Post

Replies

Boosts

Views

Activity

Reply to Xcode 26 + iOS 26 debugging is extremely slow
I refuse to believe they didn’t test it before, so I can only assume they don’t care. It’s crazy, but it looks like every other Xcode version ships with a broken debugger. In some versions, it works better, but then the next one arrives with a broken debugger. Beside Wi-Fi issues, the debugger sometimes gets stuck, fails to debug async code, and so on. This forum is full of debugger bugs, it makes me wonder which debugger they use to debug the internal apps.
3w
Reply to NSFileVersion.currentVersionOfItem not consistent across devices after simultaneous edit
On device B otherVersionsOfItem(at:) returns [fileVerB], no sign of fileVerA at this point. It seems that fileVerA is not yet synchronized to device B. However, I expect that on device B, the system should not call NSFilePresenter’s presentedItemDidGain(_:) before the file from device A has been downloaded, or at least before the fileVerA URL has been created on disk (so that I can then read its contents using NSFileCoordinator). In presentedItemDidGain I check if the URL passed in is in a conflict state by using 'newVersionURL.isConflict', and it returns true, and then I run the merge. Is this the right way to check for a pending conflict? [quote='862396022, DTS Engineer, /thread/804253?answerId=862396022#862396022'] Also, after the newer version from the Device A arrives on Device B, does currentVersionOfItem still return fileVerB on device B? [/quote] I no longer check for currentVersionOfItem because the system calls presentedItemDidChange (a split second after presentedItemDidGain), and in this method, I assume there’s no conflict, so I don’t run the merge algorithm. Instead, I simply read the file contents, which happen to be fileVerA, and redisplay it. I noticed that in case of conflict, the system calls presentedItemDidChange then shortly after presentedItemDidGain. So in presentedItemDidChange I always check for conflict using unresolvedConflictVersionsOfItem(at:), and bail out if positive, because I expect to receive a presentedItemDidGain next. When I receive it, I run the merge then write the resulting file on disk. The system then calls again presentedItemDidChange, I check again for conflict, this time is negative, and I proceed to display the merged file. I tried to reproduce this, and call 'currentVersionOfItem', but I can’t trigger the case right now, it only happens occasionally.
Oct ’25
Reply to Popovers are broken on macCatalyst
Apple expect us to create bland list apps only; anything beyond that breaks down. SwiftUI is broken from a design point of view, use the old trusty UIKit, designed before Tim, and call it a day. As you can see, Apple doesn't care, no one has answered you for a week.
Topic: UI Frameworks SubTopic: SwiftUI Tags:
Sep ’25
Reply to lldb issue: 'self cannot be reconstructed'
They don't seem to care, I found this post from 2022 https://developer.apple.com/forums/thread/720519, so this bug is 3 years old now 😕 And then this one, where an Apple staff even asks for sample projects to reproduce the bug, they get at least two of them, and then they fell silent, never responding to additional questions: https://developer.apple.com/forums/thread/771788. Maybe you can ping this guy on his latest comment,I don't know what else to do: https://developer.apple.com/forums/thread/771788?answerId=826459022#826459022
May ’25
Reply to error: type for self cannot be reconstructed: type for typename
They don't seem to care, I found this post from 2022 https://developer.apple.com/forums/thread/720519, so this bug is 3 years old now 😕 And then this one, where an Apple staff even asks for sample projects to reproduce the bug, they get at least two of them, and then they fell silent, never responding to additional questions: https://developer.apple.com/forums/thread/771788. Maybe you can ping this guy on his latest comment,I don't know what else to do: https://developer.apple.com/forums/thread/771788?answerId=826459022#826459022
May ’25