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     try_get,
  16     urljoin,
  17     compat_str,
  18 )
  19
  20
  21 class SVTBaseIE(InfoExtractor):
  22     _GEO_COUNTRIES = ['SE']
  23
  24     def _extract_video(self, video_info, video_id):
  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_native',
  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         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  49             self.raise_geo_restricted(
  50                 'This video is only available in Sweden',
  51                 countries=self._GEO_COUNTRIES)
  52         self._sort_formats(formats)
  53
  54         subtitles = {}
  55         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  56         if isinstance(subtitle_references, list):
  57             for sr in subtitle_references:
  58                 subtitle_url = sr.get('url')
  59                 subtitle_lang = sr.get('language', 'sv')
  60                 if subtitle_url:
  61                     if determine_ext(subtitle_url) == 'm3u8':
  62                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  63                         continue
  64
  65                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  66
  67         title = video_info.get('title')
  68
  69         series = video_info.get('programTitle')
  70         season_number = int_or_none(video_info.get('season'))
  71         episode = video_info.get('episodeTitle')
  72         episode_number = int_or_none(video_info.get('episodeNumber'))
  73
  74         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  75         age_limit = None
  76         adult = dict_get(
  77             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  78             skip_false_values=False)
  79         if adult is not None:
  80             age_limit = 18 if adult else 0
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'formats': formats,
  86             'subtitles': subtitles,
  87             'duration': duration,
  88             'age_limit': age_limit,
  89             'series': series,
  90             'season_number': season_number,
  91             'episode': episode,
  92             'episode_number': episode_number,
  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'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)'
 138     _TESTS = [{
 139         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 140         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 141         'info_dict': {
 142             'id': '5996901',
 143             'ext': 'mp4',
 144             'title': 'Flygplan till Haile Selassie',
 145             'duration': 3527,
 146             'thumbnail': r're:^https?://.*[\.-]jpg$',
 147             'age_limit': 0,
 148             'subtitles': {
 149                 'sv': [{
 150                     'ext': 'wsrt',
 151                 }]
 152             },
 153         },
 154     }, {
 155         # geo restricted to Sweden
 156         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 157         'only_matching': True,
 158     }, {
 159         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 160         'only_matching': True,
 161     }]
 162
 163     def _real_extract(self, url):
 164         video_id = self._match_id(url)
 165
 166         webpage = self._download_webpage(url, video_id)
 167
 168         data = self._parse_json(
 169             self._search_regex(
 170                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 171                 group='json'),
 172             video_id, fatal=False)
 173
 174         thumbnail = self._og_search_thumbnail(webpage)
 175
 176         if data:
 177             video_info = try_get(
 178                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 179                 dict)
 180             if video_info:
 181                 info_dict = self._extract_video(video_info, video_id)
 182                 info_dict.update({
 183                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 184                     'thumbnail': thumbnail,
 185                 })
 186                 return info_dict
 187
 188         video_id = self._search_regex(
 189             r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 190             webpage, 'video id', default=None)
 191
 192         if video_id:
 193             data = self._download_json(
 194                 'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 195                 video_id, headers=self.geo_verification_headers())
 196             info_dict = self._extract_video(data, video_id)
 197             if not info_dict.get('title'):
 198                 info_dict['title'] = re.sub(
 199                     r'\s*\|\s*.+?$', '',
 200                     info_dict.get('episode') or self._og_search_title(webpage))
 201             return info_dict
 202
 203
 204 class SVTSeriesIE(SVTPlayBaseIE):
 205     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)'
 206     _TESTS = [{
 207         'url': 'https://www.svtplay.se/rederiet',
 208         'info_dict': {
 209             'id': 'rederiet',
 210             'title': 'Rederiet',
 211             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 212         },
 213         'playlist_mincount': 318,
 214     }, {
 215         'url': 'https://www.svtplay.se/rederiet?tab=sasong2',
 216         'info_dict': {
 217             'id': 'rederiet-sasong2',
 218             'title': 'Rederiet - Säsong 2',
 219             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 220         },
 221         'playlist_count': 12,
 222     }]
 223
 224     @classmethod
 225     def suitable(cls, url):
 226         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 227
 228     def _real_extract(self, url):
 229         series_id = self._match_id(url)
 230
 231         qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
 232         season_slug = qs.get('tab', [None])[0]
 233
 234         if season_slug:
 235             series_id += '-%s' % season_slug
 236
 237         webpage = self._download_webpage(
 238             url, series_id, 'Downloading series page')
 239
 240         root = self._parse_json(
 241             self._search_regex(
 242                 self._SVTPLAY_RE, webpage, 'content', group='json'),
 243             series_id)
 244
 245         season_name = None
 246
 247         entries = []
 248         for season in root['relatedVideoContent']['relatedVideosAccordion']:
 249             if not isinstance(season, dict):
 250                 continue
 251             if season_slug:
 252                 if season.get('slug') != season_slug:
 253                     continue
 254                 season_name = season.get('name')
 255             videos = season.get('videos')
 256             if not isinstance(videos, list):
 257                 continue
 258             for video in videos:
 259                 content_url = video.get('contentUrl')
 260                 if not content_url or not isinstance(content_url, compat_str):
 261                     continue
 262                 entries.append(
 263                     self.url_result(
 264                         urljoin(url, content_url),
 265                         ie=SVTPlayIE.ie_key(),
 266                         video_title=video.get('title')
 267                     ))
 268
 269         metadata = root.get('metaData')
 270         if not isinstance(metadata, dict):
 271             metadata = {}
 272
 273         title = metadata.get('title')
 274         season_name = season_name or season_slug
 275
 276         if title and season_name:
 277             title = '%s - %s' % (title, season_name)
 278         elif season_slug:
 279             title = season_slug
 280
 281         return self.playlist_result(
 282             entries, series_id, title, metadata.get('description'))