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 (
   8     compat_parse_qs,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     dict_get,
  14     int_or_none,
  15     orderedSet,
  16     strip_or_none,
  17     try_get,
  18     urljoin,
  19     compat_str,
  20 )
  21
  22
  23 class SVTBaseIE(InfoExtractor):
  24     _GEO_COUNTRIES = ['SE']
  25
  26     def _extract_video(self, video_info, video_id):
  27         is_live = dict_get(video_info, ('live', 'simulcast'), default=False)
  28         m3u8_protocol = 'm3u8' if is_live else 'm3u8_native'
  29         formats = []
  30         for vr in video_info['videoReferences']:
  31             player_type = vr.get('playerType') or vr.get('format')
  32             vurl = vr['url']
  33             ext = determine_ext(vurl)
  34             if ext == 'm3u8':
  35                 formats.extend(self._extract_m3u8_formats(
  36                     vurl, video_id,
  37                     ext='mp4', entry_protocol=m3u8_protocol,
  38                     m3u8_id=player_type, fatal=False))
  39             elif ext == 'f4m':
  40                 formats.extend(self._extract_f4m_formats(
  41                     vurl + '?hdcore=3.3.0', video_id,
  42                     f4m_id=player_type, fatal=False))
  43             elif ext == 'mpd':
  44                 if player_type == 'dashhbbtv':
  45                     formats.extend(self._extract_mpd_formats(
  46                         vurl, video_id, mpd_id=player_type, fatal=False))
  47             else:
  48                 formats.append({
  49                     'format_id': player_type,
  50                     'url': vurl,
  51                 })
  52         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  53             self.raise_geo_restricted(
  54                 'This video is only available in Sweden',
  55                 countries=self._GEO_COUNTRIES)
  56         self._sort_formats(formats)
  57
  58         subtitles = {}
  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                     if determine_ext(subtitle_url) == 'm3u8':
  66                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  67                         continue
  68
  69                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  70
  71         title = video_info.get('title')
  72
  73         series = video_info.get('programTitle')
  74         season_number = int_or_none(video_info.get('season'))
  75         episode = video_info.get('episodeTitle')
  76         episode_number = int_or_none(video_info.get('episodeNumber'))
  77
  78         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  79         age_limit = None
  80         adult = dict_get(
  81             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  82             skip_false_values=False)
  83         if adult is not None:
  84             age_limit = 18 if adult else 0
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'formats': formats,
  90             'subtitles': subtitles,
  91             'duration': duration,
  92             'age_limit': age_limit,
  93             'series': series,
  94             'season_number': season_number,
  95             'episode': episode,
  96             'episode_number': episode_number,
  97             'is_live': is_live,
  98         }
  99
 100
 101 class SVTIE(SVTBaseIE):
 102     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 103     _TEST = {
 104         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 105         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 106         'info_dict': {
 107             'id': '2900353',
 108             'ext': 'mp4',
 109             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 110             'duration': 27,
 111             'age_limit': 0,
 112         },
 113     }
 114
 115     @staticmethod
 116     def _extract_url(webpage):
 117         mobj = re.search(
 118             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 119         if mobj:
 120             return mobj.group('url')
 121
 122     def _real_extract(self, url):
 123         mobj = re.match(self._VALID_URL, url)
 124         widget_id = mobj.group('widget_id')
 125         article_id = mobj.group('id')
 126
 127         info = self._download_json(
 128             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 129             article_id)
 130
 131         info_dict = self._extract_video(info['video'], article_id)
 132         info_dict['title'] = info['context']['title']
 133         return info_dict
 134
 135
 136 class SVTPlayBaseIE(SVTBaseIE):
 137     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 138
 139
 140 class SVTPlayIE(SVTPlayBaseIE):
 141     IE_DESC = 'SVT Play and Öppet arkiv'
 142     _VALID_URL = r'''(?x)
 143                     (?:
 144                         svt:(?P<svt_id>[^/?#&]+)|
 145                         https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)
 146                     )
 147                     '''
 148     _TESTS = [{
 149         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 150         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 151         'info_dict': {
 152             'id': '5996901',
 153             'ext': 'mp4',
 154             'title': 'Flygplan till Haile Selassie',
 155             'duration': 3527,
 156             'thumbnail': r're:^https?://.*[\.-]jpg$',
 157             'age_limit': 0,
 158             'subtitles': {
 159                 'sv': [{
 160                     'ext': 'wsrt',
 161                 }]
 162             },
 163         },
 164     }, {
 165         # geo restricted to Sweden
 166         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 167         'only_matching': True,
 168     }, {
 169         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 170         'only_matching': True,
 171     }, {
 172         'url': 'https://www.svtplay.se/kanaler/svt1',
 173         'only_matching': True,
 174     }, {
 175         'url': 'svt:1376446-003A',
 176         'only_matching': True,
 177     }, {
 178         'url': 'svt:14278044',
 179         'only_matching': True,
 180     }]
 181
 182     def _adjust_title(self, info):
 183         if info['is_live']:
 184             info['title'] = self._live_title(info['title'])
 185
 186     def _extract_by_video_id(self, video_id, webpage=None):
 187         data = self._download_json(
 188             'https://api.svt.se/video/%s' % video_id,
 189             video_id, headers=self.geo_verification_headers())
 190         info_dict = self._extract_video(data, video_id)
 191         if not info_dict.get('title'):
 192             title = dict_get(info_dict, ('episode', 'series'))
 193             if not title and webpage:
 194                 title = re.sub(
