I add a series of buttons in the UIToolbar of accessoryView. Before liquidGlass (Xcode 26.6, iOS 26), I get this: Buttons are properly stacked, allowing up to 8 buttons, which is what I need. But with liquidGlass, buttons are now enclosed in a useless bubble, and thus take much more space. No way to accommodate 8 buttons anymore, they are displayed in the "…" followup section, which is very inconvenient. In iOS 26, I can still set UIDesignRequiresCompatibility to true and get the expected presentation. But that does not work in Xcode 27 / iOS 27. Code is essentially the following: let bar = UIToolbar(frame:CGRect(x: 0, y: 0, width: 200, height: 44)) let letterBack = UIImage(named: "letterBackground")! let targetSize = CGSize(width: 36, height: 36) let scaledImage = letterBack.scalePreservingAspectRatio( targetSize: targetSize ) let letterI = UIBarButtonItem(title: "I", style: .plain, target: self, action: #selector(letterTapped(_:))) letterI.setBackgroundImage(scaledImage, for: .normal, barMetrics: .default) // Same for other letters bar.items = [letterI, letterV, letterX, letterL, letterC, flexibleSpace, peseudoReturn] bar.sizeToFit() aTextField.inputAccessoryView = bar I have tried to use barMetrics: .compact to no avail So, a few questions: Is it possible, with liquidGlass, to have the buttons displayed without their ellipse bubble, so that they stack against each other ? Is there other appearence setting to set ? Or should I give up toolbar and create a collectionView that I will place atop keyboard ? In anycase, liquidGlass is really problematic in term of screen estate use.

I'm building a custom keyboard extension that offers voice dictation. Because keyboard extensions are constrained (memory cap ~30–48 MB, restricted audio session access), I delegate recording to my container app: User in a host app (e.g., Safari) taps the mic in my keyboard extension. The keyboard calls extensionContext.open(URL("myapp://dictation")) to launch the container app. The container app records audio via AVAudioEngine + SFSpeechRecognizer, writes the final transcript to the App Group, and signals completion via a Darwin notification. 4. The user is expected to be returned to the original host app (Safari) automatically so they can keep typing. The problem (step 4): On iOS 26.4 I can no longer identify which app was the host. Every previously-known path returns nil for the keyboard extension's host: parent.value(forKey: "_hostBundleID") → returns the literal string parent.value(forKey: "_hostApplicationBundleIdentifier") → returns NSNull xpc_connection_copy_bundle_id on the underlying XPC connection (via PKService.defaultService.personalities[…]) → returns NULL NSXPCConnection.processBundleIdentifier on extensionContext._extensionHostProxy._connection → returns nil proc_pidpath(hostPID, …) → EPERM from the keyboard sandbox LSApplicationWorkspace.frontmostApplication → selector unavailable from the extension RBSProcessHandle.handleForIdentifier:error: → returns an RBSServiceErrorDomain error Without the host's bundle ID, the container app has no way to call LSApplicationWorkspace.openApplicationWithBundleID: (the technique that worked on iOS 25 and earlier). UIApplication.suspend() correctly sends the container to background, but iOS treats us as a "fresh launch" — it returns the user to the Home Screen instead of Safari, because the container app was launched by an extension, not directly by Safari. KeyboardKit's maintainer reached the same conclusion (issue #1014) and shipped 10.4 without the feature. My questions: Is there a public, App-Store-safe API in iOS 26+ for a custom keyboard extension to identify its host application, or for the container app (launched via the extension's openURL) to identify which app initially hosted the extension that opened it? UIOpenURLContext.options.sourceApplication reports the extension's own container, not the actual host. 2. Is there a public mechanism for "return to source app" when the container app was launched by an extension's openURL? Equivalent to the ← Source affordance iOS shows for normal inter-app openURL, but triggered programmatically by the launched app. 3. Some popular keyboards (e.g., 微信输入法 / WeChat Keyboard) still appear to round-trip through their container app on iOS 26.4 and return the user to the original host — including the iOS ← WeChat back affordance in the host's status bar afterward. What's the recommended approach to achieve this? If it requires a specific scene-activation flow, NSUserActivity pattern, or extension-context configuration, please point at the relevant docs. 4. If there is no public path today, is FB22247647 (or a related radar) the right place to track this? Should developers in this position migrate to in-extension audio capture (which has its own significant constraints in keyboard extensions)? I'd much rather not rely on private APIs. Concrete guidance — or even an acknowledgment of which direction Apple intends — would help thousands of custom-keyboard developers who currently have a degraded voice-input experience on iOS 26.4+. Tested on iPhone 12 Pro Max running iOS 26.4.2 (build 23E261), Xcode 26.x, Swift 5. Thanks!

