Post

Replies

Boosts

Views

Activity

Testing app intents and forcing Shortcuts app to refresh
The Verify the behavior of your intent in Simulator or on-device documentation says that it's sufficient to build and run the app and open the Shortcuts app to test an app intent, but that doesn't seem to be the case. I always needed to at least move the debugged app to the Applications folder before the app intents showed up in Shortcuts, and even then, most of the times I also need to wait a lot or restart the Mac. When updating an existing app intent (for instance by changing its title), building the app, overwriting the existing one in the Applications folder and restarting Shortcuts is not sufficient to make the new title appear in Shortcuts. Is there an efficient way to test app intents in the Shortcuts app? I already created FB15638502 one month ago but got no response.
0
0
418
Dec ’24
Camera zoom in to 3D point in SceneKit scene
I would like to implement zoom functionality in my SceneKit game: when the user performs the pinch gesture on a point on the screen, the scene zooms in to make that point larger. Until now I simply changed SCNCamera.focalLength, but this simply zooms in to the center of what is currently visible on screen. Is it somehow possible to implement the zoom functionality described above by perhaps interactively rotating the camera at the same time towards the pinched point? Is there a formula for this? I would like to avoid suddenly rotating the camera to face the pinched point when the pinch gesture begins and then zoom in while the pinch is in progress.
0
0
630
Dec ’24
Notification when mounting volume on iOS
On macOS, we have didMountNotification but there doesn't seem to be an equivalent for iOS. Is there a way to be notified when a volume is mounted on iOS? I would like to use it in my iOS app I'm currently porting from macOS, which starts a synchronization from the volume (which has been previously selected in a NSOpenPanel) as soon as it's mounted.
3
0
442
Feb ’25
Resolving URL from bookmark data doesn't automatically mount SMB volume on iOS
On macOS, the Finder allows to connect to a server and store the login credentials. When creating a bookmark to a file on a server and resolving it again, the server is mounted automatically (unless I provide the option URL.BookmarkResolutionOptions.withoutMounting). I just tried connecting to my Mac from my iPad via SMB in the Files app and storing a bookmark to a file on the server, but disconnecting the server and trying to resolve the bookmark throws the error (I translated the English text from Italian): Error Domain=NSFileProviderErrorDomain Code=-2001 "No file provider was found with the identifier "com.apple.SMBClientProvider.FileProvider"'" UserInfo={NSLocalizedDescription=No file provider was found with the identifier "com.apple.SMBClientProvider.FileProvider"., NSUnderlyingError=0x302a1a340 {Error Domain=NSFileProviderErrorDomain Code=-2013 "(null) "}} Every time I disconnect and reconnect to the server, selecting the same file returns a different path. The first time I got /private/var/mobile/Library/LiveFiles/com.apple.filesystems.smbclientd/WtFD3Ausername/path/to/file.txt The next time WtFD3A changed to EqHc2g and so on. Is it not possible to automatically mount a server when resolving a bookmark on iOS? The following code allows to reproduce the issue: struct ContentView: View { @State private var isPresentingFilePicker = false @AppStorage("bookmarkData") private var bookmarkData: Data? @State private var url: URL? @State private var stale = false @State private var error: Error? var body: some View { VStack { Button("Open") { isPresentingFilePicker = true } if let url = url { Text(url.path) } else if bookmarkData != nil { Text("couldn't resolve bookmark data") } else { Text("no bookmark data") } if stale { Text("bookmark is stale") } if let error = error { Text("\(error)") .foregroundStyle(.red) } } .padding() .fileImporter(isPresented: $isPresentingFilePicker, allowedContentTypes: [.data]) { result in do { let url = try result.get() if url.startAccessingSecurityScopedResource() { bookmarkData = try url.bookmarkData() } } catch { self.error = error } } .onChange(of: bookmarkData, initial: true) { _, bookmarkData in if let bookmarkData = bookmarkData { do { url = try URL(resolvingBookmarkData: bookmarkData, bookmarkDataIsStale: &stale) } catch { self.error = error } } } } }
2
0
522
Feb ’25
Crash when assigning NSImage to `@objc dynamic var` property
Xcode downloaded a crash report for my app which I don't quite understand. It seems the following line caused the crash: myEntity.image = newImage where myEntity is of type MyEntity: class MyEntity: NSObject, Identifiable { @objc dynamic var image: NSImage! ... } The code is called on the main thread. According to the crash report, thread 0 makes that assignment, and at the same time thread 16 is calling [NSImageView asynchronousPreparation:prepareResultUsingParameters:]. What could cause such a crash? Could I be doing something wrong or is this a bug in macOS? crash.crash
11
0
266
Aug ’25
