We develop an SDK that requires sharing a device-specific identifier with our web API, in order to guarantee that certain artifacts are only used on the correct device. For the device-specific identifier, we use UIDevice.currentDevice.identifierForVendor which should not be restricted under ATT. In production, many developers are getting back to us with complaints of web requests being blocked: nw_endpoint_handler_path_change [C1 [our url]:443 waiting parent-flow (satisfied (Path is satisfied), interface: en0[802.11], ipv4, dns, uses wifi)] blocked tracker Connection 1: received failure notification Connection 1: failed to connect 1:50, reason -1 Connection 1: encountered error(1:50) Task <FA03088C-DDFC-437E-A06F-E05CC930E3E0>.<1> HTTP load failed, 0/0 bytes (error code: -1009 [1:50]) Task <FA03088C-DDFC-437E-A06F-E05CC930E3E0>.<1> finished with error [-1009] Error Domain=NSURLErrorDomain Code=-1009 "The Internet connection appears to be offline." UserInfo={_kCFStreamErrorCodeKey=50, NSUnderlyingError=0x3031118f0 {Error Domain=kCFErrorDomainCFNetwork Code=-1009 "(null)" UserInfo={_NSURLErrorBlockedTrackerFailureKey=true, _kCFStreamErrorDomainKey=1, _kCFStreamErrorCodeKey=50, _NSURLErrorNWPathKey=satisfied (Path is satisfied), interface: en0[802.11], ipv4, dns, uses wifi}}, _NSURLErrorFailingURLSessionTaskErrorKey=LocalDataTask <FA03088C-DDFC-437E-A06F-E05CC930E3E0>.<1>, _NSURLErrorRelatedURLSessionTaskErrorKey=( "LocalDataTask <FA03088C-DDFC-437E-A06F-E05CC930E3E0>.<1>" ), NSLocalizedDescription=The Internet connection appears to be offline., NSErrorFailingURLStringKey=..., NSErrorFailingURLKey=..., _kCFStreamErrorDomainKey=1} Interestingly, I've made a few observations: The blacklist seems to be persistent, across devices. The blacklist stays in place regardless of whether we send no identifiable data in the web request (in fact, an empty ping request to our URL still gets blocked) The only way to get past the block is to use ATT, and request from the user that we track them across websites. This is false, because we don't track any user data whatsoever; and iOS disables ATT by default (in the settings app, users have to opt-in). Our iOS SDK already has an xcprivacy manifest mentioning the fact that we use a device-specific identifier, and that we send it to our web API URL. Still, we get blocked. How can we fix this? We can standup a proxy URL but I'd imagine it's only a matter of time before that also gets blocked. Apple has not provided any guidance on the specifics of how domains get blocked, and how they can be unblocked.

So I have 2 schemes: prod and staging. And both have their own Debug and Release configuration settings (one of which is the bundle ID, eg. prod-Release: com.mycomp.myApp, prod-Debug: com.myComp.myApp.debug, staging-Release: com.myComp.myApp.staging.debug, staging-Debug: com.myComp.myApp.staging.debug). Now, there is a framework that is bound to each bundle ID. The framework is not from a So I need 4 instances of the same framework for each bundle ID. I tried separating the frameworks into folders and just link them whenever I need to run the appropriate scheme. Basically like so: \{PROJECT_ROOT}\Frameworks\MyFramework\prod-Debug\MyFramework.xcframework \{PROJECT_ROOT}\Frameworks\MyFramework\prod-Release\MyFramework.xcframework \{PROJECT_ROOT}\Frameworks\MyFramework\staging-Debug\MyFramework.xcframework \{PROJECT_ROOT}\Frameworks\MyFramework\staging-Release\MyFramework.xcframework I've added a Run script to create the link before Compile Sources step like so: cd "${SRCROOT}/zDefend" if [ "${CONFIGURATION}" = "prod-Debug" ]; then rm -rf ./MyFramework.xcframework ln -s ./prod-Debug/MyFramework.xcframework MyFramework.xcframework elif [ "${CONFIGURATION}" = "prod-Release" ]; then rm -rf ./MyFramework.xcframework ln -s ./prod-Release/MyFramework.xcframework/ MyFramework.xcframework elif [ "${CONFIGURATION}" = "staging-Debug" ]; then rm -rf ./MyFramework.xcframework ln -s ./staging-Debug/MyFramework.xcframework MyFramework.xcframework elif [ "${CONFIGURATION}" = "staging-Release" ]; then rm -rf ./MyFramework.xcframework ln -s ./staging-Release/MyFramework.xcframework/ MyFramework.xcframework fi I only have 1 target, myApp. But in the Build Settings, I set it so that every configurations will create 4 different binaries with 4 different bundle ID like I've mentioned above. I've added the framework by its link and set it to "Embed and Sign". The framework itself is not from Cocoapods, Carthage or SwiftPM. We just need to compile it ourselves into 4 and bind them with their own bundle ID in the code. Then we manually added them into the project. It will throw error if the app's bundle ID is not the same with the bundle ID bound to the framework. This can be compiled just fine. But when in created 2 or more binaries in the simulator, it crashes. So this is possibly a linking issue. How do I solve this? Thank you.

Hello ! This will be my first blog post. I hope I will find the solution. So, I try to use the new iOS 17+ API for ScrollView. I have an array of messages that I want to show in scroll view. And I need to scroll to bottom when I tap to Send button. When I use scrollPosition(id:anchor:), it freeze the UI. The link to the screencast to demonstrate. I think the problem might be the keyboard when it shows and hides. But I can't understand what can I do. struct ChatView: View { @ObservedObject var viewModel: CoachViewModel @State private var scrollTo: Message.ID? var body: some View { ScrollView { LazyVStack { ForEach(viewModel.messages) { message in MessageView(viewModel: viewModel, message: message) } } .scrollTargetLayout() } .scrollPosition(id: $scrollTo) .background(.primaryBackground) .onChange(of: viewModel.scrollToBottom) { if $1, let lastMessage = viewModel.messages.last { withAnimation { scrollTo = lastMessage.id } viewModel.scrollToBottom = false } } } } struct Message: Identifiable, Hashable { let id: UUID let author: MessageAuthor let text: String } final class CoachViewModel: ObservableObject { @Published var messageTextInput = "" @Published var scrollToBottom = false @Published var messages: [Message] = [...] // ... more code ... func sendMessage() { messages.append( .init( id: .init(), author: .user, text: messageTextInput ) ) messageTextInput.removeAll() scrollToBottom = true } } struct CoachView: View { @StateObject private var viewModel = CoachViewModel() @State private var isSearching = false var body: some View { VStack(spacing: 0) { Group { CoachTitleView(viewModel: viewModel, isSearching: $isSearching) ChatView(viewModel: viewModel) } .onTapGesture { UIApplication.shared.endEditing() } if isSearching { MatchboxView(viewModel: viewModel) } else { InputTextFieldView(viewModel: viewModel) } } } }

Our company is going to develop an iOS library of functionality for inclusion in several apps, some of the apps will be developed in-house and some externally. When the library is included in external apps the source code shouldn't be visible. When the library is included in internal apps, having the option to view and step through the code within the library will be invaluable for debugging. How could I set things up so that two forms of the library can be easily generated - one exposing source code and the other not. Ideally they should be capable of being incorporated within the apps in the same manner regardless of if the source code is expose or not Is that possible? I looked at Swift Packages, but it seems its for distribution of source code only, and its not possible to hide the source code. However, conversely, a framework only contains binary and no source code. Is there any other option?

I’m looking into building a closed source XCFramework from a local Swift package that has dependencies on other packages, which can later be distributed via Swift Package Manager. In initial discussions, we thought xcodebuild does not support linking the dependencies externally, it always includes them statically in the built framework. It's my understanding this is because we're asking xcodebuild to build a framework from a local Swift Package. Is there another way this can be achieved? To explain in more detail: I have built a closed source SDK for other developers to integrate in their apps, currently distributed as an XCFramework. The interesting thing about the SDK is it has dependencies on other libraries, which need to be resolved when adding this SDK as a dependency to an app. The SDK’s dependencies should not be baked into our XCFramework. CocoaPods has worked well for that but we want to instead use SPM. The current project setup is an iOS framework Xcode project and an app Xcode workspace. The framework project is included in the app workspace and is in the same repo as the app, which allows me to modify the framework source code then run the app to test it. The framework project can also be opened independently and built to verify it doesn’t have any errors, but to verify it’s working I run it with the app. To distribute a new release I use xcodebuild to create an XCFramework and then deploy that. For this to work with CocoaPods I had to add a Podfile to the app directly as well as the framework directory so both have the dependencies available. This means I have an xcworkspace for the framework and not just a xcodeproj. I specify the framework workspace file in the xcodebuild command. To switch to a setup that utilizes Swift Package Manager, I created a Package.swift in the iOS framework project’s directory that specifies its dependencies, removed CocoaPods integration including deleting the workspace file, removed the framework project from the app’s workspace, added the Package as a local package to the app project, and added the framework directory via + > Add Files to “App” which adds the package to the top of the sidebar, making its source code available to edit within the app workspace. Everything is working when I run the app. Xcode properly resolves the dependencies for the local package and I can run the app to develop it. Now to create an XCFramework I run the following command in the framework directory (which contains the Package.swift): xcodebuild archive -workspace . -scheme FrameworkName -configuration Release -destination 'generic/platform=iOS' -archivePath './build/FrameworkName.framework-iphoneos.xcarchive' SKIP_INSTALL=NO BUILD_LIBRARIES_FOR_DISTRIBUTION=YES ENABLE_USER_SCRIPT_SANDBOXING=NO This succeeds however the dependencies have been linked statically thus included in our SDK. We need to only include the code from our framework and link to external dependencies, like it does with our current CocoaPods setup. I'm wondering what options there are to achieve this. Even if I need to change the project setup locally, for example to continue using a framework project/workspace instead of a local Swift package. It seems I just need xcodebuild to be able to create an XCFramework which can then be distributed with its own Package.swift file that specifies its dependencies. If it's not possible to link the dependencies externally, could you help me to understand the implications of including them statically? I don't know what problems could arise as a result of that or other concerns this would bring. Thanks!

