- Beginners - Haskell.org

Problem installing Haskell platform on Ubuntu Karmic
by Kim Stebel 10 May '10

10 May '10

Hello List, I can't compile the haskell platform. I followed this guide http://davidsiegel.org/haskell-platform-in-karmic-koala/ and downloaded http://hackage.haskell.org/platform/2010.1.0.0/haskell-platform-2010.1.0.0.… as suggested in comment 5. configure runs fine, but when I try to "make" it, it fails to compile the network package. Preprocessing library network-2.2.1.7... Building network-2.2.1.7... [1 of 5] Compiling Network.URI ( Network/URI.hs, dist/build/Network/URI.o ) [2 of 5] Compiling Network.Socket.Internal ( dist/build/Network/Socket/Internal.hs, dist/build/Network/Socket/Internal.o ) [3 of 5] Compiling Network.Socket ( dist/build/Network/Socket.hs, dist/build/Network/Socket.o ) [4 of 5] Compiling Network.BSD ( dist/build/Network/BSD.hs, dist/build/Network/BSD.o ) [5 of 5] Compiling Network ( Network.hs, dist/build/Network.o ) Network.hs:274:15: Warning: Pattern match(es) are overlapped In a case alternative: a -> ... Network/URI.hs:108:7: Could not find module `Text.ParserCombinators.Parsec': Perhaps you haven't installed the profiling libraries for package `parsec-2.1.0.1'? Use -v to see a list of the files searched for. Error: Building the network-2.2.1.7 package failed make: *** [build.stamp] Error 2 I'd really appreciate any help.

3 2

When, if ever, does Haskell "calculate once"?
by Travis Erdman 10 May '10

10 May '10

Two questions here, I think they might be related, perhaps even the same, but I am not sure, so I will ask both: Q1: Re the function f below, I like the first implementation as it's "cleaner", but is the second implementation necessary for performance purposes? -- g = some CPU intensive function -- alternate 1 f a b = c + (g b) where c = dosomethingelse a (g b) -- alternate 2 f a b = c + saveit where saveit = g b c = dosomethingelse a saveit Q2: Imagine that I might be creating a custom data structure. I need to access both the data in that structure, but also "features" (don't know proper comp sci term) of the data structure. For instance, consider a Haskell list. The stuff in the list is the data, whereas the length of the list is what I am calling a "feature" here. Some of my features might be quite a bit more computationally intensive than "length", and may be referenced multiple times in many different places in my code. For performance purposes, should I consider modifying my data structure to embed these "features" as part of the data to ensure that it is only calculated once? Or can I rely on Haskell to do that for me? For instance, do I need to create the equivalent of: data MyList a = MyList {mylist::[a], mylength::Int} Thanks again for all your generous advice.

5 8

Question about FRP trigger based programming
by Mathew de Detrich 10 May '10

10 May '10

Basically Im designing an interface which will (eventually) be customizable through scripts run through HSPlugin, and the scripting is going to be very reactive based (its for a game). Im basically trying to make a Trigger based API, so this example should show simply what I am trying to do (you can also see it here http://pastebin.com/ZNCD75KN) Obviously some of the haskell isn't legitimate (I just hope im conveying what I am trying to do) data Event = OnPress | OnRelease data Bindings = (a,b,..z) -- This isn't legal haskell, but basically a tuple with any -- amount of types, this would have to be done with generics or something along those lines onKey :: Event Char Bindings (Char → Bindings → ReturnVal) onKeyPress = error "stub" testFunc :: Char → Bindings → ReturnVal testFunc a _ = do putStrLn [a] testFunc2 :: Char → Bindings → ReturnVal testFunc2 a _ = do putStrLn [z] where z = a++a testFunc3 :: Char → Bindings → ReturnVal testFunc3 a b = do putStrLn ([a] ++ show bX ++ bZ) where bX = b.x -- x binding from the tuple (x,z) bZ = b.z -- z binding from the tuple (x,z) main = do onKey Press 'A' Nothing testFunc onKey Release 'B' Nothing testFunc2 onKey Press 'C' (x,z) testFunc3 where x = 2 z = "cheese" Basically the premise is that you make "Triggers" which are based on 2 data types. One being the event (i.e. in this example a key being pressed down or up) and the actual data type that is being triggered (in this case a key). In the example above, the onKey is the actual function that the users creating the scripts will use (so it has to be as clear as possible) and the testFuncX are the functions that the users will create to bind onto the triggers. What happens is you attach functions to those triggers, and those functions execute (for now consecutively but it can be changed) whenever those events happen. There also needs to be the ability of binding variables to those functions if you need them (this is what happens for testFunc3 with (x,y)), in case users need to send extra data to the functions being bounded onto the triggers. The supposed above would output this A BB C2cheese I have been reading through two FRP libraries (both Yampa and Reactive) however the papers are very generic and there seems to be limited examples (at least in the area of binding functions to triggers instead of the more generic events and behaviors based on values that change over time). Would the basic principle above be able to coded using either Reactive or Yampa or would I have to create my own library to do such a thing, and more importantly how would I be able to do it (and a working version of the example above) Thanks

