Re: CALayer transform, setting the Y value of rotation causes side-effects
Re: CALayer transform, setting the Y value of rotation causes side-effects
- Subject: Re: CALayer transform, setting the Y value of rotation causes side-effects
- From: Frédéric Testuz <email@hidden>
- Date: Mon, 14 Jul 2008 21:19:04 +0200
Le 14 juil. 08 à 18:57, John Clayton a écrit :
Hi All,
I'm setting the rotation values of a CALayer, and notice that when
setting the Y component of rotation of a CATransform3D structure via
the :
[layer setValue:someValue forKeyPath:@"transform.rotation.y"]
call, that in certain cases of 'someValue' (the y rotation), the X
and Z rotation values are modified as well.
I set the value of the transform.rotation.Y using the standard
setValue:forKeyPath: method, and logged the resulting X, Y and Z
rotation values. Here's the code and logging results:
- (void) awakeFromNib {
CALayer* testLayer = [CALayer new];
NSLog(@"before: x, y, x = %@ / %@ / %@",
[testLayer valueForKeyPath:@"transform.rotation.x"],
[testLayer valueForKeyPath:@"transform.rotation.y"],
[testLayer valueForKeyPath:@"transform.rotation.z"]);
// but this breaks the X and Z, how come?
[testLayer setValue:[NSNumber numberWithDouble:2.12041]
forKeyPath:@"transform.rotation.y"];
NSLog(@"after: x, y, x = %@ / %@ / %@",
[testLayer valueForKeyPath:@"transform.rotation.x"],
[testLayer valueForKeyPath:@"transform.rotation.y"],
[testLayer valueForKeyPath:@"transform.rotation.z"]);
[testLayer release];
}
2008-07-14 17:48:23.555 Annotate[42997:10b] before: x, y, x = 0 /
-0 / 0
2008-07-14 17:48:23.562 Annotate[42997:10b] after: x, y, x =
3.141593 / 1.021183 / 3.141593
Huh?
Question 1: Why do X and Z get very close to PI?
Question 2: Why does Y end up being 1.02 when I set it to 2.12041?
Does anyone know what's going on here? I'm a little stumped by this
one.
Hello,
I never used CALayer but if I understand correctly the rotation, there
is no difference between
rotation (A) : x, y, x = 0 / 2.12041 / 0
and
rotation (B) : x, y, x = 3.141593 / 1.021183 / 3.141593
It's because a rotation of pi is a half-circle rotation and 1.021183
is equal to pi-2.12041. So the two transformation are equals. But why
the layer transform the rotation like this. It has to be a internal
check like in 2D when you want the angle of rotation to stay between -
pi and pi.
You can check the equality by doing the transformation for an
arbitrary point, or calculating the transformation matrix.
rot(A) : x,y,z = 0 , a , 0
(cos a 0 -sin a 0)
R(A) = ( 0 1 0 0)
(sin a 0 cos a 0)
( 0 0 0 1)
rot(B) : x,y,z = pi, pi-a, pi
(1 0 0 0) (cos pi-a 0 -sin pi-a 0) ( cos pi
sin pi 0 0)
R(B) = (0 cos pi sin pi 0) ( 0 1 0 0) (-sin pi
cos pi 0 0)
(0 -sin pi cos pi 0) (sin pi-a 0 cos pi-a 0)
( 0 0 1 0)
(0 0 0 1) ( 0 0 0 1)
( 0 0 0 1)
(1 0 0 0) (-cos a 0 -sin a 0) (-1 0 0 0)
R(B) = (0 -1 0 0) ( 0 1 0 0) ( 0 -1 0 0)
(0 0 -1 0) ( sin a 0 -cos a 0) ( 0 0 1 0)
(0 0 0 1) ( 0 0 0 1) ( 0 0 0 1)
(-cos a 0 -sin a 0) (-1 0 0 0)
R(B) = ( 0 -1 0 0) ( 0 -1 0 0)
(-sin a 0 cos a 0) ( 0 0 1 0)
( 0 0 0 1) ( 0 0 0 1)
(cos a 0 -sin a 0)
R(B) = ( 0 1 0 0)
(sin a 0 cos a 0)
( 0 0 0 1)
And you have R(A) = R(B)
Frédéric Testuz_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden