[extractor/youtube] Add extractor-arg `include_duplicate_formats`

diff --git a/README.md b/README.md
index f28bf8e12d93f2f28d74ed0dbc1417c956d95815..de83e421fb388baa9312bb548c3190f68e7a77d6 100644 (file)
--- a/README.md
+++ b/README.md
@@ -1787,6 +1787,7 @@ #### youtube
 * `comment_sort`: `top` or `new` (default) - choose comment sorting mode (on YouTube's side)
 * `max_comments`: Limit the amount of comments to gather. Comma-separated list of integers representing `max-comments,max-parents,max-replies,max-replies-per-thread`. Default is `all,all,all,all`
     * E.g. `all,all,1000,10` will get a maximum of 1000 replies total, with up to 10 replies per thread. `1000,all,100` will get a maximum of 1000 comments, with a maximum of 100 replies total
+* `include_duplicate_formats`: Extract formats with identical content but different URLs or protocol. This is useful if some of the formats are unavailable or throttled.
 * `include_incomplete_formats`: Extract formats that cannot be downloaded completely (live dash and post-live m3u8)
 * `innertube_host`: Innertube API host to use for all API requests; e.g. `studio.youtube.com`, `youtubei.googleapis.com`. Note that cookies exported from one subdomain will not work on others
 * `innertube_key`: Innertube API key to use for all API requests

diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py
index 79174b882b917ea2ba0cee044fbb6f3f7096115b..48f822e445fac6492f173b09081d83a7ff966165 100644 (file)
--- a/yt_dlp/extractor/youtube.py
+++ b/yt_dlp/extractor/youtube.py
@@ -3640,6 +3640,7 @@ def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, l
             'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
         ])
         streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
+        all_formats = self._configuration_arg('include_duplicate_formats')
 
         for fmt in streaming_formats:
             if fmt.get('targetDurationSec'):
@@ -3648,8 +3649,9 @@ def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, l
             itag = str_or_none(fmt.get('itag'))
             audio_track = fmt.get('audioTrack') or {}
             stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
-            if stream_id in stream_ids:
-                continue
+            if not all_formats:
+                if stream_id in stream_ids:
+                    continue
 
             quality = fmt.get('quality')
             height = int_or_none(fmt.get('height'))
@@ -3739,7 +3741,7 @@ def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, l
                     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',
-                    self.get_param('verbose') and client_name,
+                    (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
                 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
@@ -3762,26 +3764,28 @@ def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, l
             if mime_mobj:
                 dct['ext'] = mimetype2ext(mime_mobj.group(1))
                 dct.update(parse_codecs(mime_mobj.group(2)))
-
+            if itag:
+                itags[itag].add(('https', dct.get('language')))
+                stream_ids.append(stream_id)
             single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
             if single_stream and dct.get('ext'):
                 dct['container'] = dct['ext'] + '_dash'
-            if single_stream or itag == '17':
-                CHUNK_SIZE = 10 << 20
-                dct.update({
+
+            CHUNK_SIZE = 10 << 20
+            if dct['filesize']:
+                yield {
+                    **dct,
+                    'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
                     'protocol': 'http_dash_segments',
                     'fragments': [{
                         'url': update_url_query(dct['url'], {
                             'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, dct["filesize"])}'
                         })
                     } for range_start in range(0, dct['filesize'], CHUNK_SIZE)]
-                } if itag != '17' and dct['filesize'] else {
-                    'downloader_options': {'http_chunk_size': CHUNK_SIZE}
-                })
-
-            if itag:
-                itags[itag].add(('https', dct.get('language')))
-                stream_ids.append(stream_id)
+                }
+                if not all_formats:
+                    continue
+            dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
             yield dct
 
         needs_live_processing = self._needs_live_processing(live_status, duration)
@@ -3803,11 +3807,13 @@ def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, l
 
         def process_manifest_format(f, proto, client_name, itag):
             key = (proto, f.get('language'))
-            if key in itags[itag]:
+            if not all_formats and key in itags[itag]:
                 return False
             itags[itag].add(key)
 
-            if any(p != proto for p, _ in itags[itag]):
+            if itag and all_formats:
+                f['format_id'] = f'{itag}-{proto}'
+            elif any(p != proto for p, _ in itags[itag]):
                 f['format_id'] = f'{itag}-{proto}'
             elif itag:
                 f['format_id'] = itag