yt_dlp/extractor/gaia.py

   1 from .common import InfoExtractor
   2 from ..compat import (
   3     compat_str,
   4     compat_urllib_parse_unquote,
   5 )
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     str_or_none,
  10     strip_or_none,
  11     try_get,
  12     urlencode_postdata,
  13 )
  14
  15
  16 class GaiaIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?gaia\.com/video/(?P<id>[^/?]+).*?\bfullplayer=(?P<type>feature|preview)'
  18     _TESTS = [{
  19         'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=feature',
  20         'info_dict': {
  21             'id': '89356',
  22             'ext': 'mp4',
  23             'title': 'Connecting with Universal Consciousness',
  24             'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
  25             'upload_date': '20151116',
  26             'timestamp': 1447707266,
  27             'duration': 936,
  28         },
  29         'params': {
  30             # m3u8 download
  31             'skip_download': True,
  32         },
  33     }, {
  34         'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=preview',
  35         'info_dict': {
  36             'id': '89351',
  37             'ext': 'mp4',
  38             'title': 'Connecting with Universal Consciousness',
  39             'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
  40             'upload_date': '20151116',
  41             'timestamp': 1447707266,
  42             'duration': 53,
  43         },
  44         'params': {
  45             # m3u8 download
  46             'skip_download': True,
  47         },
  48     }]
  49     _NETRC_MACHINE = 'gaia'
  50     _jwt = None
  51
  52     def _real_initialize(self):
  53         auth = self._get_cookies('https://www.gaia.com/').get('auth')
  54         if auth:
  55             auth = self._parse_json(compat_urllib_parse_unquote(auth.value), None, fatal=False)
  56             self._jwt = auth.get('jwt')
  57
  58     def _perform_login(self, username, password):
  59         if self._jwt:
  60             return
  61         auth = self._download_json(
  62             'https://auth.gaia.com/v1/login',
  63             None, data=urlencode_postdata({
  64                 'username': username,
  65                 'password': password
  66             }))
  67         if auth.get('success') is False:
  68             raise ExtractorError(', '.join(auth['messages']), expected=True)
  69         self._jwt = auth.get('jwt')
  70
  71     def _real_extract(self, url):
  72         display_id, vtype = self._match_valid_url(url).groups()
  73         node_id = self._download_json(
  74             'https://brooklyn.gaia.com/pathinfo', display_id, query={
  75                 'path': 'video/' + display_id,
  76             })['id']
  77         node = self._download_json(
  78             'https://brooklyn.gaia.com/node/%d' % node_id, node_id)
  79         vdata = node[vtype]
  80         media_id = compat_str(vdata['nid'])
  81         title = node['title']
  82
  83         headers = None
  84         if self._jwt:
  85             headers = {'Authorization': 'Bearer ' + self._jwt}
  86         media = self._download_json(
  87             'https://brooklyn.gaia.com/media/' + media_id,
  88             media_id, headers=headers)
  89         formats = self._extract_m3u8_formats(
  90             media['mediaUrls']['bcHLS'], media_id, 'mp4')
  91
  92         subtitles = {}
  93         text_tracks = media.get('textTracks', {})
  94         for key in ('captions', 'subtitles'):
  95             for lang, sub_url in text_tracks.get(key, {}).items():
  96                 subtitles.setdefault(lang, []).append({
  97                     'url': sub_url,
  98                 })
  99
 100         fivestar = node.get('fivestar', {})
 101         fields = node.get('fields', {})
 102
 103         def get_field_value(key, value_key='value'):
 104             return try_get(fields, lambda x: x[key][0][value_key])
 105
 106         return {
 107             'id': media_id,
 108             'display_id': display_id,
 109             'title': title,
 110             'formats': formats,
 111             'description': strip_or_none(get_field_value('body') or get_field_value('teaser')),
 112             'timestamp': int_or_none(node.get('created')),
 113             'subtitles': subtitles,
 114             'duration': int_or_none(vdata.get('duration')),
 115             'like_count': int_or_none(try_get(fivestar, lambda x: x['up_count']['value'])),
 116             'dislike_count': int_or_none(try_get(fivestar, lambda x: x['down_count']['value'])),
 117             'comment_count': int_or_none(node.get('comment_count')),
 118             'series': try_get(node, lambda x: x['series']['title'], compat_str),
 119             'season_number': int_or_none(get_field_value('season')),
 120             'season_id': str_or_none(get_field_value('series_nid', 'nid')),
 121             'episode_number': int_or_none(get_field_value('episode')),
 122         }