We have a constraint (a~b) that can appear in a type signature
f :: (a~b) => a -> b
We can also have a type-level equality function
type family Eq a b where
Eq a a = True
Eq a b = False
The type-level Boolean reflects the (a~b) constraint precisely.
You sound as if you want a type family Satisfied, so you
type family Satisfied :: Constraint -> Bool
Then you could have
instance Ord T where ...
type instance Satisfied (Ord T) = True
The difficulty is that we can never return False, because you can always make T an instance of Ord "later". And you can't have Satisfied (Ord T) returning True in one place and Falsie in another. It can get stuck (not reduce) but it can't return False.
I don't know if this would be useful to you.
And for more complex instances lie
instance Ord a => Ord [a]
I don't know what you want to say for
type instance Satisfied (Ord a) = ???
Simon
From: Libraries [mailto:libraries-bounces@haskell.org] On Behalf Of Ryan Newton
Sent: 15 October 2013 04:03
To: Pedro Magalhães (dreixel@gmail.com)
Cc: Haskell Libraries; Andres Löh; Ganesh Sittampalam; ghc-devs@haskell.org
Subject: Re: Proposal: GHC.Generics marked UNSAFE for SafeHaskell
Hey, that's an awesome formulation! Thanks Pedro.
Any idea how much work this would be to implement in GHC, if it did garner approval?
On Tue, Oct 8, 2013 at 3:48 AM, José Pedro Magalhães