W3C home > Mailing lists > Public > www-svg@w3.org > June 2006

RE: [SVGMobile12] script element processing

From: Ian Hickson <ian@hixie.ch>
Date: Wed, 28 Jun 2006 21:44:26 +0000 (UTC)
To: Doug Schepers <doug.schepers@vectoreal.com>
Cc: 'Anne van Kesteren' <annevk@opera.com>, www-svg@w3.org
Message-ID: <Pine.LNX.4.62.0606282139450.4826@dhalsim.dreamhost.com>

On Tue, 20 Jun 2006, Doug Schepers wrote:
> Ian Hickson
> |
> | HTML5 and XBL2 define how you handle this case (as does 
> | CSS, in the external case). HTML4, XHTML1, and SVG don't, 
> | however, as you point out.
> Has this proven to be an issue with implementations in the past? Are 
> there UAs that treat the contents of script blocks differently?

Yes, for XHTML at least there was one time where all the UAs I tested had 
widely varying handling of cases like:

    bla bla
    <x> bla bla </x>
    bla bla

In fact, I have even seen browsers differ on their handling of <script> 
and <style> blocks that contain only text nodes and CDATA nodes, since 
the specs don't define exactly how that should work either.

I haven't tested browsers recently, but as Anne mentioned, for HTML and 
XHTML this is no longer ambiguous at the spec level (since the Web Apps 
spec now defines this for HTML5 and XHTML5).

> | It would be lovely for SVG to define things like this, 
> | especially since the SVG group announced their intent 
> | to define error handling.
> Could you please define "things like this"?

The handling of unexpected content.

> My reason for this is simple. I agree that well-defined error handling 
> and parsing rules are a laudable goal, but I don't think that this is 
> something specific to SVG. Instead of describing such things in the SVG 
> specification, when they are needed in CSS, HTML4, XHTML1, and SVG alike 
> (as well as many other W3C languages, I'm sure), I propose that we try 
> to create a normatively referenceable document that can concentrate on 
> exactly these issues, and be broadly applicable to all relevant W3C 
> documents.

I don't really have an opinion on how this should get defined, so long as 
implementations have a clear and unambiguous (set of) specification(s) to 
work from.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 28 June 2006 22:56:18 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:08 UTC