Is there something special that I have to do when unwrapping objects and getting access to the structs inside of them? This currently works with forced unwrapping, but not with optional binding/chaining. Check out lines 74 through 78 in ViewController.swift

AppDelegate.swift

//
//  AppDelegate.swift
//  Test
//
//  Created by Jon Friskics on 9/29/14.
//  Copyright (c) 2014 Code School. All rights reserved.
//

import UIKit

@UIApplicationMain
class AppDelegate: UIResponder, UIApplicationDelegate {

    // MARK: ------ Property declarations
    
    var window: UIWindow?
    
    // MARK: ------ App delegate methods
    
    func application(application: UIApplication!, didFinishLaunchingWithOptions launchOptions: NSDictionary!) -> Bool {
        
        window = UIWindow(frame: UIScreen.mainScreen().bounds)
        
        let comicVC = ViewController()
        
        let navController = UINavigationController(rootViewController: comicVC)
        
        window?.rootViewController = navController
        window?.makeKeyAndVisible()
        
        return true
    }
}

ViewController.swift

//
//  ViewController.swift
//  Test
//
//  Created by Jon Friskics on 9/29/14.
//  Copyright (c) 2014 Code School. All rights reserved.
//

import UIKit

class ViewController: UIViewController, UIScrollViewDelegate {

    lazy var scrollView:UIScrollView? = {
        let sv = UIScrollView()
        sv.delegate = self
        sv.maximumZoomScale = 5.0
        sv.backgroundColor = UIColor.lightGrayColor()
        return sv
    }()
    
    lazy var imageView:UIImageView? = {
        let iv = UIImageView()
        iv.backgroundColor = UIColor.redColor()
        iv.userInteractionEnabled = true
        return iv
    }()
    
    func makeImage(color: UIColor) -> UIImage? {
        let rect = CGRect(x: 0, y: 0, width: 250, height: 250)
        UIGraphicsBeginImageContext(rect.size);
        
        let context = UIGraphicsGetCurrentContext()
        CGContextSetFillColorWithColor(context, UIColor.blueColor().CGColor)
        CGContextFillRect(context, rect);

        let image = UIGraphicsGetImageFromCurrentImageContext()

        UIGraphicsEndImageContext();
        
        return image;
    }
    
    override init() {
        super.init(nibName: nil, bundle: nil)
        
        self.automaticallyAdjustsScrollViewInsets = false
    }

    required init(coder aDecoder: NSCoder) {
        fatalError("init(coder:) has not been implemented")
    }
    
    override func viewDidLoad() {
        super.viewDidLoad()
        
        view.backgroundColor = UIColor.greenColor()
        
        let image = makeImage(UIColor.blueColor())
        imageView!.image = image
        scrollView!.addSubview(imageView!)
        if let sv = scrollView {
            view.addSubview(sv)
        }
    }

    override func viewWillLayoutSubviews() {
        super.viewWillLayoutSubviews()

        imageView?.sizeToFit()
        scrollView?.frame = CGRect(x: 0, y: topLayoutGuide.length, width: 200, height: 200)
        
        println("\(imageView!.frame.size.width) \(imageView!.frame.size.height)")
        
        // works
        scrollView?.contentSize = CGSize(width: imageView!.frame.size.width, height: imageView!.frame.size.height)
        
        // doesn't work
        // scrollView?.contentSize = CGSize(width: imageView?.frame.size.width, height: imageView?.frame.size.height)
    }
}

I guess it comes down to this - why is the frame CGRect an optional in addition to the UIImageView?

This code:

    override func viewWillLayoutSubviews() {
        super.viewWillLayoutSubviews()

        imageView?.sizeToFit()
        scrollView?.frame = CGRect(x: 0, y: topLayoutGuide.length, width: 200, height: 200)

        println("\(imageView!.frame.size.width) \(imageView!.frame.size.height)")

        let frame = imageView?.frame
        println("frame: \(frame)")
        let size = frame!.size
        println("size: \(size)")
        let width = size.width
        let height = size.height
        println("width: \(width), height: \(height)")

        scrollView?.contentSize = CGSize(width: imageView!.frame.size.width, height: imageView!.frame.size.height)
    }

Gives this output:

250.0 250.0
frame: Optional((0.0,0.0,250.0,250.0))
size: (250.0,250.0)
width: 250.0, height: 250.0

Which shows that the frame returned from the imageView optional is itself an optional, which just doesn't seem quite right. Especially since the UIView docs don't have frame:CGRect marked as an optional.

After researching a little bit more, it sounds like maybe the way to go is to declare the lazy objects as implicity unwrapped optionals, like this:

    lazy var imageView:UIImageView! = {
        let iv = UIImageView()
        iv.backgroundColor = UIColor.redColor()
        iv.userInteractionEnabled = true
        return iv
    }()

This makes some sense, because UIImageView() should never return nil. By declaring it as implicitly unwrapped after it has been created, the frame-setting is simplified to this:

        scrollView.contentSize = CGSize(width: imageView.frame.size.width, height: imageView.frame.size.height)