W3C home > Mailing lists > Public > public-fx@w3.org > July to September 2011

Re: [filters] notes to CSS filters

From: Dirk Schulze <vbs85@gmx.de>
Date: Fri, 23 Sep 2011 07:02:19 +0200
Cc: public-fx@w3.org
Message-Id: <A803022B-61FB-46D0-912F-66BF38615B35@gmx.de>
To: robert@ocallahan.org

Am 23.09.2011 um 01:51 schrieb Robert O'Callahan:

> On Thu, Sep 22, 2011 at 10:15 PM, Dirk Schulze <vbs85@gmx.de> wrote:
> For the current CSS shorthand filter effects it is quite easy to omit filter regions. The implementations can calculate the boundaries
> themselves. What is the status on the SVG part? Do we find a way to limit the filter region size automatically?
> 
> Yes.
>  
> Right now one of the filter effects that can have unlimited size is lightning (same for feTurbulence IIRC).
> However, Firefox limits the size of a lighting effect automatically! You'll see that if you compare the example [3] between Firefox and Chrome/Opera. The size of the filter result on Firefox is smaller. Even if it is a bug in Firefox at the moment, we could add a rule to the filter specification that would specify this behavior. Once we've done that, we can omit filterRegions (or make it optional) and don't need to auto clip to (-10%, -10%, 120%, 120%) anymore.
> (Firefox takes the size of the previous filter effect and uses the same size for the lighting effect if no subregion was specified).
> 
> That's a bug. Maybe we don't need to do anything in the spec; if primitive regions are not specified the unbounded primitives can just fill the entire visible (clipped) area.
But that is exactly the point. If we don't have a filter region, the filter effect would be unbound! Right now just the filter region bounds lightning effects. What else can clip the visible area?

Dirk
Received on Friday, 23 September 2011 05:02:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 23 September 2011 05:02:50 GMT