[ie/gem.cbc.ca:playlist] Fix extractor

Authored by: bashonly

yt_dlp/extractor/cbc.py

@@ -518,7 +518,41 @@ def entries():
         return self.playlist_result(entries(), playlist_id)
 
 
-class CBCGemIE(InfoExtractor):
+class CBCGemBaseIE(InfoExtractor):
+    _NETRC_MACHINE = 'cbcgem'
+    _GEO_COUNTRIES = ['CA']
+
+    def _call_show_api(self, item_id, display_id=None):
+        return self._download_json(
+            f'https://services.radio-canada.ca/ott/catalog/v2/gem/show/{item_id}',
+            display_id or item_id, query={'device': 'web'})
+
+    def _extract_item_info(self, item_info):
+        episode_number = None
+        title = traverse_obj(item_info, ('title', {str}))
+        if title and (mobj := re.match(r'(?P<episode>\d+)\. (?P<title>.+)', title)):
+            episode_number = int_or_none(mobj.group('episode'))
+            title = mobj.group('title')
+
+        return {
+            'episode_number': episode_number,
+            **traverse_obj(item_info, {
+                'id': ('url', {str}),
+                'episode_id': ('url', {str}),
+                'description': ('description', {str}),
+                'thumbnail': ('images', 'card', 'url', {url_or_none}, {update_url(query=None)}),
+                'episode_number': ('episodeNumber', {int_or_none}),
+                'duration': ('metadata', 'duration', {int_or_none}),
+                'release_timestamp': ('metadata', 'airDate', {unified_timestamp}),
+                'timestamp': ('metadata', 'availabilityDate', {unified_timestamp}),
+                'age_limit': ('metadata', 'rating', {trim_str(start='C')}, {parse_age_limit}),
+            }),
+            'episode': title,
+            'title': title,
+        }
+
+
+class CBCGemIE(CBCGemBaseIE):
     IE_NAME = 'gem.cbc.ca'
     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>[0-9a-z-]+/s(?P<season>[0-9]+)[a-z][0-9]+)'
     _TESTS = [{
@@ -574,9 +608,7 @@ class CBCGemIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    _GEO_COUNTRIES = ['CA']
     _TOKEN_API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
-    _NETRC_MACHINE = 'cbcgem'
     _claims_token = None
 
     def _new_claims_token(self, email, password):
@@ -639,9 +671,7 @@ def _real_initialize(self):
 
     def _real_extract(self, url):
         video_id, season_number = self._match_valid_url(url).group('id', 'season')
-        video_info = self._download_json(
+        video_info = self._call_show_api(video_id)
-            f'https://services.radio-canada.ca/ott/catalog/v2/gem/show/{video_id}',
-            video_id, query={'device': 'web'})
         item_info = traverse_obj(video_info, (
             'content', ..., 'lineups', ..., 'items',
             lambda _, v: v['url'] == video_id, any, {require('item info')}))
@@ -691,38 +721,21 @@ def _real_extract(self, url):
                 if 'descriptive' in fmt['format_id'].lower():
                     fmt['preference'] = -2
 
-        episode_number = None
-        title = traverse_obj(item_info, ('title', {str}))
-        if title and (mobj := re.match(r'(?P<episode>\d+)\. (?P<title>.+)', title)):
-            episode_number = int_or_none(mobj.group('episode'))
-            title = mobj.group('title')
-
         return {
             'season_number': int_or_none(season_number),
-            'episode_number': episode_number,
             **traverse_obj(video_info, {
                 'series': ('title', {str}),
                 'season_number': ('structuredMetadata', 'partofSeason', 'seasonNumber', {int_or_none}),
                 'genres': ('structuredMetadata', 'genre', ..., {str}),
             }),
-            **traverse_obj(item_info, {
+            **self._extract_item_info(item_info),
-                'description': ('description', {str}),
-                'thumbnail': ('images', 'card', 'url', {url_or_none}, {update_url(query=None)}),
-                'episode_number': ('episodeNumber', {int_or_none}),
-                'duration': ('metadata', 'duration', {int_or_none}),
-                'release_timestamp': ('metadata', 'airDate', {unified_timestamp}),
-                'timestamp': ('metadata', 'availabilityDate', {unified_timestamp}),
-                'age_limit': ('metadata', 'rating', {trim_str(start='C')}, {parse_age_limit}),
-            }),
             'id': video_id,
             'episode_id': video_id,
-            'episode': title,
-            'title': title,
             'formats': formats,
         }
 
