- Haskell-Cafe - Haskell.org

Cabal --enable-tests
by satvik chauhan 10 Sep '13

10 Sep '13

Hi cafe, I wanted to ask this as I couldn't find this in cabal documentation. Is there any CCP macro set when a package is configured with --enable-testing? If not is there a way to do that? -Satvik

2 1

An Extra Empty Line
by Hong Yang 10 Sep '13

10 Sep '13

Hi Dear Haskellers, I have this small program to grep recursively in parallel. It works fine, but generates the last line empty. Is this empty line coming from mapConcurrently()? Thanks, Hong -- mygrepr.hs -- lrf.pl is an old Perl script to get all non-duplicate files recursively under the current directory, since some of directories have tons of links -- the program runs concurrently on either specified number of threads or 3/4 of the number of available cores module Main (main) where import Control.Concurrent.Async import Control.Monad import Data.List import Data.List.Split import GHC.Conc import System.Environment ( getArgs ) import System.Exit import System.Process import Text.Regex.Posix main :: IO () main = do hs_argv <- getArgs if null hs_argv || any (=~ "-h") hs_argv || any (=~ "--h") hs_argv then putStrLn "mygrepr [+RTS -N[x] -RTS] [OPTION]... PATTERN" else do numCores <- getNumProcessors numCapas <- getNumCapabilities let numT | numCapas > 1 = numCapas | otherwise = max numCapas (numCores `div` 4 * 3) _ <- setNumCapabilities numT let numThreads = fromIntegral numT :: Double findResult <- readProcess "lrf.pl" [] [] let files = lines findResult let num_of_files = fromIntegral $ length files :: Double let chunks = chunksOf (ceiling (num_of_files/numThreads)) files results <- mapConcurrently (grep hs_argv) chunks let (_, grepResult, _) = unzip3 results putStr $ unlines $ nub $ filter (\line -> not (line =~ "Binary file .* matches")) $ lines (concat grepResult) grep :: [String] -> [String] -> IO (ExitCode, String, String) grep hs_argv files = readProcessWithExitCode "/tool/pandora64/.package/grep-2.5.4/bin/grep" (hs_argv ++ files) []

1 0

FLTK GUI Binding in progress. Call for participation.
by aditya siram 09 Sep '13

09 Sep '13

Hi all, I'm working on a FLTK [1] GUI binding [2]. The attraction of FLTK is that it is portable across many platforms, meaning it is easy to: 1. provide a binary for your application that works without installing anything else. eg. no need to install X on Mac. Just double-click! 2. develop on other platforms because both the library and the bindings have few dependencies and compile from scratch without a fuss. 3. provide native solutions that don't require the browser. To my knowledge no current bindings have all those advantages. Most of the work of wrapping the C++ in C is done and a number of the examples that ship with FLTK have been ported to C and seem to be compatible. I have only just started the Haskell bindings. This is a great time to: 1. help with the low-level porting. Most of it is just busy work. 2. significantly influence what the Haskelly abstraction layer will look like. 3. tell me if I'm doing something wrong in the C/C++ code. I'm a newbie to both languages. I've learned what I needed to know to get this far. Thanks for your attention. -deech [1] http://www.fltk.org/index.php [2] http://github.com/deech/fltkhs

3 3

a little parsec enhancement
by Petr Pudlák 09 Sep '13

09 Sep '13

Hi, when thinking about this SO question <http://stackoverflow.com/q/18583416/1333025>, I couldn't find a combinator that allows a parser to /optionally/ fail without consuming input, or consume input and return its value. So I'm suggesting such a function: |-- | @emptyIf p@ parses @p@ and if its return value is @Nothing@, pretends -- that an error has occured with no input consumed. -- -- If @p@ fails and consumes some input, so does @emptyIf p@. Combine with -- 'try' if this is undesirable. emptyIf :: (Stream s m t) =>ParsecT s u m (Maybe a) ->ParsecT s u m a emptyIf p =ParsecT $ \s cok cerr eok eerr -> let cok' (Just x) s e = cok x s e cok'Nothing _ e = eerr e eok' (Just x) s e = eok x s e eok'Nothing _ e = eerr e in unParser p s cok' cerr eok' eerr| With this function, the answer to the SO question becomes really easy: |rcomb :: (Stream s m t) =>ParsecT s u m a ->ParsecT s u m b ->ParsecT s u m b rcomb p q = emptyIf $ runMaybeT (opt p *> opt q) where opt =MaybeT . optional-- optional from Control.Applicative!| Whenever |p| or |q| fails without consuming input, then |rcomb p q| fails without consuming input. Unfortunately |ParsecT| constructor isn't exported so I'm not able to implement it outside /parsec/. (Perhaps it would make sense to export |ParsecT| in some module such as |Text.Parsec.Internal|?) Therefore I'm suggesting to add such a function to /parsec/ (darcs patch included). Perhaps change the name, I couldn't think of anything better. Best regards, Petr

3 4

New Functional Programming Job Opportunities
by Functional Jobs 09 Sep '13

09 Sep '13

Here are some functional programming job opportunities that were posted recently: Haskell Programmer at Zalora Southeast Asia http://functionaljobs.com/jobs/8645-haskell-programmer-at-zalora-southeast-… Cheers, Sean Murphy FunctionalJobs.com

