April 2015 - ghc-tickets

[GHC] #10252: Abstract newtype in hs-boot
by GHC 07 Apr '15

07 Apr '15

#10252: Abstract newtype in hs-boot -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: feature | Status: new request | Milestone: Priority: normal | Version: 7.11 Component: Compiler | Operating System: Unknown/Multiple (Type checker) | Type of failure: None/Unknown Keywords: backpack | Blocked By: Architecture: | Related Tickets: Unknown/Multiple | Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Currently, the following syntax is not supported in an `hs-boot` file: {{{ module A where newtype F }}} This functionality would be pretty useful, since often the RHS of a newtype is intended to be abstract, and would be a pain to place in the boot file. This is also useful for Backpack. It's a bit hard for me to tell what this should do under the hood: we would have newtypes without the coercions. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10252> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #10188: prefix type-level cons can't be parsed
by GHC 07 Apr '15

07 Apr '15

#10188: prefix type-level cons can't be parsed -------------------------------------+------------------------------------- Reporter: Kinokkory | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.4 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Whereas other type operators (whether used as type families or used as data constructors) can be safely used in a prefix form, only the type- level cons of the built-in list kind can't be used in a prefix form. It really is a problem because we don't have any means to express type- level cons itself in source code. This for example prevents us from writing a generic type-level foldr that takes the built-in list and other similar data types alike. In a ghci: {{{#!hs > :set -XDataKinds -XTypeOperators > data Listy a = a ::: Listy a > :kind! (:::) (:::) :: k -> Listy k -> Listy k = forall (k :: BOX). (':::) > :kind! (:) <interactive>:1:2: parse error on input ‘:’ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10188> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #9964: GHC crash with NOINLINE and weird IO stuff
by GHC 07 Apr '15

07 Apr '15

#9964: GHC crash with NOINLINE and weird IO stuff -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- {{{#!hs {-# LANGUAGE UnboxedTuples #-} module Crash where import GHC.Base crash :: IO Int crash = IO (\s -> let {-# NOINLINE s' #-} s' = s in (# s', 1::Int #)) }}} This compiles under 7.6.3, but both 7.8.3 and head choke: {{{ ghc: panic! (the 'impossible' happened) (GHC version 7.8.3 for x86_64-unknown-linux): StgCmmEnv: variable not found s{v sLw} [lid] local binds for: Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9964> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 15

[GHC] #10214: parser: TransStmt has incorrect SrcSpan
by GHC 07 Apr '15

07 Apr '15

#10214: parser: TransStmt has incorrect SrcSpan -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 (Parser) | Operating System: Unknown/Multiple Keywords: | Type of failure: Other ApiAnnotations | Blocked By: Architecture: | Related Tickets: Unknown/Multiple | Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- The production for `squals` is. {{{#!hs squals :: { Located [LStmt RdrName (LHsExpr RdrName)] } -- In reverse order, because the last -- one can "grab" the earlier ones : squals ',' transformqual {% addAnnotation (gl $ last $ unLoc $1) AnnComma (gl $2) >> return (sLL $1 $> [L (getLoc $3) ((unLoc $3) (reverse (unLoc $1)))]) } }}} The location of the whole statement is set to that `transformqual` when it fact it should encompass the whole production. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10214> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #10207: parser: ParStmt has incorrect SrcSpan
by GHC 07 Apr '15

07 Apr '15

#10207: parser: ParStmt has incorrect SrcSpan -------------------------------------+------------------------------------- Reporter: alanz | Owner: alanz Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 (Parser) | Operating System: Unknown/Multiple Keywords: | Type of failure: Other ApiAnnotations | Blocked By: Architecture: | Related Tickets: Unknown/Multiple | Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- The production for pquals is {{{#!hs pquals :: { Located [[LStmt RdrName (LHsExpr RdrName)]] } : squals '|' pquals {% addAnnotation (gl $ last $ unLoc $1) AnnVbar (gl $2) >> return (L (getLoc $2) (reverse (unLoc $1) : unLoc $3)) } | squals { L (getLoc $1) [reverse (unLoc $1)] } }}} This assigns the location of the '|' to the entire returned value. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10207> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10253: Variable name with special characters
by GHC 07 Apr '15

