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         subtitles = {}
  27         for vr in video_info['videoReferences']:
  28             player_type = vr.get('playerType') or vr.get('format')
  29             vurl = vr['url']
  30             ext = determine_ext(vurl)
  31             if ext == 'm3u8':
  32                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  33                     vurl, video_id,
  34                     ext='mp4', entry_protocol=m3u8_protocol,
  35                     m3u8_id=player_type, fatal=False)
  36                 formats.extend(fmts)
  37                 self._merge_subtitles(subs, target=subtitles)
  38             elif ext == 'f4m':
  39                 formats.extend(self._extract_f4m_formats(
  40                     vurl + '?hdcore=3.3.0', video_id,
  41                     f4m_id=player_type, fatal=False))
  42             elif ext == 'mpd':
  43                 fmts, subs = self._extract_mpd_formats_and_subtitles(
  44                     vurl, video_id, mpd_id=player_type, fatal=False)
  45                 formats.extend(fmts)
  46                 self._merge_subtitles(subs, target=subtitles)
  47             else:
  48                 formats.append({
  49                     'format_id': player_type,
  50                     'url': vurl,
  51                 })
  52         rights = try_get(video_info, lambda x: x['rights'], dict) or {}
  53         if not formats and rights.get('geoBlockedSweden'):
  54             self.raise_geo_restricted(
  55                 'This video is only available in Sweden',
  56                 countries=self._GEO_COUNTRIES, metadata_available=True)
  57         self._sort_formats(formats)
  58
  59         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  60         if isinstance(subtitle_references, list):
  61             for sr in subtitle_references:
  62                 subtitle_url = sr.get('url')
  63                 subtitle_lang = sr.get('language', 'sv')
  64                 if subtitle_url:
  65                     sub = {
  66                         'url': subtitle_url,
  67                     }
  68                     if determine_ext(subtitle_url) == 'm3u8':
  69                         # XXX: no way of testing, is it ever hit?
  70                         sub['ext'] = 'vtt'
  71                     subtitles.setdefault(subtitle_lang, []).append(sub)
  72
  73         title = video_info.get('title')
  74
  75         series = video_info.get('programTitle')
  76         season_number = int_or_none(video_info.get('season'))
  77         episode = video_info.get('episodeTitle')
  78         episode_number = int_or_none(video_info.get('episodeNumber'))
  79
  80         timestamp = unified_timestamp(rights.get('validFrom'))
  81         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  82         age_limit = None
  83         adult = dict_get(
  84             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  85             skip_false_values=False)
  86         if adult is not None:
  87             age_limit = 18 if adult else 0
  88
  89         return {
  90             'id': video_id,
  91             'title': title,
  92             'formats': formats,
  93             'subtitles': subtitles,
  94             'duration': duration,
  95             'timestamp': timestamp,
  96             'age_limit': age_limit,
  97             'series': series,
  98             'season_number': season_number,
  99             'episode': episode,
 100             'episode_number': episode_number,
 101             'is_live': is_live,
 102         }
 103
 104
 105 class SVTIE(SVTBaseIE):
 106     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 107     _TEST = {
 108         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 109         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 110         'info_dict': {
 111             'id': '2900353',
 112             'ext': 'mp4',
 113             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 114             'duration': 27,
 115             'age_limit': 0,
 116         },
 117     }
 118
 119     @staticmethod
 120     def _extract_url(webpage):
 121         mobj = re.search(
 122             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 123         if mobj:
 124             return mobj.group('url')
 125
 126     def _real_extract(self, url):
 127         mobj = self._match_valid_url(url)
 128         widget_id = mobj.group('widget_id')
 129         article_id = mobj.group('id')
 130
 131         info = self._download_json(
 132             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 133             article_id)
 134
 135         info_dict = self._extract_video(info['video'], article_id)
 136         info_dict['title'] = info['context']['title']
 137         return info_dict
 138
 139
 140 class SVTPlayBaseIE(SVTBaseIE):
 141     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 142
 143
 144 class SVTPlayIE(SVTPlayBaseIE):
 145     IE_DESC = 'SVT Play and Öppet arkiv'
 146     _VALID_URL = r'''(?x)
 147                     (?:
 148                         (?:
 149                             svt:|
 150                             https?://(?:www\.)?svt\.se/barnkanalen/barnplay/[^/]+/
 151                         )
 152                         (?P<svt_id>[^/?#&]+)|
 153                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 154                         (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))?
