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 )
  10
  11
  12 class SVTBaseIE(InfoExtractor):
  13     def _extract_video(self, url, video_id):
  14         info = self._download_json(url, video_id)
  15
  16         title = info['context']['title']
  17         thumbnail = info['context'].get('thumbnailImage')
  18
  19         video_info = info['video']
  20         formats = []
  21         for vr in video_info['videoReferences']:
  22             player_type = vr.get('playerType')
  23             vurl = vr['url']
  24             ext = determine_ext(vurl)
  25             if ext == 'm3u8':
  26                 formats.extend(self._extract_m3u8_formats(
  27                     vurl, video_id,
  28                     ext='mp4', entry_protocol='m3u8_native',
  29                     m3u8_id=player_type, fatal=False))
  30             elif ext == 'f4m':
  31                 formats.extend(self._extract_f4m_formats(
  32                     vurl + '?hdcore=3.3.0', video_id,
  33                     f4m_id=player_type, fatal=False))
  34             elif ext == 'mpd':
  35                 if player_type == 'dashhbbtv':
  36                     formats.extend(self._extract_mpd_formats(
  37                         vurl, video_id, mpd_id=player_type, fatal=False))
  38             else:
  39                 formats.append({
  40                     'format_id': player_type,
  41                     'url': vurl,
  42                 })
  43         self._sort_formats(formats)
  44
  45         subtitles = {}
  46         subtitle_references = video_info.get('subtitleReferences')
  47         if isinstance(subtitle_references, list):
  48             for sr in subtitle_references:
  49                 subtitle_url = sr.get('url')
  50                 if subtitle_url:
  51                     subtitles.setdefault('sv', []).append({'url': subtitle_url})
  52
  53         duration = video_info.get('materialLength')
  54         age_limit = 18 if video_info.get('inappropriateForChildren') else 0
  55
  56         return {
  57             'id': video_id,
  58             'title': title,
  59             'formats': formats,
  60             'subtitles': subtitles,
  61             'thumbnail': thumbnail,
  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': '9648197555fc1b49e3dc22db4af51d46',
  72         'info_dict': {
  73             'id': '2900353',
  74             'ext': 'flv',
  75             'title': 'Här trycker Jagr till Giroux (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 _real_extract(self, url):
  89         mobj = re.match(self._VALID_URL, url)
  90         widget_id = mobj.group('widget_id')
  91         article_id = mobj.group('id')
  92         return self._extract_video(
  93             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
  94             article_id)
  95
  96
  97 class SVTPlayIE(SVTBaseIE):
  98     IE_DESC = 'SVT Play and Öppet arkiv'
  99     _VALID_URL = r'https?://(?:www\.)?(?P<host>svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
 100     _TEST = {
 101         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
 102         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
 103         'info_dict': {
 104             'id': '5996901',
 105             'ext': 'mp4',
 106             'title': 'Flygplan till Haile Selassie',
 107             'duration': 3527,
 108             'thumbnail': 're:^https?://.*[\.-]jpg$',
 109             'age_limit': 0,
 110             'subtitles': {
 111                 'sv': [{
 112                     'ext': 'wsrt',
 113                 }]
 114             },
 115         },
 116     }
 117
 118     def _real_extract(self, url):
 119         mobj = re.match(self._VALID_URL, url)
 120         video_id = mobj.group('id')
 121         host = mobj.group('host')
 122         return self._extract_video(
 123             'http://www.%s.se/video/%s?output=json' % (host, video_id),
 124             video_id)