- Beginners - Haskell.org

doing state right
by Floptical Logic 24 Apr '09

24 Apr '09

I am using a PPM library to generate a square image where each white pixel represents a prime number. The PPM library takes a function (Int -> Int -> Colour) to create the image. This interface isn't ideal but it is what I have to work with. I am convinced that using a sieve is faster than testing every pixel in the image for primality, but the (Int -> Int -> Colour) interface makes this awkward. The code below attempts to treat the list of prime numbers as a stack via global mutable state, popping the head whenever the pixel is prime. Obviously this is not idiomatic Haskell. What is the correct approach of dealing with state here? Thanks for reading. import ONeillPrimes import PPM6 import Colour import Data.IORef import System.IO.Unsafe limit = 2000 slimit = limit*limit primeList = takeWhile (<=slimit) primes p :: IORef [Int] p = unsafePerformIO (newIORef primeList) pcol n = unsafePerformIO $ do xs <- readIORef p if null xs then return black else if n == head xs then do writeIORef p (tail xs) return white else return black main = quick_ppm "foo.ppm" (\i j -> pcol ((i-1)*limit+j)) limit limit -- quick_ppm :: FilePath -> (Int -> Int -> Colour.Colour) -> Int -> Int -> IO ()

3 4

computation vs function
by Daniel Carrera 23 Apr '09

23 Apr '09

Hello, I have finished the tutorial at http://ertes.de/articles/monads.html and my understanding of monads has increased greatly. I still need to cement some concepts in my mind. What exactly is the difference between a computation and a function? Monads revolve around computations, so I'd like to understand computations better. Thanks for the help. Daniel.

6 10

Type families with kind * -> *
by Marco Túlio Gontijo e Silva 23 Apr '09

23 Apr '09

Hello, I read the type families example at http://haskell.org/haskellwiki/GHC/Indexed_types and I wanted to do something similar to the Collects example, but using a type of kind * -> *: > class StateFunctor sf where > type SFMonad sf > type SFValue sf > sfmap :: (SFValue sf -> SFValue sf) -> sf -> SFMonad sf () I wrote the instance as: > instance StateFunctor (ListStore a) where > type SFMonad (ListStore a) = IO > type SFValue (ListStore a) = a > sfmap function listStore > = listStoreGetSize listStore >>= listStoreSfmap function listStore > listStoreSfmap :: (a -> a) -> ListStore a -> Int -> IO () > listStoreSfmap _function _listStore 0 = [] > listStoreSfmap function listStore size > = listStoreGetValue listStore index > >>= listStoreSetValue listStore index . function > >> listStoreSfmap function listStore index > where > index :: Int > index = pred size I'm getting: DistroCreator/GUI/List.hs:51:47: Kind error: `SFMonad' is applied to too many type arguments In the type `SFMonad sf ()' In the type `sf -> SFMonad sf ()' In the type `(SFValue sf -> SFValue sf) -> sf -> SFMonad sf ()' Failed, modules loaded: none. How can I use type families with types * -> *? Greetings. -- marcot http://marcot.iaaeee.org/

3 2

Help with types
by Daniel Carrera 23 Apr '09

23 Apr '09

Hi, I want to make a simple "average" function: average xs = (sum xs)/(length xs) When I try that on ghci I get the following error: No instance for (Fractional Int) arising from a use of `/' at <interactive>:1:17-36 Possible fix: add an instance declaration for (Fractional Int) In the expression: (sum xs) / (length xs) In the definition of `average': average xs = (sum xs) / (length xs) Ok, so 'sum' is of type (Num a => [a] -> a), length is of type ([a] -> Int) and (/) is of type (Fractional a => a -> a -> a). Can someone explain how I can "add an instance declaration" to my function so as to make all these types compatible? Thanks, Daniel.

