- Haskell-Cafe - Haskell.org

is anyone familiar with "Comprehending Queries" by Torsten Grust?
by Günther Schmidt 28 Mar '10

28 Mar '10

Hi all, is anyone here familiar with the paper "Comprehending Queries" by Thorsten Grust? http://kops.ub.uni-konstanz.de/volltexte/1999/312/pdf/312_1.pdf As the paper dates from 1999 I wonder if its content may have been "obsoleted" by later works or if it still is pretty much current. I ask this because I'm still struggling to put my program with lots of queries on a more sound code base. Günther

2 2

First Ghent FPG Meeting on April 1, 19h: Program and Final Details
by Jeroen Janssen 28 Mar '10

28 Mar '10

Dear all, We are nearing the First Ghent Functional Programming Group (Ghent FPG) Meeting, and, judging by the number of people that registered on the google form (28!), it will be quite crowded! The final formal program is as follows: 1. Jeroen Janssen - Welcome and short introduction to FP 2. Jasper Van der Jeugt - "BlazeHtml: a blazingly fast Html generator in Haskell" 3. Tom Schrijvers - "Functional Pearl: The Monad Zipper" 4. Romain Slootmaekers - "Functional Programming at Amplidata" (not yet fully confirmed) 5. Denis Defreyne, Pieter De Baets - "An introduction to Phunctional Programming" 6. Short discussion on the direction of the Ghent Functional Programming Group The meeting takes place on Thursday, April 1, starting at 19h, in meeting room "Shannon" of the Technicum Building of Ghent University (Sint-Pietersnieuwstraat 41, 9000 Gent). Since the automatic sliding doors will be closed, we will put up a note on the sliding doors on the front left, with a phone number you can call to get in. If you are uncertain whether you'll be able to find the venue, just ask for our cellphone number by e-mail, so you can call us if you get lost. Hope to see you all on Thursday, Bart Coppens (Bart.Coppens(a)elis.ugent.be) Jeroen Janssen (jejansse(a)gmail.com) The Ghent FPG organizing committee.

1 0

Refactoring of OAuth Library - Need Review
by Diego Souza 28 Mar '10

28 Mar '10

Hi, I've done a complete refactoring of oauth library. This new version is incompatible (in terms of interface) with the previous one. An example of use is available here: http://projects.bitforest.org/hoauth/dist/doc/html/hoauth/Network-OAuth.html Probably the major change is that, along the functions that deal with oauth authentication, there is now an HttpClient typeclass. The library now is able to perform http requests and fully deal with the oauth protocol. I'm planning to upload to hackage soon enough, but before that I'd like to ask if someone could review it and send feedbacks. That would be much appreciated. To get the code: $ darcs get http://projects.bitforest.org/hoauth/ To run tests: $ haskell -isrc/main/haskell -isrc/test/haskell src/test/haskell/Tests.hs Thanks in advance, ~dsouza

1 1

Are there any male haskellers?
by Günther Schmidt 28 Mar '10

28 Mar '10

Hi everyone, are there any male haskellers? I hope this question is not considered inappropriate .... Günther

2 1

chp-plus doesn't install
by Colin Paul Adams 28 Mar '10

28 Mar '10

I'm getting these errors (ghc 6.10.4 on Linux x86_64): Building chp-plus-1.1.0... [1 of 9] Compiling Control.Concurrent.CHP.Test ( Control/Concurrent/CHP/Test.hs, dist/build/Control/Concurrent/CHP/Test.o ) [2 of 9] Compiling Control.Concurrent.CHP.Console ( Control/Concurrent/CHP/Console.hs, dist/build/Control/Concurrent/CHP/Console.o ) [3 of 9] Compiling Control.Concurrent.CHP.Connect ( Control/Concurrent/CHP/Connect.hs, dist/build/Control/Concurrent/CHP/Connect.o ) Control/Concurrent/CHP/Connect.hs:146:67: Couldn't match expected type `ChanOpts a' against inferred type `ConnectableParam (Chanout a)' In the first argument of `oneToOneChannel'', namely `o' In the second argument of `(<$>)', namely `oneToOneChannel' o' In the first argument of `(>>=)', namely `((writer &&& reader) <$> oneToOneChannel' o)' Control/Concurrent/CHP/Connect.hs:152:67: Couldn't match expected type `ChanOpts a' against inferred type `ConnectableParam (Chanin a)' In the first argument of `oneToOneChannel'', namely `o' In the second argument of `(<$>)', namely `oneToOneChannel' o' In the first argument of `(>>=)', namely `((reader &&& writer) <$> oneToOneChannel' o)' etc. -- Colin Adams Preston Lancashire

2 3

Internet game servers in Haskell?
by Colin Paul Adams 28 Mar '10

28 Mar '10

Has anyone ever written a server in Haskell for managing live game-playing (any game) across the internet? -- Colin Adams Preston Lancashire

3 2

Testing for valid data
by michael rice 28 Mar '10

28 Mar '10

When I'm learning a new language I like to translate old programs into the new language as a test of my understanding. However, many of the old programs are from old programming texts, many written in the time of punch-cards for batch processing, and many containing significant amounts of code that only tests for valid data. Should we still be writing programs in this fashion, or acknowledge the fact that better tools for pre-screening data are now available and code only for the problem at hand? Michael

