Bug 69273 – CSS 2.1 failure: inline-replaced-height-008.htm

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Robert Hogan]

inline-replaced-height-008.htm
inline-block-replaced-height-008.htm
floating-inline-replaced-height-008.htm

in the unstable test suite all fail

[Robert Hogan]

Created attachment 109504 [details]
Patch

[WebKit Review Bot]

Comment on attachment 109504 [details]
Patch

Attachment 109504 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/9938169

New failing tests:
css2.1/20110323/inline-replaced-height-008.htm
css2.1/20110323/inline-block-replaced-height-008.htm
css2.1/20110323/floating-replaced-height-008.htm

[Robert Hogan]

Created attachment 109650 [details]
Patch

[Robert Hogan]

(In reply to comment #2)

> New failing tests:
> css2.1/20110323/inline-replaced-height-008.htm
> css2.1/20110323/inline-block-replaced-height-008.htm
> css2.1/20110323/floating-replaced-height-008.htm

Forgot to add a support file to the tests.

[Simon Fraser (smfr)]

Comment on attachment 109650 [details]
Patch

r=me, but those don't look like mac pixel results. What platform are you generating them on?

[Robert Hogan]

(In reply to comment #5)
> (From update of attachment 109650 [details])
> r=me, but those don't look like mac pixel results. What platform are you generating them on?

Chromium accepts mac pixel results for a lot of these tests, so I thought generating them from chromium would be OK. Sounds like that's not the case. I will land them to chromium and let Mac generate its own.

[Robert Hogan]

Committed r97378: <http://trac.webkit.org/changeset/97378>

[Nikolas Zimmermann]

I've stumbled across this work and read

 LayoutUnit RenderImage::computeReplacedLogicalHeight() const
 {
+    // FIXME: This function should use RenderReplaced::computeReplacedLogicalHeight()
...
in the commit.

I just wanted to denote that my patch in bug 47156 (which is not ready yet, depends on another one to be landed first) fixes all of the inline-height* bugs in the CSS 2.1 testsuite.
It also removes a lot of duplicated functionality from RenderImage, which is now present in RenderReplaced.

I want to avoid that both of us do the same work!