5 11

generating the set of all finite-valued functions on a finite space
by Erik Quaeghebeur 23 Apr '09

23 Apr '09

Hi, I'd like to lazily generate the set of all {-1,0,1}-valued functions on {'a','b','c'}? How should I best approach this. I was thinking about generalizing the power set definition powerset :: [a] -> [[a]] powerset = filterM (const [True, False]) but clearly don't know enough about filterM and the like to do it this way. Erik

1 0

problem cabal install'ing hmatrix
by Erik Quaeghebeur 22 Apr '09

22 Apr '09

$ cabal install hmatrix Resolving dependencies... Configuring hmatrix-0.5.1.1... Preprocessing library hmatrix-0.5.1.1... running dist/build/Numeric/GSL/Special/Internal_hsc_make failed command was: dist/build/Numeric/GSL/Special/Internal_hsc_make >dist/build/Numeric/GSL/Special/Internal.hs cabal: Error: some packages failed to install: hmatrix-0.5.1.1 failed during the building phase. The exception was: exit: ExitFailure 1 What should I do to diagnose/resolve this problem? I'm on Gentoo/amd64 and have also posted this question in the gentoo forums. TIA for any help and pointers, Erik

2 1

Design questions for a Pascal interpreter
by Giuliano Vilela 21 Apr '09

21 Apr '09

Hello, I've been doing some research for a upcoming homework project, a Pascal interpreter written in Haskell. As a beginner to the language, I have a few design questions I'd like to share with you. I started playing with Parsec, getting a feel for parsing with monadic combinators. My assumption here is that i'm going to take the source, transform it in an AST (built with normal data constructors present in Haskell), maybe build a symbol table while doing that (using the implicity state in Parsec), do some checking (mainly type checking I think), and the evaluation itself. Here are a few questions I had so far: - Keeping the whole AST in memory for the evalution phase seems overkill. Is there a better way? - The evalution, I think, would be a set of nice pure mutually recursive functions that do some pattern matching on the program AST. I would pass the current stack and heap for those functions to use and modify. Is the State monad a good fit for this task? Wouldn't the code become "too imperative"? - And finally, the big question. Consider the following pascal source: program HelloWorld; begin writeln('Hello World'); end. To evaluate the AST, I would eventually arrive at something like: eval (FunctionCall func_name func_args) = ... Obviously, to evaluate writeln I need to be in the IO monad. Here, my whole scheme went down. Do I really have to mix my own state (stack, heap) within the IO monad along my evaluation functions? PS: I could keep a list of things to print, that I would eventually do after I traversed the whole tree, but that wouldn't be "realistic". Thank you! -- []'s Giuliano Vilela.

3 2

typeclass error
by Amitava Shee 21 Apr '09

21 Apr '09

I have the following code -- foo.hs module Foo where class Foo a where bar :: a -> a instance Foo Int where bar i = i + 10 ------------------ It fails to compile withe following error *Foo> bar 10 <interactive>:1:4: Ambiguous type variable `t' in the constraints: `Num t' arising from the literal `10' at <interactive>:1:4-5 `Foo t' arising from a use of `bar' at <interactive>:1:0-5 Probable fix: add a type signature that fixes these type variable(s) *Foo> - Thanks Amitava Shee

3 4

Parsec 3.0 problems
by Ian Duncan 21 Apr '09

21 Apr '09

Hello everyone, I have the simple applicative parser combinator: parseLabel = Label <$> (between spaces (char ':') (many1 alphaNum)) This gives me the error: No instance for (Stream s m Char) arising from a use of `spaces' at tree.hs:18:32-37 Possible fix: add an instance declaration for (Stream s m Char) In the first argument of `between', namely `spaces' In the second argument of `(<$>)', namely `(between spaces (char ':') (many1 alphaNum))' In the expression: Label <$> (between spaces (char ':') (many1 alphaNum)) When I test this function in ghci, I have no problems, yet whenever I try to open it in a file it fails. What am I doing wrong? For those who are curious, Label is a data constructor in the following data type: data Loc = Label String | Addr Int deriving Show Thanks! Ian Duncan

2 1

Question on Lists
by Nathan Holden 19 Apr '09

19 Apr '09

I have been reading Chris Okasaki's PhD thesis on Purely Functional Data Structures (www.cs.cmu.edu/~rwh/theses/*okasaki*.pdf) and it discusses his idea of lazy lists (He uses Standard ML in the paper), and it raised some questions for me. To cut to the chase, my question is this: Say I have list x of length n, and I have a single piece of data y. Does x ++ [y] take more cycles, or the same as if I'd said y:[x] (which would get it on the wrong side?)

2 1