NSView.knowsPageRange(_:) called twice when showing print panel
When creating a default macOS document-based Xcode project and using the code below (and wiring the File menu's Print item to printDocument: instead of the default print:, which does nothing), opening the print panel causes PrintView.knowsPageRange(_:) to be called twice. Is this a bug? My app populates PrintView dynamically, and for large documents it can be quite inefficient to populate it once, only for the contents to be immediately discarded and populated again. A workaround that came to my mind would be to check if the print options have changed, though I'm not sure if it's a reliable indicator that the print preview is effectively the same. I created FB17018494. class Document: NSDocument { override func makeWindowControllers() { addWindowController(NSStoryboard(name: NSStoryboard.Name("Main"), bundle: nil).instantiateController(withIdentifier: NSStoryboard.SceneIdentifier("Document Window Controller")) as! NSWindowController) } override func printOperation(withSettings printSettings: [NSPrintInfo.AttributeKey : Any]) throws -> NSPrintOperation { return NSPrintOperation(view: PrintView(frame: CGRect(x: 0, y: 0, width: 100, height: 100)), printInfo: NSPrintInfo(dictionary: printSettings)) } } class PrintView: NSView { override func knowsPageRange(_ range: NSRangePointer) -> Bool { print("knowsPageRange") range.pointee = NSRange(location: 1, length: 1) return true } }
Topic: UI Frameworks SubTopic: AppKit
1
0
87
Mar ’25
NSTableView.clickedRow sometimes is greater than number of rows
Xcode has been downloading many similar crash reports for my app for some time now, related to an index out of range runtime exception when accessing a Swift array. The crashes always happen in methods triggered by user input or during menu item validation when I try to access the data source array by using the following code to determine the indexes of the relevant table rows: let indexes = clickedRow == -1 || selectedRowIndexes.contains(clickedRow) ? selectedRowIndexes : IndexSet(integer: clickedRow) I was never able to reproduce the crash until today. When the app crashed in the Xcode debugger, I examined the variables clickedRow and selectedRowIndexes.first, which were 1 and 0 respectively. What's interesting: the table view only contained one row, so clickedRow was effectively invalid. I tried to reproduce the issue several times afterwards, but it never happened again. What could cause this issue? What are the circumstances where it is invalid? Do I always have to explicitly check if clickedRow is within the data source range?
Topic: UI Frameworks SubTopic: AppKit Tags:
4
0
120
Apr ’25
Crash in IndexSet.map during menu item validation in client report downloaded by Xcode
For many years I've had the following code to access the active objects of a table view in my App Store app: class MyViewController: NSViewController: NSMenuItemValidation { private var tableView: NSTableView! private var objects = [MyObject]() func numberOfRows(in tableView: NSTableView) -> Int { return objects.count } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { // make view for row } private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.map({ objects[$0] }) ?? [] } func validateMenuItem(_ menuItem: NSMenuItem) -> Bool { let activeObjects = self.activeObjects ... } } extension NSTableView { var activeRowIndexes: IndexSet { return clickedRow == -1 || selectedRowIndexes.contains(clickedRow) ? selectedRowIndexes : IndexSet(integer: clickedRow) } } In one of the recent updates, I wanted to add some kind of header to the table view, so I decided to add a row at the beginning and offset the indexes by 1. func numberOfRows(in tableView: NSTableView) -> Int { return objects.count + 1 } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { if row == 0 { // make header view } else { // make view for row - 1 } } private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.subtracting(IndexSet(integer: 0)).map({ objects[$0 - 1] }) ?? [] } But since I added this change, Xcode regularly downloads crash reports from clients crashing during menu item validation in IndexSet.map with reason Code 5 Trace/BPT trap: 5. I assumed that I was accessing an invalid array index, so I added some debug code: the crash report would then show the invalid index beside the crashed thread's name. private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.subtracting(IndexSet(integer: 0)).map({ i in if !objects.indices.contains(i - 1) { Thread.current.name = (Thread.current.name ?? "") + ". Invalid index \(i - 1) for count \(objects.count)" preconditionFailure() } return objects[i - 1] }) ?? [] } But the crash reports for this new app version look just like the old ones and the thread name is not changed. Indeed, when recreating an invalid index access on my Mac, the crash report mentions Array._checkSubscript(_:wasNativeTypeChecked:), which does not appear in the crash reports downloaded by Xcode. Manually symbolicating the crash report also doesn't give any more information: all lines referring to my app code are resolved to either /<compiler-generated>:0 or MyViewController.swift:0. Apparently the problem is not an invalid array index, but something else. Does anybody have a clue what the problem could be? (Note: the crash report mentions Sequence.compactMap because now I'm effectively calling tableView?.activeRowIndexes.compactMap, but the same crash happened before when calling IndexSet.map, which would appear in the crash report as Collection.map.) crash2.crash
