Fwd: [Moderator Action] getCTM / SVGVector

You must be subscribed to post.

----- Forwarded message from Alex Fritze <alex@croczilla.com> -----

X-Envelope-From: www-svg-request@tux.w3.org  Tue Sep 25 10:39:16 2001
From: "Alex Fritze" <alex@croczilla.com>
To: "Www-Svg" <www-svg@w3.org>
Old-Date: Tue, 25 Sep 2001 15:29:05 +0100
Importance: Normal
X-Diagnostic: Not on the accept list
Subject: [Moderator Action] getCTM / SVGVector
X-Diagnostic: Mail coming from a daemon, ignored
X-Envelope-To: www-svg
X-UIDL: 543b12fc4c968019111e13adc1cb7bdb

Hi,

I have a couple of questions about the SVG specs:

1.
Should SVGSVGElement::getCTM() return a transformation from its viewport
coordinate system (and to the nearestVieportElement's viewport coordinate
system) or should it return a transformation from its viewbox coordinate
system?
To rephrase: Does the 'current user space' of a viewport element include the
'viewbox to viewport' transformation? The specs say that the viewbox to
viewport transformation doesn't apply to some parameters (x, y, width,
height), so I'm not quite sure.
Also what should happen if getCTM() is called on the outermost element, i.e.
when nearestViewportElement is null? Should it return the identity matrix?
Or throw an exception?

2.
One data type I would have found very useful in SVG is an SVGVector. I.e.
something that has the same interface as a point, but doesn't translate when
transformed. Has it been deliberately omitted?


Thanks,
Alex


----- End forwarded message -----

Received on Tuesday, 25 September 2001 10:50:12 UTC