[dojo-contributors] Re: Draft: Unified 2D graphics subsystem for Dojo

gavin at dfmm.org gavin at dfmm.org
Thu Jun 8 09:06:24 EDT 2006

> There is a problem with "something manipulatable". I think it is
> possible to preserve a node name during the transformation, but it
> doesn't do a lot of good for manipulation purposes because VML and SVG
> handle it differently.

I would assume that if we instantiate a "dojoRectangle" from a node, be it
svg, vml, or a div, that the dojoRectangle implementation on that browser
would be specific in its implementation (though generic in its interface).

> It may be even more hairy to change graphical attributes or apply a
> transformation because it may involve adding more nodes or even changing
> the hierarchy of existing nodes, e.g., putting them under a parent to
> facilitate some group changes.

My initial attempts suggest that using the native group transforms is
often *not* the way to go. For example, if you group several SVG shapes
together and scale them up, the stroke weight is scaled also, often
unevenly. You can see this behavior in my drawing editor prototype. I had
to do some silly "re-apply the group xform to the elements inside the
group" when ungrouping them as well.

>> Already have the paper and have digested a number of things from it,
>> thanks.  The main thing I'm looking at is the need to include script
>> in the transformation process; I don't think I can load it on the fly
>> though, which means there's probably be some duplication of JS stuff
>> :(   I may see if I can figure out a way of loading that code, but
>> we'll see.

Yeah, I didn't get that either. Possibly was using jscript to emulate SVG
features not in VML?? I also emailed the Italian guy who was doing the VML
translator XSLT but haven't heard back yet.

(Still in Florida with lousy internet, but will be busy on this stuff all

More information about the dojo-contributors mailing list