- Hugs-Bugs - Haskell.org

[Hugs-bugs] Wrapper FFI appears broken on Hugs/Windows
by Brian Smith 05 Sep '05

05 Sep '05

The tests that use the "wrapper" FFI scheme all fail with an unexpected signal: ffi/Sin.input ffi/Sparc.input ffi/Callback.input This leads me to believe that "wrapper" FFI declarations are broken on Windows. Unfortunately, I have no idea how to fix it. Can somebody help? Regards, Brian

1 0

[Hugs-bugs] [Fwd: Re: To Paul Hudak / Alastair Reid / hugs-bugs-request....]
by Paul Hudak 20 Jun '05

20 Jun '05

1 0

[Hugs-bugs] Re: [Haskell] ANNOUNCE: Hugs interim release
by Ross Paterson 14 Mar '05

14 Mar '05

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/sgml-docbook-4.2.cat > Using stylesheet: /home/test/hugs98-Mar2005/docs/users_guide/hugs-ug.dsl > Working on: /home/test/hugs98-Mar2005/docs/users_guide/users_guide.sgml > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1166:13:E: document type does not allow element "VARLISTENTRY" here > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1176:13:E: document type does not allow element "VARLISTENTRY" here > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1189:14:E: end tag for "VARLISTENTRY" omitted, but OMITTAG NO was specified > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1149:0: start tag was here > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1156:22:E: [xref to VARLISTENTRY unsupported] > jade:/home/test/hugs98-Mar2005/docs/users_guide/options.sgml:1193:5:E: [xref to VARLISTENTRY unsupported] > make[2]: *** [users_guide/license.html] Error 8 > make[2]: Leaving directory `/home/test/hugs98-Mar2005/docs/users_guide' > make[1]: *** [all] Error 2 > make[1]: Leaving directory `/home/test/hugs98-Mar2005/docs' > make: *** [all] Error 2 > > It's probably a missing or outdated package. Can you direct me in the > right direction (couldn't find any hints in the docs). No, the document is malformed: a </varlistentry> is missing before line 1165 of docs/users_guide/options.xml.

1 0

[Hugs-bugs] make_bootlib
by Victor Secarin 09 Mar '05

09 Mar '05

To whom it may concern: I have finally succeeded in building a recent snapshot of Hugs on my Fedora 1 box. I have built the late February snapshot and then the March 7 and the March 8 ones. All builds required the following same fix: - 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 have learned to expect, used to be ABC...XYZabc...xyz. In that case, saying [a-z] means "all the lowercase letters" However, in Fedora 1, if you do for instance an 'ls', you can easily see that the new collating sequence is now AaBbCc....XxYyZz. In this case, [a-z] now means "all letters, uppercase or lowercase, except A". Since [a-z] is in the argument to '-prune', the result is that it prunes everything except Array.hs. Using [abcdefghijklmnopqrstuvwxyz] instead is by all means ugly, but then shell scripting is ugly by default. Just as long as it works. I cannot imagine many sillier things than changing the collating sequence on everyone. So who do I tell? I guess we all want to be like Windows and we just don't know it... yours truly, Victor Secarin

2 1

[Hugs-bugs] MacOSX install problem
by Rich Loft 28 Feb '05

28 Feb '05

Dear Hugs-bugs: Excuse me in advance if this is a known problem. I am new to Haskell and I am trying to install HUGS on a MacOSX 10.3 system with gcc3.3. I get the following ld errors: gcc hugs.o edit.o observe.o builtin.o compiler.o errors.o evaluator.o ffi.o goal.o input.o machdep.o machine.o module.o opts.o output.o plugin.o script.o static.o storage.o strutil.o subst.o type.o version.o -lncurses -o hugs ld: Undefined symbols: _doubleFromParts _part1Double _part2Double Any ideas for a workaround? Rich Loft

2 1

RE: [Hugs-bugs] Bug with fun-deps
by Simon Peyton-Jones 23 Feb '05

23 Feb '05

I've added this as a GHC test too! (typecheck/should_compile/tc187) | -----Original Message----- | From: hugs-bugs-bounces(a)haskell.org [mailto:hugs-bugs-bounces@haskell.org] On Behalf Of Iavor | Diatchki | Sent: 19 February 2005 01:19 | To: hugs-bugs(a)haskell.org | Subject: [Hugs-bugs] Bug with fun-deps | | Hello, | | I seem to have run into a problem involving functional dependencies. | Below is an excerpt of a program I was writing that illustrates the bug. | | > data N0 | > newtype Succ n = Succ n | | > class Plus a b c | a b -> c | > instance Plus N0 n n | > instance Plus a b c => Plus (Succ a) b (Succ c) | | > ( # ) :: Plus x y z => x -> y -> z | > ( # ) = undefined | | > class BitRep t n | t -> n where | > toBits :: t -> n | | > instance BitRep Bool (Succ N0) | | > instance BitRep (Bool,Bool,Bool) (Succ (Succ (Succ N0))) where | > toBits (x,y,z) = | | The program works with the line below, but if I remove it | I get a type error --- Hugs complains that it cannot solve | the constraint: | | Plus (Succ N0) (Succ N0) (Succ (Succ N0)) | | > toBits' where toBits' = | | > toBits x | > # toBits y | > # toBits z | | The program works with GHC in either form, and it looks like | Hugs should be able to solve the constraint using the instances above. | | -Iavor | _______________________________________________ | Hugs-Bugs mailing list | Hugs-Bugs(a)haskell.org | http://www.haskell.org/mailman/listinfo/hugs-bugs

