kPOPViewScaleXY issue when view's scale set to [0.0, 0.0] to start about pop HOT 5 CLOSED

This is not a pop specific behaviour.

You can't create a transform matrix of 0 like this. CGAffineTransformMakeScale(0.0, 0.0);

I believe that the problem is that your matrix cannot be inverted when it has a 0.0 scale, but it CAN when you have a 0.001 scale. You'll run in all kinds of nasty divide by zero errors when inverting that matrix.

from pop.

Thanks for the report. The best workaround, and in fact what we've employed in Paper, is to replace zero with a minuscule, but not-quite-zero scale value.

from pop.

@depl0y I'm converting some code that currently uses JNWSpringAnimation just to get my feet a little wet and it works fine with a starting scale of 0.0.

myButton.transform = CGAffineTransformMakeScale(0.0, 0.0);
JNWSpringAnimation *decay = [JNWSpringAnimation animationWithKeyPath:@"transform.scale"];
decay.damping = 35;
decay.stiffness = 1000;
decay.mass = 3;
decay.fromValue = @(0);
decay.toValue = @(1);
[myButton.layer addAnimation:decay forKey:@"transform.scale"];

Of course I'm by no means a matrix transformation expert so perhaps Jonathan has a special case in his code to change a 0 to a very small but non-zero number. Not sure.

@kimon Thanks! I'd say that a nice-to-have would be to notice this starting-at-0.0f-scale use case and adjust to something very small automatically, but that's probably more of a hack than something that'd belong in a library like this. Feel free to re-open if you think it's worth updating.

from pop.

Ran into this and the above was helpful! Wanted to add an additional case for those who find it in the future:

When animating using a POPSpringAnimation, the spirngyness causes the scale to go below (0,0), which inverts things as described above.

To remedy this, you can apply a clampMode to prevent it from dipping below 0,0

thanks!

from pop.

不明觉厉

from pop.