Nathan Faubion

In general, I think that's a really good proposal. A few questions: > Preceding a module declaration, on its own line Is there an issue with scoping? For example, can...

> type Deprecated = String -> Annotation type NoInline = Annotation type NoSpecialize = Annotation type IgnoreWarning = String -> Annotation > In more sophisticated cases, polymorphism and type classes...

I missed that the arguments are expr4. How does this interact with scoping (runtime arguments, type variables, given constraints)? I'm personally wary of using runtime syntax for this purpose. Would...

If the annotation takes constraints, how does that elaborate in the CoreFn annotation syntax? Are constraints omitted or are they elaborated to annotation arguments? Is it useful to have them...

What do you expect this to evaluate to?

Oh sorry, you mean when both are known strings. Yes, I think we could handle that.

What's going on with the stack overflow in your output?

I didn't notice the `--tokens` flag. As long as a full parse works OK, then I'm going to assume that the `--tokens` bench is naive in some way.

https://github.com/natefaubion/purescript-language-cst-parser/blob/67323ef7038ee6514c1912684398e3da5a3fd207/bench/ParseFile.purs#L63-L69 Yeah, because of the `step >>> ...` part, the main compiler no longer sees this as tail recursive, so it doesn't generate a loop and it stack overflows.

See https://github.com/purescript/registry-dev/issues/662 I think we are waiting on a new ecosystem import to get correct dependencies.