May 2008 - Haskell-Cafe

Couple of formal questions
by Creighton Hogg 17 May '08

17 May '08

Hello Haskell, So there's two questions that have been bothering me lately & while they are, as usual, a little off topic I figured this might be a good forum: Where could I find a good treatment on data vs. codata & the difference between well-founded recursion & well-founded(?) corecursion? Where could I find a proof that the initial algebras & final coalgebras of CPO coincide? I saw this referenced in the "Bananas.." paper as a fact, but am not sure where this comes from. Thanks, Creighton Hogg

6 8

alex + happy parser problem
by Sean McLaughlin 17 May '08

17 May '08

Hi, To learn alex and happy, I'm trying to write a parser for a simple expression language. When I wrote my own lexer and just used happy, it was fine. When I used the basic wrapper of alex it was also fine. However, when I use the posn wrapper to get position information, I get a strange exception when the parse error occurs at the end of the input. For example, parsing "1 + " yields "Internal Happy error" rather than something like "Parse error at line 1, column 5" The lexer and parser are attached. Can anyone see what I'm doing wrong? calling parse "1+" yields a "Internal Happy error" instead of a parse error as I would expect. Thanks, Sean ------------------ -- Lexer ------------------ { module ExprLexer ( Token(..), AlexPosn(..), alexScanTokens, tokenPosn ) where } %wrapper "posn" $digit = 0-9 tokens :- $digit+ { (\p s -> Int p (read s)) } [\+] { (\p s -> Sym p (head s)) } { data Token = Sym AlexPosn Char | Int AlexPosn Int deriving (Eq, Show) tokenPosn (Sym p _) = p tokenPosn (Int p _) = p } -------------------------- --- Parser -------------------------- { module ExprParser where import ExprLexer (Token(..), alexScanTokens, tokenPosn, AlexPosn(..)) } %name parseExp %tokentype { Token } %token int { Int _ $$ } '+' { Sym _ '+' } %right '+' %% Exp : Exp '+' Exp { Add $1 $3 } | int { Const $1 } { data Expr = Const Int | Add Expr Expr deriving Show parse :: String -> Expr parse = parseExp . alexScanTokens happyError :: [Token] -> a happyError tks = error ("Parse error at " ++ lcn ++ "\n") where lcn = case tks of [] -> "end of file" tk:_ -> "line " ++ show l ++ ", column " ++ show c where AlexPn _ l c = tokenPosn tk }

1 0

Proving my point
by Achim Schneider 16 May '08

16 May '08

"test.l" (line 7, column 1): unexpected end of input expecting "(", Lambda abstraction, Let binding, Atom, end of input or Function application I obviously don't know anything about Parsec's inner workings. I'm going to investigate as soon as I stopped despairing. -- (c) this sig last receiving data processing entity. Inspect headers for past copyright information. All rights reserved. Unauthorised copying, hiring, renting, public performance and/or broadcasting of this signature prohibited.

4 14

How to use arrays efficiently?
by Lauri Oksanen 16 May '08

16 May '08

Hi, I have a list of index-value pairs and I want to get a list which is sorted by the index and which contains index-wise sums of all the values. Here is a function that does this (and a bit more). --- pixelHistogram samples = let index s t = let x = round $ (fromIntegral imageWidth) * (1-s) y = round $ (fromIntegral imageHeight) * (1-t) in y*imageWidth + x indexedSamples = [(index s t, color) | (s, t, color) <- samples] pixelArray :: Array Int Color pixelArray = accumArray (\+) black (0, imageWidth*imageHeight) indexedSamples in elems pixelArray --- The problem is that this function is very inefficient. It seems that the whole indexedSamples list is stored in memory when creating pixelArray. When I do some profiling I see that the heap consumption of this function grows linearly. If I just write the samples list to a file instead of first summing them, I get a nice and flat heap profile. So how to do this efficiently? Thanks, Lauri

6 11

fgl Data.Graph examples
by Anatoly Yakovenko 16 May '08

16 May '08

Can someone post some simple examples of using the Data.Graph library? So I can define a simple graph let g = buildG (1,2) [(1,2), (2,1)] but how do i label my edges and nodes? i want to be able to name my nodes, and add weights to the edges and be able to update those weights as well. Thanks, Anatoly

2 2

ChessLibrary project
by Andrew Wagner 16 May '08

16 May '08

All, I've made my first code drop for a new project I'm working on (though it's really an extension of work I've been doing for about 5 years). Information and code can be found at http://code.haskell.org/ChessLibrary/ . Comments and suggestions warmly welcomed. Also, I'm looking for an experienced haskeller to serve as a mentor for this project - just to provide regular code reviews and architectural advice, mainly. Thanks for looking!

