W3C home > Mailing lists > Public > public-css-testsuite@w3.org > December 2010

Re: Proposed improvements to c414-flt-ln-002

From: fantasai <fantasai.lists@inkedblade.net>
Date: Mon, 06 Dec 2010 21:52:27 -0500
Message-ID: <4CFDA16B.4070502@inkedblade.net>
To: css21testsuite@gtalbot.org
CC: Řyvind Stenhaug <oyvinds@opera.com>, Geoffrey Sneddon <gsneddon@opera.com>, "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
On 12/06/2010 03:13 PM, "Gérard Talbot" wrote:
>
> Le Lun 6 décembre 2010 0:53, fantasai a écrit :
>> On 10/06/2010 04:15 PM, "Gérard Talbot" wrote:
>>> Řyvind and Geoffrey,
>>>
>>> http://test.csswg.org/suites/css2.1/20101001/html4/c414-flt-ln-002.htm
>>>
>>> We discussed the c414-flt-ln-002 test before.
>>
>> I've altered the test to make it easier to determine the pass/fail
>> conditions:
>>     http://test.csswg.org/source/approved/css2.1/src/css1/c414-flt-ln-002.xht
>>
>> Let me know if this addresses the problem.
>>
>> ~fantasai
>
> If unstyled body text uses the default font-size of 16px, then the
> following values can and may well cause fractional pixel issues:
>
>     .test { position: relative; font: 1em/1.4 serif;
>             border: 5px solid gray; margin: 1em; color: silver; }
>
>     .test p, .test span { color: white; border: solid 0.2em; }
>
> line-height: 1.4 will be resolved as 16px mult 1.4 == 22.4px
>
> border-width: 0.2em will be resolved as 16px mult 0.2 == 3.2px
>
> There are several ways to workaround this.
>
> I used line-height: 1.25 (resolved as 20px, which creates the usual and
> normal 2px half-leading for a normal<p>)

I took this approach.

~fantasai
Received on Tuesday, 7 December 2010 02:53:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 7 December 2010 02:54:32 GMT