[americastestkitchen] Improve metadata extraction for ATK episodes (#27860)

youtube_dl/extractor/aenetworks.py

@@ -256,7 +256,7 @@ class AENetworksShowIE(AENetworksListBaseIE):
             'title': 'Ancient Aliens',
             'description': 'md5:3f6d74daf2672ff3ae29ed732e37ea7f',
         },
-        'playlist_mincount': 168,
+        'playlist_mincount': 150,
     }]
     _RESOURCE = 'series'
     _ITEMS_KEY = 'episodes'

youtube_dl/extractor/aljazeera.py

@@ -1,13 +1,16 @@
 from __future__ import unicode_literals
 
+import json
+import re
+
 from .common import InfoExtractor
 
 
 class AlJazeeraIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?:programmes|video)/.*?/(?P<id>[^/]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?P<type>program/[^/]+|(?:feature|video)s)/\d{4}/\d{1,2}/\d{1,2}/(?P<id>[^/?&#]+)'
 
     _TESTS = [{
-        'url': 'http://www.aljazeera.com/programmes/the-slum/2014/08/deliverance-201482883754237240.html',
+        'url': 'https://www.aljazeera.com/program/episode/2014/9/19/deliverance',
         'info_dict': {
             'id': '3792260579001',
             'ext': 'mp4',
@@ -20,14 +23,34 @@ class AlJazeeraIE(InfoExtractor):
         'add_ie': ['BrightcoveNew'],
         'skip': 'Not accessible from Travis CI server',
     }, {
-        'url': 'http://www.aljazeera.com/video/news/2017/05/sierra-leone-709-carat-diamond-auctioned-170511100111930.html',
+        'url': 'https://www.aljazeera.com/videos/2017/5/11/sierra-leone-709-carat-diamond-to-be-auctioned-off',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.aljazeera.com/features/2017/8/21/transforming-pakistans-buses-into-art',
         'only_matching': True,
     }]
-    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/665003303001/default_default/index.html?videoId=%s'
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
 
     def _real_extract(self, url):
-        program_name = self._match_id(url)
-        webpage = self._download_webpage(url, program_name)
-        brightcove_id = self._search_regex(
-            r'RenderPagesVideo\(\'(.+?)\'', webpage, 'brightcove id')
-        return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)
+        post_type, name = re.match(self._VALID_URL, url).groups()
+        post_type = {
+            'features': 'post',
+            'program': 'episode',
+            'videos': 'video',
+        }[post_type.split('/')[0]]
+        video = self._download_json(
+            'https://www.aljazeera.com/graphql', name, query={
+                'operationName': 'SingleArticleQuery',
+                'variables': json.dumps({
+                    'name': name,
+                    'postType': post_type,
+                }),
+            }, headers={
+                'wp-site': 'aje',
+            })['data']['article']['video']
+        video_id = video['id']
+        account_id = video.get('accountId') or '665003303001'
+        player_id = video.get('playerId') or 'BkeSH5BDb'
+        return self.url_result(
+            self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id),
+            'BrightcoveNew', video_id)

youtube_dl/extractor/americastestkitchen.py

@@ -6,8 +6,10 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
+    int_or_none,
     try_get,
     unified_strdate,
+    unified_timestamp,
 )
 
 
@@ -22,8 +24,8 @@ class AmericasTestKitchenIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'md5:64e606bfee910627efc4b5f050de92b3',
             'thumbnail': r're:^https?://',
-            'timestamp': 1523664000,
-            'upload_date': '20180414',
+            'timestamp': 1523318400,
+            'upload_date': '20180410',
             'release_date': '20180410',
             'series': "America's Test Kitchen",
             'season_number': 18,
@@ -33,6 +35,27 @@ class AmericasTestKitchenIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # Metadata parsing behaves differently for newer episodes (705) as opposed to older episodes (582 above)
+        'url': 'https://www.americastestkitchen.com/episode/705-simple-chicken-dinner',
+        'md5': '06451608c57651e985a498e69cec17e5',
+        'info_dict': {
+            'id': '5fbe8c61bda2010001c6763b',
+            'title': 'Simple Chicken Dinner',
+            'ext': 'mp4',
+            'description': 'md5:eb68737cc2fd4c26ca7db30139d109e7',
+            'thumbnail': r're:^https?://',
+            'timestamp': 1610755200,
+            'upload_date': '20210116',
+            'release_date': '20210116',
+            'series': "America's Test Kitchen",
+            'season_number': 21,
+            'episode': 'Simple Chicken Dinner',
+            'episode_number': 3,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'https://www.americastestkitchen.com/videos/3420-pan-seared-salmon',
         'only_matching': True,
@@ -60,7 +83,10 @@ class AmericasTestKitchenIE(InfoExtractor):
             'url': 'https://player.zype.com/embed/%s.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ' % video['zypeId'],
             'ie_key': 'Zype',
             'description': clean_html(video.get('description')),
+            'timestamp': unified_timestamp(video.get('publishDate')),
             'release_date': unified_strdate(video.get('publishDate')),
+            'episode_number': int_or_none(episode.get('number')),
+            'season_number': int_or_none(episode.get('season')),
             'series': try_get(episode, lambda x: x['show']['title']),
             'episode': episode.get('title'),
         }

