I need to darken a UIImageView when it gets touched, almost exactly like icons on the springboard (home screen).
Should I be added UIView with a 0.5 alpha and black background. This seems clumsy. Should I be using Layers or something (CALayers).
I would let a UIImageView handle the actual drawing of the image, but toggle the image to one that's been darkened in advance. Here's some code I've used to generate darkened images with alpha maintained:
+ (UIImage *)darkenImage:(UIImage *)image toLevel:(CGFloat)level
{
// Create a temporary view to act as a darkening layer
CGRect frame = CGRectMake(0.0, 0.0, image.size.width, image.size.height);
UIView *tempView = [[UIView alloc] initWithFrame:frame];
tempView.backgroundColor = [UIColor blackColor];
tempView.alpha = level;
// Draw the image into a new graphics context
UIGraphicsBeginImageContext(frame.size);
CGContextRef context = UIGraphicsGetCurrentContext();
[image drawInRect:frame];
// Flip the context vertically so we can draw the dark layer via a mask that
// aligns with the image's alpha pixels (Quartz uses flipped coordinates)
CGContextTranslateCTM(context, 0, frame.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGContextClipToMask(context, frame, image.CGImage);
[tempView.layer renderInContext:context];
// Produce a new image from this context
CGImageRef imageRef = CGBitmapContextCreateImage(context);
UIImage *toReturn = [UIImage imageWithCGImage:imageRef];
CGImageRelease(imageRef);
UIGraphicsEndImageContext();
[tempView release];
return toReturn;
}
How about subclassing UIView and adding a UIImage ivar (called image)? Then you could override -drawRect: something like this, provided you had a boolean ivar called pressed that was set while touched.
- (void)drawRect:(CGRect)rect
{
[image drawAtPoint:(CGPointMake(0.0, 0.0))];
// if pressed, fill rect with dark translucent color
if (pressed)
{
CGContextRef ctx = UIGraphicsGetCurrentContext();
CGContextSaveGState(ctx);
CGContextSetRGBFillColor(ctx, 0.5, 0.5, 0.5, 0.5);
CGContextFillRect(ctx, rect);
CGContextRestoreGState(ctx);
}
}
You would want to experiment with RGBA values above. And, of course, non-rectangular shapes would require a bit more work - like a CGMutablePathRef.
UIImageView can have multiple images; you could have two versions of the image and switch to the darker one when needed.
I know it's possible to tint a rectangular image by drawing a CGContextFillRect over it and setting the blend mode. However, I can't figure out how to do a tint on a transparent image such as an icon. It must be possible since the SDK does it itself on tab-bars in such. Would anyone be able to provide a snippet?
UPDATE:
Lots of great suggestions have been given for this problem since I originally asked. Be sure to read through all the answers to figure out what suits you best.
UPDATE (Apr 30, 2015):
With iOS 7.0, I can now just do the following, which would satisfy the needs of my original question. But if you have more complicated cases, check out all the answers.
UIImage *iconImage = [[UIImage imageNamed:#"myImageName"] imageWithRenderingMode:UIImageRenderingModeAlwaysTemplate];
UIImageView *icon = [[UIImageView alloc] initWithImage:iconImage];
icon.tintColor = [UIColor redColor];
Update: Here is a Gist for a Swift UIColor extension using the code below.
If you have a greyscale image and want white become the tinting color, kCGBlendModeMultiply is the way to go. With this method, you cannot have highlights lighter than your tinting color.
On the contrary, if you have either a non-greyscale image, OR you have highlights and shadows that should be preserved, the blend mode kCGBlendModeColor is the way to go. White will stay white and black will stay black as the lightness of the image is preserved. This mode is just made for tinting - it is the same as Photoshop's Color layer blend mode (disclaimer: slightly differing results may happen).
Note that tinting alpha-pixels does not work correctly neither in iOS nor in Photoshop - half-transparent black pixels would not stay black. I updated the answer below to work around that issue, it took quite a long time to find out.
You can also use one of the blend modes kCGBlendModeSourceIn/DestinationIn instead of CGContextClipToMask.
