|Summary:||Added RenderMathMLBlock as based object for MathML|
|Product:||WebKit||Reporter:||Alex Milowski <alex>|
|Component:||WebCore Misc.||Assignee:||Nobody <webkit-unassigned>|
|Version:||528+ (Nightly build)|
|OS:||OS X 10.5|
|Bug Depends on:|
Description Alex Milowski 2010-01-21 12:07:48 PST
This patch adds a new base render object for inline-block elements of MathML. It also sets up the necessary bits for all the rendering objects to follow.
Comment 1 Alex Milowski 2010-01-21 12:08:43 PST
Created attachment 47135 [details] RenderMathMLBlock Patch
Comment 2 Dave Hyatt 2010-01-21 14:25:24 PST
Comment on attachment 47135 [details] RenderMathMLBlock Patch In RenderMathMLBlock::styleDidChange, why are you marking the block as needing layout and pref width recalc always? That's really inefficient. I'm also not clear on wny it's necessary. Isn't the style diff doing the right thing in the base classes? In isChildAllowed you don't have to do ? true : false. Just return the conditional itself. stretchToHeight would be cleaner using a for loop to walk the children in my opinion. for (RenderObject* current = firstChild(); current; current = current->nextSibling()); It seems like makeBlockStyle should return a PassRefPtr, since you're wanting to have it be set on some RenderObject presumably.
Comment 3 Alex Milowski 2010-01-21 15:42:28 PST
Created attachment 47154 [details] Update to patch to address issues The styleDidChange method was removed. It isn't necessary in the code. The other changes have been made.
Comment 4 Dave Hyatt 2010-01-22 10:01:54 PST
Comment on attachment 47154 [details] Update to patch to address issues r=me
Comment 5 WebKit Commit Bot 2010-01-22 23:12:06 PST
Comment on attachment 47154 [details] Update to patch to address issues Clearing flags on attachment: 47154 Committed r53764: <http://trac.webkit.org/changeset/53764>
Comment 6 WebKit Commit Bot 2010-01-22 23:12:11 PST
All reviewed patches have been landed. Closing bug.