1 0

A basic misunderstanding of how to program with IO
by Ken Overton 10 May '10

10 May '10

Sorry for such a beginner-y question, but is there a way to make a function like: interact :: String -> Resp interact txt = putStrLn txt rsp <- getLine return parseResp rsp parseResp :: String -> Resp Or is that simply a wrong way of programming in Haskell with IO? Thanks (and apologies), -- kov

5 8

comprehending monad execution
by rick＠rickmurphy.org 08 May '10

08 May '10

All: I've been analyzing monads for a while now and have a question on execution. In the snippet below the ghc debugger shows that in just one single step Discriminate has been substituted for g. Even with :trace turned on in the debugger there is no evidence of an execution step to carry out the substitution which leads me to assume a compile time operation identifies Discriminate as a substitute for g. Could someone correct my assumption or explain how this works? Note: this snippet was extracted from [1]. > {-# LANGUAGE EmptyDataDecls, Rank2Types #-} > module LEM where > import Control.Monad.Identity > > data F -- no constructors > > data Discriminate a b = L a | R b > instance Monad (Discriminate e) where > return = R > R x >>= f = f x > L x >>= f = L x > > h :: (forall m. Monad m => ((a -> m F) -> m F)) -> a > h g = case g (\x -> L x) of > R x -> error "Not reachable" > L x -> x > > -- A proof term for a -> NOT NOT a > lift :: forall a m. Monad m => a -> ((a -> m F) -> m F) > lift x = \k -> k x > > t = h $ lift True > ghci trace ... C:\haskell\lem>ghci lem.lhs GHCi, version 6.8.3: http://www.haskell.org/ghc/ :? for help Loading package base ... linking ... done. [1 of 1] Compiling LEM ( lem.lhs, interpreted ) Ok, modules loaded: LEM. *LEM> :b h Breakpoint 0 activated at lem.lhs:(14,2)-(16,17) *LEM> :trace t Loading package mtl-1.1.0.1 ... linking ... done. Stopped at lem.lhs:(14,2)-(16,17) _result :: a = _ [lem.lhs:(14,2)-(16,17)] *LEM> :step Stopped at lem.lhs:(14,8)-(16,17) _result :: a = _ g :: (a -> Discriminate a F) -> Discriminate a F = _ [lem.lhs:(14,8)-(16,17)] *LEM> :hist -1 : h (lem.lhs:(14,2)-(16,17)) -2 : t (lem.lhs:22:6-18) <end of history> [lem.lhs:(14,8)-(16,17)] *LEM> My previous analysis includes running single step execution on Wadler's I monad here [2], where I am able to observe execution of the I monad. [essence.lhs:32:33-37] *Essence> :hist -1 : unitI (essence.lhs:7:34) -2 : unitI (essence.lhs:7:2-34) -3 : interp (essence.lhs:32:26-38) -4 : interp (essence.lhs:(31,2)-(39,44)) -5 : test (essence.lhs:54:41-51) -6 : showval (essence.lhs:(26,2)-(28,45)) -7 : showI (essence.lhs:9:34-42) -8 : showI (essence.lhs:9:2-42) -9 : test (essence.lhs:54:34-52) -10 : test (essence.lhs:54:2-52) <end of history> [1] http://okmij.org/ftp/Computation/lem.html [2] http://homepages.inf.ed.ac.uk/wadler/papers/essence/essence.ps Rick

