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

Closes #7417
This commit is contained in:
pukkandan 2023-06-25 20:10:00 +05:30
parent e59e20744e
commit 58786a10f2
No known key found for this signature in database
GPG Key ID: 7EEE9E1E817D0A39
2 changed files with 17 additions and 8 deletions

View File

@ -1805,8 +1805,7 @@ The following extractors use this feature:
* `comment_sort`: `top` or `new` (default) - choose comment sorting mode (on YouTube's side) * `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` * `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 * 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. * `formats`: Change the types of formats to return. `dashy` (convert http to DASH), `duplicate` (identical content but different URLs or protocol; includes `dashy`), `incomplete` (cannot be downloaded completely - live dash and post-live m3u8)
* `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_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 * `innertube_key`: Innertube API key to use for all API requests

View File

@ -3752,7 +3752,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres' 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
]) ])
streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...)) streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
all_formats = self._configuration_arg('include_duplicate_formats') format_types = self._configuration_arg('formats')
all_formats = 'duplicate' in format_types
if self._configuration_arg('include_duplicate_formats'):
all_formats = True
self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
'Use formats=duplicate extractor argument instead')
def build_fragments(f): def build_fragments(f):
return LazyList({ return LazyList({
@ -3892,18 +3897,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if single_stream and dct.get('ext'): if single_stream and dct.get('ext'):
dct['container'] = dct['ext'] + '_dash' dct['container'] = dct['ext'] + '_dash'
if all_formats and dct['filesize']: if (all_formats or 'dashy' in format_types) and dct['filesize']:
yield { yield {
**dct, **dct,
'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'], 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
'protocol': 'http_dash_segments', 'protocol': 'http_dash_segments',
'fragments': build_fragments(dct), 'fragments': build_fragments(dct),
} }
dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE} if all_formats or 'dashy' not in format_types:
yield dct dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
yield dct
needs_live_processing = self._needs_live_processing(live_status, duration) needs_live_processing = self._needs_live_processing(live_status, duration)
skip_bad_formats = not self._configuration_arg('include_incomplete_formats') skip_bad_formats = 'incomplete' not in format_types
if self._configuration_arg('include_incomplete_formats'):
skip_bad_formats = False
self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
'Use formats=incomplete extractor argument instead')
skip_manifests = set(self._configuration_arg('skip')) skip_manifests = set(self._configuration_arg('skip'))
if (not self.get_param('youtube_include_hls_manifest', True) if (not self.get_param('youtube_include_hls_manifest', True)
@ -3915,7 +3925,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
skip_manifests.add('dash') skip_manifests.add('dash')
if self._configuration_arg('include_live_dash'): if self._configuration_arg('include_live_dash'):
self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. ' self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
'Use include_incomplete_formats extractor argument instead') 'Use formats=incomplete extractor argument instead')
elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live': elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
skip_manifests.add('dash') skip_manifests.add('dash')