My own iOS app is using AVFoundation to do barcode scanning. Under Mac Catalyst I'm getting an error. Turns out the Apple provided AVCam demo app has the exact same issue.Both the Swift and Objective-C versions of the AVCam demo app fail but the Objective-C version gives a better error so I'll refer to that in the rest of the question.In the AVCamCameraViewController, the configureSession method fails on the line:AVCaptureDeviceInput* videoDeviceInput = [AVCaptureDeviceInput deviceInputWithDevice:videoDevice error:&error];The error is:Could not create video device input: Error Domain=AVFoundationErrorDomain Code=-11814 "Cannot Record" UserInfo={NSLocalizedDescription=Cannot Record, NSLocalizedRecoverySuggestion=Try recording again.}It turns out that none of the three attempts to call [AVCaptureDevice defaultDeviceWithDeviceType:... are succeeding.This is the same error I get in my own app though I only attempt to call [AVCaptureDevice defaultDeviceWithMediaType:AVMediaTypeVideo]; which is also returning nil.Permission was granted for the app to access the camera.There is nothing related to this issue in the release notes for macOS 10.15. I am running the GM version of 10.15 with Xcode 11.1 on a 2018 15" Macbook Pro.How do we successfully get a video capture device under Mac Catalyst?

Just watched the Targeting Content with Multiple Windows video. My app uses UIApplicationShortcutItem and UIApplication.shortcutItems.So I thought it would be useful to set the targetContentIdentifier property of each shortcut item so when the user selects one of the shortcuts, they can be brought to the mose appropriate scene in the app.But UIApplicationShortcurItem targetContentIdentifer is a read-only property. Is this is mistake?The documentation for UISceneActiviationConditions states (emphesis mine):Many different objects contain a targetContentIdentifier property, including NSUserActivity, UNNotificationContent, and UIApplicationShortcutItem. When creating those objects, fill that property with a value that uniquely describes the event and matches your scenes' predicates.Kind of hard to do when the property is read-only.Am I missing something or is this an API bug?