youtube_dl/extractor/ard.py

@@ -187,13 +187,13 @@ class ARDMediathekIE(ARDMediathekBaseIE):
             if doc.tag == 'rss':
                 return GenericIE()._extract_rss(url, video_id, doc)
 
-        title = self._html_search_regex(
+        title = self._og_search_title(webpage, default=None) or self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
              r'<meta name="dcterms\.title" content="(.*?)"/>',
              r'<h4 class="headline">(.*?)</h4>',
              r'<title[^>]*>(.*?)</title>'],
             webpage, 'title')
-        description = self._html_search_meta(
+        description = self._og_search_description(webpage, default=None) or self._html_search_meta(
             'dcterms.abstract', webpage, 'description', default=None)
         if description is None:
             description = self._html_search_meta(
@@ -249,18 +249,18 @@ class ARDMediathekIE(ARDMediathekBaseIE):
 
 
 class ARDIE(InfoExtractor):
-    _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos(?:extern)?/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
+    _VALID_URL = r'(?P<mainurl>https?://(?:www\.)?daserste\.de/[^?#]+/videos(?:extern)?/(?P<display_id>[^/?#]+)-(?:video-?)?(?P<id>[0-9]+))\.html'
     _TESTS = [{
-        # available till 14.02.2019
-        'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
-        'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49',
+        # available till 7.01.2022
+        'url': 'https://www.daserste.de/information/talk/maischberger/videos/maischberger-die-woche-video100.html',
+        'md5': '867d8aa39eeaf6d76407c5ad1bb0d4c1',
         'info_dict': {
-            'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
-            'id': '102',
+            'display_id': 'maischberger-die-woche',
+            'id': '100',
             'ext': 'mp4',
-            'duration': 4435.0,
-            'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
-            'upload_date': '20180214',
+            'duration': 3687.0,
+            'title': 'maischberger. die woche vom 7. Januar 2021',
+            'upload_date': '20210107',
             'thumbnail': r're:^https?://.*\.jpg$',
         },
     }, {
@@ -315,17 +315,17 @@ class ARDIE(InfoExtractor):
 class ARDBetaMediathekIE(ARDMediathekBaseIE):
     _VALID_URL = r'https://(?:(?:beta|www)\.)?ardmediathek\.de/(?P<client>[^/]+)/(?:player|live|video)/(?P<display_id>(?:[^/]+/)*)(?P<video_id>[a-zA-Z0-9]+)'
     _TESTS = [{
-        'url': 'https://ardmediathek.de/ard/video/die-robuste-roswita/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
-        'md5': 'dfdc87d2e7e09d073d5a80770a9ce88f',
+        'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/',
+        'md5': 'a1dc75a39c61601b980648f7c9f9f71d',
         'info_dict': {
             'display_id': 'die-robuste-roswita',
-            'id': '70153354',
+            'id': '78566716',
             'title': 'Die robuste Roswita',
-            'description': r're:^Der Mord.*trüber ist als die Ilm.',
+            'description': r're:^Der Mord.*totgeglaubte Ehefrau Roswita',
             'duration': 5316,
-            'thumbnail': 'https://img.ardmediathek.de/standard/00/70/15/33/90/-1852531467/16x9/960?mandant=ard',
-            'timestamp': 1577047500,
-            'upload_date': '20191222',
+            'thumbnail': 'https://img.ardmediathek.de/standard/00/78/56/67/84/575672121/16x9/960?mandant=ard',
+            'timestamp': 1596658200,
+            'upload_date': '20200805',
             'ext': 'mp4',
         },
     }, {

