- Haskell-Cafe - Haskell.org

Re: Why is type 'b' forced to be type 'm a' and not possibly 'm a -> m a' (Anatoly Zaretsky)
by Vivian McPhail 18 Sep '06

18 Sep '06

> Message: 6 > Date: Fri, 15 Sep 2006 18:36:35 +0300 > From: "Anatoly Zaretsky" <anatoly.zaretsky(a)gmail.com> > Subject: Re: [Haskell-cafe] Why is type 'b' forced to be type 'm a' > and not possibly 'm a -> m a' > To: "Vivian McPhail" <vivian.mcphail(a)paradise.net.nz> > Cc: Haskell Cafe <haskell-cafe(a)haskell.org> > Message-ID: > <a21cb60a0609150836u5695cf41jfbb31530105bec3a(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > On 9/15/06, Vivian McPhail <vivian.mcphail(a)paradise.net.nz> wrote: > > > > class Forkable a where > > fork :: String -> a -> a -> a > > > > ... > > {- > > instance (Monad m, Forkable (m a), Forkable b) => Forkable (m a -> b) where > > fork n a1 a2 a = do > > a' <- a > > fork n (a1 $ return a') (a2 $ return a') > > -} > > > > Let's do manual type checking. > First, fork :: Forkable a => String -> a -> a -> a > So for Forkable (m a -> b) > fork :: String -> (m a -> b) -> (m a -> b) -> m a -> b > Then > fork n a1 a2 a :: b > But you define it as > fork n a1 a2 a = do {...} > So it should be of type Monad t => t a, not just any `b'. > > Instead, you can define > instance (Monad m, Forkable (m b)) => Forkable (m a -> m b) where > ... > Well, I can partially instantiate what I am trying to achieve by enumerating cases. Note that when the first type is a monadic type the computation gets evaluated and then forked, but when the first type is a function it merely gets passed. My problem is that there are a very large number of possible cases. So in the case Forkable (m a -> b), a number of instances of which I can implement (e.g. Forkable (m a -> m a -> m a), Forkable ((m a -> m a) -> m a), and Forkable (m a -> (m a -> m a) -> m a)), I don't see why 'b' should necessarily typecheck to 't t1'. What I would like to be able to do is differentiate between Forkable (m a -> b) and Forkable (<function type> -> b). By the way, the following code typechecks and runs correctly, my problem is that enumerating all possible types requires five factorial (120) different instances, and to a lazy functional programmer who can 'see' the pattern it seems that there must be a nicer way of achieving my end. \begin{code} instance (Monad m, Forkable (m a)) => Forkable (m a -> m a) where fork n a1 a2 a = do a' <- a fork n (a1 $ return a') (a2 $ return a') instance (Monad m, Forkable (m a)) => Forkable (m a -> m a -> m a) where fork n a1 a2 a b = do a' <- a fork n (a1 $ return a') (a2 $ return a') b instance (Monad m, Forkable (m a)) => Forkable ((m a -> m a) -> m a) where fork n a1 a2 a = do fork n (a1 a) (a2 a) instance (Monad m, Forkable (m a)) => Forkable (m a -> (m a -> m a) -> m a) where fork n a1 a2 a b = do a' <- a fork n (a1 $ return a') (a2 $ return a') b \end{code} > Note that to compile it you also need -fallow-undecidable-instances > and -fallow-overlapping-instances. > > -- > Tolik > Thanks for your help so far! > > Cheers, Vivian -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.405 / Virus Database: 268.12.4/449 - Release Date: 15/09/2006

1 0

Problems interpreting
by Carajillu 18 Sep '06

18 Sep '06

Hi, I'm a student and I have to do a program with Haskell. This is the first time I use this languaje, and I'm having problems with the indentation. I want to check if this function is correct, but when I try to make the GHCi interpret it, I get "line 18:parse error (possibly incorrect indentation)" The function is: -- Replaces a wildcard in a list with the list given as the third argument substitute :: Eq a => a -> [a] -> [a] -> [a] substitute e l1 l2= [check_elem c | c <- l1] where check_elem::Eq a => a -> [a] check_elem x = if x == e then return l2 -- Tries to match two lists. If they match, the result consists of the sublist -- bound to the wildcard in the pattern list. (line 18) match :: Eq a => a -> [a] -> [a] -> Maybe [a] match _ _ _ = Nothing {- TO BE WRITTEN -} Thank you for your attention! -- View this message in context: http://www.nabble.com/Problems-interpreting-tf2290155.html#a6360687 Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

