<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - AX: Crash at AccessibilityRenderObject::computeAccessibilityIsIgnored const + 552"
href="https://bugs.webkit.org/show_bug.cgi?id=161276#c6">Comment # 6</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - AX: Crash at AccessibilityRenderObject::computeAccessibilityIsIgnored const + 552"
href="https://bugs.webkit.org/show_bug.cgi?id=161276">bug 161276</a>
from <span class="vcard"><a class="email" href="mailto:cfleizach@apple.com" title="chris fleizach <cfleizach@apple.com>"> <span class="fn">chris fleizach</span></a>
</span></b>
<pre>Maybe we can just do
RenderObject protect(mrenderer)
Then access it everywhere with
protect.isBR() and likewise with dot operator
(In reply to <a href="show_bug.cgi?id=161276#c5">comment #5</a>)
<span class="quote">> Comment on <span class=""><a href="attachment.cgi?id=287168&action=diff" name="attach_287168" title="Patch">attachment 287168</a> <a href="attachment.cgi?id=287168&action=edit" title="Patch">[details]</a></span>
> Patch
>
> View in context:
> <a href="https://bugs.webkit.org/attachment.cgi?id=287168&action=review">https://bugs.webkit.org/attachment.cgi?id=287168&action=review</a>
>
> >>> Source/WebCore/accessibility/AccessibilityRenderObject.cpp:1158
> >>> + RenderObject* protectedRenderer(m_renderer);
> >>
> >> does this hold onto the object? or just make a copy of the pointer?
> >
> > Good point, I'll double check. The RenderObject is not ref counted.
>
> Since m_renderer is just a raw pointer, we have no way to know anything
> about the ownership and hold onto it easily (I did a lot of searches on
> smart pointers but I don't think we want to declare sole ownership within
> this function). A copy of the RenderObject might work but I suspect it will
> affect the performance quite a lot. The easiest way I can think of (since we
> are not sure what is causing the m_renderer being null) is to null check at
> all the place within this function when we are accessing m_renderer, but
> that seems to be very ugly.</span ></pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>