Hear me out, developers at Apple! You've done an amazing job with privacy so far. The introduction of ContactAccess and the Contact Access Button this year is truly an amazing work of art. Well done. However, all the way back to the first iPhone I got in 2007, I wished Apple would make API's available to access the users installed apps (in a privacy way). This would (further) open up apps in the category of App Managers and App Launchers. Back in the early days the trick was to use deep link, but that was also awkward, since you needed to know the deeplink upfront and the apps needed to support it. Not ideal. It also had a side effect that Apple noticed, that evil party's used it to scan your device to know which apps you had installed and create a profile from that. Apple fixed that by using LSApplicationQueriesSchemes. Now you have to tell the App upfront which apps you will be calling (up to 50), or use Universal Link to be freed from this limitation again. Again not ideal. Why not turn it around and let the user decide? User central. Imagine an App Launcher app. There is a button in the launcher app where you as a user can click on to add an app. It calls an api available from Apple to launch a privacy shielded sheet with the apps the user got installed on its device. These are not exposed to the launcher app. When the user clicks on the installed app (or apps) it likes to add, the launcher app will get identifiers to launch the app. That's it. This would be limited access, perhaps the api could also expose full access, so permission could be granted once and the app will get all identifiers available. The choice will be up to the user. One step further, but this would be more nice-to-have, is the ability to access meta data of the app, such as icon, title, last launched, app size, etc. This way App Launchers can make decisions such as putting the most recent launched app in front or App Managers can use this to decide which apps you are not using and can advise to remove them to clean up space. Love to hear everyone's opinion. So let it be the start of the discussion.

1.The code structure is as follows: The main project imports an SDK. 2.The main project defines a struct A with a length of 1400 bytes, and the SDK also defines a struct A with a length of 1000 bytes. 3.The SDK does not expose the definition of struct A, but it is used in the implementation of the exposed API in the SDK. 4.In the usage process, the main project first calls the SDK's API, which uses struct A of SDK's and initializes it with memset. Then, in the main project, memset is used again to initialize the struct A declared in the main project. In the above scenario, it works fine on iOS versions lower than iOS18, but on iOS18 beat1, the app crashes randomly, and the crash scenarios are not fixed. We need to know whether iOS18 has made any improvements to the API for operating memory such as memset. Is memory management more stringent? Why do versions below iOS18 not cause problems? Thanks！

Related to https://developer.apple.com/forums/thread/756767 and https://feedbackassistant.apple.com/feedback/13893850 which was dismissed. Hi, I'm trying to install multiple Xcode versions with simulators on a single mac The problem is: xcrun simctl or any other simulator related process would hang several minutes after installing Xcode 16 beta additional components. If I delete /Library/PrivateFrameworks dir and reinstall additional components of Xcode 15.4, xcrun works fine without any hang. When hanging on CLI, simdiskimaged process skyrockets in CPU usage %, so that's why I assume CoreSimulator Framework is the main source of hang. Is there any way to use additional components(technically CoreSimulator Framework) of Xcode 15.4 on Xcode 16 beta? ※Plus, as I described in related thread, the hang disappears if I run Simulator.app on GUI. I already tried xattr -r -d com.apple.quarantine /Library/PrivateFrameworks or Xcode 16 beta with no use. Is there any command I can open Simulator.app headlessly on CLI?

