Vladislav Zavialov

No problem, it's by no means urgent. I'm not even sure that HPC will work at all because today I've separated the test suite itself into a separate package as...

> The typeclass in that module should actually redefine rput, rget, etc. as well That's a useful insight, thanks. I thought about unifying `RElem` and `UElem`, but my initial ideas...

Type-indexed patterns: ``` f = OpenUnion '[Integer, Bool, Char] -> r f = \case OU (a :: Integer) -> ... OU (b :: Bool) -> ... OU (c :: Char)...

There is a `SetLength` now, but it doesn't always work.

I think we can have both `threadDelay @Second 5` and `threadDelay 5 @Second` work if we change the type signature like this: ``` threadDelay :: forall (unit :: Rat) m....

https://ghc.haskell.org/trac/ghc/ticket/13773

Consider [turtle](http://hackage.haskell.org/package/turtle) as an alternative to Shelly, too.

> any strong opinions? I feel rather strongly that we should retain the option of generating safe code without `unsafeCoerce`. The remaining configuration space can be reduced.

Sounds like an issue with the `happy` that you have in your `$PATH`. Similar issue: https://github.com/haskell/happy/issues/262. I can't reproduce the failure.

> Do you mean you can only bootstrap happy-2 by happy-2 ? Not at all. I mean that your installation of `happy-1` might be somehow misconfigured.