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

Microsoft line-height-111.htm (line-height: normal assumed to only be 2 values)

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Mon, 11 Oct 2010 13:21:12 -0700
Message-ID: <fcc23f045806c5745b539445f0b540db.squirrel@cp3.shieldhost.com>
To: "Řyvind Stenhaug" <oyvinds@opera.com>
Cc: "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>, "Arron Eicholz" <arron.eicholz@microsoft.com>
> (Author: Microsoft)

> 2)
> http://test.csswg.org/suites/css2.1/20101001/html4/line-height-111.htm
Wrong pass condition. Refers to "the two boxes", but there are
actually
> three of them.

Agreed. This would have to be reworded if the testcase was actually
correct, trustworthy.

> It looks like the test is indended to pass iff [height of middle box]
<=
> [height of leftmost box] <= [height of rightmost box], i.e. the  used
value of 'normal' is between 1 and 1.2, but in the CSS 2.1 spec this is
> rather explicitly non-normative.

Agreed.

As far as I can see this in my own tests[1], IE8, Opera 10.62, Chrome
6.0.472.63, Safari 5.0 and Konqueror 4.5.1 implements 'line-height:
normal' as 'line-height: 1.0' while Firefox 3.6.x implements
'line-height: normal' as 'line-height: 1.2'. But your point is correct.
Any CSS 2.1 compliant browser could implement any value, anything
between 1.0 and 1.2 as 'line-height: normal'. So, this makes testing
more
difficult. It makes line-height-111.htm unpredictable and rejectable.

<usability rant>
Firefox implementation helps legibility and it has been proven that
line-height 1.3 or 1.4 is better for reading (at least on a widescreen;
it is different for handheld small-screen devices ). This is furthermore
true for sentences in wide blocks or columns, for font with a relatively
tall x-height. When a webpage has link underlining, diacritic marks,
subscripts, superscripts, then a taller line-height is definitely
desirable.

IMHO, the best accessibility-wise choice for all CSS 2.1 compliant
browsers on widescreen would be to resolve 'line-height: normal' as
'line-height: 1.25' because such setting contributes/promotes
legibility, readability and avoids fractional pixel at the same time.
There is a consensus in web authoring (html, stylesheet, etc)
newsgroups that the default line-height setting in browsers (for
widescreen) is too small.
</usability rant>

There is also another phenomenon happening: fractional pixel calculation
and ensuing truncation or rounding up. 1.2em is 19.2px.

[1]: Revealing experiments on content box height, vertical-align and
line-height
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/experiments-va-lineheight-02.html

regards, Gérard
-- 
Contributions to the CSS 2.1 test suite:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/

CSS 2.1 test suite (RC2; October 1st 2010):
http://test.csswg.org/suites/css2.1/20101001/html4/toc.html

CSS 2.1 test suite contributors:
http://test.csswg.org/source/contributors/
Received on Monday, 11 October 2010 20:21:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 11 October 2010 20:21:58 GMT