yt_dlp/extractor/curiositystream.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import ExtractorError, int_or_none, urlencode_postdata
   6
   7
   8 class CuriosityStreamBaseIE(InfoExtractor):
   9     _NETRC_MACHINE = 'curiositystream'
  10     _auth_token = None
  11
  12     def _handle_errors(self, result):
  13         error = result.get('error', {}).get('message')
  14         if error:
  15             if isinstance(error, dict):
  16                 error = ', '.join(error.values())
  17             raise ExtractorError(
  18                 f'{self.IE_NAME} said: {error}', expected=True)
  19
  20     def _call_api(self, path, video_id, query=None):
  21         headers = {}
  22         if not self._auth_token:
  23             auth_cookie = self._get_cookies('https://curiositystream.com').get('auth_token')
  24             if auth_cookie:
  25                 self.write_debug('Obtained auth_token cookie')
  26                 self._auth_token = urllib.parse.unquote(auth_cookie.value)
  27         if self._auth_token:
  28             headers['X-Auth-Token'] = self._auth_token
  29         result = self._download_json(
  30             self._API_BASE_URL + path, video_id, headers=headers, query=query)
  31         self._handle_errors(result)
  32         return result['data']
  33
  34     def _perform_login(self, username, password):
  35         result = self._download_json(
  36             'https://api.curiositystream.com/v1/login', None,
  37             note='Logging in', data=urlencode_postdata({
  38                 'email': username,
  39                 'password': password,
  40             }))
  41         self._handle_errors(result)
  42         CuriosityStreamBaseIE._auth_token = result['message']['auth_token']
  43
  44
  45 class CuriosityStreamIE(CuriosityStreamBaseIE):
  46     IE_NAME = 'curiositystream'
  47     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/video/(?P<id>\d+)'
  48     _TESTS = [{
  49         'url': 'http://app.curiositystream.com/video/2',
  50         'info_dict': {
  51             'id': '2',
  52             'ext': 'mp4',
  53             'title': 'How Did You Develop The Internet?',
  54             'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
  55             'channel': 'Curiosity Stream',
  56             'categories': ['Technology', 'Interview'],
  57             'average_rating': float,
  58             'series_id': '2',
  59             'thumbnail': r're:https://img.curiositystream.com/.+\.jpg',
  60             'tags': [],
  61             'duration': 158,
  62         },
  63         'params': {
  64             # m3u8 download
  65             'skip_download': True,
  66         },
  67     }]
  68
  69     _API_BASE_URL = 'https://api.curiositystream.com/v1/media/'
  70
  71     def _real_extract(self, url):
  72         video_id = self._match_id(url)
  73
  74         formats = []
  75         for encoding_format in ('m3u8', 'mpd'):
  76             media = self._call_api(video_id, video_id, query={
  77                 'encodingsNew': 'true',
  78                 'encodingsFormat': encoding_format,
  79             })
  80             for encoding in media.get('encodings', []):
  81                 playlist_url = encoding.get('master_playlist_url')
  82                 if encoding_format == 'm3u8':
  83                     # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
  84                     formats.extend(self._extract_m3u8_formats(
  85                         playlist_url, video_id, 'mp4',
  86                         m3u8_id='hls', fatal=False))
  87                 elif encoding_format == 'mpd':
  88                     formats.extend(self._extract_mpd_formats(
  89                         playlist_url, video_id, mpd_id='dash', fatal=False))
  90                 encoding_url = encoding.get('url')
  91                 file_url = encoding.get('file_url')
  92                 if not encoding_url and not file_url:
  93                     continue
  94                 f = {
  95                     'width': int_or_none(encoding.get('width')),
  96                     'height': int_or_none(encoding.get('height')),
  97                     'vbr': int_or_none(encoding.get('video_bitrate')),
  98                     'abr': int_or_none(encoding.get('audio_bitrate')),
  99                     'filesize': int_or_none(encoding.get('size_in_bytes')),
 100                     'vcodec': encoding.get('video_codec'),
 101                     'acodec': encoding.get('audio_codec'),
