': True,
 197     }, {
 198         'url': 'svt:14278044',
 199         'only_matching': True,
 200     }, {
 201         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 202         'only_matching': True,
 203     }, {
 204         'url': 'svt:eWv5MLX',
 205         'only_matching': True,
 206     }]
 207
 208     def _extract_by_video_id(self, video_id, webpage=None):
 209         data = self._download_json(
 210             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 211             video_id, headers=self.geo_verification_headers())
 212         info_dict = self._extract_video(data, video_id)
 213         if not info_dict.get('title'):
 214             title = dict_get(info_dict, ('episode', 'series'))
 215             if not title and webpage:
 216                 title = re.sub(
 217                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 218             if not title:
 219                 title = video_id
 220             info_dict['title'] = title
 221         return info_dict
 222
 223     def _real_extract(self, url):
 224         mobj = self._match_valid_url(url)
 225         video_id = mobj.group('id')
 226         svt_id = mobj.group('svt_id') or mobj.group('modal_id')
 227
 228         if svt_id:
 229             return self._extract_by_video_id(svt_id)
 230
 231         webpage = self._download_webpage(url, video_id)
 232
 233         data = self._parse_json(
 234             self._search_regex(
 235                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 236                 group='json'),
 237             video_id, fatal=False)
 238
 239         thumbnail = self._og_search_thumbnail(webpage)
 240
 241         if data:
 242             video_info = try_get(
 243                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 244                 dict)
 245             if video_info:
 246                 info_dict = self._extract_video(video_info, video_id)
 247                 info_dict.update({
 248                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 249                     'thumbnail': thumbnail,
 250                 })
 251                 return info_dict
 252
 253             svt_id = try_get(
 254                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 255                 compat_str)
 256
 257         if not svt_id:
 258             svt_id = self._search_regex(
 259                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 260                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id),
 261                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 262                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 263                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 264                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 265                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 266                 webpage, 'video id')
 267
 268         info_dict = self._extract_by_video_id(svt_id, webpage)
 269         info_dict['thumbnail'] = thumbnail
 270
 271         return info_dict
 272
 273
 274 class SVTSeriesIE(SVTPlayBaseIE):
 275     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 276     _TESTS = [{
 277         'url': 'https://www.svtplay.se/rederiet',
 278         'info_dict': {
 279             'id': '14445680',
 280             'title': 'Rederiet',
 281             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 282         },
 283         'playlist_mincount': 318,
 284     }, {
 285         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 286         'info_dict': {
 287             'id': 'season-2-14445680',
 288             'title': 'Rederiet - Säsong 2',
 289             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 290         },
 291         'playlist_mincount': 12,
 292     }]
 293
 294     @classmethod
 295     def suitable(cls, url):
 296         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 297
 298     def _real_extract(self, url):
 299         series_slug, season_id = self._match_valid_url(url).groups()
 300
 301         series = self._download_json(
 302             'https://api.svt.se/contento/graphql', series_slug,
 303             'Downloading series page', query={
 304                 'query': '''{
 305   listablesBySlug(slugs: ["%s"]) {
 306     associatedContent(include: [productionPeriod, season]) {
 307       items {
 308         item {
 309           ... on Episode {
 310             videoSvtId
 311           }
 312         }
 313       }
 314       id
 315       name
 316     }
 317     id
 318     longDescription
 319     name
 320     shortDescription
 321   }
 322 }''' % series_slug,
 323             })['data']['listablesBySlug'][0]
 324
 325         season_name = None
 326
 327         entries = []
 328         for season in series['associatedContent']:
 329             if not isinstance(season, dict):
 330                 continue
 331             if season_id:
 332                 if season.get('id') != season_id:
 333                     continue
 334                 season_name = season.get('name')
 335             items = season.get('items')
 336             if not isinstance(items, list):
 337                 continue
 338             for item in items:
 339                 video = item.get('item') or {}
 340                 content_id = video.get('videoSvtId')
 341                 if not content_id or not isinstance(content_id, compat_str):
 342                     continue
 343                 entries.append(self.url_result(
 344                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 345
 346         title = series.get('name')
 347         season_name = season_name or season_id
 348
 349         if title and season_name:
 350             title = '%s - %s' % (title, season_name)
 351         elif season_id:
 352             title = season_id
 353
 354         return self.playlist_result(
 355             entries, season_id or series.get('id'), title,
 356             dict_get(series, ('longDescription', 'shortDescription')))
 357
 358
 359 class SVTPageIE(InfoExtractor):
 360     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 361     _TESTS = [{
 362         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 363         'info_dict': {
 364             'id': '25298267',
 365             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 366         },
 367         'playlist_count': 4,
 368     }, {
 369         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 370         'info_dict': {
 371             'id': '24243746',
 372             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 373         },
 374         'playlist_count': 2,
 375     }, {
 376         # only programTitle
 377         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 378         'info_dict': {
 379             'id': '8439V2K',
 380             'ext': 'mp4',
 381             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 382             'duration': 27,
 383             'age_limit': 0,
 384         },
 385     }, {
 386         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 387         'only_matching': True,
 388     }, {
 389         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 390         'only_matching': True,
 391     }]
 392
 393     @classmethod
 394     def suitable(cls, url):
 395         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 396
 397     def _real_extract(self, url):
 398         path, display_id = self._match_valid_url(url).groups()
 399
 400         article = self._download_json(
 401             'https://api.svt.se/nss-api/page/' + path, display_id,
 402             query={'q': 'articles'})['articles']['content'][0]
 403
 404         entries = []
 405
 406         def _process_content(content):
 407             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 408                 video_id = compat_str(content['image']['svtId'])
 409                 entries.append(self.url_result(
 410                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 411
 412         for media in article.get('media', []):
 413             _process_content(media)
 414
 415         for obj in article.get('structuredBody', []):
 416             _process_content(obj.get('content') or {})
 417
 418         return self.playlist_result(
 419             entries, str_or_none(article.get('id')),
 420             strip_or_none(article.get('title')))