[Webkit-unassigned] [Bug 89000] [CSS Regions] Fix the lifecycle for the flow objects and their renderers

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=89000





--- Comment #22 from Tim Horton <timothy_horton at apple.com>  2012-06-21 01:36:11 PST ---
(In reply to comment #21)
> > It seems vaguely odd to me that the filenames are WebKit-prefixed as well, but it seems like this isn't even remotely the first one, so I guess it's OK. Class names, too.
> I thought the decision to prefix the NamedFlow class had some good reasons behind it so I've made the same thing for the collection.

Ok, seems reasonable.

> > I wonder if you meant for all this VCCLCompilerTool stuff to be here?
> I hand-crafted this part to be similar with the configuration for the WebKitNamedFlow.* files :). You're saying that they are not actually required?

I'm not positive, but maybe! Would be worth checking briefly, but not much more than that.

> > Can you reassure my innate fear of const_cast in this case? (i.e. explain why it's safe)
> This code was inherited from the old version :). However, I think the const_cast should be safe because the original Node* HashSet on the renderer doesn't have any const-ness attached to it.

Sure!

> Thanks for the review! Update coming later today.

Excellent.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.