July 2008 - Beginners

data type design question
by Markus Barenhoff 31 Jul '08

31 Jul '08

Hi, I have written a parsec parser for reading a marshallaled dataformat. The returned data structure is based on the following data type: > data T = TString String > | TInt Integer > | TList [TorrentT] > | TDict [(TorrentT, TorrentT)] I think TString and TInt are clear. The elements of a TList always have the same "type" (same constructor). The TDict is a dictionary where the key is always a TString but the value can be of any of the other "types", even in the same dictionary. F.e the key "foo" may map to a TInt while the key "bar" maps to another TDict. I'am not happy with this declaration, but I'am not sure how to express this better. One haskell data type for each of the four and then using type classes? Maybe something like this? : > type TString = String > type TInt = Integer > type TList = TC t => [t] > type TDict = (TC t) => [(TString, t)] > class TC where ... > instance TC TString > instance TC TInt > instance TC TList > instance TC TDict Thnx for some inseperation! Markus -- Markus Barenhoff - Münster - Germany - Europe - Earth e-mail: alios(a)alios.org - jabber: alios(a)jabber.ccc.de - icq: 27998346

6 9

[Haskell-begin] Looking for cunning ways to update a tree
by Quergle Quergle 29 Jul '08

29 Jul '08

Hello all, This is a really helpful list: I've learned half a dozen new things just by reading this month's traffic. Anyway...I have the following bit of code that updates a tree structure given a route to a leaf: data Tree a = Leaf a | Node (Tree a) (Tree a) deriving (Show, Eq) data PathSelector = GoLeft | GoRight deriving (Show, Eq) type Path = [PathSelector] selectChild (Node left _) GoLeft = left selectChild (Node _ right ) GoRight = right updateNode (Node _ right) GoLeft newLeft = Node newLeft right updateNode (Node left _) GoRight newRight = Node left newRight updateLeaf new (Leaf previous) = Leaf new updateTree :: Tree a -> Path -> a -> Tree a updateTree tree path newValue = case path of [] -> updateLeaf newValue tree (p:ps) -> updateNode tree p (updateTree' (selectChild tree p) ps newValue) I wanted to rewrite updateTree without using explicit recursion. Unfortunately, the best I could come up with is: upDownRecurse :: (a -> b -> a) -> (a -> c) -> (a -> b -> c -> c) -> a -> [b] -> c upDownRecurse down bottoming up = upDownRecurse' where upDownRecurse' acc [] = bottoming acc upDownRecurse' acc (x:xs) = up acc x (upDownRecurse' (down acc x) xs) updateTree' :: Tree a -> Path -> a -> Tree a updateTree' tree path newValue = upDownRecurse selectChild (updateLeaf newValue) updateNode tree path So what's the sexier way of doing this? Cheers, -- Matt

3 5

[Haskell-begin] Maximum of a List?
by Steve Klabnik 29 Jul '08

29 Jul '08

Hello everyone- I'm working on some project Euler problems today, and I'm stuck on one. It's not the problem itself that's the problem, it's that finding the maximum of a list makes me run out of heap space! http://projecteuler.net/index.php?section=problems&id=14 my code: import Data.List f :: Int -> Int -> Int f acc x | x == 1 = acc | even x = f (acc + 1) (x `div` 2) | otherwise = f (acc + 1) (3 * x + 1) answer = (foldl' max 0) $ map (f 1) [1 .. 999999] I tried using 'foldl' max ' instead of 'maximum' because I thought that foldl' was supposed to work better than foldl or something...I could be confused on that point. Anyway, here's what I get... / _ \ /\ /\/ __(_) / /_\// /_/ / / | | GHC Interactive, version 6.4, for Haskell 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ \____/\/ /_/\____/|_| Type :? for help. Loading package base-1.0 ... linking ... done. Prelude> :l ..\test.hs Compiling Main ( ..\test.hs, interpreted ) Ok, modules loaded: Main. *Main> answer GHC's heap exhausted: current limit is 268435456 bytes; Use the `-M<size>' option to increase the total heap size. I also tried writing a simple 'max' function that I thought was tail recursive, and that that would help. Same error. I'm finding it hard to believe that finding a maximum of a list of a million small integers would cause this kind of overflow...is it really that big of a problem? Any help would be appreciated. I have a feeling I'll run into this problem again in the future.

5 15

New List
by Chad Wilson 28 Jul '08

28 Jul '08

And the new list is now begun for people like me, n00bs. -- Chad Wilson

15 23

[Haskell-begin] Cabal-Install Update Problems
by Jeff Wheeler 26 Jul '08

26 Jul '08

Hi, (Note: I also posted this a few days ago to libraries(a)haskell.org without much response, and thought I might have more success here. Please correct me if this is inappropriate.) I'm trying to use cabal-install on two machines, Debian Sid and OS X 10.5, but I am unable to run "update" on either. Most other commands seem to work fine, but the networking for "update" gives an error on both machines, completely independently: Downloading package list from server 'http://hackage.haskell.org/packages/archive' cabal: recv: resource vanished (Connection reset by peer) Running with the -v3 option does not provide much more information: Downloading package list from server 'http://hackage.haskell.org/packages/archive' Sending: GET http://hackage.haskell.org/packages/archive/00-index.tar.gz HTTP/1.1 User-Agent: cabal-install/0.5.1 Creating new connection to hackage.haskell.org cabal: recv: resource vanished (Connection reset by peer) Both machines are using a vanilla Netgear ProSafe VPN Firewall FVS318v3, a very simple prosumer router. Obviously, the router works fine for everything else, and has been running flawlessly for ages. This is using cabal-install from darcs and the cabal 1.4 branch. Thanks a bunch for helping, Jeff Wheeler

