yt_dlp/extractor/svt.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     determine_ext,
  10     dict_get,
  11     int_or_none,
  12     unified_timestamp,
  13     str_or_none,
  14     strip_or_none,
  15     try_get,
  16 )
  17
  18
  19 class SVTBaseIE(InfoExtractor):
  20     _GEO_COUNTRIES = ['SE']
  21
  22     def _extract_video(self, video_info, video_id):
  23         is_live = dict_get(video_info, ('live', 'simulcast'), default=False)
  24         m3u8_protocol = 'm3u8' if is_live else 'm3u8_native'
  25         formats = []
  26         for vr in video_info['videoReferences']:
  27             player_type = vr.get('playerType') or vr.get('format')
  28             vurl = vr['url']
  29             ext = determine_ext(vurl)
  30             if ext == 'm3u8':
  31                 formats.extend(self._extract_m3u8_formats(
  32                     vurl, video_id,
  33                     ext='mp4', entry_protocol=m3u8_protocol,
  34                     m3u8_id=player_type, fatal=False))
  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                 if player_type == 'dashhbbtv':
  41                     formats.extend(self._extract_mpd_formats(
  42                         vurl, video_id, mpd_id=player_type, fatal=False))
  43             else:
  44                 formats.append({
  45                     'format_id': player_type,
  46                     'url': vurl,
  47                 })
  48         rights = try_get(video_info, lambda x: x['rights'], dict) or {}
  49         if not formats and rights.get('geoBlockedSweden'):
  50             self.raise_geo_restricted(
  51                 'This video is only available in Sweden',
  52                 countries=self._GEO_COUNTRIES, metadata_available=True)
  53         self._sort_formats(formats)
  54
  55         subtitles = {}
  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                     if determine_ext(subtitle_url) == 'm3u8':
  63                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  64                         continue
  65
  66                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  67
  68         title = video_info.get('title')
  69
  70         series = video_info.get('programTitle')
  71         season_number = int_or_none(video_info.get('season'))
  72         episode = video_info.get('episodeTitle')
  73         episode_number = int_or_none(video_info.get('episodeNumber'))
  74
  75         timestamp = unified_timestamp(rights.get('validFrom'))
  76         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  77         age_limit = None
  78         adult = dict_get(
  79             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  80             skip_false_values=False)
  81         if adult is not None:
  82             age_limit = 18 if adult else 0
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'formats': formats,
  88             'subtitles': subtitles,
  89             'duration': duration,
  90             'timestamp': timestamp,
  91             'age_limit': age_limit,
  92             'series': series,
  93             'season_number': season_number,
  94             'episode': episode,
  95             'episode_number': episode_number,
  96             'is_live': is_live,
  97         }
  98
  99
 100 class SVTIE(SVTBaseIE):
 101     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 102     _TEST = {
 103         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 104         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 105         'info_dict': {
 106             'id': '2900353',
 107             'ext': 'mp4',
 108             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 109             'duration': 27,
 110             'age_limit': 0,
 111         },
 112     }
 113
 114     @staticmethod
 115     def _extract_url(webpage):
 116         mobj = re.search(
 117             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 118         if mobj:
 119             return mobj.group('url')
 120
 121     def _real_extract(self, url):
 122         mobj = self._match_valid_url(url)
 123         widget_id = mobj.group('widget_id')
 124         article_id = mobj.group('id')
 125
 126         info = self._download_json(
 127             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 128             article_id)
 129
 130         info_dict = self._extract_video(info['video'], article_id)
 131         info_dict['title'] = info['context']['title']
 132         return info_dict
 133
 134
 135 class SVTPlayBaseIE(SVTBaseIE):
 136     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 137
 138
 139 class SVTPlayIE(SVTPlayBaseIE):
 140     IE_DESC = 'SVT Play and Öppet arkiv'
 141     _VALID_URL = r'''(?x)
 142                     (?:
 143                         (?:
 144                             svt:|
 145                             https?://(?:www\.)?svt\.se/barnkanalen/barnplay/[^/]+/
 146                         )
 147                         (?P<svt_id>[^/?#&]+)|
 148                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 149                         (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))?
