August 2006 - Haskell-Cafe

state and exception or types again...
by Andrea Rossato 30 Aug '06

30 Aug '06

Hello! Sorry if I keep bothering, but I'm still trying to understand types and monads. Now I'm trying to create a statefull evaluator, with output and exception, but I'm facing a problem I seem not to be able to conceptually solve. Take the code below. Now, in order to get it run (and try to debug) the Eval_SOI type has a Raise constructor that produces the same type of SOIE. Suppose instead it should be constructing something like Raise "something". Moreover, I wrote a second version of >>=, commented out. This is just to help me illustrate to problem I'm facing. Now, >>= is suppose to return Raise if "m" is matched against Raise (second version commented out). If "m" matches SOIE it must return a SOIE only if "f a" does not returns a Raise (output must be concatenated). I seem not to be able to find a way out. Moreover, I cannot understand if a way out can be possibly found. Something suggests me it could be related to that Raise "something". But my feeling is that functional programming could be something out of the reach of my mind... by the way, I teach Law, so perhaps you'll forgive me...;-) If you can help me to understand this problem all I can promise is that I'll mention your help in the tutorial I'm trying to write on "the monadic way"... that seems to lead me nowhere. Thanks for your kind attention. Andrea the code: data Eval_SOI a = Raise { unPackMSOIandRun :: State -> (a, State, Output) } | SOIE { unPackMSOIandRun :: State -> (a, State, Output) } instance Monad Eval_SOI where return a = SOIE (\s -> (a, s, "")) m >>= f = SOIE (\x -> let (a, y, s1) = unPackMSOIandRun m x in case f a of SOIE nextRun -> let (b, z, s2) = nextRun y in (b, z, s1 ++ s2) Raise e1 -> e1 y --Is it right? ) -- (>>=) m f = case m of -- Raise e -> error "ciao" -- why this is not going to happen? -- SOIE a -> SOIE (\x -> -- let (a, y, s1) = unPackMSOIandRun m x in -- let (b, z, s2) = unPackMSOIandRun (f a) y in -- (b, z, s1 ++ s2)) incSOIstate :: Eval_SOI () incSOIstate = SOIE (\s -> ((), s + 1, "")) print_SOI :: Output -> Eval_SOI () print_SOI x = SOIE (\s -> ((),s, x)) raise x e = Raise (\s -> (x,s,e)) eval_SOI :: Term -> Eval_SOI Int eval_SOI (Con a) = do incSOIstate print_SOI (formatLine (Con a) a) return a eval_SOI (Add t u) = do a <- eval_SOI t b <- eval_SOI u incSOIstate print_SOI (formatLine (Add t u) (a + b)) if (a + b) == 42 then raise (a+b) " = The Ultimate Answer!!" else return (a + b) runEval exp = case eval_SOI exp of Raise a -> a 0 SOIE p -> let (result, state, output) = p 0 in (result,state,output) --"Result = " ++ show result ++ " Recursions = " ++ show state ++ " Output = " ++ output --runEval (Add (Con 10) (Add (Con 28) (Add (Con 40) (Con 2))))

3 10

'Compiling' expression graph into Arrows
by Daniel McAllansmith 29 Aug '06

29 Aug '06

Hello, I have a graph of function applications which I would like to 'compile' into an Arrow, specifically the SF Arrow from Yampa. I'd appreciate any advice on how I might go about this. The graphs, of which there will be many, will be constructed at runtime and will be executed for extended periods of time so execution time will dominate compilation time. The graphs might also need to be passed around and executed 'elsewhere', where 'elsewhere' will almost certainly be the same GHC version but maybe on different architectures. The hand-written 'compilation' in the attached file gives a rough idea of what I start with and what I need to end up with. I could just automate the hand-written process by generating the SFs and arrow-syntax blocks, write them into a file then call GHC... seems a bit cowboyish. Or, write the graph into a template file that has Template Haskell to generate the SFs and arrow-syntax blocks then call GHC... almost as cowboyish? Or, fold the graph up using loop and arr... probably the simplest, once I get rid of my sugar dependency and figure out how to use loop that is ;). Or, maybe I can actually get the expression graph into a GHC api representation somehow and let it work marvels of optimisation on the expression before turning it into an arrow. Any idea on the relative merits of these, or other ideas? Can GHC optimise arrow code much during compilation, or will code folded up at runtime using loop/arr be just as efficient? Could GHC condense the functions of multiple graph nodes into a single function for conversion to an SF arrow? Will the necessity of introducing delays rule out using GHC to simplify the raw expression? Thanks Daniel