3
0
132
Jul ’25
Xcode has high CPU usage when apparently doing nothing for hours
In 2020 I created FB7719215, which I updated several times (including just now) and in 2021 I created FB9204092, but the issue is still there: when I keep Xcode open (currently version 16.3), my battery drains much quicker, even when it's apparently idle. For instance, today I barely did anything in Xcode, but still it has been at a constant 90% CPU for the last hours, and I keep checking the battery percentage to check how much time I have left. Does anyone at Apple has an explanation, workaround and/or fix?
2
0
163
Jun ’25
NSTableView.reloadData(forRowIndexes:columnIndexes:) causes wrong subview layout when usesAutomaticRowHeights = true
I have a table view where each row has two labels, one left-aligned and one right-aligned. I would like to reload a single row, but doing so causes the right-aligned label to hug the left-aligned label. Before the reload: After the reload: Reloading the whole table view instead, or disabling automatic row height, solves the issue. Can a single row be reloaded without resorting to these two workarounds? I created FB13534100 1.5 years ago but got no response. class ViewController: NSViewController, NSTableViewDataSource, NSTableViewDelegate { override func loadView() { let tableView = NSTableView() tableView.translatesAutoresizingMaskIntoConstraints = false tableView.dataSource = self tableView.delegate = self tableView.usesAutomaticRowHeights = true let column = NSTableColumn() column.width = 400 tableView.addTableColumn(column) let scrollView = NSScrollView(frame: CGRect(x: 0, y: 0, width: 500, height: 500)) scrollView.translatesAutoresizingMaskIntoConstraints = false scrollView.documentView = tableView view = scrollView Timer.scheduledTimer(withTimeInterval: 2, repeats: false) { _ in print("reload") tableView.reloadData(forRowIndexes: IndexSet(integer: 2), columnIndexes: IndexSet(integer: 0)) // tableView.reloadData() } } func numberOfRows(in tableView: NSTableView) -> Int { return 5 } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { let cell = NSTableCellView() let textField1 = NSTextField(labelWithString: "hello") textField1.translatesAutoresizingMaskIntoConstraints = false let textField2 = NSTextField(wrappingLabelWithString: "world") textField2.translatesAutoresizingMaskIntoConstraints = false textField2.alignment = .right let stack = NSStackView(views: [ textField1, textField2 ]) stack.translatesAutoresizingMaskIntoConstraints = false stack.distribution = .fill cell.addSubview(stack) NSLayoutConstraint.activate([stack.topAnchor.constraint(equalTo: cell.topAnchor, constant: 0), stack.leadingAnchor.constraint(equalTo: cell.leadingAnchor, constant: 0), stack.bottomAnchor.constraint(equalTo: cell.bottomAnchor, constant: 0), stack.trailingAnchor.constraint(equalTo: cell.trailingAnchor, constant: 0)]) return cell } }
Topic: UI Frameworks SubTopic: AppKit Tags:
2
0
151
Aug ’25
Tapping tab item in Xcode UI test doesn't always work
While preparing automated screenshots with Xcode UI tests for the iOS 26 release, I noticed that this simple line of code app.buttons["myTabItem"].tap() doesn't always work, both on iPhone and iPad Simulator. In fact, it rarely works. Even when repeating the same test, mostly it fails on that line, but a few times it works and I can see the tab item change in the simulator. My main view looks like this: TabView { MyTab1() .tag(tag1) .tabItem { Label("label1", systemImage: "image1") } MyTab2() .tag(tag2) .tabItem { Label("label2", systemImage: "image2") .accessibilityIdentifier("myTabItem") } The error I get is Failed to tap "myTabItem" Button: No matches found for Elements matching predicate '"myTabItem" IN identifiers' from input In the given list of buttons, I see the tab items with their labels, but none of them seem to have an identifier, while other buttons have the correct identifier. I wonder how this can only sometimes work. I tried isolating the issue by iteratively commenting out parts of the SwiftUI code, but unfortunately the behaviour seems erratic. When a change results in the issue not happening anymore, undoing the last X changes often causes the issue to stay away, even with configurations that previously had the issue. I've already spent almost a whole day trying to find the root cause, but with such apparently random behaviour it has proven impossible. Of course, I cannot reproduce the issue with a fresh test project, so the only way to reproduce it with 95% probability is running my original project. Has anyone had the same issue, or does anyone know how I could debug this to find out what causes my UI test to not be able to tap another tab item?