 195                     r'\s*\|\s*.+?$', '', self._og_search_title(webpage))
 196             if not title:
 197                 title = video_id
 198             info_dict['title'] = title
 199         self._adjust_title(info_dict)
 200         return info_dict
 201
 202     def _real_extract(self, url):
 203         mobj = re.match(self._VALID_URL, url)
 204         video_id, svt_id = mobj.group('id', 'svt_id')
 205
 206         if svt_id:
 207             return self._extract_by_video_id(svt_id)
 208
 209         webpage = self._download_webpage(url, video_id)
 210
 211         data = self._parse_json(
 212             self._search_regex(
 213                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 214                 group='json'),
 215             video_id, fatal=False)
 216
 217         thumbnail = self._og_search_thumbnail(webpage)
 218
 219         if data:
 220             video_info = try_get(
 221                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 222                 dict)
 223             if video_info:
 224                 info_dict = self._extract_video(video_info, video_id)
 225                 info_dict.update({
 226                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 227                     'thumbnail': thumbnail,
 228                 })
 229                 self._adjust_title(info_dict)
 230                 return info_dict
 231
 232         svt_id = self._search_regex(
 233             r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 234             webpage, 'video id')
 235
 236         return self._extract_by_video_id(svt_id, webpage)
 237
 238
 239 class SVTSeriesIE(SVTPlayBaseIE):
 240     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)'
 241     _TESTS = [{
 242         'url': 'https://www.svtplay.se/rederiet',
 243         'info_dict': {
 244             'id': 'rederiet',
 245             'title': 'Rederiet',
 246             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 247         },
 248         'playlist_mincount': 318,
 249     }, {
 250         'url': 'https://www.svtplay.se/rederiet?tab=sasong2',
 251         'info_dict': {
 252             'id': 'rederiet-sasong2',
 253             'title': 'Rederiet - Säsong 2',
 254             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 255         },
 256         'playlist_count': 12,
 257     }]
 258
 259     @classmethod
 260     def suitable(cls, url):
 261         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 262
 263     def _real_extract(self, url):
 264         series_id = self._match_id(url)
 265
 266         qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
 267         season_slug = qs.get('tab', [None])[0]
 268
 269         if season_slug:
 270             series_id += '-%s' % season_slug
 271
 272         webpage = self._download_webpage(
 273             url, series_id, 'Downloading series page')
 274
 275         root = self._parse_json(
 276             self._search_regex(
 277                 self._SVTPLAY_RE, webpage, 'content', group='json'),
 278             series_id)
 279
 280         season_name = None
 281
 282         entries = []
 283         for season in root['relatedVideoContent']['relatedVideosAccordion']:
 284             if not isinstance(season, dict):
 285                 continue
 286             if season_slug:
 287                 if season.get('slug') != season_slug:
 288                     continue
 289                 season_name = season.get('name')
 290             videos = season.get('videos')
 291             if not isinstance(videos, list):
 292                 continue
 293             for video in videos:
 294                 content_url = video.get('contentUrl')
 295                 if not content_url or not isinstance(content_url, compat_str):
 296                     continue
 297                 entries.append(
 298                     self.url_result(
 299                         urljoin(url, content_url),
 300                         ie=SVTPlayIE.ie_key(),
 301                         video_title=video.get('title')
 302                     ))
 303
 304         metadata = root.get('metaData')
 305         if not isinstance(metadata, dict):
 306             metadata = {}
 307
 308         title = metadata.get('title')
 309         season_name = season_name or season_slug
 310
 311         if title and season_name:
 312             title = '%s - %s' % (title, season_name)
 313         elif season_slug:
 314             title = season_slug
 315
 316         return self.playlist_result(
 317             entries, series_id, title, metadata.get('description'))
 318
 319
 320 class SVTPageIE(InfoExtractor):
 321     _VALID_URL = r'https?://(?:www\.)?svt\.se/(?:[^/]+/)*(?P<id>[^/?&#]+)'
 322     _TESTS = [{
 323         'url': 'https://www.svt.se/sport/oseedat/guide-sommartraningen-du-kan-gora-var-och-nar-du-vill',
 324         'info_dict': {
 325             'id': 'guide-sommartraningen-du-kan-gora-var-och-nar-du-vill',
 326             'title': 'GUIDE: Sommarträning du kan göra var och när du vill',
 327         },
 328         'playlist_count': 7,
 329     }, {
 330         'url': 'https://www.svt.se/nyheter/inrikes/ebba-busch-thor-kd-har-delvis-ratt-om-no-go-zoner',
 331         'info_dict': {
 332             'id': 'ebba-busch-thor-kd-har-delvis-ratt-om-no-go-zoner',
 333             'title': 'Ebba Busch Thor har bara delvis rätt om ”no-go-zoner”',
 334         },
 335         'playlist_count': 1,
 336     }, {
 337         # only programTitle
 338         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
 339         'info_dict': {
 340             'id': '2900353',
 341             'ext': 'mp4',
 342             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 343             'duration': 27,
 344             'age_limit': 0,
 345         },
 346     }, {
 347         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
 348         'only_matching': True,
 349     }, {
 350         'url': 'https://www.svt.se/vader/manadskronikor/maj2018',
 351         'only_matching': True,
 352     }]
 353
 354     @classmethod
 355     def suitable(cls, url):
 356         return False if SVTIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 357
 358     def _real_extract(self, url):
 359         playlist_id = self._match_id(url)
 360
 361         webpage = self._download_webpage(url, playlist_id)
 362
 363         entries = [
 364             self.url_result(
 365                 'svt:%s' % video_id, ie=SVTPlayIE.ie_key(), video_id=video_id)
 366             for video_id in orderedSet(re.findall(
 367                 r'data-video-id=["\'](\d+)', webpage))]
 368
 369         title = strip_or_none(self._og_search_title(webpage, default=None))
 370
 371         return self.playlist_result(entries, playlist_id, title)