November 2005 - Haskell-Cafe

Compiled hugs - on "overload" crashes
by Dusan Kolar 03 Nov '05

03 Nov '05

Hello again, without the --endable-timer option, I have managed to compile hugs. Nevertheless, I was not successfull with doc. OK, I don't care about it, but anyway... I've run serveral tests. One of them is appended - for trial just type cesty (1,1) - it's a kind of state-space search implementation (horse problem) made by a student long time ago, without any optimization, for testing very good. 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 printing some results. Nevertheless, the new hugs compilation stops with this: Segmentation Fault(coredump) And that's probably not the correct end. May I do something wrong? I've thought that configure + make + make install_all_but_docs (no doc compiled) should work. Sorry for bothering if this is trivial or off topic. Regards, hoping for help, Dusan -- Dusan Kolar tel: +420 54 114 1238 UIFS FIT VUT Brno fax: +420 54 114 1270 Bozetechova 2 e-mail: kolar(a)fit.vutbr.cz Brno 612 66 Czech Republic --

1 0

Vector/matrix arithmetic
by James McNeill 03 Nov '05

03 Nov '05

I'm attempting to learn Haskell by writing some orbital mechanics code. I'd like to use vector and matrix arithmetic to simplify it. The Haskell book I've got, Simon Thompson's "Haskell: The Craft of Functional Programming," has one approach, which is to alias a list of numbers to Vector, and a list of vectors to Matrix, and then define functions to perform inner and outer products (dot/cross products) and so forth. On http://haskell.org/hawiki/FunDeps another approach is sketched out that looks far more appealing to me. This is to create Vector and Matrix types that can use overloaded arithmetic operators. It uses functional dependencies, and the resulting syntax looks a lot like what you can do using C++ or Matlab. I don't know enough Haskell to finish out the code that is started there, and I haven't been able to find any vector/matrix libraries for Haskell that look like this. I did find a port of a linear algebra library (BLAS) to Haskell, but it looked like it did everything in imperative style. Performance is not a major concern for me at this point so I'd prefer to do things functionally. Does anyone know what the best way to do vector/matrix arithmetic in Haskell might be? I don't need arbitrary dimension; all my work will be confined to 2D and 3D space. Thanks!

2 1

Compiling hugs - stop on missing symbol
by Dusan Kolar 03 Nov '05

03 Nov '05

Hello all, I don't know if this is the right place to ask about how to compile hugs, but I'll try. :-) My machine identifies this way: Linux mymachine 2.6.14 #1 SMP Mon Oct 31 10:55:16 CET 2005 i686 athlon i386 GNU/Linux I've downloaded this SRC package of hugs: hugs98-Mar2005-patched.tar.gz Download, unpack - OK ./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.o char.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 -lreadline -lncurses -lm -ldl -o runhugs evaluator.o(.text+0x137): In function `evaluator': /mnt/data/FLP/Downloads/hugs98-Mar2005-patched/src/evaluator.c:98: undefined reference to `updateTimers' collect2: ld returned 1 exit status make: *** [runhugs] Error 1 Well, yes, the symbol is definitely missing, as timer.c, which defines the symbol, is included in hugs.c, which is not in the list of linked files. Nevertheless, I don't think including the file manually would help. Moreover, probably configure has produced a wrong Makefile. Any suggestions how to make it work? Dusan -- Dusan Kolar tel: +420 54 114 1238 UIFS FIT VUT Brno fax: +420 54 114 1270 Bozetechova 2 e-mail: kolar(a)fit.vutbr.cz Brno 612 66 Czech Republic --

1 0

Haskell Workshop Steering Committee
by Johan Jeuring 02 Nov '05

02 Nov '05

At the Haskell workshop in Tallinn in September it was decided to set up a Haskell Workshop Steering Committee. The main purpose of the Haskell Workshop Steering Committee is to provide continuity of the workshop and to offer help and advice to the current organizer(s) of the workshop. I'm pleased to announce the Haskell Workshop Steering Committee, which has been approved by ACM SIGPLAN. You can find more information about the Haskell Workshop and its Steering Committee on http://www.haskell.org/haskell-workshop/ -- Johan Jeuring

