W3C home > Mailing lists > Public > www-style@w3.org > October 2005

Re: [validator] Proprietary extensions

From: Mikko Rantalainen <mikko.rantalainen@peda.net>
Date: Tue, 11 Oct 2005 16:46:43 +0300
Message-ID: <434BC243.5050200@peda.net>
To: www-style@w3.org

Jens Meiert wrote:
>>>They work in IE, but they're not valid...hence the validator is 
>>>behaving correctly. "Clearly work" does not necessarily mean valid.
>>
>>Maybe validator could be extended to add such description when 
>>reporting error? (same for behavior, filter, -moz-, -o-, etc.)
> 
> The validator aims to look for specification compliance, at least this is
> what it should do. Though proprietary extensions are generally allowed [1],
> they are not part of the specification, so the validator would then only
> check for correct syntax.
> 
> Also, how would you decide what proprietary properties to be added? Only
> Gecko supported properties using the "-moz-" prefix? Gecko and Microsoft
> properties? Others as well? Where to get the information on changes and
> additions?
> 
> [1] http://www.w3.org/TR/CSS21/syndata.html#q4

I think validator should regognize proprietary extensions as 
described in the spec. That is, scrollbar-related stuff should *not* 
be regognized as a proprietary extension. On the other hand, stuff 
like -xyz-property-name and _xyz-property-name should be marked as 
proprietary. For example, it currently gives error

   Property -moz-border-radius doesn't exist : 0.5em

whereas it could return only warning

   Only syntax checked for proprietary extension: -moz-border-radius

Validator could also strip the vendor identifier from proprietary 
properties and then try to validate the remaining property. For 
example, "-moz-border-radius: 1em" would be mapped to 
"border-radius: 1em". This way, some of the vendor extensions could 
be sort-of-validated.

Also, the warning about proprietary "-vnd-border-radius" could be 
silenced if the same rule also has declaration for "border-radius", 
especially if both have the same value... In this case, it should be 
safe to assume that vendor extension has been used for better 
compatibility only.

-- 
Mikko
Received on Tuesday, 11 October 2005 13:46:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:41 GMT