youtube_dl/extractor/extractors.py

@@ -651,6 +651,11 @@ from .microsoftvirtualacademy import (
     MicrosoftVirtualAcademyIE,
     MicrosoftVirtualAcademyCourseIE,
 )
+from .minds import (
+    MindsIE,
+    MindsChannelIE,
+    MindsGroupIE,
+)
 from .ministrygrid import MinistryGridIE
 from .minoto import MinotoIE
 from .miomio import MioMioIE
@@ -1116,6 +1121,10 @@ from .stitcher import (
 from .sport5 import Sport5IE
 from .sportbox import SportBoxIE
 from .sportdeutschland import SportDeutschlandIE
+from .spotify import (
+    SpotifyIE,
+    SpotifyShowIE,
+)
 from .spreaker import (
     SpreakerIE,
     SpreakerPageIE,

youtube_dl/extractor/minds.py

@@ -0,0 +1,196 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    clean_html,
+    int_or_none,
+    str_or_none,
+    strip_or_none,
+)
+
+
+class MindsBaseIE(InfoExtractor):
+    _VALID_URL_BASE = r'https?://(?:www\.)?minds\.com/'
+
+    def _call_api(self, path, video_id, resource, query=None):
+        api_url = 'https://www.minds.com/api/' + path
+        token = self._get_cookies(api_url).get('XSRF-TOKEN')
+        return self._download_json(
+            api_url, video_id, 'Downloading %s JSON metadata' % resource, headers={
+                'Referer': 'https://www.minds.com/',
+                'X-XSRF-TOKEN': token.value if token else '',
+            }, query=query)
+
+
+class MindsIE(MindsBaseIE):
+    IE_NAME = 'minds'
+    _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'(?:media|newsfeed|archive/view)/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://www.minds.com/media/100000000000086822',
+        'md5': '215a658184a419764852239d4970b045',
+        'info_dict': {
+            'id': '100000000000086822',
+            'ext': 'mp4',
+            'title': 'Minds intro sequence',
+            'thumbnail': r're:https?://.+\.png',
+            'uploader_id': 'ottman',
+            'upload_date': '20130524',
+            'timestamp': 1369404826,
+            'uploader': 'Bill Ottman',
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'tags': ['animation'],
+            'comment_count': int,
+            'license': 'attribution-cc',
+        },
+    }, {
+        # entity.type == 'activity' and empty title
+        'url': 'https://www.minds.com/newsfeed/798025111988506624',
+        'md5': 'b2733a74af78d7fd3f541c4cbbaa5950',
+        'info_dict': {
+            'id': '798022190320226304',
+            'ext': 'mp4',
+            'title': '798022190320226304',
+            'uploader': 'ColinFlaherty',
+            'upload_date': '20180111',
+            'timestamp': 1515639316,
+            'uploader_id': 'ColinFlaherty',
+        },
+    }, {
+        'url': 'https://www.minds.com/archive/view/715172106794442752',
+        'only_matching': True,
+    }, {
+        # youtube perma_url
+        'url': 'https://www.minds.com/newsfeed/1197131838022602752',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        entity_id = self._match_id(url)
+        entity = self._call_api(
+            'v1/entities/entity/' + entity_id, entity_id, 'entity')['entity']
+        if entity.get('type') == 'activity':
+            if entity.get('custom_type') == 'video':
+                video_id = entity['entity_guid']
+            else:
+                return self.url_result(entity['perma_url'])
+        else:
+            assert(entity['subtype'] == 'video')
+            video_id = entity_id
+        # 1080p and webm formats available only on the sources array
+        video = self._call_api(
+            'v2/media/video/' + video_id, video_id, 'video')
+
+        formats = []
+        for source in (video.get('sources') or []):
+            src = source.get('src')
+            if not src:
+                continue
+            formats.append({
+                'format_id': source.get('label'),
