It's certainly true that CSS is a fragile area, but I would argue that
one of the reasons it's fragile is because the implementation is a bit
unsound in many places. Cleaning up tech debt could make this
progressively more robust in the long run. Note that I'm also not
proposing a refactor to make the code more elegant or modern, but to
solve a systematic issue.


On Mon, Aug 26, 2024 at 8:52 PM Kevin Rushforth
<kevin.rushfo...@oracle.com> wrote:
>
> Lots of things might be good ideas. I'm rather skeptical of the cost / 
> benefit of implementing a new CSS parser. This would be a large effort and 
> run a significant risk of regression, especially since CSS is a fragile area 
> (although parsing less so than the runtime CSS processing).
>
> -- Kevin

Reply via email to