W3C home > Mailing lists > Public > www-validator-css@w3.org > January 2003

Re: Re: Fwd: selector grammer

From: Alex Geller <alexgeller@web.de>
Date: Wed, 22 Jan 2003 04:12:30 -0500 (EST)
Message-Id: <200301220911.h0M9BvO01915@mailgate5.cinetic.de>
To: www-validator-css@w3.org




>>>
>>> By the original grammer even the following should be valid
>>> #myid#anotherid { font-size:12pt }
>>> Under "5.9 ID selectors"
>> 
>> This one I wouldn't worry too much about. By prose CSS is aware of the 
>> uniqueness of IDs, but there is nothing in CSS to prevent impossible 
>> selectors like :link:visited, nor is there any need for that. It is like 
>> selecting triangles with four corners. I see no philosophical or practical 
>> reasons why a selector that can't return any element should be disallowed.

>Indeed that example isn't even impossible -- if an element has ID
>attributes from two namespaces, for instance, or if it is in a non-XML
>language. Also, using the same ID twice is a quick way of increasing the
>specificity.
Sorry if I was unprecise, my point was less about whether or not these selectors make sense (this would be the wrong forum then). My point was, that the validator find them both erronous but for different reasons. #myid#otherid is rejected for grammatical reasons while
:first-child#myid#otherid is rejected for sematic reasons. My suggestion was to change the grammer so that both are rejected for the same reason.   
________________________________________________________________
Mit der Grupppen-SMS von WEB.DE FreeMail können Sie eine SMS an alle 
Freunde gleichzeitig schicken: http://freemail.web.de/features/?mc=021179
Received on Friday, 24 January 2003 12:41:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 June 2012 00:14:11 GMT