Tests for Ratio fail.

[kantp]

The tests for Ratio Int8 and Ratio Int64 fail with an arithmetic overflow. This is rather strange, given that the implementations of poke, peek, and size do not seem to be performing any kind of arithmetic.

Failures:

  test/Data/StoreSpec.hs:237:
  1) Data.Store, Manually listed polymorphic store instances, Roundtrips (GHC.Real.Ratio GHC.Int.Int8)
       uncaught exception: ArithException (arithmetic overflow)

  test/Data/StoreSpec.hs:237:
  2) Data.Store, Manually listed polymorphic store instances, Roundtrips (GHC.Real.Ratio GHC.Int.Int64)
       uncaught exception: ArithException (arithmetic overflow)

[mgsloan]

Yes, this is why we have a submodule dep on smallcheck. For some reason roman has not accepted this PR https://github.com/feuerbach/smallcheck/pull/34

This is among the reasons I'm considering moving away from using smallcheck - https://github.com/fpco/store/issues/17

[kantp]

Oh, I see. That's inconvenient.

[mgsloan]

We actually should maybe consider changing the instance for Ratio to use Storable. We have two choices:

• Enable serialization of invalid values, violating Ratio invariants (current behavior)
• Have both peek and poke throw exceptions for this

I think when this decision was made I was working on making all the unit tests pass. I think the other behavior is likely more sensible

[bitonic]

I've mitigated this in e9aa2be52fa9e8c37b74f06522507604bd135d85 . This became necessary because tests were failing while building docker images.

[bitonic]

@mgsloan I think that this means that the smallcheck submodule is not needed anymore.

[mgsloan]

Yup, feel free to remove it!