- Glasgow-haskell-users

lambda calculus theory
by Hans N Beck 08 Nov '05

08 Nov '05

Hi, I'm searching for a good mathematical oriented introduction to the theory of lambda calculus or other theoretical foundations of Lisp/ Haskell, i.e. monads or such (of course in the web there are much hints, but what is the best for mathematicans foreign to this field) Regards Hans

3 3

RE: -ferror-spans and ghci
by Simon Marlow 08 Nov '05

08 Nov '05

On 07 November 2005 05:01, Ian Zimmerman wrote: > itz@unicorn:~/src/mindent$ ghci -ferror-spans > ___ ___ _ > / _ \ /\ /\/ __(_) > / /_\// /_/ / / | | GHC Interactive, version 6.4, for Haskell > 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ > \____/\/ /_/\____/|_| Type :? for help. > > Loading package base-1.0 ... linking ... done. > Prelude> :a SimpleHaskell > Compiling Main ( SimpleHaskell.hs, interpreted ) > > SimpleHaskell.hs:3:0: parse error (possibly incorrect indentation) > Failed, modules loaded: none. > Prelude> Leaving GHCi. > > wth? > > It is in ghci where this option would be most useful, imho. Is it > implemented? Yes, it is implemented. Perhaps there was no reasonable span that GHC could determine for that particular error - can you show us the source file? Does 'ghc -ferror-spans' give a more detailed error? Cheers, Simon

2 1

class aliases
by John Meacham 08 Nov '05

08 Nov '05

If I were to implement my class alias proposal for ghc in an acceptably clean way, would it be accepted? There appeared to be enough interest to support at least experimentation with it and I would have no issues with it being dropped in the future if it turned out to not be useful. (emotional attachment to code is not good practice :) ) John -- John Meacham - ⑆repetae.net⑆john⑈

1 0

-ferror-spans and ghci
by Ian Zimmerman 07 Nov '05

07 Nov '05

itz@unicorn:~/src/mindent$ ghci -ferror-spans ___ ___ _ / _ \ /\ /\/ __(_) / /_\// /_/ / / | | GHC Interactive, version 6.4, for Haskell 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ \____/\/ /_/\____/|_| Type :? for help. Loading package base-1.0 ... linking ... done. Prelude> :a SimpleHaskell Compiling Main ( SimpleHaskell.hs, interpreted ) SimpleHaskell.hs:3:0: parse error (possibly incorrect indentation) Failed, modules loaded: none. Prelude> Leaving GHCi. wth? It is in ghci where this option would be most useful, imho. Is it implemented? -- A true pessimist won't be discouraged by a little success.

1 0

INSTANCE_TYPEABLEx now missing in GHC
by Vadim Konovalov 06 Nov '05

06 Nov '05

Dear all, I installed latest GHC version 6.5.20051102, and then, when compiling 'hs-plugins', I got an error about INSTANCE_TYPEABLE. Those were previously defined in 'Typeable.h' file, and copying its definition from older version makes things work. Is that change intentional? Best regards, Vadim.

2 1

HSQL installation woes
by Ed Osinski 06 Nov '05

06 Nov '05

Hi, I'm trying to install GHC and related software under Solaris 9 and am stuck at HSQL. I initially used the binary package of ghc 6.4.1. It installed w/o any problems, and I was able to build alex, happy, haddock and cpphs. However, when trying to install HSQL-1.6, it complained about "ld" not recognizing "-x". I guess this is due to the binary package for Solaris being built on a machine that used the GNU linker, which recognizes "-x"? Anyway, this led me to build ghc 6.4.1 from source, with gcc 4.0.2, which was successful. :-) However, when I try to install HSQL-1.6, it dies during the build step. :-( Here is the output: /sware/build/HSQL-1.6/HSQL |> runghc Setup.lhs configure Warning: No license-file field. Configuring hsql-1.6... configure: searching for ghc in path. configure: found ghc at /usr/local/bin/ghc configure: looking for package tool: ghc-pkg near compiler in /usr/local/bin/ghc configure: found package tool in /usr/local/bin/ghc-pkg configure: Using install prefix: /usr/local configure: Using compiler: /usr/local/bin/ghc configure: Compiler flavor: GHC configure: Compiler version: 6.4.1 configure: Using package tool: /usr/local/bin/ghc-pkg configure: Using haddock: /usr/local/bin/haddock configure: Using happy: /usr/local/bin/happy configure: Using alex: /usr/local/bin/alex configure: Using hsc2hs: /usr/local/bin/hsc2hs configure: Using cpphs: /usr/local/bin/cpphs configure: Reading installed packages... configure: Dependency base-any: using base-1.0 /sware/build/HSQL-1.6/HSQL |> runghc Setup.lhs build Preprocessing library hsql-1.6... Building hsql-1.6... Chasing modules from: Database.HSQL,Database.HSQL.Types Compiling Database.HSQL.Types ( ./Database/HSQL/Types.hs, dist/build/./Database/HSQL/Types.o ) Compiling Database.HSQL ( ./Database/HSQL.hs, dist/build/./Database/HSQL.o ) *** Exception: waitForProcess: interrupted (Interrupted system call) This error happens consistently, but sometimes one line earlier. Googling reveals that at least one person encountered this exception on Solaris in the recent past: http://www.mail-archive.com/glasgow-haskell-users@haskell.org/msg07794.html Is this a showstopper, or is there some fix or workaround? - Ed

