[lecturio] Improve subtitles extraction (closes #18488)

This commit is contained in:
Sergey M․ 2018-12-11 23:45:02 +07:00
parent 13e17cd28e
commit 0a05cfabb6
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -136,9 +136,15 @@ def _real_extract(self, url):
cc_url = url_or_none(cc_url) cc_url = url_or_none(cc_url)
if not cc_url: if not cc_url:
continue continue
sub_dict = automatic_captions if 'auto-translated' in cc_label else subtitles
lang = self._search_regex( lang = self._search_regex(
r'/([a-z]{2})_', cc_url, 'lang', default=cc_label.split()[0]) r'/([a-z]{2})_', cc_url, 'lang',
default=cc_label.split()[0] if cc_label else 'en')
original_lang = self._search_regex(
r'/[a-z]{2}_([a-z]{2})_', cc_url, 'original lang',
default=None)
sub_dict = (automatic_captions
if 'auto-translated' in cc_label or original_lang
else subtitles)
sub_dict.setdefault(self._CC_LANGS.get(lang, lang), []).append({ sub_dict.setdefault(self._CC_LANGS.get(lang, lang), []).append({
'url': cc_url, 'url': cc_url,
}) })