If you want to create a UIImage, each of the following code sections can be surrounded by the following code:
UIGraphicsBeginImageContextWithOptions (myIconImage.size, NO, myIconImage.scale); // for correct resolution on retina, thanks #MobileVet
CGContextRef context = UIGraphicsGetCurrentContext();
CGContextTranslateCTM(context, 0, myIconImage.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGRect rect = CGRectMake(0, 0, myIconImage.size.width, myIconImage.size.height);
// image drawing code here
UIImage *coloredImage = UIGraphicsGetImageFromCurrentImageContext();
UIGraphicsEndImageContext();
So here's the code for tinting a transparent image with kCGBlendModeColor:
// draw black background to preserve color of transparent pixels
CGContextSetBlendMode(context, kCGBlendModeNormal);
[[UIColor blackColor] setFill];
CGContextFillRect(context, rect);
// draw original image
CGContextSetBlendMode(context, kCGBlendModeNormal);
CGContextDrawImage(context, rect, myIconImage.CGImage);
// tint image (loosing alpha) - the luminosity of the original image is preserved
CGContextSetBlendMode(context, kCGBlendModeColor);
[tintColor setFill];
CGContextFillRect(context, rect);
// mask by alpha values of original image
CGContextSetBlendMode(context, kCGBlendModeDestinationIn);
CGContextDrawImage(context, rect, myIconImage.CGImage);
If your image has no half-transparent pixels, you could also do it the other way around with kCGBlendModeLuminosity:
// draw tint color
CGContextSetBlendMode(context, kCGBlendModeNormal);
[tintColor setFill];
CGContextFillRect(context, rect);
// replace luminosity of background (ignoring alpha)
CGContextSetBlendMode(context, kCGBlendModeLuminosity);
CGContextDrawImage(context, rect, myIconImage.CGImage);
// mask by alpha values of original image
CGContextSetBlendMode(context, kCGBlendModeDestinationIn);
CGContextDrawImage(context, rect, myIconImage.CGImage);
If you don't care for luminosity, as you just have got an image with an alpha channel that should be tinted with a color, you can do it in a more efficient way:
// draw tint color
CGContextSetBlendMode(context, kCGBlendModeNormal);
[tintColor setFill];
CGContextFillRect(context, rect);
// mask by alpha values of original image
CGContextSetBlendMode(context, kCGBlendModeDestinationIn);
CGContextDrawImage(context, rect, myIconImage.CGImage);
or the other way around:
// draw alpha-mask
CGContextSetBlendMode(context, kCGBlendModeNormal);
CGContextDrawImage(context, rect, myIconImage.CGImage);
// draw tint color, preserving alpha values of original image
CGContextSetBlendMode(context, kCGBlendModeSourceIn);
[tintColor setFill];
CGContextFillRect(context, rect);
Have fun!
I had most success with this method, because the others I tried caused distorted colors for semi-transparent pixels for certain color-combinations. This should also be a bit better on the performance side.
+ (UIImage *) imageNamed:(NSString *) name withTintColor: (UIColor *) tintColor {
UIImage *baseImage = [UIImage imageNamed:name];
CGRect drawRect = CGRectMake(0, 0, baseImage.size.width, baseImage.size.height);
UIGraphicsBeginImageContextWithOptions(baseImage.size, NO, 0);
CGContextRef context = UIGraphicsGetCurrentContext();
CGContextTranslateCTM(context, 0, baseImage.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
// draw original image
CGContextSetBlendMode(context, kCGBlendModeNormal);
CGContextDrawImage(context, drawRect, baseImage.CGImage);
// draw color atop
CGContextSetFillColorWithColor(context, tintColor.CGColor);
CGContextSetBlendMode(context, kCGBlendModeSourceAtop);
CGContextFillRect(context, drawRect);
UIImage *tintedImage = UIGraphicsGetImageFromCurrentImageContext();
UIGraphicsEndImageContext();
return tintedImage;
}
After searching around, the best solution I've come to thus far is to use a combination of blend mode and the clipping mask to achieve colorizing/tinting a transparent PNG:
CGContextSetBlendMode (context, kCGBlendModeMultiply);
CGContextDrawImage(context, rect, myIconImage.CGImage);
CGContextClipToMask(context, rect, myIconImage.CGImage);
CGContextSetFillColorWithColor(context, tintColor);
CGContextFillRect(context, rect);
I can get results very close to the tint in the Apple navigation bar by using kCGBlendModeOverlay. Taking excelent #fabb answer and combining #omz approach in this post https://stackoverflow.com/a/4684876/229019 I came with this solution that helds the results I was expecting:
- (UIImage *)tintedImageUsingColor:(UIColor *)tintColor;
{
UIGraphicsBeginImageContextWithOptions (self.size, NO, [[UIScreen mainScreen] scale]);
CGContextRef context = UIGraphicsGetCurrentContext();
CGRect rect = CGRectMake(0, 0, self.size.width, self.size.height);
// draw original image
[self drawInRect:rect blendMode:kCGBlendModeNormal alpha:1.0f];
// tint image (loosing alpha).