1 0

[ ghc-Feature Requests-1349036 ] Compiling multiple executables with -make
by SourceForge.net 05 Nov '05

05 Nov '05

Feature Requests item #1349036, was opened at 2005-11-05 15:20 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358032&aid=1349036&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Priority: 5 Submitted By: Mads Lindstrøm (supermule) Assigned to: Nobody/Anonymous (nobody) Summary: Compiling multiple executables with -make Initial Comment: Hi I am very happy with the -make option. However, I think it could be improved even further. I would like ghc to be able to compile multiple executables with one invocation of ghc, using the -make option. Compiling multiple executables in one invokation could safe a lot of compile-time, when the executables shares most of their source files. I believe the time would be saved, as it would remove the need to check all source files for possible recompilation for each executable. Furthermore, we would also only need to build the dependency graph once. /Mads Lindstrøm ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358032&aid=1349036&group_…

1 0

RE: Silly IO problem
by Simon Marlow 04 Nov '05

04 Nov '05

On 03 November 2005 12:57, skaller wrote: > On Thu, 2005-11-03 at 09:49 +0100, Mirko Rahn wrote: >>> It prints n, rather than tak(3n,2n,n). >> >> ??? No: >> >> *Main> :set args 0 >> *Main> main >> 0.0 >> *Main> :set args 1 >> *Main> main >> 2.0 >> *Main> :set args 2 >> *Main> main >> 3.0 >> *Main> :set args 3 >> *Main> main >> 6.0 >> >> The same with some versions of ghc. > > Ah .. it seems to be a bug in gcc 4.0 > > The Glorious Glasgow Haskell Compilation System, version 6.4 > [running on Ubuntu which uses Debian package] > > gcc (GCC) 4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9) > x86_64 > > Summary: gcc does not work with GHC 6.4 on x86_64 with > optimisation higher than -O1 > > Perhaps the Evil Mangler is not working, or, possibly > the version isn't even registered. Thanks for the report, I'll look into gcc 4.x before we release 6.4.2. Cheers, Simon

1 0

RE: Rules for use of unsafeThaw...
by Simon Marlow 04 Nov '05

04 Nov '05

On 03 November 2005 17:08, Jan-Willem Maessen wrote: > I've recently been experimenting with unsafeFreeze/unsafeThaw in > GHC. Judicious use of these functions vastly reduces GC overhead in > Data.HashTable. However, a slightly mis-timed GC will cause the > whole mess to crash. I am attempting to understand the invariants > required to safely use unsafeFreeze/unsafeThaw. > > I believe the following usage ought to be 100% safe: > > 1) Take the last and only reference to a mutable array, and call > unsafeFreeze to obtain an immutable array. > > 2) Take the last and only reference to an immutable array, and call > unsafeThaw to obtain a mutable array. I would like it to be the case that no usage of unsafe{Freeze,Thaw} can crash the GC. That perhaps isn't true right now, but I'd like to understand why not. Multiple unsafeThaws on the same array might cause problems in 6.4.1 (but not the HEAD), due to the way the old generation mutable list works. That's fixable, if it is the problem. Multiple unsafeFreezes should be harmless. Can you give me a test program? Cheers, Simon

1 0

Rules for use of unsafeThaw...
by Jan-Willem Maessen 04 Nov '05

04 Nov '05

I've recently been experimenting with unsafeFreeze/unsafeThaw in GHC. Judicious use of these functions vastly reduces GC overhead in Data.HashTable. However, a slightly mis-timed GC will cause the whole mess to crash. I am attempting to understand the invariants required to safely use unsafeFreeze/unsafeThaw. I believe the following usage ought to be 100% safe: 1) Take the last and only reference to a mutable array, and call unsafeFreeze to obtain an immutable array. 2) Take the last and only reference to an immutable array, and call unsafeThaw to obtain a mutable array. Are there any other usage modes which are GC-safe? In particular, is it possible for a reference to the frozen array to exist somewhere on the heap when a call to unsafeThaw occurs? Is there a particular set of circumstances under which we might get away with this (eg, no allocation during the interval where the ambiguous references occur)? -Jan-Willem Maessen

2 2