Comment between type annotation and declaration results in compiler seeing 2 declarations

[erwinkn]

Any comment added between a type annotation and the associated declaration results in the compiler thinking the two are separate declarations.

Reproduction In any of the "Hello world" examples, write:

main: Str
# comment
main = "I'm still the same declaration!"

Expected behavior The program runs fine and prints I'm still the same declaration!

Actual behavior I get the following error:

emitted runtime error NoImplementationNamed { def_symbol: `8.IdentId(0)` }
🔨 Rebuilding host... 
── DUPLICATE NAME ──────────────────────────────────────────────────────────────

The main name is first defined here:

6│  main: Str
    ^^^^

But then it's defined a second time here:

8│  main = "Hello!" 
    ^^^^

Since these variables have the same name, it's easy to use the wrong
one on accident. Give one of them a new name.

────────────────────────────────────────────────────────────────────────────────

Done!
Roc hit a panic: NoImplementationNamed { def_symbol: `8.IdentId(0)` }

[rtfeldman]

The compiler is interpreting this as intended, but we could definitely give a helpful tip in this situation!

[robbym]

I get the same runtime error if the definition is omitted altogether. I was surprised when it compiled. Usually that would give me a link-time error in C for example.

Example:

app "helloweb"
    packages { pf: "https://github.com/roc-lang/basic-webserver/releases/download/0.3.0/gJOTXTeR3CD4zCbRqK7olo4edxQvW5u3xGL-8SSxDcY.tar.br" }
    imports [
        pf.Task.{ Task },
        pf.Http.{ Request, Response },
    ]
    provides [main] to pf

main : Request -> Task Response []