1 0

elem of infinite set of tuple
by leledumbo 16 May '08

16 May '08

I don't know how Haskell should behave on this. Consider this function: elemOf (x,y) = (x,y) `elem` [ (a,b) | a <- [0..], b <- [0..] ] If I try to query elemOf (1,1), the program keeps searching and searching but it never makes it. But if I query elemOf (0,1) (or anything as long as the first element is 0), it can find it easily. I wonder how it's handled. >From my point of view, instead of starting from (1,0), the program starts from (0,0), which will never finish since the limit of the second element is infinite. -- View this message in context: http://www.nabble.com/elem-of-infinite-set-of-tuple-tp17272802p17272802.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

6 6

runInteractiveCommand: program ends before writing or reading all the output
by Olivier Boudry 16 May '08

16 May '08

Hi all, It's the first time I use the runInteractiveCommand and I was probably bitten by laziness. When I run the following program and send its output to a file using '>' redirection I get the full output of the called process. But if I run it in the console I get only half of the output. As console is slower than disk I assume the called process terminates before all data has been read from it or the main process terminates before data has been written to stdout. I thought using waitForProcess, closing called process output and flushing stdout would solve the problem but it doesn't. > -- Compile with -threaded option > module Main where > > import Control.Concurrent (forkIO) > import System.Environment (getArgs) > import System.FilePath (dropExtension, takeFileName) > import System.IO (Handle, hClose, hFlush, hGetContents, stdout) > import System.Process (runInteractiveCommand, waitForProcess) > > main :: IO () > main = do > (file:_) <- getArgs > (_, out, _, pid) <- runInteractiveCommand $ "dumpbin /EXPORTS " ++ file > forkIO (createDefFile file out) > waitForProcess pid > hClose out > hFlush stdout > > createDefFile :: String -> Handle -> IO () > createDefFile file inp = do > putStrLn $ "LIBRARY " ++ (dropExtension . takeFileName) file ++ ".dll" > putStrLn "EXPORTS" > text <- hGetContents inp > mapM_ writeExport $ keepExports $ map words $ lines text > where > keepExports :: [[String]] -> [String] > keepExports = map head > . filter (not . null) > . takeWhile (["Summary"]/=) > . drop 1 > . dropWhile (["ordinal","name"]/=) > writeExport ('_':xs) = putStrLn xs > writeExport xs = putStrLn xs Any idea regarding the cause of this problem? Thanks, Olivier.

5 8

bottomless Top?
by Conor McBride 16 May '08

16 May '08

Folks [statutory warning: functors, terminal objects, unsafeCoerce] I had a peculiar notion this morning to wonder whether I could get away with shape :: Functor f => f a -> f () shape fa = unsafeCoerce fa Sure enough, ghci gives me *Top> shape "moo" [(),(),()] *Top> shape [undefined] [*** Exception: Prelude.undefined which shows that pattern matching does just enough work to check that an element of () isn't bottom. I'm just wondering if that's really what's happening, what other implementations do, and how the garbage collector would react: would all those a's hang around for as long as the shape? I'm also wondering whether it makes sense to have a "bottomless Top" type, with constructor _ and lazy pattern _ (with (undefined :: Top) equal to _). Then the constant-time shape operator makes the same sort of sense as the constant-time inflate :: Functor f => f Zero -> f a Any thoughts on either or both? Cheers Conor

2 3

getting output from shell commands
by Philip Müller 15 May '08

15 May '08

Hi, I would like to be able to run a shell command like curl and process the output in my Haskell program, but I couldn't find anything helpful about it. Something like main = do args <- getArgs inp <- exec "curl" args putStrLn (processInput inp) would be very helpful. Regards Philip

3 2