(Objective-C) Difference between accessing the navigation stack through the window.rootView and keyWindow

Does anyone know what the difference is between

TabBarController* tabBar = (TabBarController *)_window.rootViewController;
UINavigationController* navigationController = tabBar.selectedViewController;
ViewController* initialViewController = (ViewController *)navigationController.topViewController;

and this

UINavigationController* navigationController = [UIApplication sharedApplication].keyWindow.rootViewController.navigationController;

ViewController* initialViewController = (ViewController *)navigationController.topViewController;

My assumptions:

• In example A I am using a specific nav stack (can't rely on that stack because I am trying to handle presenting a VC upon returning to the app after universal linking and it won't always be available).

• In example B I am attempting to access the navigation stack I am currently on.

• Last line of both examples, I am trying to push the new VC on top of the stack.

My results :

• example A succeeds at pushing my new VC onto the stack while example B doesn't.

Putting aside design principles (example B seems hacky right now, but it is all I have thus far), I don't really understand the difference in these 2 examples, shouldn't they perform the same? Could anyone shine a light on why these are different ?

1. The navigationController property on UIViewController is nullable, so it may return nil.

2. The topViewController property on UINavigationController is also nullable, and returns a UIViewController?, not a specific class. Simply casting any UIViewController to ViewController is not safe, so you should be checking against isKindOfClass.

3. Same thing applies for rootViewController, it may not always be a TabBarController, so casting it unconditionally is unsafe.

• also, found out that if rootViewController is of type UITableViewController, the navigationController is nil.