W3C home > Mailing lists > Public > www-style@w3.org > April 1998

Re: Preview: CSS1 Test Suite

From: Eric A. Meyer <emeyer@sr71.lit.cwru.edu>
Date: Tue, 7 Apr 1998 13:19:01 -0500
Message-Id: <v03102806b1500b1a7b6d@[]>
To: www-style@w3.org
>Err, hi. Me again. Having read the whole suite (minus a few of the margin
>pages which were getting a bit boring)..

   Yeah, those and the border tests get a little weary.

>I now have a longer list of problems
>to report. On the whole, however, the tests are great! Good work!

   Thank you, and thanks for all the reports!

>On page:
>Using the style sheet:
>>HTML BODY TABLE P {color: navy; font-family: sans-serif;}
>>EM, UL LI LI {color: green;}
>There is something very wrong with the first paragraph:
>>This text is normal for this page: navy (dark
>>blue) in color and a sans-serif font.
>>This sentence should be navy, except for the
>>last word, which should be green.
>Presumably, the TABLE in the first line of the style sheet is superfluous.

   A couple of people have commented on this one.  I think my original
intent was to comma-separate the selectors and I just forgot to do so.
I'll go back to the page and invest some thought in just how I want to
alter things.

>On page:
>I question the validity of the following statement:
>>The word quoted word "anchor" should NOT be a different
>>color, even though it's part of an anchor. It's a named
>>anchor, and styles declared for the A tag are not applied
>>to them under CSS1.
>I can't find this explicitly given in the CSS1 spec. Surely this is only
>true of As with a pseudo-class? Section 2.1 applies only to Anchor
>pseudo-classes, not to an A on it's own. Anybody have any other views on

   I believe the authors have said that anchor selectors don't apply to
named anchors.  I'll get a confirmation and update the text, if necessary.

>On page:
>Maybe it should be mentioned that support for this is optional. Also, is
>first-letter tested? I didn't see it anywhere.

   sec24.htm is a test for first-letter.  I made a conscious decision,
early on, to ignore considerations of the CSS1 core and optional
properties.  Either a property is supported, or it isn't.

>On page:
>Quite a few problems here, first of all in the style sheet:
>UL LI.mar {color: #66000;}
>There is a missing "0" in the hex (should be three or six characters).

   Oops.  I'll fix that right away.

>Secondly, the keyword "orange" is not CSS1, it is an IE4 invention (note
>that this appears in a few other pages too).

   Drat.  I was hoping I could get away with that one.  I'll put this on
the list of things to do.

>On page:
>In the table bit, the second square should be half the width of the *table
>cell*, not the UA window as written. This may not be the same (IE4, for
>example, only makes the table as big as is needed to accommodate the text).

   Yes, the text needs to be improved.  This was one of those tests I wrote
before I did the duplication-into-a-table thing, so the text is misleading.

>On page:
>In CSS1, percentage values on the height property are not applicable (in
>fact, they are absurd when we're not talking about CSS-P). Therefore this
>test is irrelevant. Note - IE4 does this as percentage of browser window
>height, which must have been interesting to program... Any comments on this

   Ironic that I would forget this, considering that a large chunk of the
last WG teleconference was devoted to percentage heights in positioning.
I'll take that statement out.

>On all pages:
>All the [SPEC] links are actually wrong, they point to "/PR/REC-CSS1/"
>instead of "/TR/REC-CSS1".
>(afternote - did this change while I was writing this or something? The
>links are correct now...)

   I fixed that typo yesterday, probably while you were going through the
   Excellent feedback, sir!

Eric A. Meyer  -  eam3@po.cwru.edu  -  http://www.cwru.edu/home/eam3.html
 Hypermedia Systems Manager
 Digital Media Services                http://www.cwru.edu/dms/dms.html
 Case Western Reserve University       http://www.cwru.edu/
Received on Tuesday, 7 April 1998 13:31:20 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:26:47 UTC