[applepodcasts] Support new AMP-ish page structure

This commit is contained in:
dirkf 2021-12-08 23:37:54 +00:00
parent 23ad6402a6
commit 005339d637

View File

@ -3,7 +3,9 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
clean_html,
clean_podcast_url, clean_podcast_url,
get_element_by_class,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
try_get, try_get,
@ -14,15 +16,15 @@ class ApplePodcastsIE(InfoExtractor):
_VALID_URL = r'https?://podcasts\.apple\.com/(?:[^/]+/)?podcast(?:/[^/]+){1,2}.*?\bi=(?P<id>\d+)' _VALID_URL = r'https?://podcasts\.apple\.com/(?:[^/]+/)?podcast(?:/[^/]+){1,2}.*?\bi=(?P<id>\d+)'
_TESTS = [{ _TESTS = [{
'url': 'https://podcasts.apple.com/us/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777', 'url': 'https://podcasts.apple.com/us/podcast/207-whitney-webb-returns/id1135137367?i=1000482637777',
'md5': 'df02e6acb11c10e844946a39e7222b08', 'md5': '41dc31cd650143e530d9423b6b5a344f',
'info_dict': { 'info_dict': {
'id': '1000482637777', 'id': '1000482637777',
'ext': 'mp3', 'ext': 'mp3',
'title': '207 - Whitney Webb Returns', 'title': '207 - Whitney Webb Returns',
'description': 'md5:13a73bade02d2e43737751e3987e1399', 'description': 'md5:75ef4316031df7b41ced4e7b987f79c6',
'upload_date': '20200705', 'upload_date': '20200705',
'timestamp': 1593921600, 'timestamp': 1593932400,
'duration': 6425, 'duration': 6454,
'series': 'The Tim Dillon Show', 'series': 'The Tim Dillon Show',
} }
}, { }, {
@ -39,17 +41,38 @@ class ApplePodcastsIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
episode_id = self._match_id(url) episode_id = self._match_id(url)
webpage = self._download_webpage(url, episode_id) webpage = self._download_webpage(url, episode_id)
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( ember_data = self._parse_json(self._search_regex(
r'id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<', r'(?s)id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<',
webpage, 'ember data'), episode_id) webpage, 'ember data'), episode_id) or {}
ember_data = ember_data.get(episode_id) or ember_data ember_data = ember_data.get(episode_id) or ember_data
episode = ember_data['data']['attributes'] episode_data = try_get(ember_data, lambda x: x['data'], dict)
episode = episode_data['attributes']
description = episode.get('description') or {} description = episode.get('description') or {}
series = None 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': if inc.get('type') == 'media/podcast':
series = try_get(inc, lambda x: x['attributes']['name']) series = try_get(inc, lambda x: x['attributes']['name'])
series = series or clean_html(get_element_by_class('podcast-header__identity', webpage))
return { return {
'id': episode_id, 'id': episode_id,