[webkit-reviews] review requested: [Bug 57604] Remove nonstandard noresize attribute from HTML FrameSet Element : [Attachment 97573] Propose Patch

[bugzilla-daemon at webkit.org]

sarapkarthik <sarap.karthik at gmail.com> has asked  for review:
Bug 57604: Remove nonstandard noresize attribute from HTML FrameSet Element
https://bugs.webkit.org/show_bug.cgi?id=57604

Attachment 97573: Propose Patch
https://bugs.webkit.org/attachment.cgi?id=97573&action=review

------- Additional Comments from sarapkarthik <sarap.karthik at gmail.com>
#1>	Currently we have made the HTMLFrameSetElement::m_noresize = false so
that it ignores the "noresize" attributes value being mentioned in the html
page & always allow the frames to be resized.

There are two other ways it could have been resolved as described below:

#2>	We can remove the HTMLFrameSetElement::m_noresize &
HTMLFrameSetElement::noResize(). But this will give problem for FrameEdgeInfo
which is being used for both RenderFrame & RenderFrameSet, therefore we can't
remove the m_preventResize from FrameEdgeInfo.[noresize is a standard attribute
for frame element which needs to be respected]

We can continue to use FrameEdgeInfo for RenderFrame & introduce one more
struct FrameSetEdgeInfo {}, which will only contain m_allowBorder & NOT
m_preventResize.

#3>	If we don't want to introduce any struct as discussed above, we can
still maintain same FrameEdgeInfo. But we have to remove
RenderFrameSet::GridAxis::m_preventResize & modify
RenderFrameSet::fillFromEdgeInfo() accordingly so that we don't set the
m_preventResize for RenderFrameSet::m_rows & RenderFrameSet::m_cols.

Considering the fact that the code changes for #2 Or #3 discussed above would
be little more, we have implemented #1. Any suggestions/comments are always
welcome.