0
0
218
Sep ’25
Testing app intents and forcing Shortcuts app to refresh
The Verify the behavior of your intent in Simulator or on-device documentation says that it's sufficient to build and run the app and open the Shortcuts app to test an app intent, but that doesn't seem to be the case. I always needed to at least move the debugged app to the Applications folder before the app intents showed up in Shortcuts, and even then, most of the times I also need to wait a lot or restart the Mac. When updating an existing app intent (for instance by changing its title), building the app, overwriting the existing one in the Applications folder and restarting Shortcuts is not sufficient to make the new title appear in Shortcuts. Is there an efficient way to test app intents in the Shortcuts app? I already created FB15638502 one month ago but got no response.
Replies
0
Boosts
0
Views
418
Activity
Dec ’24
Camera zoom in to 3D point in SceneKit scene
I would like to implement zoom functionality in my SceneKit game: when the user performs the pinch gesture on a point on the screen, the scene zooms in to make that point larger. Until now I simply changed SCNCamera.focalLength, but this simply zooms in to the center of what is currently visible on screen. Is it somehow possible to implement the zoom functionality described above by perhaps interactively rotating the camera at the same time towards the pinched point? Is there a formula for this? I would like to avoid suddenly rotating the camera to face the pinched point when the pinch gesture begins and then zoom in while the pinch is in progress.
Replies
0
Boosts
0
Views
630
Activity
Dec ’24
Xcode downloads client crash report but the stacktraces don't contain any of my app's symbols (2)
All the threads only contain system calls. The crashed thread only contains a single call to my app's code which is main.swift:12. What could cause such a crash? crash.txt
Replies
2
Boosts
0
Views
463
Activity
Feb ’25
Notification when mounting volume on iOS
On macOS, we have didMountNotification but there doesn't seem to be an equivalent for iOS. Is there a way to be notified when a volume is mounted on iOS? I would like to use it in my iOS app I'm currently porting from macOS, which starts a synchronization from the volume (which has been previously selected in a NSOpenPanel) as soon as it's mounted.
Replies
3
Boosts
0
Views
442
Activity
Feb ’25
Resolving URL from bookmark data doesn't automatically mount SMB volume on iOS
On macOS, the Finder allows to connect to a server and store the login credentials. When creating a bookmark to a file on a server and resolving it again, the server is mounted automatically (unless I provide the option URL.BookmarkResolutionOptions.withoutMounting). I just tried connecting to my Mac from my iPad via SMB in the Files app and storing a bookmark to a file on the server, but disconnecting the server and trying to resolve the bookmark throws the error (I translated the English text from Italian): Error Domain=NSFileProviderErrorDomain Code=-2001 "No file provider was found with the identifier "com.apple.SMBClientProvider.FileProvider"'" UserInfo={NSLocalizedDescription=No file provider was found with the identifier "com.apple.SMBClientProvider.FileProvider"., NSUnderlyingError=0x302a1a340 {Error Domain=NSFileProviderErrorDomain Code=-2013 "(null) "}} Every time I disconnect and reconnect to the server, selecting the same file returns a different path. The first time I got /private/var/mobile/Library/LiveFiles/com.apple.filesystems.smbclientd/WtFD3Ausername/path/to/file.txt The next time WtFD3A changed to EqHc2g and so on. Is it not possible to automatically mount a server when resolving a bookmark on iOS? The following code allows to reproduce the issue: struct ContentView: View { @State private var isPresentingFilePicker = false @AppStorage("bookmarkData") private var bookmarkData: Data? @State private var url: URL? @State private var stale = false @State private var error: Error? var body: some View { VStack { Button("Open") { isPresentingFilePicker = true } if let url = url { Text(url.path) } else if bookmarkData != nil { Text("couldn't resolve bookmark data") } else { Text("no bookmark data") } if stale { Text("bookmark is stale") } if let error = error { Text("\(error)") .foregroundStyle(.red) } } .padding() .fileImporter(isPresented: $isPresentingFilePicker, allowedContentTypes: [.data]) { result in do { let url = try result.get() if url.startAccessingSecurityScopedResource() { bookmarkData = try url.bookmarkData() } } catch { self.error = error } } .onChange(of: bookmarkData, initial: true) { _, bookmarkData in if let bookmarkData = bookmarkData { do { url = try URL(resolvingBookmarkData: bookmarkData, bookmarkDataIsStale: &amp;stale) } catch { self.error = error } } } } }
Replies
2
Boosts
0
Views
522
Activity
Feb ’25
Crash (segmentation fault) on _swift_release_dealloc
This crash report for one of my apps was downloaded by Xcode. Apparently the app crashed while releasing an object of type Scan.File, which is a Swift class held in an array in the Scan.Directory class. I'm not doing any manual reference counting or low-level stuff with that object. What could cause such a crash? crash.crash
Replies
4
Boosts
0
Views
423
Activity
Feb ’25
Xcode downloads client crash report for NSOutlineView._updateDisclosureButtonForRowView with reason "Segmentation fault" but the stacktraces don't contain any of my app's symbols
All the threads only contain system calls. The crashed thread only contains a single call to my app's code which is main.swift:13. What could cause such a crash? crash.crash
Replies
2
Boosts
0
Views
379
Activity
Apr ’25
Crash when assigning NSImage to `@objc dynamic var` property
Xcode downloaded a crash report for my app which I don't quite understand. It seems the following line caused the crash: myEntity.image = newImage where myEntity is of type MyEntity: class MyEntity: NSObject, Identifiable { @objc dynamic var image: NSImage! ... } The code is called on the main thread. According to the crash report, thread 0 makes that assignment, and at the same time thread 16 is calling [NSImageView asynchronousPreparation:prepareResultUsingParameters:]. What could cause such a crash? Could I be doing something wrong or is this a bug in macOS? crash.crash
Replies
11
Boosts
0
Views
266
Activity
Aug ’25
NSView.knowsPageRange(_:) called twice when showing print panel
When creating a default macOS document-based Xcode project and using the code below (and wiring the File menu's Print item to printDocument: instead of the default print:, which does nothing), opening the print panel causes PrintView.knowsPageRange(_:) to be called twice. Is this a bug? My app populates PrintView dynamically, and for large documents it can be quite inefficient to populate it once, only for the contents to be immediately discarded and populated again. A workaround that came to my mind would be to check if the print options have changed, though I'm not sure if it's a reliable indicator that the print preview is effectively the same. I created FB17018494. class Document: NSDocument { override func makeWindowControllers() { addWindowController(NSStoryboard(name: NSStoryboard.Name("Main"), bundle: nil).instantiateController(withIdentifier: NSStoryboard.SceneIdentifier("Document Window Controller")) as! NSWindowController) } override func printOperation(withSettings printSettings: [NSPrintInfo.AttributeKey : Any]) throws -> NSPrintOperation { return NSPrintOperation(view: PrintView(frame: CGRect(x: 0, y: 0, width: 100, height: 100)), printInfo: NSPrintInfo(dictionary: printSettings)) } } class PrintView: NSView { override func knowsPageRange(_ range: NSRangePointer) -> Bool { print("knowsPageRange") range.pointee = NSRange(location: 1, length: 1) return true } }
Topic: UI Frameworks SubTopic: AppKit
Replies
1
Boosts
0
Views
87
Activity
Mar ’25
NSTableView.clickedRow sometimes is greater than number of rows
Xcode has been downloading many similar crash reports for my app for some time now, related to an index out of range runtime exception when accessing a Swift array. The crashes always happen in methods triggered by user input or during menu item validation when I try to access the data source array by using the following code to determine the indexes of the relevant table rows: let indexes = clickedRow == -1 || selectedRowIndexes.contains(clickedRow) ? selectedRowIndexes : IndexSet(integer: clickedRow) I was never able to reproduce the crash until today. When the app crashed in the Xcode debugger, I examined the variables clickedRow and selectedRowIndexes.first, which were 1 and 0 respectively. What's interesting: the table view only contained one row, so clickedRow was effectively invalid. I tried to reproduce the issue several times afterwards, but it never happened again. What could cause this issue? What are the circumstances where it is invalid? Do I always have to explicitly check if clickedRow is within the data source range?
