Re: [commit: packages/base] master: Add fmapCoerce to Functor class. (3da4fd9)

older
Incorrect Binding in GHCi (trac...

Johan Tibell

25 Sep 2013 25 Sep '13

12:41 a.m.

Having coerce (!) show up in such a commonly used type class is a bit alarming. Was there a libraries discussion about this? At least we should add some docs to the method. On Wed, Sep 25, 2013 at 12:37 AM, wrote:

...

Repository : ssh://git@git.haskell.org/base

On branch : master Link : http://ghc.haskell.org/trac/ghc/changeset/3da4fd92e005cb9348ab2852d9268ad61a...

...
---------------------------------------------------------------

commit 3da4fd92e005cb9348ab2852d9268ad61ae1b347 Author: Austin Seipp Date: Tue Sep 24 23:20:35 2013 -0500

Add fmapCoerce to Functor class.

This allows coercions to work under functors that are not locally known.

Authored-by: Edward Kmett Signed-off-by: Austin Seipp

...
---------------------------------------------------------------

3da4fd92e005cb9348ab2852d9268ad61ae1b347 GHC/Base.lhs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/GHC/Base.lhs b/GHC/Base.lhs index d876202..55172af 100644 --- a/GHC/Base.lhs +++ b/GHC/Base.lhs @@ -105,7 +105,6 @@ module GHC.Base module GHC.Err -- import it explicitly ) where - import GHC.Types import GHC.Classes import GHC.CString @@ -186,6 +185,9 @@ class Functor f where (<$) :: a -> f b -> f a (<$) = fmap . const

+ fmapCoerce :: Coercible a b => f a -> f b + fmapCoerce = fmap coerce + {- | The 'Monad' class defines the basic operations over a /monad/, a concept from a branch of mathematics known as /category theory/. From the perspective of a Haskell programmer, however, it is best to

_______________________________________________ ghc-commits mailing list ghc-commits@haskell.org http://www.haskell.org/mailman/listinfo/ghc-commits

Show replies by date

Edward Kmett

25 Sep 25 Sep

3:04 a.m.

New subject: [commit: packages/base] master: Add fmapCoerce to Functor class. (3da4fd9)

The discussion was back and forth with SPJ and Richard Eisenberg during the discussion about the creation of the roles machinery and Simon suggested the shape it currently has to me last night. Having this in the class is the only way that this can lift over an 'unknown' Functor in local scope, and permits users of a functor that is polymorphic to get access to the O(1) coercion machinery. Without it Coercible is almost completely useless in any code that has any sort of polymorphism involved. This is analogous to (<$), which was added to Functor explicitly because it made an asymptotic performance difference in a large number of scenarios. The reason this has to live in the class is analogous to how Data has to hold the at-first-glance redundant dataCast1 and dataCast2, because it is the only place the right things are known, despite them always being gcast1 and gcast2. See 7.4 in Scrap more boilerplate: reflection, zips, and generalised castshttp://research.microsoft.com/en-us/um/people/simonpj/papers/hmap/ for a discussion of an almost identical issue. There is still some work to be done here. I definitely agree that we should add some docs to it. One of the reasons why we didn't in this initial commit is because there is some of it still up in the air in terms of what is possible. The current default definition is in terms of 'fmap coerce', so it'll currently work with even 'illegal' functors that don't have a representational argument, but the user doesn't see any improvement unless One of the reasons the docs didn't go in with the first commit was that there is room for improvement. *If* we can improve some things about Coercible inference we may be able to upgrade it to actually use 'coerce' via a default signature. -Edward On Wed, Sep 25, 2013 at 12:41 AM, Johan Tibell wrote:

...

Having coerce (!) show up in such a commonly used type class is a bit alarming. Was there a libraries discussion about this? At least we should add some docs to the method.

On Wed, Sep 25, 2013 at 12:37 AM, wrote:

...
Repository : ssh://git@git.haskell.org/base

On branch : master Link : http://ghc.haskell.org/trac/ghc/changeset/3da4fd92e005cb9348ab2852d9268ad61a...

...
---------------------------------------------------------------

commit 3da4fd92e005cb9348ab2852d9268ad61ae1b347 Author: Austin Seipp Date: Tue Sep 24 23:20:35 2013 -0500

Add fmapCoerce to Functor class.

This allows coercions to work under functors that are not locally known.

