I have noticed that in iOS 14 the UIPickerView has by default a light grey background on the selected Row like shown here. https://developer.apple.com/design/human-interface-guidelines/ios/controls/pickers/ I noticed also that pickerView.showsSelectionIndicator is deprecated on iOS 14. Is there a way to change the background color to white and add separators to achieve a pre iOS 14 UIPickerView style? Thank you

I'm developing a rhythm game for iOS which has four buttons spanning the width of the screen in portrait. I noticed that my testers were having some missed inputs on the buttons on the left and right due to the fact that iOS, by default, tries to ignore accidental touches on the edges of the screen. So I enabled "Defer System Gestures" on the left and right edges, but then quickly started to notice a new, very specific, issue. Description of the issue If you have finger #1 touching and holding anywhere in the middle of the screen, and finger #2 touches on the far right or left edge of the screen just below the horizontal position of finger #1, those touches are inconsistently not recognized. If finger #1 is not present, this issue does not occur. If finger #2 is above or well below finger #1, this issue also does not occur. A dead zone is created on the right and left edges of the screen just below the horizontal position of the first touch. Here is a rough representative example of where touches #1 and #2 need to be for this issue to manifest, in case the text above is not clear. |				| |				| |				| |				| |	 1		| |			 2| |				| It just so happens that this issue is causing major usability problems with my game, as it results in what the user sees as sporadic and inconsistent response when the game calls for two notes to be played at the same time. Steps to recreate the issue Here are the steps if you want to recreate the problem yourself using the "Create New Gesture" pane in "Assistive Touch" (Note that this problem is not specific to the Settings app, but rather is an issue across the system—however this panel defers system gestures and shows where touches are being read, so it is a great place to demonstrate): (1) Go to Settings > Accessibility > Touch > Assistive Touch > Create New Gesture...; (2) With one finger, touch the middle of the screen and hold it through step 3; (3) With a second finger, tap 4 times along the right (or left) edge of the screen in the following places: (a) well above the vertical position of the first touch, (b) just above the vertical position of the first touch, (c) just below the vertical position of the first touch, and (d) well below the vertical position of the first touch; (4) Notice how, more than half the time, touch (c) does not register. I have found that this problem is more replicatable when the first touch is on the lower half of the screen, but I have been able to replicate it when the finger is higher as well, just not as consistently. Here are the four positions described in the steps above: Position a: both touches register |				| |				| |				| |			 2| |	 1		| |				| |				| Position b: both touches usually register |				| |				| |				| |				| |	 1	 2| |				| |				| Position c: only touch 1 registers |				| |				| |				| |				| |	 1		| |			 2| |				| Position d: both touches register |				| |				| |				| |				| |	 1		| |				| |			 2| Is there anything I can do to resolve this behavior? My app requires gesture deferment to be on for the expected experience by the user, and this bug is causing other issues for my testers that kind of need to be resolved before I can confidently release the game.

Why is the UIKeyboard implementation still holding a reference to this UITextField, thus keeping it from being deallocated? The memory debugger shows: UIKeyboardImpl -> UIKBAutofillController -> NSMutableDictionary -> NSMutable...(Storage) -> UITextField Any idea what's going on there?

