raindrops: Rewrite tests to use hspec with fail-fast.

[rbasso]

• Rewrite test to use hspec
• Remove test not matching x-common.

Related to #211.

[petertseng]

👍 as-is, but while we're thinking about it, how about a potential question:

For all cases, it is true that description is just show number. Given this consideration, is description really necessary? Then again, I imagine these were generated straight from the JSON file, so this question may have been better for the JSON file in x-common.

So 👍 here.

[rbasso]

I agree! It's ugly to have the descriptions equal to the numbers. Considering that they don't add any useful information, It would be nice to remove them in x-common, but I can also remove it now from xhaskell. Should I do it?

Edit: Done. :)

[petertseng]

Ah, right. 👍 then, no question.

Optional: This would be a candidate to use tuples since the two fields should be unambiguous, as in prime-factors the other PR, but as I've said I do not object to a record even if there is an opportunity to use a tuple.

[rbasso]

That was fast, @petertseng!

I'm followed your advice and changed to tuples. 👍

[petertseng]

👍