_ Git - youtube-dl/blob - 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 ..utils import (
   8     determine_ext,
   9     dict_get,
  10     int_or_none,
  11     try_get,
  12 )
  13
  14
  15 class SVTBaseIE(InfoExtractor):
  16     _GEO_COUNTRIES = ['SE']
  17     def _extract_video(self, video_info, video_id):
  18         formats = []
  19         for vr in video_info['videoReferences']:
  20             player_type = vr.get('playerType') or vr.get('format')
  21             vurl = vr['url']
  22             ext = determine_ext(vurl)
  23             if ext == 'm3u8':
  24                 formats.extend(self._extract_m3u8_formats(
  25                     vurl, video_id,
  26                     ext='mp4', entry_protocol='m3u8_native',
  27                     m3u8_id=player_type, fatal=False))
  28             elif ext == 'f4m':
  29                 formats.extend(self._extract_f4m_formats(
  30                     vurl + '?hdcore=3.3.0', video_id,
  31                     f4m_id=player_type, fatal=False))
  32             elif ext == 'mpd':
  33                 if player_type == 'dashhbbtv':
  34                     formats.extend(self._extract_mpd_formats(
  35                         vurl, video_id, mpd_id=player_type, fatal=False))
  36             else:
  37                 formats.append({
  38                     'format_id': player_type,
  39                     'url': vurl,
  40                 })
  41         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  42             self.raise_geo_restricted(
  43                 'This video is only available in Sweden',
  44                 countries=self._GEO_COUNTRIES)
  45         self._sort_formats(formats)
  46
  47         subtitles = {}
  48         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  49         if isinstance(subtitle_references, list):
  50             for sr in subtitle_references:
  51                 subtitle_url = sr.get('url')
  52                 subtitle_lang = sr.get('language', 'sv')
  53                 if subtitle_url:
  54                     if determine_ext(subtitle_url) == 'm3u8':
  55                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  56                         continue
  57
  58                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  59
  60         title = video_info.get('title')
  61
  62         series = video_info.get('programTitle')
  63         season_number = int_or_none(video_info.get('season'))
  64         episode = video_info.get('episodeTitle')
  65         episode_number = int_or_none(video_info.get('episodeNumber'))
  66
  67         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  68         age_limit = None
  69         adult = dict_get(
  70             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  71             skip_false_values=False)
  72         if adult is not None:
  73             age_limit = 18 if adult else 0
  74
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'formats': formats,
  79             'subtitles': subtitles,
  80             'duration': duration,
  81             'age_limit': age_limit,
  82             'series': series,
  83             'season_number': season_number,
  84             'episode': episode,
  85             'episode_number': episode_number,
  86         }
  87
  88
  89 class SVTIE(SVTBaseIE):
  90     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
  91     _TEST = {
  92         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
  93         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
  94         'info_dict': {
  95             'id': '2900353',
  96             'ext': 'mp4',
  97             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
  98             'duration': 27,
  99             'age_limit': 0,
 100         },
 101     }
 102
 103     @staticmethod
 104     def _extract_url(webpage):
 105         mobj = re.search(
 106             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 107         if mobj:
 108             return mobj.group('url')
 109
 110     def _real_extract(self, url):
 111         mobj = re.match(self._VALID_URL, url)
 112         widget_id = mobj.group('widget_id')
 113         article_id = mobj.group('id')
 114
 115         info = self._download_json(
 116             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 117             article_id)
 118
 119         info_dict = self._extract_video(info['video'], article_id)
 120         info_dict['title'] = info['context']['title']
 121         return info_dict
 122
 123
 124 class SVTPlayIE(SVTBaseIE):
 125     IE_DESC = 'SVT Play and Öppet arkiv'
 126     _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)'
 127     _TESTS = [{
 128         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 129         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 130         'info_dict': {
 131             'id': '5996901',
 132             'ext': 'mp4',
 133             'title': 'Flygplan till Haile Selassie',
 134             'duration': 3527,
 135             'thumbnail': r're:^https?://.*[\.-]jpg$',
 136             'age_limit': 0,
 137             'subtitles': {
 138                 'sv': [{
 139                     'ext': 'wsrt',
 140                 }]
 141             },
 142         },
 143     }, {
 144         # geo restricted to Sweden
 145         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 146         'only_matching': True,
 147     }, {
 148         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 149         'only_matching': True,
 150     }]
 151
 152     def _real_extract(self, url):
 153         video_id = self._match_id(url)
 154
 155         webpage = self._download_webpage(url, video_id)
 156
 157         data = self._parse_json(
 158             self._search_regex(
 159                 r'root\["__svtplay"\]\s*=\s*([^;]+);',
 160                 webpage, 'embedded data', default='{}'),
 161             video_id, fatal=False)
 162
 163         thumbnail = self._og_search_thumbnail(webpage)
 164
 165         if data:
 166             video_info = try_get(
 167                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 168                 dict)
 169             if video_info:
 170                 info_dict = self._extract_video(video_info, video_id)
 171                 info_dict.update({
 172                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 173                     'thumbnail': thumbnail,
 174                 })
 175                 return info_dict
 176
 177         video_id = self._search_regex(
 178             r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 179             webpage, 'video id', default=None)
 180
 181         if video_id:
 182             data = self._download_json(
 183                 'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id)
 184             info_dict = self._extract_video(data, video_id)
 185             if not info_dict.get('title'):
 186                 info_dict['title'] = re.sub(
 187                     r'\s*\|\s*.+?$', '',
 188                     info_dict.get('episode') or self._og_search_title(webpage))
 189             return info_dict