generics | concept implementation of my generics proposal

I'm not sure that the critical problem is orphan instances and modules.

I think the main issue is that it's unexpectedly difficult to maintain coherence of instance definitions and efficiency of compilation in the face of constraint disjunctions. The problem is that the hypothetical eitherConstraint in your imagined syntax can't be simply, locally resolved at compile time. For example, consider the functions:

The difference is the category. Nat is an initial algebra in the category of types. Rep is an initial algebra in the category of indexed types. The category of indexed types has as objects type constructors of kind * -> *, and as morphisms from f ~> g, functions of type forall t. f t -> g t.

Then Rep is the initial algebra for the functor RepF defined as follows:

The issue is that your isAny yields boolean, which means both of these assignments will pass:

because you want

t = append(t, 0)

the data type can be int or float group.

this code should work

Edit: see blackgreen's answer, which I also found later on my own while scanning through the same documentation they linked. I was going to edit this answer to update based on that, but now I don't have to. :-)

There is probably a better way—this one seems a bit clumsy—but I was able to work around this with reflect:

I think the old interface{} is enough to do this.

Like this:

This seems to be a bug in TypeScript; see microsoft/TypeScript#27709 for details. Nested generic indexed access types seem to "forget" their constraint and won't let you index into them with a specific key anymore. The bug has been open for a long time with no sign that it will be addressed soon, so you might want to work around it.

In cases where you have a type T where you know T extends U is true but the compiler doesn't, you can often replace T with Extract using the Extract utility type. The compiler will accept that Extract extends U, and assuming you're correct about T extends U, then eventually (when the generics are specified), Extract will evaluate to just T. (Note that this is an "off-label" use of Extract which is used primarily to filter unions in T to just those members which are assignable to U.)

In order to index into some type T with the key type "c", it means the type should be assignable to {c?: any}. If T["c"] fails to compile, then Extract should succeed. Let's try it here:

I've passed your code through gotip that uses a more evolved implementation of the proposal and it does not complain about that part of the code, so I would assume that the problem is with the go2go initial implementation.

Please note that your implementation will not work since you can definitely use parametric interfaces in type assertion expressions, but you can't use interfaces with type lists as you are doing in testDifference[intOrString]

As mentioned in the comments of the question, the described behavior is a bug in Go 1.18beta1 and is being tracked by issue 50419.

I have confirmed that the bug in question is fixed in 1.18beta2, which was released on 31 January, 2022.