Merge a6a2eba51d into 6d8a53d870

2 weeks ago · 26bdeec626
parent 6d8a53d870 a6a2eba51d
commit 26bdeec626
2 changed files with 24 additions and 12 deletions
--- a/README.md
+++ b/README.md
@ -1760,7 +1760,7 @@ The following extractors use this feature:
 #### youtube
 * `lang`: Prefer translated metadata (`title`, `description` etc) of this language code (case-sensitive). By default, the video primary language metadata is preferred, with a fallback to `en` translated. See [youtube.py](https://github.com/yt-dlp/yt-dlp/blob/c26f9b991a0681fd3ea548d535919cec1fbbd430/yt_dlp/extractor/youtube.py#L381-L390) for list of supported content language codes
 * `skip`: One or more of `hls`, `dash` or `translated_subs` to skip extraction of the m3u8 manifests, dash manifests and [auto-translated subtitles](https://github.com/yt-dlp/yt-dlp/issues/4090#issuecomment-1158102032) respectively
-* `player_client`: Clients to extract video data from. The main clients are `web`, `android` and `ios` with variants `_music`, `_embedded`, `_embedscreen`, `_creator` (e.g. `web_embedded`); and `mweb`, `mweb_embedscreen`, `mediaconnect` and `tv_embedded` (agegate bypass) with no variants. By default, `ios,android,web` is used, but `tv_embedded` and `creator` variants are added as required for age-gated videos. Similarly, the music variants are added for `music.youtube.com` urls. You can use `all` to use all the clients, and `default` for the default clients.
+* `player_client`: Clients to extract video data from. The main clients are `web`, `ios` and `android`, with variants `_music`, `_embedded`, `_embedscreen`, `_creator` (e.g. `web_embedded`); and `mweb`, `mweb_embedscreen` and `tv_embedded` (agegate bypass) with no variants. By default, `ios,web` is used, but `tv_embedded` and `creator` variants are added as required for age-gated videos. Similarly, the music variants are added for `music.youtube.com` urls. The `android` clients will always be given lowest priority since their formats are broken. You can use `all` to use all the clients, and `default` for the default clients.
 * `player_skip`: Skip some network requests that are generally needed for robust extraction. One or more of `configs` (skip client configs), `webpage` (skip initial webpage), `js` (skip js player). While these options can help reduce the number of requests needed or avoid some rate-limiting, they could cause some issues. See [#860](https://github.com/yt-dlp/yt-dlp/pull/860) for more details
 * `player_params`: YouTube player parameters to use for player requests. Will overwrite any default ones set by yt-dlp.
 * `comment_sort`: `top` or `new` (default) - choose comment sorting mode (on YouTube's side)
--- a/yt_dlp/extractor/youtube.py
+++ b/yt_dlp/extractor/youtube.py
@ -2353,6 +2353,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                'format': '17',  # 3gp format available on android
                'extractor_args': {'youtube': {'player_client': ['android']}},
            },
+            'skip': 'android client broken',
        },
        {
            # Skip download of additional client configs (remix client config in this case)
@ -2730,7 +2731,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                'heatmap': 'count:100',
            },
            'params': {
-                'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
+                'extractor_args': {'youtube': {'player_client': ['ios'], 'player_skip': ['webpage']}},
            },
        },
    ]
@ -3614,8 +3615,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
        yt_query = {
            'videoId': video_id,
        }
-        if _split_innertube_client(client)[0] in ('android', 'android_embedscreen'):
-            yt_query['params'] = 'CgIIAQ=='

        pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
        if pp_arg:
@ -3631,19 +3630,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor):

    def _get_requested_clients(self, url, smuggled_data):
        requested_clients = []
-        default = ['ios', 'android', 'web']
+        android_clients = []
+        default = ['ios', 'web']
        allowed_clients = sorted(
            (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
            key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
        for client in self._configuration_arg('player_client'):
-            if client in allowed_clients:
-                requested_clients.append(client)
-            elif client == 'default':
+            if client == 'default':
                requested_clients.extend(default)
            elif client == 'all':
                requested_clients.extend(allowed_clients)
-            else:
+            elif client not in allowed_clients:
                self.report_warning(f'Skipping unsupported client {client}')
+            elif client.startswith('android'):
+                android_clients.append(client)
+            else:
+                requested_clients.append(client)
+        # Force deprioritization of broken Android clients for format de-duplication
+        requested_clients.extend(android_clients)
        if not requested_clients:
            requested_clients = default

@ -3862,6 +3866,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                    f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)

            client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
+            # Android client formats are broken due to integrity check enforcement
+            # Ref: https://github.com/yt-dlp/yt-dlp/issues/9554
+            is_broken = client_name and client_name.startswith(short_client_name('android'))
+            if is_broken:
+                self.report_warning(
+                    f'{video_id}: Android client formats are broken and may yield HTTP Error 403. '
+                    'They will be deprioritized', only_once=True)
+
            name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
            fps = int_or_none(fmt.get('fps')) or 0
            dct = {
@ -3874,7 +3886,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                    name, fmt.get('isDrc') and 'DRC',
                    try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
                    try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
-                    throttled and 'THROTTLED', is_damaged and 'DAMAGED',
+                    throttled and 'THROTTLED', is_damaged and 'DAMAGED', is_broken and 'BROKEN',
                    (self.get_param('verbose') or all_formats) and client_name,
                    delim=', '),
                # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
@ -3892,8 +3904,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                'language': join_nonempty(audio_track.get('id', '').split('.')[0],
                                          'desc' if language_preference < -1 else '') or None,
                'language_preference': language_preference,
-                # Strictly de-prioritize damaged and 3gp formats
-                'preference': -10 if is_damaged else -2 if itag == '17' else None,
+                # Strictly de-prioritize broken, damaged and 3gp formats
+                'preference': -20 if is_broken else -10 if is_damaged else -2 if itag == '17' else None,
            }
            mime_mobj = re.match(
                r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')