functional dependency + subclass

I expected the following to work: > class Foo x where foo :: x > class Foo x => FooBar x where bar :: x > class FD b c | b -> c where fun :: b -> c > instance (Foo a, FD b c) => Foo (Either a b) where > foo = undefined > instance (FooBar a, FD b c) => FooBar (Either a b) where >

Re: functional dependency + subclass

Thanks for the patch (though it needs a few more '#if EXPLAIN_INSTANCE_RESOLUTION's). It gets me a little further, up to > class Foo a where foo :: a > class Foo a => FooBar a where bar :: a > class FD b c | b -> c where fun :: b -> c > instance (Foo a, FD b c) => Foo (Either a b) where >

existentials + contexts + universals

Try the following: data Showable = forall a. Show a => MkShowable (forall b. b -> a) showit (MkShowable f) = show (f 1) and you get ERROR "foo.hs" (line 3): Use of MkShowable requires at least 2 arguments Change the context to (Show a, Read a), and it wants 3 arguments, usw.

Re: Trex empty row

> It seems to be impossible to denote an empty row, so the type Rec () is > ill-formed. Perhaps you want EmptyRec :: Rec EmptyRow which is there. Pity about the type synonyms though.

Re: Type synonyms and Trex

On second thought, a bit more is needed: row synonyms in tail position must also be expanded as required. The following patch works for me. --- ../src/subst.c Tue Oct 10 10:05:49 2000 +++ subst.c Tue Oct 10 10:21:25

Re: Perhaps a Very Small Bug

On Thu, Jun 28, 2001 at 02:57:48PM +0100, Iain McNaughton wrote: > I was using WinHugs as a calculator. From the main command prompt in > WinHugs, I typed the following: > > 102.0*201.0 + 103.0*202.0 + 102.0*201.0 +102.0*202.0 + 102.5*200.5 > > I then hit 'return', and was presented with the a

missing braces

>From hugs.c: static Void local addScriptName(s,sch) /* Add script to list of scripts */ String s; /* to be read in ... */ Bool sch; { /* TRUE => requires pathname search*/ if (namesUpto>=NUM_SCRIPTS) { ERRMSG(

copyright notice

Most of the files in src/winhugs still have the old copyright notice. ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman/listinfo/hugs-bugs

constraints inside classes

Hugs and GHC (but not NHC) erroneously accept the following: class Foo a where f :: Eq a => a -> Bool ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman/listinfo/hugs-bugs

semicolons

Haskell 98 allows things separated by semicolons to be empty, unlike earlier versions. Hugs implements this for decls and ldecls, but not other contexts, e.g. it rejects stuff like module Foo where {import A;;x = 1;;y = 2;} do {;;;x <- [1];;;return x;;} The attached patch fixes

transitive imports

Given these modules: module A(Stuff(Foo, Bar)) where data Stuff = Foo | Bar module B(Stuff(..)) where import A module C where import B loading C gives "INTERNAL ERROR: importEntity" Changing the import in C gives different internal errors. The

standard input not a tty

When hugs's standard input is not a tty, this patch stops it prompting or attempting to use readline. It also disables $$ in that situation; this could be changed if desired, but is probably the right thing. If it is put it in, the --with-readline option of configure can probably be ditched (i.e

expandSynFully in subst.c

expandSynFully() is declared and defined inside #ifdef TREX but used outside of it (in getProgType). ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman/listinfo/hugs-bugs

Re: transitive imports

A variation on the previous bug. Given these modules: module A(Stuff(Foo, Bar)) where data Stuff = Foo | Bar module B(Stuff(..)) where import A module C where import B f = Foo loading C gives Dependency analysis ERROR "C.hs":4 - Undefined constructor function "Foo" __

hierarchical libraries

I've hacked up an experimental arrangement to make most of the hierarchical libraries distributed with GHC work with Hugs too. The aim is to provide source-level compatibility with both GHC 5.04 and current Hugs. The main omission is modules related to the FFI, as well as various GHC-specific stu

preprocessor & file not found

If I ask Hugs to read a non-existent file, e.g. "hugs Foo", I get Reading file "/usr/local/share/hugs/lib/Prelude.hs": Reading file "Foo": ERROR "Foo" - Unable to open file "Foo" Prelude> But if I say "hugs -Fcat Foo", I get Reading file "/usr/local/share/hugs/lib/Prelude.hs": Reading file "Foo

transitive imports & hiding

With these modules: module A(function, Class(method)) where function = id class Class a where method :: a module B(function, Class(..)) where import A module C where import B hiding (function) f = method hugs C.hs yields ERROR "C.

hiding (Stuff(..))

