youtube_dl/extractor/googlepodcasts.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     clean_podcast_url,
  10     int_or_none,
  11     try_get,
  12     urlencode_postdata,
  13 )
  14
  15
  16 class GooglePodcastsBaseIE(InfoExtractor):
  17     _VALID_URL_BASE = r'https?://podcasts\.google\.com/feed/'
  18
  19     def _batch_execute(self, func_id, video_id, params):
  20         return json.loads(self._download_json(
  21             'https://podcasts.google.com/_/PodcastsUi/data/batchexecute',
  22             video_id, data=urlencode_postdata({
  23                 'f.req': json.dumps([[[func_id, json.dumps(params), None, '1']]]),
  24             }), transform_source=lambda x: self._search_regex(r'(?s)(\[.+\])', x, 'data'))[0][2])
  25
  26     def _extract_episode(self, episode):
  27         return {
  28             'id': episode[4][3],
  29             'title': episode[8],
  30             'url': clean_podcast_url(episode[13]),
  31             'thumbnail': episode[2],
  32             'description': episode[9],
  33             'creator': try_get(episode, lambda x: x[14]),
  34             'timestamp': int_or_none(episode[11]),
  35             'duration': int_or_none(episode[12]),
  36             'series': episode[1],
  37         }
  38
  39
  40 class GooglePodcastsIE(GooglePodcastsBaseIE):
  41     IE_NAME = 'google:podcasts'
  42     _VALID_URL = GooglePodcastsBaseIE._VALID_URL_BASE + r'(?P<feed_url>[^/]+)/episode/(?P<id>[^/?&#]+)'
  43     _TEST = {
  44         'url': 'https://podcasts.google.com/feed/aHR0cHM6Ly9mZWVkcy5ucHIub3JnLzM0NDA5ODUzOS9wb2RjYXN0LnhtbA/episode/MzBlNWRlN2UtOWE4Yy00ODcwLTk2M2MtM2JlMmUyNmViOTRh',
  45         'md5': 'fa56b2ee8bd0703e27e42d4b104c4766',
  46         'info_dict': {
  47             'id': '30e5de7e-9a8c-4870-963c-3be2e26eb94a',
  48             'ext': 'mp3',
  49             'title': 'WWDTM New Year 2021',
  50             'description': 'We say goodbye to 2020 with Christine Baranksi, Doug Jones, Jonna Mendez, and Kellee Edwards.',
  51             'upload_date': '20210102',
  52             'timestamp': 1609606800,
  53             'duration': 2901,
  54             'series': "Wait Wait... Don't Tell Me!",
  55         }
  56     }
  57
  58     def _real_extract(self, url):
  59         b64_feed_url, b64_guid = re.match(self._VALID_URL, url).groups()
  60         episode = self._batch_execute(
  61             'oNjqVe', b64_guid, [b64_feed_url, b64_guid])[1]
  62         return self._extract_episode(episode)
  63
  64
  65 class GooglePodcastsFeedIE(GooglePodcastsBaseIE):
  66     IE_NAME = 'google:podcasts:feed'
  67     _VALID_URL = GooglePodcastsBaseIE._VALID_URL_BASE + r'(?P<id>[^/?&#]+)/?(?:[?#&]|$)'
  68     _TEST = {
  69         'url': 'https://podcasts.google.com/feed/aHR0cHM6Ly9mZWVkcy5ucHIub3JnLzM0NDA5ODUzOS9wb2RjYXN0LnhtbA',
  70         'info_dict': {
  71             'title': "Wait Wait... Don't Tell Me!",
  72             'description': "NPR's weekly current events quiz. Have a laugh and test your news knowledge while figuring out what's real and what we've made up.",
  73         },
  74         'playlist_mincount': 20,
  75     }
  76
  77     def _real_extract(self, url):
  78         b64_feed_url = self._match_id(url)
  79         data = self._batch_execute('ncqJEe', b64_feed_url, [b64_feed_url])
  80
  81         entries = []
  82         for episode in (try_get(data, lambda x: x[1][0]) or []):
  83             entries.append(self._extract_episode(episode))
  84
  85         feed = try_get(data, lambda x: x[3]) or []
  86         return self.playlist_result(
  87             entries, playlist_title=try_get(feed, lambda x: x[0]),
  88             playlist_description=try_get(feed, lambda x: x[2]))