07 Apr '15

#10253: Variable name with special characters -------------------------------------+------------------------------------- Reporter: songzh | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- Declaring a variable in GHCi with $ . is allowed, but they are not referable. I suppose that the parser should impose more restrictions at this. {{{#!hs *Main> let a$1 = 10 *Main> a$1 <interactive>:20:1: Not in scope: ‘a’ Prelude> let a.1 = 10 Prelude> a.1 <interactive>:9:1: Not in scope: ‘a’ Prelude> let a%1 = 1 Prelude> a%1 1 Prelude> let a^1 = 1 Prelude> a^1 1 Prelude> a^1 + 1 2 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10253> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #9958: System.IO.Error: Fix a documentation link to Control.Exception.Exception
by GHC 07 Apr '15

07 Apr '15

#9958: System.IO.Error: Fix a documentation link to Control.Exception.Exception -------------------------------------+------------------------------------- Reporter: mineo | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.8.4 libraries/base | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- On http://hackage.haskell.org/package/base-4.7.0.2/docs/System-IO- Error.html, the link to Control.Exception.Exception in the sentence "For a more general type of exception, including also those that arise in pure code, see Control.Exception.Exception." is broken because (in the source) it's a link to a module, but C.E.E is a type. Changing the double to single quotes fixes that. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9958> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #10197: Release note links to Trac are wrong
by GHC 06 Apr '15

06 Apr '15

#10197: Release note links to Trac are wrong -------------------------------------+------------------------------------- Reporter: Lemming | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.10.1 Documentation | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- In https://downloads.haskell.org/~ghc/7.10.1/docs/html/users_guide/release-7-1… the links to "issues" actually point to the Release Notes and not to Trac tickets. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10197> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9225: Missing syntax error for package qualified import with version number
by GHC 06 Apr '15

06 Apr '15

#9225: Missing syntax error for package qualified import with version number -------------------------+------------------------------------------------- Reporter: | Owner: ezyang ezyang | Status: new Type: bug | Milestone: Priority: | Version: 7.8.2 normal | Operating System: Unknown/Multiple Component: | Type of failure: Incorrect warning at Compiler (Parser) | compile-time Keywords: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- Consider the following test file: {{{ {-# LANGUAGE PackageImports #-} module Foo where import "containers-0.5.0.0" Data.Map }}} where we have {{{ ezyang@sabre:~$ ghc-pkg list | grep containers containers-0.5.0.0 unordered-containers-0.2.3.0 }}} GHC will fail to compile this, but in an unusual way: {{{ ezyang@sabre:~$ ghc test.hs [1 of 1] Compiling Foo ( test.hs, test.o ) test.hs:3:1: Failed to load interface for `Data.Map' It is not a module in the current program, or in any known package. }}} That's misleading: the real reason the import failed is because this is a package qualified import (not package ID qualified imports). For my purposes, it would be useful to also support package ID imports, but in the meantime, it would be a good idea to run Cabal's syntax check on the package name, which requires every hyphenated component to contain an alphabetic character. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9225> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #8057: PackageImports with versions for ghci
by GHC 06 Apr '15

06 Apr '15

#8057: PackageImports with versions for ghci ------------------------------------+------------------------------------- Reporter: Lemming | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Imports with explicit package names and versions would be nice to have in GHCi, because you cannot use the -package and -hide-package options from within GHCi. This is what currently happens: {{{ $ ghci-7.6.3 -XPackageImports GHCi, version 7.6.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Prelude> import Control.Monad.Trans.List <no location info>: Ambiguous module name `Control.Monad.Trans.List': it was found in multiple packages: transformers-0.3.0.0 List-0.5.1 Prelude> import "transformers-0.3.0.0" Control.Monad.Trans.List <no location info>: Could not find module `Control.Monad.Trans.List' It is not a module in the current program, or in any known package. Prelude> :q Leaving GHCi. $ ghci-7.6.3 -hide-package List GHCi, version 7.6.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Prelude> import Control.Monad.Trans.List Prelude Control.Monad.Trans.List> }}} However, I now see that the import with explicit package name works, if I omit the package version. I think GHCi should at least warn that providing a version does not work. See also #2362 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8057> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4