Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/discoverygo.py

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