Re: object-oriented access?

Steve,
I believe you are asking why the only way which we allow you to modify the
attributes on the animation elements is via the XML DOM (i.e., getAttribute
and setAttribute), whereas nearly every other attribute in SVG can be
accessed on an object-oriented approaches such as 'MyRect.x'.

The reasons why the only access to the animation attributes is via the XML
DOM:

1) The whole animation facility in SVG was jointly developed by the SYMM
and SVG working groups and is meant to be a standard facility for animating
any XML grammar, including XHTML. Thus, the owners of the specification for
the core animation facilities (i.e., SMIL Animation) is the SYMM working
group. The SYMM working group did not give a priority to defining special
DOM facilities for the animation elements, and the SVG working group did
not feel that these facilities were so critical as to warrant the SVG group
doing something on its own.

2) The animation elements are very complicated. Defining object-oriented
access to the various attributes would take lots of work, and there is
already plenty of things to do.

3) The SYMM working group has stated that the semantics of changing the
animation elements via the DOM is not well-defined at this time (what
exactly should happen if you change an animation while the animation is
running), and thus they are discouraging people from doing this until these
issues can be sorted through.

Jon Ferraiolo
SVG Editor
Adobe Systems Incorporated


At 01:24 PM 5/23/00 -0700, Steve Dickson wrote:
>In several places in the WD of 2000/03/03, the sentence "Object-oriented
>access to the attributes of the <X> element via the SVG DOM is not
>available' occurs.  An example can be found in the definition of the
>SVGSetElement interface.  Does this mean that the user is expected to
>access and/or modify these attributes through DOM calls such as
>getAttribute() and setAttribute()?  If so, what was the rationale behind
>not providing access/mutator functions for these attributes in the SVGDOM
>itself?
>S.
>

Received on Thursday, 25 May 2000 01:08:15 UTC