[acast] fix extraction(closes #21444)(closes #27612)(closes #27613)

youtube_dl/extractor/acast.py

@@ -2,21 +2,47 @@
 from __future__ import unicode_literals
 
 import re
-import functools
 
 from .common import InfoExtractor
-from ..compat import compat_str
 from ..utils import (
     clean_html,
-    float_or_none,
     int_or_none,
-    try_get,
+    parse_iso8601,
-    unified_timestamp,
-    OnDemandPagedList,
 )
 
 
-class ACastIE(InfoExtractor):
+class ACastBaseIE(InfoExtractor):
+    def _extract_episode(self, episode, show_info):
+        title = episode['title']
+        info = {
+            'id': episode['id'],
+            'display_id': episode.get('episodeUrl'),
+            'url': episode['url'],
+            'title': title,
+            'description': clean_html(episode.get('description') or episode.get('summary')),
+            'thumbnail': episode.get('image'),
+            'timestamp': parse_iso8601(episode.get('publishDate')),
+            'duration': int_or_none(episode.get('duration')),
+            'filesize': int_or_none(episode.get('contentLength')),
+            'season_number': int_or_none(episode.get('season')),
+            'episode': title,
+            'episode_number': int_or_none(episode.get('episode')),
+        }
+        info.update(show_info)
+        return info
+
+    def _extract_show_info(self, show):
+        return {
+            'creator': show.get('author'),
+            'series': show.get('title'),
+        }
+
+    def _call_api(self, path, video_id, query=None):
+        return self._download_json(
+            'https://feeder.acast.com/api/v1/shows/' + path, video_id, query=query)
+
+
+class ACastIE(ACastBaseIE):
     IE_NAME = 'acast'
     _VALID_URL = r'''(?x)
                     https?://
@@ -28,15 +54,15 @@ class ACastIE(InfoExtractor):
                     '''
     _TESTS = [{
         'url': 'https://www.acast.com/sparpodcast/2.raggarmordet-rosterurdetforflutna',
-        'md5': '16d936099ec5ca2d5869e3a813ee8dc4',
+        'md5': 'f5598f3ad1e4776fed12ec1407153e4b',
         'info_dict': {
             'id': '2a92b283-1a75-4ad8-8396-499c641de0d9',
             'ext': 'mp3',
             'title': '2. Raggarmordet - Röster ur det förflutna',
-            'description': 'md5:4f81f6d8cf2e12ee21a321d8bca32db4',
+            'description': 'md5:a992ae67f4d98f1c0141598f7bebbf67',
             'timestamp': 1477346700,
             'upload_date': '20161024',
-            'duration': 2766.602563,
+            'duration': 2766,
             'creator': 'Anton Berg & Martin Johnson',
             'series': 'Spår',
             'episode': '2. Raggarmordet - Röster ur det förflutna',
@@ -45,7 +71,7 @@ class ACastIE(InfoExtractor):
         'url': 'http://embed.acast.com/adambuxton/ep.12-adam-joeschristmaspodcast2015',
         'only_matching': True,
     }, {
-        'url': 'https://play.acast.com/s/rattegangspodden/s04e09-styckmordet-i-helenelund-del-22',
+        'url': 'https://play.acast.com/s/rattegangspodden/s04e09styckmordetihelenelund-del2-2',
         'only_matching': True,
     }, {
         'url': 'https://play.acast.com/s/sparpodcast/2a92b283-1a75-4ad8-8396-499c641de0d9',
@@ -54,40 +80,14 @@ class ACastIE(InfoExtractor):
 
     def _real_extract(self, url):
         channel, display_id = re.match(self._VALID_URL, url).groups()
-        s = self._download_json(
+        episode = self._call_api(
-            'https://feeder.acast.com/api/v1/shows/%s/episodes/%s' % (channel, display_id),
+            '%s/episodes/%s' % (channel, display_id),
-            display_id)
+            display_id, {'showInfo': 'true'})
-        media_url = s['url']
+        return self._extract_episode(
-        if re.search(r'[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12}', display_id):
+            episode, self._extract_show_info(episode.get('show') or {}))
-            episode_url = s.get('episodeUrl')
-            if episode_url:
-                display_id = episode_url
-            else:
-                channel, display_id = re.match(self._VALID_URL, s['link']).groups()
-        cast_data = self._download_json(
-            'https://play-api.acast.com/splash/%s/%s' % (channel, display_id),
-            display_id)['result']
-        e = cast_data['episode']
-        title = e.get('name') or s['title']
-        return {
-            'id': compat_str(e['id']),
-            'display_id': display_id,
-            'url': media_url,
-            'title': title,
-            'description': e.get('summary') or clean_html(e.get('description') or s.get('description')),
-            'thumbnail': e.get('image'),
-            'timestamp': unified_timestamp(e.get('publishingDate') or s.get('publishDate')),
-            'duration': float_or_none(e.get('duration') or s.get('duration')),
-            'filesize': int_or_none(e.get('contentLength')),
-            'creator': try_get(cast_data, lambda x: x['show']['author'], compat_str),
-            'series': try_get(cast_data, lambda x: x['show']['name'], compat_str),
-            'season_number': int_or_none(e.get('seasonNumber')),
-            'episode': title,
-            'episode_number': int_or_none(e.get('episodeNumber')),
-        }
 
 
-class ACastChannelIE(InfoExtractor):
+class ACastChannelIE(ACastBaseIE):
     IE_NAME = 'acast:channel'
     _VALID_URL = r'''(?x)
                     https?://
@@ -102,34 +102,24 @@ class ACastChannelIE(InfoExtractor):
         'info_dict': {
             'id': '4efc5294-5385-4847-98bd-519799ce5786',
             'title': 'Today in Focus',
-            'description': 'md5:9ba5564de5ce897faeb12963f4537a64',
+            'description': 'md5:c09ce28c91002ce4ffce71d6504abaae',
         },
-        'playlist_mincount': 35,
+        'playlist_mincount': 200,
     }, {
         'url': 'http://play.acast.com/s/ft-banking-weekly',
         'only_matching': True,
     }]
-    _API_BASE_URL = 'https://play.acast.com/api/'
-    _PAGE_SIZE = 10
 
     @classmethod
     def suitable(cls, url):
         return False if ACastIE.suitable(url) else super(ACastChannelIE, cls).suitable(url)
 
-    def _fetch_page(self, channel_slug, page):
-        casts = self._download_json(
-            self._API_BASE_URL + 'channels/%s/acasts?page=%s' % (channel_slug, page),
-            channel_slug, note='Download page %d of channel data' % page)
-        for cast in casts:
-            yield self.url_result(
-                'https://play.acast.com/s/%s/%s' % (channel_slug, cast['url']),
-                'ACast', cast['id'])
-
     def _real_extract(self, url):
-        channel_slug = self._match_id(url)
+        show_slug = self._match_id(url)
-        channel_data = self._download_json(
+        show = self._call_api(show_slug, show_slug)
-            self._API_BASE_URL + 'channels/%s' % channel_slug, channel_slug)
+        show_info = self._extract_show_info(show)
-        entries = OnDemandPagedList(functools.partial(
+        entries = []
-            self._fetch_page, channel_slug), self._PAGE_SIZE)
+        for episode in (show.get('episodes') or []):
-        return self.playlist_result(entries, compat_str(
+            entries.append(self._extract_episode(episode, show_info))
-            channel_data['id']), channel_data['name'], channel_data.get('description'))
+        return self.playlist_result(
+            entries, show.get('id'), show.get('title'), show.get('description'))