- Haskell-Cafe - Haskell.org

RE: [Haskell-cafe] Adding Ord constraint to instance Monad Set?
by Simon Peyton-Jones 31 Mar '04

31 Mar '04

Alex | Following the declaration of "instance Monad []" | in the prelude, and puzzling over the absence of | its equivalent from Data.Set, I naively typed: | | instance Monad Set where | m >>= k = concatSets (mapSet k m) | return x = unitSet x | fail s = emptySet | | concatSets sets = foldl union emptySet (setToList sets) | instance (Eq b,Ord b) => Ord (Set b) where | compare set1 set2 = compare (setToList set1) (setToList set2) | | and got the following error: | | Could not deduce (Ord b) from the context (Monad Set) | arising from use of `concatSets' at dbMeta3.hs:242 I'm sorry to say that you just can't make Set an instance of Monad. To make an instance of Monad for a type constructor T you must have a function bindT :: forall a b. T a -> (a -> T b) -> T b And there just *is* no such function for Set, because of the need for ordering. There's a less polymorphic function bindSet :: forall a b. (Ord a, Ord b) => T a -> (a -> T b) -> T b To see why this is no good, consider sequence :: Monad m => [m a] -> m [a] The code for sequence calls (>>=) a lot. You can't supply bindSet as the function to call because bindSet takes two Ord parameters at run-time, whereas ordinary bind does not. It's not just a quirk of the type system -- there is a real problem! Simon

3 2

[Haskell-cafe] RE: [Haskell] Programming language shootout (completing theHaskell entry)
by Simon Marlow 31 Mar '04

31 Mar '04

> When compiling via gcc, there's the __builtin_prefetch function: > > http://gcc.gnu.org/onlinedocs/gcc-3.3.3/gcc/Other-Builtins.html > (about 2/3rds of the way down the page) Thanks, I wasn't aware of that. > see also: > http://gcc.gnu.org/projects/prefetch.html > > I didn't read it very carefully, but it's not clear (on the > x86 cpus) if > there is a prefetch instruction that is common between the amd & intel > flavours, ie would a ghc binary with prefetch be portable between > P3/P4/Athlon. P3/P4 and Athlon have different prefetch instructions. However, Athlon64 supports the P3/P4 instructions. Cheers, Simon

1 0

Re: [Haskell-cafe] Context for type parameters of type constructors
by MR K P SCHUPKE 30 Mar '04

30 Mar '04

>I didn't get the point yet why the context for 'data' >is not sufficient for the 'instance' method definition. Not sure who's relpy you're replying to, but I thought my response was pretty clear (I would though)... The problem is that to define 'zero' the type of the 'zero' must unify with the vector type. If you define vector operations as (v a) then a is a polymorphic variable. However the definition of zero implies that a is an Integral .. these are not the same type: (forall a . a) is not (forall a . Integral a => a) and thats where the type error comes from. You have two choices, drop the definition of zero, or use multi-parameter type classes. You could get rid of zero and have: class VectorSpace v where set :: Num a => a -> v a add :: Num a => v a -> v a -> v a scale :: Num a => a -> v a -> v a Should do the trick! (zero becomes "set 0") Regards, Keean.

3 3

What are Kind errors and how do you fix them?
by S. Alexander Jacobson 30 Mar '04

30 Mar '04

Implementing Reverse from before, I am running into this weird error: type ReverseType a string = (string ->(string,a)) data Reverse a string = Reverse (ReverseType a string) instance Monad (Reverse a s) where return x = Reverse (\text -> (text,x)) (Reverse p) >>= k = Reverse p3 where p3 s0 = p2 s1 where (Reverse p2) = k a (s1,a)=p s0 Produces the error: Kind error: Expecting kind `* -> *', but `Reverse a s' has kind `*' When checking kinds in `Monad (Reverse a s)' In the instance declaration for `Monad (Reverse a s)' I have no clue what this error message means. -Alex- _________________________________________________________________ S. Alexander Jacobson mailto:me@alexjacobson.com tel:917-770-6565 http://alexjacobson.com

5 6

Re: [Haskell-cafe] Context for type parameters of type constructors
by MR K P SCHUPKE 29 Mar '04

29 Mar '04

>Could not deduce (Num a) from the context (VectorSpace VList) The problem is in the definition: zero = VList (repeat 0) Is "0" an Int or an Integer? To define zero, instances need to be parameterised by vector type: EG: class VectorSpace v a where zero :: v a add :: v a -> v a -> v a scale :: a -> v a -> v a instance VectorSpace VList Int where zero = VList (repeat 0) ... instance VectorSpace VList Float where zero = VList (repeat 0.0) etc... >Could not deduce (Num a) from the context (VectorSpace VList) The problem is in the definition: zero = VList (repeat 0) Is "0" an Int or an Integer? To define zero, instances need to be parameterised by vector type: EG: class VectorSpace v a where zero :: v a add :: v a -> v a -> v a scale :: a -> v a -> v a instance VectorSpace VList Int where zero = VList (repeat 0) ... instance VectorSpace VList Float where zero = VList (repeat 0.0) etc... Regards, Keean.

