mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-09-10 00:58:17 +02:00
[udemy] Fix subtitles extraction (closes #13812)
This commit is contained in:
parent
11a6793f80
commit
57a38a38c3
@ -15,6 +15,7 @@
|
||||
ExtractorError,
|
||||
float_or_none,
|
||||
int_or_none,
|
||||
js_to_json,
|
||||
sanitized_Request,
|
||||
unescapeHTML,
|
||||
urlencode_postdata,
|
||||
@ -268,6 +269,25 @@ def extract_formats(source_list):
|
||||
f = add_output_format_meta(f, format_id)
|
||||
formats.append(f)
|
||||
|
||||
def extract_subtitles(track_list):
|
||||
if not isinstance(track_list, list):
|
||||
return
|
||||
for track in track_list:
|
||||
if not isinstance(track, dict):
|
||||
continue
|
||||
if track.get('kind') != 'captions':
|
||||
continue
|
||||
src = track.get('src')
|
||||
if not src or not isinstance(src, compat_str):
|
||||
continue
|
||||
lang = track.get('language') or track.get(
|
||||
'srclang') or track.get('label')
|
||||
sub_dict = automatic_captions if track.get(
|
||||
'autogenerated') is True else subtitles
|
||||
sub_dict.setdefault(lang, []).append({
|
||||
'url': src,
|
||||
})
|
||||
|
||||
download_urls = asset.get('download_urls')
|
||||
if isinstance(download_urls, dict):
|
||||
extract_formats(download_urls.get('Video'))
|
||||
@ -315,23 +335,16 @@ def extract_formats(source_list):
|
||||
extract_formats(data.get('sources'))
|
||||
if not duration:
|
||||
duration = int_or_none(data.get('duration'))
|
||||
tracks = data.get('tracks')
|
||||
if isinstance(tracks, list):
|
||||
for track in tracks:
|
||||
if not isinstance(track, dict):
|
||||
continue
|
||||
if track.get('kind') != 'captions':
|
||||
continue
|
||||
src = track.get('src')
|
||||
if not src or not isinstance(src, compat_str):
|
||||
continue
|
||||
lang = track.get('language') or track.get(
|
||||
'srclang') or track.get('label')
|
||||
sub_dict = automatic_captions if track.get(
|
||||
'autogenerated') is True else subtitles
|
||||
sub_dict.setdefault(lang, []).append({
|
||||
'url': src,
|
||||
})
|
||||
extract_subtitles(data.get('tracks'))
|
||||
|
||||
if not subtitles and not automatic_captions:
|
||||
text_tracks = self._parse_json(
|
||||
self._search_regex(
|
||||
r'text-tracks=(["\'])(?P<data>\[.+?\])\1', view_html,
|
||||
'text tracks', default='{}', group='data'), video_id,
|
||||
transform_source=lambda s: js_to_json(unescapeHTML(s)),
|
||||
fatal=False)
|
||||
extract_subtitles(text_tracks)
|
||||
|
||||
self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user