I'm wondering if this should have been `>=` instead of `>` . The `min_value_and_waste` change policy was previously returning change for excess equal to the min value.

[LLFourn]

          I'm wondering if this should have been `>=` instead of `>` . The `min_value_and_waste` change policy was previously returning change for excess equal to the min value.

Originally posted by @jp1ac4 in #14 (comment)

Create a test for this.

[jp1ac4]

For this test, should it also check that if the change policy min value is 0 and the drain value is 0, then drain_value still returns None (i.e. we don't include a drain with zero value)?

EDIT: Perhaps it would be simpler to keep it as > and let the user set the min value accordingly.

[LLFourn]

Well if we keep it as > we have to change the name since I read min as inclusive.

[jp1ac4]

Well if we keep it as > we have to change the name since I read min as inclusive.

I agree, I think it's better to change it back to >=. I can try working on this if you like.

[jp1ac4]

If we use >= and the ChangePolicy has min_value of 0, do you think it should be possible to create a change output with value 0?

(Realistically, a user would not set 0 as the min value, but I'm wondering whether this should be explicitly prevented.)

[jp1ac4]

If we use >= and the ChangePolicy has min_value of 0, do you think it should be possible to create a change output with value 0?

I think a change ouput with value 0 would be even less likely if the metric controlled the change amount as per #20.