I have been trying to understand and utilize intrinsicSize on a custom UIView for some days now. So far with little success. The post is quite long, sorry for that :-) Problem is, that the topic is quite complex. While I know that there might be other solutions, I simply want to understand how intrinsicSize can be used correctly. So when someone knows a good source for a in depth explanation on how to use / implement intrinsicSize you can skip all my questions and just leave me link. My goal: Create a custom UIView which uses intrinsicSize to let AutoLayout automatically adopt to different content. Just like a UILabel which automatically resizes depending on its text content, font, font size, etc. As an example assume a simple view RectsView which does nothing but drawing a given number of rects of a given size with given spacing. If not all rects fit into a single row, the content is wrapped and drawing is continued in another row. Thus the height of the view depends on the different properties (number of rects, rects size, spacing, etc.) This is very much like a UILabel but instead of words or letters simple rects are drawn. However, while UILabel works perfectly I was not able to achive the same for my RectsView. Why intrinsicSize I do not have to use intrinsicSize to achieve my goal. I could also use subviews and add constraints to create such a rect pattern. Or I could use a UICollectionView, etc. While this might certainly work, I think it would add a lot of overhead. If the goal would be to recreate a UILabel class, one would not use AutoLayout or a CollectionView to arrange the letters to words, would one? Instead one would certainly try to draw the letters manually... Especially when using the RectsView in a TableView or a CollectionView a plain view with direct drawing is certainly better than a complex solution compiled of tons of subviews arranged using AutoLayout. Of course this is an extreme example. However, at the bottom line there are cases where using intrinsicSize is certainly the better option. Since UILabel and other build in views uses intrinsicSize perfectly, there has to be a way to get this working and I just want to know how :-) My understanding of intrinsic Size The problem is that I found no source which really explains it... Thus I have spend several hours trying to understand how to correctly use intrinsicSize without little progress. This is what I have learned [from the docs][1]: intrinsicSize is a feature used in AutoLayout. Views which offer an intrinsic height and/or width do not need to specify constraints for these values. There is no guarantee that the view will exactly get its intrinsicSize. It is more like a way to tell autoLayout which size would be best for the view while autoLayout will calculate the actual size. The calculation is done using the intrinsicSize and the Compression Resistance + Content Hugging properties. The calculation of the intrinsicSize should only depend on the content, not of the views frame. What I do not understand: How can the calculation be independent from the views frame? Of course the UIImageView can use the size of its image but the height of a UILabel can obviously only be calculated depending on its content AND its width. So how could my RectsView calculate its height without considering the frames width? When should the calculation of the intrinsicSize happen? In my example of the RectsView the size depends on rect size, spacing and number. In a UILabel the size also depends on multiple properties like text, font, font size, etc. If the calculation is done when setting each property it will be performed multiple times which is quite inefficient. So what is the right place to do it? I will continue the question a second post due to the character limit...

unifieldFlexKeyboard[53102:24145174] [External] -[UIInputViewController needsInputModeSwitchKey] was called before a connection was established to the host application. This will produce an inaccurate result. Please make sure to call this after your primary view controller has been initialized. hey

Dear random Apple UIKit engineer. This is a question for you. Today let's speak about keyboard notifications. In particular, UIResponder.keyboardWillShowNotification and UIResponder.keyboardWillHideNotification. While working with those, I noticed some undocumented behaviour. First, let me give you some context: extension UIViewController { func registerForKeyboardNotifications() { NotificationCenter.default.addObserver(self, selector: #selector(keyboardNotification), name: UIResponder.keyboardWillShowNotification, object: nil) NotificationCenter.default.addObserver(self, selector: #selector(keyboardNotification), name: UIResponder.keyboardWillHideNotification, object: nil) } /// Override this method to handle keyboard notifications. @objc func keyboardNotification(_ notification: Notification) { ... } } Eventually, I found that latter method with 3 dots has an implicit animation inside it's scope. Here is the [proof.](https://medium.com /uptech-team/why-does-uiresponder-keyboard-notification-handler-animate-10cc96bce372) Another thing I noticed, is that this property definition is perfectly valid let curve = UIView.AnimationCurve(rawValue: 7)!. The 7 btw comes from UIResponder.keyboardAnimationCurveUserInfoKey as a default value during my tests. So, the enum with 4 possible values (0...3) can be initialized with a value out of enum's cases range. Also, how can I initialize UIView.AnimationOption from 7? I will pollute my OptionSet which I feed to options parameter on UIView.animate(...) My questions: Why implicit animation is not documented and can I trust it or it's a subject to change. Why UIView.AnimationCurve(rawValue: 7)! does not crash. How can I convert UIResponder.keyboardAnimationCurveUserInfoKey's value into UIView.AnimationOption properly if I don't want to use implicit value. I don't encroach on UIKit secrets. I just need to know how to work with the API. Thank you!

