Compiler in infinite loop about lux HOT 2 CLOSED

My bad, I found my errors!

from lux.

Hi, @Arjuna144

I am still using luxc3.1.jar since I am getting the following error with luxc.3.2.jar and the following

There were some important changes to Lux as a language in v0.3.2 that cause the v0.3.1 stdlib to not compile (which was probably the error you got, since I see the _lux_def form you showed there lacks meta-data, which was a v0.3.2 addition).

I'm waiting until I'm done with v0.3.3 to document those changes, since some stuff requires a bit more polishing (I fixed a bug with tuples and I'm working on fixing one with variants, both of which were introduced in a major change during v0.3.2).

I've also abstained from talking about JVM interop because I'm still adding a bit more features on that front, plus polishing what's already there, since it's usable, but not terribly comfortable as of v0.3.2 (importing them is great, but defining new classes is still somewhat of a hassle).

Anyway... regarding the infinite loop!
You didn't post the error, so I can only speculate, but I'm aware of a stack-overflow error that sometimes happens in the type-checking phase that gets solved by just adding a bit more type annotations.
I imagine you got that one and fixed it in that way.

By the way, when you define this:

(def (echo)
  (IO (,))
  (>>= IO/Monad read-line tellWhat))

You don't need to put echo in parentheses. You can, if you want, but it's not mandatory.

The parentheses syntax on def is just there as a convenience to save people from having to add an extra lambda form.

Consider this:

(def (>>= m ma f)
  (All [m a b]
    (-> (Monad m) (m a) (-> a (m b)) (m b)))
  (using m
    (join (map f ma))))

Versus this:

(def >>=
  (All [m a b]
    (-> (Monad m) (m a) (-> a (m b)) (m b)))
  (lambda [m ma f]
    (using m
      (join (map f ma)))))

Oh... and remember that to run a program you need a (program args ...) form, which is basically your "main" method.

from lux.