April 2012 - Beginners

Design of Webmachine in Haskell
by Petar Radosevic 17 Apr '12

17 Apr '12

Hi, I would love to create a port of Webmachine [1] in Haskell so I can learn the language and in the meanwhile create something useful. The problem is, I don't know how I would design such a system in Haskell. Simply put, webmachine is a library in Erlang which let's you create a REST API by binding resources to URI's. The resources have sane defaults, defined with functions, like `resource_exists`, `service_available` and `is_authorized`. You can override these functions per resource to create your own API. Webmachine will then create the appropriate response when a request comes in. I have been thinking how to implement such a system in Haskell, and I have come up with the following. First of, use WAI with Warp to handle the `Request` -> `Response` cycle. That was easy to come up with, but I'm still struggling with the following question: What would a `Resource` be in Haskell? I need something that lets me define sane defaults which I can override per resource. Should I have a `Resource` typeclass which defines all the default functions? Or is there a better way to accomplish this? I hope the above was a bit clear, would appreciate any insights! [1]: http://wiki.basho.com/Webmachine-Resource.html -- Petar Radosevic | @wunki

2 3

Tying the knot with binary trees
by Michael Schober 16 Apr '12

16 Apr '12

Hi folk, as an exercise I'm trying to write a binary tree whose nodes also include a reference to its parent. I've got the data structure I want to use and some helper functions, but there seems to be a bug in insert or find or both (although I assume it's in insert). Here's what I got so far: data BinTree a = Leaf | Node { value :: a , left :: BinTree a , right :: BinTree a , father :: BinTree a } instance Show a => Show (BinTree a) where show Leaf = "[]" show (Node v l r _) = "(Node " ++ show v ++ " " ++ show l ++ " " ++ show r ++ ")" mkRoot :: a -> BinTree a mkRoot value = let root = Node value Leaf Leaf root in root member :: Ord a => a -> BinTree a -> Bool member v Leaf = False member v (Node v' l r _) = if v == v' then True else if v <= v' then member v l else member v r find :: Ord a => a -> BinTree a -> Maybe (BinTree a) find v Leaf = Nothing find v n@(Node v' l r _) = if v == v' then Just n else if v <= v' then find v l else find v r insert :: Ord a => a -> BinTree a -> BinTree a insert v' Leaf = mkRoot v' insert v' n@(Node v l r f) = insert' v' n f where insert' :: Ord a => a -> BinTree a -> BinTree a -> BinTree a insert' v' Leaf f' = Node v' Leaf Leaf f' insert' v' n@(Node v l r f) f' = if v' == v then n else if v' <= v then let inserted = insert' v' l result result = Node v inserted r f in result else let inserted = insert' v' r result result = Node v l inserted f in result I thought this should do the trick, but here's what I get in ghci: *Main> find 3 (insert 7 (insert 3 (insert 5 Leaf))) >>= return . parent Just (Node 5 (Node 3 [] []) []) I'm expecting to see Just (Node 5 (Node 3 [] []) (Node 7 [] [])) Inserting into an empty tree (i.e. a leaf) works fine, as does mkRoot. Also, it seems as inserting an existing value works fine as well - but obviously I couldn't test that one exhaustingly so far. I'm grateful for any pointers towards a solution. Best regards, Michael P.S.: For those unfamiliar with this problem, here is a list of URLs of what I read of the subject: [1] http://www.haskell.org/haskellwiki/Tying_the_Knot#Migrated_from_the_old_wiki [2] http://debasishg.blogspot.de/2009/02/learning-haskell-solving-josephus.html [3] http://blog.sigfpe.com/2006/12/tying-knots-generically.html