Hi, In a Mac Catalyst app, I need to allow the user insert a passcode using a UITextField. The field is used to insert a one time passcode and I want to keep the content hidden. For this reason I set the isSecureTextEntry property to true. passcodeTextField.isSecureTextEntry = true By doing this, a button to allow the user to pick a password from the keychain is displayed: This option in my case should not appear because the password is a one time password that change every time. For that reason I set the textContentType to oneTimeCode. passcodeTextField.textContentType = .oneTimeCode This actually removes the password button, but introduce something weird. If the user type something and then delete everything, a big empty box appear under the field: I have no idea what this box is and why it appears. Does anyone know why it appears and how I can remove it? Thank you

Facing issue of NSUserDefaults data got cleared on update new version from app store. so all data of user got cleared and user have to login again. I was having app in store with version 1.1.3 and then we have uploaded new version to store with version 1.1.4 but when user download app from store their data was cleared from UserDefaults.so this is big issue in update. need to help. is that apple updated any process on update. Question How Apple change version build in device when we update app from app store. is it uninstall old build and install new build ? Or it is install new version on old version ? Should UserDefaults.standard will delete on update ? Updating Provisioning profile from manual to automatically cause issue on UserDefaults.standard Like UserDefaults got cleared and uuid will change ? Do we have anythings that we can identify specific user as uniq user ? What is better solution to store login data into SQLite store or in UserDefaults as model ? Will appreciate for your help. this is very big issue and login again on update will make app user more effect

Hello 👋🏽 I am a new iOS developer and I am having a hard time understanding the behavior of the new UIPasteControl to avoid showing the system prompt. Does anyone has any example code I could look at to understand the behavior. Idk how to set the target of the button to the general pasteboard. also I am using objective-c . thanks cristian

We are currently trying to adopt the newly introduced find bar in our app. The app: The app is a text editor with a main text view. However it includes nested views (for text like footnotes) that are presented as modal sheets. So you tap on the footnote within the main text, a form sheet is presented with the contents of the footnote ready to be edited. We have an existing search implementation, but are eager to move to the system-provided UI. Connecting the find bar through a custom UIFindSession with our existing implementation is working without any issues. The Problem: Searching for text does not only work in the main text view, but also nested text (like footnotes). Let's say I have a text containing the word "iPhone" both in the main text and the footnote. In our existing implementation, stepping from the search match to the next one would open the modal and highlight the match in the nested text. The keyboard would stay open. With the new UIFindInteraction this is not working however. As soon as a modal form sheet is presented, the find interaction closes. By looking at the stack trace I can see a private class called UIInputWindowController that cleans up input accessory views after the modal gets presented. I believe it is causing the find panel to give up its first responder state. I noticed that opening popovers appears to be working fine. Is there a way to alter the presentation of the nested text so that the view is either not modal or able to preserve the current find session? Or is this unsupported behavior and we should try and look for a different way? The thing that really confuses me is that this appears to work without issue in Notes.app. There the find bar is implemented as well. There are multiple views that can be presented while the find bar is open. Move Note is one of them. The view appears as a modal sheet. It keeps the find bar open and active, though its tint color matches the deactivated one of the main Notes view. The find bar is still functional with the text field being active and the overlay updating in the background. This behavior appears to be a bug in the Notes app, but is exactly what we want for our use case. I attached some images: Two are from the Notes app, two from a test project demonstrating the problem. Opening a modal view closes the find bar there.

