Java Enums cannot be used in case branches (but can in if statements)

[radeusgd]

Currently, if I try to pattern match a Java enum:

package foo;

import org.graalvm.polyglot.Value;

public enum Foo {
    A, B, C;

    public static Foo generate(int i) {
        if (i == 0) return A;
        return B;
    }
}

from Standard.Base import all

polyglot java import foo.Foo

convert enum_value =
    case enum_value of
        Foo.A -> "aaa"
        Foo.B -> "bbb"
        Foo.C -> "ccc"

main =
    IO.println (convert Foo.C)
    IO.println (convert (Foo.generate 0))
    IO.println (convert (Foo.generate 1))

I get

C:\NBO\repr\PolyProject\src\Main.enso: error: The name `Foo.A` could not be found.
Aborting due to 1 errors and 0 warnings.
Execution finished with an error: Compilation aborted due to errors.

I can convert the Enso code to:

from Standard.Base import all
import Standard.Base.Errors.Common.No_Such_Method

polyglot java import foo.Foo

convert enum_value =
    if enum_value == Foo.A then "aaa" else
        if enum_value == Foo.B then "bbb" else
            if enum_value == Foo.C then "ccc" else 
                Nothing

main =
    IO.println (convert Foo.C)
    IO.println (convert (Foo.generate 0))
    IO.println (convert (Foo.generate 1))

and then it works:

ccc
aaa
bbb

Nonetheless, it would be nice if the Java enums were also supported in regular pattern matching. It may be a bit surprising to users that if works but case does not.

Of course this is lower priority, because most of our users won't be using Java polyglot as much and even less will care about enums. But it may be worth to improve this at some point.

[radeusgd]

Another small issue related to this:

As you can see the error message C:\NBO\repr\PolyProject\src\Main.enso: error: The name Foo.A could not be found. - it does not have any indication of the position of the error in the file. In my small example it is easy to find it, but in a big library file it may be problematic when it is unclear where the error is located.

[enso-bot]

Jaroslav Tulach reports a new STANDUP for yesterday (2023-08-21):

Progress: - JavaEnums merged: b9702d1

• EpbLanguage rewrite merged: Eliminate compile time dependency on runtime-language-epb #7611
• PR for python support: Special support for Python Date/Time/Zone interop #7617
• GitHub: addressing BigInteger comment: BigInteger support in PInt oracle/graalpython#346 (comment)
• standups, emails, demo meetings It should be finished by 2023-08-21.

Next Day: More of Python Interop & TCK