 
-class CBCGemPlaylistIE(InfoExtractor):
+class CBCGemPlaylistIE(CBCGemBaseIE):
     IE_NAME = 'gem.cbc.ca:playlist'
     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>(?P<show>[0-9a-z-]+)/s(?P<season>[0-9]+))/?(?:[?#]|$)'
     _TESTS = [{
@@ -732,70 +745,35 @@ class CBCGemPlaylistIE(InfoExtractor):
         'info_dict': {
             'id': 'schitts-creek/s06',
             'title': 'Season 6',
-            'description': 'md5:6a92104a56cbeb5818cc47884d4326a2',
             'series': 'Schitt\'s Creek',
             'season_number': 6,
             'season': 'Season 6',
-            'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/season/perso/cbc_schitts_creek_season_06_carousel_v03.jpg?impolicy=ott&im=Resize=(_Size_)&quality=75',
         },
     }, {
         'url': 'https://gem.cbc.ca/schitts-creek/s06',
         'only_matching': True,
     }]
-    _API_BASE = 'https://services.radio-canada.ca/ott/cbc-api/v2/shows/'
+
+    def _entries(self, season_info):
+        for episode in traverse_obj(season_info, ('items', lambda _, v: v['url'])):
+            yield self.url_result(
+                f'https://gem.cbc.ca/media/{episode["url"]}', CBCGemIE,
+                **self._extract_item_info(episode))
 
     def _real_extract(self, url):
-        match = self._match_valid_url(url)
+        season_id, show, season = self._match_valid_url(url).group('id', 'show', 'season')
-        season_id = match.group('id')
+        show_info = self._call_show_api(show, display_id=season_id)
-        show = match.group('show')
+        season_info = traverse_obj(show_info, (
-        show_info = self._download_json(self._API_BASE + show, season_id, expected_status=426)
+            'content', ..., 'lineups',
-        season = int(match.group('season'))
+            lambda _, v: v['seasonNumber'] == int(season), any, {require('season info')}))
 
-        season_info = next((s for s in show_info['seasons'] if s.get('season') == season), None)
+        return self.playlist_result(
-
+            self._entries(season_info), season_id,
-        if season_info is None:
+            **traverse_obj(season_info, {
-            raise ExtractorError(f'Couldn\'t find season {season} of {show}')
+                'title': ('title', {str}),
-
+                'season': ('title', {str}),
-        episodes = []
+                'season_number': ('seasonNumber', {int_or_none}),
-        for episode in season_info['assets']:
+            }), series=traverse_obj(show_info, ('title', {str})))
-            episodes.append({
-                '_type': 'url_transparent',
-                'ie_key': 'CBCGem',
-                'url': 'https://gem.cbc.ca/media/' + episode['id'],
-                'id': episode['id'],
-                'title': episode.get('title'),
-                'description': episode.get('description'),
-                'thumbnail': episode.get('image'),
-                'series': episode.get('series'),
-                'season_number': episode.get('season'),
-                'season': season_info['title'],
-                'season_id': season_info.get('id'),
-                'episode_number': episode.get('episode'),
-                'episode': episode.get('title'),
-                'episode_id': episode['id'],
-                'duration': episode.get('duration'),
-                'categories': [episode.get('category')],
-            })
-
-        thumbnail = None
-        tn_uri = season_info.get('image')
-        # the-national was observed to use a "data:image/png;base64"
-        # URI for their 'image' value. The image was 1x1, and is
-        # probably just a placeholder, so it is ignored.
-        if tn_uri is not None and not tn_uri.startswith('data:'):
-            thumbnail = tn_uri
-
-        return {
-            '_type': 'playlist',
-            'entries': episodes,
-            'id': season_id,
-            'title': season_info['title'],
-            'description': season_info.get('description'),
-            'thumbnail': thumbnail,
-            'series': show_info.get('title'),
-            'season_number': season_info.get('season'),
-            'season': season_info['title'],
-        }
 
 
 class CBCGemLiveIE(InfoExtractor):