How do I Represent Success Status Codes in Linux - c

Question

Windows has different types of status codes, HRESULT and NTSTATUS for instance, that allow for representing both success and failure values. This allows for macros like FAILED and NT_SUCCESS. One example of a success status code is STATUS_PENDING that indicates an IO is not yet complete.
I'm trying to map similar style status codes in Linux. I have a (WIP) cross platform library that uses the above type of status codes on Windows (user mode and kernel mode). I have my own "FAILED" macro that wraps the above macros. I need an equivalent for Linux for it all.
I've looked around and haven't found anything interesting/helpful. For any system/platform function (sockets, thread, locks) that might return and error, I need to make sure it is converted to the appropriate status code type. Windows has lots of helpers for things like this, but I'm not sure about Linux.
Psuedocode example:
MY_STATUS SendAsync(...) {
ASYNC_OPERATION* Operation = MY_ALLOC(...)
if (Operation == NULL) {
return MY_OUT_OF_MEMORY_ERROR; // Error status
}
// Build the operation
MY_STATUS Status = QueueOperation(Operation);
if (MY_FAILED(Status)) {
return Status;
}
return MY_STATUS_PENDING; // Success status
}
The caller should then be able to have the following code:
MY_STATUS Status = SendAsync(...)
if (MY_FAILED(Status)) {
// Bail
}
// Handle success
I think the crux of the problem is that I don't know how to stuff status codes from system/platform function calls into a generic type that supports success/fail. Generally, it looks like all the status codes are positive, so I could just say my success error codes are <= 0. But I'm not sure that's a great solution.
To be clear, since it was proposed that this question might the answer, I'm not trying to force the Linux status/error codes into HRESULT format explicitly. I'm just trying to find some way to represent error codes as either success or failure. If I have to convert system error codes (negate them?) along the way, that's acceptable. If there is no real solution to this problem, then that would be an acceptable answer (though not desired), and I'll have to find some work around.

Answer 1

For error codes in linux there are the constants defined in errno.h header file. The error constants are available from man page man 3 errno. Also there is: What errno means
As for exit status codes, there is no standard set of codes. The value 0 has always denoting success, and non-zero value has denoted an error state. Bash, for example has this Bash exit status information.
All of the Bash builtins return an exit status of zero if they succeed
and a non-zero status on failure, so they may be used by the
conditional and list constructs. All builtins return an exit status of
2 to indicate incorrect usage, generally invalid options or missing
arguments.
There is sysexits which was an attempt to define a set of status codes, but it has been largely ignored in the field of linux software.
For more info: Exit Status codes in Linux

Answer 2

To get the error return code from a function in Linux, use the function's return value (usually nonzero on error) and the value of the errno variable, from errno.h.
Search for "linux errno codes" or run man errno on Linux to get more information.
errno (and return statuses) are usually represented as an int, so that's what I'd go with if you're looking for a type to use.
For example, checking to see if the return code of non-blocking connect is negative one and errno is EINPROGRESS is the same as checking for STATUS_PENDING in Windows.
Here's some pseudocode to respond to your pseudocode example:
#ifdef _WIN32
typedef /* Windows stuff */ MY_STATUS;
# define MY_FAILED(x) /* Windows stuff */
#else /* POSIX, other reasonable operating systems, etc. */
typedef int MY_STATUS;
# define MY_FAILED(x) ((x) == -1)
#endif
Of course, this assumes that your functions will return int or something similar, which may not be the case. But you get the idea.