I just wanted to add:
On 4/8/2013 2:19 PM, Jim Graham wrote:
It's already done and the code is written, but in retrospect, I'm not sure how
much the optimizations in getTransform/setTransform are worth compared to the
easy readability of having the transforms done linearly in a LIFO manner, such
as:
setTransform:
transform.setToIdentity();
if (cX|cY != 0) { transform.translate(cX, cY); }
if (dS != 1) transform.scale(dS, dS); }
transform.concatenate(Tx);
getTransform:
tx = new AT(transform);
if (dS != 1) { tx.scale(invDs, invDs); }
if (cX|cY != 0) { tx.translate(-cX, -cY); }
return tx;
(I don't think it should be changed, but I'm throwing it out there as food for
thought as it may be easier to maintain in the future, but it could also change
the performance as I'm just speculating from how I remember the optimizations
work...)
I had a sudden "uh oh" about the commutativity of the operations done in
setTransform/getTransform (the ones in the webrev), but I just ran the math and it does
appear to be a correct inverse operation (probably because it is a uniform scale and a
translate which are more benign than other matrix ops). So, there's a double check on
the new math...
...jim