1 0

ANN: unix-bytestring 0.3.7
by wren ng thornton 09 Sep '13

09 Sep '13

-------------------------------------------- -- unix-bytestring 0.3.7 -------------------------------------------- The unix-bytestring package offers a full selection of Unix/Posix-specific functions for reading and writing ByteStrings to Fds. -------------------------------------------- -- Changes (since 0.3.4) -------------------------------------------- Most of the changes are for portability with newer versions of GHC, though the 0.3.7 release corrects a major bug on certain systems. I'd like to thank John Lato, Kazu Yamamoto, and Gergely Risko for their help with these patches. 0.3.5 (2011-06-29): - System.Posix.IO.ByteString: added fdSeek, tryFdSeek. - Foreign.C.Error.Safe: added. 0.3.5.3 (2012-02-21): - Foreign.C.Error.Safe: changed (Num a) to (Eq a, Num a) for GHC 7.4 0.3.5.4 (2012-08-23): - System.Posix.IO.ByteString: adjusted some imports for GHC 7.6 - Removed the custom Setup.hs, for compatibility with GHC 7.6 0.3.6 (2013-05-29): - System.Posix.Types.Iovec: adjusted unsafeForeignPtrToPtr for GHC 7.8 0.3.7 (2013-08-08): - System.Posix.IO.ByteString: corrected a bug in fdSeek and tryFdSeek where the return value was always 0 on largefile i686 systems due to the underlying C function being (undocumentedly) implemented as a macrocall. This had been fixed in unix-2.6.0.1, and I've just backported their fix, cf: * <https://github.com/ghc/packages-base/blob/master/System/Posix/Internals.hs#…> * <https://github.com/ghc/packages-unix/blob/master/System/Posix/IO/Common.hsc…> -------------------------------------------- -- Links -------------------------------------------- Homepage: http://code.haskell.org/~wren/ Hackage: http://hackage.haskell.org/package/unix-bytestring Darcs: http://community.haskell.org/~wren/unix-bytestring Haddock (Darcs version): http://community.haskell.org/~wren/unix-bytestring/dist/doc/html/unix-bytes… -- Live well, ~wren

1 0

Impredicative types and Lens?
by Artyom Kazak 08 Sep '13

08 Sep '13

Here’s a small example, which, when compiled, gives an error. Why? {-# LANGUAGE FlexibleInstances, ImpredicativeTypes, TemplateHaskell #-} import Control.Lens class Item a where name :: a -> String instance Item (String, Int) where name = fst type ItemFilter = Item a => a -> Bool data ItemBox = ItemBox { _itemFilter :: ItemFilter } makeLenses ''ItemBox The error is Couldn't match type `a0 -> Bool' with `forall a. Item a => a -> Bool' Expected type: ItemFilter Actual type: a0 -> Bool In the expression: b_aaZE In the first argument of `iso', namely `\ (ItemBox b_aaZE) -> b_aaZE' In the expression: iso (\ (ItemBox b_aaZE) -> b_aaZE) ItemBox I’m using GHC 7.6.2, if it’s important.

2 1

Any Haskell events in Madrid next week?
by Joachim Breitner 07 Sep '13

07 Sep '13

Hi, I’ll be visiting Madrid next week (research visit) and I’m wondering if there are any Haskell or FP Group meeting or other events that might be interesting? I could possibly contribute a talk. (Both preferably in English.) Wednesday or Thursday evening might would most convenient. Greetings, Joachim -- Joachim “nomeata” Breitner mail(a)joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata(a)joachim-breitner.de • GPG-Key: 0x4743206C Debian Developer: nomeata(a)debian.org

3 2

Unary functions and infix notation
by Johannes Emerich 06 Sep '13

06 Sep '13

As is well known, any binary function f can be turned into an infix operator by surrounding it with backticks: f a b -- prefix application a `f` b -- infix application It is then possible to take left and right sections, i.e. partially applying f: (a `f`) -- equivalent to \b -> a `f` b (`f` b) -- equivalent to \a -> a `f` b This extends relatively naturally to functions of arity greater than two, where usage of a function in infix notation produces a binary operator that returns a function of arity n-2. Weirdly, however, infix notation can also be used for unary functions with polymorphic types, as the following ghci session shows: Prelude> :t (`id` 1) (`id` 1) :: Num a => (a -> t) -> t Prelude> (`id` 1) (\y -> show y ++ ".what") "1.what" Desugaring of an equivalent source file shows that id is applied to the anonymous function, which is then applied to 1. The following example of a function that is not polymorphic in its return type behaves closer to what I would have expected: It does not work. Prelude> let z = (\y -> True) :: a -> Bool Prelude> :t (`z` True) <interactive>:1:2: The operator `z' takes two arguments, but its type `a0 -> Bool' has only one In the expression: (`z` True) What is the purpose/reason for this behaviour? Thank you, --Johannes

5 4

sequence causing stack overflow on pretty small lists
by Niklas Hambüchen 06 Sep '13

06 Sep '13

On #haskell we recently had a discussion about the following: import System.Random list <- replicateM 1000000 randomIO :: IO [Int] I would think that this gives us a list of a million random Ints. In fact, this is what happens in ghci. But with ghc we get: Stack space overflow: current size 8388608 bytes. Use `+RTS -Ksize -RTS' to increase it. This is because sequence is implemented as sequence (m:ms) = do x <- m xs <- sequence ms return (x:xs) and uses stack space when used on some [IO a]. >From a theoretical side, this is an implementation detail. From the software engineering side this disastrous because the code is * obviously correct by itself * the first thing people would come up with * not exaggerating: a million elements is not much * used a lot of places: mapM, replicateM are *everywhere* and yet it will kill our programs, crash our airplanes, and give no helpful information where the problem occurred. Effectively, sequence is a partial function. (Note: We are not trying to obtain a lazy list of random numbers, use any kind of streaming or the likes. We want the list in memory and use it.) We noticed that this problem did not happen if sequence were implemented with a difference list. What do you think about this? Should we "fix" functions like this, probably trading off a small performance hit, or accept that idiomatic Haskell code can crash at any time?

6 12