yt_dlp/extractor/allstar.py

@@ -5,8 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     OnDemandPagedList,
+    UnsupportedError,
     int_or_none,
-    join_nonempty,
     parse_qs,
     traverse_obj,
     urljoin,
@@ -60,44 +60,36 @@ _QUERIES = {
 class AllstarBaseIE(InfoExtractor):
     @staticmethod
     def _parse_video_data(video_data):
-        def media_url_or_none(path):
+        def _media_url_or_none(path):
             return urljoin('https://media.allstar.gg/', path)
 
-        info = traverse_obj(video_data, {
+        def _profile_url_or_none(path):
+            return urljoin('https://allstar.gg/u/', path)
+
+        return traverse_obj(video_data, {
             'id': ('_id', {str}),
             'display_id': ('shareId', {str}),
             'title': ('clipTitle', {str}),
-            'url': ('clipLink', {media_url_or_none}),
+            'url': ('clipLink', {_media_url_or_none}),
-            'thumbnail': ('clipImageThumb', {media_url_or_none}),
+            'thumbnail': ('clipImageThumb', {_media_url_or_none}),
             'duration': ('clipLength', {int_or_none}),
             'filesize': ('clipSizeBytes', {int_or_none}),
-            'timestamp': ('createdDate', {functools.partial(int_or_none, scale=1000)}),
+            'timestamp': ('createdDate', {int_or_none}),
             'uploader': ('username', {str}),
             'uploader_id': ('user', '_id', {str}),
+            'uploader_url': ('user', '_id', {_profile_url_or_none}),
             'view_count': ('views', {int_or_none}),
         })
 
-        if info.get('id') and info.get('url'):
+    def _send_query(self, query, variables={}, path=(), video_id=None, note=None):
-            basename = 'clip' if '/clips/' in info['url'] else 'montage'
-            info['webpage_url'] = f'https://allstar.gg/{basename}?{basename}={info["id"]}'
-
-        info.update({
-            'extractor_key': AllstarIE.ie_key(),
-            'extractor': AllstarIE.IE_NAME,
-            'uploader_url': urljoin('https://allstar.gg/u/', info.get('uploader_id')),
-        })
-
-        return info
-
-    def _call_api(self, query, variables, path, video_id=None, note=None):
         response = self._download_json(
             'https://a1.allstar.gg/graphql', video_id, note=note,
             headers={'content-type': 'application/json'},
             data=json.dumps({'variables': variables, 'query': query}).encode())
 
-        errors = traverse_obj(response, ('errors', ..., 'message', {str}))
+        errors = response.get('errors')
         if errors:
-            raise ExtractorError('; '.join(errors))
+            raise ExtractorError(errors, expected=True)
 
         return traverse_obj(response, path)
 
@@ -116,10 +108,8 @@ class AllstarIE(AllstarBaseIE):
             'ext': 'mp4',
             'duration': 20,
             'filesize': 21199257,
-            'timestamp': 1682451501,
+            'timestamp': 1682451501555,
             'uploader_id': '62b8bdfc9021052f7905882d',
-            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
-            'upload_date': '20230425',
             'view_count': int,
         }
     }, {
@@ -133,11 +123,9 @@ class AllstarIE(AllstarBaseIE):
             'thumbnail': 'md5:90564b121f5fd7a4924920ef45614634',
             'duration': 16,
             'filesize': 30175859,
-            'timestamp': 1688333419,
+            'timestamp': 1688333419392,
             'uploader': 'cherokee',
             'uploader_id': '62b8bdfc9021052f7905882d',
-            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
-            'upload_date': '20230702',
             'view_count': int,
         }
     }, {
@@ -149,11 +137,9 @@ class AllstarIE(AllstarBaseIE):
             'url': 'md5:a3ee356022115db2b27c81321d195945',
             'thumbnail': 'md5:f1a5e811864e173f180b738d956356f4',
             'ext': 'mp4',
-            'timestamp': 1681810448,
+            'timestamp': 1681810448040,
             'uploader': 'cherokee',
             'uploader_id': '62b8bdfc9021052f7905882d',
-            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
-            'upload_date': '20230418',
             'view_count': int,
         }
     }, {
@@ -164,12 +150,10 @@ class AllstarIE(AllstarBaseIE):
             'title': 'cherokee Rapid Fire Snipers Montage',
             'url': 'md5:d5672e6f88579730c2310a80fdbc4030',
             'thumbnail': 'md5:60872f0d236863bb9a6f3dff1623403c',
-            'ext': 'mp4',
-            'timestamp': 1688365434,
             'uploader': 'cherokee',
+            'ext': 'mp4',
+            'timestamp': 1688365434271,
             'uploader_id': '62b8bdfc9021052f7905882d',
-            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
-            'upload_date': '20230703',
             'view_count': int,
         }
     }]
@@ -178,8 +162,9 @@ class AllstarIE(AllstarBaseIE):
         query_id, video_id = self._match_valid_url(url).group('type', 'id')
 
