Bug 61117 – Inline non-replaced elements are reported to have zero width and height

• Component: CSS
• Status: RESOLVED FIXED

[Alexander Pavlov (apavlov)]

Firebug reports "auto" for the width and height of inline elements.

[Alexander Pavlov (apavlov)]

@annevk: which of the reported results ("0" vs "auto") do you consider correct? Or do you think it should be something else?

[Alexander Pavlov (apavlov)]

annevk's response:

----8<----
If 'display' is inline the 'width' (or 'height') property does not apply. Therefore the computed value needs to be returned. Which would be auto in this case as they are not set and auto is the initial value. Does that help?

(See http://www.w3.org/TR/CSS21/visudet.html#the-width-property for when 'width' applies.)
----8<----

Hence, the computed width/height should be "auto" not "0".

[Alexander Pavlov (apavlov)]

*** Bug 70252 has been marked as a duplicate of this bug. ***

[Alexander Pavlov (apavlov)]

Created attachment 116732 [details]
Patch

[Antti Koivisto]

Comment on attachment 116732 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=116732&action=review

> Source/WebCore/css/CSSComputedStyleDeclaration.cpp:361
> +static bool isInlineLevel(RenderStyle* style)
> +{
> +    switch (style->display()) {
> +    case INLINE:
> +    case INLINE_BLOCK:
> +    case INLINE_TABLE:
> +        return true;
> +    default:
> +        return false;
> +    }
> +}
> +

This information should be available from the renderer.

[Alexander Pavlov (apavlov)]

Created attachment 116753 [details]
Patch

[Alexander Pavlov (apavlov)]

(In reply to comment #5)
> (From update of attachment 116732 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=116732&action=review
> 
> > Source/WebCore/css/CSSComputedStyleDeclaration.cpp:361
> > +static bool isInlineLevel(RenderStyle* style)
> > +{
> > +    switch (style->display()) {
> > +    case INLINE:
> > +    case INLINE_BLOCK:
> > +    case INLINE_TABLE:
> > +        return true;
> > +    default:
> > +        return false;
> > +    }
> > +}
> > +
> 
> This information should be available from the renderer.

Made use of RenderObject::isInline() instead.

[Antti Koivisto]

Comment on attachment 116753 [details]
Patch

r=me

[Alexander Pavlov (apavlov)]

Committed r101787: <http://trac.webkit.org/changeset/101787>

[Eric Seidel (no email)]

This is causing a failure on Lion:
http://build.webkit.org/results/Lion%20Intel%20Release%20(Tests)/r101787%20(3165)/svg/css/getComputedStyle-basic-pretty-diff.html

[Alexander Pavlov (apavlov)]

(In reply to comment #10)
> This is causing a failure on Lion:
> http://build.webkit.org/results/Lion%20Intel%20Release%20(Tests)/r101787%20(3165)/svg/css/getComputedStyle-basic-pretty-diff.html

Could I have failed to update a Lion-specific expectation for this test? I have fixed one general and 2 chromium-specific ones. Are there more to fix?

[Darin Adler]

(In reply to comment #11)
> (In reply to comment #10)
> > This is causing a failure on Lion:
> > http://build.webkit.org/results/Lion%20Intel%20Release%20(Tests)/r101787%20(3165)/svg/css/getComputedStyle-basic-pretty-diff.html
> 
> Could I have failed to update a Lion-specific expectation for this test? I have fixed one general and 2 chromium-specific ones. Are there more to fix?

Yes, there is.

platform/mac/svg/css/getComputedStyle-basic-expected.txt

I used the "find" command to find all the copies of this test result. Apparently the Mac version exists because CSS_GRID_LAYOUT is on for Mac and off for other platforms.

[Alexander Pavlov (apavlov)]

(In reply to comment #12)
> (In reply to comment #11)
> > (In reply to comment #10)
> > > This is causing a failure on Lion:
> > > http://build.webkit.org/results/Lion%20Intel%20Release%20(Tests)/r101787%20(3165)/svg/css/getComputedStyle-basic-pretty-diff.html
> > 
> > Could I have failed to update a Lion-specific expectation for this test? I have fixed one general and 2 chromium-specific ones. Are there more to fix?
> 
> Yes, there is.
> 
> platform/mac/svg/css/getComputedStyle-basic-expected.txt
> 
> I used the "find" command to find all the copies of this test result. Apparently the Mac version exists because CSS_GRID_LAYOUT is on for Mac and off for other platforms.

Ouch... I used the same command and did not find this file. Also, http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/svg/css/getComputedStyle-basic-expected.txt results in
"No node /trunk/LayoutTests/platform/mac/svg/css/getComputedStyle-basic-expected.txt at revision 101951". Any clues what's going on here?

[Darin Adler]

Between the last two comments, I landed <http://trac.webkit.org/changeset/101941>, which removes the Mac-specific file and fixes the issue.

[Alexander Pavlov (apavlov)]

(In reply to comment #14)
> Between the last two comments, I landed <http://trac.webkit.org/changeset/101941>, which removes the Mac-specific file and fixes the issue.

Thanks for handling this, Darin. I should be more attentive next time...