yt_dlp/extractor/gaia.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_str,
   8     compat_urllib_parse_unquote,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13     str_or_none,
  14     strip_or_none,
  15     try_get,
  16     urlencode_postdata,
  17 )
  18
  19
  20 class GaiaIE(InfoExtractor):
  21     _VALID_URL = r'https?://(?:www\.)?gaia\.com/video/(?P<id>[^/?]+).*?\bfullplayer=(?P<type>feature|preview)'
  22     _TESTS = [{
  23         'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=feature',
  24         'info_dict': {
  25             'id': '89356',
  26             'ext': 'mp4',
  27             'title': 'Connecting with Universal Consciousness',
  28             'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
  29             'upload_date': '20151116',
  30             'timestamp': 1447707266,
  31             'duration': 936,
  32         },
  33         'params': {
  34             # m3u8 download
  35             'skip_download': True,
  36         },
  37     }, {
  38         'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=preview',
  39         'info_dict': {
  40             'id': '89351',
  41             'ext': 'mp4',
  42             'title': 'Connecting with Universal Consciousness',
  43             'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
  44             'upload_date': '20151116',
  45             'timestamp': 1447707266,
  46             'duration': 53,
  47         },
  48         'params': {
  49             # m3u8 download
  50             'skip_download': True,
  51         },
  52     }]
  53     _NETRC_MACHINE = 'gaia'
  54     _jwt = None
  55
  56     def _real_initialize(self):
  57         auth = self._get_cookies('https://www.gaia.com/').get('auth')
  58         if auth:
  59             auth = self._parse_json(
  60                 compat_urllib_parse_unquote(auth.value),
  61                 None, fatal=False)
  62         if not auth:
  63             username, password = self._get_login_info()
  64             if username is None:
  65                 return
  66             auth = self._download_json(
  67                 'https://auth.gaia.com/v1/login',
  68                 None, data=urlencode_postdata({
  69                     'username': username,
  70                     'password': password
  71                 }))
  72             if auth.get('success') is False:
  73                 raise ExtractorError(', '.join(auth['messages']), expected=True)
  74         if auth:
  75             self._jwt = auth.get('jwt')
  76
  77     def _real_extract(self, url):
  78         display_id, vtype = self._match_valid_url(url).groups()
  79         node_id = self._download_json(
  80             'https://brooklyn.gaia.com/pathinfo', display_id, query={
  81                 'path': 'video/' + display_id,
  82             })['id']
  83         node = self._download_json(
  84             'https://brooklyn.gaia.com/node/%d' % node_id, node_id)
  85         vdata = node[vtype]
  86         media_id = compat_str(vdata['nid'])
  87         title = node['title']
  88
  89         headers = None
  90         if self._jwt:
  91             headers = {'Authorization': 'Bearer ' + self._jwt}
  92         media = self._download_json(
  93             'https://brooklyn.gaia.com/media/' + media_id,
  94             media_id, headers=headers)
  95         formats = self._extract_m3u8_formats(
  96             media['mediaUrls']['bcHLS'], media_id, 'mp4')
  97         self._sort_formats(formats)
  98
  99         subtitles = {}
 100         text_tracks = media.get('textTracks', {})
 101         for key in ('captions', 'subtitles'):
 102             for lang, sub_url in text_tracks.get(key, {}).items():
 103                 subtitles.setdefault(lang, []).append({
 104                     'url': sub_url,
 105                 })
 106
 107         fivestar = node.get('fivestar', {})
 108         fields = node.get('fields', {})
 109
 110         def get_field_value(key, value_key='value'):
 111             return try_get(fields, lambda x: x[key][0][value_key])
 112
 113         return {
 114             'id': media_id,
 115             'display_id': display_id,
 116             'title': title,
 117             'formats': formats,
 118             'description': strip_or_none(get_field_value('body') or get_field_value('teaser')),
 119             'timestamp': int_or_none(node.get('created')),
 120             'subtitles': subtitles,
 121             'duration': int_or_none(vdata.get('duration')),
 122             'like_count': int_or_none(try_get(fivestar, lambda x: x['up_count']['value'])),
 123             'dislike_count': int_or_none(try_get(fivestar, lambda x: x['down_count']['value'])),
 124             'comment_count': int_or_none(node.get('comment_count')),
 125             'series': try_get(node, lambda x: x['series']['title'], compat_str),
 126             'season_number': int_or_none(get_field_value('season')),
 127             'season_id': str_or_none(get_field_value('series_nid', 'nid')),
 128             'episode_number': int_or_none(get_field_value('episode')),
 129         }