[Webkit-unassigned] [Bug 11667] SVG: method .getTransformToElement(elt) in SVGLocatable not implemented

[bugzilla-daemon at webkit.org]

http://bugs.webkit.org/show_bug.cgi?id=11667





------- Comment #6 from rwlbuis at gmail.com  2006-11-27 03:17 PDT -------
Hi Eric,

(In reply to comment #5)
> (From update of attachment 11636 [edit])
> Ok, I have two concerns.
> 
> 1. We're mallocing when not necessary:
>  +    SVGMatrix* ctm = SVGSVGElement::createSVGMatrix();
> +    ctm->multiply(startctm);
> 
> In the error condition, no malloc is needed.  Ideally we would just move off of
> SVGMatrix internally (this function could take/return AffineTransforms, and let
> something outside of it do the wrapping).  I'm not sure it's worth going
> through contortions to avoid mallocs in error conditions however.  Consider
> making this an AffineTransform in/out function though.

I agree in theory, this is AFAIK a gray area since I don't think it is
specified anywhere
what to return in error conditions, ie. null or some "real" matrix. Also it
could wait until
we adopt AffineTransform more for internal use, where we use SVGMatrix now.

> 2.  We need tests for the error conditions.  I see this behavior:
> +        if (ec)
> +            return ctm;
> 
> Does that match the spec?  Does that match firefox/opera?  It would be good to
> test.

I added it in the test (now othermaciej pointed out how to do it).  Will attach
a new patch shortly.

> Otherwise it looks fine.

Great.
Cheers,

Rob.


-- 
Configure bugmail: http://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.