_ 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 )
  11
  12
  13 class SVTBaseIE(InfoExtractor):
  14     def _extract_video(self, info, video_id):
  15         video_info = self._get_video_info(info)
  16
  17         formats = []
  18         for vr in video_info['videoReferences']:
  19             player_type = vr.get('playerType')
  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         self._sort_formats(formats)
  41
  42         subtitles = {}
  43         subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
  44         if isinstance(subtitle_references, list):
  45             for sr in subtitle_references:
  46                 subtitle_url = sr.get('url')
  47                 subtitle_lang = sr.get('language', 'sv')
  48                 if subtitle_url:
  49                     if determine_ext(subtitle_url) == 'm3u8':
  50                         # TODO(yan12125): handle WebVTT in m3u8 manifests
  51                         continue
  52
  53                     subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
  54
  55         duration = video_info.get('materialLength')
  56         age_limit = 18 if video_info.get('inappropriateForChildren') else 0
  57
  58         return {
  59             'id': video_id,
  60             'formats': formats,
  61             'subtitles': subtitles,
  62             'duration': duration,
  63             'age_limit': age_limit,
  64         }
  65
  66
  67 class SVTIE(SVTBaseIE):
  68     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
  69     _TEST = {
  70         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
  71         'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
  72         'info_dict': {
  73             'id': '2900353',
  74             'ext': 'mp4',
  75             'title': 'Stjärnorna skojar till det - under SVT-intervjun',
  76             'duration': 27,
  77             'age_limit': 0,
  78         },
  79     }
  80
  81     @staticmethod
  82     def _extract_url(webpage):
  83         mobj = re.search(
  84             r'(?:<iframe src|href)="(?P<url>%s[^"]*)"' % SVTIE._VALID_URL, webpage)
  85         if mobj:
  86             return mobj.group('url')
  87
  88     def _get_video_info(self, info):
  89         return info['video']
  90
  91     def _real_extract(self, url):
  92         mobj = re.match(self._VALID_URL, url)
  93         widget_id = mobj.group('widget_id')
  94         article_id = mobj.group('id')
  95
  96         info = self._download_json(
  97             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
  98             article_id)
  99
 100         info_dict = self._extract_video(info, article_id)
 101         info_dict['title'] = info['context']['title']
 102         return info_dict
 103
 104
 105 class SVTPlayIE(SVTBaseIE):
 106     IE_DESC = 'SVT Play and Öppet arkiv'
 107     _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
 108     _TEST = {
 109         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 110         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 111         'info_dict': {
 112             'id': '5996901',
 113             'ext': 'mp4',
 114             'title': 'Flygplan till Haile Selassie',
 115             'duration': 3527,
 116             'thumbnail': 're:^https?://.*[\.-]jpg$',
 117             'age_limit': 0,
 118             'subtitles': {
 119                 'sv': [{
 120                     'ext': 'wsrt',
 121                 }]
 122             },
 123         },
 124     }
 125
 126     def _get_video_info(self, info):
 127         return info['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video']
 128
 129     def _real_extract(self, url):
 130         video_id = self._match_id(url)
 131
 132         webpage = self._download_webpage(url, video_id)
 133
 134         data = self._parse_json(self._search_regex(
 135             r'root\["__svtplay"\]\s*=\s*([^;]+);', webpage, 'embedded data'), video_id)
 136
 137         thumbnail = self._og_search_thumbnail(webpage)
 138
 139         info_dict = self._extract_video(data, video_id)
 140         info_dict.update({
 141             'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
 142             'thumbnail': thumbnail,
 143         })
 144
 145         return info_dict