_ 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     urljoin,
  13     compat_str,
  14 )
  15
  16
  17 class SVTBaseIE(InfoExtractor):
  18     _GEO_COUNTRIES = ['SE']
  19
  20     def _extract_video(self, video_info, video_id):
  21         formats = []
  22         for vr in video_info['videoReferences']:
  23             player_type = vr.get('playerType') or vr.get('format')
  24             vurl = vr['url']
  25             ext = determine_ext(vurl)
  26             if ext == 'm3u8':
  27                 formats.extend(self._extract_m3u8_formats(
  28                     vurl, video_id,
  29                     ext='mp4', entry_protocol='m3u8_native',
  30                     m3u8_id=player_type, fatal=False))
  31             elif ext == 'f4m':
  32                 formats.extend(self._extract_f4m_formats(
  33                     vurl + '?hdcore=3.3.0', video_id,
  34                     f4m_id=player_type, fatal=False))
  35             elif ext == 'mpd':
  36                 if player_type == 'dashhbbtv':
  37                     formats.extend(self._extract_mpd_formats(
  38                         vurl, video_id, mpd_id=player_type, fatal=False))
  39             else:
  40                 formats.append({
  41                     'format_id': player_type,
  42                     'url': vurl,
  43                 })
  44         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  45             self.raise_geo_restricted(
  46                 'This video is only available in Sweden',
  47                 countries=self._GEO_COUNTRIES)
  48         self._sort_formats(formats)
  49
  50         subtitles = {}
  51         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  52         if isinstance(subtitle_references, list):
  53             for sr in subtitle_references:
  54                 subtitle_url = sr.get('url')
  55                 subtitle_lang = sr.get('language', 'sv')
  56                 if subtitle_url:
  57                     if determine_ext(subtitle_url) == 'm3u8':
  58                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  59                         continue
  60
  61                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  62
  63         title = video_info.get('title')
  64
  65         series = video_info.get('programTitle')
  66         season_number = int_or_none(video_info.get('season'))
  67         episode = video_info.get('episodeTitle')
  68         episode_number = int_or_none(video_info.get('episodeNumber'))
  69
  70         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  71         age_limit = None
  72         adult = dict_get(
  73             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  74             skip_false_values=False)
  75         if adult is not None:
  76             age_limit = 18 if adult else 0
  77
  78         return {
  79             'id': video_id,
  80             'title': title,
  81             'formats': formats,
  82             'subtitles': subtitles,
  83             'duration': duration,
  84             'age_limit': age_limit,
  85             'series': series,
  86             'season_number': season_number,
  87             'episode': episode,
  88             'episode_number': episode_number,
  89         }
  90
  91
  92 class SVTIE(SVTBaseIE):
  93     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
  94     _TEST = {
  95         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
  96         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
  97         'info_dict': {
  98             'id': '2900353',
  99             'ext': 'mp4',
 100             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 101             'duration': 27,
 102             'age_limit': 0,
 103         },
 104     }
 105
 106     @staticmethod
 107     def _extract_url(webpage):
 108         mobj = re.search(
 109             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 110         if mobj:
 111             return mobj.group('url')
 112
 113     def _real_extract(self, url):
 114         mobj = re.match(self._VALID_URL, url)
 115         widget_id = mobj.group('widget_id')
 116         article_id = mobj.group('id')
 117
 118         info = self._download_json(
 119             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 120             article_id)
 121
 122         info_dict = self._extract_video(info['video'], article_id)
 123         info_dict['title'] = info['context']['title']
 124         return info_dict
 125
 126
 127 class SVTPlayIE(SVTBaseIE):
 128     IE_DESC = 'SVT Play and Öppet arkiv'
 129     _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp)/(?P<id>[0-9]+)'
 130     _TESTS = [{
 131         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 132         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 133         'info_dict': {
 134             'id': '5996901',
 135             'ext': 'mp4',
 136             'title': 'Flygplan till Haile Selassie',
 137             'duration': 3527,
 138             'thumbnail': r're:^https?://.*[\.-]jpg$',
 139             'age_limit': 0,
 140             'subtitles': {
 141                 'sv': [{
 142                     'ext': 'wsrt',
 143                 }]
 144             },
 145         },
 146     }, {
 147         # geo restricted to Sweden
 148         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 149         'only_matching': True,
 150     }, {
 151         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 152         'only_matching': True,
 153     }]
 154
 155     def _real_extract(self, url):
 156         video_id = self._match_id(url)
 157
 158         webpage = self._download_webpage(url, video_id)
 159
 160         data = self._parse_json(
 161             self._search_regex(
 162                 r'root\["__svtplay"\]\s*=\s*([^;]+);',
 163                 webpage, 'embedded data', default='{}'),
 164             video_id, fatal=False)
 165
 166         thumbnail = self._og_search_thumbnail(webpage)
 167
 168         if data:
 169             video_info = try_get(
 170                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 171                 dict)
 172             if video_info:
 173                 info_dict = self._extract_video(video_info, video_id)
 174                 info_dict.update({
 175                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 176                     'thumbnail': thumbnail,
 177                 })
 178                 return info_dict
 179
 180         video_id = self._search_regex(
 181             r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 182             webpage, 'video id', default=None)
 183
 184         if video_id:
 185             data = self._download_json(
 186                 'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 187                 video_id, headers=self.geo_verification_headers())
 188             info_dict = self._extract_video(data, video_id)
 189             if not info_dict.get('title'):
 190                 info_dict['title'] = re.sub(
 191                     r'\s*\|\s*.+?$', '',
 192                     info_dict.get('episode') or self._og_search_title(webpage))
 193             return info_dict
 194
 195
 196 class SVTPlaylistIE(InfoExtractor):
 197     IE_DESC = 'SVT Play serie'
 198     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)'
 199     IE_NAME = 'svtplay:serie'
 200     _TESTS = [{
 201         'url': 'https://www.svtplay.se/rederiet',
 202         'info_dict': {
 203             'id': 'rederiet',
 204             'title': 'Rederiet',
 205             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 206         },
 207         'playlist_mincount': 318,
 208     }]
 209
 210     @classmethod
 211     def suitable(cls, url):
 212         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPlaylistIE, cls).suitable(url)
 213
 214     def _real_extract(self, url):
 215         video_id = self._match_id(url)
 216
 217         page = self._download_webpage(
 218             url, video_id,
 219             note='Downloading serie page',
 220             errnote='unable to fetch serie page')
 221
 222         root_json = self._search_regex(
 223             r'root\[\'__svtplay\'\]\s*=(.+);\n',
 224             page, 'root')
 225         root = self._parse_json(root_json, video_id)
 226
 227         metadata = root.get('metaData', {})
 228         related_videos_accordion = root['relatedVideoContent']['relatedVideosAccordion']
 229
 230         entries = []
 231         for season in related_videos_accordion:
 232             videos = season.get('videos')
 233             if not isinstance(videos, list):
 234                 continue
 235
 236             for video in videos:
 237                 content_url = video.get('contentUrl')
 238                 if not isinstance(content_url, compat_str):
 239                     continue
 240                 entries.append(
 241                     self.url_result(
 242                         urljoin(url, content_url),
 243                         ie=SVTPlayIE.ie_key(),
 244                         video_title=video.get('title')
 245                     ))
 246
 247         return self.playlist_result(
 248             entries, video_id, metadata.get('title'), metadata.get('description'))