1. CATransform3D结构成员的意义。
struct CATransform3D
{
CGFloat m11（x缩放）, m12（y切变）, m13（旋转）, m14（）;
CGFloat m21（x切变）, m22（y缩放）, m23（）, m24（）;
CGFloat m31（旋转）, m32（）, m33（）, m34（透视效果，要操作的这个对象要有旋转的角度，否则没有效果。正直/负值都有意义）;
CGFloat m41（x平移）, m42（y平移）, m43（z平移）, m44（）;
};
ps:整体比例变换时，也就是m11==m22时，若m33>1，图形整体缩小，若0<m33<1，图形整体放大，若s<0,发生关于原点的对称等比变换。
（）空的地方以后补充。
2. CATransform3DMakeTranslation
CATransform3DMakeTranslation(0, 0, 0) 创建了一个4*4的单位矩阵。
3. CATransform3DMakeRotation And CATransform3DRotate
CATransform3DMakeRotation()
_transformedLayer = [CALayer layer];
_transformedLayer.frame = self.bounds;
_transformedLayer.anchorPoint = CGPointMake(0.5f, 0.5f);
CATransform3D sublayerTransform = CATransform3DIdentity;
// Set perspective
sublayerTransform.m34 = kPerspective;
[_transformedLayer setSublayerTransform:sublayerTransform];
[self.layer addSublayer:_transformedLayer];
//init Sublayers
CATransform3D t = CATransform3DMakeTranslation(0, 0, 0);
// take snapshot of the current view
[_transformedLayer addSublayer:[self snapshot:t
withView:_contentView
isMasked:YES]];
// 暂时先支持一个方向翻转
RotateDirection direction = RotateFromBottom;
if (YES || direction == RotateFromBottom)
{
CGFloat height = self.bounds.size.height;
//CGFloat cubeSize = 100.0f;
t = CATransform3DRotate(t, D2R(90.0), 1, 0, 0);【1】
t = CATransform3DTranslate(t, 0, height, 0);
CALayer *subLayer = [self snapshot:t withView:view isMasked:YES];
[_transformedLayer addSublayer:subLayer];
}
else
{
}
_newContentView = view;
[self animationCubeRotate:direction withDuration:duration];
4. 翻转的动画
- (void)animationCubeRotate:(RotateDirection)direction
withDuration:(float)duration
{
[CATransaction flush];
CGFloat height = self.bounds.size.height;
CABasicAnimation *rotation;
// CABasicAnimation *translationX; // 如果沿X轴翻转,则用不到这个变量.
CABasicAnimation *translationY; // 如果沿Y轴翻转,则用不到这个变量.
CABasicAnimation *translationZ;
CAAnimationGroup *animationGroup = [CAAnimationGroup animation];
animationGroup.delegate = self;
animationGroup.duration = duration;
if ( direction == RotateFromBottom )
{
// 创建(某方向)关键帧动画.
translationY = [CABasicAnimation animationWithKeyPath:
@"sublayerTransform.translation.y"];
translationY.toValue = [NSNumber numberWithFloat:-(height / 2)];【2】
rotation = [CABasicAnimation animationWithKeyPath:
@"sublayerTransform.rotation.x"];
rotation.toValue = [NSNumber numberWithFloat:D2R(-90.0f)];
}
else if ( direction == RotateFromTop )
{
}
// 处理Z轴
translationZ = [CABasicAnimation animationWithKeyPath:
@"sublayerTransform.translation.z"];
translationZ.toValue = [NSNumber numberWithFloat:height / 2];【3】
animationGroup.animations =
[NSArray arrayWithObjects: rotation, translationY, translationZ, nil];
animationGroup.fillMode = kCAFillModeForwards;
animationGroup.removedOnCompletion = NO;
[_transformedLayer addAnimation:animationGroup forKey:kAnimationKey];
}
made, 我发现这个东西确实很难讲清楚，主要是因为我理论薄弱，
【1】针对X轴旋转，就是1，0，0，针对Y轴旋转，就是0，1，0...下面那行也要进行正确的转换。
【2】此处应该是和 anchorPoint有关系的。
【3】这个值会影响类似于深度的东西，比如说Cube会离我们更近，或者是更远。（但是，似乎不算是透视关系）
