RenderTableRow should check if it has access to its ancestor chain.

[zalan]

It might be in detached state.

[zalan]

Created attachment 246687 [details]
Patch

[Andreas Kling]

Comment on attachment 246687 [details]
Patch

r=me

[Chris Dumez]

Comment on attachment 246687 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=246687&action=review

r=me % nits.

> Source/WebCore/rendering/RenderTableRow.h:46
> +    RenderTable* table() const

Now that this is no longer a one-liner, we may want to move it outside the class and mark it as inline.

> Source/WebCore/rendering/RenderTableRow.h:121
> +    RenderTableSection* section() const { return downcast<RenderTableSection>(parent()); }

Even though they are no outside callers now, I could imagine this getter being useful. Why not keep it public?

[Chris Dumez]

Andreas beat me to it :)

[zalan]

(In reply to comment #3)
> Comment on attachment 246687 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=246687&action=review
> 
> r=me % nits.
> 
> > Source/WebCore/rendering/RenderTableRow.h:46
> > +    RenderTable* table() const
> 
> Now that this is no longer a one-liner, we may want to move it outside the
> class and mark it as inline.
> 
Sure.

> > Source/WebCore/rendering/RenderTableRow.h:121
> > +    RenderTableSection* section() const { return downcast<RenderTableSection>(parent()); }
> 
> Even though they are no outside callers now, I could imagine this getter
> being useful. Why not keep it public?
I prefer the not public until required policy.

[zalan]

Created attachment 246690 [details]
Patch

[zalan]

Created attachment 246692 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 246692 [details]
Patch

Clearing flags on attachment: 246692

Committed r180190: <http://trac.webkit.org/changeset/180190>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.