Infer context for Data instance of (data Foo f = Foo (f Bool) (f Int))

(based on WIP -XQuantifiedConstraints)

Could GHC infer the context for deriving Data (Foo f) (#12245 (closed))?

data Foo f = Foo (f Bool) (f Int)
  deriving stock Data

{-
Generates:

instance (Typeable f, (forall xx. Data xx => Data (f xx))) => Data (Foo f)
-}

Trac metadata

Trac field	Value
Version
Type	FeatureRequest
TypeOfFailure	OtherFailure
Priority	normal
Resolution	Unresolved
Component	Compiler
Test case
Differential revisions
BlockedBy
Related
Blocking
CC
Operating system
Architecture

changed weight to 5

added Tfeature request Trac import labels

I'm not sure. When doing

data T a = MkT a [a] deriving( Eq )

the deriving mechanism says "oh, we need Eq on each field, that is Eq a, Eq [a]". Then it simplifies that to arrive at the instance context.

For your example, it'd say "I need Data (f Bool), Data (f Int) and then get stuck. There are two ways to get unglued. One is simply to put those in the instance context

instance (Data (f Int), Data (f Bool)) => Data (Foo f) where ...

The other is to use quantified constraints. It's not clear which is best in general. And even figuring out the appropriate quantified-constraint generalisation of a set of ground constraints is far from trivial.

added deriving label

I too am skeptical that there is a robust, systematic way to do what you're seeking. I think the intent of #12245 (closed)#ticket:14748#comment:148487 was to combine quantified constraints with StandaloneDeriving, not have the quantified constraints be inferred.

closed

Closing, since this is neither how we want to tackle this problem, nor a solution that will work most of the time, as per ticket:14748#comment:148487.

Trac metadata

Trac field	Value
Resolution	Unresolved → ResolvedWon'tFix

added 1 deleted label

added Pnormal label