W3C home > Mailing lists > Public > public-css-testsuite@w3.org > August 2005

Possible issues with working copy of CSS 2.1 Test Suite

From: Jacqueline Weber <jweber@ftrd.us>
Date: Fri, 5 Aug 2005 16:32:31 -0400
Message-ID: <E7D674F13365EB489DDE409C298124B504EC0DD6@valiant.orange.us>
To: <public-css-testsuite@w3.org>
Cc: "Josh Randall" <jrandall@ftrd.us>
In going through the working copy of the CSS 2.1 Test Suite, I found  

a few possible issues with the tests:

 

 

 

__________________________________________________________

 

 

 

http://www.hixie.ch/tests/evil/css/css21/tests/t1506-c525-font-wt-00-b.x
ht

 

 

In this test, most of the lines state that they should be bolder or the
same as the previous line. The sentence "This sentence should be very
light" lacks that caveat. 

 

If there isn't a lighter font, isn't it ok for the font to be normal?

 

Here's the spec:

 

http://www.w3.org/TR/CSS21/fonts.html#font-boldness

 

"If there are fewer then 9 weights in the family, the default algorithm
for filling the "holes" is as 

follows. If '500' is unassigned, it will be assigned the same font as
'400'. If any of the values 

'600', '700', '800' or '900' remains unassigned, they are assigned to
the same face as the next 

darker assigned keyword, if any, or the next lighter one otherwise. If
any of '300', '200' or 

'100' remains unassigned, it is assigned to the next lighter assigned
keyword, if any, or the next 

darker otherwise."

 

So maybe the line could say something like "This sentence should be
lighter (or the same)" ?

 

__________________________________________________________

 

 

http://www.hixie.ch/tests/evil/css/css21/tests/t1508-c527-font-07- 

b.xht

 

 

 

This test states that "The lines should be one inch apart." The  

test says this because it defines the line-height to be one inch.

 

This has several possible problems.

 

First of all, the word 'apart' is interpreted by many people to mean  

the distance between two things, non-inclusive.  These people would  

measure from the bottom of the upper line to the top of the lower  

line. However, I believe line-height should actually be measured from  

baseline to baseline.

 

Secondly, doesn't line-height specify the minimum value between  

baselines for block-level elements?

 

     "On a block-level, table-cell, table-caption or inline-block  

element whose content is composed of inline-level elements, 'line- 

height' specifies the minimal height of line boxes within the  

element."  (http://www.w3.org/TR/CSS21/visudet.html#propdef-line-height)

 

In my opinion, the test would be more accurate if it read: "The  

distance from baseline to baseline of consecutive lines should be at  

least one inch."  Or something to that effect, anyway.

 

 __________________________________________________________

 

 

 

 http://www.hixie.ch/tests/evil/css/css21/tests/t1508-c527-font-08- 

 b.xht

 

 

This test seems to have an error in it.  The test reads: "This text  

should be 18px sans-serif, in small-caps, oblique, and bold."

 

However, I can't find where in the source code the font is defined to  

be small-caps.  In fact, the font variant is explicitly set to 'normal'.

 

 From the source code:

    .nine {font: oblique normal 700 18px/200% sans-serif;}

 

 Changing 'normal' to 'small-caps' makes this test pass in Deer Park  

 and Opera.

 

 

 

 

 __________________________________________________________

 

 

 

 

The labels on the following tests seem to be incorrect (in both .xht  

and .htm test files as well as the index):

 

t170602-bdr-conflct-w-14-d.htm
<http://www.hixie.ch/tests/evil/css/css21/tests/t170602-bdr-conflct-w-14
-d.htm>      10-Jul-2005 16:46     1k  CSS 2.1 Test Suite: Border
Conflict Resolution (width) - double/dashed

 

I think that the label should be 'double/dotted.'

 

 
 
t170602-bdr-conflct-w-96-d.htm
<http://www.hixie.ch/tests/evil/css/css21/tests/t170602-bdr-conflct-w-96
-d.htm>      10-Jul-2005 16:46     1k  CSS 2.1 Test Suite: Border
Conflict Resolution (width) - inset/outset

 

I believe the label should be 'none/outset.'

 

 

 

 

 
Received on Saturday, 6 August 2005 09:06:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 20 September 2010 17:51:54 GMT