Hugs accepts the following modules: module A(Stuff(Foo, Bar)) where data Stuff = Foo | Bar module B(Stuff) where import A hiding(Stuff(..)) but Stuff should also be hidden. ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskel

Re: hiding (Stuff(..))

On Tue, Sep 10, 2002 at 08:21:28AM -0700, Sigbjorn Finne wrote: > Thanks, now fixed. Here's a trickier one: module A(Class(foo), bar) where class Class a where foo :: a bar :: a module B(Class(..)) where import A module C

Re: floating point arithmetic

On Fri, Sep 27, 2002 at 12:53:01PM +0200, Juan Manuel Urbano Blanco wrote: > Hi, I inform you of the following bug: > > Prelude> (4.72*25) > 118.0 > Prelude> floor (4.72*25) > 117 Prelude> 4.72*25 - 118 -7.62939e-06 Prelude> round (4.72*25::Rational) 118 Not a bug, just the old problem that Dou

Re: Hugs98 Oct2002 release candidate available

On Wed, Oct 23, 2002 at 03:19:57PM +0200, Oliver Braun wrote: > BTW, I am no auto{conf|make|...} crack, but on FreeBSD HAVE_LIMITS_H is > not defined, but limits.h has to be included for building ffihugs > without errors. I use something like > s/#ifdef HAVE_LIMITS_H/#if defined(__FreeBSD__)/ > f

Re: import ambiguity

On Wed, Oct 23, 2002 at 01:14:37PM -0400, Dean Herington wrote: > With the program: > > import Exception > main = let catch = Prelude.catch in catch (print "ok") print > > hugs 98 version 20021021 gives: > [...] > Dependency analysis > ERROR "ImportAmbiguity.hs":2 - Ambiguous variable occurrence

Re: convert_libraries chokes on Posix.hsc

On Sat, Oct 12, 2002 at 11:00:13AM -0700, Levent Erkok wrote: > Typing "make" on a freshly checked out hugs yields: > > > Converting base/Text/Regex/Posix.hsc > /usr/local/ghc-5.04/bin/i386-unknown-linux/hsc2hs: /hsc2hs-bin: No such file or >directory > sed: can't read ../../librarie

export/import of fields (in RC)

Given the modules module A(Type(..)) where data Type a = Constr { field :: a } module B where import A(field) Hugs says ERROR "B.hs" - Unknown entity "field" imported from module "A" ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://w

ambiguous occurrences (in RC1)

Hugs accepts the following: module A where x = 1 module B where import A x = 'a' y = x but the reference to x in B is ambiguous. ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman

Re: Directory checksum error in hugs98-Nov2002.tar.gz

On Thu, Nov 28, 2002 at 12:15:39PM +, Alastair Reid wrote: > > > Hi, I have a problem with the hugs generic unix source code. When > > extracting the tar file, a checksum error is reported. > > > cksum hugs98-Nov2002.tar.gz > > 31504372 1598806 hugs98-Nov2002.tar.gz > > Thanks for a detaile

Re: entities imported from multiple modules

On Wed, Dec 04, 2002 at 02:50:02PM +0100, Stavros Tripakis wrote: > I would like to report what I believe is a difference of Hugs > with the Haskell'98 report, in particular the section : > > http://haskell.cs.yale.edu/onlinereport/modules.html#sect5 > > In this section, it is mentionned that >

Re: signum (recip (-1 :% 1)) should not be 1 % 1

On Thu, Dec 05, 2002 at 07:04:34PM +0100, Ralf Welter wrote: > I think I have found a bug in the hugs prelude: > Bug: definition of recip in the instance declaration of breaks silent > assumptions about the internal data in Ratio types. > > Description: > data Integral a => Ratio a = !a :% !a der

