1158653 - 'clip: rect()' and vertical-rl writing-mode

Status: RESOLVED DUPLICATE of bug 1079151

(Core :: Web Painting, defect)

Description

[Gérard Talbot]

Tests (with vendor-prefixes)
----------------------------

A)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-010.xht

B)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-012.xht

C)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-014.xht

D)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-016.xht

Expected results
----------------

A)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-010-ref.xht

B)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-012-ref.xht

C)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-014-ref.xht

D)
http://www.gtalbot.org/BrowserBugsSection/CSS3WritingModes/s76-clip-vrl-016-ref.xht


Explanation
-----------

'clip: rect()' should make a rectangular area visible and the rest invisible. Invisible in the CSS sense of 'visibility: hidden'; therefore transparent but still affecting layout.

In the 4 tests, the clipping "removes" the clipped area (removes in the CSS sense of 'display: none').

In the 4 tests, the invisible area should still affect layout. 

In tests s76-clip-vrl-010.xht and s76-clip-vrl-012.xht, the green square should not be overlapping the pass-fail-conditions sentence (expected).

"
A clipping region defines what portion of an element's border box is visible.
"
CSS2.1, §11.1.2 Clipping: the 'clip' property
http://www.w3.org/TR/CSS21/visufx.html#clipping

Notes
-----
- Those tests (along with 12 other 'clip: rect()' tests) will eventually be submitted to the test.csswg.org repository 
- IE11 and Chrome 44.0.2376.0 (dev) pass these 4 tests. Prince 10.1 fails those 4  tests.
- I am using Firefox 40.0a1 buildID=20150422165714
- I use Linux 3.13.0-49-generic x86_64, Qt: 4.8.6, KDE 4.13.3; Kubuntu (trusty) 14.04.02 LTS
- I've searched for duplicates and did not find any.

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Is the spec clear on whether rect() is intended to be physical or logical?

Comment 2

[Simon Montagu :smontagu]

Yes, it's explicitly called out as physical at http://dev.w3.org/csswg/css-writing-modes/#physical-only

Comment 3

[Jonathan Kew [:jfkthame]]

AFAICS, the clipping is working OK in these tests; in each case, I see the expected green square portion of the test image. The problem I'm seeing is that position:absolute is positioning the test image incorrectly.

Comment 4

[Jonathan Kew [:jfkthame]]

As such, I think this may well be a dupe of bug 1079151, but I'll mark it as dependent for now. Once 1079151 is fixed we can re-check these testcases.

Comment 5

[Jonathan Kew [:jfkthame]]

I've checked that these testcases all pass as expected with the patches from bug 1079151, just landed on inbound. Closing this as a dupe, as I don't think there's anything further to do here.