search

fpco / inline-c

284 stars 51 forks source link

Fix spurious build failure #118

Closed roberth closed 3 years ago

roberth commented 3 years ago

I've encountered quickcheck giving in the tests. I've had a look at the cause but didn't find a good way to improve example generation. It's a good approach because apparently this isn't encountered often. This change doubles the discard ratio, which is still good and fixes test case generation. IIUC this shift the cut-off on a bell curve to a point where the probability of failure is extremely low.

log section ``` Failures: test/Language/C/Types/ParseSpec.hs:33:3: 1) Language.C.Types.Parse parses everything which is pretty-printable (C) Gave up after 94 tests; 1000 discarded! To rerun use: --match "/Language.C.Types.Parse/parses everything which is pretty-printable (C)/" test/Language/C/Types/ParseSpec.hs:44:3: 2) Language.C.Types.Parse parses everything which is pretty-printable (Haskell) Gave up after 94 tests; 1000 discarded! To rerun use: --match "/Language.C.Types.Parse/parses everything which is pretty-printable (Haskell)/" Randomized with seed 1695730547 ```

Happy building!

roberth commented 3 years ago

@bitonic This fixes CI

Third time's a charm.

bitonic commented 3 years ago

Thanks, released as 0.9.1.3.