go-sumtype VS exhaustive

I find this implementation to be quite minimal and less clumsy than alternatives. Sure, you don't get nice exhaustive pattern matching. Also, type inference gets in the way when instantiating UserKey (though you can wrap it in constructor functions). But expressing your intent using types still makes your code much more convenient and easier to understand.

The call out to sum types is something I feel. I've been using Rust daily for almost 10 years now, and sum types are absolutely still one of the things I love most about it. It's easily one of the things I miss the most in other languages. I'm usually a proponent of "using languages as they're intended," but I missed exhaustiveness checking so much that I ported a version of it to Go[1] as a sort of lint.

[1]: https://github.com/BurntSushi/go-sumtype

A Google search for golang sum types currently shows my project as a second hit: https://github.com/BurntSushi/go-sumtype

I've been writing Go and Rust nearly daily for about a decade now (Go is more than a decade, Rust is about 8 years). You are not going to teach me anything about the pros and cons of either language in a reddit comment. I do not need to be taught about the "iota mess" when I've written tooling for exhaustiveness checking in Go.

https://github.com/BurntSushi/go-sumtype forces exhaustive type switches for interfaces specifically annotated to need that.

We can fix these two problems by relying on static analyzers such as go-sumtypes

See also https://github.com/BurntSushi/go-sumtype

go-sumtype[0] has completeness checking for sealed interfaces.

[0] https://github.com/BurntSushi/go-sumtype

Pretty easy to achieve outside of the compiler: https://github.com/BurntSushi/go-sumtype

This is an analyzer that will catch this: https://github.com/nishanths/exhaustive

I believe it's in golangci-lint.

I agree linters in general are quite useful for Go though. The default suite from golangci-lint is quite good. I would also recommend enabling exhaustive if you're working with a codebase that uses "enums" (full disclosure, I contributed a bit to that project).

there’s a linter for exhaustive matching: https://github.com/nishanths/exhaustive

I tried to find that linter and found this: exhaustive

And in Go you'd use a linter, like this one.

this is AST go vet analyzer that performs just that: https://github.com/nishanths/exhaustive (too bad it can not do struct based enums..)

>> the main thing missing from Go is ADT's. After using these in Rust and Swift, a programming language doesn't really feel complete without them

What are the differences between an ADT (plus pattern matching i’d reckon?) in Rust/Swift vs the equiv in Go (tagged interfaces + switch statement)?

One has exhaustive matching at compile time, the other has a default clause (non exhaustive matching), although there’s an important nub here with respect to developer experience; it would be idiomatic in Go to use static analysis tooling (e.g. Rob Pike is on record saying that various checks - inc this one - don’t belong in the compiler and should live in go vet). I’ve been playing with Go in a side project and using golint-ci which invokes https://github.com/nishanths/exhaustive - net result, in both go and rust, i get a red line of text annotated at the switch in vscode if i miss a case.

Taking a step back, there isn’t a problem you can solve with one that you can’t solve with the other, or is there?

To take a step further back, why incomplete?

Use a linter.

For an exhaustive linter, were you referring to this? It looks pretty nice. If it's possible to check this with static analysis, is it something that could be in the compiler itself in the future?

https://github.com/nishanths/exhaustive

here, have fun. You’re gonna write some tests, make new types to satisfy interfaces for testing, and then wind up with branches for your test paths in your live code, but go for it, I guess. You know everything! I am but a simple blubbite, too dim, too dim to get it.