On Mon, Dec 14, 2015 at 08:03:12PM +0100, Frédéric WANG wrote:
> Le 14/12/2015 19:39, Zalan Bujtas a écrit :
> > It’s great to hear that MathML is getting some proper cleanup. I think the 
> > proposal sounds good and I am looking forward to the patches. 
> >
> > Minor observation: I noticed that, since MathML block is no longer a 
> > flexbox, but it still needs some of the flexbox functionality, the 
> > following pattern has been introduced to the generic renderbox/block code.  
> >    
> > -    if (parent()->isFlexibleBox()) {
> > +    if (parent()->isFlexibleBox() || parent()->isRenderMathMLBlock()) {
> > I just wish there was another way to handle this.
> >
> > Zalan.
> Alex will correct me if I'm wrong, but I think the changes in RenderBox
> are just a temporary solution to avoid breaking tests while we continue
> to rewrite the MathML layout code. Hopefully, these won't be needed
> anymore when we are done.
> 

Exactly, we need those to keep the width and height overrides working
like the flexbox did. This is the way layouts such us the flexbox or
the grid can force sizes in the children even if they do not request
it. Until we remove the old nodes structure in every renderer we have
to keep it, but I think we can remove all of them because MathML is
not a general layout so we control every layout and we can change its
size based on what their parent sets, even if that means using the
size of a sibling.

Alex
_______________________________________________
webkit-dev mailing list
webkit-dev@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-dev

Reply via email to