youtube-dl

commit: 005339d6375f2d2a4cec962b1c1a157c1dffbf8f
parent 23ad6402a6966dd09e4c854f32c33f69be1a064e
Author: dirkf <fieldhouse@gmx.net>
Date:   Wed,  8 Dec 2021 23:37:54 +0000

[applepodcasts] Support new AMP-ish page structure

Diffstat:
M
youtube_dl/extractor/applepodcasts.py
43
+++++++++++++++++++++++++++++++++----------

1 file changed, 33 insertions(+), 10 deletions(-)
diff --git a/youtube_dl/extractor/applepodcasts.py b/youtube_dl/extractor/applepodcasts.py
@@ -3,7 +3,9 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
     clean_podcast_url,
+    get_element_by_class,
     int_or_none,
     parse_iso8601,
     try_get,
@@ -14,15 +16,15 @@ class ApplePodcastsIE(InfoExtractor):
     _VALID_URL = r'https?://podcasts\.apple\.com/(?:[^/]+/)?podcast(?:/[^/]+){1,2}.*?\bi=(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://podcasts.apple.com/us/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777',
-        'md5': 'df02e6acb11c10e844946a39e7222b08',
+        'md5': '41dc31cd650143e530d9423b6b5a344f',
         'info_dict': {
             'id': '1000482637777',
             'ext': 'mp3',
             'title': '207 - Whitney Webb Returns',
-            'description': 'md5:13a73bade02d2e43737751e3987e1399',
+            'description': 'md5:75ef4316031df7b41ced4e7b987f79c6',
             'upload_date': '20200705',
-            'timestamp': 1593921600,
-            'duration': 6425,
+            'timestamp': 1593932400,
+            'duration': 6454,
             'series': 'The Tim Dillon Show',
         }
     }, {
@@ -39,17 +41,38 @@ class ApplePodcastsIE(InfoExtractor):
     def _real_extract(self, url):
         episode_id = self._match_id(url)
         webpage = self._download_webpage(url, episode_id)
-        ember_data = self._parse_json(self._search_regex(
-            r'id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<',
-            webpage, 'ember data'), episode_id)
-        ember_data = ember_data.get(episode_id) or ember_data
-        episode = ember_data['data']['attributes']
+        episode_data = {}
+        ember_data = {}
+        # new page type 2021-11
+        amp_data = self._parse_json(self._search_regex(
+            r'(?s)id="shoebox-media-api-cache-amp-podcasts"[^>]*>\s*({.+?})\s*<',
+            webpage, 'AMP data', default='{}'), episode_id, fatal=False) or {}
+        amp_data = try_get(amp_data,
+                           lambda a: self._parse_json(
+                               next(a[x] for x in iter(a) if episode_id in x),
+                               episode_id),
+                           dict) or {}
+        amp_data = amp_data.get('d') or []
+        episode_data = try_get(
+            amp_data,
+            lambda a: next(x for x in a
+                           if x['type'] == 'podcast-episodes' and x['id'] == episode_id),
+            dict)
+        if not episode_data:
+            # try pre 2021-11 page type: TODO: consider deleting if no longer used
+            ember_data = self._parse_json(self._search_regex(
+                r'(?s)id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<',
+                webpage, 'ember data'), episode_id) or {}
+            ember_data = ember_data.get(episode_id) or ember_data
+            episode_data = try_get(ember_data, lambda x: x['data'], dict)
+        episode = episode_data['attributes']
         description = episode.get('description') or {}
 
         series = None
-        for inc in (ember_data.get('included') or []):
+        for inc in (amp_data or ember_data.get('included') or []):
             if inc.get('type') == 'media/podcast':
                 series = try_get(inc, lambda x: x['attributes']['name'])
+        series = series or clean_html(get_element_by_class('podcast-header__identity', webpage))
 
         return {
             'id': episode_id,