winder - Developer IT

Gradle Support in NetBeans IDE 7.2

- by Geertjan

Russel Winder and Steve Chin spent half an hour, and then gave up, setting up NetBeans IDE to use Gradle, because they couldn't find the NetBeans Gradle plugin, during Steve's NightHacking tour. That need happen no more because Attila Kelemen's NetBeans Gradle plugin is now available in the Plugin Manager in NetBeans IDE 7.2: Aside from opening Gradle-based applications, you can now also create new ones: Details and documentation: https://github.com/kelemen/netbeans-gradle-project

Read the article

How to load a UIView from a NIB?

- by Winder

I have been using UIViewControllers and initWithNibName with much success, basically using them as a convenient way to design the view with Interface Builder. Unfortunately I have built a hierarchy of views before noticing this line in the UIViewController documentation: Note: You should not use view controllers to manage views that fill only a part of their window My question is this: Having a very simple NIB that only has a UIView in addition to the default First Responder and Owning Object, what is the simplest way to load the UIView into my code? I have not been able to get loadNibNamed:owner:options: to work at this point, but suspect the answer will involve it somehow.

Read the article

Objective-C Pointer to class that implements a protocol

- by Winder

I have three classes which implement the same protocol, and have the same parent class which doesn't implement the protocol. Normally I would have the protocol as pure virtual functions in the parent class but I couldn't find an Objective-C way to do that. How can I utilize polymorphism on these subclasses and call the functions implemented in the protocol without warnings? Some pseudocode if that didn't make sense: @interface superclass: NSObject {} @interface child1: superclass<MyProtocol> {} @interface child2: superclass<MyProtocol> {} The consumer of these classes: @class child1 @class child2 @class superclass @interface SomeViewController: UIViewController { child1 *oneView; child2 *otherView; superclass *currentView; } -(void) someMethod { [currentView protocolFunction]; } The only nice way I've found to do pure virtual functions in Objective-C is a hack by putting [self doesNotRecognizeSelector:_cmd]; in the parent class, but it isn't ideal.

Read the article

Adding C++ Object to Objective-C Class

- by Winder

I'm trying to mix C++ and Objective-C, I've made it most of the way but would like to have a single interface class between the Objective-C and C++ code. Therefore I would like to have a persistent C++ object in the ViewController interface. This fails by forbidding the declaration of 'myCppFile' with no type: #import <UIKit/UIKit.h> #import "GLView.h" #import "myCppFile.h" @interface GLViewController : UIViewController <GLViewDelegate> { myCppFile cppobject; } @end However this works just fine in the .mm implementation file (It doesn't work because I want cppobject to persist between calls) #import "myCppFile.h" @implementation GLViewController - (void)drawView:(UIView *)theView { myCppFile cppobject; cppobject.draw(); }

Read the article

Objective-C @class / import best practice

- by Winder

I've noticed that a lot of Objective-C examples will forward declare classes with @class, then actually import the class in the .m file with an import. I understand that this is considered a best practice, as explained in answers to question: http://stackoverflow.com/questions/322597/objective-c-class-vs-import Coming from C++ this feels backwards. I would normally include all needed .h files in the new classes header file. This seems useful since it would make the compiler generate a warning when two classes include each other, at which point I can decide whether this is a bad thing or not then use the same Objective-C style and forward declare the class in the header and include it in the .cpp file. What is the benefit of forward declaring @class and importing in the implementation file? Should it be a best practice in C++ to forward declare classes rather than including the header file? Or is it wrong to think of Objective-C and C++ in these similar terms to begin with?

Read the article

Objective-C Memory Question

- by Winder

Is it a leak if I have a view controller and allocate the view like this: self.view = [[UIView alloc] initWithFrame:frame]; Do I need to do something like this: UIView *v = [[UIView alloc] initWithFrame:frame]; self.view = v; [v release];

Read the article

Load a UIView from a NIB

- by Winder

I have been using UIViewControllers and initWithNibName with much success, basically using them as a convenient way to design the view with Interface Builder. Unfortunately I have built a hierarchy of views before noticing this line in the UIViewController documentation: Note: You should not use view controllers to manage views that fill only a part of their window My question is this: Having a very simple NIB that only has a UIView in addition to the default First Responder and Owning Object, what is the simplest way to load the UIView into my code? I have not been able to get loadNibNamed:owner:options: to work at this point, but suspect the answer will involve it somehow.

Read the article

UIImageView Fade-In Dissapears

- by Winder

I have this code which should create a splash image with either no animation or a fade in, then call code to dismiss the image out after a delay. The SplashViewAnimationNone works fine and creates the full screen image, but the Fade code fades the image in but then immediately disappears. - (void)startSplash { [[[[UIApplication sharedApplication] windows] objectAtIndex:0] addSubview:self]; splashImage = [[UIImageView alloc] initWithImage:self.image]; if (self.animationIn == SplashViewAnimationNone) { [self addSubview:splashImage]; } else if (self.animationIn == SplashViewAnimationFade) { [self addSubview:splashImage]; CABasicAnimation *animSplash = [CABasicAnimation animationWithKeyPath:@"opacity"]; animSplash.duration = self.animationDelay; animSplash.removedOnCompletion = NO; animSplash.fillMode = kCAFillModeForwards; animSplash.fromValue = [NSNumber numberWithFloat:0.0]; animSplash.toValue = [NSNumber numberWithFloat:1.0]; animSplash.delegate = self; [self.layer addAnimation:animSplash forKey:@"animateOpacity"]; } // Dismiss after delay. [self performSelector:@selector(dismissSplash) withObject:self afterDelay:self.delay]; }

Read the article

C++ Pointer Issue

- by Winder

_stuckVertices is an array of pointers and I would like to update one index of that array without using _stuckVertices[ (row * _cols) + column ] 3 times. The reason it is an array of pointers is because the vast majority of the time the pointer will be NULL. The following code works but I need to dereference a each time I use it: void Cloth::stickPoint(int column, int row) { Anchor **a = &_stuckVertices[ (row * _cols) + column ]; if (!*a) *a = new Anchor(this, column, row); (*a)->stick(); } I originally had it written like this, but the _stuckVertices pointer doesn't get updated: void Cloth::stickPoint(int column, int row) { Anchor *a = _stuckVertices[ (row * _cols) + column ]; if (!a) a = new Anchor(this, column, row); a->stick(); } Is there a way to write Anchor *a = _stuckVertices[ index ] so that a is like an alias into the array that I can update, or is something like the first piece of code how I should do this? Thanks

Developer IT