Authored-by: Edward Kmett Signed-off-by: Austin Seipp

...
---------------------------------------------------------------

3da4fd92e005cb9348ab2852d9268ad61ae1b347 GHC/Base.lhs | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/GHC/Base.lhs b/GHC/Base.lhs index d876202..55172af 100644 --- a/GHC/Base.lhs +++ b/GHC/Base.lhs @@ -105,7 +105,6 @@ module GHC.Base module GHC.Err -- import it explicitly ) where - import GHC.Types import GHC.Classes import GHC.CString @@ -186,6 +185,9 @@ class Functor f where (<$) :: a -> f b -> f a (<$) = fmap . const

+ fmapCoerce :: Coercible a b => f a -> f b + fmapCoerce = fmap coerce + {- | The 'Monad' class defines the basic operations over a /monad/, a concept from a branch of mathematics known as /category theory/. From the perspective of a Haskell programmer, however, it is best to

_______________________________________________ ghc-commits mailing list ghc-commits@haskell.org http://www.haskell.org/mailman/listinfo/ghc-commits

ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

Joachim Breitner

4:18 p.m.

New subject: [commit: packages/base] master: Add fmapCoerce to Functor class. (3da4fd9)

Hi, Am Mittwoch, den 25.09.2013, 03:04 -0400 schrieb Edward Kmett:

...

The discussion was back and forth with SPJ and Richard Eisenberg during the discussion about the creation of the roles machinery and Simon suggested the shape it currently has to me last night.

given the recently intensified urge to make GHC a more community driven project, may I suggest to reduce the amount of off-list discussions? I find it very confusing if some discussions happen on list, some discussion on list with certain people explictily CC’ed, and some discusisons off-list, and some alternating between these. I believe we are all able to ignore discussions on the list that are not relevant for us, but on-list discussions are easily referenced and re-read later.

...

Having this in the class is the only way that this can lift over an 'unknown' Functor in local scope, and permits users of a functor that is polymorphic to get access to the O(1) coercion machinery.

Is the Coercion feature really mature enough to be embedded in Functor? There are still some open questions (e.g. the in-scope-requirements for it). I would feel more comfortable if people have one release’s time to break it before we put it so easily in every programmers hands. Greetings to ICFP from the Heidelberg Laureates Forum, Joachim -- Joachim “nomeata” Breitner mail@joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata@joachim-breitner.de • GPG-Key: 0x4743206C Debian Developer: nomeata@debian.org

Ross Paterson

6:22 p.m.

New subject: [commit: packages/base] master: Add fmapCoerce to Functor class. (3da4fd9)

On Wed, Sep 25, 2013 at 10:18:51PM +0200, Joachim Breitner wrote:

...

given the recently intensified urge to make GHC a more community driven project, may I suggest to reduce the amount of off-list discussions? I find it very confusing if some discussions happen on list, some discussion on list with certain people explictily CC’ed, and some discusisons off-list, and some alternating between these. I believe we are all able to ignore discussions on the list that are not relevant for us, but on-list discussions are easily referenced and re-read later.

I agree with all that, and also share Johan's alarm at the sudden appearance of coerce in this class. On the substance, I don't think Functor is the right place for this, as it's about lifting isomorphisms rather than functions, and type constructors don't have to be functors to support that.

Richard Eisenberg

8:38 p.m.

New subject: fmapCoerce / changes in 7.8

In light of the conversations I've been having with Edward and others, I, too, wonder if we should wait a little before doing this. There is no burning fire here. Though we have Coercible in 7.8, that doesn't mean we *have* to use it. The old trick to use unsafeCoerce in the situations Edward wants to improve will still work, and it will remain as safe as it was before (i.e., not at all). In the end, I agree with Joachim that Coercible should try to have a low profile in 7.8. I do think that the ideas behind fmapCoerce -- that is, making sure that the new `coerce` primitive is used to the best effect -- are great. But I am not convinced that we've found the best possible way to do it. Richard On Sep 25, 2013, at 6:22 PM, Ross Paterson wrote:

...

On Wed, Sep 25, 2013 at 10:18:51PM +0200, Joachim Breitner wrote:

...
given the recently intensified urge to make GHC a more community driven project, may I suggest to reduce the amount of off-list discussions? I find it very confusing if some discussions happen on list, some discussion on list with certain people explictily CC’ed, and some discusisons off-list, and some alternating between these. I believe we are all able to ignore discussions on the list that are not relevant for us, but on-list discussions are easily referenced and re-read later.

