[QuickCheck] 2.7 is a huge pain in the ass

[Nick Smallbone]

Hi Bryan,

Sorry to hear this. I do try hard to avoid making breaking changes to
QuickCheck, but as there hadn't been a significant QuickCheck release
for nearly two years (my fault) there were a couple of API changes
this time. Can I ask what exactly you ran into?

I also have a mirror of Hackage that I grep to assess the impact of
any API change, and back out if it's too severe, but this time I
must've missed something. I think in the future I will also build HP
with the new version to stop this happening again.

Nick

On Monday 28 April, 2014 at 04:57 am, Bryan O'Sullivan wrote:
> Hey folks,
> 
> As QuickCheck is a widely used library, it's very bad form to make sweeping
> breaking changes to QuickCheck. I'll probably be spending the rest of
> tonight fixing up test suites so that they work with 2.7. This makes me
> quite unhappy.

> _______________________________________________
> QuickCheck mailing list
> QuickCheck at projects.haskell.org
> http://projects.haskell.org/cgi-bin/mailman/listinfo/quickcheck