// kCGBlendModeOverlay is the closest I was able to match the
// actual process used by apple in navigation bar
CGContextSetBlendMode(context, kCGBlendModeOverlay);
[tintColor setFill];
CGContextFillRect(context, rect);
// mask by alpha values of original image
[self drawInRect:rect blendMode:kCGBlendModeDestinationIn alpha:1.0f];
UIImage *tintedImage = UIGraphicsGetImageFromCurrentImageContext();
UIGraphicsEndImageContext();
return tintedImage;
}
Here's an example tinting several grayscale images with transparency:
:
The first line is the apple toolbar tinted [UIColor orangeColor].
The second line is the same gradient tinted in several colors starting with clear color (= the actual gradient) and ending with the same orange.
The third is a simple circle with transparency (the linen is the background color)
The forth line is a complex dark noisy texture
You could create an UIImage category and do it like this:
- (instancetype)tintedImageWithColor:(UIColor *)tintColor {
UIGraphicsBeginImageContextWithOptions(self.size, NO, 0.0);
CGContextRef context = UIGraphicsGetCurrentContext();
CGRect rect = (CGRect){ CGPointZero, self.size };
CGContextSetBlendMode(context, kCGBlendModeNormal);
[self drawInRect:rect];
CGContextSetBlendMode(context, kCGBlendModeSourceIn);
[tintColor setFill];
CGContextFillRect(context, rect);
UIImage *image = UIGraphicsGetImageFromCurrentImageContext();
UIGraphicsEndImageContext();
return image;
}
In iOS7, they've introduced tintColor property on UIImageView and renderingMode on UIImage. See my example at https://stackoverflow.com/a/19125120/1570970
With iOS 7.0, you can also just do this to tint a basic UIImageView:
UIImage *iconImage = [[UIImage imageNamed:#"myImageName"] imageWithRenderingMode:UIImageRenderingModeAlwaysTemplate];
UIImageView *icon = [[UIImageView alloc] initWithImage:iconImage];
icon.tintColor = [UIColor redColor];
Note that in the accepted answer by fabb, the "surrounding" code for making a UIImage gave me the wrong resolution of images on retina screen. To fix, change:
UIGraphicsBeginImageContext(myIconImage.size);
to:
UIGraphicsBeginImageContextWithOptions(myIconImage.size, NO, 0.0);
The last parameter which is set to 0.0 is scale, and according to Apple documentation:
"If you specify a value of 0.0, the scale factor is set to the scale
factor of the device’s main screen".
Dont have the permission to comment, and editing seems a bit rude, so I mention this in an answer.
Just in case someone encounters this same problem.
UIImageView (or any view for that matter) has a background color which is RGBA. The alpha in the color may do what you need without inventing something new.
Not my work, but i've successfully used this approach:
http://coffeeshopped.com/2010/09/iphone-how-to-dynamically-color-a-uiimage
I wanted to shade my image views in my custom UIButton subclass and the other solutions didn't do what I wanted. I needed to darken "tint" the image color. Here's how to change the brightness using CoreImage.
