The existing code in the shim layer returns 0 or 1 as a proxy for a boolean. I wonder if there's opportunity to change those to bool. It seems based on this PR that bool is effectively the FFI type for itself, which is why there's nothing in ffi for it. I'm not 100% sure, but I'm having a hard time imagining a situation where this could cause any problems. Although, on second glance the 0 and 1 values are (at least sometimes) reversed - where 0 means success and 1 means failure. So we'd have to be careful to get the logic correct.
The existing code in the shim layer returns 0 or 1 as a proxy for a boolean. I wonder if there's opportunity to change those to bool. It seems based on this PR that bool is effectively the FFI type for itself, which is why there's nothing in ffi for it. I'm not 100% sure, but I'm having a hard time imagining a situation where this could cause any problems. Although, on second glance the 0 and 1 values are (at least sometimes) reversed - where 0 means success and 1 means failure. So we'd have to be careful to get the logic correct.