 155                     )
 156                     '''
 157     _TESTS = [{
 158         'url': 'https://www.svtplay.se/video/30479064',
 159         'md5': '2382036fd6f8c994856c323fe51c426e',
 160         'info_dict': {
 161             'id': '8zVbDPA',
 162             'ext': 'mp4',
 163             'title': 'Designdrömmar i Stenungsund',
 164             'timestamp': 1615770000,
 165             'upload_date': '20210315',
 166             'duration': 3519,
 167             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 168             'age_limit': 0,
 169             'subtitles': {
 170                 'sv': [{
 171                     'ext': 'vtt',
 172                 }]
 173             },
 174         },
 175         'params': {
 176             # skip for now due to download test asserts that segment is > 10000 bytes and svt uses
 177             # init segments that are smaller
 178             # AssertionError: Expected test_SVTPlay_jNwpV9P.mp4 to be at least 9.77KiB, but it's only 864.00B
 179             'skip_download': True,
 180         },
 181     }, {
 182         'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA',
 183         'only_matching': True,
 184     }, {
 185         'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa',
 186         'only_matching': True,
 187     }, {
 188         # geo restricted to Sweden
 189         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 190         'only_matching': True,
 191     }, {
 192         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 193         'only_matching': True,
 194     }, {
 195         'url': 'https://www.svtplay.se/kanaler/svt1',
 196         'only_matching': True,
 197     }, {
 198         'url': 'svt:1376446-003A',
 199         'only_matching': True,
 200     }, {
 201         'url': 'svt:14278044',
 202         'only_matching': True,
 203     }, {
 204         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 205         'only_matching': True,
 206     }, {
 207         'url': 'svt:eWv5MLX',
 208         'only_matching': True,
 209     }]
 210
 211     def _extract_by_video_id(self, video_id, webpage=None):
 212         data = self._download_json(
 213             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 214             video_id, headers=self.geo_verification_headers())
 215         info_dict = self._extract_video(data, video_id)
 216         if not info_dict.get('title'):
 217             title = dict_get(info_dict, ('episode', 'series'))
 218             if not title and webpage:
 219                 title = re.sub(
 220                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 221             if not title:
 222                 title = video_id
 223             info_dict['title'] = title
 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                 return info_dict
 255
 256             svt_id = try_get(
 257                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 258                 compat_str)
 259
 260         if not svt_id:
 261             svt_id = self._search_regex(
 262                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 263                  r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id),
 264                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 265                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 266                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 267                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 268                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 269                 webpage, 'video id')
 270
 271         info_dict = self._extract_by_video_id(svt_id, webpage)
 272         info_dict['thumbnail'] = thumbnail
 273
 274         return info_dict
 275
 276
 277 class SVTSeriesIE(SVTPlayBaseIE):
 278     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 279     _TESTS = [{
 280         'url': 'https://www.svtplay.se/rederiet',
 281         'info_dict': {
 282             'id': '14445680',
 283             'title': 'Rederiet',
 284             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 285         },
 286         'playlist_mincount': 318,
 287     }, {
 288         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 289         'info_dict': {
 290             'id': 'season-2-14445680',
 291             'title': 'Rederiet - Säsong 2',
 292             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 293         },
 294         'playlist_mincount': 12,
 295     }]
 296
 297     @classmethod
 298     def suitable(cls, url):
 299         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 300
 301     def _real_extract(self, url):
 302         series_slug, season_id = self._match_valid_url(url).groups()
 303
 304         series = self._download_json(
 305             'https://api.svt.se/contento/graphql', series_slug,
 306             'Downloading series page', query={
 307                 'query': '''{
 308   listablesBySlug(slugs: ["%s"]) {
 309     associatedContent(include: [productionPeriod, season]) {
 310       items {
 311         item {
 312           ... on Episode {
 313             videoSvtId
 314           }
 315         }
 316       }
 317       id
 318       name
 319     }
 320     id
 321     longDescription
 322     name
 323     shortDescription
 324   }
 325 }''' % series_slug,
 326             })['data']['listablesBySlug'][0]
 327
 328         season_name = None
 329
 330         entries = []
 331         for season in series['associatedContent']:
 332             if not isinstance(season, dict):
 333                 continue
 334             if season_id:
 335                 if season.get('id') != season_id:
 336                     continue
 337                 season_name = season.get('name')
 338             items = season.get('items')
 339             if not isinstance(items, list):
 340                 continue
 341             for item in items:
 342                 video = item.get('item') or {}
 343                 content_id = video.get('videoSvtId')
 344                 if not content_id or not isinstance(content_id, compat_str):
 345                     continue
 346                 entries.append(self.url_result(
 347                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 348
 349         title = series.get('name')
 350         season_name = season_name or season_id
 351
 352         if title and season_name:
 353             title = '%s - %s' % (title, season_name)
 354         elif season_id:
 355             title = season_id
 356
 357         return self.playlist_result(
 358             entries, season_id or series.get('id'), title,
 359             dict_get(series, ('longDescription', 'shortDescription')))
 360
 361
 362 class SVTPageIE(InfoExtractor):
 363     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 364     _TESTS = [{
 365         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 366         'info_dict': {
 367             'id': '25298267',
 368             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 369         },
 370         'playlist_count': 4,
 371     }, {
 372         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 373         'info_dict': {
 374             'id': '24243746',
 375             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 376         },
 377         'playlist_count': 2,
 378     }, {
 379         # only programTitle
 380         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 381         'info_dict': {
 382             'id': '8439V2K',
 383             'ext': 'mp4',
 384             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 385             'duration': 27,
 386             'age_limit': 0,
 387         },
 388     }, {
 389         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 390         'only_matching': True,
 391     }, {
 392         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 393         'only_matching': True,
 394     }]
 395
 396     @classmethod
 397     def suitable(cls, url):
 398         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 399
 400     def _real_extract(self, url):
 401         path, display_id = self._match_valid_url(url).groups()
 402
 403         article = self._download_json(
 404             'https://api.svt.se/nss-api/page/' + path, display_id,
 405             query={'q': 'articles'})['articles']['content'][0]
 406
 407         entries = []
 408
 409         def _process_content(content):
 410             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 411                 video_id = compat_str(content['image']['svtId'])
 412                 entries.append(self.url_result(
 413                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 414
 415         for media in article.get('media', []):
 416             _process_content(media)
 417
 418         for obj in article.get('structuredBody', []):
 419             _process_content(obj.get('content') or {})
 420
 421         return self.playlist_result(
 422             entries, str_or_none(article.get('id')),
 423             strip_or_none(article.get('title')))