youtube_dl/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     str_or_none,
  13     strip_or_none,
  14     try_get,
  15 )
  16
  17
  18 class SVTBaseIE(InfoExtractor):
  19     _GEO_COUNTRIES = ['SE']
  20
  21     def _extract_video(self, video_info, video_id):
  22         is_live = dict_get(video_info, ('live', 'simulcast'), default=False)
  23         m3u8_protocol = 'm3u8' if is_live else 'm3u8_native'
  24         formats = []
  25         for vr in video_info['videoReferences']:
  26             player_type = vr.get('playerType') or vr.get('format')
  27             vurl = vr['url']
  28             ext = determine_ext(vurl)
  29             if ext == 'm3u8':
  30                 formats.extend(self._extract_m3u8_formats(
  31                     vurl, video_id,
  32                     ext='mp4', entry_protocol=m3u8_protocol,
  33                     m3u8_id=player_type, fatal=False))
  34             elif ext == 'f4m':
  35                 formats.extend(self._extract_f4m_formats(
  36                     vurl + '?hdcore=3.3.0', video_id,
  37                     f4m_id=player_type, fatal=False))
  38             elif ext == 'mpd':
  39                 if player_type == 'dashhbbtv':
  40                     formats.extend(self._extract_mpd_formats(
  41                         vurl, video_id, mpd_id=player_type, fatal=False))
  42             else:
  43                 formats.append({
  44                     'format_id': player_type,
  45                     'url': vurl,
  46                 })
  47         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  48             self.raise_geo_restricted(
  49                 'This video is only available in Sweden',
  50                 countries=self._GEO_COUNTRIES)
  51         self._sort_formats(formats)
  52
  53         subtitles = {}
  54         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  55         if isinstance(subtitle_references, list):
  56             for sr in subtitle_references:
  57                 subtitle_url = sr.get('url')
  58                 subtitle_lang = sr.get('language', 'sv')
  59                 if subtitle_url:
  60                     if determine_ext(subtitle_url) == 'm3u8':
  61                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  62                         continue
  63
  64                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  65
  66         title = video_info.get('title')
  67
  68         series = video_info.get('programTitle')
  69         season_number = int_or_none(video_info.get('season'))
  70         episode = video_info.get('episodeTitle')
  71         episode_number = int_or_none(video_info.get('episodeNumber'))
  72
  73         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  74         age_limit = None
  75         adult = dict_get(
  76             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  77             skip_false_values=False)
  78         if adult is not None:
  79             age_limit = 18 if adult else 0
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'formats': formats,
  85             'subtitles': subtitles,
  86             'duration': duration,
  87             'age_limit': age_limit,
  88             'series': series,
  89             'season_number': season_number,
  90             'episode': episode,
  91             'episode_number': episode_number,
  92             'is_live': is_live,
  93         }
  94
  95
  96 class SVTIE(SVTBaseIE):
  97     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
  98     _TEST = {
  99         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 100         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 101         'info_dict': {
 102             'id': '2900353',
 103             'ext': 'mp4',
 104             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 105             'duration': 27,
 106             'age_limit': 0,
 107         },
 108     }
 109
 110     @staticmethod
 111     def _extract_url(webpage):
 112         mobj = re.search(
 113             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 114         if mobj:
 115             return mobj.group('url')
 116
 117     def _real_extract(self, url):
 118         mobj = re.match(self._VALID_URL, url)
 119         widget_id = mobj.group('widget_id')
 120         article_id = mobj.group('id')
 121
 122         info = self._download_json(
 123             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 124             article_id)
 125
 126         info_dict = self._extract_video(info['video'], article_id)
 127         info_dict['title'] = info['context']['title']
 128         return info_dict
 129
 130
 131 class SVTPlayBaseIE(SVTBaseIE):
 132     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 133
 134
 135 class SVTPlayIE(SVTPlayBaseIE):
 136     IE_DESC = 'SVT Play and Öppet arkiv'
 137     _VALID_URL = r'''(?x)
 138                     (?:
 139                         svt:(?P<svt_id>[^/?#&]+)|
 140                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 141                     )
 142                     '''
 143     _TESTS = [{
 144         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 145         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 146         'info_dict': {
 147             'id': '5996901',
 148             'ext': 'mp4',
 149             'title': 'Flygplan till Haile Selassie',
 150             'duration': 3527,
 151             'thumbnail': r're:^https?://.*[\.-]jpg$',
 152             'age_limit': 0,
 153             'subtitles': {
 154                 'sv': [{
 155                     'ext': 'wsrt',
 156                 }]
 157             },
 158         },
 159     }, {
 160         # geo restricted to Sweden
 161         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 162         'only_matching': True,
 163     }, {
 164         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 165         'only_matching': True,
 166     }, {
 167         'url': 'https://www.svtplay.se/kanaler/svt1',
 168         'only_matching': True,
 169     }, {
 170         'url': 'svt:1376446-003A',
 171         'only_matching': True,
 172     }, {
 173         'url': 'svt:14278044',
 174         'only_matching': True,
 175     }]
 176
 177     def _adjust_title(self, info):
 178         if info['is_live']:
 179             info['title'] = self._live_title(info['title'])
 180
 181     def _extract_by_video_id(self, video_id, webpage=None):
 182         data = self._download_json(
 183             'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 184             video_id, headers=self.geo_verification_headers())
 185         info_dict = self._extract_video(data, video_id)
 186         if not info_dict.get('title'):
 187             title = dict_get(info_dict, ('episode', 'series'))
 188             if not title and webpage:
 189                 title = re.sub(
 190                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 191             if not title:
 192                 title = video_id
 193             info_dict['title'] = title
 194         self._adjust_title(info_dict)
 195         return info_dict
 196
 197     def _real_extract(self, url):
 198         mobj = re.match(self._VALID_URL, url)
 199         video_id, svt_id = mobj.group('id', 'svt_id')
 200
 201         if svt_id:
 202             return self._extract_by_video_id(svt_id)
 203
 204         webpage = self._download_webpage(url, video_id)
 205
 206         data = self._parse_json(
 207             self._search_regex(
 208                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 209                 group='json'),
 210             video_id, fatal=False)
 211
 212         thumbnail = self._og_search_thumbnail(webpage)
 213
 214         if data:
 215             video_info = try_get(
 216                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 217                 dict)
 218             if video_info:
 219                 info_dict = self._extract_video(video_info, video_id)
 220                 info_dict.update({
 221                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 222                     'thumbnail': thumbnail,
 223                 })
 224                 self._adjust_title(info_dict)
 225                 return info_dict
 226
 227             svt_id = try_get(
 228                 data, lambda x: x['statistics']['dataLake']['content']['id'],
 229                 compat_str)
 230
 231         if not svt_id:
 232             svt_id = self._search_regex(
 233                 (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 234                  r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"'),
 235                 webpage, 'video id')
 236
 237         return self._extract_by_video_id(svt_id, webpage)
 238
 239
 240 class SVTSeriesIE(SVTPlayBaseIE):
 241     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)(?:.+?\btab=(?P<season_slug>[^&#]+))?'