Hi all, I had previously posted this on the Media Technologies section but didn't receive any replies so thought I would try my luck here. Apologies if re-posting questions on the forum in a short-ish time span is against the rules, but would greatly appreciate some assistance with the following situation regarding GroupSessionJournal and loading attachments. I'm currently working on a shareplay feature that allows users to pull 3d models from icloud and view it via volumes/immersive space on the vision pro. Was able to get the sharing working with multiple windows recently so now all that's left is to be able to sync/share the model in the SharePlay session. As I understand it, we should generally use GroupSessionMessenger for commands and light data like model positioning/syncing properties. Whereas for bigger pieces of data (images/videos/models), we should send these through GroupSessionJournal which the group session manages and syncs it for all users in the call. I have a button to get the current user's model data and add it to the journal via /// modelData is type `Data` try await journal.add(modelData) I have also set up a task to observe/receive updates to the journal's attachments in when receiving a group session. for await groupSession in MyModelActivity.sessions() { ... tasks.insert { Task { for await attachments in journal.attachments { for attachment in attachments { do { let modelData = try await attachment.load(Data.self) // throws error here - `notSupported` let modelUrl = writeModelDataToTempDirectory(modelData: modelData) self.modelUrlToLoadForGroupSession = modelUrl } catch let error { print("Error: \(error)") } } } } } } Not quite sure why I'm running into an error being thrown when attempting to load the attachment data on the other devices, any thoughts? The documentation for add(_:) and load(_:) say that the attachment should conform to Transferable but Data.Type should already conform to Transferable

When I turn on the PAC proxy configuration on my computer, the URL request part in the developed app will overflow the memory, but it will not happen if I turn off PAC

We have an implementation in which we use QUIC via a connection group, server are client are on Swift using the Network framework. Our use case is, the server should send data buffers to the client as fast and as much as possible, now the pace to call the send method from the server should be carefully done, because if we send too much data of course the client is not gonna be able to receive it. The question would be, is there a way to query the congestion window so we know on the server side, how much data we should be able to send at some point? Asking because we are not getting all the data we are sending from the server on our client side... We are using these settings: let options = NWProtocolQUIC.Options(alpn: ["h3"]) options.direction = .bidirectional // options.idleTimeout = 86_400_000 options.maxUDPPayloadSize = Int.max options.initialMaxData = Int.max options.initialMaxStreamDataBidirectionalLocal = Int.max options.initialMaxStreamDataBidirectionalRemote = Int.max options.initialMaxStreamDataUnidirectional = Int.max options.initialMaxStreamsBidirectional = 400 options.initialMaxStreamsUnidirectional = 400 Questions: 1.- Can we get a little more detail in above options, specifically on their impact to the actual connection? 2.- IsinitialMaxData the actual congestion window value 3.- Are we missing something or making incorrect assumptions? Thanks in advance.

I have a unique need here and hope there is someone out there that might be of help. There is a backend server that will send an x509 certificate and private key (as strings) after the mobile apps on-boarding process. Additionally, the app includes an AWS SDK that is used to talk to their IoT system. This SDK requires PKCS12 certificate format to pass authentication. (I believe the common method is to have bundled the cert into the app which is not an option for me here sadly) I suspect it may be possible to use some openSSL iOS framework to do this conversion at runtime but have not personally tried it yet as my go-to is usually trying things first with Apples APIs. So my question becomes is there a way to meet this requirement using any of the security APIs or other APIs that apple has like swift-nio-ssl? Thank you very much for your time. Best, Michael

I have an app that uses the ImageCaptureCore's ICDeviceBrowser to find and connect to external digital cameras. Prior to iOS 18 this worked just fine, the device browser would start up and find any cameras connected via USB. However since the update the device browser fails to ever detect any connected device or to trigger any delegate events at all after browser start. I noticed that the Contents authorization in iOS 18 is undetermined, where in previous iOS versions it would default to authorized. I tried to resolve this by requesting authorization, however this immediately returns denied without ever having prompted the app user for permission. I do have the Camera Privacy Usage description setup, and also am able to request permission for the iOS camera successfully. How can I successfully request contents authorization via ICC or otherwise? Or are there alternative Apple libraries I can use for finding and connecting to external digital cameras on iOS?

i build apps that act as Screen Readers to 1) add Vim motions everywhere on macOS 2) click (and more) AX Elements through the keyboard 3) scroll through the keyboard. it works extremely well with native apps. with non-native apps, i need to blast them with some extra AX Attributes (AXManualAccessibility, AXEnhancedUserInterface) to get them to expose their AX Elements. but there are a couple of apps tho which i can't get them to expose their AX Elements programmatically. now the weird thing is as soon as i start VoiceOver, those apps open up. or for some, if i use the Accessibility Inspector to go through their AX Elements, then they start opening up. so i'm wondering, is there one public known way that i'm missing to open up those apps, or is Apple using private APIs? any way i could make my apps behave like VoiceOver or the Accessibility Inspector to force those recalcitrant apps to open up? thanks in advance.