We went through Apple's CarPlay documentation and, based on what we found, CarPlay provides templates for both Navigation and Audio, and our app has entitlements for both. One thing we would like to confirm before proceeding with implementation is whether a single app can expose both Navigation and Audio experiences in CarPlay, using both entitlements. Specifically, can the same CarPlay app provide a navigation experience (e.g., using CPMapTemplate) while also exposing an Audio/"Now Playing" experience, or does Apple expect these experiences to be implemented as separate apps or otherwise treated as separate CarPlay experiences? We'd also like to confirm whether CarPlay now allows an app to display a split-screen interface with navigation/map content on one side and audio controls or "Now Playing" controls on the other side. If so, is this something a single app with both Navigation and Audio entitlements can implement? The documentation explains the available templates and entitlements, but we couldn't find explicit guidance on whether the same app can leverage both Navigation and Audio experiences or whether these can be presented together in a split-screen layout. We'd like to clarify this before implementing the "Now Playing" screen, to ensure our approach aligns with Apple's CarPlay requirements and guidelines.

The context is partially expressed in an earlier post. In summary: There is an iOS App target that contains minimal code, only to load a Framework explicitly at runtime using dlopen and dlsym, instead of the usual load-time imports in Apple platforms. For iOS app (C++ (primary) and Swift), the entry point is a UIApplicationDelegate conformer class - AppDelegate, marked with @main. But the problem is, the AppDelegate class cannot remain in the App target, which has barely any logic. The App target is a thin loader. The AppDelegate contains some methods such as application(_:didRegisterForRemoteNotificationsWithDeviceToken:) that needs some logical processing, which is not present in the App target. Instead of using dlsym (to hand over to the Framework) for every AppDelegate event that doesn't have a broadcast notification, the thought was to move the AppDelegate class into the Framework, and the entry point in App target is now main.swift. This keeps the Framework clean and minimal with the following steps: Interop to C++ Explicitly loading the MachO binary inside the Framework using dlopen Loading the symbol using dlsym Invoking the Framework entry point Then, the Framework entry point in C++ creates the UIApplication class and the UIApplicationDelegate using UIApplicationMain(_:_:_:_:) method, which doesn't return as it transfers control to the UIApplicationDelegate. This is against the recommended @main entry point, but based on research, @main seems like syntactic sugar to avoid writing boilerplate code. But in my case, which needs to avoid instantiating the UIApplicationDelegate in the App target, using main.swift, even for an iOS app, is the best fit. I understand that main thread has to be returned back to the OS asap for processing user events etc., and the intent is to not execute the entire startup logic of the app in main thread. Wanted to confirm if this approach of using main.swift entry point is valid for iOS, iPadOS and tvOS apps too and in which case, these flows can converge to macOS, which is already using main.swift approach.

I'm working on a suite of apps supporting macOS, iOS and iPadOS (potentially tvOS, watchOS and visionOS in the future). Each of these App targets contain minimal code to only load the framework dynamically instead of the recommended load-time imports for Apple platforms. The rationales for runtime loading of framework (using dlopen and dlsym) is expressed in earlier post - fyi. Each app can load multiple frameworks at runtime. Instead of naming the frameworks like this - AppName_purpose1.framework, AppName_purpose2.framework, AppName_purpose3.framework, can it be named as AppName.purpose1, AppName.purpose2, AppName.purpose3 etc? Basically, change the Framework bundle extension name from .framework to a custom name based on purpose. The folder's extension name is changed, but it's still a framework bundle. The advantage of this approach is, in my project, Frameworks belonging to each of the apps cleanly distinguish themselves with a concise name. While this post is about Apple platforms, I'm also checking if other platforms allow to change the names of dynamic libraries (windows allows change). Using my custom extension can standardize the dynamic library's extension name across all platforms. Easy framework name construction - The Framework name is now the same as the App bundle name, which can be queried and this string can be appended with an appropriate extension to load all the Frameworks. Is it possible to change the extension name of the Framework bundle from the default .framework? If yes, how?