I agree with all that, and also share Johan's alarm at the sudden appearance of coerce in this class.

On the substance, I don't think Functor is the right place for this, as it's about lifting isomorphisms rather than functions, and type constructors don't have to be functors to support that. _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

Edward A Kmett

26 Sep 26 Sep

12:39 a.m.

New subject: fmapCoerce / changes in 7.8

I agree with Richard. We're still working up the right way to get this functionality to work, and sadly the simple solution doesn't cover enough cases. As I noted during the day, this approach doesn't scale. Let's revert. -Edward

...

On Sep 25, 2013, at 8:38 PM, Richard Eisenberg wrote:

In light of the conversations I've been having with Edward and others, I, too, wonder if we should wait a little before doing this. There is no burning fire here. Though we have Coercible in 7.8, that doesn't mean we *have* to use it. The old trick to use unsafeCoerce in the situations Edward wants to improve will still work, and it will remain as safe as it was before (i.e., not at all).

In the end, I agree with Joachim that Coercible should try to have a low profile in 7.8. I do think that the ideas behind fmapCoerce -- that is, making sure that the new `coerce` primitive is used to the best effect -- are great. But I am not convinced that we've found the best possible way to do it.

Richard

...
On Sep 25, 2013, at 6:22 PM, Ross Paterson wrote:

...
On Wed, Sep 25, 2013 at 10:18:51PM +0200, Joachim Breitner wrote: given the recently intensified urge to make GHC a more community driven project, may I suggest to reduce the amount of off-list discussions? I find it very confusing if some discussions happen on list, some discussion on list with certain people explictily CC’ed, and some discusisons off-list, and some alternating between these. I believe we are all able to ignore discussions on the list that are not relevant for us, but on-list discussions are easily referenced and re-read later.

I agree with all that, and also share Johan's alarm at the sudden appearance of coerce in this class.

On the substance, I don't think Functor is the right place for this, as it's about lifting isomorphisms rather than functions, and type constructors don't have to be functors to support that. _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

David Luposchainsky

3:59 a.m.

New subject: fmapCoerce / changes in 7.8

I'm also sceptical about this, regardless of whether it sounds like a good idea now. Adding a function to a typeclass is trivial, removing it later far from it (see fail, some, many). On the other hand, are there reasons not to create

...

class Functor f => FunctorC f where fmapC :: Coercible a b => f a -> f b

instead? David/quchen

Edward A Kmett

8:13 a.m.

New subject: fmapCoerce / changes in 7.8

A very real problem with making such a class is it basically forces a tension on any library to make two versions of literally everything. Consider the case of lens, which makes incredibly aggressive use of such fmap-likes to put on and take off newtypes and does very little else. All the combinators that do so would have to duplicate to work in such a world. Because now such an instance may not exist. This is mapM vs traverse for library authors, but it is an unforced error here. There are three agents involved in consumption and production of such an instance: The author of the library that demands it, the author of the library that needs to supply it and the end user of both. In my experience, making a separate class like this only really works when these three roles are partitioned among two people. You need to support data types not supplied by the library, and not supplied by the end user. The rest of the time it leads to a ratsnest of orphans and missing instances for something that should be supportable for every functor. I've tried this. The result isn't maintainable for the library author or the user. The reason we had Austin revert the patch was because this approach wasn't sufficient to solve the problem, so we backed off of it until Richard can get us better tools for composing Coercible witnesses, to let us upgrade this to something that can give you O(1) lifting over abstract transformer stacks. When it is exiled to some other class, that sort of code just becomes something that isn't safe to use, and in many cases like while writing a transformer, isn't possible to use. We introduce an unnecessary tension and duplication for authors and users. -Edward

...

On Sep 26, 2013, at 3:59 AM, David Luposchainsky wrote:

I'm also sceptical about this, regardless of whether it sounds like a good idea now. Adding a function to a typeclass is trivial, removing it later far from it (see fail, some, many). On the other hand, are there reasons not to create

...
class Functor f => FunctorC f where fmapC :: Coercible a b => f a -> f b

instead?

David/quchen _______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

4254

Age (days ago)

4255

Last active (days ago)

List overview

Download

7 comments

7 participants

participants (7)

David Luposchainsky
Edward A Kmett
Edward Kmett
Joachim Breitner
Johan Tibell
Richard Eisenberg
Ross Paterson