@Developer Tools Engineer Thank you for providing useful suggestions! I found that the occurrence of NOT_FOUND errors in October is 10 times that of the past. And the logs collected on user devices also show a relation to CKError.Code.unknownItem. I also observed that CKFetchSubscriptionsOperation would return an internal CKPrettyError (not a CKError), which should also be a problem. I will submit a FB after further research.

👍

@NSCruiser No update :(

A user just reported exact same error when trying refresh the receipt.

Thanks!

Thanks for the response. Here are the steps to generate the "helpful diagnostics": (1) Clone my demo project at https://github.com/gongzhang/rkassets-preview-issue-demo (2) Open preview window for ContentView.swift (failed) (3) Click the "Diagnostics" button in the error banner

typo: I can => I can't

Thanks for sharing the information!

FB12900696

Thanks for the reply! After careful consideration, I think this modification is in line with the convention of os() conditional compilation, even if it will bring a lot of code changes. Because #if os(iOS)/os(macOS)/os(watchOS)/os(tvOS)/os(visionOS) should be mutually exclusive.

@aviwad Thanks for the update! Confirmed contentMarginsDisabled() is back deployed to iOS 15/macOS 12/watchOS 9.0.

and the warnings appeared in watchOS target.

-ld64 saves me a day. And thanks for reporting a FB! I hope Apple can fix this issue in the next beta...

Cool. The canvas setting is really helpful. Thank you very much!

I sincerely believe that the best way to solve a problem is to provide a step-by-step guide or sample code that can 100% reproduce the issue, rather than diagnostic data. Although I encountered a different issue, I noticed that every time it was related to Preview, the request was always for diagnostic data feedback. I believe this is not very efficient.