1 0

MaybeT, guards and run-time pattern matching failure
by Joel Reymont 01 Nov '05

01 Nov '05

I'm trying to optimize this and I thought that things that return Nothing in the Maybe monad should fail the whole computation sequence. Something subtle is going wrong, I think. MaybeT is at the end of this message. ---- timeout :: forall a.Int -> IO a -> IO (Maybe a) connect_ :: HostName -> Int -> IO Handle startSSL :: IO (SSL, BIO, BIO) type EngineState = MaybeT (WriterT [Log] (StateT World IO)) connect :: [Prop] -> HostName -> Int -> EngineState () connect env h p = do h <- liftIO $ timeout 0 $ connect_ h p -- no complains about this (ssl', tob, fromb) <- liftIO $ timeout 0 startSSL -- bummer, see below ./Script/Engine.hs:101:7: Couldn't match `Maybe (SSL, BIO, BIO)' against `(a, b, c)' Expected type: Maybe (SSL, BIO, BIO) Inferred type: (a, b, c) When checking the pattern: (ssl', tob, fromb) In a 'do' expression: (ssl', tob, fromb) <- liftIO $ (timeout 0 startSSL) ---- I thought I would change it like this: ---- connect env h p = do (Just h) <- liftIO $ timeout 0 $ connect_ h p -- no complains about this ---- but when the connection times out (i'm making it with a delay of 0) I get the following runtime error: *** Exception: Pattern match failure in do expression at ./Script/ Engine.hs:103: So am I stuck with this? ---- connect :: [Prop] -> HostName -> Int -> EngineState () connect env h p = do m'h <- liftIO $ timeout 0 $ connect_ h p m'ssl <- liftIO $ timeout 0 startSSL guard (m'h /= Nothing && m'ssl /= Nothing) -- do I need this?! let h = case m'h of (Just h) -> h let (ssl', tob, fromb) = case m'ssl of a -> a ---- And this is MaybeT: newtype MaybeT m a = MaybeT { runMaybeT :: m (Maybe a) } instance (Monad m) => Monad (MaybeT m) where (MaybeT mon) >>= f = MaybeT (mon >>= maybe (return Nothing) (runMaybeT . f)) return = MaybeT . return . Just instance MonadTrans MaybeT where lift mon = MaybeT (mon >>= return . Just) instance (Monad m) => MonadPlus (MaybeT m) where mzero = MaybeT (return Nothing) mplus (MaybeT a) (MaybeT b) = MaybeT $ do ma <- a mb <- b return $ ma `mplus` mb instance MonadState s m => MonadState s (MaybeT m) where get = lift get put s = lift $ put s instance (MonadIO m) => MonadIO (MaybeT m) where liftIO = lift . liftIO Thanks, Joel -- http://wagerlabs.com/

2 1

Interactive Haskell and hs-plugins
by Fraser Wilson 01 Nov '05

01 Nov '05

Hi there, I would like to use evaluate arbitrary expressions in the context of a mixed-language (Ada, Haskell and about twelve lines of C glue) applications. Is it possible to use dynload from hs-plugins to load a module that references symbols in the loading program? For example, the application contains a Haskell module that looks a bit like this: > module TestKeys.NamedObject where > tt :: Int > tt = 2 I want the application to be able to offer the following sort of interaction: haskell> tt 2 haskell> But what actually happens is this: haskell> tt test_haskell3: /tmp/MDnum15633.o: unknown symbol `TestKeysziNamedObject_tt_closure' test_haskell3: user error (resolveObjs failed with <<False>>) Compiling Haskell code in advance works (i.e. if I create and compile a module that evalutes tt and link it in, everything runs fine). This is how I try to evaluate the expression: > evaluate :: String -> IO () > evaluate "" = return () > evaluate e = do writeFile "temp.hs" fileContents > status <- makeWith "LeanderStub.hs" "temp.hs" ["-c"] > case status of MakeSuccess code path -> loadAndEval path > MakeFailure error -> print error > where fileContents = "module Temp where\n\ > \ result = " ++ e ++ "\n" > > loadAndEval :: FilePath -> IO () > loadAndEval path = do mv <- dynload path [] [] "result" > case mv of > LoadFailure msg -> print msg > LoadSuccess _ v -> putStrLn v LeanderStub.hs is a module containing the necessary imports for the expression to evaluate in the right context. I was hoping that passing "-c" to makeWith would create an object file whose missing dependencies would be resolved when it was loaded into an application which contained them. Is this a dumb expectation? The alternative -- linking to the entire application for each expression evaluated -- seems a bit over the top, and I can't see how state would be maintained. I originally used unsafeEval_, but this has the same problem. I can't help but think that this must have been done before. Any advice? If you know Guile, then the equivalent of gh_repl is what I'm really after. cheers, Fraser.

2 1

DARCS and case of file name letters
by Dimitry Golubovsky 01 Nov '05

01 Nov '05

I have a file in my repo (HSX11/Xauthority.hs) and I want to rename it changing just case of one letter: bash$ darcs mv HSX11/Xauthority.hs HSX11/XAuthority.hs darcs failed: A file or dir named HSX11/XAuthority.hs already exists in working directory. bash$ ls HSX11/X[aA]*.hs HSX11/Xauthority.hs bash$ darcs --version 1.0.3 (release) DARCS was compiled by GHC 6.2.2. bash$ uname -a Linux dmghome 2.2.16 #9 Mon Sep 16 22:43:25 EDT 2002 i686 unknown Is it DARCS that ignores the case of letters in filenames, or the Haskell runtime? Dimitry Golubovsky Middletown, CT

2 1