W3C home > Mailing lists > Public > public-fx@w3.org > January to March 2011

Re: Some practical issues integrating the SVG transform attribute with CSS transform properties

From: Dirk Schulze <vbs85@gmx.de>
Date: Sat, 26 Mar 2011 18:53:55 +0100
Cc: Anthony Grasso <Anthony.Grasso@cisra.canon.com.au>, Jonathan Watt <jwatt@jwatt.org>, Cameron McCormack <cam@mcc.id.au>, "public-fx@w3.org" <public-fx@w3.org>
Message-Id: <221E370C-015F-47D3-9EFD-7B30D84CE46D@gmx.de>
To: robert@ocallahan.org

Am 25.03.2011 um 22:51 schrieb Robert O'Callahan:

> On Fri, Mar 25, 2011 at 7:50 PM, Anthony Grasso <Anthony.Grasso@cisra.canon.com.au> wrote:
> I remember from past discussions in the FX Group that the “transform” attribute and “transform” property behave differently. At least for “transform-origin” it’s currently defined that way [1] . We originally decided that the CSS property will override the SVG attribute (which is how it works with current SVG I think).
> Have you got a reference to those discussions? It's not clear to me what "CSS property overrides the SVG attribute" would actually mean. Does it mean something different from the way presentational attributes normally map into CSS?
I guess he means the normal mapping like on presentation properties.

>  People have commented to me that would like the current proposal changed such that CSS transform property is additive to the SVG transform attribute. That is, the CSS transform property will apply to SVG elements based on the bounding box of the SVG element with respect to the top-level document. This differs from the current proposal for the CSS transform property to override the SVG transform attribute. It resolves the syntax/parsing issue and makes sense in developer scenarios, particularly concerning animation.
> Can you be more specific? This would mean an element actually has two transforms: an SVG transform and a separate CSS transform on top. I guess that would work. However, it means that setting an SVG transform attribute behaves differently to setting the CSS transform property.
> I'd be reluctant to do something different to the normal way presentational attributes are mapped into CSS.
If we have two different transformations, SVG and CSS, it would solve our problems with different kinds of transformations between the both candidates (which still exist IIRC) as well as the transform-origin issue. On the other side, if you want to know the transformation of the current object, you always  have to check both, CSS and SVG transformations. Also, which transformation would get applied to the object first? Do authors know of the differences between the two kinds of transformations? The normal mapping like mentioned above is less confusing in my eyes.

Received on Saturday, 26 March 2011 18:01:16 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:49:38 UTC