yt_dlp/extractor/tunein.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     OnDemandPagedList,
   6     determine_ext,
   7     parse_iso8601,
   8     traverse_obj,
   9 )
  10
  11
  12 class TuneInBaseIE(InfoExtractor):
  13     _VALID_URL_BASE = r'https?://(?:www\.)?tunein\.com'
  14
  15     def _extract_metadata(self, webpage, content_id):
  16         return self._search_json(r'window.INITIAL_STATE=', webpage, 'hydration', content_id, fatal=False)
  17
  18     def _extract_formats_and_subtitles(self, content_id):
  19         streams = self._download_json(
  20             f'https://opml.radiotime.com/Tune.ashx?render=json&formats=mp3,aac,ogg,flash,hls&id={content_id}',
  21             content_id)['body']
  22
  23         formats, subtitles = [], {}
  24         for stream in streams:
  25             if stream.get('media_type') == 'hls':
  26                 fmts, subs = self._extract_m3u8_formats_and_subtitles(stream['url'], content_id, fatal=False)
  27                 formats.extend(fmts)
  28                 self._merge_subtitles(subs, target=subtitles)
  29             elif determine_ext(stream['url']) == 'pls':
  30                 playlist_content = self._download_webpage(stream['url'], content_id)
  31                 formats.append({
  32                     'url': self._search_regex(r'File1=(.*)', playlist_content, 'url', fatal=False),
  33                     'abr': stream.get('bitrate'),
  34                     'ext': stream.get('media_type'),
  35                 })
  36             else:
  37                 formats.append({
  38                     'url': stream['url'],
  39                     'abr': stream.get('bitrate'),
  40                     'ext': stream.get('media_type'),
  41                 })
  42
  43         return formats, subtitles
  44
  45
  46 class TuneInStationIE(TuneInBaseIE):
  47     _VALID_URL = TuneInBaseIE._VALID_URL_BASE + r'(?:/radio/[^?#]+-|/embed/player/)(?P<id>s\d+)'
  48     _EMBED_REGEX = [r'<iframe[^>]+src=["\'](?P<url>(?:https?://)?tunein\.com/embed/player/s\d+)']
  49
  50     _TESTS = [{
  51         'url': 'https://tunein.com/radio/Jazz24-885-s34682/',
  52         'info_dict': {
  53             'id': 's34682',
  54             'title': 're:^Jazz24',
  55             'description': 'md5:d6d0b89063fd68d529fa7058ee98619b',
  56             'thumbnail': 're:^https?://[^?&]+/s34682',
  57             'location': 'Seattle-Tacoma, US',
  58             'ext': 'mp3',
  59             'live_status': 'is_live',
  60         },
  61         'params': {
  62             'skip_download': True,
  63         },
  64     }, {
  65         'url': 'https://tunein.com/embed/player/s6404/',
  66         'only_matching': True,
  67     }, {
  68         'url': 'https://tunein.com/radio/BBC-Radio-1-988-s24939/',
  69         'info_dict': {
  70             'id': 's24939',
  71             'title': 're:^BBC Radio 1',
  72             'description': 'md5:f3f75f7423398d87119043c26e7bfb84',
  73             'thumbnail': 're:^https?://[^?&]+/s24939',
  74             'location': 'London, UK',
  75             'ext': 'mp3',
  76             'live_status': 'is_live',
  77         },
  78         'params': {
  79             'skip_download': True,
  80         },
  81     }]
  82
  83     def _real_extract(self, url):
  84         station_id = self._match_id(url)
  85
  86         webpage = self._download_webpage(url, station_id)
  87         metadata = self._extract_metadata(webpage, station_id)
  88
  89         formats, subtitles = self._extract_formats_and_subtitles(station_id)
  90         return {
  91             'id': station_id,
  92             'title': traverse_obj(metadata, ('profiles', station_id, 'title')),
  93             'description': traverse_obj(metadata, ('profiles', station_id, 'description')),
  94             'thumbnail': traverse_obj(metadata, ('profiles', station_id, 'image')),
  95             'timestamp': parse_iso8601(
  96                 traverse_obj(metadata, ('profiles', station_id, 'actions', 'play', 'publishTime'))),
  97             'location': traverse_obj(
  98                 metadata, ('profiles', station_id, 'metadata', 'properties', 'location', 'displayName'),
  99                 ('profiles', station_id, 'properties', 'location', 'displayName')),
 100             'formats': formats,
 101             'subtitles': subtitles,
 102             'is_live': traverse_obj(metadata, ('profiles', station_id, 'actions', 'play', 'isLive')),
 103         }
 104
 105
 106 class TuneInPodcastIE(TuneInBaseIE):
 107     _VALID_URL = TuneInBaseIE._VALID_URL_BASE + r'/(?:podcasts/[^?#]+-|embed/player/)(?P<id>p\d+)/?(?:#|$)'
 108     _EMBED_REGEX = [r'<iframe[^>]+src=["\'](?P<url>(?:https?://)?tunein\.com/embed/player/p\d+)']
 109
 110     _TESTS = [{
 111         'url': 'https://tunein.com/podcasts/Technology-Podcasts/Artificial-Intelligence-p1153019',
 112         'info_dict': {
 113             'id': 'p1153019',
 114             'title': 'Lex Fridman Podcast',
 115             'description': 'md5:bedc4e5f1c94f7dec6e4317b5654b00d',
 116         },
