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)
  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                     )
 150                     '''
 151     _TESTS = [{
 152         'url': 'https://www.svtplay.se/video/26194546/det-har-ar-himlen',
 153         'md5': '2382036fd6f8c994856c323fe51c426e',
 154         'info_dict': {
 155             'id': 'jNwpV9P',
 156             'ext': 'mp4',
 157             'title': 'Det här är himlen',
 158             'timestamp': 1586044800,
 159             'upload_date': '20200405',
 160             'duration': 3515,
 161             'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$',
 162             'age_limit': 0,
 163             'subtitles': {
 164                 'sv': [{
 165                     'ext': 'vtt',
 166                 }]
 167             },
 168         },
 169         'params': {
 170             'format': 'bestvideo',
 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         # geo restricted to Sweden
 178         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 179         'only_matching': True,
 180     }, {
 181         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 182         'only_matching': True,
 183     }, {
 184         'url': 'https://www.svtplay.se/kanaler/svt1',
 185         'only_matching': True,
 186     }, {
 187         'url': 'svt:1376446-003A',
 188         'only_matching': True,
 189     }, {
 190         'url': 'svt:14278044',
 191         'only_matching': True,
 192     }, {
 193         'url': 'https://www.svt.se/barnkanalen/barnplay/kar/eWv5MLX/',
 194         'only_matching': True,
 195     }, {
 196         'url': 'svt:eWv5MLX',
 197         'only_matching': True,
 198     }]
 199
 200     def _adjust_title(self, info):
 201         if info['is_live']:
 202             info['title'] = self._live_title(info['title'])
 203
 204     def _extract_by_video_id(self, video_id, webpage=None):
 205         data = self._download_json(
 206             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 207             video_id, headers=self.geo_verification_headers())
 208         info_dict = self._extract_video(data, video_id)
 209         if not info_dict.get('title'):
 210             title = dict_get(info_dict, ('episode', 'series'))
 211             if not title and webpage:
 212                 title = re.sub(
 213                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 214             if not title:
 215                 title = video_id
 216             info_dict['title'] = title
 217         self._adjust_title(info_dict)
 218         return info_dict
 219
 220     def _real_extract(self, url):
 221         mobj = re.match(self._VALID_URL, url)
 222         video_id, svt_id = mobj.group('id', 'svt_id')
 223
 224         if svt_id:
 225             return self._extract_by_video_id(svt_id)
 226
 227         webpage = self._download_webpage(url, video_id)
 228
 229         data = self._parse_json(
 230             self._search_regex(
 231                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 232                 group='json'),
 233             video_id, fatal=False)
 234
 235         thumbnail = self._og_search_thumbnail(webpage)
 236
 237         if data:
 238             video_info = try_get(
 239                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 240                 dict)
 241             if video_info:
 242                 info_dict = self._extract_video(video_info, video_id)
 243                 info_dict.update({
 244                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 245                     'thumbnail': thumbnail,
 246                 })
 247                 self._adjust_title(info_dict)
 248                 return info_dict
 249
 250             svt_id = try_get(
 251                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 252                 compat_str)
 253
 254         if not svt_id:
 255             svt_id = self._search_regex(
 256                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 257                  r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 258                  r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)',
 259                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"',
 260                  r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)',
 261                  r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'),
 262                 webpage, 'video id')
 263
 264         info_dict = self._extract_by_video_id(svt_id, webpage)
 265         info_dict['thumbnail'] = thumbnail
 266
 267         return info_dict
 268
 269
 270 class SVTSeriesIE(SVTPlayBaseIE):
 271     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 272     _TESTS = [{
 273         'url': 'https://www.svtplay.se/rederiet',
 274         'info_dict': {
 275             'id': '14445680',
 276             'title': 'Rederiet',
 277             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 278         },
 279         'playlist_mincount': 318,
 280     }, {
 281         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 282         'info_dict': {
 283             'id': 'season-2-14445680',
 284             'title': 'Rederiet - Säsong 2',
 285             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 286         },
 287         'playlist_mincount': 12,
 288     }]
 289
 290     @classmethod
 291     def suitable(cls, url):
 292         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 293
 294     def _real_extract(self, url):
 295         series_slug, season_id = re.match(self._VALID_URL, url).groups()
 296
 297         series = self._download_json(
 298             'https://api.svt.se/contento/graphql', series_slug,
 299             'Downloading series page', query={
 300                 'query': '''{
 301   listablesBySlug(slugs: ["%s"]) {
 302     associatedContent(include: [productionPeriod, season]) {
 303       items {
 304         item {
 305           ... on Episode {
 306             videoSvtId
 307           }
 308         }
 309       }
 310       id
 311       name
 312     }
 313     id
 314     longDescription
 315     name
 316     shortDescription
 317   }
 318 }''' % series_slug,
 319             })['data']['listablesBySlug'][0]
 320
 321         season_name = None
 322
 323         entries = []
 324         for season in series['associatedContent']:
 325             if not isinstance(season, dict):
 326                 continue
 327             if season_id:
 328                 if season.get('id') != season_id:
 329                     continue
 330                 season_name = season.get('name')
 331             items = season.get('items')
 332             if not isinstance(items, list):
 333                 continue
 334             for item in items:
 335                 video = item.get('item') or {}
 336                 content_id = video.get('videoSvtId')
 337                 if not content_id or not isinstance(content_id, compat_str):
 338                     continue
 339                 entries.append(self.url_result(
 340                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 341
 342         title = series.get('name')
 343         season_name = season_name or season_id
 344
 345         if title and season_name:
 346             title = '%s - %s' % (title, season_name)
 347         elif season_id:
 348             title = season_id
 349
 350         return self.playlist_result(
 351             entries, season_id or series.get('id'), title,
 352             dict_get(series, ('longDescription', 'shortDescription')))
 353
 354
 355 class SVTPageIE(InfoExtractor):
 356     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 357     _TESTS = [{
 358         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 359         'info_dict': {
 360             'id': '25298267',
 361             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 362         },
 363         'playlist_count': 4,
 364     }, {
 365         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 366         'info_dict': {
 367             'id': '24243746',
 368             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 369         },
 370         'playlist_count': 2,
 371     }, {
 372         # only programTitle
 373         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 374         'info_dict': {
 375             'id': '8439V2K',
 376             'ext': 'mp4',
 377             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 378             'duration': 27,
 379             'age_limit': 0,
 380         },
 381     }, {
 382         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 383         'only_matching': True,
 384     }, {
 385         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 386         'only_matching': True,
 387     }]
 388
 389     @classmethod
 390     def suitable(cls, url):
 391         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 392
 393     def _real_extract(self, url):
 394         path, display_id = re.match(self._VALID_URL, url).groups()
 395
 396         article = self._download_json(
 397             'https://api.svt.se/nss-api/page/' + path, display_id,
 398             query={'q': 'articles'})['articles']['content'][0]
 399
 400         entries = []
 401
 402         def _process_content(content):
 403             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 404                 video_id = compat_str(content['image']['svtId'])
 405                 entries.append(self.url_result(
 406                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 407
 408         for media in article.get('media', []):
 409             _process_content(media)
 410
 411         for obj in article.get('structuredBody', []):
 412             _process_content(obj.get('content') or {})
 413
 414         return self.playlist_result(
 415             entries, str_or_none(article.get('id')),
 416             strip_or_none(article.get('title')))