I suggest doing a major release of transformers at the same time as the next major GHC release, with Data.Functor.Identity removed and added to

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: new Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Comment (by hvr): Replying to [comment:1 ross]: base. This version would depend on the new version of base to force an upgrade. Terrific! I'll draft a `transformers`-patch in our locally forked repo over at http://git.haskell.org/packages/transformers.git and migrate `Data.Functor.Identity` into GHC HEAD's `base` so we can start using that in GHC HEAD. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: patch Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D313 | -------------------------------------+------------------------------------- Changes (by hvr): * cc: core-libraries-committee@… (added) Comment: @ross, any comments? -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:4 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

Changing to have Data.Functor.Class import Data.Functor.Identity reduces

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: patch Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D313 | -------------------------------------+------------------------------------- Comment (by hvr): Replying to [comment:5 ross]: the ifdefs. Sure, I can easily change that (at the cost of some minor code duplication)

Ideally I'd like to avoid a release with conditional exports - is there any way to detect a GHC bootstrap build?

As for this I'm not sure how that helps here. Assuming we have a way of setting a Cabal flag when we're bootstrapping GHC (i.e. building stage 1), how would this improve the situation of having the `Data.Functor.Identity` module conditionally exposed (which now would be depending on that Cabal flag setting rather than predicating on `impl(ghc>=7.9)`)? -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: patch Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D313 | -------------------------------------+------------------------------------- Comment (by hvr): @ross, I noticed you already reversed the Identity<->Class import-dependency in recent `transformers` commmits I've adapted http://git.haskell.org/packages/transformers.git/commitdiff/refs/heads/wip/T... which is now a rather minimal change to `transformers` (i.e. moving the `Identity.hs` file into a different source folder, and adapting `.cabal`) You write you "like to avoid a release with conditional exports". What problems do you anticipate that could occur by making use of such a conditional (non)export of `Data.Functor.Identity`? Specifically, I used `impl(ghc>=7.9)` instead of Cabal flags in order to avoid any Cabal-related problems I could think of. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: patch Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D313 | -------------------------------------+------------------------------------- Comment (by Herbert Valerio Riedel ): In [changeset:"87101364e0c2db5e472c6331ad35503028b2ec3c/ghc"]: {{{ #!CommitTicketReference repository="ghc" revision="87101364e0c2db5e472c6331ad35503028b2ec3c" Move Data.Functor.Identity from transformers to base This also updates the `transformers` submodule to the just released `transformers-0.4.2.0` package version. See #9664 for more details Reviewed By: austin, ekmett Differential Revision: https://phabricator.haskell.org/D313 }}} -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:10 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#9664: Add identity functor to base -------------------------------------+------------------------------------- Reporter: hvr | Owner: ekmett Type: task | Status: closed Priority: normal | Milestone: 7.10.1 Component: Core | Version: Libraries | Keywords: Resolution: fixed | Architecture: Unknown/Multiple Operating System: | Difficulty: Easy (less than 1 Unknown/Multiple | hour) Type of failure: | Blocked By: None/Unknown | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D313 | -------------------------------------+------------------------------------- Comment (by Herbert Valerio Riedel ): In [changeset:"8cbd25a49051171da7c73db57ebd87bb0296c2f7/ghc"]: {{{ #!CommitTicketReference repository="ghc" revision="8cbd25a49051171da7c73db57ebd87bb0296c2f7" Make Data.Functor.Identity trustworthy again Alas `{-# LANGUAGE Safe #-}` can't be used since `Data.Coerce` isn't "safe". However, we use `coerce` just as an optimisation (see also 4ba884bdd3a9521ea92fcda8f601a7d0f3537bc1 which broke the safe-inferred status of `Data.Functor.Identity`), so this module at least deserves `{-# LANGUAGE Trustworthy #-}`. NOTE: `Data.Functor.Identity` was added to `base` in the context of #9664 Reviewed By: luite Differential Revision: https://phabricator.haskell.org/D507 }}} -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9664#comment:12 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler