yt_dlp/extractor/applepodcasts.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     clean_html,
   4     clean_podcast_url,
   5     get_element_by_class,
   6     int_or_none,
   7     parse_iso8601,
   8     try_get,
   9 )
  10
  11
  12 class ApplePodcastsIE(InfoExtractor):
  13     _VALID_URL = r'https?://podcasts\.apple\.com/(?:[^/]+/)?podcast(?:/[^/]+){1,2}.*?\bi=(?P<id>\d+)'
  14     _TESTS = [{
  15         'url': 'https://podcasts.apple.com/us/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777',
  16         'md5': '41dc31cd650143e530d9423b6b5a344f',
  17         'info_dict': {
  18             'id': '1000482637777',
  19             'ext': 'mp3',
  20             'title': '207 - Whitney Webb Returns',
  21             'description': 'md5:75ef4316031df7b41ced4e7b987f79c6',
  22             'upload_date': '20200705',
  23             'timestamp': 1593932400,
  24             'duration': 6454,
  25             'series': 'The Tim Dillon Show',
  26             'thumbnail': 're:.+[.](png|jpe?g|webp)',
  27         }
  28     }, {
  29         'url': 'https://podcasts.apple.com/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777',
  30         'only_matching': True,
  31     }, {
  32         'url': 'https://podcasts.apple.com/podcast/207-whitney-webb-returns?i=1000482637777',
  33         'only_matching': True,
  34     }, {
  35         'url': 'https://podcasts.apple.com/podcast/id1135137367?i=1000482637777',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         episode_id = self._match_id(url)
  41         webpage = self._download_webpage(url, episode_id)
  42         episode_data = {}
  43         ember_data = {}
  44         # new page type 2021-11
  45         amp_data = self._parse_json(self._search_regex(
  46             r'(?s)id="shoebox-media-api-cache-amp-podcasts"[^>]*>\s*({.+?})\s*<',
  47             webpage, 'AMP data', default='{}'), episode_id, fatal=False) or {}
  48         amp_data = try_get(amp_data,
  49                            lambda a: self._parse_json(
  50                                next(a[x] for x in iter(a) if episode_id in x),
  51                                episode_id),
  52                            dict) or {}
  53         amp_data = amp_data.get('d') or []
  54         episode_data = try_get(
  55             amp_data,
  56             lambda a: next(x for x in a
  57                            if x['type'] == 'podcast-episodes' and x['id'] == episode_id),
  58             dict)
  59         if not episode_data:
  60             # try pre 2021-11 page type: TODO: consider deleting if no longer used
  61             ember_data = self._parse_json(self._search_regex(
  62                 r'(?s)id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<',
  63                 webpage, 'ember data'), episode_id) or {}
  64             ember_data = ember_data.get(episode_id) or ember_data
  65             episode_data = try_get(ember_data, lambda x: x['data'], dict)
  66         episode = episode_data['attributes']
  67         description = episode.get('description') or {}
  68
  69         series = None
  70         for inc in (amp_data or ember_data.get('included') or []):
  71             if inc.get('type') == 'media/podcast':
  72                 series = try_get(inc, lambda x: x['attributes']['name'])
  73         series = series or clean_html(get_element_by_class('podcast-header__identity', webpage))
  74
  75         return {
  76             'id': episode_id,
  77             'title': episode.get('name'),
  78             'url': clean_podcast_url(episode['assetUrl']),
  79             'description': description.get('standard') or description.get('short'),
  80             'timestamp': parse_iso8601(episode.get('releaseDateTime')),
  81             'duration': int_or_none(episode.get('durationInMilliseconds'), 1000),
  82             'series': series,
  83             'thumbnail': self._og_search_thumbnail(webpage),
  84             'vcodec': 'none',
  85         }