sieve: Rewrite tests to use hspec with fail-fast. #295
Conversation
|
👍 It makes me realize an interesting thing about the x-common data - "limit is prime" is already covered by 2 case! |
| , 919, 929, 937, 941, 947, 953, 967, 971, 977, 983, 991, 997 | ||
| ] @=? primesUpTo 1000 | ||
| , testCase "first thousand primes" $ | ||
| 1000 @=? length (primesUpTo 7919) |
There was a problem hiding this comment.
did you want to keep this as a track-specific test?
I'm happy regardless of what the answer is
There was a problem hiding this comment.
Nice catch!
I just copied the test from x-common and forgot about this one. If you think this adds anything to the exercise, I can put it back without problems. 😄
There was a problem hiding this comment.
I think most other tracks do not have it, and I don't think it adds a whole lot for us (if they can go up to 1000, they surely can go up to 7919 without problems, right?), so I wouldn't have been sad to lose it. Unless there is a Haskell-specific consideration that I am not aware of :o
There was a problem hiding this comment.
and in case you wanted to look through history, it's been here since the start at 2fa545b
There was a problem hiding this comment.
This is not the first exercise in the Haskell track that tests big numbers without an obvious reason.
If it is there to test performance, it should go in a benchmark suite (this is something that's easy to do now), if it isn't, the test should have a explicit goal.
Let's just drop it! 😄
Related to #211.