I'm using UIDocumentPickerViewController to open a url. Works fine in debug mode but version on the App Store is failing. Code to create the document picker is like: NSArray *theTypes = [UTType typesWithTag:@"docxtensionhere" tagClass:UTTagClassFilenameExtension conformingToType:nil]; UIDocumentPickerViewController *documentPicker = [[UIDocumentPickerViewController alloc]initForOpeningContentTypes:theTypes]; documentPicker.delegate = self;   [self presentViewController:documentPicker animated:YES completion:nil]; So in debug mode this is all gravy. -documentPicker:didPickDocumentsAtURLs: passes back a URL and I can read the file. In release mode I get a URL but my app is denied access to read the file. After inspecting some logging it appears the sandbox is not granting my app permission. error Domain=NSCocoaErrorDomain Code=257 "The file “Filename.fileextensionhere” couldn’t be opened because you don’t have permission to view it." UserInfo={NSFilePath=/private/var/mobile/Library/Mobile Documents/comappleCloudDocs/Filename.fileextensionhere, NSUnderlyingError=0x2834c9da0 {Error Domain=NSPOSIXErrorDomain Code=1 "Operation not permitted"}} -- If I'm doing something wrong with UIDocumentPickerViewController it is a real shame that permission is not being denied in Debug mode, as devs are more likely to catch in prior to release. Anyone know where I'm going wrong and if not have a workaround? Thanks in advance.

In Xcode 14 RC, I'm seeing this in the Console: [Assert] UINavigationBar decoded as unlocked for UINavigationController, or navigationBar delegate set up incorrectly. Inconsistent configuration may cause problems. navigationController=<MasterNavigationController: 0x135016200>, navigationBar=<UINavigationBar: 0x134f0aec0; frame = (0 20; 0 50); opaque = NO; autoresize = W; layer = <CALayer: 0x600000380be0>> delegate=0x135016200 The above message displays exactly four times immediately at app launch (top of the console) then does not repeat. MasterNavigationController is the internal class for the app's navigation controller. It is in a Storyboard, with very minimal ObjC code. I am not setting any specific size for the nav bar. I don't remember seeing this in earlier builds of Xcode, but I can't swear to it that this is new. No assertion actually fires.

In iOS 16, VoiceOver no longer speaks numbers such as “¥1,230" in Japanese, which worked correctly in iOS 15 and earlier. This is a very important bug in Accessibility support. Steps to reproduce Create iOS App Project on Xcode. (I used Xcode 14) Set the Development Localization to "Japanese" on the project settings. (to make VoiceOver speak Japanese) For example, write the following code. Build and run the App targeting the actual device, not the simulator. Actual devices running iOS 15 or earlier will correctly read out loud with VoiceOver, while iOS 16 will have some reading problems. import SwiftUI @main struct VoiceOverProblemApp: App {   var body: some Scene {     WindowGroup {       // on project settings, set the development localization to Japanese.       VStack {         // ✅ said "Hello, world!"         Text("Hello, world!")                   // ✅ said "残高 (ざんだか, zan-daka)"         Text("残高")                   // ❌ said nothing (until iOS 15 or earlier, said "千二百三十円 (せんにひゃくさんじゅうえん, sen-nihyaku-san-ju-yen)"         Text("¥1,230")                   // ❌ said wrong         //  correct (iOS 15 or earlier):  "残高は千二百三十円です (zan-daka wa sen-nihyaku-san-ju-yen desu)"         //  incorrect (iOS 16)      : "残高はです (zan-daka wa desu)"         Text("残高は ¥1,230 です")       }     }   } } The above sample code uses SwiftUI's Text but the same problem occurs with UIKit's UILabel.

I am trying to understand if what I am seeing is expected behavior or not with the following UIKit components. 1.I have a view controller "A" embedded in a navigation controller (part of a multi-step flow). Large titles are active on this navigation controller. In this view controller "A", I have a container view that contains another view controller "B" (I want to reuse the contents of B in other flows) Inside view controller "B" I have a UICollectionView using a diffable data source. When you load view controller "A" it appears to work fine. My collection view loads data, I see a nice list and when I scroll it... ... the expectation is it scrolls inside it's container and has no impact on the parent controller "B" However, the navigation bar and title in "A" reflect the content offset of the collection view. Scroll a couple lines, the large title turns small and centered on top. If I turn off large title, I still see the background color of the navigation bar change as it would if you were scrolling a view directly inside controller "A" without the container view. Am I supposed to be manually capturing the gesture recognizer in B and somehow preventing the gesture to bubble up to A? It seems like strange behavior to have to correct. Any suggestions? Thanks!