4 3

Microsoftish Haskell
by Kaveh Shahbazian 17 Sep '06

17 Sep '06

1 - Prologue I am writing C# WEB/Desktop/Embeded applications for 3 years by now and before that, I was writing Visual C++/MFC/COM sort of things. I can say now that everything that come from microsoft is a handcraft of a saleman that has no guaranty and will break before you get home and even try it! For example see generics in C#. These thing are just a code generatore utility that is embeded in Visual Studio 2005 and Microsoft "JUST" has named it "GENERICS". For comparation see generics in Java 5.0. 2 - Question Will Haskell become another pet for Microsoft? If Yes/No, Why? (There are many issues around licensing GHC as you'v seen in this mailing list and I think Haskell already HAS some big problems that prevent others to use it confidently.)

2 3

Serialising types with existential data constructors
by Misha Aizatulin 16 Sep '06

16 Sep '06

hello all, I've been using existentially quantified data constructors like > data Box = forall a. Cxt a => Box a quite successfully for a while. But now I am trying to implement the Load/Save mechanism and getting stuck with that. It's not hard to write a Box into a file, but how do I get it back? Has anyone solved the same problem before? I would be very thankful for any suggestions! If Template Haskell would support finding out all instances of a given class, I could generate a function that would map Names of types to appropriate Box readers. In the file I would write entries like <NameOfTypeInTheBox> : <BoxContents> But sadly, TH doesn't allow that yet. Cheers, Misha

5 7

How to use clipboard in Gtk2Hs
by Yu, Jae-Myoung 16 Sep '06

16 Sep '06

I'm making a text editor with Gtk2Hs, and I want to implement copy & paste with TextView. I read the api document but I can't find how to use the Gtk.Clipboard object. Without context menu basically included in the TextView widget, How can I implement copy & paste?

2 1

Deferred instance declarations (context without type variables)
by Misha Aizatulin 16 Sep '06

16 Sep '06

hello, I have a question about context in type signature. I would like to write a function, say (f :: T -> T) which also relies on an instance of class C being defined for T. The problem is, I don't want this instance defined at the same time f is defined, instead I would like to defer this definition until f is called (in some other module). Naively the code would look like this: ================================= module DefineF where class C a where fC :: a -> a data T = T f :: (C T) => T -> T f T = fC T ================================= module CallF where instance C T where fC = id call = f T ================================= The definition of f in DefineF won't compile because "All of the type variables .. are already in scope". Could you recommend any other way to achieve what I am trying to do? An interesting thing is that I can easily make the code compile with the same meaning by changing the declaration of T to include a dummy type variable: > data T a = T it's just a pity that I have to trick the compiler in such an ugly way. Cheers, Misha

1 0

Re: [Haskell-cafe] Re: ambiguous partially defined type problem
by Maarten 15 Sep '06

