Herbert Valerio Riedel

...what about a more or less verbose variable called `CABAL_SANDBOX`?

> The size of data loaded in memory may be 2-4 times bigger than in the database because of boxed representation in Haskell. Yeah, but I'm surprised that the majority...

@lykahb What's the ETA for `groundhog-pipes` btw?

> Oddly, I need to use the `ghc802` branch for 8.2, and `master` for 8.0. It's less confusing if you're already used to `__GLASGOW_HASKELL__`'s numbering convention... :-)

@DanBurton I did mean to place InfoMod a4 -> InfoMod Options • Perhaps you meant one of these: ‘’ (imported from Options.Applicative), ‘*>’ (imported from Options.Applicative), ‘ Mod f1 Verbosity...

@DanBurton The simplest explanation I can think of is the stackage nightly buildbot failed to re-check the build-status for `angel-0.6.2` after `optparse-applicative-0.13` entered the snapshot

@MichaelXavier yes, that would be the proper fix to make `angel` compatible with `optparse-applicative-0.13` Btw, you also should add a lower bound on `optparse-applicative`, as `angel` fails to compile with...

Here's the output from today's GHC w/o `?`-marks: ``` src/Control/Bifunctor.hs:38:30: The first argument of ‘Bifunctor’ should have kind ‘x1 -> y1 -> z1’, but ‘p’ has kind ‘x2 -> y2...

Fwiw, if you `M-x package-install haskell-snippets`, you get autocompletion via https://github.com/haskell/haskell-snippets/blob/master/snippets/haskell-mode/lang-pragma

please migrate to https://github.com/ekmett/ansi-wl-pprint/issues