Re: [CSS21] response to issue 115 (and 44)

On Sunday, February 22, 2004, 2:22:43 PM, Bert wrote:


BB> Boris Zbarsky writes:
>> I'll start maybe believing that when the W3C site sheets happen to have some
>> encoding information attached to them.... 

BB> Working on it... :-)

BB> Note that HTML files are correctly labeled on the W3C site and have
BB> been for a long time. I already asked internally if it is feasable for
BB> CSS files as well.

I suggested internally that writing a script to prepend an @charset
onto any CSS file that does not have it is not difficult, and then
calling this script from 'find' is an easy way to do a site-wide
update.

The advantage of this method is that one does not need to change the
server config, and also the CSS files continue to work correctly when
read over the filesystem instead of HTTP (for example, when making a
zipped version of all the files in a specification).

BB> Anybody knows what browsers will break if we send a "charset="
BB> parameter after the "text/css" MIME type?. And if there would be any
BB> work-arounds?

The CSS 2.1 testsuite will presumably test this?




-- 
 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 Member, W3C Technical Architecture Group

Received on Sunday, 22 February 2004 19:14:51 UTC