I\'m trying to use CGAffineTransformMakeRotation, and I refer Apple document which says
The angle, in radians, by which this matrix rotates the coordi
To get anti-clockwise rotation.. make ur angle in -ve eg:
*
imageToBeRotated.transform = CGAffineTransformMakeRotation(-angle); // for anti-clockwise rotation
imageToBeRotated.transform = CGAffineTransformMakeRotation(angle); // for clockwise rotation
*
Don't get confuse. Its wrong in apple documentation.
CGAffineTransformMakeRotation(-angle) // for anti-clockwise rotation CGAffineTransformMakeRotation(angle) // for clockwise rotation
This is actually correct behavior, and the documentation describes it correctly.
You're dealing with two different coordinate systems here. Core Graphics' coordinate system has (0, 0) on the bottom left, but UIKit's coordinate system has (0, 0) on the top left. Both coordinate systems have an X-axis with 0 on the left. In other words, UIKit's coordinate system is the Core Graphics coordinate system flipped vertically. The following image demonstrates the difference:
Now, let's look at your quote:
The angle, in radians, by which this matrix rotates the coordinate system axes. In iOS, a positive value specifies counterclockwise rotation and a negative value specifies clockwise rotation.
At first glance that seems wrong, but you've missed something. This is a quote from the CGAffineTransformMakeRotation reference. This is part of the Core Graphics reference, not the UIKit reference. It's referring to the rotation as counterclockwise in Core Graphics' coordinate system.
You can picture this rotation in your head by flipping what you're seeing on screen vertically and imagining what the rotation would look like.
In fact, the CGAffineTransformMakeRotation reference has a discussion item which goes on to document this:
The actual direction of rotation is dependent on the coordinate system orientation of the target platform, which is different in iOS and macOS.
The discussion item on CGContext's rotate(by:) method reference describes this behavior a little better:
The direction that the context is rotated may appear to be altered by the state of the current transformation matrix prior to executing this function. For example, on iOS, a UIView applies a transformation to the graphics context that inverts the Y-axis (by multiplying it by -1). Rotating the user coordinate system on coordinate system that was previously flipped results in a rotation in the opposite direction (that is, positive values appear to rotate the coordinate system in the clockwise direction).
I have a simple code to rotate image in diffrent direction.
int Count==1;
- (IBAction)Btn_Image:(id)sender {
if (Count==1) {
Count = Count + 1;
UIImage *image = [UIImage imageWithCGImage:self.Img_Selected.CGImage scale:1.0 orientation:UIImageOrientationRight];
self.Img_Canvas.image = image;
}
else if (Count == 2) {
Count = Count + 1;
UIImage *image = [UIImage imageWithCGImage:self.Img_Selected.CGImage scale:1.0 orientation:UIImageOrientationDown];
self.Img_Canvas.image = image;
}
else if (Count == 3) {
Count = Count + 1;
UIImage *image = [UIImage imageWithCGImage:self.Img_Selected.CGImage scale:1.0 orientation:UIImageOrientationLeft];
self.Img_Canvas.image = image;
}
else if (Count == 4) {
Count = 1;
UIImage *image = [UIImage imageWithCGImage:self.Img_Selected.CGImage scale:1.0 orientation:UIImageOrientationUp];
self.Img_Canvas.image = image;
}}
This is an error in documentation.
Here they describe matrix and provide equations which they use to do the transform:
If you do the math, you'll see that for positive angles, it gives rotation counterclockwise when Y axis direction is upwards as in OS X and clockwise when Y axis directed downwards as in iOS
For example you may try to transform point (5,0) with 90 degrees, you will get (0,5), which means clockwise for iOS and counterclockwise for OS X.
The same piece of documentation still have the (false) part you quoted:
In iOS, a positive value specifies counterclockwise rotation and a negative value specifies clockwise rotation. In OS X, a positive value specifies clockwise rotation and a negative value specifies counterclockwise rotation.
Which is clearly wrong, because the equations (which are the part of the same document and are pretty standard rotation equations) say the opposite.