Re: IRIX 5.3 and import qualified Prelude
Graeme Moss writes: > > GHC fails on the following script: > > >>> > import qualified Prelude > > main :: Prelude.IO () > main = Prelude.print Prelude.True > <<< > > with > > >>> > importPrel.hs:3: Type constructor or class not in scope: `()' > Yep, a renaming problem. For now, play games such as import qualified Prelude hiding ( () ) import Prelude ( () ) Thanks, --Sigbjorn
Re: Importing symbolic constructors
Graeme Moss writes: > (Apologies if you're aware of/fixed this problem, I have GHC 2.05 and > haven't kept up with bug fix news.) > > Exporting a constructor like (:..:) seems to cause an interface file > that fails to be parsed. Hand-editing out the offending portion that > looks something like: > Sigh, fixed in the next release. For now, use something other than '.' as the second char in your constructor. Thanks for the report, --Sigbjorn
Importing symbolic constructors
(Apologies if you're aware of/fixed this problem, I have GHC 2.05 and haven't kept up with bug fix news.) Exporting a constructor like (:..:) seems to cause an interface file that fails to be parsed. Hand-editing out the offending portion that looks something like: >>> _exports_ Auburn_CoreGen core guardResultBool guardResultList makeSig IntSubset(All :..: FiniteSet None) Op Phase(Phase) Profile(Profile) Role(Generator Mutator Observer) ShadowClass ShadowInfoClass Sig; <<< to remove the constructor: >>> _exports_ Auburn_CoreGen core guardResultBool guardResultList makeSig IntSubset(All FiniteSet None) Op Phase(Phase) Profile(Profile) Role(Generator Mutator Observer) ShadowClass ShadowInfoClass Sig; <<< lets the interface-file be parsed, but of course the missing constructor now causes the expected problem. Putting parantheses around the constructor in the interface-file didn't work either. -=GEM=-
IRIX 5.3 and import qualified Prelude
The Good News - I've tried GHC for IRIX 6.2 on my IRIX 5.3 machine and all seems well so far, hurrah! (Thanks go to Fuad Abdallah for pointing this possibility out and the work I know he did in helping to bring GHC 2.xx to IRIX.) The version I used was obtained at: ftp://ftp.dcs.gla.ac.uk/pub/haskell/glasgow/working/ghc-2.05-prof-mips-sgi-irix6.2.tar.gz The Bad News GHC fails on the following script: >>> import qualified Prelude main :: Prelude.IO () main = Prelude.print Prelude.True <<< with >>> importPrel.hs:3: Type constructor or class not in scope: `()' Compilation had errors <<< Thanks again to Fuad and those who helped put the IRIX 6.2 version together, it's been a long time coming but I finally have access to a GHC for Haskell 1.3/1.4 and it's worth the wait! Graeme.
Re: 2.07 binary installation
This suggests to me that you for some reason don't have a #! added to the top of the driver. To make this a little bit easier to track down, could you reply with the contents of config.cache and Makefile? Thx, --Sigbjorn Justin Cormack writes: > > > > > > There's a tpyo in Makefile.in, change PACKAGE_LIBSCRIPTS to > > PACKAGE_LIB_SCRIPTS. > > > > --Sigbjorn > > > > ok, so far so good, install was happy, then testing gives: > > /homes/jpc1/bin/ghc: =2-space GC: not found > /homes/jpc1/bin/ghc: =dual-mode GC: not found > /homes/jpc1/bin/ghc: =: not found
Re: 2.07 binary installation
> > > There's a tpyo in Makefile.in, change PACKAGE_LIBSCRIPTS to > PACKAGE_LIB_SCRIPTS. > > --Sigbjorn > ok, so far so good, install was happy, then testing gives: /homes/jpc1/bin/ghc: =2-space GC: not found /homes/jpc1/bin/ghc: =dual-mode GC: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =-prof -fscc-profiling -DPROFILING -optc-DPROFILING: not found /homes/jpc1/bin/ghc: =-fticky-ticky -DTICKY_TICKY -optc-DTICKY_TICKY: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =-concurrent -fstack-check -fconcurrent -D__CONCURRENT_HASKELL__ -optc-DCONCURRENT: not found /homes/jpc1/bin/ghc: =-prof -concurrent -fstack-check -fconcurrent -fscc-profiling -D__CONCURRENT_HASKELL__ -DPROFILING -optc-DCONCURRENT -optc-DPROFILING: not found /homes/jpc1/bin/ghc: =-ticky -concurrent -fstack-check -fconcurrent -fticky-ticky -D__CONCURRENT_HASKELL__ -DTICKY-TICKY -optc-DCONCURRENT -optc-DTICKY_TICKY: not found /homes/jpc1/bin/ghc: =-parallel -fstack-check -fconcurrent -D__PARALLEL_HASKELL__ -optc-DPAR -optc-DCONCURRENT: not found /homes/jpc1/bin/ghc: =-gransim -fstack-check -fconcurrent -fgransim -D__GRANSIM__ -D__CONCURRENT_HASKELL__ -optc-DCONCURRENT -optc-DGRAN: not found /homes/jpc1/bin/ghc: =-1s -optc-DGC1s: not found /homes/jpc1/bin/ghc: =-2s -optc-DGC2s: not found /homes/jpc1/bin/ghc: =-du -optc-DGCdu: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: =: not found /homes/jpc1/bin/ghc: syntax error at line 114: `=~' unexpected
Re: 2.07 binary installation
There's a tpyo in Makefile.in, change PACKAGE_LIBSCRIPTS to PACKAGE_LIB_SCRIPTS. --Sigbjorn Justin Cormack writes: > > > > Justin Cormack <[EMAIL PROTECTED]> writes: > > > > > The configure script is missing from the 2.07 Solaris binary > > > distribution it seems; presumably I can just steal it from the > > > source distribution ... ? > > > > No; you'll have to run autoconf to get configure for the binary > > distribution. > > ok, found autoconf, ran that, changed the Makefile because it was set to > version=2.03 not 2.07, but then running gmake install gives > > Configuring ghc, version 2.07, on sparc-sun-solaris2 ... > rm -f bin/sparc-sun-solaris2/ghc-2.07/ghc-2.07 > Creating a configured version of ghc .. > Done. > Creating a configured version of stat2resid .. > Done. > Creating a configured version of hstags .. > Done. > Creating a configured version of mkdependHS .. > Done. > /bin/sh: syntax error at line 1: `;' unexpected > gmake: *** [config-pkgs] Error 2
Re: 2.07 binary installation
> > Justin Cormack <[EMAIL PROTECTED]> writes: > > > The configure script is missing from the 2.07 Solaris binary > > distribution it seems; presumably I can just steal it from the > > source distribution ... ? > > No; you'll have to run autoconf to get configure for the binary > distribution. ok, found autoconf, ran that, changed the Makefile because it was set to version=2.03 not 2.07, but then running gmake install gives Configuring ghc, version 2.07, on sparc-sun-solaris2 ... rm -f bin/sparc-sun-solaris2/ghc-2.07/ghc-2.07 Creating a configured version of ghc .. Done. Creating a configured version of stat2resid .. Done. Creating a configured version of hstags .. Done. Creating a configured version of mkdependHS .. Done. /bin/sh: syntax error at line 1: `;' unexpected gmake: *** [config-pkgs] Error 2
Re: ghc-2.07 on HPUX calamaties.