1 0

state and exception or types again...
by tpledger＠ihug.co.nz 28 Aug '06

28 Aug '06

Andrea Rossato wrote: > > Now I'm trying to create a statefull evaluator, with output and > exception, but I'm facing a problem I seem not to be able to > conceptually solve. If a computation fails in your monad, do you still want to return a value of the result type? I'd expect not, and hence remove the 'a' from the 'Raise' constructor. data Eval_SOI a = Raise (State -> (State, Output)) | SOIE (State -> (a, State, Output)) The above is very similar to using the monad foundation classes: import Control.Monad.Error import Control.Monad.State hiding (State) import Control.Monad.Writer type Eval_SOI = ErrorT String (StateT State (Writer Output)) ...assuming that you're happy to send back a descriptive String when a computation fails. Have a look at the Control.Monad.Error source code, to see how 'instance Error a => Monad (Either a)' is defined. It's the sort of thing you were trying to do in your 'instance Monad Eval_SOI'. Regards, Tom

1 0

Win32-package and Windows error codes in exceptions
by Esa Ilari Vuokko 28 Aug '06

28 Aug '06

Hi, Recently, I've been annoyed how Win32-package throws user errors, ie String, on failures. The System.Win32-being rather low level, it often makes sense for caller to catch some specific errors. I don't really understand how, or if, I could extend System.IO.Error. unix- package uses it via Foreign.C.Error, which makes me wonder if the underlying functionality in there should be extended. Is the only or best choice to go for Dynamic exceptions? That introduces a subtle difference for (already existing) programs that rely on exceptions to be presentable via Show and default handler. Best regards, --Esa Ilari Vuokko

1 0

Re: [Haskell] Exists any null instruction in Haskell?
by Neil Mitchell 26 Aug '06

26 Aug '06

[Moving to haskell-cafe] Hi, > I want to know if it exists a null instruction (a instruction > that do anything Not really, since instructions don't "do anything" anyway - they only compute values. If you are in the IO monad then "return ()" is the do nothing. In the function world, "id" is the do nothing. In recursive list processing, then something like [] is probably the "base case" in a recursive function, so stops the recursivity. If you give a more concrete example, then maybe we can give you a better answer. Thanks Neil

1 0

OOP vs type classes: Everything is object?
by Bulat Ziganshin 26 Aug '06

26 Aug '06

Hello haskell-cafe, i've added section http://haskell.org/haskellwiki/OOP_vs_type_classes#Everything_is_object.3F which lists cases when we DON'T use type classes to emulate OOP classes features -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0

Re: [Haskell-cafe] monads once again: a newbie perspective
by Andrea Rossato 26 Aug '06

26 Aug '06

Il Fri, Aug 25, 2006 at 01:13:58PM -0400, Cale Gibbard ebbe a scrivere: > Hey cool, a new monad tutorial! :) > > Just out of interest, have you seen my Monads as Containers article? > http://www.haskell.org/haskellwiki/Monads_as_Containers > > Let me know what you think of it. I find that often newcomers to > monads will find the container perspective easier to grasp before > moving on to treating monads as an abstraction of computation, but > that side of things needs coverage too. :) Sure I've read it! I must confess I find it difficult, though. I mean, the exemplification part is very interesting but, for me, it was too difficult to connect it to the code I was looking at. This is way I decided this approach: let's start building a monad and see what it is by actually looking it at work. The evaluator is a very simple piece of code, you can clearly see what it does. And then you start building up your knowledge by expanding it. Take into account that, for me, *writing* that tutorial is *the* way to get to grasp all the concepts behind the type system (and monads). As you can see in a thread below, I'm studying hard in order to find out the proper continuation of the tutorial, and for my learning. So far I'll be able to describe the code of a statefull evaluator that produces output. Then I'd like to add exception handling. Let's see if I'll be able to get that far... By the way, I'll soon add links to the other important tutorials on monads: yours, IO Inside, All about monads. At the I'd like to be able to link A Gentle introduction, the Haskell 98 Report and, why not, "Write yourself a Scheme in 48 hours". That will be it. Thanks for your kind attention. Andrea

