Reject incomplete MM title markup before HTML parsing
Metadata integrity checks / metadata (push) Successful in 6m59s

This commit is contained in:
2026-11-18 04:21:31 +08:00
parent 9864632774
commit 12f95cff91
@@ -155,6 +155,11 @@ class _TitleFormatting(HTMLParser):
def formatting_text(title):
# HTMLParser may silently discard a trailing incomplete start tag.
# Check lexical coverage before parsing so no title content can disappear.
literal=re.sub(r'</?(?:i|scp|sup|sub)\s*>','',title,flags=re.I)
if '<' in literal or '>' in literal:
return None
parser=_TitleFormatting()
parser.feed(title);parser.close()
text=''.join(parser.parts)