- ghc-tickets - Haskell.org

[GHC] #9273: TypeNats and record syntax don't compile
by GHC 06 Jul '14

06 Jul '14

#9273: TypeNats and record syntax don't compile -------------------------------------+------------------------------------- Reporter: br1 | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler (Type | Version: 7.8.2 checker) | Operating System: Unknown/Multiple Keywords: | Type of failure: GHC rejects Architecture: Unknown/Multiple | valid program Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- {{{ #!haskell data SNat :: Nat -> * where SNatZ :: SNat 0 SNatS :: {sNatPred :: SNat n} -> SNat (n+1) }}} gives error {{{ rec_gadt_nat.hs:13:13: Could not deduce (n1 ~ n) from the context ((n + 1) ~ (n1 + 1)) bound by a pattern with constructor SNatS :: forall (n :: Nat). SNat n -> SNat (n + 1), in an equation for ‘sNatPred’ at rec_gadt_nat.hs:13:13-20 ‘n1’ is a rigid type variable bound by a pattern with constructor SNatS :: forall (n :: Nat). SNat n -> SNat (n + 1), in an equation for ‘sNatPred’ at rec_gadt_nat.hs:13:13 ‘n’ is a rigid type variable bound by the type signature for sNatPred :: SNat (n + 1) -> SNat n at rec_gadt_nat.hs:13:13 Expected type: SNat n Actual type: SNat n1 Relevant bindings include sNatPred :: SNat n1 (bound at rec_gadt_nat.hs:13:13) sNatPred :: SNat (n + 1) -> SNat n (bound at rec_gadt_nat.hs:13:13) In the expression: sNatPred In an equation for ‘sNatPred’: sNatPred (SNatS {sNatPred = sNatPred}) = sNatPred }}} while {{{ #!haskell data SNat :: Nat -> * where SNatZ :: SNat 0 SNatS :: SNat n -> SNat (n+1) }}} works. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9273> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

Re: [GHC] #3178: Fix linking -lpthread for semaphores
by GHC 04 Jul '14

04 Jul '14

#3178: Fix linking -lpthread for semaphores -----------------------------------+------------------------------------ Reporter: sthibaul | Owner: Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: libraries/unix | Version: 6.10.2 Resolution: fixed | Keywords: Operating System: Other | Architecture: Unknown/Multiple Type of failure: None/Unknown | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: #4523 -----------------------------------+------------------------------------ Changes (by thomie): * status: new => closed * resolution: => fixed * related: => #4523 Comment: This issue should be fixed together with ticket #4523, commit [http://git.haskell.org/packages/unix.git/commit/dd0178b4803924d832622436261… dd0178b]. Proper autoconf test for sem_close's library; fixes trac #4523 Note: the current issue tracker for the `unix` library is at https://github.com/haskell/unix/issues -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3178#comment:7> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #6018: Injective type families
by GHC 04 Jul '14

04 Jul '14

#6018: Injective type families -------------------------------+------------------------------------------- Reporter: lunaris | Owner: simonpj Type: feature | Status: new request | Milestone: 7.10.1 Priority: normal | Version: 7.4.1 Component: Compiler | Keywords: TypeFamilies, Injective Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Unknown Unknown/Multiple | Blocked By: Type of failure: | Related Tickets: #4259 None/Unknown | Test Case: | Blocking: | -------------------------------+------------------------------------------- Changes (by Artyom.Kazak): * cc: Artyom.Kazak@… (removed) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/6018#comment:35> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #6018: Injective type families
by GHC 04 Jul '14

