yt_dlp/extractor/gaia.py

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