I have an app that needs accessibility trust to install an active global event tap. The only way I found to reliably detect whether I have that trust is to just try calling CGEventTapCreate and see whether it returns NULL or not.

Go to the download area of developer.apple.com and click the More link.

If you select the field and then look at the size inspector, what does it say next to Layout, "Autoresizing Mask" or "Inferred (Autoresizing Mask)"? In my experience, if I set all my items to "Autoresizing Mask", then Xcode won't bug me about constraints.

By the way, experimentally, Finder won't let me set a file name with more than 255 (ascii) characters on an APFS volume. But maybe that's just Finder, not the file system.

One of the references of the Wikipedia article is the Apple File System Reference . It’s pretty technical, but as far as I can tell, the name of an APFS volume is stored in 256 bytes, but the name of a file is a null terminated string, and elsewhere the name of the string is stored as a 10 bit unsigned number.

Whoops, never mind! I needed to set com.apple.security.inherit to true in the entitlements. Now I see more documentation: Embedding a command-line tool in a sandboxed app

Go to the downloads area at developer.apple.com, click the More link, and you'll find it in the list. I can't promise that it will run on a recent version of macOS.

The frontmostApplication property of the NSWorkspace class can be observed with key-value observing (KVO).

I tried your project and it does reproduce. I notice that I don't see the glitch if I increase the step parameters to CGPatternCreate, putting gaps between the cells. You should send a bug report with Feedback Assistant. This forum is not an official channel for reporting bugs. By the way, you can simplify your code sample a bit. Instead of passing the size as a dictionary, pass &_size in the CGPatternCreate call, and then say CGFloat size = *(CGFloat*) info; in the callback.

You do need to notarize your app. Otherwise users will not be able to start it the first time.

Thanks, Quinn!

I don't think the second snippet would be fixed by writing NSString *ns = (__bridge_transfer NSString *)CFBridgingRelease(string); because as far as I know, __bridge_transfer is basically the same as CFBridgingRelease. The problem with the second snippet is that when the local variable ns goes out of scope, the string will be released, but the local variable string is not set to nil, hence CFRelease will be passed a bad pointer.

For notarizing, you should use a "Developer ID Application" certificate.

You should send that request to Apple using Feedback Assistant, a more official channel for such things.

I've never seen the problem you describe, but if the Issue Navigator panel isn't working, you could look in the Report Navigator. But also use Feedback Assistant to file a bug report.