For a UIKit app based on scenes (UIScene), is it safe to reference UserDefaults in code that is executed from UIApplicationDelegate/application(_: didFinishLaunchingWithOptions:) ? I've read that in iOS 15, there were undocumented scenarios involving app prewarming that would cause UserDefaults reads to fail within a window of time after device reboots, as described at https://christianselig.com/2024/10/beware-userdefaults/ The failure mode is that an app would be released, and months later, a small fraction of users would report failures consistent with UserDefaults reads unexpectedly returning nil, causing a loss of data. The user experience is bad, and debugging this behavior is then challenging because of how rarely it occurs. Apple's https://developer.apple.com/documentation/uikit/app_and_environment/responding_to_the_launch_of_your_app/about_the_app_launch_sequence#3894431 seems to suggest that prewarming only executes an app "up until, but not including when main() calls UIApplicationMain(_:_:_:_:), but https://stackoverflow.com/questions/71025205/ios-15-prewarming-causing-appwilllaunch-method-when-prewarm-is-done documents that UIApplicationDelegate/application(_: didFinishLaunchingWithOptions:) has in fact been observed executing during app prewarming in scene-based apps. So, my question: In an app based on scenes, if I'd like to reference UserDefaults within UIApplicationDelegate/application(_: didFinishLaunchingWithOptions:), when is it safe to do this? I'm guessing the answer is one of these: Never. Only in apps that don't support scenes. Only in iOS 16 or later. Only in IOS 17 or later. Is it guaranteed safe to reference UserDefaults in UIWindowSceneDelegate/scene(_:willConnectTo:options:) or later? Is there documentation from Apple regarding this issue? Thank you.

With the 27 OSes using UserDefaults.standard.integer(forKey: ) can cause a crash with EXC_BAD_ACCESS (code=1, address=0x0) It has been seen on a Multiplatform app, up to now tested on iOS/iPadOS and visionOS 27 Beta 1. In our code we use UserDefaults.standard.integer(forKey: ) from a singleton called during the SwiftUI app init(), and we don't know yet if this is the only moment there is a crash as we can't go farther. The API should return 0 if it can't get a value. There is no reason the app should crash if the API conforms to its contract. Running the same code from Xcode on iOS 26 runs it without issue. FeedBack FB23310748

I have an app on the App Store that takes user content and creates a Generable struct out of it. In the last couple weeks I have started getting complains from my users that the part of the app leveraging FoundationModels isn't working properly. In my testing I noticed that the same request that would've worked a couple weeks ago is now getting errors with guardrails violation. I'm initializing my model this way LanguageModelSession(model: SystemLanguageModel(guardrails: .permissiveContentTransformations)) // I'm aware that .permissiveContentTransformations does not apply to Generable, but I'd really really really really love it, if it did!. This started around the iOS 26.5/macOS 26.5 releases and I wonder if there's a way to fix it.

When a user uses the "Add Name" feature on an unknown number in their Call Logs, the name appears under "Other Known Contacts" in the native iOS Contacts app. The Problem: CNContactStore completely ignores these contacts during a standard fetch/enumeration. When user gives limited permission they can search for that contact and select it, but it won't be visible in my App as it's not technically a contact. Is CNContactStore intentionally blocked from reading "Other Known Contacts" for privacy reasons or are there any future plans to expose API so that third party apps can access it?

I am building an interactive application. My application architecture looks like this: There will be a LoaderBinary that will load 1 or more shared libraries (.framework in this case). This is also where the entry point from an OS perspective lies. There will be 1 or more frameworks built. This framework is expected to have large part of my logic as shared code so that multiple flows like the application flow, widget, notifications etc can reuse code by loading this framework itself. Now, I want to achieve the following: Building a framework independently - I believe this is doable and works fine too. Building a mach-o binary - This is what we are not clear if it is allowed or not to build just a mach-o binary ? Yes, there is an option to build a command-line tool but as this is an interactive binary, what should be the path to take ? Building a macOS bundle (.app) using 1 and 2 - Now, as I have a PRE-BUILT framework and a PRE-BUILT mach-o binary, can I create a application bundle using these ? Some directions here will help to take this forward - in alignment with both my architecture as well as how Apple Build system works. Thanks!

