This sample code exhibits two issues: struct ContentView: View { @State private var myColor = Color.red var body: some View { VStack() { List() { Text("Object") Text("Object") Text("Object") .listRowSeparatorTint(myColor) Text("Object") } Button(action:{myColor = Color.green}) {Text("Change color")} } .foregroundColor(myColor) } } the row separator isn't redraws when the @State property change listRowSeparatorTint apply to two lines The first point is really disappointing. Is there anyone which know if this is a bug or there is a more correct way to use listRowSeparatorTint with changing parameter?

Hi, I was wondering if it's possible (and advisable) to use the new glass effects available in iOS 26 in Swift Charts? For example, in a chart like the one in the image I've attached to this post, I was looking to try adding a .glassEffect modifier to the BarMarks to see how that would look and feel. However, it seems it's not available directly on the BarMark (ChartContent) type, and I'm having trouble adding it in other ways too, such as using in on the types I supply to modifiers like foregroundStyle or clipShape. Am I missing anything? Maybe it's just not advisable or necessary to use glass effects within Charts?

I have a memory leak for SVG image that located in Assets.xcassets file when using SwiftUI Image, but when I use UIImage then convert it to SwiftUI Image the issue is not found. import SwiftUI struct ContentView: View { var body: some View { NavigationStack { VStack { NavigationLink("Show", destination: SecondView()) } .padding() } } } struct SecondView: View { @Environment(\.dismiss) var dismiss var body: some View { NavigationStack { VStack { IM.svgImage .resizable() .scaledToFit() .frame(width: 200, height: 200) Button("Dismiss") { dismiss() } } } } } enum IM { static let testImage: Image = "test_image".image static let svgImage: Image = "svgImage".image } extension String { var image: Image { Image(self) // Memory leak } var imageFromUIImage: Image { guard let uiImage = UIImage(named: self) else { return Image(self) } return Image(uiImage: uiImage) // No Memory leak } } Environment that produces the issue: Xcode: 16.2 Simulator: iPhone 15 Pro (iOS 17.5)

Verbatim of a feedback report (FB18431713) I submitted, duplicated here since we can't see each other's feedbacks, and I wanted a centralized place to track the resolution of this as I'm surely not the only one facing this. When building the app using Xcode 26 beta 2 and running it in an iOS 26 simulator, I'm experiencing a retain cycle in the UINavigationController. From the data I saw in Xcode's memory graph debugger, it seems that _UIViewControllerOneToOneTransitionContext is retaining it. I base this on the fact that the line connecting a view controller and _UIViewControllerOneToOneTransitionContext has a "strong" reference, as indicated in Xcode. (However, I'm reporting this as a retain cycle in UINavigationController, as that's what seems to hold onto this transition-context.)

More specifically, if I have 2 pinned tabs that are displayed in different order for compact and regular and switch between compact and regular, when it gets to regular mode, the images for both tabs will not be rendered. The tabs can still be pressed, just the icon is missing.

The issue is, I cannot auto acquire bluetooth keyboard focus in PHPickerViewController after enabling 'Full Keyboard Access' in my IPhone 14 with iOS version 18.3.1. The keyboard focus in PHPickerViewController will show, however, after I tapped on the blank space of the PHPickerViewController. How to make the focus on at the first place then? I'm using UINavigationController and calling setNavigationBarHidden(true, animated: false). Then I use this controller to present PHPickerViewController using some configuration setup below. self.configuration = PHPickerConfiguration() configuration.filter = .any(of: filters) configuration.selectionLimit = selectionLimit if #available(iOS 15.0, *), allowOrdering { configuration.selection = .ordered } configuration.preferredAssetRepresentationMode = .current Finally I set the delegate to PHPickerViewController and call UINavigationController.present(PHPickerViewController, animated: true) to render it. Also I notice animation showing in first video then disappear.

Hello, My goal is to have a NavigationStack whose root view is determined based on its height and width. To do so, I'm using ViewThatFits, which should choose the right view to display. It is working fine, but unexpectedly both views trigger onAppear, whereas only the appropriate one should. This causes the logic in both closures to be executed, which is not intended. The code below demonstrates the problem: struct NavigationStackContentView: View { var body: some View { NavigationStack { ViewThatFits(in: .vertical) { Color.yellow .onAppear { print("|-> on appear: yellow") } .onDisappear { print("|-> on disappear: yellow") } Color.red .frame(width: 1500, height: 1500) .onAppear { print("|-> on appear: red") } .onDisappear { print("|-> on disappear: red") } } } } } this produces: |-> on appear: red |-> on disappear: red |-> on appear: yellow When ViewThatFits is not nested within NavigationStack, the problem does not occur — only the yellow view (in this sample) triggers onAppear, which is the expected behavior. I also checked the macOS version, and the problem does not occur at all, whether within NavigationStack or not. This example is simple and demonstrates that the larger view is the second one. When I switch their places, the problem does not occur because it recognizes that the first view would not fit at this point. However, in my case I will have these views without knowing which one will not fit, so switching their order is not a viable solution if this works without NavigationStack. Am I doing something wrong, or is this a bug? // iOS: 18.3.1 Xcode: 16.2