Alex Ferguson writes: > > Juan Jose Quintela: > > I am sorry, but I press the repply button and I don't read the response > > address. Teh problem appears to be that I obtain the following > > "repply" adresses: > > Indeed, I think it's a problem with the list, not with your message, > Juan; I think a similar problem was mentioned before. Seems to me > axiomatic that list-servers shouldn't generate addresses in headers > that will make them ill if they're replied too. ;-) > OK, I better stick up for the local majordomo setup :) The headers generated are fine, the relevant ones are
Re: ghc-2.07 on HPUX calamaties.
> From [EMAIL PROTECTED] Thu Oct 2 08:05:35 1997 > Resent-Message-Id: <[EMAIL PROTECTED]> > Old-Received: from lionsden.informatik.uni-muenchen.de by vanuata with SMTP (MMTA) with ESMTP; Thu, 2 Oct 1997 08:05:26 +0100 > Old-Received: from tahiti.pms.informatik.uni-muenchen.de (tahiti.pms.informatik.uni-muenchen.de [129.187.214.96]) by lionsden.informatik.uni-muenchen.de (8.7.5/8.6.9) with SMTP id JAA22695 for <[EMAIL PROTECTED]>; Thu, 2 Oct 1997 09:05:23 +0200 (MESZ) > Old-Received: from liesl.pms.informatik.uni-muenchen.de by tahiti.pms.informatik.uni-muenchen.de with SMTP (1.38.193.4/16.2) id AA10927; Thu, 2 Oct 1997 09:05:22 +0200 Sven Panne on: > > [...] sh: 16173 Memory fault(coredump) [...] > You didn't tell which ghc version you used for compilation, but anyway > here is my guess if you used 0.29: I had the same problem here: hsc > built without any sign of trouble, but when it was used to compile the > libs, it coredumped. Sorry: yes, you're right, twas 0.29. I'll try again later with either 2.02, which built fine, just to be confusing, or with your kindly- offered later versions. Juan Jose Quintela: > I am sorry, but I press the repply button and I don't read the response > address. Teh problem appears to be that I obtain the following > "repply" adresses: Indeed, I think it's a problem with the list, not with your message, Juan; I think a similar problem was mentioned before. Seems to me axiomatic that list-servers shouldn't generate addresses in headers that will make them ill if they're replied too. ;-) Cheers all, Alex.
Re: ghc-2.07 and ghc-0.29
Juan Jose Quintela <[EMAIL PROTECTED]> writes: > Yes, but the problem is that the Parser.hs that cames with the sources > of ghc-2.07 is not ghc-0.29 friendly (it has newtype declarations) and you > can't compile another Parser.hs becouse you dont have happy, and so on ... Yes, that's a bug. Thanks for pointing it out. -- Simon Marlow [EMAIL PROTECTED] University of Glasgow http://www.dcs.gla.ac.uk/~simonm/ finger for PGP public key
Re: ghc-2.07 on HPUX calamaties.
Alex Ferguson wrote: > Anyone got any bright ideas on this on? OK, I'm trying to be bright... :-) > hsc seems to build OK on a HPUX box, but as soon as it tries to > compile PrelBase: > [...] sh: 16173 Memory fault(coredump) [...] You didn't tell which ghc version you used for compilation, but anyway here is my guess if you used 0.29: I had the same problem here: hsc built without any sign of trouble, but when it was used to compile the libs, it coredumped. Building 2.07 with 2.05 succeeded and so did bootstrapping. There are 0.29, 2.05 and 2.06 versions for HP-UX on ftp://ftp.informatik.uni-muenchen.de/pub/local/pms/ I didn't have the time yet to make a 2.07 HP-UX bundle (though there's a 2.07 Linux version). -- Sven PanneTel.: +49/89/2178-2235 LMU, Institut fuer Informatik FAX : +49/89/2178-2211 LFE Programmier- und Modellierungssprachen Oettingenstr. 67 mailto:[EMAIL PROTECTED]D-80538 Muenchen http://www.pms.informatik.uni-muenchen.de/mitarbeiter/panne
