|Version 18 (modified by dreixel, 4 years ago) (diff)|
The new Generic Deriving mechanism (ongoing work)
GHC includes a new (in 2010) mechanism to let you write generic functions. It is described in A generic deriving mechanism for Haskell, by Magalhães, Dijkstra, Jeuring and Löh. This page sketches the specifics of the implementation; we assume you have read the paper.
This mechanism replaces the previous generic classes implementation. The code is in a branch of GHC; you can get it with darcs get http://darcs.haskell.org/ghc-generic-15Feb11/ghc.
Changes from the paper
In the paper we describe the implementation in UHC. The implementation in GHC is slightly different:
- We are using type families, so the Representable0 and Representable1 type classes have only one type argument. So, in GHC the classes look like what we describe in "Avoiding extensions" part of Section 2.3 of the paper. This change affects only a generic function writer, and not a generic function user.
- Default definitions (Section 3.3) work differently. In GHC we don't use a DERIVABLE pragma; instead, a type class can declare a generic default method, which is akin to a standard default method, but includes a generic type signature. For example, the Encode class of Section 3.1 is now:
class Encode a where encode :: a -> [Bit] generic encode :: (Representable0 a, Encode1 (Rep a)) => a -> [Bit] encode = encode1 . from0This removes the need for a separate default definition and a pragma.
- To derive generic functionality to a user type, the user no longer uses deriving instance (Section 4.6.1). Instead, the user gives an instance without defining the method; GHC then uses the generic default. For instance:
instance Encode [a] -- works if there is an instance Representable0 [a]
- TcDeriv.tcDeriving generates an InstInfo for each data type that fulfills the isRep0 predicate. This InstInfo is the Representable0 instance for that type, allowing it to be handled generically (by kind-* generic functions).
- The representation types and core functionality of the library live on GHC.Generics (on the ghc-prim package).
- Many names have been added as known in prelude/PrelNames
- Most of the code generation is handled by types/Generics
What already works
- Representable0 instances are automatically generated when -XGenerics is enabled.
- Remove all of the old deriving mechanism stuff
- Properly deal with isTuple information for constructors
- Generate Representable1 instances
- What about base types like , Maybe, etc.?
- For temporary testing, a file test/Main.hs is available with sample datatypes.
- Currently, in TcDeriv.genGenericRepBind we generate instances using mkLocalInstance. Is this right, or should we use mkImportedInstance instead? SLPJ: mkLocalInstance: it's as if the instance declaration was in this module, right?