 117         'playlist_mincount': 200,
 118     }, {
 119         'url': 'https://tunein.com/embed/player/p191660/',
 120         'only_matching': True,
 121     }, {
 122         'url': 'https://tunein.com/podcasts/World-News/BBC-News-p14/',
 123         'info_dict': {
 124             'id': 'p14',
 125             'title': 'BBC News',
 126             'description': 'md5:1218e575eeaff75f48ed978261fa2068',
 127         },
 128         'playlist_mincount': 200,
 129     }]
 130
 131     _PAGE_SIZE = 30
 132
 133     def _real_extract(self, url):
 134         podcast_id = self._match_id(url)
 135
 136         webpage = self._download_webpage(url, podcast_id, fatal=False)
 137         metadata = self._extract_metadata(webpage, podcast_id)
 138
 139         def page_func(page_num):
 140             api_response = self._download_json(
 141                 f'https://api.tunein.com/profiles/{podcast_id}/contents', podcast_id,
 142                 note=f'Downloading page {page_num + 1}', query={
 143                     'filter': 't:free',
 144                     'offset': page_num * self._PAGE_SIZE,
 145                     'limit': self._PAGE_SIZE,
 146                 })
 147
 148             return [
 149                 self.url_result(
 150                     f'https://tunein.com/podcasts/{podcast_id}?topicId={episode["GuideId"][1:]}',
 151                     TuneInPodcastEpisodeIE, title=episode.get('Title'))
 152                 for episode in api_response['Items']]
 153
 154         entries = OnDemandPagedList(page_func, self._PAGE_SIZE)
 155         return self.playlist_result(
 156             entries, playlist_id=podcast_id, title=traverse_obj(metadata, ('profiles', podcast_id, 'title')),
 157             description=traverse_obj(metadata, ('profiles', podcast_id, 'description')))
 158
 159
 160 class TuneInPodcastEpisodeIE(TuneInBaseIE):
 161     _VALID_URL = TuneInBaseIE._VALID_URL_BASE + r'/podcasts/(?:[^?&]+-)?(?P<podcast_id>p\d+)/?\?topicId=(?P<id>\w\d+)'
 162
 163     _TESTS = [{
 164         'url': 'https://tunein.com/podcasts/Technology-Podcasts/Artificial-Intelligence-p1153019/?topicId=236404354',
 165         'info_dict': {
 166             'id': 't236404354',
 167             'title': '#351 \u2013 MrBeast: Future of YouTube, Twitter, TikTok, and Instagram',
 168             'description': 'md5:e1734db6f525e472c0c290d124a2ad77',
 169             'thumbnail': 're:^https?://[^?&]+/p1153019',
 170             'timestamp': 1673458571,
 171             'upload_date': '20230111',
 172             'series_id': 'p1153019',
 173             'series': 'Lex Fridman Podcast',
 174             'ext': 'mp3',
 175         },
 176     }]
 177
 178     def _real_extract(self, url):
 179         podcast_id, episode_id = self._match_valid_url(url).group('podcast_id', 'id')
 180         episode_id = f't{episode_id}'
 181
 182         webpage = self._download_webpage(url, episode_id)
 183         metadata = self._extract_metadata(webpage, episode_id)
 184
 185         formats, subtitles = self._extract_formats_and_subtitles(episode_id)
 186         return {
 187             'id': episode_id,
 188             'title': traverse_obj(metadata, ('profiles', episode_id, 'title')),
 189             'description': traverse_obj(metadata, ('profiles', episode_id, 'description')),
 190             'thumbnail': traverse_obj(metadata, ('profiles', episode_id, 'image')),
 191             'timestamp': parse_iso8601(
 192                 traverse_obj(metadata, ('profiles', episode_id, 'actions', 'play', 'publishTime'))),
 193             'series_id': podcast_id,
 194             'series': traverse_obj(metadata, ('profiles', podcast_id, 'title')),
 195             'formats': formats,
 196             'subtitles': subtitles,
 197         }
 198
 199
 200 class TuneInShortenerIE(InfoExtractor):
 201     IE_NAME = 'tunein:shortener'
 202     IE_DESC = False  # Do not list
 203     _VALID_URL = r'https?://tun\.in/(?P<id>[A-Za-z0-9]+)'
 204
 205     _TEST = {
 206         # test redirection
 207         'url': 'http://tun.in/ser7s',
 208         'info_dict': {
 209             'id': 's34682',
 210             'title': 're:^Jazz24',
 211             'description': 'md5:d6d0b89063fd68d529fa7058ee98619b',
 212             'thumbnail': 're:^https?://[^?&]+/s34682',
 213             'location': 'Seattle-Tacoma, US',
 214             'ext': 'mp3',
 215             'live_status': 'is_live',
 216         },
 217         'params': {
 218             'skip_download': True,  # live stream
 219         },
 220     }
 221
 222     def _real_extract(self, url):
 223         redirect_id = self._match_id(url)
 224         # The server doesn't support HEAD requests
 225         urlh = self._request_webpage(
 226             url, redirect_id, note='Downloading redirect page')
 227
 228         url = urlh.url
 229         url_parsed = urllib.parse.urlparse(url)
 230         if url_parsed.port == 443:
 231             url = url_parsed._replace(netloc=url_parsed.hostname).url
 232
 233         self.to_screen(f'Following redirect: {url}')
 234         return self.url_result(url)