1 0

[Fwd: Binary tree algorithm]
by legajid 07 May '10

07 May '10

Oops, mistake in sender name. Hi, i wanted to write an algorithm for loading a binary tree, then get the values in order and other operations. So, i created a data type : data AR a = R a (AR a) (AR a) | N deriving (Show, Ord, Eq) for an example : R 30 (R 10 (N) (R 20 N N )) (R 40 (N) (R 50 N N )) the main problem i had is : how to update such a complex value, to add another one. After hours of trying (and error !) my solution is : find where to insert the new value, then read the whole tree and copy its values, except for the one for insert. The tree must be read entirely as long as there are values to insert. For large numbers of values, i guess calculation time will increase seriously. I wrote it using "basic" Haskell capabilities. Are there some modules that manage this in a quicker way? using mutable variables ? Below is my code. How can it be improved ? Several limitations to my program (i could solve them) : . only Int values (strings should not be a problem) . minimum 2 values to make a tree . no duplicate values :it's used to find the node where to insert a new value. To allow duplicates, i should generate a unique "node id". -- cre [30, 10, 40, 50, 20] cre :: [Int] -> AR Int cre val = -- add values except the first one cre' (tail val) arb where -- init tree with first value arb=R (head val) N N -- Add values to the tree cre' :: [Int] -> AR Int -> AR Int cre' [] arb = arb -- at the end, get last value for the tree data cre' (x:xs) arb = cre' xs (cre'' x arb) -- Find the node where to insert the new value and proceed cre'' :: Int -> AR Int -> AR Int cre'' n arb = cre''' n arb (r, gd) where (r, gd)= trouve n arb -- insert at node value r, g(auche) ou d(roite) -- Read the whole tree and update the specified node to insert the new value cre''' :: Int -> AR Int -> (Int, String) -> AR Int cre''' _ N _ = N cre''' n (R x y z) (r, gd) = R x (cre''' n y' (r,gd)) (cre''' n z' (r,gd)) where -- left value (<) y' = if x == r && gd=="g" then R n N N else y -- right value (>) z' = if x == r && gd=="d" then R n N N else z -- find the node where to insert the value trouve :: Int -> AR Int -> (Int, String) trouve n (R x y z) | n <= x && y == N = (x, "g") | n <= x = trouve n y | n > x && z == N = (x, "d") | n > x = trouve n z trouve n (N) = (0, " ") Thanks for your comments and help. Didier.

