W3C home > Mailing lists > Public > public-css-testsuite@w3.org > January 2011

Re: [RC4] and [RC5beta] containing-block-017 and other containing-block-01*: section 10.1 (inline versus block)

From: Alan Gresley <alan@css-class.com>
Date: Thu, 13 Jan 2011 04:54:54 +1100
Message-ID: <4D2DEAEE.7030409@css-class.com>
To: css21testsuite@gtalbot.org
CC: Arron Eicholz <Arron.Eicholz@microsoft.com>, "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
On 8/01/2011 8:05 AM, "Gérard Talbot" wrote:
> Arron and Alan,
>
> http://test.csswg.org/suites/css2.1/20101210/html4/containing-block-017.htm
> (RC4)
>
> http://test.csswg.org/source/contributors/microsoft/submitted/Chapter_10/containing-block-017.htm
> (RC5beta)
>
> There is a confusion here.


Yes there is.


> The original containing-block-017 testcase (RC3) was trying to test
> section 10.1, bullet 4, sub-bullet 2 by explictly styling span as
> 'display: block'.


This is correct.


> Now, suddenly, the RC5beta is testing section 10.1, bullet 4, sub-bullet
> 1, sub-sub-bullet 1


This is wrong and due to my error. My test is better for 
containing-block-031 which Opera passes but fails my test. I will write 
a new list message for containing-block-031.


> Whatever Microsoft chooses to do as testcases here, I hope the testcases
> are accurate, without duplication and that the test asserts accurate as
> well. I think the previous versions (RC4) had some coding to change and
> the test assert needed to be corrected.


Yes RC4 were correct for containing-block-017 and containing-block-018.

I would like to propose some different test for section 10.1 since the 
direction of ltr or rtl does not have much bearing to section 10.1 apart 
from issues.

1. Positioned elements with a offset value of left or right as auto and 
bi-direction.
2. Positioned elements and the direction of visible or hidden overflow 
and bi-direction.


These parts are relevant in section 10.1. Emphasis added.

   | The *position* and *size* of an element's box(es) are *sometimes* 
calculated
   | relative to a certain rectangle, called the containing block of the 
element.


I propose something like this to replace containing-block-017.


<http://css-class.com/test/css21testsuite/containing-block-017a.htm>


Removing position: absolute from the innermost <span> will result in the 
<div> box showing with a red background-color. I have rewritten the title.

    CSS Test: Containing Blocks - Block-level elements position and size 
based on
    the nearest block-level element that is position relative


I propose something like this to replace containing-block-018.


<http://css-class.com/test/css21testsuite/containing-block-018b.htm>


Both WebKit (Safari 5.0.3) and IE9 beta fails this. I'm not sure why IE9 
beta fails it. I know the bug with Safari since in rtl bi-direction, 
offset left: auto should be ignored thus having offset right: auto as 
the used value for positioning. This bug with Safari is seen in all the 
rtl test in this series with the aqua box with auto offset.


<http://css-class.com/test/css/bidi/visible-overflow-containing-block-rtl-ap-left.htm>


-- 
Alan http://css-class.com/

Armies Cannot Stop An Idea Whose Time Has Come. - Victor Hugo
Received on Wednesday, 12 January 2011 17:56:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 12 January 2011 17:56:45 GMT