- Glasgow-haskell-users

Multi-parameter OOP
by George Russell 25 Oct '01

25 Oct '01

Recently I've been experimenting with a sort of OOP with GHC, using existential types and (overlapping, undecidable) multi-parameter type classes, but it doesn't seem to work as you might expect because of the way GHC resolves overloaded functions at compile-time. For example, given class A a data WrappedA = forall a . A a => WrappedA a data A1 = A1 instance A A1 class B b data WrappedB = forall b . B b => WrappedB b data B1 = B1 instance B B1 class AB a b where toBool :: a -> b -> Bool instance (A a,B b) => AB a b where toBool _ _ = False instance AB A1 B1 where toBool _ _ = True instance AB WrappedA WrappedB where toBool (WrappedA a) (WrappedB b) = toBool a b a naive user (like me a month ago) might expect that this to work, so that toBool (WrappedA a) (WrappedB b) will return False unless a is an A1, and b a B1, in which case it returns True. In fact ghc5.02 (rightly) gives an error message with the second instance declaration: Could not unambiguously deduce (AB a b) from the context (A a, B b) The choice of (overlapping) instance declaration depends on the instantiation of `a, b' So is there any other way of doing this sort of dynamic lookup at runtime, in a reasonably neat way?

2 2

RE: Thread-Safety through FFI
by Simon Marlow 25 Oct '01

25 Oct '01

> >It isn't safe to call foreign exported > >functions from more that one OS thread simultaneously. > > OK, can I call a foreign exported function from some other > thread while > the main thread is in 'foreign import' native code? I could > have calls to > the foreign exported function block until the main thread called a > special foreign imported 'yield' function. This might work, but no guarantees. Make sure that at least the foreign import yield doesn't have an unsafe annotation. The reason I'm slightly worried is that the RTS isn't written to be used in a threaded environment - it doesn't use thread-friendly library calls, and I don't even want to think about signal handling. Cheers, Simon

1 0

qforeign-0.65: Bzip2: problems with [Word8]
by Michael Marte 25 Oct '01

25 Oct '01

Hello *, I am just trying to catch up with the lastest developments. >From qforeign-0.64 to qforeign-0.65, some types were changed: "Zlib and Bzip2 express byte lists as [Word8] instead of String." But I need String -> String! (Reading from a file, decompressing, and feeding it to the XML parser.) How can I do this? I found out that fromIntegral does the conversion from Word8 to Char but I do not know how to convert from Char to Word8. Since files are pretty large, the conversions should have no operational impact. Thanks, Michael

1 1

RE: gtk+hs and libgmp
by Simon Marlow 25 Oct '01

25 Oct '01

> When trying to compile the examples in gtk+hs, I get the > following link > errors: > /opt/ghc/lib/ghc-5.02/libHSstd.a(PrelFloat__413.o): In > function `shP2_2_alt': > PrelFloat__413.o(.text+0x2b0): undefined reference to `__gmpz_cmp_si' > etc.. > When linking simple programs, this doesn't occur. > I have the following gmp's installing: > /usr/lib/libgmp.so.2.0.2 > /usr/local/lib/libgmp.so.3.1.1 > /opt/ghc/lib/ghc-5.02/libgmp.a > It seems that the programs only link with the version in > /usr/local/lib. > I tried putting this path in 'library_dirs' for package 'gmp' in > 'package.conf', but to no avail. Only when I explicitly put > '/usr/local/lib/libgmp.so' in 'extra_ld_opts' for package 'gmp' > did it work. The version in /opt/ghc/lib/ghc-5.02 should work, as should /usr/local/lib/libgmp.so.3.1.1. The version in /usr/lib looks like GMP 2, so that would cause the link errors. Make sure the link command line doesn't have -L/usr/lib on it, or at least not before -L/usr/local/lib and -L/opt/ghc/lib/ghc-5.02. Cheers, Simon

1 0

RE: Socket library
by Simon Marlow 25 Oct '01

25 Oct '01

> I just wanted to play a bit with the Socket library comming > with Haskell. However, whatever I try to do, I get the following > runtime error, eg in ghci: > > Socket> connectTo "pcsep" (PortNumber 80) > *** Exception: does not exist > Action: getProtocolByName > Reason: no such protocol entry > > > Seems that I am doing something wrong here? Which platform are you on? (there is a bug in 5.02's networking library which means that Socket.conncetTo might not work - but I haven't seen it produce the error that you quote above). Simon

