_ 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 ..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         is_live = dict_get(video_info, ('live', 'simulcast'), default=False)
  26         m3u8_protocol = 'm3u8' if is_live else 'm3u8_native'
  27         formats = []
  28         for vr in video_info['videoReferences']:
  29             player_type = vr.get('playerType') or vr.get('format')
  30             vurl = vr['url']
  31             ext = determine_ext(vurl)
  32             if ext == 'm3u8':
  33                 formats.extend(self._extract_m3u8_formats(
  34                     vurl, video_id,
  35                     ext='mp4', entry_protocol=m3u8_protocol,
  36                     m3u8_id=player_type, fatal=False))
  37             elif ext == 'f4m':
  38                 formats.extend(self._extract_f4m_formats(
  39                     vurl + '?hdcore=3.3.0', video_id,
  40                     f4m_id=player_type, fatal=False))
  41             elif ext == 'mpd':
  42                 if player_type == 'dashhbbtv':
  43                     formats.extend(self._extract_mpd_formats(
  44                         vurl, video_id, mpd_id=player_type, fatal=False))
  45             else:
  46                 formats.append({
  47                     'format_id': player_type,
  48                     'url': vurl,
  49                 })
  50         if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
  51             self.raise_geo_restricted(
  52                 'This video is only available in Sweden',
  53                 countries=self._GEO_COUNTRIES)
  54         self._sort_formats(formats)
  55
  56         subtitles = {}
  57         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  58         if isinstance(subtitle_references, list):
  59             for sr in subtitle_references:
  60                 subtitle_url = sr.get('url')
  61                 subtitle_lang = sr.get('language', 'sv')
  62                 if subtitle_url:
  63                     if determine_ext(subtitle_url) == 'm3u8':
  64                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  65                         continue
  66
  67                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  68
  69         title = video_info.get('title')
  70
  71         series = video_info.get('programTitle')
  72         season_number = int_or_none(video_info.get('season'))
  73         episode = video_info.get('episodeTitle')
  74         episode_number = int_or_none(video_info.get('episodeNumber'))
  75
  76         duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
  77         age_limit = None
  78         adult = dict_get(
  79             video_info, ('inappropriateForChildren', 'blockedForChildren'),
  80             skip_false_values=False)
  81         if adult is not None:
  82             age_limit = 18 if adult else 0
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'formats': formats,
  88             'subtitles': subtitles,
  89             'duration': duration,
  90             'age_limit': age_limit,
  91             'series': series,
  92             'season_number': season_number,
  93             'episode': episode,
  94             'episode_number': episode_number,
  95             'is_live': is_live,
  96         }
  97
  98
  99 class SVTIE(SVTBaseIE):
 100     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
 101     _TEST = {
 102         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
 103         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
 104         'info_dict': {
 105             'id': '2900353',
 106             'ext': 'mp4',
 107             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
 108             'duration': 27,
 109             'age_limit': 0,
 110         },
 111     }
 112
 113     @staticmethod
 114     def _extract_url(webpage):
 115         mobj = re.search(
 116             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
 117         if mobj:
 118             return mobj.group('url')
 119
 120     def _real_extract(self, url):
 121         mobj = re.match(self._VALID_URL, url)
 122         widget_id = mobj.group('widget_id')
 123         article_id = mobj.group('id')
 124
 125         info = self._download_json(
 126             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
 127             article_id)
 128
 129         info_dict = self._extract_video(info['video'], article_id)
 130         info_dict['title'] = info['context']['title']
 131         return info_dict
 132
 133
 134 class SVTPlayBaseIE(SVTBaseIE):
 135     _SVTPLAY_RE = r'root\s*\[\s*(["\'])_*svtplay\1\s*\]\s*=\s*(?P<json>{.+?})\s*;\s*\n'
 136
 137
 138 class SVTPlayIE(SVTPlayBaseIE):
 139     IE_DESC = 'SVT Play and Öppet arkiv'
 140     _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+)'
 141     _TESTS = [{
 142         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 143         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 144         'info_dict': {
 145             'id': '5996901',
 146             'ext': 'mp4',
 147             'title': 'Flygplan till Haile Selassie',
 148             'duration': 3527,
 149             'thumbnail': r're:^https?://.*[\.-]jpg$',
 150             'age_limit': 0,
 151             'subtitles': {
 152                 'sv': [{
 153                     'ext': 'wsrt',
 154                 }]