Re: Hugs ffi, sizeOf operator

On Mon, Jan 06, 2003 at 03:42:27PM -0500, R. Legere wrote: > I ran into a possible problem while testing some ffi code. > This is with hugs compiled for cygwin or with the windows (msi) version. > Hugs says: sizeOf x::CDouble is 4, > wheras 'C' says sizeof(double) is 8. > > I ~~think~~ I see wha

Re: Trex

On Fri, Jan 10, 2003 at 12:50:14PM +, Stephen Eldridge wrote: > Using version Nov 2002 in Windows XP I find that import Trex gives me > > Reading file "c:\progra~1\hugs98\lib\hugs\Trex.hs": > ERROR "c:\progra~1\hugs98\lib\hugs\Trex.hs" - Illegal export of a lone data > cons > tructor "EmptyRe

Re: module bug?

On Sun, Feb 02, 2003 at 11:33:52AM -0600, Hamilton Richards wrote: > Version: November 2002 > > Platform: Mac OS X 10.2.3 > > Problem: Redefining Prelude names no longer requires explicit hiding. > > Example: > > script.lhs: > > > module M where > > > length = 5 >

Re: bug (?) in type checking of record update

On Tue, Feb 04, 2003 at 11:00:16AM +0100, Johannes Waldmann wrote: > wanna see a large Int in hugs? I mean Int, not Integer. > the program below gives this nice result: > > Main> contents bug > 100 :: Int > > data Container a = > Container { contents :: a >

Re: bug (?) in type checking of record update

On Tue, Feb 04, 2003 at 10:14:19AM +, Ross Paterson wrote: > On Tue, Feb 04, 2003 at 11:00:16AM +0100, Johannes Waldmann wrote: > > data Container a = > > Container { contents :: a > > , info :: String > > } > > Definitely

Re: hugs makefiles and Debian

On Tue, Feb 04, 2003 at 11:09:46AM -0500, Isaac Jones wrote: > [...] > Is there a way you or I can alter the build system so ffihugs looks at > the right place at the right time? (It looks at $DESTDIR during > installation and in the correct place afterward). I'd suggest cd src && HUGSDIR

Re: Nov2002 Hugs doesn't build properly on Solaris

On Wed, Feb 12, 2003 at 10:48:24AM -0800, Brian Rogoff wrote: > Hi, > Looks like a Makefile problem. If you don't have time or a Solaris > system, I'll look at this when I have time. FYI, The last release built > fine on this system. A lot of stuff that should be under fptools/libraries ap

Re: hugs makefiles and Debian

On Wed, Feb 12, 2003 at 10:04:23PM -0500, Isaac Jones wrote: > I still seem to have this problem: during the installation section of > the build, ffihugs is looking in /usr/share for the libraries it > thinks it just installed (Actually, it just installed them in > DESTDIR). The following steps wo

Re: Trex and module system

On Tue, Feb 18, 2003 at 11:51:06AM -0800, Iavor S. Diatchki wrote: > the recent updates of the module system seem to have broken Trex. > the problem is that Hugs.Trex tries to export "EmptyRec" which is a data > constructor and so cannot be named explicitly in an export list. > (interestingly enou

Re: failure to check completeness in instance declarations

On Fri, Mar 21, 2003 at 10:37:59AM -0800, Andrew Tolmach wrote: > The following file is accepted by hugs (Nov.02 version on linux) without > complaint. > > data Foo a = Foo a > > instance Eq (Foo a) where > > z = Foo 1 == Foo 2 Incomplete instances are legal in Haskell 98, but this one isn't, b

Re: hugs install info

On Tue, May 27, 2003 at 07:27:02PM +0100, Alastair Reid wrote: > Any thoughts on how I can find out information like > > version number of hugs program > location of HsFFI.h It would be trivial to make a shell script created by configure to do that. __

Re: fromInteger-Problem

On Fri, Jun 20, 2003 at 01:46:48PM +0200, Elke Kasimir wrote: > I use a hugs version from November 2002, which > happens to correspond to the unstable version of > the debian hugs-package, and > encountered the following: > > Prelude> fromInteger 4 :: Integer > 2 Strange -- I can't re

Re: A small problem

On Fri, Jul 25, 2003 at 08:17:37PM +0100, Shalom Lappin wrote: > I am a (very) new Haskell user, having spent many years working in Prolog. I > have just installed HUGS on my laptop in Windows XP, and it is working well > except for one strange problem. I cannot load more than one file at a time.