04 Jul '14

#6018: Injective type families -------------------------------+------------------------------------------- Reporter: lunaris | Owner: simonpj Type: feature | Status: new request | Milestone: 7.10.1 Priority: normal | Version: 7.4.1 Component: Compiler | Keywords: TypeFamilies, Injective Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Unknown Unknown/Multiple | Blocked By: Type of failure: | Related Tickets: #4259 None/Unknown | Test Case: | Blocking: | -------------------------------+------------------------------------------- Comment (by goldfire): I thinking about this ticket, I've imagined some syntax like this: {{{ type family FullyInjective a | Result -> a -- use a fun-dep-like syntax, where "Result" is a keyword, capitalized to avoid name-clashes with type variables type family Plus a b | Result a -> b, Result b -> a type family Only1stInjective a b | Result -> a }}} I could even imagine having dependencies among the arguments, though that might be a little strange. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/6018#comment:34> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #4259: Relax restrictions on type family instance overlap
by GHC 04 Jul '14

04 Jul '14

#4259: Relax restrictions on type family instance overlap --------------------------------------------+------------------------------ Reporter: lilac | Owner: Type: feature request | Status: new Priority: normal | Milestone: 7.10.1 Component: Compiler (Type checker) | Version: 6.12.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: None/Unknown | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: | Blocked By: | Related Tickets: #8423 --------------------------------------------+------------------------------ Changes (by jstolarek): * cc: jan.stolarek@… (added) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4259#comment:32> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #3569: ghci can't handle utf-8 chinese char correctly when modify.
by GHC 04 Jul '14

04 Jul '14

#3569: ghci can't handle utf-8 chinese char correctly when modify. -------------------------------------+---------------------------------- Reporter: guest | Owner: judah Type: bug | Status: closed Priority: normal | Milestone: ⊥ Component: GHCi | Version: 6.10.4 Resolution: fixed | Keywords: Operating System: Unknown/Multiple | Architecture: x86_64 (amd64) Type of failure: None/Unknown | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: -------------------------------------+---------------------------------- Changes (by thomie): * cc: hvr (added) * status: new => closed * resolution: => fixed Comment: Haskeline ticket [http://trac.haskell.org/haskeline/ticket/81 81], an my own testing, leads me to believe this issue has been fixed, at least on Linux. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3569#comment:4> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #6018: Injective type families
by GHC 04 Jul '14

04 Jul '14

#6018: Injective type families -------------------------------+------------------------------------------- Reporter: lunaris | Owner: simonpj Type: feature | Status: new request | Milestone: 7.10.1 Priority: normal | Version: 7.4.1 Component: Compiler | Keywords: TypeFamilies, Injective Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Unknown Unknown/Multiple | Blocked By: Type of failure: | Related Tickets: #4259 None/Unknown | Test Case: | Blocking: | -------------------------------+------------------------------------------- Changes (by jstolarek): * cc: jan.stolarek@… (added) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/6018#comment:33> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #6018: Injective type families
by GHC 04 Jul '14

04 Jul '14

#6018: Injective type families -------------------------------+------------------------------------------- Reporter: lunaris | Owner: simonpj Type: feature | Status: new request | Milestone: 7.10.1 Priority: normal | Version: 7.4.1 Component: Compiler | Keywords: TypeFamilies, Injective Resolution: | Architecture: Unknown/Multiple Operating System: | Difficulty: Unknown Unknown/Multiple | Blocked By: Type of failure: | Related Tickets: #4259 None/Unknown | Test Case: | Blocking: | -------------------------------+------------------------------------------- Comment (by jstolarek): I spent some time recently thinking about this one. It is not clear to me what should happen when we know that a type family `F` is injective. It would certainly allow us to infer from `(F a ~ F b)` that `a ~ b` (modulo corner cases). So instead of first reducing injective type family application and then working on the result we could instead reason about type family's arguments. We could even invert type families. Is there anything else we could reason about with injective type families? The discussion here is only about open type families (obviously, closed type families were not implemented two years ago). I guess that now this change would affect both open and closed type families? Replying to [comment:2 simonpj]: > Some functions might be injective in one argument but not another. For example: > {{{ > F a b ~ F c d ===> a ~ c > but not b ~ d > }}} That is not true according to the mathematical definition of injectivity (as stated in [comment:29 comment 29]). Unless we want to introduce definition of type families injective only in some of the arguments, [http://haskell.1045720.n5.nabble.com/Injective-type-families- tp3385084p3385684.html as proposed by Iavor]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/6018#comment:32> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #9175: Bad interaction between Pattern Synonyms and Text
by GHC 03 Jul '14

