Before this commit:
`lxml.html.clean.Cleaner` strips all ARIA attributes (e.g., aria-label,
aria-hidden, role) by default. This is because they are not included in
the "safe attributes" whitelist.
This behavior caused sanitized HTML to lose semantic meaning for
assistive technologies (screen readers), making the output non-compliant
with WCAG standards.
After this commit:
ARIA attributes are added to `defs.safe_attrs` so the Cleaner will no
longer filter them out. This change ensures that developers using lxml
for sanitization will not accidentally strip accessibility features
while maintaining protection against XSS.
Bug report: 2136144

This was added in lxml#486, but was already in the set on line 49.

This was added in #486, but was already in the set on line 49.

Before this commit:
`lxml.html.clean.Cleaner` striped all ARIA attributes (e.g., aria-label, aria-hidden, role) by default.
This is because they were not included in the "safe attributes" whitelist.
This behavior caused sanitized HTML to lose semantic meaning for assistive technologies
(screen readers), making the output non-compliant with WCAG standards.
After this commit:
ARIA attributes are added to `defs.safe_attrs` so the Cleaner will no longer filter them out.
This change ensures that developers using lxml for sanitization will not accidentally strip
accessibility features while maintaining protection against XSS.
Closes https://bugs.launchpad.net/bugs/2136144

This was added in #486, but was already in the set on line 49.