Topic: UI Frameworks SubTopic: AppKit Tags:
Replies
4
Boosts
0
Views
120
Activity
Apr ’25
Crash in IndexSet.map during menu item validation in client report downloaded by Xcode
For many years I've had the following code to access the active objects of a table view in my App Store app: class MyViewController: NSViewController: NSMenuItemValidation { private var tableView: NSTableView! private var objects = [MyObject]() func numberOfRows(in tableView: NSTableView) -> Int { return objects.count } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { // make view for row } private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.map({ objects[$0] }) ?? [] } func validateMenuItem(_ menuItem: NSMenuItem) -> Bool { let activeObjects = self.activeObjects ... } } extension NSTableView { var activeRowIndexes: IndexSet { return clickedRow == -1 || selectedRowIndexes.contains(clickedRow) ? selectedRowIndexes : IndexSet(integer: clickedRow) } } In one of the recent updates, I wanted to add some kind of header to the table view, so I decided to add a row at the beginning and offset the indexes by 1. func numberOfRows(in tableView: NSTableView) -> Int { return objects.count + 1 } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { if row == 0 { // make header view } else { // make view for row - 1 } } private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.subtracting(IndexSet(integer: 0)).map({ objects[$0 - 1] }) ?? [] } But since I added this change, Xcode regularly downloads crash reports from clients crashing during menu item validation in IndexSet.map with reason Code 5 Trace/BPT trap: 5. I assumed that I was accessing an invalid array index, so I added some debug code: the crash report would then show the invalid index beside the crashed thread's name. private var activeObjects: [MyObject] { return tableView?.activeRowIndexes.subtracting(IndexSet(integer: 0)).map({ i in if !objects.indices.contains(i - 1) { Thread.current.name = (Thread.current.name ?? "") + ". Invalid index \(i - 1) for count \(objects.count)" preconditionFailure() } return objects[i - 1] }) ?? [] } But the crash reports for this new app version look just like the old ones and the thread name is not changed. Indeed, when recreating an invalid index access on my Mac, the crash report mentions Array._checkSubscript(_:wasNativeTypeChecked:), which does not appear in the crash reports downloaded by Xcode. Manually symbolicating the crash report also doesn't give any more information: all lines referring to my app code are resolved to either /<compiler-generated>:0 or MyViewController.swift:0. Apparently the problem is not an invalid array index, but something else. Does anybody have a clue what the problem could be? (Note: the crash report mentions Sequence.compactMap because now I'm effectively calling tableView?.activeRowIndexes.compactMap, but the same crash happened before when calling IndexSet.map, which would appear in the crash report as Collection.map.) crash2.crash
Replies
3
Boosts
0
Views
132
Activity
Jul ’25
Xcode has high CPU usage when apparently doing nothing for hours
In 2020 I created FB7719215, which I updated several times (including just now) and in 2021 I created FB9204092, but the issue is still there: when I keep Xcode open (currently version 16.3), my battery drains much quicker, even when it's apparently idle. For instance, today I barely did anything in Xcode, but still it has been at a constant 90% CPU for the last hours, and I keep checking the battery percentage to check how much time I have left. Does anyone at Apple has an explanation, workaround and/or fix?
Replies
2
Boosts
0
Views
163
Activity
Jun ’25
Updating macOS Tahoe 26 to beta 5 on separate partition boots into main partition
I successfully installed macOS Tahoe 26 in June on a separate external partition. Now I want to update to beta 5, but when the Mac restarts to install the update, it shows the progress indicator for a minute or two, then boots again into the main partition which has macOS 15.6. Am I required to install it from scratch?