4 6

cabal update behaviour
by Hector 15 Apr '12

15 Apr '12

Hi all, these are the steps I followed to get Haskell installed in my machine[1]: https://gist.github.com/2389324 The installation completed successfully, and gave me these instructions: Now do "cabal update" to initialize the package list But when I do, I get this message: Note: there is a new version of cabal-install available. To upgrade, run: cabal install cabal-install Which I do, but after another "cabal update" I'm given the same message, meaning that cabal-install was not actually updated. Any ideas on why this is happening? Best, -- Héctor [1] Specs for my machine: $ lsb_release -a No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 10.04.4 LTS Release: 10.04 Codename: lucid $ uname -a Linux 2.6.32-33-generic #70-Ubuntu SMP x86_64 GNU/Linux

2 1

error: Not in scope: data constructor `BinTree'
by Kak Dod 14 Apr '12

14 Apr '12

if i compile the following code I get "bintree.hs:3:13: Not in scope: data constructor `BinTree'" data BinTree a = Node BinTree a BinTree a | EmptyBinTree deriving Show b1 = (Node (BinTree 3) EmptyBinTree) please help -kak

4 7

return (user defined type) confusion?
by umptious 13 Apr '12

13 Apr '12

I'm bashing together a simple command line geometry thing that let's you create shapes, move them around them, delete and edit them, test for intersections, etc. The motivation is to get a grip on Haskell IO and how it interacts with user defined datatypes. I'm using GHC. The problem, which I've tried to show as simply as possible in the abstracted code below, is that something seems to go horribly wrong with un-boxing (is this the correct terminology?) user defined type X's from IO(X)'s. In the example below: - enterCircle makes a Shape using the Circle cstr and boxes it in a return. And the type checker is happy with the function. - But when the Shape is un-boxed in mainLoop then Haskell's type system (or GHC's?) has lost track of its Shapeness. Instead it sees it as a Pt2->Float. Which makes some sort of sense, but is useless (or at least needs ugly hacks to overcome) if you want to put your Circle in a List of Shapes. - Even worse then the loss of Shapeness is that Haskell is letting me type enterCircle as returning an IO(Shape) - indeed that's what :t gives me in GCHI - but then disagrees with this type in mainLoop! This isn't what I was expecting from the vaunted Haskell type system; can someone explain this contradiction? So. Does return fail to wrap up type info about user defined types in general? Is this correct Haskell behaviour? Or have I done something wrong? If I haven't done something wrong, is there a standard idiom for overcoming this? I suppose I could use a show to convert the Circle to a string and then a read to convert it back in mainLoop's where, but this will mean uglier branching than I planned (the idea was to have a Map of fns that return shapes and just one branch in mainLoop to handle every kind of shape.) And it seems to make something of a nonsense of Haskell's type system as the entire Shape hierarchy will be reduced to IO(String) for these operations. import Data.Map as Map data Pt2 = Pt2 {x::Float, y::Float} deriving (Show, Read) data Shape = Circle {origin::Pt2, radius::Float} | Square {origin::Pt2, side ::Float} | Rect {origin::Pt2, other ::Pt2} deriving (Show, Read) enterCircle :: IO(Shape) enterCircle = return (r) -- needs to rtn io (shape) because will take input here where r = Circle{origin=Pt2{x=1,y=2}, radius=4} mainLoop :: [Shape] -> IO (String) mainLoop shapes = do putStrLn "?" cmd <- getLine if cmd=="q" then return ("done") else do y <- enterCircle mainLoop shapes' where shapes' = y : shapes main = mainLoop []

2 1

A "show" error
by bahadýr altan 12 Apr '12

12 Apr '12

Hi, I'm trying to process on a tree with this function : f (Branch x (Branch y y1 y2) (Branch z z1 z2)) = (x,y,y1,y2,z,z1,z2) and my tree declaration is this : data Tree = Empty | Branch Integer Tree Tree deriving (Show) And I'm getting this error : Main> f (Branch 12 (Branch 15 Empty Empty) (Branch 28 Empty Empty)) ERROR - Cannot find "show" function for: *** Expression : f (Branch 12 (Branch 15 Empty Empty) (Branch 28 Empty Empty)) *** Of type : (Integer,Integer,Tree,Tree,Integer,Tree,Tree) I'll be happy if you help me to get rid of this error..

