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