Could you please reconsider the addition of @RangeSource?

[k163377]

I have a heavy test for a very large range and am trying to parallelize this.

In doing so, I have noticed that if I use a @MethodSource, it will not be parallelized.

If this is a limitation due to circumstances specific to @MethodSource, could we introduce a @RangeSource?

In my layman's opinion, I think it would be easier to implement since the internal processing can be done in the same way as a @ValueSource.

Also, I feel that @MethodSource is a bit difficult to use in several ways, so I would like a @RangeSource in that sense.

I realize that this is a feature that has already been denied once with #851, but could you please consider it again?

[sbrannen]

Have you tried JUnit Pioneer's support for Range Sources?

Also, I feel that @MethodSource is a bit difficult to use in several ways, so I would like a @RangeSource in that sense.

How about using a @FieldSource?

@ParameterizedTest
@FieldSource
void range(int argument) {
	assertTrue(argument >= 110 && argument <= 200);
}

static Supplier<IntStream> range = () -> IntStream.rangeClosed(100, 200).skip(10);

I find that rather compelling and straightforward.

Thoughts?

[k163377]

Have you tried JUnit Pioneer's support for Range Sources?

Sorry, I had missed this.
This is exactly what I was looking for, thank you.

How about using a @FieldSource?

Sorry, I missed this as well.
If it can run concurrently, this would be useful for my use case.

Thoughts?

I've made the point a bit vague, but I think the problem remains that the MethodSource is not executed concurrently.
However, is this a duplication of #3373?
If so, I will close it.

[sbrannen]

Thanks for the feedback, @k163377.

Have you tried JUnit Pioneer's support for Range Sources?

Sorry, I had missed this. This is exactly what I was looking for, thank you.

👍

How about using a @FieldSource?

Sorry, I missed this as well.

No worries. It's "new" since JUnit Jupiter 5.11 which was just recently released.

I've made the point a bit vague, but I think the problem remains that the MethodSource is not executed concurrently.

I don't think there's anything particular about @MethodSource with regards to parallel execution: every ArgumentsProvider for a parameterized test ends up producing a Stream of Arguments.

However, is this a duplication of #3373? If so, I will close it.

Indeed, I am closing this as a duplicate of the following.

• Introduce @RangeSource for parameterized tests #851
• Cannot run @ParameterizedTest method invocations in parallel and other test sequentially #3373