Can't use value ranges from python

[luxaritas]

Describe the bug An error is encountered when attempting to annotate a ValueRangeProvider as a CountableValueRange

Expected behavior No error is encountered

Actual behavior

java.lang.java.lang.IllegalArgumentException: java.lang.IllegalArgumentException: The entityClass (class <domainobject>) has a @PlanningVariable annotated property (<property>) that refers to a @ValueRangeProvider annotated member (method $method$get_range on class <domainobject>) that does not return a Collection, an array or a ValueRange

To Reproduce Attach the following to a planning entity and attempt to run the solver

val: Annotated[int, PlanningVariable(value_range_provider_refs=['val'])] = 0
def get_range(self) -> Annotated[CountableValueRange, ValueRangeProvider(id='val')]:
        return ValueRangeFactory.create_int_value_range(0, 10)

Environment

Timefold Solver Version or Git ref: 1.17.0

Output of java -version:

openjdk version "23" 2024-09-17
OpenJDK Runtime Environment (build 23)
OpenJDK 64-Bit Server VM (build 23, mixed mode, sharing)

Output of uname -a or ver:

Linux 6.6.63-1-MANJARO #1 SMP PREEMPT_DYNAMIC Sat Nov 23 02:15:34 UTC 2024 x86_64 GNU/Linux

[triceo]

CountableValueRange is not meant to be used as an annotation. Instead, a value range provider may return instances of this type as a means of providing values, if a collection is not used. (Which is what you're doing.)

Value ranges should only be annotated with ValueRangeProvider - and if there is only one value range, it doesn't need to be annotated at all.

Please remove the annotation and let us know if the issue persists.

[luxaritas]

I’m not using it as an annotation here - I’m using it as the return type. This matches how the docs say to specify the return type of a function using ValueRangeFactory. (Nb: Annotated[ValueRangeProvider] would mean that the return type of the function is ValueRangeProvider)

[triceo]

OK, thanks for clarifying. In that case, this is probably a bug.

[triceo]

As of today, Timefold halted any further development of the Python solver. Find out more here: https://github.com/TimefoldAI/timefold-solver/discussions/1698

Closing this issue.