Re: INTERNAL ERROR: Import Entity

On Fri, Aug 01, 2003 at 11:39:11AM -0700, Jeffrey R Lewis wrote: > I've recently started to try using hugs again (HEAD), and have run into > the above error on several different programs. > > Anyone else have experience with the above beastie? Yes, but not for several months. I thought Sigbjorn

Re: runhugs args

On Tue, Aug 12, 2003 at 09:58:15AM -0700, Jeffrey R Lewis wrote: > Runhugs as of Oct 2002 handled command-line args supplied by #! > scripts. HEAD however, doesn't. That would be my fault. I re-instated option processing for the server after it had been accidentally deleted, but did it too simp

Re: bugs in Time.hs

On Tue, Aug 26, 2003 at 05:53:32PM +0200, Härmel Nestra wrote: > I found several divergences of the module Time of Hugs (v. Nov > 2002 for Linux) from the Haskell Report which can't be found neither > in the list of known Hugs bugs nor in the list of differences from > Haskell 98 (the latter is

Re: regex library misbehaviour

On Tue, Aug 26, 2003 at 10:02:22PM +0200, Ferenc Wagner wrote: > loading this: > -8< > import Text.Regex > > re2 = mkRegexWithOpts "was" True True > match = matchRegexAll re2 "this was wrong" > -8< > into hugs +N (Nove

Re: Minor BSD fixes for Hugs Aug03

On Wed, Aug 27, 2003 at 11:21:26AM +0200, Volker Stolz wrote: > The first patch to prelude.h avoids loads of gratuitious warnings about > using malloc.h (which is deprecated). Thanks -- I've applied a variant of this. > The second patch to configure.in caters for FreeBSDs different pthread-flags.

Re: Ldfs.hs

On Thu, Aug 28, 2003 at 01:18:31PM +0100, [EMAIL PROTECTED] wrote: > I think there is a typo in the definition of the function "tree", in the > module "Ldfs.hs", which is in the demos directory of the hugs distribution. > The original definition is: > > tree :: Bounds -> Forest Vertex

Re: enumeration oddity

On Tue, Sep 02, 2003 at 09:07:31AM -0400, Doug McIlroy wrote: > > Though I'm abusing enumerations, I'm surprised that a and b > give different answers (in Hugs 1.4 on Linux). Is that intended? > >default (Integer,Rational) >a = take 3 [1/1,3/2 .. 2/1] -- [1%1,3%2,2%1] >

Re: illegal export of lone data constructor

On Sat, Sep 13, 2003 at 04:57:34PM +0200, Andrew Frank wrote: > i am moving fom the nov 99 to the nov 02 hugs.exe (under windows) > when recompiling old files i find two problems which i do not understand: > > i want to export (:) from the prelude because i do not use the standard > prelude usuall

Re: -W option

On Wed, Sep 17, 2003 at 04:03:07PM -0400, Paul Hudak wrote: > This isn't really a bug report, but I wasn't sure where else to ask the > following question. Hugs used to handle all of my SOE code with no > problem. Now, however, I get the error message: > > ERROR "C:\Program Files\Hugs98\lib\

Re: -W option

On Thu, Sep 18, 2003 at 05:37:34PM +0100, Alastair Reid wrote: > I'd say the problem is in Hugs. Why does it have a setting which prevents it > from being able to load standard libraries? It's now something like 2 years > since we agreed to change the syntax, isn't it time that we removed suppo

Re: -W option

On Thu, Sep 18, 2003 at 05:37:34PM +0100, Alastair Reid wrote: > The error report comes when loading MarshalUtils.hs which is one of the new > standard FFI libraries and defines a function called 'with'. It's an > incredibly useful library that would be painful to avoid. > > I'd say the problem

Re: qualified import of prelude

On Sun, Sep 21, 2003 at 09:16:19PM +0200, Andrew Frank wrote: > i seem to have a problem with a qualified import of the prelude in hugs > nov03, haskell extensions, under windows: > > Reading file "C:\gisTheoryCodeSept03\lib\Language\Value240.hs": > ERROR "C:\gisTheoryCodeSept03\lib\Language\Value