I just put the TextField on UI and call the NSTextField setString, but it is memory usage is increasing. StoryBoard Objective C put TextField and button to UI set TextField variable to "ABC" in ViewController.h @property (weak) IBOutlet NSTextView* ABC; on button event function //dispatch_sync(dispatch_get_main_queue(), ^{ [_ABC setString:str]; //}); How to block the memory usage increase? Also I was check on Instruments app, and there are many malloc 48bytes, its count is almost same with setString count. Thank you!

I am trying to implement the NSTextViewDelegate function textViewDidChangeSelection(_ notification: Notification). My text view's delegate is the Coordinator of my NSViewRepresentable. I've found that this delegate function never fires, but any other delegate function that I implement, as long as it doesn't take a Notification as an argument, does fire (e.g., textView(:willChangeSelectionFromCharacterRange:toCharacterRange:), fires and is called on the delegate exactly when it should be). For context, I've verified all of the below: textView.isSelectable = true textView.isEditable = true textView.delegate === my coordinator I can call textViewDidChangeSelection(:) directly on the delegate without issue. I can select and edit text without issues. I.e., the selections are being set correctly. But the delegate method is never called when they are. I am able to add the intended delegate as an observer for the selector textViewDidChangeSelection via NotificationCenter. If I do this, the function executes when it should, but fires for every text view in my view hierarchy, which can number in the hundreds. I'm using an NSLayoutManager, so I figure this should only fire once. I've added a check within my code: func textViewDidChangeSelection(_ notification: Notification) { guard let textView = notification.object as? NSTextView, textView === layoutManager.firstTextView else { return } // Any code I want to execute... } But the above guard check lets through every notification, so, no matter what, my closure executes hundreds of times if I have hundreds of text views, all of them being sent by textView === layoutManager.firstTextView, but once for each and every text view managed by that layoutManager. Does anyone know why this method isn't ever called on the delegate, while seemingly all other delegate methods are? I could go the NotificationCenter route, but I'd love to know why this won't execute as a delegate method when documentation says that it should, and I don't want to have to implement a counter to make sure my code only executes once per selection update. And for more reasons than that, implementing via delegate method is preferable to using notifications for my use case. Thanks for any help!

Overview Starting with macOS 26 beta 1, a new NSGlassContainerView is added inside NSToolbarView. This view intercepts mouse events, so any SwiftUI Button (or other interactive view) overlaid on the title‑bar / toolbar area no longer receives clicks. (The same code works fine on macOS 15 and earlier.) Filed as FB18201935 via Feedback Assistant. Reproduction (minimal project) macOS 15 or earlier → button is clickable macOS 26 beta → button cannot be clicked (no highlight, no action call) @main struct Test_macOS26App: App { init() { // Uncomment to work around the issue (see next section) // enableToolbarClickThrough() } var body: some Scene { WindowGroup { ContentView() } .windowStyle(.hiddenTitleBar) // ⭐️ hide the title bar } } struct ContentView: View { var body: some View { NavigationSplitView { List { Text("sidebar") } } detail: { HSplitView { listWithOverlay listWithOverlay } } } private var listWithOverlay: some View { List(0..<30) { Text("item: \($0)") } .overlay(alignment: .topTrailing) { // ⭐️ overlay in the toolbar area Button("test") { print("test") } .glassEffect() .ignoresSafeArea() } } } Investigation In Xcode View Hierarchy Debugger, a layer chain NSToolbarView > NSGlassContainerView sits in front of the button. -[NSView hitTest:] on NSGlassContainerView returns itself, so the event never reaches the SwiftUI layer. Swizzling hitTest: to return nil when the result is the view itself makes the click go through: func enableToolbarClickThrough() { guard let cls = NSClassFromString("NSGlassContainerView"), let m = class_getInstanceMethod(cls, #selector(NSView.hitTest(_:))) else { return } typealias Fn = @convention(c)(AnyObject, Selector, NSPoint) -> Unmanaged<NSView>? let origIMP = unsafeBitCast(method_getImplementation(m), to: Fn.self) let block: @convention(block)(AnyObject, NSPoint) -> NSView? = { obj, pt in guard let v = origIMP(obj, #selector(NSView.hitTest(_:)), pt)?.takeUnretainedValue() else { return nil } return v === (obj as AnyObject) ? nil : v // ★ make the container transparent } method_setImplementation(m, imp_implementationWithBlock(block)) } Questions / Call for Feedback Is this an intentional behavioral change? If so, what is the recommended public API or pattern for allowing clicks to reach views overlaid behind the toolbar? Any additional data points or confirmations are welcome—please reply if you can reproduce the issue or know of an official workaround. Thanks in advance!