Make sure you add CoreImage.framework to your project's libraries. (Link Binary with Libraries)
UIImage shade method
- (UIImage *)shadeImage:(UIImage *)image {
CIImage *inputImage = [CIImage imageWithCGImage:image.CGImage];
CIContext *context = [CIContext contextWithOptions:nil];
CIFilter *filter = [CIFilter filterWithName:#"CIColorControls"
keysAndValues:kCIInputImageKey, inputImage,
#"inputBrightness", [NSNumber numberWithFloat:-.5], nil];
CIImage *outputImage = [filter outputImage];
CGImageRef cgImage = [context createCGImage:outputImage fromRect:[outputImage extent]];
UIImage *newImage = [UIImage imageWithCGImage:cgImage scale:image.scale orientation:image.imageOrientation];
CGImageRelease(cgImage);
return newImage;
}
You'll want to store a copy of the context as an ivar, rather than recreate it.
No answers will help me on stack overflow: our designers draw UI elements with various forms, various alpha values (and "alpha-holes"). In most cases, this is 32-Bit PNG file with alpha channel, which comprises black & white pixels (of all possible intensities). After tinting such a picture I had to get this tinted result: white pixels - tinted over more, and dark pixels - less. And all this in view of alpha channel. And i wrote this method for my UIImage category. Maybe it not high efficient, but it work as clock:) Here it is:
- (UIImage *)imageTintedWithColor:(UIColor *)color {
UIGraphicsBeginImageContextWithOptions(self.size, NO, self.scale);
CGContextRef context = UIGraphicsGetCurrentContext();
CGRect rect = CGRectMake(0, 0, self.size.width, self.size.height);
CGContextSetBlendMode(context, kCGBlendModeCopy);
[color setFill];
CGContextFillRect(context, rect);
[self drawInRect:rect blendMode:kCGBlendModeXOR alpha:1.0];
CGContextSetBlendMode(context, kCGBlendModeXOR);
CGContextFillRect(context, rect);
[self drawInRect:rect blendMode:kCGBlendModeMultiply alpha:1.0];
UIImage *coloredImage = UIGraphicsGetImageFromCurrentImageContext();
UIGraphicsEndImageContext();
return coloredImage;
}
First I want to thank fabb for his exceptional solution which has helped me to accomplish my task to tint half transparent Icons. Because I needed a solution for C# (Monotouch) I had to translate his code. Here is what I came up with. Just copy paste this into your code and add your half transparent Image and your done.
So again all credits go to fabb. This is just to kick start C# users :)
//TINT COLOR IMAGE
UIImageView iImage = new UIImageView(new RectangleF(12, 14, 24,24));
iImage.ContentMode = UIViewContentMode.ScaleAspectFit;
iImage.Image = _dataItem.Image[0] as UIImage;
UIGraphics.BeginImageContextWithOptions(iImage.Bounds.Size, false, UIScreen.MainScreen.Scale);
CGContext context = UIGraphics.GetCurrentContext();
context.TranslateCTM(0, iImage.Bounds.Size.Height);
context.ScaleCTM(1.0f, -1.0f);
RectangleF rect = new RectangleF(0,0, iImage.Bounds.Width, iImage.Bounds.Height);
// draw black background to preserve color of transparent pixels
context.SetBlendMode(CGBlendMode.Normal);
UIColor.Black.SetFill();
context.FillRect(rect);
// draw original image
context.SetBlendMode(CGBlendMode.Normal);
context.DrawImage(rect, iImage.Image.CGImage);
// tint image (loosing alpha) - the luminosity of the original image is preserved
context.SetBlendMode(CGBlendMode.Color);
UIColor.Orange.SetFill();
context.FillRect(rect);
// mask by alpha values of original image
context.SetBlendMode(CGBlendMode.DestinationIn);
context.DrawImage(rect, iImage.Image.CGImage);
UIImage coloredImage = UIGraphics.GetImageFromCurrentImageContext();
UIGraphics.EndImageContext();
iImage = new UIImageView(coloredImage);
iImage.Frame = new RectangleF(12, 14, 24,24);
//END TINT COLOR IMAGE
cell.Add(iImage);
Is there a way to add an image in the background of UINavigationBar. There is no direct option through interface builder but I have seen apps implemented this thing.
Any help would be appreciated.