Re: small hugs98-Sep2003 build problem on FreeBSD 4.8

On Mon, Sep 29, 2003 at 12:57:24AM +0200, Reinhard Speyerer wrote: > Trying to install hugs98-Sep2003 on FreeBSD 4.8 produced the following error: > > /usr/libexec/elf/ld: cannot find -ldl I believe this one is fixed in http://cvs.haskell.org/Hugs/downloads/Sep2003/hugs98-Sep2003-rc1.tar.gz Ple

Re: problems with export declarations in hugs nov 2002

On Sun, Sep 14, 2003 at 11:11:24PM +0200, Andrew Frank wrote: > i am moving (under windows) from hugs nov 99 to hugs nov 02. i observe the > following difference: > > i understand that for classes and data types, an export of the form X (..) > exports the class/data type with all the methods/const

Re: defining unix

On Thu, Oct 02, 2003 at 06:35:10PM -0700, Iavor Diatchki wrote: > hugs crashes if one tries to define unix: > > > unix = undefined > > INTERNAL ERROR: compileGlobalFunction I presume you have -F set to run the C preprocessor on source files (and unix is predefined in there to 1). So the real b

Re: parsing strings

[trawling through old bug reports] On Wed, Jul 30, 2003 at 06:44:52PM +, Iavor Diatchki wrote: > there seems to be a problem when parsing multi-line strings: > in the first string there is no space after the \, > while in the second there is. i tried this on linux, > with the july version of

Re: INTERNAL ERROR: parseInput

On Tue, Oct 07, 2003 at 04:21:12PM -0700, Iavor Diatchki wrote: > apologies for the many bug-posts. Keep them coming. However Jeff is probably the only one who can handle the fundeps ones. > hugs craches with an internal error when it encounters > an unfinished pattern like this: > > > test ( >

Re: hugsIsEOF in Mac OS X

On Wed, Oct 08, 2003 at 08:31:56AM -0500, Hamilton Richards wrote: > In Hugs Version: November 2002 running on Mac OS X (10.2), CTRL-D > does not result in hugsIsEOF returning True. Yes, stdin doesn't work very well in Hugs. This is unlikely to be fixed in the upcoming release, unfortunately. __

Re: module system problem

On Mon, Oct 06, 2003 at 09:38:13AM -0700, Iavor Diatchki wrote: > there seems to be a bug in the module system implementation: > > > module Test where > > import Prelude hiding (negate) > > negate = not > > test= negate > > ERROR "Test.lhs":7 - Ambiguous variable occurrence "negate" > *** Co

Re: IO problems

On Fri, Oct 10, 2003 at 01:36:20PM -0700, Iavor Diatchki wrote: > while looking at the program henk posted a few days ago, i run into a > rather weird bug in hugs, where it seems to be mixing IO handles. > the problem seems to happen with versions of hugs as early as Nov 2002. > it occurs on linux

Re: IO problems

On Mon, Oct 13, 2003 at 05:54:20PM +0100, Alastair Reid wrote: > > > The handle table was being clobbered after it was resized to fit the > > 41st handle. Now fixed in CVS. > > It seems like it is overly hard to test this kind of thing - you pretty much > have to know there is a bug before you'

Re: Type error for function main

On Thu, Oct 16, 2003 at 02:35:26PM +0200, Josef Svenningsson wrote: > I was just trying to write the smallest valid Haskell program. One try was > the following: > > > main = return 0 Much too long -- 9 characters are enough. > But I get the following error message: > ERROR "foo.hs":1 - Unresolv

Re: None

On Thu, Oct 16, 2003 at 03:54:51PM -0400, Doug McIlroy wrote: > In Hugs 1.4, ++ was an operator of class Monad; in Hugs 98 it is > an operator on lists. I've looked in the Hugs documentation and > haven't found anything about the change (probably a failing of > mine, not the documentation). >

Re: INTERNAL ERROR: parseInput

On Sun, Oct 26, 2003 at 04:58:57PM +0100, Henk-Jan.van.Tuyl wrote: > The attached program leads to the message: > INTERNAL ERROR: parseInput > at load time. > import IO > x, > y :: String Thanks for the report. This is a syntax error (y should be indented more than x) and should have been rep

Re: Fwd: when is defaulting supposed to occur?