1 0

RE: [Haskell-cafe] looking for optimization advice
by Simon Marlow 29 Mar '04

29 Mar '04

> I think that adding the extra check to see if the pointers > are identical > sped this up enough that it's probably no longer a major > issue--I'm pretty > certain that the problem was large strings that were > identical, so every > byte had to be checked, so probably scary non-portable home-made > ForeignPtrs would not be worth the effort. (Although I'm > somewhat curious as to how it would be done...) A ForeignPtr allocated with mallocForeignPtr in GHC is heap-allocated garbage-collectable storage, and doesn't need a finalizer. The other kind of ForeignPtr created with newForeignPtr is associated with some external storage and needs a finalizer. Finalizers are expensive, and so is malloc(), but mallocForeignPtr is very cheap. This gives rise to a sum-type in the implementation of ForeignPtr: see the source code (libraries/base/GHC/ForeignPtr.hs). You could make a specialised version by ripping out the non-malloc ForeignPtr bits, and make an unboxable ForeignPtr type. A mallocForeignPtr-style ForeignPtr is essentially the same as an IOUArray. In fact, the underlying implementation type is the same (MutableByteArray#), except that a ForeignPtr is allocated in pinned (immovable) storage so that it can be passed to foreign functions. There could easily be a conversion from a "malloc" ForeignPtr to IOUArray. There is also some overlap with the StorableArray type. A StorableArray has a ForeignPtr inside it. It really looks like the framework could be simplified here, but I remember putting some thought into it and not coming up with a simple solution. Cheers, Simon

1 0

dimension of arrays
by Fred Nicolier 29 Mar '04

29 Mar '04

Is there a way to get the number of dimension of an array ? i.e. something like : >dims :: (Ix a) => Array a b -> Int >dims = ... >a = listArray (1,10) [1,2..] >b = listArray ((1,1),(10,10)) [1,2..] >dims a -- should be equal to 1 >dims b -- should be equal to 2 The key is somewhere in the Ix class but where ? Fred -- ------------------------------------------------------- Dr. Frederic Nicolier Maitre de conferences, laboratoire LAM - URCA Animateur du projet ANITA : http://f.nicolier.free.fr/recherches/ Dept. GE&II, IUT de Troyes 9 rue de Quebec 10026 Troyes Cedex Tel: 03 25 42 71 01 Std: 03 25 42 46 46 Fax: 03 25 42 46 43 email: f.nicolier(a)iut-troyes.univ-reims.fr ---------------------------------------------------------

2 2

HToolkit HSQL on Windows/GHC
by ATWOOD,JOHN (HP-Corvallis,ex1) 29 Mar '04

29 Mar '04

Clive, I see that HaskellDb requires htoolkit, so that's no help. Doesn't MySql listen on a port? Is that protocol documented? I've used postgresql that way from Haskell. Another thought, which I hesitate to even mention, would be to call the commandline tool via System.Cmd, pipe the results to a file, and read the file, and parse the results. Slow and ugly, but you could have it working in an hour. Seems like we could even use the higher level abstractions of HSQL or HaskellDB, but sub in this SlowAndUglyButPortable for the lower level access. John >From: "Clive Brettingham-Moore" <c(a)brettingham-moore.net> >Subject: [Haskell-cafe] HToolkit HSQL on Windows/GHC >To: <haskell-cafe(a)haskell.org> > >I realise that this is a little technical for this list, but I thought >that I'd check here before starting on the hard-core lists. > >Is there an easy way to build the HToolkit HSQL library for GHC (6.0.1 when >I >tried, 6.2.1 now) on windows (particularly MySQL/Windows XP, but I wont >say no to postgres, ODBC & sqlite)? >I assume people have tired and succeded, but I got lost [I think] in the >nasty place between the unix/cygwin libraries in GHC and the Windows focus >of the compiled MySQL client library. > >Since haskell is mainly a hobby for me, and I don't have a lot of free >time I haven't dug very deep, but not being able to interface my two of my >favourite constructs (functional programming and relational databases) is >getting to me. > >So any takers? > >C

4 3

operating on nested monads
by Marco Righele 29 Mar '04

29 Mar '04

Hello everyone, I have some operations that have to be done in sequence, with each one having the result of the previous as input. They can fail, so they have signature a -> Maybe b Checking for error can be quite tedious so I use monadic operations: f :: a -> Maybe b do y <- foo x z <- boo y moo z The problems arise when I try to do the same thing within the IO Monad, i.e. the functions have signature a->IO (Maybe b) How can I achieve the same effect (if it is ever possible)? I feel like it should be something almost trivial, but I really can't get it. Thanks Marco

4 3

RE: [Haskell-cafe] looking for optimization advice
by MR K P SCHUPKE 27 Mar '04

27 Mar '04

Would it ba faster to store the strings as UArrays and do the comparison in Haskell? Regards, Keean.

2 1