_ Git - youtube-dl/blob - youtube_dl/extractor/discoverygo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     extract_attributes,
   9     ExtractorError,
  10     int_or_none,
  11     parse_age_limit,
  12     remove_end,
  13     unescapeHTML,
  14 )
  15
  16
  17 class DiscoveryGoBaseIE(InfoExtractor):
  18     _VALID_URL_TEMPLATE = r'''(?x)https?://(?:www\.)?(?:
  19             discovery|
  20             investigationdiscovery|
  21             discoverylife|
  22             animalplanet|
  23             ahctv|
  24             destinationamerica|
  25             sciencechannel|
  26             tlc|
  27             velocitychannel
  28         )go\.com/%s(?P<id>[^/?#&]+)'''
  29
  30     def _extract_video_info(self, video, stream, display_id):
  31         title = video['name']
  32
  33         if not stream:
  34             if video.get('authenticated') is True:
  35                 raise ExtractorError(
  36                     'This video is only available via cable service provider subscription that'
  37                     ' is not currently supported. You may want to use --cookies.', expected=True)
  38             else:
  39                 raise ExtractorError('Unable to find stream')
  40         STREAM_URL_SUFFIX = 'streamUrl'
  41         formats = []
  42         for stream_kind in ('', 'hds'):
  43             suffix = STREAM_URL_SUFFIX.capitalize() if stream_kind else STREAM_URL_SUFFIX
  44             stream_url = stream.get('%s%s' % (stream_kind, suffix))
  45             if not stream_url:
  46                 continue
  47             if stream_kind == '':
  48                 formats.extend(self._extract_m3u8_formats(
  49                     stream_url, display_id, 'mp4', entry_protocol='m3u8_native',
  50                     m3u8_id='hls', fatal=False))
  51             elif stream_kind == 'hds':
  52                 formats.extend(self._extract_f4m_formats(
  53                     stream_url, display_id, f4m_id=stream_kind, fatal=False))
  54         self._sort_formats(formats)
  55
  56         video_id = video.get('id') or display_id
  57         description = video.get('description', {}).get('detailed')
  58         duration = int_or_none(video.get('duration'))
  59
  60         series = video.get('show', {}).get('name')
  61         season_number = int_or_none(video.get('season', {}).get('number'))
  62         episode_number = int_or_none(video.get('episodeNumber'))
  63
  64         tags = video.get('tags')
  65         age_limit = parse_age_limit(video.get('parental', {}).get('rating'))
  66
  67         subtitles = {}
  68         captions = stream.get('captions')
  69         if isinstance(captions, list):
  70             for caption in captions:
  71                 subtitle_url = caption.get('fileUrl')
  72                 if (not subtitle_url or not isinstance(subtitle_url, compat_str) or
  73                         not subtitle_url.startswith('http')):
  74                     continue
  75                 lang = caption.get('fileLang', 'en')
  76                 subtitles.setdefault(lang, []).append({'url': subtitle_url})
  77
  78         return {
  79             'id': video_id,
  80             'display_id': display_id,
  81             'title': title,
  82             'description': description,
  83             'duration': duration,
  84             'series': series,
  85             'season_number': season_number,
  86             'episode_number': episode_number,
  87             'tags': tags,
  88             'age_limit': age_limit,
  89             'formats': formats,
  90             'subtitles': subtitles,
  91         }
  92
  93
  94 class DiscoveryGoIE(DiscoveryGoBaseIE):
  95     _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % r'(?:[^/]+/)+'
  96     _GEO_COUNTRIES = ['US']
  97     _TEST = {
  98         'url': 'https://www.discoverygo.com/bering-sea-gold/reaper-madness/',
  99         'info_dict': {
 100             'id': '58c167d86b66d12f2addeb01',
 101             'ext': 'mp4',
 102             'title': 'Reaper Madness',
 103             'description': 'md5:09f2c625c99afb8946ed4fb7865f6e78',
 104             'duration': 2519,
 105             'series': 'Bering Sea Gold',
 106             'season_number': 8,
 107             'episode_number': 6,
 108             'age_limit': 14,
 109         },
 110     }
 111
 112     def _real_extract(self, url):
 113         display_id = self._match_id(url)
 114
 115         webpage = self._download_webpage(url, display_id)
 116
 117         container = extract_attributes(
 118             self._search_regex(
 119                 r'(<div[^>]+class=["\']video-player-container[^>]+>)',
 120                 webpage, 'video container'))
 121
 122         video = self._parse_json(
 123             container.get('data-video') or container.get('data-json'),
 124             display_id)
 125
 126         stream = video.get('stream')
 127
 128         return self._extract_video_info(video, stream, display_id)
 129
 130
 131 class DiscoveryGoPlaylistIE(DiscoveryGoBaseIE):
 132     _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % ''
 133     _TEST = {
 134         'url': 'https://www.discoverygo.com/bering-sea-gold/',
 135         'info_dict': {
 136             'id': 'bering-sea-gold',
 137             'title': 'Bering Sea Gold',
 138             'description': 'md5:cc5c6489835949043c0cc3ad66c2fa0e',
 139         },
 140         'playlist_mincount': 6,
 141     }
 142
 143     @classmethod
 144     def suitable(cls, url):
 145         return False if DiscoveryGoIE.suitable(url) else super(
 146             DiscoveryGoPlaylistIE, cls).suitable(url)
 147
 148     def _real_extract(self, url):
 149         display_id = self._match_id(url)
 150
 151         webpage = self._download_webpage(url, display_id)
 152
 153         entries = []
 154         for mobj in re.finditer(r'data-json=(["\'])(?P<json>{.+?})\1', webpage):
 155             data = self._parse_json(
 156                 mobj.group('json'), display_id,
 157                 transform_source=unescapeHTML, fatal=False)
 158             if not isinstance(data, dict) or data.get('type') != 'episode':
 159                 continue
 160             episode_url = data.get('socialUrl')
 161             if not episode_url:
 162                 continue
 163             entries.append(self.url_result(
 164                 episode_url, ie=DiscoveryGoIE.ie_key(),
 165                 video_id=data.get('id')))
 166
 167         return self.playlist_result(
 168             entries, display_id,
 169             remove_end(self._og_search_title(
 170                 webpage, fatal=False), ' | Discovery GO'),
 171             self._og_search_description(webpage))