Thanks
Create a UIView and add it as a subview.
Edit: You can now use setBackgroundImage:forBarMetrics:. Source: https://stackoverflow.com/a/7765102/313875
Use this code
UIImage *backgroundImage = [UIImage imageNamed:#"strip.png"];
[upnavbar setBackgroundImage:backgroundImage forBarMetrics:UIBarMetricsDefault];
this wil work.
Just add your background as a subview, as jrtc27 suggested and change the tint color according your needs.
You can also override the drawLayer:inContext: method in a UINavigationBar category class. Inside the drawLayer:inContext: method, you can draw the background image you want to use. You can also use different sized images for portrait and landscape orientations if you'd like to.
- (void) drawLayer:(CALayer *)layer inContext:(CGContextRef)context
{
if ([self isMemberOfClass:[UINavigationBar class]] == NO) {
return;
}
UIImage *image = (self.frame.size.width > 320) ?
[UINavigationBar bgImageLandscape] : [UINavigationBar bgImagePortrait];
CGContextClip(context);
CGContextTranslateCTM(context, 0, image.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGContextDrawImage(context, CGRectMake(0, 0, self.frame.size.width, self.frame.size.height), image.CGImage);
}
And as a complete demo Xcode project on customizing the appearance of UINavigationBar this and this might be helpful.
I recently wrote an article about customization of background for UINavigatioBar and UIToolbar. You can find a code sample and categories for seamless integration of this functionality in your app by following link — http://leonov.co/2011/04/uinavigationbar-and-uitoolbar-customization-ultimate-solution/
You can also override the drawRect function.
#implementation UINavigationBar (UINavigationBarCustom)
- (void)drawRect:(CGRect)rect {
UIImage *image = [UIImage imageNamed:#"nav_bar_logo.png"];
[image drawInRect:rect];
}
#end
My original question:
I'm creating a simple drawing
application and need to be able to
draw over existing, previously drawn content in my drawRect.
What is the proper way to draw on top of existing content
without entirely replacing it?
Based on answers received here and elsewhere, here is the deal.
You should be prepared to redraw the
entire rectangle whenever drawRect
is called.
You cannot prevent the contents from being erased by doing the following:
[self setClearsContextBeforeDrawing: NO];
This is merely a hint to the graphics engine that there is no point in having it pre-clear the view for you, since you will likely need to re-draw the whole area anyway. It may prevent your view from being automatically erased, but you cannot depend on it.
To draw on top of your view without erasing, do your drawing to an off-screen bitmap context (which is never cleared by the system.) Then in your drawRect, copy from this off-screen buffer to the view.
Example:
- (id) initWithCoder: (NSCoder*) coder {
if (self = [super initWithCoder: coder]) {
self.backgroundColor = [UIColor clearColor];
CGSize size = self.frame.size;
drawingContext = [self createDrawingBufferContext: size];
}
return self;
}
- (CGContextRef) createOffscreenContext: (CGSize) size {
CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceRGB();
CGContextRef context = CGBitmapContextCreate(NULL, size.width, size.height, 8, size.width*4, colorSpace, kCGImageAlphaPremultipliedLast);
CGColorSpaceRelease(colorSpace);
CGContextTranslateCTM(context, 0, size.height);
CGContextScaleCTM(context, 1.0, -1.0);
return context;
}
- (void)drawRect:(CGRect) rect {
UIGraphicsPushContext(drawingContext);
CGImageRef cgImage = CGBitmapContextCreateImage(drawingContext);
UIImage *uiImage = [[UIImage alloc] initWithCGImage:cgImage];
UIGraphicsPopContext();
CGImageRelease(cgImage);
[uiImage drawInRect: rect];
[uiImage release];
}
TODO: can anyone optimize the drawRect so that only the (usually tiny) modified rectangle region is used for the copy?
It is fairly common to draw everything in an offscreen image, and simply display this image when drawing the screen. You can read: Creating a Bitmap Graphics Context.