+                'height': int_or_none(source.get('size')),
+                'url': src,
+            })
+        self._sort_formats(formats)
+
+        entity = video.get('entity') or entity
+        owner = entity.get('ownerObj') or {}
+        uploader_id = owner.get('username')
+
+        tags = entity.get('tags')
+        if tags and isinstance(tags, compat_str):
+            tags = [tags]
+
+        thumbnail = None
+        poster = video.get('poster') or entity.get('thumbnail_src')
+        if poster:
+            urlh = self._request_webpage(poster, video_id, fatal=False)
+            if urlh:
+                thumbnail = urlh.geturl()
+
+        return {
+            'id': video_id,
+            'title': entity.get('title') or video_id,
+            'formats': formats,
+            'description': clean_html(entity.get('description')) or None,
+            'license': str_or_none(entity.get('license')),
+            'timestamp': int_or_none(entity.get('time_created')),
+            'uploader': strip_or_none(owner.get('name')),
+            'uploader_id': uploader_id,
+            'uploader_url': 'https://www.minds.com/' + uploader_id if uploader_id else None,
+            'view_count': int_or_none(entity.get('play:count')),
+            'like_count': int_or_none(entity.get('thumbs:up:count')),
+            'dislike_count': int_or_none(entity.get('thumbs:down:count')),
+            'tags': tags,
+            'comment_count': int_or_none(entity.get('comments:count')),
+            'thumbnail': thumbnail,
+        }
+
+
+class MindsFeedBaseIE(MindsBaseIE):
+    _PAGE_SIZE = 150
+
+    def _entries(self, feed_id):
+        query = {'limit': self._PAGE_SIZE, 'sync': 1}
+        i = 1
+        while True:
+            data = self._call_api(
+                'v2/feeds/container/%s/videos' % feed_id,
+                feed_id, 'page %s' % i, query)
+            entities = data.get('entities') or []
+            for entity in entities:
+                guid = entity.get('guid')
+                if not guid:
+                    continue
+                yield self.url_result(
+                    'https://www.minds.com/newsfeed/' + guid,
+                    MindsIE.ie_key(), guid)
+            query['from_timestamp'] = data['load-next']
+            if not (query['from_timestamp'] and len(entities) == self._PAGE_SIZE):
+                break
+            i += 1
+
+    def _real_extract(self, url):
+        feed_id = self._match_id(url)
+        feed = self._call_api(
+            'v1/%s/%s' % (self._FEED_PATH, feed_id),
+            feed_id, self._FEED_TYPE)[self._FEED_TYPE]
+
+        return self.playlist_result(
+            self._entries(feed['guid']), feed_id,
+            strip_or_none(feed.get('name')),
+            feed.get('briefdescription'))
+
+
+class MindsChannelIE(MindsFeedBaseIE):
+    _FEED_TYPE = 'channel'
+    IE_NAME = 'minds:' + _FEED_TYPE
+    _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'(?!(?:newsfeed|media|api|archive|groups)/)(?P<id>[^/?&#]+)'
+    _FEED_PATH = 'channel'
+    _TEST = {
+        'url': 'https://www.minds.com/ottman',
+        'info_dict': {
+            'id': 'ottman',
+            'title': 'Bill Ottman',
+            'description': 'Co-creator & CEO @minds',
+        },
+        'playlist_mincount': 54,
+    }
+
+
+class MindsGroupIE(MindsFeedBaseIE):
+    _FEED_TYPE = 'group'
+    IE_NAME = 'minds:' + _FEED_TYPE
+    _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'groups/profile/(?P<id>[0-9]+)'
+    _FEED_PATH = 'groups/group'
+    _TEST = {
+        'url': 'https://www.minds.com/groups/profile/785582576369672204/feed/videos',
+        'info_dict': {
+            'id': '785582576369672204',
+            'title': 'Cooking Videos',
+        },
+        'playlist_mincount': 1,
+    }