 242     _TESTS = [{
 243         'url': 'https://www.svtplay.se/rederiet',
 244         'info_dict': {
 245             'id': '14445680',
 246             'title': 'Rederiet',
 247             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 248         },
 249         'playlist_mincount': 318,
 250     }, {
 251         'url': 'https://www.svtplay.se/rederiet?tab=season-2-14445680',
 252         'info_dict': {
 253             'id': 'season-2-14445680',
 254             'title': 'Rederiet - Säsong 2',
 255             'description': 'md5:d9fdfff17f5d8f73468176ecd2836039',
 256         },
 257         'playlist_mincount': 12,
 258     }]
 259
 260     @classmethod
 261     def suitable(cls, url):
 262         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 263
 264     def _real_extract(self, url):
 265         series_slug, season_id = re.match(self._VALID_URL, url).groups()
 266
 267         series = self._download_json(
 268             'https://api.svt.se/contento/graphql', series_slug,
 269             'Downloading series page', query={
 270                 'query': '''{
 271   listablesBySlug(slugs: ["%s"]) {
 272     associatedContent(include: [productionPeriod, season]) {
 273       items {
 274         item {
 275           ... on Episode {
 276             videoSvtId
 277           }
 278         }
 279       }
 280       id
 281       name
 282     }
 283     id
 284     longDescription
 285     name
 286     shortDescription
 287   }
 288 }''' % series_slug,
 289             })['data']['listablesBySlug'][0]
 290
 291         season_name = None
 292
 293         entries = []
 294         for season in series['associatedContent']:
 295             if not isinstance(season, dict):
 296                 continue
 297             if season_id:
 298                 if season.get('id') != season_id:
 299                     continue
 300                 season_name = season.get('name')
 301             items = season.get('items')
 302             if not isinstance(items, list):
 303                 continue
 304             for item in items:
 305                 video = item.get('item') or {}
 306                 content_id = video.get('videoSvtId')
 307                 if not content_id or not isinstance(content_id, compat_str):
 308                     continue
 309                 entries.append(self.url_result(
 310                     'svt:' + content_id, SVTPlayIE.ie_key(), content_id))
 311
 312         title = series.get('name')
 313         season_name = season_name or season_id
 314
 315         if title and season_name:
 316             title = '%s - %s' % (title, season_name)
 317         elif season_id:
 318             title = season_id
 319
 320         return self.playlist_result(
 321             entries, season_id or series.get('id'), title,
 322             dict_get(series, ('longDescription', 'shortDescription')))
 323
 324
 325 class SVTPageIE(InfoExtractor):
 326     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
 327     _TESTS = [{
 328         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
 329         'info_dict': {
 330             'id': '25298267',
 331             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
 332         },
 333         'playlist_count': 4,
 334     }, {
 335         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
 336         'info_dict': {
 337             'id': '24243746',
 338             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
 339         },
 340         'playlist_count': 2,
 341     }, {
 342         # only programTitle
 343         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 344         'info_dict': {
 345             'id': '8439V2K',
 346             'ext': 'mp4',
 347             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 348             'duration': 27,
 349             'age_limit': 0,
 350         },
 351     }, {
 352         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 353         'only_matching': True,
 354     }, {
 355         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 356         'only_matching': True,
 357     }]
 358
 359     @classmethod
 360     def suitable(cls, url):
 361         return False if SVTIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 362
 363     def _real_extract(self, url):
 364         path, display_id = re.match(self._VALID_URL, url).groups()
 365
 366         article = self._download_json(
 367             'https://api.svt.se/nss-api/page/' + path, display_id,
 368             query={'q': 'articles'})['articles']['content'][0]
 369
 370         entries = []
 371
 372         def _process_content(content):
 373             if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
 374                 video_id = compat_str(content['image']['svtId'])
 375                 entries.append(self.url_result(
 376                     'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
 377
 378         for media in article.get('media', []):
 379             _process_content(media)
 380
 381         for obj in article.get('structuredBody', []):
 382             _process_content(obj.get('content') or {})
 383
 384         return self.playlist_result(
 385             entries, str_or_none(article.get('id')),
 386             strip_or_none(article.get('title')))