3 2

gtk+hs and libgmp
by Gérard Milmeister 25 Oct '01

25 Oct '01

When trying to compile the examples in gtk+hs, I get the following link errors: /opt/ghc/lib/ghc-5.02/libHSstd.a(PrelFloat__413.o): In function `shP2_2_alt': PrelFloat__413.o(.text+0x2b0): undefined reference to `__gmpz_cmp_si' etc.. When linking simple programs, this doesn't occur. I have the following gmp's installing: /usr/lib/libgmp.so.2.0.2 /usr/local/lib/libgmp.so.3.1.1 /opt/ghc/lib/ghc-5.02/libgmp.a It seems that the programs only link with the version in /usr/local/lib. I tried putting this path in 'library_dirs' for package 'gmp' in 'package.conf', but to no avail. Only when I explicitly put '/usr/local/lib/libgmp.so' in 'extra_ld_opts' for package 'gmp' did it work. -- Gérard Milmeister Tannenrauchstr. 35 8038 Zürich gemi(a)bluewin.ch

1 0

RE: Thread-Safety through FFI
by Ashley Yakeley 24 Oct '01

24 Oct '01

At 2001-10-24 02:52, Simon Marlow wrote: >If you're talking about OS threads (eg. POSIX threads), Yes >then GHC's RTS certainly isn't thread safe. Good. This means I know why it's crashing... >It isn't safe to call foreign exported >functions from more that one OS thread simultaneously. OK, can I call a foreign exported function from some other thread while the main thread is in 'foreign import' native code? I could have calls to the foreign exported function block until the main thread called a special foreign imported 'yield' function. The alternative is to have all Haskell in one thread, which means other threads would have to queue up calls to foreign exported functions, which would get executed by the main thread whenever it called 'yield'. -- Ashley Yakeley, Seattle WA

1 0

Re: 5.02 MonadError Broke My Code
by Marcin 'Qrczak' Kowalczyk 24 Oct '01

24 Oct '01

Sun, 21 Oct 2001 19:38:55 -0700, Ashley Yakeley <ashley(a)semantic.org> pisze: > MonadError seems to have been redefined in 5.02 to have a fundep: > > 5.00.2: > > class (Monad m) => MonadError e m > > 5.02: > > class (Monad m) => MonadError e m | m -> e > > Why? It allows some practical functions which would be awkward to write without the fundep. For example: data ParseError = ... class HasParseError e where inject :: ParseError -> e parseInteger :: (MonadError m e, HasParseError e) => m Integer -- This type is ambiguous without the fundep because the main part of -- the type doesn't mention e. You could argue that we should write this instead: parseString :: MonadError m ParseError => String -> m () but this style isn't always feasible. It would force to have a concrete error type in all places where the error type is not directly visible in the type of the whole function. I don't want to commit to a particular error type in generic parsing library for example - the library only specifies properties of the error type using classes and gives several choices for it; details of the error type are specific to places where the library is used. Also a constraint like 'MonadError m ParseError' which mentions a concrete type can't be put in an instance context without -fallow-undecidable-instances. I have an example where putting this in an instance context is useful: there is a large set of constraints (six) which is used in many functions, so they are packed in a subclass with no methods which serves as a "class synonym". > Anyway, because of GHC's naive instance overlapping checking, it broke my > code: > > instance (JVMMonad m) => MonadError ThrowableRef m where Sorry about that. A single MonadError class can't simultaneously allow both ways of associating error types with monads. I guess you can move 'MonadError ThrowableRef m' to superclasses of JVMMonad and make MonadError instances for all types in this class separately. -- __("< Marcin Kowalczyk * qrczak(a)knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTÊPCZA QRCZAK

1 0

Socket library
by Sven Eric Panitz 24 Oct '01

24 Oct '01

I just wanted to play a bit with the Socket library comming with Haskell. However, whatever I try to do, I get the following runtime error, eg in ghci: Socket> connectTo "pcsep" (PortNumber 80) *** Exception: does not exist Action: getProtocolByName Reason: no such protocol entry Seems that I am doing something wrong here? Sven Eric

1 0

RE: Thread-Safety through FFI
by Simon Marlow 24 Oct '01

24 Oct '01

> Is there a way of calling exported 'dynamic export' functions in a > thread-safe manner? I have this problem in JVM-Bridge... > > This is what the main thread does, from Haskell: > > 1. starts the Java VM, which starts a bunch of other threads, > including > the AWT thread; > 2. creates a pointer to a some IO function 'paint', using a 'dynamic > export' callback-maker; > 3. creates a new Java Frame (i.e. a GUI window) with the > function-pointer; > 4. repeatedly calls the Java Thread.yield function for 20 seconds. > (the program then ends) > > Now, once the Java Frame has been created, the AWT thread will > occasionally call the 'paint' method, which ends up calling the > function-pointer. However, it seems GHC 5.02 'dynamic export' > calls are > not thread-safe, and sooner or later it crashes, either with a seg > violation or a complaint about unknown closure types. If you're talking about OS threads (eg. POSIX threads), then GHC's RTS certainly isn't thread safe. It isn't safe to call foreign exported functions from more that one OS thread simultaneously. We have some partially working thread support, but it has been on the back burner for some time now. Cheers, Simon

1 0