youtube_dl/extractor/spotify.py

@@ -0,0 +1,156 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_podcast_url,
+    float_or_none,
+    int_or_none,
+    strip_or_none,
+    try_get,
+    unified_strdate,
+)
+
+
+class SpotifyBaseIE(InfoExtractor):
+    _ACCESS_TOKEN = None
+    _OPERATION_HASHES = {
+        'Episode': '8276d4423d709ae9b68ec1b74cc047ba0f7479059a37820be730f125189ac2bf',
+        'MinimalShow': '13ee079672fad3f858ea45a55eb109553b4fb0969ed793185b2e34cbb6ee7cc0',
+        'ShowEpisodes': 'e0e5ce27bd7748d2c59b4d44ba245a8992a05be75d6fabc3b20753fc8857444d',
+    }
+    _VALID_URL_TEMPL = r'https?://open\.spotify\.com/%s/(?P<id>[^/?&#]+)'
+
+    def _real_initialize(self):
+        self._ACCESS_TOKEN = self._download_json(
+            'https://open.spotify.com/get_access_token', None)['accessToken']
+
+    def _call_api(self, operation, video_id, variables):
+        return self._download_json(
+            'https://api-partner.spotify.com/pathfinder/v1/query', video_id, query={
+                'operationName': 'query' + operation,
+                'variables': json.dumps(variables),
+                'extensions': json.dumps({
+                    'persistedQuery': {
+                        'sha256Hash': self._OPERATION_HASHES[operation],
+                    },
+                })
+            }, headers={'authorization': 'Bearer ' + self._ACCESS_TOKEN})['data']
+
+    def _extract_episode(self, episode, series):
+        episode_id = episode['id']
+        title = episode['name'].strip()
+
+        formats = []
+        audio_preview = episode.get('audioPreview') or {}
+        audio_preview_url = audio_preview.get('url')
+        if audio_preview_url:
+            f = {
+                'url': audio_preview_url.replace('://p.scdn.co/mp3-preview/', '://anon-podcast.scdn.co/'),
+                'vcodec': 'none',
+            }
+            audio_preview_format = audio_preview.get('format')
+            if audio_preview_format:
+                f['format_id'] = audio_preview_format
+                mobj = re.match(r'([0-9A-Z]{3})_(?:[A-Z]+_)?(\d+)', audio_preview_format)
+                if mobj:
+                    f.update({
+                        'abr': int(mobj.group(2)),
+                        'ext': mobj.group(1).lower(),
+                    })
+            formats.append(f)
+
+        for item in (try_get(episode, lambda x: x['audio']['items']) or []):
+            item_url = item.get('url')
+            if not (item_url and item.get('externallyHosted')):
+                continue
+            formats.append({
+                'url': clean_podcast_url(item_url),
+                'vcodec': 'none',
+            })
+
+        thumbnails = []
+        for source in (try_get(episode, lambda x: x['coverArt']['sources']) or []):
+            source_url = source.get('url')
+            if not source_url:
+                continue
+            thumbnails.append({
+                'url': source_url,
+                'width': int_or_none(source.get('width')),
+                'height': int_or_none(source.get('height')),
+            })
+
+        return {
+            'id': episode_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'description': strip_or_none(episode.get('description')),
+            'duration': float_or_none(try_get(
+                episode, lambda x: x['duration']['totalMilliseconds']), 1000),
+            'release_date': unified_strdate(try_get(
+                episode, lambda x: x['releaseDate']['isoString'])),
+            'series': series,
+        }
+
+
+class SpotifyIE(SpotifyBaseIE):
+    IE_NAME = 'spotify'
+    _VALID_URL = SpotifyBaseIE._VALID_URL_TEMPL % 'episode'
+    _TEST = {
+        'url': 'https://open.spotify.com/episode/4Z7GAJ50bgctf6uclHlWKo',
+        'md5': '74010a1e3fa4d9e1ab3aa7ad14e42d3b',
+        'info_dict': {
+            'id': '4Z7GAJ50bgctf6uclHlWKo',
+            'ext': 'mp3',
+            'title': 'From the archive: Why time management is ruining our lives',
+            'description': 'md5:b120d9c4ff4135b42aa9b6d9cde86935',
+            'duration': 2083.605,
+            'release_date': '20201217',
+            'series': "The Guardian's Audio Long Reads",
+        }
+    }
+
+    def _real_extract(self, url):
+        episode_id = self._match_id(url)
+        episode = self._call_api('Episode', episode_id, {
+            'uri': 'spotify:episode:' + episode_id
+        })['episode']
+        return self._extract_episode(
+            episode, try_get(episode, lambda x: x['podcast']['name']))
+
+
+class SpotifyShowIE(SpotifyBaseIE):
+    IE_NAME = 'spotify:show'
+    _VALID_URL = SpotifyBaseIE._VALID_URL_TEMPL % 'show'
+    _TEST = {
+        'url': 'https://open.spotify.com/show/4PM9Ke6l66IRNpottHKV9M',
+        'info_dict': {
+            'id': '4PM9Ke6l66IRNpottHKV9M',
+            'title': 'The Story from the Guardian',
+            'description': 'The Story podcast is dedicated to our finest audio documentaries, investigations and long form stories',
+        },
+        'playlist_mincount': 36,
+    }
+
+    def _real_extract(self, url):
+        show_id = self._match_id(url)
+        podcast = self._call_api('ShowEpisodes', show_id, {
+            'limit': 1000000000,
+            'offset': 0,
+            'uri': 'spotify:show:' + show_id,
+        })['podcast']
+        podcast_name = podcast.get('name')
+
+        entries = []
+        for item in (try_get(podcast, lambda x: x['episodes']['items']) or []):
+            episode = item.get('episode')
+            if not episode:
+                continue
+            entries.append(self._extract_episode(episode, podcast_name))
+
+        return self.playlist_result(
+            entries, show_id, podcast_name, podcast.get('description'))