I can guess what might be the problem here, because I've done it:
I've found that often when I add init code to loadView, I end up with an infinite stack trace
Don't read self.view in -loadView. Only set it, don't get it.
The self.view property accessor calls -loadView if the view isn't currently loaded. There's your infinite recursion.
The usual way to build the view programmatically in -loadView, as demonstrated in Apple's pre-Interface-Builder examples, is more like this:
UIView *view = [[UIView alloc] init...];
...
[view addSubview:whatever];
[view addSubview:whatever2];
...
self.view = view;
[view release];
And I don't blame you for not using IB. I've stuck with this method for all of Instapaper and find myself much more comfortable with it than dealing with IB's complexities, interface quirks, and unexpected behind-the-scenes behavior.
The bounds of an UIView is the rectangle, expressed as a location (x,y) and size (width,height) relative to its own coordinate system (0,0).
The frame of an UIView is the rectangle, expressed as a location (x,y) and size (width,height) relative to the superview it is contained within.
So, imagine a view that has a size of 100x100 (width x height) positioned at 25,25 (x,y) of its superview. The following code prints out this view's bounds and frame:
// This method is in the view controller of the superview
- (void)viewDidLoad {
[super viewDidLoad];
NSLog(@"bounds.origin.x: %f", label.bounds.origin.x);
NSLog(@"bounds.origin.y: %f", label.bounds.origin.y);
NSLog(@"bounds.size.width: %f", label.bounds.size.width);
NSLog(@"bounds.size.height: %f", label.bounds.size.height);
NSLog(@"frame.origin.x: %f", label.frame.origin.x);
NSLog(@"frame.origin.y: %f", label.frame.origin.y);
NSLog(@"frame.size.width: %f", label.frame.size.width);
NSLog(@"frame.size.height: %f", label.frame.size.height);
}
And the output of this code is:
bounds.origin.x: 0
bounds.origin.y: 0
bounds.size.width: 100
bounds.size.height: 100
frame.origin.x: 25
frame.origin.y: 25
frame.size.width: 100
frame.size.height: 100
So, we can see that in both cases, the width and the height of the view is the same regardless of whether we are looking at the bounds or frame. What is different is the x,y positioning of the view. In the case of the bounds, the x and y coordinates are at 0,0 as these coordinates are relative to the view itself. However, the frame x and y coordinates are relative to the position of the view within the parent view (which earlier we said was at 25,25).
There is also a great presentation that covers UIViews. See slides 1-20 which not only explain the difference between frames and bounds but also show visual examples.
Best Answer
The only difference is in where the view is added: whether it is the frontmost view (
addSubview:
), or it is before the 5th subview, (insertSubview:atIndex:
) or if it is immediately behind another subview (insertSubview:aboveSubview:
).