On Thu, Oct 30, 2003 at 04:46:10PM -0800, Mark Tullsen wrote: > So, it looks like it is hugs which is not following the report > (section 4.5.5 (rule 2) as Iavor noted), which seems to indicate > that defaulting only occurs after an entire module is type-checked. Yes. It's among the divergences f

Re: recip bug?!

On Sun, Nov 23, 2003 at 02:21:09PM +0100, Robert Reitmeier wrote: > HUGS says: > > > Main> :version > > -- Hugs Version December 2001 > > Main> (-1)%1 > > -1 % 1 > > Main> recip ((-1)%1) > > 1 % -1 > > Main> (recip ((-1)%1))+(1%1) > > 0 % -1 > > Main> ((recip ((-1)%1))+(1%1))==0 > > False > > It'

Re: Fran1.16 and Hugs98-Nov2003

On Mon, Dec 01, 2003 at 04:44:13PM -0500, Antony Courtney wrote: > > ...examining the body of Fran.hs, it appears to be using a previous Hugs > feature for loading a module from a sub-directory which clashes > syntactically with hierarchical module support: > > $ cat Fran.hs > -

Re: FLOAT PRECISSION

On Mon, Dec 15, 2003 at 03:47:22PM +1100, mike V wrote: > What is the precission with which floating point operations are performed ? > The following example shows an incorrect answer in the second case ? > How can this problem be addressed i.e. the apparent precission seems to be > a sevre limita

Re: Failure to build hugs98-Nov2003 on Mac OS X 10.3.2

Thanks for the detailed report. Until this is fixed, you could change the line /* #undef HAVE_LIBM */ in config.h to #define HAVE_LIBM 1 and make again. ___ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman/listinfo/hugs-bugs

Re: Failure to build hugs98-Nov2003 on Mac OS X 10.3.2

On Fri, Dec 19, 2003 at 07:07:00PM +, Alwyn wrote: > the build succeeded, resulting in what seems to be a working. However, > towards the end, I got the following problems: > [...] OK, will check. You should have a working system though, except for the modules System.Posix.Signals, System.Po

Re: error messages in Nov 2003 release

On Tue, Jan 06, 2004 at 02:09:42PM +, Alastair Reid wrote: > I'm not quite sure when it happened but error messages in the Nov 2003 > release are much less informative than they used to be. In particular, > I was hoping to see a line number and filename in this message. > > Program error: p

Re: problems building for AIX without gcc

On Mon, Jan 12, 2004 at 03:30:38PM -0500, herington, dean wrote: > In trying to build Hugs using > > hugs98-Nov2003.tar.gz on an AIX machine without gcc, I get many errors. > I've attached a sample of some of them. Any ideas? T

Re: New moderation policy

It seems that the policy of rejecting non-member email has had the unexpected consequence of an increase in postings being sent to the wrong list, because people find it too much trouble to subscribe to the right list (and unsubscribe again) just to send one message. Thus we see Hugs bugs in the ha

[Hugs-bugs] Re: [Haskell] More on HXML toolbox with Hugs/Windows: matchRegex

On Mon, Jan 26, 2004 at 06:29:45PM +, Graham Klyne wrote: > I think I've managed to create a version of POpen to go with Hugs under > Windows. (It all compiles/links, but I haven't yet got a stand-alone test > case for it). > > So now I'm pushing ahead with trying to get the HXML toolbox ru

Re: [Hugs-bugs] Winhugs startup error

On Mon, Jan 26, 2004 at 08:45:23AM -0800, Jim Kerr wrote: > I just installed Hugs from the file hugs98-Nov2003-1.msi. The installation > seemed to go just fine, and I can start the command-line version of Hugs > without any problem. If I try to run winhugs, however, this error message > appears i

[Hugs-bugs] Re: the woes of non-cvs haskellers

On Wed, Feb 04, 2004 at 01:37:32PM -, Claus Reinke wrote: > To give just one concrete example, what packages do I > recommend to someone wanting to learn a bit of Haskell, and > do some graphics and perhaps windowing, on Windows? > > - Hugs (Winhugs bug in current binary release, so better av

