[vvvvid] fix season metadata extraction(#18130)

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'))

youtube_dl/extractor/stitcher.py

@@ -4,25 +4,28 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
+    clean_html,
+    ExtractorError,
     int_or_none,
-    js_to_json,
+    str_or_none,
-    unescapeHTML,
+    try_get,
 )
 
 
 class StitcherIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?stitcher\.com/podcast/(?:[^/]+/)+e/(?:(?P<display_id>[^/#?&]+?)-)?(?P<id>\d+)(?:[/#?&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?stitcher\.com/(?:podcast|show)/(?:[^/]+/)+e(?:pisode)?/(?:(?P<display_id>[^/#?&]+?)-)?(?P<id>\d+)(?:[/#?&]|$)'
     _TESTS = [{
         'url': 'http://www.stitcher.com/podcast/the-talking-machines/e/40789481?autoplay=true',
-        'md5': '391dd4e021e6edeb7b8e68fbf2e9e940',
+        'md5': 'e9635098e0da10b21a0e2b85585530f6',
         'info_dict': {
             'id': '40789481',
             'ext': 'mp3',
             'title': 'Machine Learning Mastery and Cancer Clusters',
-            'description': 'md5:55163197a44e915a14a1ac3a1de0f2d3',
+            'description': 'md5:547adb4081864be114ae3831b4c2b42f',
             'duration': 1604,
             'thumbnail': r're:^https?://.*\.jpg',
+            'upload_date': '20180126',
+            'timestamp': 1516989316,
         },
     }, {
         'url': 'http://www.stitcher.com/podcast/panoply/vulture-tv/e/the-rare-hourlong-comedy-plus-40846275?autoplay=true',
@@ -38,6 +41,7 @@ class StitcherIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'Page Not Found',
     }, {
         # escaped title
         'url': 'http://www.stitcher.com/podcast/marketplace-on-stitcher/e/40910226?autoplay=true',
@@ -45,37 +49,39 @@ class StitcherIE(InfoExtractor):
     }, {
         'url': 'http://www.stitcher.com/podcast/panoply/getting-in/e/episode-2a-how-many-extracurriculars-should-i-have-40876278?autoplay=true',
         'only_matching': True,
+    }, {
+        'url': 'https://www.stitcher.com/show/threedom/episode/circles-on-a-stick-200212584',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id, audio_id = re.match(self._VALID_URL, url).groups()
-        audio_id = mobj.group('id')
-        display_id = mobj.group('display_id') or audio_id
 
-        webpage = self._download_webpage(url, display_id)
+        resp = self._download_json(
+            'https://api.prod.stitcher.com/episode/' + audio_id,
+            display_id or audio_id)
+        episode = try_get(resp, lambda x: x['data']['episodes'][0], dict)
+        if not episode:
+            raise ExtractorError(resp['errors'][0]['message'], expected=True)
 
-        episode = self._parse_json(
+        title = episode['title'].strip()
-            js_to_json(self._search_regex(
+        audio_url = episode['audio_url']
-                r'(?s)var\s+stitcher(?:Config)?\s*=\s*({.+?});\n', webpage, 'episode config')),
-            display_id)['config']['episode']
 
-        title = unescapeHTML(episode['title'])
+        thumbnail = None
-        formats = [{
+        show_id = episode.get('show_id')
-            'url': episode[episode_key],
+        if show_id and episode.get('classic_id') != -1:
-            'ext': determine_ext(episode[episode_key]) or 'mp3',
+            thumbnail = 'https://stitcher-classic.imgix.net/feedimages/%s.jpg' % show_id
-            'vcodec': 'none',
-        } for episode_key in ('episodeURL',) if episode.get(episode_key)]
-        description = self._search_regex(
-            r'Episode Info:\s*</span>([^<]+)<', webpage, 'description', fatal=False)
-        duration = int_or_none(episode.get('duration'))
-        thumbnail = episode.get('episodeImage')
 
         return {
             'id': audio_id,
             'display_id': display_id,
             'title': title,
-            'description': description,
+            'description': clean_html(episode.get('html_description') or episode.get('description')),
-            'duration': duration,
+            'duration': int_or_none(episode.get('duration')),
             'thumbnail': thumbnail,
-            'formats': formats,
+            'url': audio_url,
+            'vcodec': 'none',
+            'timestamp': int_or_none(episode.get('date_created')),
+            'season_number': int_or_none(episode.get('season')),
+            'season_id': str_or_none(episode.get('season_id')),
         }

youtube_dl/extractor/vvvvid.py

@@ -25,7 +25,6 @@ class VVVVIDIE(InfoExtractor):
             'duration': 239,
             'series': '"Perché dovrei guardarlo?" di Dario Moccia',
             'season_id': '437',
-            'season_number': 1,
             'episode': 'Ping Pong',
             'episode_number': 1,
             'episode_id': '3334',
@@ -75,7 +74,6 @@ class VVVVIDIE(InfoExtractor):
     def _extract_common_video_info(self, video_data):
         return {
             'thumbnail': video_data.get('thumbnail'),
-            'episode_number': int_or_none(video_data.get('number')),
             'episode_id': str_or_none(video_data.get('id')),
         }
 
@@ -145,6 +143,17 @@ class VVVVIDIE(InfoExtractor):
 
             return d
 
+        info = {}
+
+        def metadata_from_url(r_url):
+            if not info and r_url:
+                mobj = re.search(r'_(?:S(\d+))?Ep(\d+)', r_url)
+                if mobj:
+                    info['episode_number'] = int(mobj.group(2))
+                    season_number = mobj.group(1)
+                    if season_number:
+                        info['season_number'] = int(season_number)
+
         for quality in ('_sd', ''):
             embed_code = video_data.get('embed_info' + quality)
             if not embed_code:
@@ -166,9 +175,12 @@ class VVVVIDIE(InfoExtractor):
             else:
                 formats.extend(self._extract_wowza_formats(
                     'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
+            metadata_from_url(embed_code)
+
         self._sort_formats(formats)
 
-        info = self._extract_common_video_info(video_data)
+        metadata_from_url(video_data.get('thumbnail'))
+        info.update(self._extract_common_video_info(video_data))
         info.update({
             'id': video_id,
             'title': title,
@@ -176,7 +188,6 @@ class VVVVIDIE(InfoExtractor):
             'duration': int_or_none(video_data.get('length')),
             'series': video_data.get('show_title'),
             'season_id': season_id,
-            'season_number': video_data.get('season_number'),
             'episode': title,
             'view_count': int_or_none(video_data.get('views')),
             'like_count': int_or_none(video_data.get('video_likes')),
@@ -211,7 +222,6 @@ class VVVVIDShowIE(VVVVIDIE):
 
         entries = []
         for season in (seasons or []):
-            season_number = int_or_none(season.get('number'))
             episodes = season.get('episodes') or []
             for episode in episodes:
                 if episode.get('playable') is False:
@@ -227,7 +237,6 @@ class VVVVIDShowIE(VVVVIDIE):
                     'url': '/'.join([base_url, season_id, video_id]),
                     'title': episode.get('title'),
                     'description': episode.get('description'),
-                    'season_number': season_number,
                     'season_id': season_id,
                 })
                 entries.append(info)