Replies
0
Boosts
0
Views
174
Activity
Aug ’25
NSTableView.reloadData(forRowIndexes:columnIndexes:) causes wrong subview layout when usesAutomaticRowHeights = true
I have a table view where each row has two labels, one left-aligned and one right-aligned. I would like to reload a single row, but doing so causes the right-aligned label to hug the left-aligned label. Before the reload: After the reload: Reloading the whole table view instead, or disabling automatic row height, solves the issue. Can a single row be reloaded without resorting to these two workarounds? I created FB13534100 1.5 years ago but got no response. class ViewController: NSViewController, NSTableViewDataSource, NSTableViewDelegate { override func loadView() { let tableView = NSTableView() tableView.translatesAutoresizingMaskIntoConstraints = false tableView.dataSource = self tableView.delegate = self tableView.usesAutomaticRowHeights = true let column = NSTableColumn() column.width = 400 tableView.addTableColumn(column) let scrollView = NSScrollView(frame: CGRect(x: 0, y: 0, width: 500, height: 500)) scrollView.translatesAutoresizingMaskIntoConstraints = false scrollView.documentView = tableView view = scrollView Timer.scheduledTimer(withTimeInterval: 2, repeats: false) { _ in print("reload") tableView.reloadData(forRowIndexes: IndexSet(integer: 2), columnIndexes: IndexSet(integer: 0)) // tableView.reloadData() } } func numberOfRows(in tableView: NSTableView) -> Int { return 5 } func tableView(_ tableView: NSTableView, viewFor tableColumn: NSTableColumn?, row: Int) -> NSView? { let cell = NSTableCellView() let textField1 = NSTextField(labelWithString: "hello") textField1.translatesAutoresizingMaskIntoConstraints = false let textField2 = NSTextField(wrappingLabelWithString: "world") textField2.translatesAutoresizingMaskIntoConstraints = false textField2.alignment = .right let stack = NSStackView(views: [ textField1, textField2 ]) stack.translatesAutoresizingMaskIntoConstraints = false stack.distribution = .fill cell.addSubview(stack) NSLayoutConstraint.activate([stack.topAnchor.constraint(equalTo: cell.topAnchor, constant: 0), stack.leadingAnchor.constraint(equalTo: cell.leadingAnchor, constant: 0), stack.bottomAnchor.constraint(equalTo: cell.bottomAnchor, constant: 0), stack.trailingAnchor.constraint(equalTo: cell.trailingAnchor, constant: 0)]) return cell } }
Topic: UI Frameworks SubTopic: AppKit Tags:
Replies
2
Boosts
0
Views
151
Activity
Aug ’25
Tapping tab item in Xcode UI test doesn't always work
While preparing automated screenshots with Xcode UI tests for the iOS 26 release, I noticed that this simple line of code app.buttons["myTabItem"].tap() doesn't always work, both on iPhone and iPad Simulator. In fact, it rarely works. Even when repeating the same test, mostly it fails on that line, but a few times it works and I can see the tab item change in the simulator. My main view looks like this: TabView { MyTab1() .tag(tag1) .tabItem { Label("label1", systemImage: "image1") } MyTab2() .tag(tag2) .tabItem { Label("label2", systemImage: "image2") .accessibilityIdentifier("myTabItem") } The error I get is Failed to tap "myTabItem" Button: No matches found for Elements matching predicate '"myTabItem" IN identifiers' from input In the given list of buttons, I see the tab items with their labels, but none of them seem to have an identifier, while other buttons have the correct identifier. I wonder how this can only sometimes work. I tried isolating the issue by iteratively commenting out parts of the SwiftUI code, but unfortunately the behaviour seems erratic. When a change results in the issue not happening anymore, undoing the last X changes often causes the issue to stay away, even with configurations that previously had the issue. I've already spent almost a whole day trying to find the root cause, but with such apparently random behaviour it has proven impossible. Of course, I cannot reproduce the issue with a fresh test project, so the only way to reproduce it with 95% probability is running my original project. Has anyone had the same issue, or does anyone know how I could debug this to find out what causes my UI test to not be able to tap another tab item?
Replies
0
Boosts
0
Views
218
Activity
Sep ’25