Funny, C has sum types (coproduct is a sum type) and it still happened.

The use of sum types may have helped, especially when combined with a tag which is trivial to do in C, but a language having sum types does not imply that the authors will use them. No language can save a reckless developer.

Memory is where C is broken. I feel like without `defer` or destructors the language meaningfully is incomplete.

the fault for sshhole isn't just "C bad", it's a series of poor choices in workarounds falling together to create a vulnerability.

being condescending about decisions made 50 years ago, before any of the better languages you're mad the developer didn't use even existed, is silly.

if you want to lay blame, lay it at the feet of more recent devs that started returning values instead of NULL because people were using crypt incorrectly. likely sha512 returned a fake value instead of a real one because prior hashes couldn't fail. the interface should have been updated then, but instead they worked around it.

I can think of two common patterns that I've seen frequently:

* Return NULL for errors (sometimes with a global errorno() function).

* Return an integer error with any return values as out parameters.

So if it is possible to improve, why doesn't anyone do it?

> * Return an integer error with any return values as out parameters.

You must first standarize all that (usually negative) integers, make docs, check for it, do not make mistakes in your if's or case's...

And errno is just that - integers with a meaning. errno is even thread local now and quite safe to use but people on internet scream loudly when someone try to use that not app global varible :) I guess their llms statistically have less new texts feeded in ;)

At some point it is just better to return pointer to struct with all info you can imagine.

And how you think other languages implemented their Error type returns ? Just a struct or union with tag. Because current CPU's operate on simple type registers there is no other sane way to do it.

Because in fact all the changes were about changing the function signatures. The function signature is enforced by the linker. The semantic of strings that go through isn't.

Returning 2 values is possible in C (via pointers input parameters). The main issue here is that libc is a strong API contract based on C function signatures, that nobody can change, only augment with new symbols. Such strong API compatibility constraints happen in any programming language, and workaround to fit new behaviours in old APIs happens everywhere.

But also, its not like there were many alternative languages when the POSIX standards were designed so it's fair that it works like this.

But there has to be something in the payload field when it is not applicable. A null is perfectly fine.

The argument for changing that to "", if I understand the above linked discussion correctly, was that some users of this api threw away errors and a dummy value would be harder to ignore. You could argue that these behaviours are less common in Rust or Go, but that's not a technical difference.

The world is not perfect and sometimes you really need to work around bad users. Even in Rust or Go. I am not convinced this was such a case. If you are convinced it was it would still have been better to select a less trivial dummy value (such as ":ERROR:*").

Not in languages with ergonomic sum types. A `Result` or `Either` will have no payload field on the error type. The compiler won't let you dereference it.

    Maybe
    ——————
    Just hash
    Nothing

    For Either
    ————————————
    Left err
    Right hash