On optimizing drawRect
Try this:
- (void)drawRect:(CGRect)rect {
CGContextRef context = UIGraphicsGetCurrentContext();
CGImageRef cgImage = CGBitmapContextCreateImage(drawingContext);
CGContextClipToRect(context, rect);
CGContextDrawImage(context, CGRectMake(0, 0, self.frame.size.width, self.frame.size.height), cgImage);
CGImageRelease(cgImage);
}
After that you should also comment these lines in your code:
//CGContextTranslateCTM(context, 0, size.height);
//CGContextScaleCTM(context, 1.0, -1.0);
Also created a separate question to be sure that it's the optimal way.
This seems like a better method than I've been using. That is, in a touches event I make a copy of the view about to be updated. Then in drawRect I take that image and draw it to the view and make my other view changes at the same time.
But this seems inefficient but the only way I figured out how to do it.
This prevents your view from being erased before drawRect is done:
[self.layer setNeedsDisplay];
Also, I find it is better to do all the drawing in the drawRect method (unless you have a good reason not to). Drawing offscreen and transferring takes more time and adds more complexity then simply drawing everything once.
I am adding a custom background for my UINavigationBar. It works fine as long as the phone is in portrait mode. As soon as I switch to landscape mode, half the bar appears blue (the default navbar color) and half of it has my image
How can I stretch the image for landscape mode and make it small again for portrait mode?
Thanks
Solution
Incase anyone is looking for an answer to how to add an image to navigation bar - here goes
UIImageView *imgView = [[UIImageView alloc] initWithFrame:CGRectMake(0.0, 0.0, 480.0, 44.0)];
[imgView setImage:[UIImage imageWithContentsOfFile:[[NSBundle mainBundle] pathForResource:#"navbar_landscape" ofType:#"png"]]];
[navigationController.navigationBar addSubview:imgView];
[imgView release];
In both screen orientation modes it's much better to use
[navigationController.navigationBar insertSubview:imgView atIndex:0];
This puts image view under all other views and all the default navigation bar elements (title, standard buttons) work OK.
After a bit of research and trail and error, I found a work around, that will not replace the navbar when you enter the movie playback mode. Hopeefully this does not cause problems with app approval, but based on this post, I think it should be fine:
http://developer.apple.com/iphone/library/qa/qa2009/qa1637.html
#implementation UINavigationBar (UINavigationBarCategory)
- (void)drawLayer:(CALayer *)layer inContext:(CGContextRef)ctx
{
if([self isMemberOfClass: [UINavigationBar class]]){
UIImage *image = [UIImage imageNamed:#"navBarBackground.png"];
CGContextClip(ctx);
CGContextTranslateCTM(ctx, 0, image.size.height);
CGContextScaleCTM(ctx, 1.0, -1.0);
CGContextDrawImage(ctx, CGRectMake(0, 0, self.frame.size.width, self.frame.size.height), image.CGImage);
}else{
[super drawLayer:layer inContext:ctx];
}
}
#end
You probably need to set the autoresizingMask of your background image view; try using UIViewAutoresizingFlexibleWidth | UIViewAutoresizingFlexibleHeight
The solution Ben gave did solve the problem but it stretches the image in landscape mode. I ended up creating two images- one for landscape and other for the portrait mode. I then added code in shouldAutoRotate to change the navbar image based on the orientation
You can change the image for both portrait and landscape orientations by supplying different images for different orientations by checking the frame size of that UINavigation bar instance:
- (void) drawLayer:(CALayer *)layer inContext:(CGContextRef)context
{
if ([self isMemberOfClass:[UINavigationBar class]] == NO) {
return;
}
UIImage *image = (self.frame.size.width > 320) ?
[UINavigationBar bgImageLandscape] : [UINavigationBar bgImagePortrait];
CGContextClip(context);
CGContextTranslateCTM(context, 0, image.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGContextDrawImage(context, CGRectMake(0, 0, self.frame.size.width, self.frame.size.height), image.CGImage);
}
And this complete demo Xcode project on customizing the appearance of UINavigationBar might be helpful. It also includes #2x versions of background images to support retina displays on iPhone 4 devices.
Try the simpler method of [UIImage imageNamed:#"navbar_landscape.png"] since a UI element is exactly what imageNamed: is intended for, as ljonesATL shows.