5 9

netwire and "dynamic" list of objects
by Nathan Hüsken 12 Apr '12

12 Apr '12

Hey, I have a simple SDL based graphic demo, which I am trying (for educational purposes) to rewrite utilizing netwire. The demo basically consists of a State which contains a list of object states. data ObjectState = ... data State = State [ObjectState] Then there is a function updateState which updates all objects, removing some and adding some new ones. updateState :: State -> State updateState (State oss) = State (catMaybes $ map updateObject $ newObjects ++ oss) where updateObject :: ObjectState -> Maybe ObjectState updateObject = ... newObjects :: [ObjectState] newObjects = ... OK, now I wrote a wire: type MyWire = Wire.Wire () (Kleisli IO) wire :: GameState -> MyWire () GameState wire initGS = proc _ -> do rec gs' <- Wire.delay initGS -< gs gs <- arr updateState -< gs' returnA -< gs Works wonderfully :). But I avoided rewriting "updateState" in wire format. But I have no Idea how to do that. How do I handle a dynamic list of objects in a wire? Thanks! Nathan

2 3

Converting a particular warning into a fatal error.
by Michael Litchard 12 Apr '12

12 Apr '12

Is there a way to tell ghc to turn only `incomplete-pattern` warnings into fatal errors, as opposed to `-Werror`?

1 0

Wrong type inferred for polymorphic function in a case alternative
by Ramin Honary 12 Apr '12

12 Apr '12

Hi everyone, I am using GHC 7.0.4. I boiled-down my problem into a simple program which I attached to this message. I don't know how to get the program to compile. I need to pass arbitrary class member functions as a parameter to a very large, complex, polymorphic function called "apply", and have this class member function applied to abitrary fields of data constructors of a very large, complex data type. What I expect to happen is this: For each case alternative, GHC will infer the type of the polymorphic function from the matched pattern in the case alternative, and use that inference to select the correct instance function for that type. So for case alternative (TA x, TA y) -> TA (fn x y), the correct instance for "fn" is selected based on the type of "x" and "y" which are determined by the pattern match, and by the type of "TA". If that case alternative does not match, then it should try the next alternative (TB x, TB y) -> TB (fn x y), and since "fn" is polymorphic, it will still be able to select the correct instance for the type given by the pattern match. So I can pass "f1" or "f2" or "f3" to "apply" as the "fn" parameter, and have the correct instance of "f1" or "f2" or "f3" be selected depending on where in the "case" expression "fn" is used. What is actually happening is: GHC arbitrarily selects one of the case alternatives (usually the last case alternative, sometimes the first) and decides that from now on this one type is the type for "fn" for every case alternative. So for every other case alternative, it decides that "fn" is of the wrong type and rejects the program, rather than trying to infer a different type for "fn" specific to this case alternative. Said another way, the type inference algorithm is being too strict, and I would like it to be more lazy, computing the type once per each case alternative, rather than computing the type only once for the entire case statement. Is there a language option for that? I have tried using language options, like -XNoMonoPatBinds -XNoMonomorphismRestriction -XMonoLocalBinds -XPolymorphicComponents -XImpredicativeTypes, but none of those help. My current solution is to use Template Haskell and make the "apply" function a splice, so the whole case statement is copied to every place it is used (which takes a very, very long for it time to comple). Is there a way to do this without Template Haskell? Can I get GHC to behave as I expected it to? Thanks, Ramin

2 2

Type of function with constant pattern
by j.romildo＠gmail.com 11 Apr '12

11 Apr '12

Hello. Given the following function definitions f 0 = True g False = True ghc infers the following types for the functions: f :: (Eq a, Num a) => a -> Bool g :: Bool -> Bool Why f has "Eq a" in the context in ts type, and g does not? As both are defined using a constant pattern, I expected none of them should require the type of the argument to be instance of Eq. Romildo

5 6