Re: Proposal: transformation matrix interface

Hi David,


On Mar 13, 2013, at 4:21 PM, David Sheets <kosmo.zb@gmail.com> wrote:

> On Sat, Mar 9, 2013 at 5:50 AM, Dirk Schulze <dschulze@adobe.com> wrote:
>> On Feb 14, 2013, at 3:21 PM, David Sheets <kosmo.zb@gmail.com> wrote:
>>> 2. Is a DOMException necessary for the error conditions? Are
>>> DOMExceptions generateable by user code? I would like to be able to
>>> provide user-space implementations of Matrix that internally leverage
>>> invariants for better performance/behavior. For example, if I know
>>> that I am only dealing with rotation matrices, I might use an
>>> implementation of Matrix that internally uses only a quaternion
>>> representation. Similarly, if I know that I am only going to use
>>> special orthogonal transformations (no scale, no skew, det=1), I might
>>> use a dual quaternion representation internally. It would be nice to
>>> be able to pass these constructions into APIs that expect Matrix
>>> objects without building a native Matrix first.
>> 
>> There is a DOMException interface to generate DOMExceptions. You can prototype Matrix with your own methods and attributes that can throw your generated exceptions. Is that sufficient for you?
> 
> Does that provide identical semantics? Has something changed since
> Marcos' SO answer
> <http://stackoverflow.com/questions/5136727/manually-artificially-throwing-a-domexception-with-javascript#answer-9856490>
> and the thread on public-script-coord
> <http://lists.w3.org/Archives/Public/public-script-coord/2012JanMar/0458.html>?

The Matrix interface does not allow you do create new individual DOMExceptions if you mean that. But you can prototype the interface like this:

window.Matrix.prototype.rotate  = function ...

and throw a DOMException inside this function.

> 
> Is there some way to make instanceof behave identically? Or do Matrix
> users need to wrap a native Matrix to poke for exceptions? What
> happens when a script in a JS environment without DOM (e.g. node.js)
> wants to use a Matrix?

The object lives on window, means you need the usual hacks on node.js.

> 
>>> 4. In DecomposedMatrix, "quaternions of type sequence<double> Is an
>>> sequence of four double items representing the two quaternions for the
>>> rotation of the matrix." In the pseudocode for the decomposition, I
>>> only see a single quaternion generated. The other quaternion that
>>> describes the rotation is simply the negation of this quaternion as H
>>> double covers SO(3). I believe that the decomposition should always
>>> return the quaternion that represents the shortest rotation which
>>> corresponds to the quaternion with a positive scalar component
>>> (quaternion[3]?). This matters when an API user wants to take the
>>> logarithm of a quaternion and retrieve consistent bivector angles.
>> 
>> At the moment Matrix follows CSS3 Transforms and the implementation behavior of browsers. The definition seems bogus, since we just have one quaternion with 4 arguments, not two. I will correct the specification text. In the meantime, can you go into more details how you suggest this operation should look like?
> 
> I think the provided pseudocode implements what I am proposing the
> prose describe. Specifically, I believe that the prose should indicate
> that a single quaternion is returned which has a non-zero scalar part
> and L_2 magnitude of 1 ("unit quaternion"). It might also mention that
> this implies that the returned quaternion represents the geodesic
> (shortest path) for interpolators. To produce the complementary arc,
> the quaternion's elements simply need scaling by -1. This requirement
> is quite cheap for implementors, nicely uniform for users, and a good
> invariant.

Do you have a link/document that I can reference?

Greetings,
Dirk

> 
> Thanks for your work.
> 
> Hope this helps,
> 
> David

Received on Friday, 15 March 2013 12:04:34 UTC