4 4

movie database
by Payan Peshraw 07 May '10

07 May '10

Hi, I have got a database with movies added into it. What Im basically coding is: - add a new film to the database - display all the films - display all the films by a given director - display a list of user ratings (together with the users' names) for a given film - show the average overall rating for a given director - allow a given user to rate (or re-rate) a film (note that only the latest rating from the user should remain recorded) - display all the films from a given year, sorted in descending order of overall rating Now I want to allow a given user to rate or be able to re-rate a film. But the latest rating from the user should remain recorder tho..and also I want to display all the films that a given user likes (i.e. the films to which she has given a positive-valued rating). I have written the code for the user liking the film but it doesn't seem to work and i'm not sure what to do with that one eiher. Below is the coding that I've written so far, but i'm stuck on the last bit which I can't seem to code. type Director = String type Title = String type Year = Int type Mark = Int type Rating = (String, Int) type Film = (Title, Director, Year, [Rating]) type Database = [Film] testDatabase :: Database testDatabase = [("Casino Royale", "Martin Campbell", 2006, [("Garry",-8),("Dave", 0)]), ("Blade Runner", "Ridley Scott", 1982, [("Amy",5)]), ("The Fly", "David Cronenberg", 1986, [("Fred",7)]) ] addFilm :: Database -> Title -> Director -> Year -> Database addFilm database title director year = (title, director, year, []): database displayFilms :: Database ->[(Title, Director, Year, Mark)] displayFilms [] = [] displayFilms ((i, j, k, l):xs) |l == [] = (i, j, k, 0) : displayFilms xs |otherwise = (i, j, k, overallRating l) : displayFilms xs overallRating :: [Rating] -> Int overallRating rating = div (sum [j | (_,j) <- rating]) (length rating) filmsByDirector :: Database -> Director -> Database filmsByDirector [] filmDirector = [] filmsByDirector((i, j, k, l):xs) filmDirector |j == filmDirector = (i, j, k, l) : filmsByDirector xs filmDirector |otherwise = filmsByDirector xs filmDirector filmRating :: Database -> Title -> [Rating] filmRating [] filmRat = [] filmRating ((i,j, k, l):xs) filmRat |i == filmRat = (l ++ filmRating xs filmRat) |otherwise = filmRating xs filmRat THIS DOESNT SEEM TO WORK? {- filmsLiked :: Database -> String -> Database filmsLiked [] user = [] filmsLiked ((i,j,k,l):xs) user | == user = i ++ filmsLiked xs user |otherwise = filmsLiked xs user person :: [Rating] -> String -> Bool person [] userName = False person ((n, s):xs) userName |n == userName && s > 0 = True |otherwise = person xs userName -} averageDirector :: Database -> Director -> Int averageDirector [] directorRating = 0 averageDirector ((i,j,k,l):xs) directorRating |j == directorRating = (overallRating l) + (averageDirector xs directorRating) |otherwise = averageDirector xs directorRating filmFromYear :: Database -> Year -> Database filmFromYear [] filmYear = [] filmFromYear ((i,j,k,l):xs) filmYear |k == filmYear = (i,j,k,l) : filmFromYear xs filmYear |otherwise = filmFromYear xs filmYear rateFilm :: [Rating] -> Rating -> [Rating] rateFilm [] rating = rating:[] rateFilm ((user, rating):xs) (newUser, newRating) |newUser == user = ((newUser, newRating):xs) |otherwise = (user, rating) : (rateFilm xs (newUser, newRating)) Stuck on this bit.. {- changeFilm :: Database -> Title -> [Rating] -> Database changeFilm ......... -} Any help is appreciated..thank you! _________________________________________________________________ http://clk.atdmt.com/UKM/go/195013117/direct/01/ We want to hear all your funny, exciting and crazy Hotmail stories. Tell us now

2 1

Can't install haskell-platform on OpenSuse 10.3 without Glut
by David Carter 07 May '10

07 May '10

Hi, I am trying to install haskell-platform on OpenSuse 10.3, i586. I installed ghc 6.12.1 OK, then unpacked haskell-platform-2010.1.0.0.tar.gz, but "configure" gives me configure: error: The GLUT C library is required I have no particular interest in graphics programming and don't want to get involved in installing glut or freeglut as that is likely to cause more installation problems for all I know. I discovered this message and its reply suggesting looking in the bug tracker, http://www.haskell.org/pipermail/haskell-cafe/2009-December/070813.html http://www.haskell.org/pipermail/haskell-cafe/2009-December/070814.html but http://trac.haskell.org is down (I don't remember ever finding it up, in fact). Can anyone suggest a workaround? It's a bit of a showstopper if the main implementation of Haskell can't be installed without having non-free software :-(. Thanks David

4 6

fgl library
by Luca Ciciriello 07 May '10

07 May '10

Hi All. I've found on Hackage site that the data graph library fgl is perfect in order to achieve my aims. I've downloaded the user guide of Martin Erwig and now I'm searching for some sample code. Where can I find some line of code using fgl library? Thanks in advance for any Ideas. Luca. _________________________________________________________________ http://clk.atdmt.com/UKM/go/195013117/direct/01/

1 0

Possible to type a function to a particular constructor?
by Ken Overton 03 May '10

03 May '10

Hi fellow beginners (and everyone else), As an exercise, I'm implementing a simple, untyped lambda calculus: -- a term is a variable, an application, or abstraction (lambda) data T = V String | A (T) (T) | L String (T) deriving (Eq) So I'm writing a function that returns a list of all the free variables in a term and descendants. I can only get it to compile with type: freev :: T -> [T] It'd be nice for the type of that function to be restricted to just variables like: freev :: T -> [V String] -- compile error: "Not in scope: type constructor or class `V'" Is there some way to express that? The error seems to suggest maybe haskell could do it if I'd just say it correctly. I mean, isn't "V String" a type constructor? Thanks, kov

4 4