- Glasgow-haskell-users

Re: FORTRAN and HASKELL
by heron_carvalho 30 Nov '01

30 Nov '01

Hello GHC users, We want to reuse FORTRAN libraries within Haskell code. How can we use FFI to implement this interface directly ? We know that we can use C to implement it indirectly with FFI, but how can we do this directly ? Any suggestion ? Thanks, Heron de Carvalho > --- clusterpoli <clusterpoli(a)bol.com.br> wrote: > > Somebody works with together FORTRAN and Haskell. Exists > > some funcao to transform codigo FORTRAN into codigo > > HASKELL? > > TXL might help, but it's its own functional language. > It can probably convert FORTRAN to Haskell given the > EBNF for each language, but I'm not sure how it works... > http://www.research.avayalabs.com/user/wadler/realworld/l s2000.html > http://gatekeeper.dec.com/pub/misc/txl/ > > Otherwise, I would look at what algorithm the FORTRAN is > implementing, the either write it in Haskell, or look for it > on the Internet using http://www.google.com/ > > For example: http://www.cs.chalmers.se/pub/haskell/library/ > http://www.cs.chalmers.se/pub/haskell/library/bevan/index > > __________________________________________________ > Do You Yahoo!? > Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month. > http://geocities.yahoo.com/ps/info1 > > _______________________________________________ > Glasgow-haskell-users mailing list > Glasgow-haskell-users(a)haskell.org > http://www.haskell.org/mailman/listinfo/glasgow- haskell-users > __________________________________________________________________________ AcessoBOL, só R$ 9,90! O menor preço do mercado! Assine já! http://www.bol.com.br/acessobol

2 1

ghc and hierarchical module names
by Tim Barbour 30 Nov '01

30 Nov '01

The release notes for ghc 5.02.1 say: 1.4.4. New experimental features Partial support for hierarchical module names. - but I could not find a document describing how to use this feature. Is it possible to use hierarchical module names with ghc at present, and if so, how does one write and use a hierarchically-named module ? Tim -- "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- Benjamin Franklin

1 0

[qforeign] CursesTest segfaults
by Jens Petersen 29 Nov '01

29 Nov '01

I built qforeign from cvs under Linux. Any idea why tests/CursesTest segfaults? Is my ncurses too new? (I have had it working previously - 6 months ago or so.) version info: ghc-5.02.1 ncurses-5.02 glibc-2.2.4 gmp-3.1.1 readline-4.2 If you need more info or traces, please let me know. Jens

1 0

Re: Casting dynamic values
by Marcin 'Qrczak' Kowalczyk 28 Nov '01

28 Nov '01

27 Nov 2001 17:31:10 -0700, Alastair David Reid <reid(a)cs.utah.edu> pisze: > If GHC is able to inline these functions, construction and > deconstruction of the pair can probably be eliminated. cast is compiled to something similar to this: coerce :: a -> b -- this generates no runtime code type Typeable a = a -> TypeRep cast :: Typeable a -> Typeable b -> a -> Maybe b cast t1 t2 x = case t1 x of rep1 -> case t2 (coerce x) of rep2 -> case eqTypeRep rep1 rep2 of True -> Just (coerce x) False -> Nothing Instances of Typeable should not look at their argument to determine TypeRep of its type but unfortunately the compiler can't in general assume that. > But I don't think GHC can do this because AFAIK, GHC will not reduce > literal string comparisions at compile time. Since Oct 1 it tries to. Unfortunately string literals seem to be floated out to toplevel before there are recognized as literals being arguments of (==), and the rule doesn't work. > [This is assuming that GHC's implementation of Typeof still uses > strings. This may not be true though since there was talk of adding > direct compiler support to make the implementation of typeof more > efficient and, more importantly, less prone to programmer error. It doesn't use strings for comparison but uniquely generated numbers. But it's not safe either: uniqueness of these numbers relies on TyCon objects being defined at the toplevel, with {-# NOINLINE #-}, and I'm not sure if common subexpression elimination doesn't mess this up (it corrupts similar cases). Anyway equality on statically known TypeRep is not done at compile time. I don't know why. -- __("< Marcin Kowalczyk * qrczak(a)knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ QRCZAK

1 0

Re: Socket library ghc 5.02.1
by Sven Eric Panitz 28 Nov '01

28 Nov '01

From: "Sigbjorn Finne" <sof(a)galois.com> Cc: <glasgow-haskell-users(a)haskell.org> Date: Tue, 27 Nov 2001 09:30:04 -0800 > > Conclusion: you're hosed with ghc-5.02.1 and its socket libs under > Win32. Sorry. > If you don't mind getting your hands a (little) bit dirty, here's a story that will work ghc-5.02.1: * edit SocketPrim.hi (and SocketPrim.p_hi), to instead of saying "Socket" in its __export section it says "Socket{MkSocket}" (you'll find the .hi file in imports/net/ inside your 5.02.1 tree). * compile up the attached NetExtra.hs as follows: foo$ ghc -c NetExtra.hs -fvia-C -fglasgow-exts -package net * import and include NetExtra with your socket code, e.g., Thanks Danke Mercie, this works (although I did not fully understand, what I was doing). Sven Eric

1 0

Casting dynamic values
by George Russell 28 Nov '01

28 Nov '01

It would occasionally be nice to have a function cast :: (Typeable a,Typeable b) => a -> Maybe b which returns Just a if a and b have the same type, and Nothing otherwise. This may seem rather a curious need, but it arises with existential types; if you have data A = forall a . (context) => A a (context including Typeable a) then this allows you to getB :: (Typeable b) => A -> Maybe b getB (A a) = cast a and so extract a value from A, if you can guess its type. Clearly we can implement cast = fromDynamic . toDyn My question is: is this the most efficient way of doing it, or is there a better way?