Hi everyone, Recently I downloaded the last XCode (Version 27.0 beta) and I'm trying to start to play with CoreAI but it's not possible so far to import the CoreAI framework. I'm using the last version of the iPhone simulator. What am I missing?

The documentation specifies that when Contacts framework returns unified contacts that each fetched unified contact object (CNContact) has its own unique identifier that’s different from any individual contact’s identifier in the set of linked contacts and that when refetching a unified contact, that this identifier should be used. There is also an analogous identifier within the list of contactRelations, but each of these don't seem to corespondent to the unified contacts. For example, is a new contact (Sheryl Zakroff) is created in the simulator Contacts and their spouse is set to Hank Zakroff. However, the GUID created for the contactRelations identifier does not correlate to the original Hank Zakroff GUID and cannot be searched. Is this a bug or what is the indent of the contactRelations identifier? Here's a debug output of walking the unifiedContacts: Name: Hank Zakroff 2E73EE73-C03F-4D5F-B1E8-44E85A70F170 - Other : (555) 766-4823 - Other : (707) 555-1854 Name: David Taylor E94CD15C-7964-4A9B-8AC4-10D7CFB791FD - Other : 555-610-6679 Name: Sheryl Zakroff DE783BC8-7917-4138-93F6-3AF0FD4CE083 - Other : (707) 555-1854 - Spouse: <CNContactRelation: 0x60000000dd60: name=Hank M. Zakroff> - 534B467D-CA00-46D3-897C-16EEA782C9CF - Looking for ["534B467D-CA00-46D3-897C-16EEA782C9CF"] []

I have a product that currently communicates to a piece of proprietary Medical hardware via USB. We have a driver created with USBDriverKit that is a hard requirement for this product to function. Currently, we are delivering an iPadOS application that utilizes the driver. Our next step is to create an SDK that includes all of the Frameworks and the driver utilized by the application. Since the driver is a system extension that lives in the application, how do we bundle it with our SDK and make it available for use by third-parties?

I am requesting technical clarification and a formal feature addition regarding the availability of per-frame calibration data for Standard RAW captures on iPhone 17 and 17 Pro. The Technical Gap: Currently, AVCaptureVideoDataOutput provides a direct path to the AVCameraCalibrationData object, allowing real-time access to the Intrinsic Matrix, Radial/Tangential Distortion coefficients, and Lens Shading Maps. However, this same level of geometric transparency is missing from the AVCapturePhoto RAW/ProRAW delegate. While standard RAW files contain some metadata, they lack the physics sidecar required for precision manual alignment. Because the lens assembly in the iPhone 17 Pro is dynamic shifted constantly by Optical Image Stabilization (OIS) and high-speed Voice Coil Motors (VCM) for focus a static or factor calibration is mathematically insufficient for high-precision workflows. The Problem: Without the 1:1 hardware state at the millisecond of exposure, we cannot perform accurate geometric reconstruction from RAW stills. We are forced to choose between the high dynamic range of a RAW sensor dump and the geometric precision of the video pipeline. Final Questions: Is there a documented, supported method to force the inclusion of the AVCameraCalibrationData object or a raw metadata sidecar in the AVCapturePhoto workflow? If not, can Apple provide parity between the Video and Photo APIs so that the "rawest" data (RAW) is accompanied by the "rawest" physics (Calibration Data)? Providing the pixels without the lens geometry limits the utility of the RAW format for any technical workflow requiring sub-pixel geometric integrity.

Hello, After adapting Liquid Glass for iOS 26.5, I encountered an issue: the leftBarButtonItem(s) shift right, leaving a large space on the left after launching and dismissing the camera. I tried adding a negative spacer and setting leading space, but neither worked. This only happens when launching the camera; it does not occur during photo selection.

Why is the WiFiAware framework not importable in Mac Catalyst? Are there any plans to support Wi-Fi Aware technology on Mac Catalyst in the future?

