#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by edsko): We could probably work around this by adding a dummy class method in `Throws` and call that in `throwChecked`. (But I think there are other use cases of method-less classes out there in the wild :) -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:1 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by ekmett): Note: Suppressing only in the absence of superclasses means that any class constraint that just gives you a law will **always** spring this warning. {{{#!hs class Bifunctor p => Braided p where braid :: p a b -> p b a -- | @braid . braid = id@ class Braided p => Symmetric p swap :: Symmetric p => p a b -> p b a swap = braid }}} (This code is slightly simplified from http://hackage.haskell.org/package/categories-1.0.7/docs/Control-Category- Braided.html) One "solution" here is to move `swap` into the `Symmetric` class, but then you need to worry about it being overridden to be different than `braid`, which operationally can't happen in the current approach. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by ekmett): * cc: ekmett (added) -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:5 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by simonpj): Indeed, a pragma would be good. If someone is going to add class-specific pragmas, making `UndecidableSuperClasses` work on a per-class (rather than per-module) basis would be good. Simon -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:7 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by rwbarton): Demoting priority since there is an obvious workaround here: disable the warning in the module that defines `unthrow`. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:9 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by bgamari):

If someone is going to add class-specific pragmas, making `UndecidableSuperClasses` work on a per-class (rather than per-module) basis would be good.

What would this look like? Any class in the cycle being marked with the pragma would allow undecidable superclasses? If we are going to do this we may want to consider doing it before the 8.0 release. Moving from per-module to per-class modules after the per-class pragmas are in the wild may be rather painful (judging by the move from `OverlappingInstances` to `{-# OVERLAPPING #-}`, although perhaps `UndecidableSuperClasses` won't see as wide use). -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:11 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by simonpj): * owner: simonpj => -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:14 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * milestone: 8.0.2 => 8.2.1 Comment: Sadly this won't be happening for 8.0.2. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:16 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler

#11369: Suppress redundant-constraint warnings in case of empty classes -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler (Type | Version: 8.1 checker) | Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: #11370 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by simonpj): I think this is just a bug in the code that infers the context for the instance declaration that arises from GND. We should tet {{{ instance {- Empty context!! -} C (Identity a) where type T (Identity a) = T a }}} Where is that code? I can help with this after the PLDI deadline. Anyway, I think this is unrelated to the original topic of the ticket. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11369#comment:18 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler