Bug 172949 – Safari doesn't load newest The Order of the Stick comic.

• Status: RESOLVED
• Resolution: FIXED
• Component: Layout and Rendering

[alan]

1. Go to http://www.giantitp.com/Comics.html
2. Click the "New" link for The Order of the Stick

[alan]

rdar://problem/32389730

[alan]

Created attachment 312046 [details]
Patch

[alan]

Created attachment 312047 [details]
Patch

[Sam Weinig]

Comment on attachment 312047 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=312047&action=review

> Source/WebCore/ChangeLog:16
> +        Need to add a testcase.

Agreed :).

[Simon Fraser (smfr)]

Comment on attachment 312047 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=312047&action=review

> Source/WebCore/ChangeLog:11
> +        chain dirty, this dirty flag on the RenderTableRows will never be cleared and we'll end up early returning from RenderTableRow::paint.

RenderTableRow::paint() doesn't early return on needsLayout does it? I don't see that in the code.

[alan]

(In reply to Simon Fraser (smfr) from comment #5)
> Comment on attachment 312047 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=312047&action=review
> 
> > Source/WebCore/ChangeLog:11
> > +        chain dirty, this dirty flag on the RenderTableRows will never be cleared and we'll end up early returning from RenderTableRow::paint.
> 
> RenderTableRow::paint() doesn't early return on needsLayout does it? I don't
> see that in the code.
Right, my mistake. It's the RenderTableSection::paint.

[Simon Fraser (smfr)]

Weird that RenderTableSection has an early return that prevents painting. I guess we were just working around other table dirty layout bugs. We should remove that, or just paint a diagnostic color or something.

[alan]

(In reply to Simon Fraser (smfr) from comment #7)
> Weird that RenderTableSection has an early return that prevents painting. I
> guess we were just working around other table dirty layout bugs. We should
> remove that, or just paint a diagnostic color or something.
Agree. I don't mind removing it in a separate patch.

[Antti Koivisto]

Comment on attachment 312047 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=312047&action=review

> Source/WebCore/rendering/RenderBlockFlow.cpp:3698
> +        if (needsLayout())
> +            return;
> +        // FIXME: We should just kick off a subtree layout here (if needed at all) see webkit.org/b/172947.
> +        setNeedsLayout();

Maybe we can just remove this at some point.

[alan]

Created attachment 312098 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 312098 [details]
Patch

Clearing flags on attachment: 312098

Committed r217848: <http://trac.webkit.org/changeset/217848>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Simon Fraser (smfr)]

Comment on attachment 312098 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=312098&action=review

> LayoutTests/fast/table/floating-table-sibling-is-invisible.html:13
> +	font-size: 10px;

There is a tab here.

[Andres Gomez Garcia]

*** Bug 169742 has been marked as a duplicate of this bug. ***