Wooster

Ah. On POSIX AFAIU the exit code is pretty much defined as `u8` right? Or at least an integer. So in that case I'd have to put this somewhere else....

This is blocked on #16135 for now.

> null literal should behave as any other zero-sized type I think that means this should all pass too, just like if it was `u0` or `void`: ```zig test {...

when navigating through search results beyond the page with arrows keys the page should be scrolled down or up before the selected result goes out of view

I started working on this a bit here: https://github.com/wooster0/zig/commits/panic/ There is only one commit right now: https://github.com/wooster0/zig/commit/279d93ab5d846af6d1051aa2ace84413c4b7a8bc It's not finished yet. I have some questions: 1. Should `std.debug.panic` be changed...

Here is a complete test case: ```zig export fn a() void { const Foo = enum { a, }; const arr = [_]Foo{.a}; _ = arr ++ .{.b}; } export...

Good to have a second opinion on that. Wasn't sure about it either. Negative integers always return false. What if we make it `assert(value > 0)`? Otherwise, this is good...

https://github.com/search?q=toTargetCpuArch+language%3AZig&type=code&l=Zig I can't see any real-world usage of toTargetCpuArch (except in zld but you said you don't use that anymore so I assume that's just outdated). Do you think we...

Oh so this thing is causing the problem here: ```zig test { std.testing.refAllDecls(Target.Cpu.Arch); } ``` Fixed that now hopefully.

Sorry are you suggesting that the current `std.Target.Cpu.Arch.toElfMachine` and `std.Target.Cpu.Arch.toCoffMachine` were good and instead I could just remove `std.coff.MachineType.fromTargetCpuArch` and merge the logic into `std.Target.Cpu.Arch.toCoffMachine`? I think that'd be good...