Re: Bounding box of a group with a transformed child

• From: Boris Zbarsky <bzbarsky@MIT.EDU>
• Date: Mon, 26 May 2014 21:47:59 -0400
• To: www-svg@w3.org
• Message-ID: <5383EECF.2030407@mit.edu>

On 5/26/14, 5:14 AM, Dr. Olaf Hoffmann wrote:
> My guess - it puts a rectangle around the ellipse and determines the
> boundingBox for this shape, not for the ellipse.

Or (more likely in terms of logic flow, but the result is the same), 
computes the bounding box for the untransformed shape and only applies 
transforms after that.  This will do the right thing for translate and 
scale transforms, but seems pretty clearly wrong for rotations.

-Boris

Received on Tuesday, 27 May 2014 01:48:29 UTC