         return self._parse_video_data(
-            self._call_api(
+            self._send_query(
-                _QUERIES.get(query_id), {'id': video_id}, ('data', 'video'), video_id))
+                _QUERIES.get(query_id), {'id': video_id},
+                ('data', 'video'), video_id))
 
 
 class AllstarProfileIE(AllstarBaseIE):
@@ -188,28 +173,28 @@ class AllstarProfileIE(AllstarBaseIE):
     _TESTS = [{
         'url': 'https://allstar.gg/profile?user=62b8bdfc9021052f7905882d',
         'info_dict': {
-            'id': '62b8bdfc9021052f7905882d-clips',
+            'id': '62b8bdfc9021052f7905882d',
             'title': 'cherokee - Clips',
         },
         'playlist_mincount': 15
     }, {
         'url': 'https://allstar.gg/u/cherokee?game=730&view=Clips',
         'info_dict': {
-            'id': '62b8bdfc9021052f7905882d-clips-730',
+            'id': '62b8bdfc9021052f7905882d',
-            'title': 'cherokee - Clips - 730',
+            'title': 'cherokee - Clips',
         },
         'playlist_mincount': 15
     }, {
         'url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d?view=Montages',
         'info_dict': {
-            'id': '62b8bdfc9021052f7905882d-montages',
+            'id': '62b8bdfc9021052f7905882d',
             'title': 'cherokee - Montages',
         },
         'playlist_mincount': 4
     }, {
         'url': 'https://allstar.gg/profile?user=cherokee&view=Mobile Clips',
         'info_dict': {
-            'id': '62b8bdfc9021052f7905882d-mobile',
+            'id': '62b8bdfc9021052f7905882d',
             'title': 'cherokee - Mobile Clips',
         },
         'playlist_mincount': 1
@@ -217,36 +202,50 @@ class AllstarProfileIE(AllstarBaseIE):
 
     _PAGE_SIZE = 10
 
+    @staticmethod
+    def _set_webpage_url(info_dict):
+        video_id = info_dict.get('id')
+        video_url = info_dict.get('url')
+
+        if video_url is None or video_id is None:
+            return info_dict
+
+        base_name = 'clip' if '/clips/' in video_url else 'montage'
+        info_dict['webpage_url'] = f'https://allstar.gg/{base_name}?{base_name}={video_id}'
+        info_dict['webpage_url_basename'] = base_name
+
+        return info_dict
+
     def _get_page(self, user_id, display_id, game, query, page_num):
         page_num += 1
 
-        for video_data in self._call_api(
+        for video_data in self._send_query(
                 query, {
                     'user': user_id,
                     'page': page_num,
                     'game': game,
                 }, ('data', 'videos', 'data'), display_id, f'Downloading page {page_num}'):
-            yield self._parse_video_data(video_data)
+            yield self._set_webpage_url(self._parse_video_data(video_data))
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
         profile_data = self._download_json(
             urljoin('https://api.allstar.gg/v1/users/profile/', display_id), display_id)
         user_id = traverse_obj(profile_data, ('data', ('_id'), {str}))
-        if not user_id:
+
-            raise ExtractorError('Unable to extract the user id')
+        if user_id is None:
+            raise ExtractorError('Can not extract the user_id')
 
         username = traverse_obj(profile_data, ('data', 'profile', ('username'), {str}))
         url_query = parse_qs(url)
-        game = traverse_obj(url_query, ('game', 0, {int_or_none}))
+        game = traverse_obj(url_query, ('game', 0, {int}))
         query_id = traverse_obj(url_query, ('view', 0), default='Clips')
 
         if query_id not in ('Clips', 'Montages', 'Mobile Clips'):
-            raise ExtractorError(f'Unsupported playlist URL type {query_id!r}')
+            raise UnsupportedError(url)
 
         return self.playlist_result(
             OnDemandPagedList(
                 functools.partial(
                     self._get_page, user_id, display_id, game, _QUERIES.get(query_id)), self._PAGE_SIZE),
-            playlist_id=join_nonempty(user_id, query_id.lower().split()[0], game),
+            user_id, f'{username or display_id} - {query_id}')
-            playlist_title=join_nonempty((username or display_id), query_id, game, delim=' - '))