1 0

[Hugs-bugs] Bug with fun-deps
by Iavor Diatchki 19 Feb '05

19 Feb '05

Hello, I seem to have run into a problem involving functional dependencies. Below is an excerpt of a program I was writing that illustrates the bug. > data N0 > newtype Succ n = Succ n > class Plus a b c | a b -> c > instance Plus N0 n n > instance Plus a b c => Plus (Succ a) b (Succ c) > ( # ) :: Plus x y z => x -> y -> z > ( # ) = undefined > class BitRep t n | t -> n where > toBits :: t -> n > instance BitRep Bool (Succ N0) > instance BitRep (Bool,Bool,Bool) (Succ (Succ (Succ N0))) where > toBits (x,y,z) = The program works with the line below, but if I remove it I get a type error --- Hugs complains that it cannot solve the constraint: Plus (Succ N0) (Succ N0) (Succ (Succ N0)) > toBits' where toBits' = > toBits x > # toBits y > # toBits z The program works with GHC in either form, and it looks like Hugs should be able to solve the constraint using the instances above. -Iavor

1 0

[Hugs-bugs] Installing Hugs
by Iavor Diatchki 16 Feb '05

16 Feb '05

Hello, I seem to be having trouble installing Hugs from CVS. I checked out a fresh copy from CVS, and typed 'make' as the instructions on the web-page suggest. The actual Hugs builds fine, but something goes wrong with the libraries. I am using a Mandrake 10.1 distribuition of linux, and I suspect at least some of the problems are due to tool incompatabilities. 'make' reports the following errors: Preprocessing Control/Arrow /tmp/cppinput26849.c:114:3: invalid preprocessing directive #- /tmp/cppinput26849.c:226:3: invalid preprocessing directive #- and few others like that for some of the libraries (I guess something is wrong with Hugs & cpp?) The next kind of error I get is like this: runhugs: Error occurred ERROR "libraries/bootlib/Data/Bits.hs":188 - Syntax error in declaration (unexpected `}', possibly due to bad layout) This error seems to be reported many many times. Finally there seem to be some problems with generating the documentation: jade:/home/diatchki/src/hugs98/docs/users_guide/users_guide.sgml:2:22:W: cannot generate system identifier for public text "-//OASIS//DTD DocBook V4.2//EN" jade:/home/diatchki/src/hugs98/docs/users_guide/users_guide.sgml:12:0:E: reference to entity "BOOK" for which no system identifier could be generated etc. What do I need to do to get it to work? Let me know what tool versions might be relevant and I'll post them. -Iavor

1 0

[Hugs-bugs] Re: fromInt
by Bob Tennent 24 Jan '05

24 Jan '05

>I don't think you said anything about which version of Hugs you are using. I did in my original post: I'm using the hugs98-Nov2003-2 rpm on Linux. By "I" I mean that that's what's installed on the workstations my class is using and what some students are installing on their own computers at home. I don't think going back to a version from years ago is a realistic option. Now that I know what's going on, I can deal with it. I've suggested to Simon Thompson that he add this issue to the Errata page for his book. 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? Bob T.

2 1

Re: [Hugs-bugs] Re: fromInt
by Alwyn 24 Jan '05

24 Jan '05

On 24 Jan 2005, at 02:19, Mark P Jones wrote: > I don't think you said anything about which version of Hugs > you are using. It might be irritating, but it shouldn't be > too surprising that you will encounter problems if you try to > use a recent release with a textbook that is, if I recall > correctly, about five years old. Well, 'fromInt' figures in Mr Pope's *Tour of the Haskell Prelude*, which dates, I believe from summer 2001; mention of it seems to have been removed from the account of Numeric library in November of the same year. <http://www.haskell.org/definition/haskell98-bugs.html> I find no evidence from the original Haskell 98 Report that 'fromInt' was ever part of the public interface, so this in fact appears to amount to a minor tidying up. Thompson's book is from 1998, before the report was published. If the author does indeed make use of this function in the book, perhaps this should be submitted to him as 'errata'. Alwyn

1 0