spell-music/data-fix

Fixpoint data types

data-fix's Issues

Existentially quantified types for hoistFix

Thank you for the library!

Why existentially quantified types for
hoistFix :: Functor f => (forall a. f a -> g a) -> Fix f -> Fix g
and
hoistFix' :: Functor g => (forall a. f a -> g a) -> Fix f -> Fix g
type annotations. But not simply
hoistFix :: Functor f => (f (Fix g) -> g (Fix g)) -> Fix f -> Fix g
and
hoistFix' :: Functor g => (f (Fix f) -> g (Fix f)) -> Fix f -> Fix g
Probably the original idea behind was to map f -> g to Fix f -> Fix g.
But in fact the proposed method are more general.

Versioning & breaking API changes

hnix broke recently because of a version change.
It’s impossible to add a version bound at the moment, because there is no distinction between API-breaking changes and additions.

Would you be okay with switching to the second digit and incrementing it when a breaking API-change happens? Then hnix could depend on e.g. ==0.2.*.

cc @jwiegley

Build error with 7.8.4

I made a revision https://hackage.haskell.org/package/data-fix-0.2.1/revisions/

Not an issue - just a question about the haskell type system that came to my mind by looking at data-fix

Please excuse if this is a dumb question 😃

Maybe I just missing some special Haskell syntax - but I don't find a good explanation for:

type MyMaybe :: Type -> Type
type MyMaybe = Maybe

type MyMaybe' :: Type -> Type
type MyMaybe' a = Maybe a

type Works = Fix MyMaybe

-- • The type synonym ‘MyMaybe'’ should have 1 argument, but has been given none
-- • In the type synonym declaration for ‘DoesntWork’
type DoesntWork = Fix MyMaybe'

Here for sure the argument can be avoided - but what if the type is nested - e.g. something like Either (Maybe a ) ()? How to apply Fix in such a case?

Add to Stackage

Nice package - have you thought about adding it to Stackage? I'm using it and it would make my life easier. See https://github.com/fpco/stackage/blob/master/MAINTAINERS.md#adding-a-package for details.

Add Hashable instance

It would be nice to have a hashable instance for Fix. It could be defined with Hashable1. I would PR this if you would take it.

What does efficient folding/unfolding mean?

The documentation for Mu says: "Efficient folding." and the documentation for Nu says: "Efficient unfolding."

What does efficient mean in this case? Does it have to do with fusion or is it just that the implementations of the respective functions are trivial? Is there a benchmark that shows the difference?

See also discussion at: https://discourse.haskell.org/t/performance-of-fixed-points-and-fusion/8878?u=jaror

Use Eq1, Show1, etc. for instances

Since base-4.9, the Data.Functor.Classes module has been available (previously, it was available through the transformers library). It would be nice if the instances for Fix relied on these instead of using FlexibleContexts. I am willing to PR this if you would accept it.

Migrate away from Eq1/Ord1/Show1

Continuing on haskell/core-libraries-committee#190 (comment), I propose to migrate from

instance Eq1 f => Eq (Fix f)
instance Ord1 f => Ord (Fix f)
instance Show1 f => Show (Fix f)

to

instance Eq (f (Fix f)) => Eq (Fix f)
instance Ord (f (Fix f)) => Ord (Fix f)
instance Show (f (Fix f)) => Show (Fix f)

or if you fancy QuantifiedContraints

instance (forall a. Eq a => Eq (f a)) => Eq (Fix f)
instance (forall a. Ord a => Ord (f a)) => Ord (Fix f)
instance (forall a. Show a => Show (f a)) => Show (Fix f)

The motivation is that I recently migrated a codebase from a homegrown newtype Fix with instance Ord (f (Fix f)) => Ord (Fix f) to data-fix and hit performance issues. They were a combination of several factors:

• I used Data.Functor.Classes.Generic to derive Ord1 instance generically. It appeared that from and to survived in Core, despite pretty aggressive optimization options. This is admittedly not a fault of data-fix, but a general issue with usability of Eq1 / Ord1 / Show1. I had to write all instances manually.

• class Ord1 contains only liftCompare, but no liftLT, liftLE, liftGT, liftGE, so it has no access to <, <=, >=, > of the base functor. This implies that < on Fix f also necessarily goes through compare and takes a performance hit.

• It does not seem that liftEq / liftCompare can be optimized away unless the base functor f is completely monomorphic and everything is forced to inline. E. g., (==) specialised to Fix (ListF Int) looks reasonable, but (==) for Fix (ListF a) goes through liftEq. The Core for

  instance Eq (f (Fix f)) => Eq (Fix f) where
    (==) = coerce ((==) @(f (Fix f)))

  looks more reasonable to me.

---

I understand that the proposed change is a breaking one. It is however not very pronounced: in recent base we have class (forall a. Eq f a) => Eq1 f, so the majority of users will not even notice. It's only if you require Eq (Fix f) as a constraint, but use liftEq inside, which will explode, because Eq1 f is no longer guaranteed. I think, however, that GHC would yell if encounters Eq (Fix f) constraint, suggesting to replace it with Eq1 f.

@phadej what do you think?