Bug 64046 – Wrong image height in absolutely positioned div in relatively positioned parent with bottom padding

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

[Silas Brill]

Safari version: 5.0.5 (7533.21.1)

Other browsers tested:
Chrome 12: FAIL
Firefox 5: OK
IE 7/8/9: OK

Steps to reproduce:
1. Create a relatively positioned div with a height and a bottom padding
2. Put in that div an absolutely positioned div with height 100%
3. Put in the second div an image with height 100%
4. Open the file in Chrome or Safari

Expected behavior:

The image should fill the height of the second div

Actual behavior:

The image inherits the height value that was set on the first div

More information:

The attached file has steps 1, 2 and 3 already completed.  I have only tested this with the XHTML Transitional doctype.

[Silas Brill]

Created attachment 99909 [details]
Test HTML file

[Silas Brill]

Created attachment 99910 [details]
Screenshot

[Alice Boxhall]

Pulling together the relevant parts of the CSS spec:
http://www.w3.org/TR/CSS2/visudet.html#propdef-height
"<percentage> Specifies a percentage height. The percentage is calculated with respect to the height of the generated box's containing block."

http://www.w3.org/TR/CSS2/visudet.html#containing-block-details
"If the element has 'position: absolute', the containing block is established by the nearest ancestor with a 'position' of 'absolute', 'relative' or 'fixed', in the following way:
In the case that the ancestor is an inline element (...)
Otherwise, the containing block is formed by the padding edge of the ancestor."

So yes, this is a bug.

[Kulanthaivel Palanichamy]

Created attachment 101384 [details]
Proposed patch.

[Kulanthaivel Palanichamy]

Created attachment 101616 [details]
Proposed patch V2

[Dave Hyatt]

Comment on attachment 101616 [details]
Proposed patch V2

r=me. Much nicer!

[WebKit Review Bot]

Comment on attachment 101616 [details]
Proposed patch V2

Clearing flags on attachment: 101616

Committed r91533: <http://trac.webkit.org/changeset/91533>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Ryosuke Niwa]

Created attachment 101681 [details]
add pixel test result

[Ryosuke Niwa]

Can someone check & see if the pixel result is good or not?

[Dave Hyatt]

Comment on attachment 101681 [details]
add pixel test result

r=me

[Dave Hyatt]

r=me to add it, but should it really be in the cross-platform dir? The changelog says that at least.

[Ryosuke Niwa]

Committed r91862: <http://trac.webkit.org/changeset/91862>