 102                     'container': encoding.get('container_type'),
 103                 }
 104                 for f_url in (encoding_url, file_url):
 105                     if not f_url:
 106                         continue
 107                     fmt = f.copy()
 108                     rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
 109                     if rtmp:
 110                         fmt.update({
 111                             'url': rtmp.group('url'),
 112                             'play_path': rtmp.group('playpath'),
 113                             'app': rtmp.group('app'),
 114                             'ext': 'flv',
 115                             'format_id': 'rtmp',
 116                         })
 117                     else:
 118                         fmt.update({
 119                             'url': f_url,
 120                             'format_id': 'http',
 121                         })
 122                     formats.append(fmt)
 123
 124         title = media['title']
 125
 126         subtitles = {}
 127         for closed_caption in media.get('closed_captions', []):
 128             sub_url = closed_caption.get('file')
 129             if not sub_url:
 130                 continue
 131             lang = closed_caption.get('code') or closed_caption.get('language') or 'en'
 132             subtitles.setdefault(lang, []).append({
 133                 'url': sub_url,
 134             })
 135
 136         return {
 137             'id': video_id,
 138             'formats': formats,
 139             'title': title,
 140             'description': media.get('description'),
 141             'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'),
 142             'duration': int_or_none(media.get('duration')),
 143             'tags': media.get('tags'),
 144             'subtitles': subtitles,
 145             'channel': media.get('producer'),
 146             'categories': [media.get('primary_category'), media.get('type')],
 147             'average_rating': media.get('rating_percentage'),
 148             'series_id': str(media.get('collection_id') or '') or None,
 149         }
 150
 151
 152 class CuriosityStreamCollectionBaseIE(CuriosityStreamBaseIE):
 153
 154     def _real_extract(self, url):
 155         collection_id = self._match_id(url)
 156         collection = self._call_api(collection_id, collection_id)
 157         entries = []
 158         for media in collection.get('media', []):
 159             media_id = str(media.get('id'))
 160             media_type, ie = ('series', CuriosityStreamSeriesIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
 161             entries.append(self.url_result(
 162                 f'https://curiositystream.com/{media_type}/{media_id}',
 163                 ie=ie.ie_key(), video_id=media_id))
 164         return self.playlist_result(
 165             entries, collection_id,
 166             collection.get('title'), collection.get('description'))
 167
 168
 169 class CuriosityStreamCollectionsIE(CuriosityStreamCollectionBaseIE):
 170     IE_NAME = 'curiositystream:collections'
 171     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/collections/(?P<id>\d+)'
 172     _API_BASE_URL = 'https://api.curiositystream.com/v2/collections/'
 173     _TESTS = [{
 174         'url': 'https://curiositystream.com/collections/86',
 175         'info_dict': {
 176             'id': '86',
 177             'title': 'Staff Picks',
 178             'description': 'Wondering where to start? Here are a few of our favorite series and films... from our couch to yours.',
 179         },
 180         'playlist_mincount': 7,
 181     }, {
 182         'url': 'https://curiositystream.com/collections/36',
 183         'only_matching': True,
 184     }]
 185
 186
 187 class CuriosityStreamSeriesIE(CuriosityStreamCollectionBaseIE):
 188     IE_NAME = 'curiositystream:series'
 189     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:series|collection)/(?P<id>\d+)'
 190     _API_BASE_URL = 'https://api.curiositystream.com/v2/series/'
 191     _TESTS = [{
 192         'url': 'https://curiositystream.com/series/2',
 193         'info_dict': {
 194             'id': '2',
 195             'title': 'Curious Minds: The Internet',
 196             'description': 'How is the internet shaping our lives in the 21st Century?',
 197         },
 198         'playlist_mincount': 16,
 199     }, {
 200         'url': 'https://curiositystream.com/collection/2',
 201         'only_matching': True,
 202     }]