September 2013 - Haskell-Cafe

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 08 Sep '13

08 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

interactive-diagrams: call for suggestions
by Daniil Frumin 06 Sep '13

06 Sep '13

Good time of day, everyone! As some of you may know, I am developing an interactive-diagrams pastebin as part of the GSoC program [1]. The demo is live at <http://paste.hskll.org>, feel free to play around with it. This week I plan to migrate to the PostgreSQL database (right now all the information is stored using sqlite). And since I am already going to write code for migration I might as well implement some changes to the DB schema. That's why I am asking you to provide any suggestions on what (front-end) features would you like to see, that are most likely require to change the DB schema. Right now I plan on implementing changes that would allow the following: - Ability to annotate pastes. Pastes may have parent pastes and child pastes. - Logins (via GitHub), in addition to pseudonymous-pastes where user can just enter any text as his nickname. This will also allow user to make collections of his pastes. - Stars. Logged in users may "star" pastes they like. Please report your suggestions here or on GitHub [2]. Thanks! [1] https://gist.github.com/co-dan/437e9694563c0c13dd8e [2] https://github.com/co-dan/interactive-diagrams/issues/16 -- Sincerely yours, -- Daniil Frumin

1 0

ghc-mod v3.0.0
by Kazu Yamamoto 06 Sep '13

06 Sep '13

Hi all, I have just uploaded ghc-mod v3.0.0 to Hackage. In this version, ghc-mod supports the sandbox feature of "cabal-install". Instead, it stopped supporting "cabal-dev". If you want to use ghc-mod v3.0.0, I would recommand to install cabal-install 1.18. The sandbox in your package is automatically detected by ghc-mod if exists. --Kazu

2 3