 155             },
 156         },
 157     }, {
 158         # geo restricted to Sweden
 159         'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
 160         'only_matching': True,
 161     }, {
 162         'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
 163         'only_matching': True,
 164     }, {
 165         'url': 'https://www.svtplay.se/kanaler/svt1',
 166         'only_matching': True,
 167     }]
 168
 169     def _real_extract(self, url):
 170         video_id = self._match_id(url)
 171
 172         webpage = self._download_webpage(url, video_id)
 173
 174         data = self._parse_json(
 175             self._search_regex(
 176                 self._SVTPLAY_RE, webpage, 'embedded data', default='{}',
 177                 group='json'),
 178             video_id, fatal=False)
 179
 180         thumbnail = self._og_search_thumbnail(webpage)
 181
 182         def adjust_title(info):
 183             if info['is_live']:
 184                 info['title'] = self._live_title(info['title'])
 185
 186         if data:
 187             video_info = try_get(
 188                 data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
 189                 dict)
 190             if video_info:
 191                 info_dict = self._extract_video(video_info, video_id)
 192                 info_dict.update({
 193                     'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 194                     'thumbnail': thumbnail,
 195                 })
 196                 adjust_title(info_dict)
 197                 return info_dict
 198
 199         video_id = self._search_regex(
 200             r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
 201             webpage, 'video id', default=None)
 202
 203         if video_id:
 204             data = self._download_json(
 205                 'https://api.svt.se/videoplayer-api/video/%s' % video_id,
 206                 video_id, headers=self.geo_verification_headers())
 207             info_dict = self._extract_video(data, video_id)
 208             if not info_dict.get('title'):
 209                 info_dict['title'] = re.sub(
 210                     r'\s*\|\s*.+?$', '',
 211                     info_dict.get('episode') or self._og_search_title(webpage))
 212             adjust_title(info_dict)
 213             return info_dict
 214
 215
 216 class SVTSeriesIE(SVTPlayBaseIE):
 217     _VALID_URL = r'https?://(?:www\.)?svtplay\.se/(?P<id>[^/?&#]+)'
 218     _TESTS = [{
 219         'url': 'https://www.svtplay.se/rederiet',
 220         'info_dict': {
 221             'id': 'rederiet',
 222             'title': 'Rederiet',
 223             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 224         },
 225         'playlist_mincount': 318,
 226     }, {
 227         'url': 'https://www.svtplay.se/rederiet?tab=sasong2',
 228         'info_dict': {
 229             'id': 'rederiet-sasong2',
 230             'title': 'Rederiet - Säsong 2',
 231             'description': 'md5:505d491a58f4fcf6eb418ecab947e69e',
 232         },
 233         'playlist_count': 12,
 234     }]
 235
 236     @classmethod
 237     def suitable(cls, url):
 238         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTSeriesIE, cls).suitable(url)
 239
 240     def _real_extract(self, url):
 241         series_id = self._match_id(url)
 242
 243         qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
 244         season_slug = qs.get('tab', [None])[0]
 245
 246         if season_slug:
 247             series_id += '-%s' % season_slug
 248
 249         webpage = self._download_webpage(
 250             url, series_id, 'Downloading series page')
 251
 252         root = self._parse_json(
 253             self._search_regex(
 254                 self._SVTPLAY_RE, webpage, 'content', group='json'),
 255             series_id)
 256
 257         season_name = None
 258
 259         entries = []
 260         for season in root['relatedVideoContent']['relatedVideosAccordion']:
 261             if not isinstance(season, dict):
 262                 continue
 263             if season_slug:
 264                 if season.get('slug') != season_slug:
 265                     continue
 266                 season_name = season.get('name')
 267             videos = season.get('videos')
 268             if not isinstance(videos, list):
 269                 continue
 270             for video in videos:
 271                 content_url = video.get('contentUrl')
 272                 if not content_url or not isinstance(content_url, compat_str):
 273                     continue
 274                 entries.append(
 275                     self.url_result(
 276                         urljoin(url, content_url),
 277                         ie=SVTPlayIE.ie_key(),
 278                         video_title=video.get('title')
 279                     ))
 280
 281         metadata = root.get('metaData')
 282         if not isinstance(metadata, dict):
 283             metadata = {}
 284
 285         title = metadata.get('title')
 286         season_name = season_name or season_slug
 287
 288         if title and season_name:
 289             title = '%s - %s' % (title, season_name)
 290         elif season_slug:
 291             title = season_slug
 292
 293         return self.playlist_result(
 294             entries, series_id, title, metadata.get('description'))