Is there a way to observe the currentEDRHeadroom property of UIScreen for changes? KVO is not working for this property... I understand that I can query the current headroom in the draw(...) method to adapt the rendering. However, our apps only render on-demand when the user changes parameters. But we would also like to re-render when the current EDR headroom changes to adapt the tone mapping to the new environment. The only solution we've found so far is to continuously query the screen for changes, which doesn't seem ideal. It would be better if the property would be observable via KVO or if there would be a system notification to listen for. Thanks!

I have find out that a UIViewRepresentable, even with a simples UIView, seems to never be dismantled when deleted from a ForEach and this can cause serious crashes. In the following example you can observe this behavior by deleting a row from the list. The dismantleUIView function of SomeUIViewRepresentable or the deinit of SomeUIView are never called. Has anyone faced this and found a solution for it? I have also filled a Feedback: FB11979117 class SomeUIView: UIView {     deinit {         print(#function)     } } struct SomeUIViewRepresentable: UIViewRepresentable {     func makeUIView(context: Context) -> SomeUIView {         let uiView = SomeUIView()         uiView.backgroundColor = .systemBlue         return uiView     }     func updateUIView(_ uiView: SomeUIView, context: Context) { }     static func dismantleUIView(_ uiView: SomeUIView, coordinator: Coordinator) {         print(#function)     } } struct Model: Identifiable {     let id = UUID() } struct ContentView: View {     @State var models = [Model(), Model(), Model(), Model(), Model()]     var body: some View {         List {             ForEach(models) { _ in                 SomeUIViewRepresentable()             }             .onDelete {                 models.remove(atOffsets: $0)             }         }     } }

I have an app that uses UITextView for some text editing. I have some custom operations I can do on the text that I want to be able to undo, and I'm representing those operations in a way that plugs into NSUndoManager nicely. For example, if I have a button that appends an emoji to the text, it looks something like this: func addEmoji() { let inserting = NSAttributedString(string: "😀") self.textStorage.append(inserting) let len = inserting.length let range = NSRange(location: self.textStorage.length - len, length: len) self.undoManager?.registerUndo(withTarget: self, handler: { view in view.textStorage.deleteCharacters(in: range) } } My goal is something like this: Type some text Press the emoji button to add the emoji Trigger undo (via gesture or keyboard shortcut) and the emoji is removed Trigger undo again and the typing from step 1 is reversed If I just type and then trigger undo, the typing is reversed as you'd expect. And if I just add the emoji and trigger undo, the emoji is removed. But if I do the sequence above, step 3 works but step 4 doesn't. The emoji is removed but the typing isn't reversed. Notably, if step 3 only changes attributes of the text, like applying a strikethrough to a selection, then the full undo chain works. I can type, apply strikethrough, undo strikethrough, and undo typing. It's almost as if changing the text invalidates the undo manager's previous operations? How do I insert my own changes into UITextView's NSUndoManager without invalidating its chain of other operations?

On iOS17, UIDevice.current.batteryLevel is returning values rounded to 0.05, such as 1, 0.95, 0.9. Which used to be a 1% granularity in iOS16. Is this a bug or a new feature?

Help,I have encountered a thorny problem! In systems with iOS 16.5 and above, there is a probability that the keyboard will not disappear after it appears. And once it appears, unless the app is restarted, all places where the keyboard is used cannot be closed. I have tried using the forced shutdown method [UIView endEditing:YES], but it didn't work. When this exception occurs, I notice that there will be two UITextEffectsWindow at the same time. Does anyone know how to solve it?

I recently was thinking about which is the best format to use for my little icons in the app, and was considering the performance of PNG versus SVG . I know that PNG decoders are hardware accelerated and parallelized, and the rendering is as simple as placing the pixels on the screen. On the other hand, SVG requires computation to determine the placement of pixels based on mathematical equations. Considering this, my current assumption is that PNG is faster to render than SVG. However, if SVG is also hardware-accelerated, it could alter the situation, although this may not be the case.