Hello everyone. I use Translation Framework in my application. During development everything was fine, Translation framework worked well, but after two or three days of using the production version (that was published in AppStore and available for others also!) - my application stopped working. Translation framework gives errors: Error sending 1 paragraphs Error Domain=TranslationErrorDomain Code=16 "Translation failed" UserInfo={NSLocalizedDescription=Translation failed, NSLocalizedFailureReason=Offline models not available for language pair} Failed to translate input 0; returning error: Error Domain=TranslationErrorDomain Code=16 "Translation failed" UserInfo={NSLocalizedDescription=Translation failed, NSLocalizedFailureReason=Offline models not available for language pair} Received unbridged NSError to API, converting to .internalError: Error Domain=TranslationErrorDomain Code=16 "Translation failed" UserInfo={NSLocalizedDescription=Translation failed, NSLocalizedFailureReason=Offline models not available for language pair} Once again - it worked when I developed it, it was released on the AppStore, and suddenly it stopped working!

Hello all, I'm currently using Apple Connect API to retrieve download info from my app. Namely I'm using App Store Installations and Deletions under the Analytics Reports API, using an ONGOING request with daily granularity. Unfortunately, the data received is not matching the data from apple store connect UI. Recently I found this article, from apple help documentation for Analytics Reports API that states the following: "(..) Some usage reports are not generated if minimum thresholds are not met (...)" and "(...) Additionally, a small amount of statistical noise is added to metrics, preventing the reverse-engineering of individual activity while preserving accuracy at scale." I kindly ask if: What is stated in 1. happens often within the report generated from App Store Installations and Deletions. If yes, do you have any idea of the percentage, so we take that into account when showing our metrics? What is stated in 2. happens often within the report generated from App Store Installations and Deletions. If yes, how to handle the statistical noise? Discard events with the field "count" < 5, for instance? Thank you very much, HH DOCs: https://developer.apple.com/documentation/analytics-reports/app-installs https://developer.apple.com/help/app-store-connect-analytics/overview/analytics-reports-api https://developer.apple.com/documentation/analytics-reports

I am working on a cross-platform application where, on Android and Windows, I explicitly load dynamic libraries at runtime (e.g., LoadLibrary/GetProcAddress on Windows and equivalent mechanisms on Android). This allows me to control when and how modules are loaded, and to transfer execution flow from the main executable into the dynamically loaded library. I want to follow a similar approach on macOS (and also iOS) and explicitly load a framework (instead of relying on implicit linking via import). From my exploration so far, I have come across the following options: Using Bundle (NSBundle) - Load framework using: let bundle = Bundle(path: path) try bundle?.load() Access functionality via NSPrincipalClass and @objc methods (class-based entry) Using dlopen + dlsym Load the framework binary and resolve symbols: let handle = dlopen(path, RTLD_NOW) let sym = dlsym(handle, "EntryPoint") Expose Swift functions using @_cdecl Using a hybrid approach (Bundle + dlsym) - Use Bundle for loading and dlsym for symbol access From what I understand: Bundle works well for class-based/plugin-style designs using the Objective-C runtime while dlopen/dlsym works at the symbol level and is closer to what I am doing on other platforms However, my requirement is specifically: Explicit runtime loading (not compile-time linking) Ability to transfer execution flow from the main executable into the dynamically loaded framework **What is the recommended approach on macOS for this kind of explicit dynamic loading, or is implicit loading the way to go? Also, would it differ for interactive and non-interactive apps? ** In what scenarios would Apple recommend using Bundle instead of dlopen? Is there any other methods best for this explicit loading of frameworks on Apple?

[FB22327481] We are observing a rare issue where alarms scheduled using AlarmKit occasionally fire exactly at 12:00 AM, even though the alarm was scheduled for a different time. This issue happens only for a very small number of users (for example, 1–2 users per several thousand per day), but multiple reports confirm that the alarm goes off exactly at midnight. We also found that other developers are experiencing the same issue: https://developer.apple.com/forums/thread/815714 Observed behavior For a small number of users: Alarm fires at exactly 00:00 (midnight) This happens even when the scheduled alarm time is something like 7:30 AM Happens rarely Hard to reproduce internally Appears to happen only on real devices in production Additional notes We are using AlarmKit fixed schedules (not relative) Dates passed to AlarmKit are correct at scheduling time We do not intentionally create midnight alarms Issue seems random Question Are there any known limitations or edge cases with AlarmKit fixed schedules that could cause alarms to fire at midnight? For example: Invalid date fallback? Schedule expiration fallback? Time zone changes? Background refresh timing? Maximum schedule window? Any guidance would be appreciated, as this is affecting real users but is difficult to reproduce.