1 0

[Haskell-begin] Morphing Endo (ICFP Contest 2007)
by Rafael Gustavo da Cunha Pereira Pinto 25 Jul '08

25 Jul '08

Did any of you tried to do a Haskell implementation of the ICFPC 2007 problem? I was thinking of using it as a learning exercise, but I am afraid of the stack. My approach is: 1) use Data.ByteString.Lazy.Char8 to read the contents of the DNA file 2) create a recursive function process::ByteString -> a that will call itself. I have a few problems: a) the DNA is 8MB long. How can I ensure the stack will hold a recursive call? b) there is an "abnormal ending" function called finish that is called anywhere in the code. Is it a good approach to return Empty to end processing? c) should I go "monadic", keeping the dna on a state monad? Thanks -- Rafael Gustavo da Cunha Pereira Pinto Electronic Engineer, MSc.

3 5

[Haskell-begin] some basic syntax questions
by Anatoly Vorobey 25 Jul '08

25 Jul '08

Hello Haskell-beginners, I'm trying to work through the "Write Yourself a Scheme in 48 Hours" tutorial. One of the first exercises calls for writing code to sum up arguments on command line and display the sum. After a few tries, this worked for me: module Main where import System.Environment main :: IO () main = do args <- getArgs putStrLn ("Hello, " ++ show (sumIt args)) where sumIt x = sum $ map read x However, I have two basic questions: 1. I initially tried putStrLn("Hello, " ++ show $ sumIt args), but that didn't compile. Why not? If I wrap (show $ sumIt args) in parens, it works, but should I have to, if I didn't have to in the original code above? 2. I initially tried where sumIt = sum $ map read (the "point-free" style, I believe it's called?) but that didn't compile. Why not? A friend suggested where sumIt = sum . map read and that does work; I guess my real problem, then, is that I don't really understand the difference between the two and why the former doesn't work. Thanks in advance, Anatoly. -- Anatoly Vorobey, avorobey(a)gmail.com http://avva.livejournal.com (Russian) http://www.lovestwell.org (English)

4 3

Re: [Haskell-begin] Morphing Endo (ICFP Contest 2007)
by Rafael Gustavo da Cunha Pereira Pinto 24 Jul '08

24 Jul '08

Ops, forgot to copy the list. On Thu, Jul 24, 2008 at 07:54, Rafael Gustavo da Cunha Pereira Pinto < rafaelgcpp(a)gmail.com> wrote: > > > > > 2008/7/23 Benjamin L. Russell <DekuDekuplex(a)yahoo.com>: > >> >> Correct me if I'm wrong, but because the prefix should not be too >> long, and chemical process of modifying Endo with the prefix should >> not consume too much energy, this appears to be an example of >> constraint programming. Since constraint programming can often be >> carried out by constraint logic programming, I might suggest a logic >> programming approach for this problem. >> >> > > The DNA is actually a representation of a turing machine (like BF > language). Using brute-force and constraints would take too long. There is a > report at http://www.cs.uu.nl/research/techreps/repo/CS-2007/2007-029.pdf, > that shows how this DNA string was made: it is the output of a compiler for > a functional programming language specially tailored for this contest. > which, in turn, was written in Haskell. > > The basic idea is to disassemble, find the function calls and add > prefxes that do the appropriate calls. > > > -- > Rafael Gustavo da Cunha Pereira Pinto > Electronic Engineer, MSc. > -- Rafael Gustavo da Cunha Pereira Pinto Electronic Engineer, MSc.

1 0

[Haskell-begin] data declarations
by Roger Myers 23 Jul '08

23 Jul '08

hi Not sure if I'm on the right track here, so as a beginner I thought I'd ask on this list. I want to declare a type 'All' that consists of the of the types 'Tinteger, Tbool, Tstring, Tarray' but I would also like to make a relationship that Tinteger and Tbool are also the type 'Simple' and that 'Tstring' and 'Tarray' are 'Composite' and that Mixed is then either Simpletype or Composite. So I can then write a function that takes a type of All and produces a list of Mixed. data All = Tinteger Int | Tbool Bool | Tstring String | Tarray [Alltypes] data Simple = Tinteger | Tbool data Composite = Tstring | Tarray data Mixed = Simple | Composite k :: All -> [Mixed] k a = case a of Simple s -> [Simple s] Composite c -> [Composite c] But this isn't valid Haskell code because I can't have multiple declarations of the same types, is there a way to specify a relationship, if so can someone point me in the right direction? Roger

4 3

[Haskell-begin] Exercises for beginners and Natural Tansformations
by Federico Brubacher 22 Jul '08

22 Jul '08

Hi all, I think i will love this list :) Yesterday I was looking at this Haskell Exercises for beginners [1] and got somewhat stuck at Ex nbr 2 , here's why: The idea of this exercises is that you don't use built in Haskell Lists, so no pattern matching, instead you have to use the data type list provided in the source of the exercise. My friend who is more advanced in Haskell said this approach is good because you get to learn natural Transformations. So I googled that and I came up with this [2]. I can read al little bit of category theory but I would love to know how to apply [2] in the sum exercise. Anyway as we all know summing all the elements in a list using Haskell lists is trivial : sum :: [Int] -> Int sum (xs) = foldr(+)0xs But how to do it with Natural transformations ??? Thanks a lot Federico [1] http://blog.tmorris.net/haskell-exercises-for-beginners/ -- Federico Brubacher www.fbrubacher.com Colonial Duty Free Shop www.colonial.com.uy

5 8