[youtube] Add ability to authenticate with cookies

[youtube-dl] / youtube_dl / extractor / yandexmusic.py

diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py
index 6e7a23782273352c757328f659ed033b0bb7b64c..eb1062142ecbc6a4702f0dc7763c414934fd3645 100644 (file)
--- a/youtube_dl/extractor/yandexmusic.py
+++ b/youtube_dl/extractor/yandexmusic.py
@@ -20,18 +20,24 @@ class YandexMusicBaseIE(InfoExtractor):
             error = response.get('error')
             if error:
                 raise ExtractorError(error, expected=True)
+            if response.get('type') == 'captcha' or 'captcha' in response:
+                YandexMusicBaseIE._raise_captcha()
+
+    @staticmethod
+    def _raise_captcha():
+        raise ExtractorError(
+            'YandexMusic has considered youtube-dl requests automated and '
+            'asks you to solve a CAPTCHA. You can either wait for some '
+            'time until unblocked and optionally use --sleep-interval '
+            'in future or alternatively you can go to https://music.yandex.ru/ '
+            'solve CAPTCHA, then export cookies and pass cookie file to '
+            'youtube-dl with --cookies',
+            expected=True)
 
     def _download_webpage(self, *args, **kwargs):
         webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
         if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
-            raise ExtractorError(
-                'YandexMusic has considered youtube-dl requests automated and '
-                'asks you to solve a CAPTCHA. You can either wait for some '
-                'time until unblocked and optionally use --sleep-interval '
-                'in future or alternatively you can go to https://music.yandex.ru/ '
-                'solve CAPTCHA, then export cookies and pass cookie file to '
-                'youtube-dl with --cookies',
-                expected=True)
+            self._raise_captcha()
         return webpage
 
     def _download_json(self, *args, **kwargs):
@@ -69,6 +75,12 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
             % storage_dir,
             track_id, 'Downloading track location JSON')
 
+        # Each string is now wrapped in a list, this is probably only temporarily thus
+        # supporting both scenarios (see https://github.com/rg3/youtube-dl/issues/10193)
+        for k, v in data.items():
+            if v and isinstance(v, list):
+                data[k] = v[0]
+
         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
         storage = storage_dir.split('.')
 
@@ -194,7 +206,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
             'id': '1036',
             'title': 'Музыка 90-х',
         },
-        'playlist_count': 310,
+        'playlist_mincount': 300,
         'skip': 'Travis CI servers blocked by YandexMusic',
     }]
 
@@ -222,13 +234,18 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
                 'overembed': 'false',
             })['playlist']
 
-        tracks, track_ids = playlist['tracks'], playlist['trackIds']
+        tracks = playlist['tracks']
+        track_ids = [compat_str(track_id) for track_id in playlist['trackIds']]
 
         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
         # missing tracks should be retrieved manually.
         if len(tracks) < len(track_ids):
-            present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
-            missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
+            present_track_ids = set([
+                compat_str(track['id'])
+                for track in tracks if track.get('id')])
+            missing_track_ids = [
+                track_id for track_id in track_ids
+                if track_id not in present_track_ids]
             missing_tracks = self._download_json(
                 'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
                 playlist_id, 'Downloading missing tracks JSON',