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 = re.match(self._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=(?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             'format': 'bestvideo',
 172             # skip for now due to download test asserts that segment is > 10000 bytes and svt uses
 173             # init segments that are smaller
 174             # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B
 175             'skip_download': True,
 176         },
 177     }, {
 178         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 179         'only_matching': True,
 180     }, {
 181         # geo restricted to Sweden
 182         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 183         'only_matching': True,
 184     }, {
 185         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 186         'only_matching': True,
 187     }, {
 188         'url': 'https://www.svtplay.se/kanaler/svt1',
 189         'only_matching': True,
 190     }, {
 191         'url': 'svt:1376446-003A',
 192         'only_matching': True,
 193     }, {
 194         'url': 'svt:14278044',
 195         'only_matching': True,
 196     }, {
 197         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 198         'only_matching': True,
 199     }, {
 200         'url': 'svt:eWv5MLX',
 201         'only_matching': True,
 202     }]
 203
 204     def _adjust_title(self, info):
 205         if info['is_live']:
 206             info['title'] = self._live_title(info['title'])
 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         self._adjust_title(info_dict)
 222         return info_dict
 223
 224     def _real_extract(self, url):
 225         mobj = re.match(self._VALID_URL, url)
 226         video_id = mobj.group('id')
 227         svt_id = mobj.group('svt_id') or mobj.group('modal_id')
 228
 229         if svt_id:
 230             return self._extract_by_video_id(svt_id)
 231
 232         webpage = self._download_webpage(url, video_id)
 233
 234         data = self._parse_json(
 235             self._search_regex(
 236                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 237                 group='json'),
 238             video_id, fatal=False)
 239
 240         thumbnail = self._og_search_thumbnail(webpage)
 241
 242         if data:
 243             video_info = try_get(
 244                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 245                 dict)
 246             if video_info:
 247                 info_dict = self._extract_video(video_info, video_id)
 248                 info_dict.update({
 249                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 250                     'thumbnail': thumbnail,
 251                 })
 252                 self._adjust_title(info_dict)
 253                 return info_dict
 254
 255             svt_id = try_get(
 256                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 257                 compat_str)
 258
 259         if not svt_id:
 260             svt_id = self._search_regex(
 261                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 262                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\bmodalId=([\da-zA-Z-]+)' % re.escape(video_id),
 263                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 264                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 265                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 266                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 267                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 268                 webpage, 'video id')
 269
 270         info_dict = self._extract_by_video_id(svt_id, webpage)
 271         info_dict['thumbnail'] = thumbnail
 272
 273         return info_dict
 274
 275
 276 class SVTSeriesIE(SVTPlayBaseIE):
 277     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 278     _TESTS = [{
 279         'url': 'https://www.svtplay.se/rederiet',
 280         'info_dict': {
 281             'id': '14445680',
 282             'title': 'Rederiet',
 283             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 284         },
 285         'playlist_mincount': 318,
 286     }, {
 287         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 288         'info_dict': {
 289             'id': 'season-2-14445680',
 290             'title': 'Rederiet - Säsong 2',
 291             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 292         },
 293         'playlist_mincount': 12,
 294     }]
 295
 296     @classmethod
 297     def suitable(cls, url):
 298         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 299
 300     def _real_extract(self, url):
 301         series_slug, season_id = re.match(self._VALID_URL, url).groups()
 302
 303         series = self._download_json(
 304             'https://api.svt.se/contento/graphql', series_slug,
 305             'Downloading series page', query={
 306                 'query': '''{
 307   listablesBySlug(slugs: ["%s"]) {
 308     associatedContent(include: [productionPeriod, season]) {
 309       items {
 310         item {
 311           ... on Episode {
 312             videoSvtId
 313           }
 314         }
 315       }
 316       id
 317       name
 318     }
 319     id
 320     longDescription
 321     name
 322     shortDescription
 323   }
 324 }''' % series_slug,
 325             })['data']['listablesBySlug'][0]
 326
 327         season_name = None
 328
 329         entries = []
 330         for season in series['associatedContent']:
 331             if not isinstance(season, dict):
 332                 continue
 333             if season_id:
 334                 if season.get('id') != season_id:
 335                     continue
 336                 season_name = season.get('name')
 337             items = season.get('items')
 338             if not isinstance(items, list):
 339                 continue
 340             for item in items:
 341                 video = item.get('item') or {}
 342                 content_id = video.get('videoSvtId')
 343                 if not content_id or not isinstance(content_id, compat_str):
 344                     continue
 345                 entries.append(self.url_result(
 346                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 347
 348         title = series.get('name')
 349         season_name = season_name or season_id
 350
 351         if title and season_name:
 352             title = '%s - %s' % (title, season_name)
 353         elif season_id:
 354             title = season_id
 355
 356         return self.playlist_result(
 357             entries, season_id or series.get('id'), title,
 358             dict_get(series, ('longDescription', 'shortDescription')))
 359
 360
 361 class SVTPageIE(InfoExtractor):
 362     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 363     _TESTS = [{
 364         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 365         'info_dict': {
 366             'id': '25298267',
 367             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 368         },
 369         'playlist_count': 4,
 370     }, {
 371         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 372         'info_dict': {
 373             'id': '24243746',
 374             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 375         },
 376         'playlist_count': 2,
 377     }, {
 378         # only programTitle
 379         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 380         'info_dict': {
 381             'id': '8439V2K',
 382             'ext': 'mp4',
 383             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 384             'duration': 27,
 385             'age_limit': 0,
 386         },
 387     }, {
 388         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 389         'only_matching': True,
 390     }, {
 391         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 392         'only_matching': True,
 393     }]
 394
 395     @classmethod
 396     def suitable(cls, url):
 397         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 398
 399     def _real_extract(self, url):
 400         path, display_id = re.match(self._VALID_URL, url).groups()
 401
 402         article = self._download_json(
 403             'https://api.svt.se/nss-api/page/' + path, display_id,
 404             query={'q': 'articles'})['articles']['content'][0]
 405
 406         entries = []
 407
 408         def _process_content(content):
 409             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 410                 video_id = compat_str(content['image']['svtId'])
 411                 entries.append(self.url_result(
 412                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 413
 414         for media in article.get('media', []):
 415             _process_content(media)
 416
 417         for obj in article.get('structuredBody', []):
 418             _process_content(obj.get('content') or {})
 419
 420         return self.playlist_result(
 421             entries, str_or_none(article.get('id')),
 422             strip_or_none(article.get('title')))