Issue while parsing valid <meta> tags #586

Activity

1. MaciejZet commented on Oct 25, 2026

   

   MaciejZet

   on Oct 25, 2026

   More actions

   Reproduced on 1.1 and master fd4f032 with etree, DOM and lxml:

   import html5lib
   
   html = ('<!doctype html><html><head><title>x</title>'
           '<meta http-equiv="refresh" content="1; url=https://example.test/?date=1&raw=1">'
           '</head><body></body></html>')
   html5lib.HTMLParser(strict=True).parse(html)
   # ParseError: Named entity expected. Got none.

   strict=False preserves the attribute value correctly, although parser.errors still records the false error. The WHATWG ambiguous ampersand state does not require an error for &raw=1; strict=True incorrectly rejects it.

   Full base/optional suites passed locally on macOS with CPython 3.8–3.12: 17,499/23,563 passes per version, retaining existing skips and xfails. Runs used setuptools 70.3.0 and chardet 5.2.0 to isolate separate dependency failures. Additional malformed-reference controls still raised as expected. No hosted test CI result was verified.