03 Jul '14

#9175: Bad interaction between Pattern Synonyms and Text -----------------------------------+--------------------------------------- Reporter: emertens | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time crash Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -----------------------------------+--------------------------------------- Compiling this file causes the impossible to happen. It is necessary to enable optimizations. The bug did not occur if the Text parameter was replaced with ByteString or String. {{{ {-# LANGUAGE PatternSynonyms #-} {-# LANGUAGE OverloadedStrings #-} module Bug where {- - Must compile with optimizations - - text-1.1.1.3 - ghc-7.8.2 - Both Darwin and Linux platforms - -} import Data.Text(Text) data T = C Text Bool f :: a -> b f _ = undefined {-# NOINLINE f #-} -- important pattern P1 a = C "sh" a -- at least two characters pattern P2 = C "x" True -- this pattern has to come last g :: Text -> T g x = case x of "" -> f (P1 undefined) -- this has to be a pattern synonym }}} {{{ [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) ghc: panic! (the 'impossible' happened) (GHC version 7.8.2 for x86_64-apple-darwin): StgCmmEnv: variable not found lvl_s2Jr{v} [lid] local binds for: main:Bug.f{v rBS} [gid] main:Bug.rewriteRule{v rBV} [gid] main:Bug.$mP2{v r17q} [gid] main:Bug.$WP2{v r17A} [gid] main:Bug.$mP1{v r17E} [gid] main:Bug.$WP1{v r17R} [gid] main:Bug.$WP3{v r3bq} [gid] main:Bug.$WP4{v r3ua} [gid] main:Bug.$WP5{v r3ub} [gid] main:Bug.$WP2_dt{v r3uc} [gid] main:Bug.$WP6{v r3ud} [gid] main:Bug.$WP7{v r3ue} [gid] main:Bug.$WP8{v r3uf} [gid] main:Bug.$w$mP1{v r3ug} [gid] main:Bug.$w$mP2{v r3uh} [gid] main:Bug.$mP3{v r3ui} [gid] main:Bug.$mP4{v r3uj} [gid] main:Bug.$mP5{v r3uk} [gid] main:Bug.$wrewriteRule{v r3ul} [gid] main:Bug.rewriteRule1{v r3um} [gid] main:Bug.rewriteRule2{v r3un} [gid] main:Bug.rewriteRule3{v r3uo} [gid] main:Bug.rewriteRule4{v r3up} [gid] main:Bug.rewriteRule5{v r3uq} [gid] lvl{v r3ur} [gid] Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9175> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #9023: Error when using empty record update on binary pattern synonym
by GHC 03 Jul '14

03 Jul '14

#9023: Error when using empty record update on binary pattern synonym --------------------------------+--------------------------------- Reporter: Iceland_jack | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Linux Architecture: x86 | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | --------------------------------+--------------------------------- Using an empty record update with patterns works here {{{ pattern Singleton a = [a] isSingleton :: [a] -> Bool isSingleton Singleton{} = True isSingleton _ = False }}} but fails in the follow example (there may exist a more minimal example) {{{ pattern P a b = Just (a, b) foo P{} = True }}} which outputs (GHC 7.8.2) {{{ $ ghci -ignore-dot-ghci tmp.g2WZtTUkhJ.hs GHCi, version 7.8.2: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling Main ( tmp.g2WZtTUkhJ.hs, interpreted ) Var/Type length mismatch: [t{tv aIW} [sk]] [] Ok, modules loaded: Main. *Main> }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9023> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17