killy971 / hpc-coveralls

coveralls.io support for haskell code coverage with hpc
http://hackage.haskell.org/package/hpc-coveralls
BSD 3-Clause "New" or "Revised" License
60 stars 20 forks source link

Add 7.8.4, 7.10.2 and 7.10.3 to travis matrix #55

Closed phadej closed 8 years ago

phadej commented 8 years ago

Last commit is optional, but I'd advice to have upper bounds.

killy971 commented 8 years ago

For some reason I can't remember (might have been an article I read a while back), I had the impression that it was not such a good idea to add upper bounds, but maybe it was my misunderstanding.

Is it still a common practice to set upper bounds these days?

Also, how should the upper bound value be defined? Is this a case by case decision?

phadej commented 8 years ago

One good answer on top of my head is https://github.com/haskell/cabal/pull/2774#issuecomment-132479035 (related discussion before and after is worth reading if you have time).

So upper bounds here are forbidding next major release of each depedency, which might introduce breaking changes affecting hpc-hoveralls, like retry-0.7 did. This for example caused travis builds of purescript fail for no reason, that's how I found an issue with hpc-hoveralls.