March 2018 - ghc-tickets

[GHC] #14442: InstanceSigs fails
by GHC 26 May '18

26 May '18

#14442: InstanceSigs fails -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.3 Keywords: InstanceSigs | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This may be expected behavior, resulting from instance signatures being allowed to be "more polymorphic" than the method `@Bool`. But better safe than sorry (can't think of a better title). This works fine: {{{#!hs {-# Language KindSignatures, GADTs, DataKinds, TypeOperators, PolyKinds, TypeFamilies, TypeFamilyDependencies, InstanceSigs #-} import Data.Kind import Data.Type.Equality type family Sing = (res :: k -> Type) | res -> k type instance Sing = SBool data SBool :: Bool -> Type where SFalse :: SBool False STrue :: SBool True class EQ a where eq :: Sing (x::a) -> Sing (y::a) -> Maybe (x :~~: y) instance EQ Bool where eq :: Sing (x :: Bool) -> Sing (y :: Bool) -> Maybe (x :~~: y) eq SFalse SFalse = Just HRefl }}} Removing the kind annotation from `x :: Bool` and/or `y :: Bool` causes it to fail: Here I have removed it from `x`: `eq :: Sing x -> Sing (y :: Bool) -> Maybe (x :~~: y)` {{{ $ ghci -ignore-dot-ghci B.hs GHCi, version 8.3.20170920: http://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( B.hs, interpreted ) B.hs:19:6: error: • Couldn't match type ‘SBool’ with ‘Sing’ Expected type: Sing x Actual type: SBool a0 • In the pattern: SFalse In an equation for ‘eq’: eq SFalse SFalse = Just HRefl In the instance declaration for ‘EQ Bool’ • Relevant bindings include eq :: Sing x -> Sing y -> Maybe (x :~~: y) (bound at B.hs:19:3) | 19 | eq SFalse SFalse = Just HRefl | ^^^^^^ B.hs:19:22: error: • Could not deduce: (x :: k1) ~~ ('False :: Bool) from the context: a0 ~ 'False bound by a pattern with constructor: SFalse :: SBool 'False, in an equation for ‘eq’ at B.hs:19:6-11 or from: y ~ 'False bound by a pattern with constructor: SFalse :: SBool 'False, in an equation for ‘eq’ at B.hs:19:13-18 ‘x’ is a rigid type variable bound by the type signature for: eq :: forall k1 (x :: k1) (y :: Bool). Sing x -> Sing y -> Maybe (x :~~: y) at B.hs:18:9-54 Expected type: Maybe (x :~~: y) Actual type: Maybe (x :~~: x) • In the expression: Just HRefl In an equation for ‘eq’: eq SFalse SFalse = Just HRefl In the instance declaration for ‘EQ Bool’ • Relevant bindings include eq :: Sing x -> Sing y -> Maybe (x :~~: y) (bound at B.hs:19:3) | 19 | eq SFalse SFalse = Just HRefl | ^^^^^^^^^^ Failed, 0 modules loaded. Prelude> }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14442> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #14246: Probably AllowAmbiguousTypes + UndecidableInstances + TypeInType
by GHC 26 May '18

26 May '18

#14246: Probably AllowAmbiguousTypes + UndecidableInstances + TypeInType -------------------------------------+------------------------------------- Reporter: Toricon | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 Keywords: | Operating System: Windows Architecture: x86_64 | Type of failure: Compile-time (amd64) | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- ghc.exe: panic! (the 'impossible' happened) (GHC version 8.0.2 for x86_64-unknown-mingw32): isInjectiveTyCon sees a TcTyCon KLN {{{#!hs {-# LANGUAGE RankNTypes, GADTs, TypeOperators, PolyKinds, DataKinds, TypeFamilies, AllowAmbiguousTypes, UndecidableInstances, TypeInType #-} module MCS.Function where import MCS.Core --import MCS.Value import Data.Kind -- necessary for * type family KLN (n :: k) :: Nat where KLN (f :: v -> k) = S (KLN (forall t. f t)) KLN (f :: *) = Z type family Reveal (n :: k) (l :: Vect (KLN n) L) :: * where Reveal (f :: v -> k) (Cons (Label (t :: v)) l) = Reveal (f t) l Reveal (a :: *) Nil = a }}} The problem *has* to be somewhere in the KLN / Reveal interaction, so I removed everything else for this. Every other weird area is commented out or unused, and the bug vanished when KLN / Reveal was commented out. KLN shouldn't even work. If it _could_ be interaction with the rest of the document, I'll put that back in. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14246> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #14172: GHC hangs during type-checking
by GHC 26 May '18

