mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-06-17 08:54:23 +00:00
Code Issues Projects Releases Wiki Activity

[ie/youtube] Fix automatic captions for some client combinations (#13268)

Browse Source 
Fix 32ed5f107c6c641958d1cd2752e130de4db55a13

Authored by: bashonly
This commit is contained in:
bashonly 2025-05-22 17:41:31 -05:00 committed by GitHub
parent 415b4c9f95
commit 53ea743a9c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 11 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
yt_dlp/extractor/youtube/_video.py
View File

@ -3950,19 +3950,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
subtitles = {} subtitles = {}
skipped_subs_clients = set() skipped_subs_clients = set()
prs = traverse_obj(player_responses, (
# Filter out initial_pr which does not have streamingData (smuggled client context)
lambda _, v: v['streamingData'] and v['captions']['playerCaptionsTracklistRenderer']))
pctrs = traverse_obj(prs, (..., 'captions', 'playerCaptionsTracklistRenderer', {dict})) # Only web/mweb clients provide translationLanguages, so include initial_pr in the traversal
translation_languages = { translation_languages = {
lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1) lang['languageCode']: self._get_text(lang['languageName'], max_runs=1)
for lang in traverse_obj(pctrs, (..., 'translationLanguages', ..., {dict}))} for lang in traverse_obj(player_responses, (
..., 'captions', 'playerCaptionsTracklistRenderer', 'translationLanguages',
lambda _, v: v['languageCode'] and v['languageName']))
}
# NB: Constructing the full subtitle dictionary is slow # NB: Constructing the full subtitle dictionary is slow
get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and ( get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles')) self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
all_captions = traverse_obj(pctrs, (..., 'captionTracks', ..., {dict})) # Filter out initial_pr which does not have streamingData (smuggled client context)
prs = traverse_obj(player_responses, (
lambda _, v: v['streamingData'] and v['captions']['playerCaptionsTracklistRenderer']))
all_captions = traverse_obj(prs, (
..., 'captions', 'playerCaptionsTracklistRenderer', 'captionTracks', ..., {dict}))
need_subs_langs = {get_lang_code(sub) for sub in all_captions if sub.get('kind') != 'asr'} need_subs_langs = {get_lang_code(sub) for sub in all_captions if sub.get('kind') != 'asr'}
need_caps_langs = { need_caps_langs = {
remove_start(get_lang_code(sub), 'a-') remove_start(get_lang_code(sub), 'a-')