I was trying to adapt memoji in my app which write by objective-c.I have a textView for user input and I need to keep allowsEditingTextAttributes == NO for some reason.Is there any other way to show memoji sticker in system emoji keyboard?Thanks!

Hello Developer Forums Team, I’ve seen that some banking apps prevent screenshots on certain sensitive screens. I’m working on a similar feature in my SDK and want to confirm if my implementation complies with App Store guidelines. Since there’s no public API to block screenshots, I’m using a workaround based on the secure rendering behavior of UITextField (isSecureTextEntry = true). I embed my custom content (e.g., a UITableView) inside the internal secure container of a UITextField, which results in blank content being captured during screenshots—similar to what some banking apps do. Approach Summary I create a UITextField I detect its internal secure container by matching UIKit internal class names as strings I embed my real UI content into that container I do not use or call any private APIs, just match view class names via strings. ScreenshotPreventingView.swift final class ScreenshotPreventingView: UIView { private let textField = UITextField() private let recognizer = HiddenContainerRecognizer() private var contentView: UIView? public var preventScreenCapture = true { didSet { textField.isSecureTextEntry = preventScreenCapture } } public init(contentView: UIView? = nil) { super.init(frame: .zero) self.contentView = contentView setupUI() } private func setupUI() { guard let container = try? recognizer.getHiddenContainer(from: textField) else { return } addSubview(container) NSLayoutConstraint.activate([ container.topAnchor.constraint(equalTo: topAnchor), container.bottomAnchor.constraint(equalTo: bottomAnchor), container.leadingAnchor.constraint(equalTo: leadingAnchor), container.trailingAnchor.constraint(equalTo: trailingAnchor) ]) if let contentView = contentView { setup(contentView: contentView, in: container) } DispatchQueue.main.async { self.preventScreenCapture = true } } private func setup(contentView: UIView) { self.contentView?.removeFromSuperview() self.contentView = contentView guard let container = hiddenContentContainer else { return } container.addSubview(contentView) container.isUserInteractionEnabled = isUserInteractionEnabled contentView.translatesAutoresizingMaskIntoConstraints = false let bottomConstraint = contentView.bottomAnchor.constraint(equalTo: container.bottomAnchor) bottomConstraint.priority = .required - 1 NSLayoutConstraint.activate([ contentView.leadingAnchor.constraint(equalTo: container.leadingAnchor), contentView.trailingAnchor.constraint(equalTo: container.trailingAnchor), contentView.topAnchor.constraint(equalTo: container.topAnchor), bottomConstraint ]) } } HiddenContainerRecognizer.swift struct HiddenContainerRecognizer { private enum Error: Swift.Error { case unsupportedOSVersion(version: Float) case desiredContainerNotFound(_ containerName: String) } func getHiddenContainer(from view: UIView) throws -> UIView { let containerName = try getHiddenContainerTypeInStringRepresentation() let containers = view.subviews.filter { subview in type(of: subview).description() == containerName } guard let container = containers.first else { throw Error.desiredContainerNotFound(containerName) } return container } private func getHiddenContainerTypeInStringRepresentation() throws -> String { if #available(iOS 15, *) { return "_UITextLayoutCanvasView" } if #available(iOS 14, *) { return "_UITextFieldCanvasView" } if #available(iOS 13, *) { return "_UITextFieldCanvasView" } if #available(iOS 12, *) { return "_UITextFieldContentView" } let currentIOSVersion = (UIDevice.current.systemVersion as NSString).floatValue throw Error.unsupportedOSVersion(version: currentIOSVersion) } } How I use it in my Screen let container = ScreenshotPreventingView() override func viewDidLoad() { super.viewDidLoad() container.preventScreenCapture = true container.setup(contentView: viewContainer) //viewContainer is UIView in storyboard, in which all other UI elements are placed in e.g. UITableView self.view.addSubview(container) container.translatesAutoresizingMaskIntoConstraints = false NSLayoutConstraint.activate([ container.topAnchor.constraint(equalTo: self.view.topAnchor), container.bottomAnchor.constraint(equalTo: self.view.bottomAnchor), container.leadingAnchor.constraint(equalTo: self.view.leadingAnchor), container.trailingAnchor.constraint(equalTo: self.view.trailingAnchor) ]) } What I’d Like to Confirm Is this approach acceptable for App Store submission? Is there a more Apple-recommended approach to prevent screen capture of arbitrary UI? Thank you for your help in ensuring compliance.