26 May '18

#14172: GHC hangs during type-checking -------------------------------------+------------------------------------- Reporter: lightandlight | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 Keywords: | Operating System: Linux Architecture: x86_64 | Type of failure: Compile-time (amd64) | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- GHC 8.0.2 doesn't terminate when type checking this file: {{{#!hs import Control.Lens import Data.Functor.Compose traverseCompose :: (a -> f b) -> g a -> f (h _) traverseCompose = _Wrapping Compose . traverse }}} This is the smallest type signature I could find that still causes the problem. GHCi infers the type without issue, and when I annotate it with the correct type signature, the file is successfully compiled. Using base-4.9.1.0, lens-4.14.1, and transformers-0.5.2.0 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14172> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #13746: Typeable changes in base-4.10 break ChasingBottoms's test suite
by GHC 26 May '18

26 May '18

#13746: Typeable changes in base-4.10 break ChasingBottoms's test suite -------------------------------------+------------------------------------- Reporter: refold | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 8.2.1-rc2 libraries/base | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Runtime crash Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- When compiled against base-4.10, `ChasingBottoms`'s test suite fails with: {{{ Approx: ChasingBottomsTestSuite: typeRepTyCon: FunTy CallStack (from HasCallStack): error, called at libraries/base/Data/Typeable/Internal.hs:308:33 in base:Data.Typeable.Internal }}} This is due to the use of `error` in the following function in [https://github.com/ghc/ghc/blob/master/libraries/base/Data/Typeable/Interna… `Data.Typeable.Internal`]: {{{#!hs -- | Observe the type constructor of a type representation typeRepTyCon :: TypeRep a -> TyCon typeRepTyCon (TrTyCon _ tc _) = tc typeRepTyCon (TrApp _ a _) = typeRepTyCon a typeRepTyCon (TrFun _ _ _) = error "typeRepTyCon: FunTy" -- TODO }}} `ChasingBottoms`'s test suite works fine when compiled with GHC 8.0.2. It seems to me that the change in behaviour constitutes a bug or a breaking change (for which there is no mention in the changelog). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13746> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #13726: Declaring GADT constructor and associated data family with the same name gives weird error
by GHC 26 May '18

26 May '18

#13726: Declaring GADT constructor and associated data family with the same name gives weird error -------------------------------------+------------------------------------- Reporter: mrkgnao | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The code this is from is really messy. I'll submit a minimal example in a short while if this isn't a known bug. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13726> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #13192: Ambiguity Caused By PolyKind and Not Helpful Error Messages
by GHC 26 May '18

26 May '18

#13192: Ambiguity Caused By PolyKind and Not Helpful Error Messages -------------------------------------+------------------------------------- Reporter: Shayan-Najd | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following code results in a confusing/wrong error message, blaming a type variable (`a`) being a fixed Skolem variable: {{{#!hs {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE PolyKinds #-} data I a type family F x type instance F (I a) = a identity :: F (I a) -> F (I a) identity x = x }}} with the error message {{{ • Couldn't match type ‘F (I a0)’ with ‘F (I a)’ Expected type: F (I a) -> F (I a) Actual type: F (I a0) -> F (I a0) NB: ‘F’ is a type function, and may not be injective The type variable ‘a0’ is ambiguous • In the ambiguity check for ‘identity’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: identity :: F (I a) -> F (I a) | 9 | identity :: F (I a) -> F (I a) | ^^^^^^^^^^^^^^^^^^ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13192> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #13118: let binding tuple of lenses error not an expression
by GHC 26 May '18