15 Sep '06

Dear Brian, Maarten wrote: > Brian Hulley wrote: > > Alternatively, you could wrap the custom part within the node as in: > > data Node = forall cust. ICustom cust => Node cust Common > > getCommon :: Node -> Common > getCommon (Node cust com) = com > Thanks. This really helped. The main thing (I think) that you put the custom part behind an interface. After this I separated the custom and common part of two 'piggy bagged' state transformers, so one can access the functionality separately. The state transformers made into active object by putting them behind a channel in a separate thread and one can invoke actions by writing to the channel. The common functionality provides the connections between the active objects. In this way I would like to create some sort of 'agent' structure, that receive message and process them in their own thread. So far this works quite neat. Wonder if this is they way to go though... Only update (see code below) is a bit ugly (I have no idea why I need fixCastUpdate) and Node itself is probably not necessary, so one level of indirection could be removed. Rest is quite straight forward. Thanks again. Maarten ... (imports) data Node = forall cust. (ICustom cust) => Node cust deriving (Typeable) instance Show Node where -- just for debugging show (Node a) = "Node (" ++ show a ++ ")" class (Show a, Typeable a) => ICustom a where getVal :: forall b cust. (Typeable b, ICustom cust) => a -> (cust -> b) -> Maybe b getVal a f = case cast a of Nothing -> Nothing Just cust -> Just (f cust) -- update :: oif -> (forall a. (ObjectIFace a) => a -> a) -> IO oif update :: a -> (forall b. (ICustom b) => b -> b) -> a update a f = f a instance ICustom Node where getVal (Node n) f = getVal n f update (Node n) f = Node (update n f) type NodeState a = StateT Node (StateT Common IO) a type Connection = Chan (NodeState ()) type Connections = [Connection] instance Show Connection where show o = "Chan (StateT Node (StateT Common IO) ())" -- common part data Common = Common { uid::Integer, connections::Connections } deriving (Show,Typeable) -- custom data data Custom = Custom { val::Integer } deriving (Show,Typeable) instance ICustom Custom where data Custom2 = Custom2 { val2::Integer } deriving (Show,Typeable) instance ICustom Custom2 where -- some function to use common functionality uidM :: NodeState Integer uidM = lift $ gets uid addNodeM :: Connection -> NodeState () addNodeM n = lift $ modify (\s -> addNode s n) where addNode (Common i ns) nn = (Common i (nn:ns)) getNodeM :: Integer -> NodeState Connection getNodeM i = do s <- lift $ get return (getNode s i) where getNode (Common _ ns) i = (ns!!(fromInteger i)) getValM f = do s <- get return (getVal s f) updateM :: forall a b. (ICustom a, ICustom b) => (a -> b) -> NodeState () updateM f = do s <- get let s' = update s (fixCastUpdate f) put s' fixCastUpdate f st = case (cast st) of Nothing -> st Just es -> case cast (f es) of Nothing -> st Just g -> g getStateM = get -- function to create active node functionality action [] = return () action (e:es) = do e s <- get -- just for debugging lift $ lift $ putStrLn $ show s action es newBaseState = do uid <- newUnique return (Common ((toInteger .hashUnique) uid) []) initAction list state = do bs <- newBaseState execStateT (execStateT (action list) state) bs return () send chan action = writeChan chan action sync chan f = do mv <- newEmptyMVar send chan (f' mv) a <- takeMVar mv return a where f' mv = do a <- f lift $ lift $ putMVar mv a newActiveObject action state = do chan <- newChan cs <- getChanContents chan forkIO (action cs state) return chan -- example main = do let n1 = Node (Custom 5) let n2 = Node (Custom2 6) let n3 = Node (Custom2 7) chan <- newActiveObject initAction n1 chan2 <- newActiveObject initAction n3 let l = [chan, chan2] mapM_ (\ch -> send ch (addNodeM chan)) l mapM_ (\ch -> send ch (addNodeM chan2)) l r <- mapM (\ch -> sync ch (getNodeM 0)) l putStrLn $ "r:" ++ show r r2 <- mapM (\ch -> sync ch (uidM)) l putStrLn $ "r2:" ++ show r2 r3 <- mapM (\ch -> sync ch (getValM val)) l putStrLn $ "r3:" ++ show r3 mapM_ (\ch -> send ch (updateM (\s -> s { val2 = 100 }))) l r5 <- mapM (\ch -> sync ch (getStateM)) l putStrLn $ "r5:" ++ show r5 getChar return ()

2 3

Why is type 'b' forced to be type 'm a' and not possibly 'm a -> m a'
by Vivian McPhail 15 Sep '06

15 Sep '06

Dear Haskell Cafe, I have a problem I can't get my head around. The code below sets the problem out. What I need to be able to do is commented out. This code works, the only problem is that what I need is that an argument will be evaluated before it is passed, so ((and fries eats) eggs) has a single `eggs` (fries1 eggs2 and3 eats4 egg2) not (fries1 eggs2 and3 eats4 eggs5). The code that doesn't work is commented out at the bottom. I'm not sure the behaviour of ghc is correct, because when it typechecks it tries to unify `b = t t1` but `b` could actually be `t t1 -> t t1`. I want to be able to specify that when the first argument of `b` is of type `m a` that fork should run it and _then_ fork the argument to the first two arguments of 'fork'. The instance for (a -> b) covers the rest of the possibilities. just type "run test[1-4]" to see results. \begin{code} {-# OPTIONS_GHC -fglasgow-exts -fno-monomorphism-restriction #-} module Fork where {--------------------------------------------------------------------------- --} import Prelude hiding (and) import Control.Monad.State {--------------------------------------------------------------------------- --} data NRef = NS0 String | NS1 String NRef | NS2 String NRef NRef deriving(Show) {--------------------------------------------------------------------------- --} data UniqueS = US { nums :: [String] } deriving(Show) type USM a = StateT UniqueS IO a newUniqueS :: UniqueS newUniqueS = US { nums = [ show x | x <- [1..] ] } freshInstance :: String -> USM String freshInstance x = do (f:fs) <- gets nums put $ US { nums = fs } return $ x ++ f {--------------------------------------------------------------------------- --} single x = do x' <- freshInstance x return $ NS0 x' unary x n = do x' <- freshInstance x n' <- n return $ NS1 x' n' binary x n1 n2 = do x' <- freshInstance x n1' <- n1 n2' <- n2 return $ NS2 x' n1' n2' {--------------------------------------------------------------------------- --} foxy = single "foxy" eggs = single "eggs" golden = unary "golden" white = unary "white" fries = binary "fries" eats = binary "eats" {--------------------------------------------------------------------------- --} class Forkable a where fork :: String -> a -> a -> a instance (Forkable a, Forkable b) => Forkable (a -> b) where fork n a1 a2 a = fork n (a1 a) (a2 a) {- instance (Monad m, Forkable (m a), Forkable b) => Forkable (m a -> b) where fork n a1 a2 a = do a' <- a fork n (a1 $ return a') (a2 $ return a') -} {--------------------------------------------------------------------------- --} instance Forkable (USM NRef) where fork n a1 a2 = do a1' <- a1 a2' <- a2 return $ NS2 n a1' a2' {--------------------------------------------------------------------------- --} and = fork "and" test1 = (and foxy eggs) test2 = (and golden white) eggs test3 = (and fries eats) foxy eggs test4 = (eats foxy (and (golden eggs) (white eggs))) run x = runStateT x newUniqueS >>= (putStrLn . show . fst) \end{code -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.405 / Virus Database: 268.12.4/448 - Release Date: 14/09/2006

