- From: Sijtsche Smeman <sijtsche@wisdom.nl>
- Date: Mon, 9 Feb 2004 09:35:47 +0100
- To: "Bjoern Hoehrmann" <derhoermi@gmx.net>, "olivier Thereaux" <ot@w3.org>
- Cc: "CSS validator list" <www-validator-css@w3.org>
Having read all the mails of the people suddenly struggling with this "property does not exist for this medium- error" bug I thought, let's spend part of my weekend fixing it. The validator now gives the same message as before, only with a warning now instead of an error, so a document can still validate as valid CSS. This fix is now in CVS for those who want to try and test it immediately. Regards, Sijtsche Smeman ----- Original Message ----- From: "Bjoern Hoehrmann" <derhoermi@gmx.net> To: "olivier Thereaux" <ot@w3.org> Cc: "CSS validator list" <www-validator-css@w3.org> Sent: Friday, February 06, 2004 3:13 PM Subject: Re: CSS validator bug - box model hack > > * olivier Thereaux wrote: > >There seems to be a recent outcry on the Web about a new bug in the CSS > >validator that would cause stylesheets using the now infamous "Box > >Model Hack", used widely to avoid unpleasant results with > >non-conforming user agents. > > Well, as far as I can tell, this is not a new bug. > > >There is, indeed, a (minor yet unpleasant) bug, causing the CSS > >validator to inconsistently return errors (and thus invalidate > >stylesheets) while it should return a simple warning. This bug has > >been, apparently, present in the production version of the CSS > >validator since at least the August 2003 release. > > Its been documented on <http://jigsaw.w3.org/css-validator/BUGS.html> > since mid 2002 and it's already known since at least mid 2001, see > <http://lists.w3.org/Archives/Public/www-validator-css/2001Jul/0086> > Tantek even links the beforementioned document and my summary report > <http://lists.w3.org/Archives/Public/www-validator-css/2001Oct/0070> > on <http://tantek.com/CSS/Examples/boxmodelhack.html> since mid 2003. >
Received on Monday, 9 February 2004 03:34:27 UTC