2 1

Re: Socket library ghc 5.02.1
by Sven Eric Panitz 27 Nov '01

27 Nov '01

From: "Sigbjorn Finne" <sof(a)galconn.com> Cc: <glasgow-haskell-users(a)haskell.org> References: <200111220959.KAA29614(a)softwareag.com> Date: Mon, 26 Nov 2001 12:30:47 -0800 "Sven Eric Panitz" <sep(a)softwareag.com> writes: > > It seems that the Socket library does still not work > with ghc 5.02.1. > I tried the simple test: > > > main = > > do > > d <- connectTo "localhost" (PortNumber 80) > > hPutStr d "GET / HTTP/1.0\n\n" > > hFlush d > > c <- hGetContents d > > putStr c > > On Windows2000 I get the known error: > > *** Exception: does not exist > Action: getProtocolByName > Reason: no such protocol entry > (You, of course, need to wrap up that code with Socket.withSocketDo to start up WinSock first). FYI, in case you're planning on doing socket programming with GHC-5.02 on a Win32 platform, stay away from using the higher-level Socket module, since its IO.Handle based view of sockets is just broken. Stick with the lower-level SocketPrim interface instead. Specifically, stay away from using the following: * Socket. connectTo * Socket.accept * Socket.sendTo * Socket. recvFrom * SocketPrim.socketToHandle --sigbjorn Thanks for the help and valuable information. I tried the following little test, which stays away from above functions: > module Main where > > import BSD > import SocketPrim > import Socket (withSocketsDo) > > main = > Socket.withSocketsDo > (do > protNum <- getProtocolNumber "tcp" > s <- socket AF_INET Stream protNum > hostAddr <- inet_addr "157.189.164.68" > let sAddr = (SockAddrInet (toEnum 8080) hostAddr) > connect s sAddr > i <- sendTo s "GET / HTTP/1.0\r\n\r\n" sAddr > (str,l,imsAddr) <- recvFrom s 1000 > putStr str > ) But something I seem to be doing wrong. During evaluation of 'recvFrom s 1000' I get the following error message (consistently for unix and windows): Fail: SocketPrim.hsc:241: Non-exhaustive patterns in case As the log of the webserver reveals the sendTo works fine. Any idea what could be my problem? Sven Eric

2 2

RE: (no subject)
by Yu Chen 26 Nov '01

26 Nov '01

hi, i've found the reason with someone's help. it is because i used "%code" instead of "safecode" in the gc file where i did callback. thanks! ---- chenyu ---------- Original Message ---------------------------------- From: "Simon Marlow" <simonmar(a)microsoft.com> Date: Mon, 26 Nov 2001 09:42:39 -0000 > >> i use green-card with ghc's FFI export to sth like >> >> Haskell -----> C ------> Haskell >> >> the program is just arithmatic operations. the program "seg fault" >> in the callback from C to Haskell. anyone whether it is a bug? > >There aren't any known bugs, but there are many known pitfalls. You'll >need to give us more information, at least: the code that causes the >crash (preferably a small example), the exact GHC command lines used to >compile it, your GHC version, your machine and OS version. > >Cheers, > Simon >

1 0

RE: GHC garbage collector
by Simon Marlow 26 Nov '01

26 Nov '01

> I have some code that uses FFI capabilities to associate > previously allocated > memory with a structure akin to a ByteArray and allows > Haskell land to read > and write to the C side of things. > > I use ForeignPtr and associate a destructor to allow the GHC's garbage > collector to clean up after it figures out that the structure > is no longer > being used. I know I am in funny territory. > > In some instances (when benchmarking) GHC takes its time > getting around to > garbage collecting. Can someone suggest hooks to encourage > GHC to clean up > things? Or better still functions to instruct GHC to garbage > collect specific > structures (by calling relevant destructors)? Also comments > on how safe this > whole business is. IOExts.performGC will cause a full garbage collection to be performed. If you want to be sure the finalizer has run to completion, you'll need to set up some synchronisation between the main thread and the finalizer using MVars. Finalizers have no special scheduling properties; they just run concurrently with the other threads in the system. Cheers, Simon

1 0

RE: (no subject)
by Simon Marlow 26 Nov '01

26 Nov '01

> i use green-card with ghc's FFI export to sth like > > Haskell -----> C ------> Haskell > > the program is just arithmatic operations. the program "seg fault" > in the callback from C to Haskell. anyone whether it is a bug? There aren't any known bugs, but there are many known pitfalls. You'll need to give us more information, at least: the code that causes the crash (preferably a small example), the exact GHC command lines used to compile it, your GHC version, your machine and OS version. Cheers, Simon

1 0