26 May '18

#13118: let binding tuple of lenses error not an expression -------------------------------------+------------------------------------- Reporter: codygman | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHCi | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This error seems to be fixed in GHC Head +/- 10 commits, but exists in GHC 8.0.1. I was told it might be a useful test case. {{{ λ> :t userId userId :: (RElem UserId rs (RIndex UserId rs), Functor f) => (Int -> f Int) -> Record rs -> f (Record rs) λ> let recLenses = userId &: Nil λ> :t recLenses recLenses :: (Functor f, RElem UserId rs (RIndex UserId rs)) => Record '[s :-> ((Int -> f Int) -> Record rs -> f (Record rs))] λ> :t recUncons recUncons :: Record (s :-> a : rs) -> (a, Record rs) λ> :t recUncons recLenses recUncons recLenses :: (Functor f, RElem UserId rs (RIndex UserId rs)) => ((Int -> f Int) -> Record rs -> f (Record rs), Record '[]) λ> let (l, ls) = recUncons recLenses <no location info>: error: not an expression: ‘let (l, ls) = recUncons recLenses’ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13118> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #13755: GHC-8.0.2+ spits out $dm names in error messages
by GHC 26 May '18

26 May '18

#13755: GHC-8.0.2+ spits out $dm names in error messages -------------------------------------+------------------------------------- Reporter: zilinc | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Poor/confusing Unknown/Multiple | error message Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Example (copied from #12881) {{{ {-# LANGUAGE FlexibleInstances #-} module Bug where class Arbitrary a where shrink :: a -> [a] shrink _ = [] instance Arbitrary a instance Arbitrary Int }}} Error message is: {{{ GHCInstance.hs:10:10: error: • Overlapping instances for Foo Int arising from a use of ‘Main.$dmfoo’ Matching instances: instance Foo a -- Defined at GHCInstance.hs:9:10 instance Foo Int -- Defined at GHCInstance.hs:10:10 • In the expression: Main.$dmfoo @Int In an equation for ‘foo’: foo = Main.$dmfoo @Int In the instance declaration for ‘Foo Int’ }}} Another example: code in #13754 description. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13755> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #12854: ghc-8 prints mangled names in error message: ‘GHC.Base.$dm<$’
by GHC 26 May '18

26 May '18

#12854: ghc-8 prints mangled names in error message: ‘GHC.Base.$dm<$’ -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Noticed on old omega codebase when built with ghc-8.0.2-rc1. Here is the minimal example: {{{#!hs {-# LANGUAGE FlexibleInstances, KindSignatures #-} {-# OPTIONS_GHC -Wall #-} data F a b instance Functor (F a) where fmap = undefined instance Functor (F String) where fmap = undefined }}} {{{ $ ghci a.hs GHCi, version 8.0.1.20161117: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/slyfox/.ghci [1 of 1] Compiling Main ( a.hs, interpreted ) a.hs:6:10: error: • Overlapping instances for Functor (F String) arising from a use of ‘GHC.Base.$dm<$’ Matching instances: instance Functor (F a) -- Defined at a.hs:5:10 instance Functor (F String) -- Defined at a.hs:6:10 • In the expression: GHC.Base.$dm<$ @F String In an equation for ‘<$’: (<$) = GHC.Base.$dm<$ @F String In the instance declaration for ‘Functor (F String)’ Failed, modules loaded: none. }}} ‘GHC.Base.$dm<$’ is an internal name. Looks like it should be a 'GHC.Base.<$'. Confusingly this does not cause overlapping error at all: {{{#!hs {-# LANGUAGE FlexibleInstances, KindSignatures #-} {-# OPTIONS_GHC -Wall #-} data F a b class C (f :: * -> *) where instance C (F a) where instance C (F String) where }}} {{{ $ ghci a.hs GHCi, version 8.0.1.20161117: http://www.haskell.org/ghc/ :? for help Loaded GHCi configuration from /home/slyfox/.ghci [1 of 1] Compiling Main ( a.hs, interpreted ) Ok, modules loaded: Main. }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12854> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #12794: Out of scope error not reported
by GHC 26 May '18

26 May '18

#12794: Out of scope error not reported -------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Consider {{{ {-# LANGUAGE TypeApplications #-} {-# OPTIONS -fdefer-type-errors #-} f = bar @Int }}} where `bar` is not in scope. We get {{{ T12768.hs:8:7: error: • Cannot apply expression of type ‘t1’ to a visible type argument ‘Int’ • In the expression: bar @Int }}} Confusingly, the "`bar` is not in scope`" error (the true error) has been suppressed. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12794> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2