2 1

difference between type and newtype
by Andrea Rossato 26 Aug '06

26 Aug '06

Hello! I cannot understand this piece of code: type Z = Int type T a = Z -> (a, Z) newtype T1 a = T1 (Z -> (a,Z)) mkT :: a -> T a mkT a = \x -> (a, x) mkT1 :: a -> b -> (a,b) mkT1 a = \x -> (a, x) why mkT is a type constructor and mkT1 seems not to be? thanks for your kind attention. Andrea

6 18

Space leak whilst implementing streams
by ephemeral.elusive＠gmail.com 26 Aug '06

26 Aug '06

Hello, I have been using arrows to implement stream processors. At first, I tried using the implementation presented in John Hughes' AFP arrows lectures. However, this appeared to have a space leak in its implementation of the left operator for ArrowChoice. I found a way to remove this space leak, however, I do not really understand why there was a space leak in the first place. I would really appreciate any light that could be shed on this. Below I include the AFP implementation (SF) and a modified implementation that no longer has the space leak (SF'). I am using ghc 6.4.2. Many thanks. import Control.Arrow import Data.Maybe main = test test = print (runSF p (repeat 1)) test' = print (runSF' p (repeat (Just 1))) -- heap profile appears to grow linearly for test, but not test' p :: ArrowChoice a => a Int (Either Int Int) p = arr Right >>> left (arr id) newtype SF a b = SF {runSF :: [a] -> [b]} instance Arrow SF where arr f = SF (map f) SF f >>> SF g = SF (f >>> g) first (SF f) = SF (unzip >>> first f >>> uncurry zip) instance ArrowChoice SF where left (SF f) = SF (\xs -> combine xs (f [y | Left y <- xs])) where combine (Left _:xs) (z:zs) = Left z :combine xs zs combine (Right r:xs) zs = Right r:combine xs zs combine [] _ = [] -- SF' does not exhibit the space leak newtype SF' a b = SF' {runSF' :: [Maybe a] -> [Maybe b]} instance Arrow SF' where arr f = SF' (map (maybe Nothing (Just . f))) SF' f >>> SF' g = SF' (f >>> g) first (SF' f) = SF' (maybe_unzip >>> first f >>> uncurry maybe_zip) where maybe_unzip = foldr mu ([],[]) where mu Nothing ~(xs,ys) = (Nothing:xs, Nothing: ys) mu (Just (x,y)) ~(xs,ys) = (Just x:xs, Just y: ys) maybe_zip = zipWith mz where mz (Just x) (Just y) = Just (x,y) mz Nothing Nothing = Nothing instance ArrowChoice SF' where left (SF' f) = SF' (\xs -> xs `combined` f (map drop_right xs)) where combined = zipWith merge_left drop_right Nothing = Nothing drop_right (Just (Left l)) = Just l drop_right (Just (Right _)) = Nothing merge_left Nothing Nothing = Nothing merge_left (Just (Left _)) (Just x) = Just (Left x) merge_left (Just (Right r)) Nothing = Just (Right r)

2 1

ghc profiling
by Jeff Polakow 25 Aug '06

25 Aug '06

Hello, When I compile my program without profiling: bash-3.1$ ghc --make -auto-all -prof -O -o analysis analysis.hs Chasing modules from: analysis.hs Compiling Main ( analysis.hs, analysis.o ) Linking ... everything works fine. However when I compile with profiling: bash-3.1$ ghc --make -auto-all -prof -O -o analysis analysis.hs Chasing modules from: analysis.hs Could not find module `Data.ByteString.Lazy.Char8': use -v to see a list of the files searched for (imported from analysis.hs) ghc complains about not finding the ByteString library. Is this supposed to hapen, and if so, is there a way to get profiling to work? thanks, Jeff -- This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.

2 1