3 3

Why is it so different between 6.12.1 and 6.10.4_1 ?
by zaxis 28 Mar '10

28 Mar '10

In 6.12.1 under archlinux >let f x y z = x + y + z > :t f f :: (Num a) => a -> a -> a -> a > :t (>>=) . f (>>=) . f :: (Num a) => a -> ((a -> a) -> a -> b) -> a -> b > ((>>=) . f) 1 (\f x -> f x) 2 5 In 6.10.4_1 under freebsd > let f x y z = x + y + z *Money> :t f f :: (Num a) => a -> a -> a -> a > :t (>>=) . f (>>=) . f :: (Monad ((->) a), Num a) => a -> ((a -> a) -> a -> b) -> a -> b > ((>>=) . f) 1 (\f x -> f x) 2 <interactive>:1:1: No instance for (Monad ((->) a)) arising from a use of `>>=' at <interactive>:1:1-5 Possible fix: add an instance declaration for (Monad ((->) a)) In the first argument of `(.)', namely `(>>=)' In the expression: ((>>=) . f) 1 (\ f x -> f x) 2 In the definition of `it': it = ((>>=) . f) 1 (\ f x -> f x) 2 Sincerely! ----- fac n = let { f = foldr (*) 1 [1..n] } in f -- View this message in context: http://old.nabble.com/Why-is-it-so-different-between-6.12.1-and-6.10.4_1---… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

6 17

Randomized N-Queens
by Ronald Guida 27 Mar '10

27 Mar '10

Hi, I'm trying to solve the N-queens problem, but with a catch: I want to generate solutions in a random order. I know how to solve the N-queens problem; my solver (below) generates all possible solutions. What I am trying to do is generate solutions in a random order by somehow randomizing the order in which "nextRow" considers the unused columns. I tried adding a random number generator to the solution state; the problem with this approach is that whenever the solver backtracks, the state of the random number generator backtracks along with it. In effect, I am selecting a random, but fixed, permutation for each row, and then I am applying that same set of permutations along all computational paths. Whenever I consider row R, regardless of which path I have taken, I am applying row R's permutation to the unused columns. This is not the behavior I want. I want each computational path to use a new, different permutation for each row. On the other hand I also want to be able to take the first few solutions without waiting for all possible solutions to be generated. How might I go about doing this? -- Ron ------------------------------------------------------------ module Main where import Control.Monad.State import Data.List import System.Environment import System.Random import System.Random.Shuffle -- from package random-shuffle newtype Location = Location {unLocation :: (Int, Int)} deriving (Show) isAttacked :: Location -> Location -> Bool isAttacked (Location (row1, column1)) (Location (row2, column2)) = or [ (row1 == row2) , (column1 == column2) , ((row1 - row2) == (column1 - column2)) , ((row1 - row2) == (column2 - column1)) ] newtype Board = Board {unBoard :: [Location]} deriving (Show) data (RandomGen g) => SolutionState g = SolutionState { solnBoard :: Board , solnUnusedColumns :: [Int] , solnRandomGen :: g } nextRow :: (RandomGen g) => Int -> Int -> StateT (SolutionState g) [] () nextRow n row = do (SolutionState (Board locs) unusedColumns gen) <- get let (ps, gen') = randShuffleSeq (length unusedColumns) gen column <- lift $ shuffle unusedColumns ps let loc = Location (row, column) guard $ all (not . isAttacked loc) locs let remainingCols = unusedColumns \\ [column] put $ (SolutionState (Board (loc : locs)) remainingCols gen') randShuffleSeq :: (RandomGen g) => Int -> g -> ([Int], g) randShuffleSeq 0 g = ([], g) randShuffleSeq 1 g = ([], g) randShuffleSeq n g = (x:xs, g2) where (x, g1) = randomR (0, n-1) g (xs, g2) = randShuffleSeq (n-1) g1 allRows :: (RandomGen g) => Int -> StateT (SolutionState g) [] () allRows n = mapM_ (nextRow n) [1..n] solve :: (RandomGen g) => Int -> g -> [Board] solve n gen = map solnBoard $ execStateT (allRows n) (SolutionState (Board []) [1..n] gen) formatSolution :: Board -> String formatSolution = show . map unLocation . unBoard main :: IO () main = do args <- getArgs let boardSize = read $ args !! 0 maxSolns = if length args > 1 then read (args !! 1) else 10 allSolns = solve boardSize (mkStdGen 42) putStrLn $ unlines $ map formatSolution $ take maxSolns allSolns

2 1

Garbage collecting pointers
by Max Rabkin 26 Mar '10

26 Mar '10

On Fri, Mar 26, 2010 at 11:21 PM, Brandon S. Allbery KF8NH <allbery(a)ece.cmu.edu> wrote: > On Mar 26, 2010, at 16:28 , Mads Lindstrøm wrote: >> >> For some time I have been thinking about an idea, which could limit >> Haskell's memory footprint. I don't know if the idea is crazy or clever, > > This is called pointer tagging. The original STG design avoided it because > of the perceived performance loss in removing the tags before using the > pointer, but as I understand it the current GHC design uses limited tagging. I'd say it's really CDR coding -- http://en.wikipedia.org/wiki/CDR_coding -- pointer tagging is just a part of it. [Sorry, Brandon, my reply all button seems elusive these days] --Max

1 0