In our project, we defined a CustomTabBar that inherits UITabBar, and we add some subviews and these subviews' frame is beyond UITabBar, see below picture(a beyond area button, this button is a subview of UITabBar, but frame is out of UITabBar's area): and in order to let this button response to click action, we override the UITabBar's hitTest method, this works well below OS 26 with Xcode version below 26: override public func hitTest(_ point: CGPoint, with event: UIEvent?) -> UIView? { let pointInCollectionView = self.beyondAreaButton.convert(point, from: self) if self.beyondAreaButton.bounds.contains(pointInCollectionView) { return self.beyondAreaButton.hitTest(pointInCollectionView, with: event) } return super.hitTest(point, with: event) } but when using Xcode26 build app on OS 26, I noticed the UITabBar is wrapped by a UIKit._UITabBarContainerView and UIKit._UITabBarContainerWrapperView, and it can not trigger the hitTest method. since the hitTest is not triggered, so the button action is no chance to trigger too. Any suggestions to solve this problem, thank you~ And I have file a feedback assistant: FB19252973

when in portrait, the UINavigationBar shows normal: when in landscape: The top is white and blank.

Double-tap three fingers and drag to change zoom” should suppress “Three Finger to Drag”. Currently these gestures are triggered simultaneously, for no real reasons. I saw different behaviors for different environments, but none is desired. Current and desired behavior: This seems an issue so I filed a feedback.

Hello everybody! TLDR: Issues with visibleItemsInvalidationHandler. Minimal code to reproduce available. I've been working with Compositional Layout for a while now and recently I've found myself needing to implement custom animation based on scroll position of UI elements. Once I found visibleItemsInvalidationHandler it felt like the exact solution that I needed. Once I implement I've found out it doesn't quite behave as you'd expect. To put it simply, it seems like the animations only work if your whole layout does not use .estimated nor .uniformAcrossSiblings. As soon as you use them then the animations will stop working, I've debugged it deeper and it seems like the invalidation context generated by it does not include the indexPath of the cells, which is always included in the version in which it works. Feel free to swap the line 51 with its comment to flip between the working and failing version of it. Playground Example My final question therefore is... Is this the expected behavior? The documentation doesn't give any clues about such behavior and although I've tried relentlessly to find a workaround for this specific hiccup I was not successful with it.

Hi everybody, I use a very old photo software called Snapseed v 1.2.1 on 2 MacBook, one runs with El Capitan and the second runs with Mojave. These app has no language selection during install, and no preferences language choice in menu when running. When I execute app on El Capitan, all menus and topics are translate in French, but on Mojave no translation is done. I looked Contents of app and found differents languages files (structured as xx.po where xx=country) located in Resources folder. I deduced translation was executed after getting language param. I checked values on both systems and there are same (LANG=fr_FR.UTF-8). So I tried to change Info.plist file to force code langage to 'fr' in CFBundleLocalizations key. Result is same. Does somebody has a idea of reason of issue and how to solve it ? Snapseed release was 2012, El Capitan 2015 and Mojave 2018, it seems framework used to code app runs differently and can't get language value.

Hello, We received a rejection on one of our IOS applications because we were doing Microsoft MSAL login through the user's browser. The representative recommended that we use Webview to do in-app logins. However when we tried to handle the custom app uri redirection (looking like myapp://auth/), Webview does not seem to send the user back to the application. Does anyone have a fix for this? Thanks!

I have an iPad app with a classic sidebar. It's been working fine for years. Now with iPadOS 26 the sidebar sometime gets this fake transparency that makes it really hard to quickly grok. A part of Liquid Glass seems to be to sometimes (but not always) take whatever is in the secondary area (the main big content), blur it, mirror it and then use as the background for the sidebar. This is silly and does not work at all for an app like mine. It maybe looks decent if your background is a photo or similar, but not for an app that manages data. Not all views cause the sidebar to get this ugly unreadable background. In most of the cases the sidebar keeps its normal opaque background that it has always had. See this example for how it looks when it's really bad: This is how it should look. Notice that the content of the "main view" is pretty similar to the case where it gets the ugly background. The difference is the segmented thing at the top, ie. a different "root view". Is there some good way for me to force the sidebar to always have an opaque background? I guess I could make a ZStack and put a solid color as the background behind the sidebar, but those kinds of hacks are better to avoid. This can not be how some UI designer envisioned that apps should look? Maybe I'm missing some new modifier or setting somewhere that would led me opt out from this aspect of Liquid Glass? Apart from this it looks pretty nice. There are some bugs where the contents of the main area gets clipped when the sidebar is shown, hidden and then shown again, but that's for another time (and it's surely known already (if the bug tracking system allowed me to search I could verify)). So, any way to make my app look nice again?

All system colors are displayed incorrectly on the popover view. Those are the same views present as a popover in light and dark mode. And those are the same views present as modal. And there is also a problem that when the popover is presented, switching to dark/light mode will not change the appearance. That affected all system apps. The following screenshot is already in dark mode. All those problem are occured on iOS 26 beta 3.