Re: [Hugs-bugs] module system/libraries bug

On Wed, Feb 04, 2004 at 10:32:56AM -0800, Iavor S. Diatchki wrote: > i am having the following problem (just tried it with the current cvs > version, on linux) > > > import Array > > import ST hiding (bounds) > > > > test x = bounds x > > ERROR "test.hs":4 - Ambiguous variable occurrence "bounds

Re: [Hugs-bugs] Hugs-bugs bug

On Wed, Apr 21, 2004 at 01:35:08AM -0700, F. Warren Burton wrote: > The March 2004 archive is missing from > http://www.haskell.org//pipermail/hugs-bugs/ > and guessing the URL from the general pattern doesn't work. > > Does anybody have a copy of the March messages? In particular, I am > looki

Re: [Hugs-bugs] Hugs-bugs bug

On Wed, Apr 21, 2004 at 08:29:51AM -0700, Ivan Matousek wrote: > I am resending a log of our problem when installing hugs on Solaris 8 > in the attachment. > > [...] > checking if '/LD' builds loadable libraries... > checking if '/LD /ML /nologo' builds loadable libraries... > checking if '-shar

Re: [Hugs-bugs] Hugs-bugs bug

On Wed, Apr 21, 2004 at 11:29:50AM -0700, Ivan Matousek wrote: > I have recompiled hugs with gcc-2.95 and the builup is now > finished. It looks that the gcc-3.3 is causing the compile > error. I'm glad you've got something working, but I'd still like to track down the original problem. Can you s

[Hugs-bugs] Re: [Haskell] Bug in experimental Unicode support for Hugs?

On Fri, May 28, 2004 at 01:20:32PM +0100, Graham Klyne wrote: > I've noticed a discrepancy in by version of Hugs with experimental Unicode > support enabled, based on the 20040109 codebase. It's exemplified by this: > > [[ > Main> '\x10' > '\1114111' > Main> maxBound::Char > '\255' > Main> >

Re: [Hugs-bugs] Compiling Hugs on Cygwin

On Mon, Oct 11, 2004 at 12:54:11PM +0200, Koen Claessen wrote: > gcc -c -g -O2 machdep.c > machdep.c: In function `scanSubDirs': > machdep.c:562: error: storage size of `findInfo' isn't known > machdep.c:577: error: `_A_SUBDIR' undeclared (first use in > this function) > machdep.c:577: error: (Eac

Re: [Hugs-bugs] Internal eror in Hugs, trying to read a non existing file

On Wed, Sep 29, 2004 at 02:18:44PM +0200, [EMAIL PROTECTED] wrote: > The attached program reads a filename ("x.x") from file "filenames.dat"; > then the program tries to read file "x.x"; > this file does not exist, in the first call to "main", an internal error > message is displayed; the second ti

Re: [Hugs-bugs] internal error

On Fri, Oct 22, 2004 at 07:45:32AM +, [EMAIL PROTECTED] wrote: > bla = (if True then (+) else (+)) 0 0 An elegant report. Here's another version: bla = (if True then id else negate) (1::Int) Unfortunately it seems to point at a whole class of longstanding bugs in the peephole optimi

Re: [Hugs-bugs] Suggested change

On Tue, Oct 26, 2004 at 11:49:52AM +0100, Alwyn wrote: > This is the third time I've broached this subject over a period of > time: I get a linking error when making shared libraries for Hugs; the > linker complains of the absence of the '_environ' symbol. Your patch has been applied. __

[Hugs-bugs] monomorphism bug

I can't track this one down, so I'll post it: Prelude> :t let f = const 1 in let g = f in g let {...} in (let {...} in g) :: Num a => b -> c The error here is that a and c should be identified. In larger examples, this can cause real trouble. More detail: f is subject to the Mon

Re: [Hugs-bugs] Bug evaluating boolean expresion

On Thu, Jan 13, 2005 at 01:02:13PM +0100, Pepe Gallardo wrote: > After loading the following script, > > b' = (if True then f else odd) 1 > f _ = True > > [...] > > Main> b' > > Program error: pattern match failure: instShow_v89_v1324 0 b' Thanks for the report. This is the same bug as ht

Re: [Hugs-bugs] Re: fromInt

On Sun, Jan 23, 2005 at 10:23:57PM -0500, Bob Tennent wrote: > I don't recall this issue arising previously; shouldn't there be a > Changelog to document language changes from one version of Hugs to the next? There is: http://cvs.haskell.org/Hugs/pages/users_guide/miscellaneous.html (pa