My SampleApp, named "myApp," already supports deep links such as "https://myApp.com/tab/record/10020." When this link is clicked, it navigates to myApp, and AppDelegate handles the further process. Now, I need to support another link (domain), "https://appcloud.com/tab/record/10020," so that clicking this link also navigates to myApp. Is there any way to solve this kind of problem without adding "apple-app-site-association"?

I'm working on an app that uses HomeKit-enabled accessories such as wall plugs to control a photographic enlarger in a darkroom. This requires precise timing when toggling the power state of the plug. For example, the timer in my app might be set to 5 seconds, so I turn on the plug using writeValue() on the plugs power state characteristic, wait 5 seconds, then turn it off again. I want to be able to measure if the plug actually responded to the command and if there was any delay in the plug actually turning on/off so I can display a warning to the user if network latency resulted in the plug being on for longer than the set time. Does writeValue() (in my case, the async/await version) only return when the plug has been turned on, or does it return as soon as the command has been sent, regardless of if it has been received/acted on? Is there a way I can (a) quickly verify that the plug has been turned on/off, and (b) measure how long it took for the plug to act on the request, that is, the time elapsed between when writeValue() is called and when the plug actually updates to the corresponding value?

Hi there, My app detects connected BlackMagic devices on a user's machine and this is done using the DeckLink SDK which first tries to load /Library/Frameworks/DeckLinkAPI.framework using CFBundleCreate. I have not been able to create a notarized app which successfully detects the devices. Either the DeckLinkAPI works or the app starts up without showing "the developer cannot be verified" on my test computer but never both. This is what I've tried so far: signed app: DeckLinkAPI available hardened runtime: DeckLinkAPI not available hardened runtime + com.apple.security.cs.disable-library-validation: DeckLinkAPI available, notarization succeeds yet the "the developer cannot be verified" I've also tried to use weak linking to DeckLinkAPI.framework instead of including the SDK's CFBundleCreate code but that made no difference: I still needed the com.apple.security.cs.disable-library-validation entitlement for that to work which caused "the developer cannot be verified". DeckLinkAPI.framework is notarized: > codesign --test-requirement="=notarized" --verify --verbose /Library/Frameworks/DeckLinkAPI.framework /Library/Frameworks/DeckLinkAPI.framework: valid on disk /Library/Frameworks/DeckLinkAPI.framework: satisfies its Designated Requirement /Library/Frameworks/DeckLinkAPI.framework: explicit requirement satisfied Is there any way to successfully notarize an app to use the DeckLink SDK or any other thirdparty notarized framework which is distributed seperately?

Is it possible to style a Picker on macOS to look like it was in a Form, but without actually being in a grouped Form? Unfortunately I can't use Form because its style is not customizable enough. I've tried setting various buttonStyles but I only got close to what I want. The biggest problem is that the picker is left-aligned and I need it right-aligned.

Hello, On iOS it is possible to use the keyboard spacebar to move the caret while editing text. The user has to longpress on the spacebar, and then the keyboard would turn into a trackpad that moves around a floating caret. Is there a way to get the position of the caret while the user uses this space bar navigation? And by the position I mean the position in an app or in a frame? Are there any other events that this action can send out? Thanks a lot!

I have an NSTextLayoutManager set up with NSTextContentStorage and NSTextContainer. To work out the height of the content, I call the method updateContentSizeIfNeeded() which contains the code textLayoutManager.enumerateTextLayoutFragments(from: textLayoutManager.documentRange.endLocation, options: [.reverse, .ensuresLayout]) { layoutFragment in height = layoutFragment.layoutFragmentFrame.maxY return false } The first time this is called, it returns the correct height. Then I add a new character to the start of the NSTextContentStorage like so textContentStorage.performEditingTransaction { storage.replaceCharacters(in: NSRange(location:0, length: 1), with: "a") } textLayoutManager.ensureLayout(for: textLayoutManager.documentRange) textLayoutManager.textViewportLayoutController.layoutViewport() updateContentSizeIfNeeded() This time, the height returned is ~600px too big. The state of the NSTextLayoutFragment is set to layoutAvailable The next time I add a character to textContentStorage using the same code above, the height returned is correct again. I can work around this by calling enumerateTextLayoutFragments from the start of the document and not in reverse, then ignoring all fragments except the last one, but I don't know if that's the correct way to do it, or if I should be doing something else