2 1

parsing long numbers
by Andrea Rossato 15 Sep '06

15 Sep '06

Hello! I'm trying to learn parsing and parser combinations in Haskell, using, as usual, Wadler's Monads in Functional Programming as my text book. Everything works fine except for a small but annoying problem related to "read". I'm sure it must be something easy, some kind of stupid faq. Still I'm not able to find a way out. I created a simple parser and made it an instance of monad and MonadPlus. I then created "iterateP" to combine recursive parsers, and "filterP" to apply some filters. Then I created a function "number", to parse numbers. This function returns a string and works fine. But I wanted to have integers back from parsing. So I created a filter, "digitS", and a new parser for numbers, "number1", that applies recursively digitS and should be returning an Int using "read". The problem is that, when I run it with strings containing a number more then 10 digit long, I get unexpected integers back: *Main> runP number1 "1234567890 and the rest" [(1234567890," and the rest")] *Main> runP number1 "12345678901 and the rest" [(-539222987," and the rest")] Obviously if I use Parsec I can parse that number perfectly. So I tried with another approach: "number2" recursively applies a filter, "digitI", that returns an Int. asNumber is a function that takes a list of single digit integers and returns the corresponding integer. *Main> runP number2 "12345678901 and the rest" [(Just 1234567890," and the rest")] *Main> runP number2 "12345678901 and the rest" [(Just (-539222987)," and the rest")] The very same result. Can you please help me understand why I seem not to be able to get the number I'd like to get? As I said, I think I'm missing something that must be pretty obvious, but still I cannot see it! Thanks for your kind attention. Andrea ps: sorry for such a long message. Moreover, here's the code: module Main where import Control.Monad import Data.Char newtype M a = S {unpack :: String -> [(a,String)]} instance Monad M where return a = S $ \s -> [(a,s)] m >>= f = S $ \s -> [(b,z) | (a,y) <- unpack m s, (b,z) <- unpack (f a) y] instance MonadPlus M where mzero = S $ \x -> [] mplus a b = a `bchoice` b bchoice (S m) (S m1) = S $ \s -> case m s of [] -> m1 s other -> other iterateP m = do { a <- m ; b <- iterateP m ; return (a:b) } `mplus` return [] filterP p = S (\xs -> case xs of [] -> [] (x:xs') -> if p x then [(x,xs')] else []) number = do { a <- filterP isDigit ; b <- number ; return (a:b) } `mplus` return [] digitS = do a <- filterP isDigit return a number1 :: M Int number1 = do a <- iterateP digitS return (read a) -- a different approach maybeAdd a b = do x <- a y <- b return (x + y) asNumber :: [Int] -> Maybe Int asNumber [] = Nothing asNumber (x:[]) = Just x asNumber (x:xs) = Just (x * 10 ^ length xs) `maybeAdd` asNumber xs digitI = do a <- filterP isDigit return $ ord a - ord '0' number2 :: M (Maybe Int) number2 = do a <- iterateP digitI return $ asNumber a runP (S f) = f

2 1

evaluate vs seq
by Michael Shulman 15 Sep '06

15 Sep '06

The GHC documentation says that (evaluate a) is not the same as (a `seq` return a). Can someone explain the difference to me, or point me to a place where it is explained? Mike

4 13