I’ve been trying out some of the view code from Erica Sadun’s book “The iPhone Developer’s Cookbook” and found some code I don’t understand. Here’s the code for a loadView method:
- (void)loadView
{
// Create the main view
UIView *contentView = [[UIView alloc] initWithFrame:
[[UIScreen mainScreen] applicationFrame]];
contentView.backgroundColor = [UIColor whiteColor];
self.view = contentView;
[contentView release];
// Get the view bounds as our starting point
CGRect apprect = [contentView bounds];
// Add each inset subview
UIView *subview = [[UIView alloc]
initWithFrame:CGRectInset(apprect, 32.0f, 32.0f)];
subview.backgroundColor = [UIColor lightGrayColor];
[contentView addSubview:subview];
[subview release];
}
My question is why does she release contentView, but then use it again in [contentView addSubview:subview]? Has self.view = contentView retained contentView?
If you look in the documentation for UIViewController, you’ll see that the view property is declared as:
@property(nonatomic, retain) UIView *view;This means that when you use the setView: method (or use .view on the left hand side of the =), then whatever value you pass in will be retained. So, if you go through the code and look at retain counts, you’ll get this:
}
I’d say that the really interesting thing is that after releasing contentView, we can still send messages to it, because the object living at the end of contentView’s pointer still exists (since it was retained by calling setView:).