Re: [Hugs-bugs] make_bootlib

On Wed, Mar 09, 2005 at 12:00:12PM -0700, Victor Secarin wrote: > - in file libraries/tools/make_bootlib - on line 77 - replace [a-z] > with [abcdefghijklmnopqrstuvwxyz] > > Explanation: they changed the collating sequence on us. > > In older distros, the Unix collating sequence, which we h

[Hugs-bugs] Re: [Haskell] ANNOUNCE: Hugs interim release

On Mon, Mar 14, 2005 at 10:52:06AM +0100, Ralf Hinze wrote: > It compiles just fine on my amd64 box, except for the user's guide: > > mkdir users_guide > /bin/cp hugs-ug.css users_guide > test -z 'docbook2html' || \ > docbook2html --dsl hugs-ug.dsl users_guide.sgml > Using catalogs: /etc/sgml/sg

Re: [Hugs-bugs] coredump on innocent code

On Thu, Sep 08, 2005 at 04:00:47PM +, [EMAIL PROTECTED] wrote: > $ cat bug.hs > main = print (let (x,(q,_)) = (1,divMod x x) in q) > $ runhugs bug.hs > Illegal instruction (core dumped) Not entirely innocent: demand for x also causes evaluation of the second pair, which depends on x: an infini

Re: [Hugs-bugs] coredump on innocent code

On Sat, Sep 10, 2005 at 03:18:16PM -, [EMAIL PROTECTED] wrote: > I previously thought that rewriting > > let = >= in > > as > > let (,) = (,) in > > is harmless, but section 3.12 of the Report says different. Yes, you'd have to write it as let (~,~) = (,) in > I sti

[Hugs-bugs] existentials

Iavor Diatchki writes: > It appears that Hugs (20050909) does not deal correctly with > existentials and multi-parameter type classes: > > > class C a b where > > f :: a -> b > > > > data T b = forall a. C a b => T a > > > > test (T a)= f a > > Hugs accepts this pr

[Hugs-bugs] Re: Compiling hugs - stop on missing symbol

[moving from haskell-cafe to hugs-bugs] On Thu, Nov 03, 2005 at 09:57:37AM +0100, Dusan Kolar wrote: > ./configure --prefix=/mnt/data/FLP --enable-timer > does not complain, suggests to perform make; make install > > Unfortunately, make stops with this error: > > gcc runhugs.o server.o builtin

[Hugs-bugs] Re: Compiled hugs - on "overload" crashes

[moving from haskell-cafe to hugs-bugs] On Thu, Nov 03, 2005 at 02:32:44PM +0100, Dusan Kolar wrote: > My older hugs interpreter stops with error: ERROR - Control stack overflow > GHCi stops with the same error: *** Exception: stack overflow > That's OK, really, that's correct behavior after prin

[Hugs-bugs] re-export bug

Bug reported by Bulat Ziganshin: load of C fails incorrectly in module A (f) where class A a where f :: a module B (module A) where import A module C where import B (f) Same if A is module A (f) where data R = R { f :: Int }

[Hugs-bugs] FDs and extistentials

On Mon, Jan 30, 2006 at 05:21:50PM +0300, Bulat Ziganshin wrote: > the attached program compiles under GHC, but not with Hugs. as far as > i see, Hugs don't use dependencies in class headers to figure out that > there is only one "vMkIOError" that can be called in the last > definition The example

Re: [Hugs-bugs] FD handling unaware of overlap resolution?

On Mon, Feb 20, 2006 at 12:16:14PM -, Claus Reinke wrote: > okay then, on to bugs:-) Here's a simpler example of this problem: class C a b | a -> b instance C a b => C (a,Int) (b,Int) instance C a b => C (a,c) (b,c) ___ Hugs-Bugs mailing list Hugs-

Re: [Hugs-bugs] FD handling unaware of overlap resolution?

On Fri, Mar 10, 2006 at 03:44:44PM +, Neil Mitchell wrote: > Is Hugs going to get a trac reporting thingy at some point, since all > the other Haskell projects seem to have them and I'm getting jealous :) There should be one next week. ___ Hugs-Bugs

  1   2   >