 150                     )
 151                     '''
 152     _TESTS = [{
 153         'url': 'https://www.svtplay.se/video/30479064',
 154         'md5': '2382036fd6f8c994856c323fe51c426e',
 155         'info_dict': {
 156             'id': '8zVbDPA',
 157             'ext': 'mp4',
 158             'title': 'Designdrömmar i Stenungsund',
 159             'timestamp': 1615770000,
 160             'upload_date': '20210315',
 161             'duration': 3519,
 162             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 163             'age_limit': 0,
 164             'subtitles': {
 165                 'sv': [{
 166                     'ext': 'vtt',
 167                 }]
 168             },
 169         },
 170         'params': {
 171             # skip for now due to download test asserts that segment is > 10000 bytes and svt uses
 172             # init segments that are smaller
 173             # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B
 174             'skip_download': True,
 175         },
 176     }, {
 177         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 178         'only_matching': True,
 179     }, {
 180         'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
 181         'only_matching': True,
 182     }, {
 183         # geo restricted to Sweden
 184         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 185         'only_matching': True,
 186     }, {
 187         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 188         'only_matching': True,
 189     }, {
 190         'url': 'https://www.svtplay.se/kanaler/svt1',
 191         'only_matching': True,
 192     }, {
 193         'url': 'svt:1376446-003A',
 194         'only_matching': True,
 195     }, {
 196         'url': 'svt:14278044',
 197         'only_matching': True,
 198     }, {
 199         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 200         'only_matching': True,
 201     }, {
 202         'url': 'svt:eWv5MLX',
 203         'only_matching': True,
 204     }]
 205
 206     def _adjust_title(self, info):
 207         if info['is_live']:
 208             info['title'] = self._live_title(info['title'])
 209
 210     def _extract_by_video_id(self, video_id, webpage=None):
 211         data = self._download_json(
 212             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 213             video_id, headers=self.geo_verification_headers())
 214         info_dict = self._extract_video(data, video_id)
 215         if not info_dict.get('title'):
 216             title = dict_get(info_dict, ('episode', 'series'))
 217             if not title and webpage:
 218                 title = re.sub(
 219                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 220             if not title:
 221                 title = video_id
 222             info_dict['title'] = title
 223         self._adjust_title(info_dict)
 224         return info_dict
 225
 226     def _real_extract(self, url):
 227         mobj = self._match_valid_url(url)
 228         video_id = mobj.group('id')
 229         svt_id = mobj.group('svt_id') or mobj.group('modal_id')
 230
 231         if svt_id:
 232             return self._extract_by_video_id(svt_id)
 233
 234         webpage = self._download_webpage(url, video_id)
 235
 236         data = self._parse_json(
 237             self._search_regex(
 238                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 239                 group='json'),
 240             video_id, fatal=False)
 241
 242         thumbnail = self._og_search_thumbnail(webpage)
 243
 244         if data:
 245             video_info = try_get(
 246                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 247                 dict)
 248             if video_info:
 249                 info_dict = self._extract_video(video_info, video_id)
 250                 info_dict.update({
 251                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 252                     'thumbnail': thumbnail,
 253                 })
 254                 self._adjust_title(info_dict)
 255                 return info_dict
 256
 257             svt_id = try_get(
 258                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 259                 compat_str)
 260
 261         if not svt_id:
 262             svt_id = self._search_regex(
 263                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 264                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id),
 265                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 266                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 267                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 268                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 269                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 270                 webpage, 'video id')
 271
 272         info_dict = self._extract_by_video_id(svt_id, webpage)
 273         info_dict['thumbnail'] = thumbnail
 274
 275         return info_dict
 276
 277
 278 class SVTSeriesIE(SVTPlayBaseIE):
 279     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 280     _TESTS = [{
