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

Re: inlines-003

From: Ian Hickson <ian@hixie.ch>
Date: Wed, 1 Dec 2010 20:38:39 +0000 (UTC)
To: fantasai <fantasai.lists@inkedblade.net>
cc: Chris Lilley <chris@w3.org>, public-css-testsuite@w3.org
Message-ID: <Pine.LNX.4.64.1012012032230.11018@ps20323.dreamhostps.com>
On Tue, 30 Nov 2010, fantasai wrote:
> On 11/24/2010 08:57 PM, Chris Lilley wrote:
> > Hello public-css-testsuite,
> > 
> >   inlines-003 has no pass criteria.
> 
> I don't actually understand the intention of this test, so I can
> only guess at how to fix it.
> 
> Hixie, do you remember what
>   http://www.hixie.ch/tests/adhoc/css/box/inline/003.xml
> was for and what it's pass conditions are supposed to be?

Its pass condition is what most browsers today do, there's nothing subtle 
about it. It's hard to do good tests for this kind of thing, though, hence 
the rather pathetic lack of a pass condition. Sorry about that.

But, try it in Opera, for instance. It's missing the left border of any 
inline starting on a new line. Other browsers used to have even more 
drastic failure conditions, historically. Not sure what the best way to 
describe the pass condition is. Maybe have two paragraphs, the second 
of which uses inline images that look like the boxed words instead of the 
words themselves, and then say that modulo font differences the two must 
look the same? Or use Ahem in some way?

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 1 December 2010 20:39:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 1 December 2010 20:39:22 GMT