 281         'url': 'https://www.svtplay.se/rederiet',
 282         'info_dict': {
 283             'id': '14445680',
 284             'title': 'Rederiet',
 285             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 286         },
 287         'playlist_mincount': 318,
 288     }, {
 289         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 290         'info_dict': {
 291             'id': 'season-2-14445680',
 292             'title': 'Rederiet - Säsong 2',
 293             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 294         },
 295         'playlist_mincount': 12,
 296     }]
 297
 298     @classmethod
 299     def suitable(cls, url):
 300         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 301
 302     def _real_extract(self, url):
 303         series_slug, season_id = self._match_valid_url(url).groups()
 304
 305         series = self._download_json(
 306             'https://api.svt.se/contento/graphql', series_slug,
 307             'Downloading series page', query={
 308                 'query': '''{
 309   listablesBySlug(slugs: ["%s"]) {
 310     associatedContent(include: [productionPeriod, season]) {
 311       items {
 312         item {
 313           ... on Episode {
 314             videoSvtId
 315           }
 316         }
 317       }
 318       id
 319       name
 320     }
 321     id
 322     longDescription
 323     name
 324     shortDescription
 325   }
 326 }''' % series_slug,
 327             })['data']['listablesBySlug'][0]
 328
 329         season_name = None
 330
 331         entries = []
 332         for season in series['associatedContent']:
 333             if not isinstance(season, dict):
 334                 continue
 335             if season_id:
 336                 if season.get('id') != season_id:
 337                     continue
 338                 season_name = season.get('name')
 339             items = season.get('items')
 340             if not isinstance(items, list):
 341                 continue
 342             for item in items:
 343                 video = item.get('item') or {}
 344                 content_id = video.get('videoSvtId')
 345                 if not content_id or not isinstance(content_id, compat_str):
 346                     continue
 347                 entries.append(self.url_result(
 348                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 349
 350         title = series.get('name')
 351         season_name = season_name or season_id
 352
 353         if title and season_name:
 354             title = '%s - %s' % (title, season_name)
 355         elif season_id:
 356             title = season_id
 357
 358         return self.playlist_result(
 359             entries, season_id or series.get('id'), title,
 360             dict_get(series, ('longDescription', 'shortDescription')))
 361
 362
 363 class SVTPageIE(InfoExtractor):
 364     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 365     _TESTS = [{
 366         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 367         'info_dict': {
 368             'id': '25298267',
 369             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 370         },
 371         'playlist_count': 4,
 372     }, {
 373         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 374         'info_dict': {
 375             'id': '24243746',
 376             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 377         },
 378         'playlist_count': 2,
 379     }, {
 380         # only programTitle
 381         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 382         'info_dict': {
 383             'id': '8439V2K',
 384             'ext': 'mp4',
 385             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 386             'duration': 27,
 387             'age_limit': 0,
 388         },
 389     }, {
 390         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 391         'only_matching': True,
 392     }, {
 393         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 394         'only_matching': True,
 395     }]
 396
 397     @classmethod
 398     def suitable(cls, url):
 399         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 400
 401     def _real_extract(self, url):
 402         path, display_id = self._match_valid_url(url).groups()
 403
 404         article = self._download_json(
 405             'https://api.svt.se/nss-api/page/' + path, display_id,
 406             query={'q': 'articles'})['articles']['content'][0]
 407
 408         entries = []
 409
 410         def _process_content(content):
 411             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 412                 video_id = compat_str(content['image']['svtId'])
 413                 entries.append(self.url_result(
 414                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 415
 416         for media in article.get('media', []):
 417             _process_content(media)
 418
 419         for obj in article.get('structuredBody', []):
 420             _process_content(obj.get('content') or {})
 421
 422         return self.playlist_result(
 423             entries, str_or_none(article.get('id')),
 424             strip_or_none(article.get('title')))