April 2010 - Haskell-Cafe

True Random Numbers
by Alex Rozenshteyn 16 Apr '10

16 Apr '10

Does haskell have a way of using /dev/random to generate random *things*? Currently I'm just reading the data into a byte string, converting it into bits, and keeping track of it in the state monad. -- Alex R

9 16

Nomic game in Haskell
by Dupont Corentin 15 Apr '10

15 Apr '10

Hello Café, do you know Nomic? It's a fabulous and strange game where you have the right to change the rules in the middle of the game! In fact, changing the rules is the goal of the game. Changing a rule is considered as a move. Of course even that could be changed! www.nomic.net I'm wondering if it could be possible to implement a Nomic (or an helper for the game) in Haskell. Haskell seems appropriate for that, since functions are first order objects, and Haskell is good at DSLs. What I think, is creating a DSL where a rule could be expressed as a function, or a monad. Many kinds of rules could be expressed: 1. - a rule that says that player #1 must be named Grand Counselor. 2. - a rule that says that the following rules must be voted at the majority, and how. 3. - a rule that says that the money is the ECU. 4. - a rule that says that if a rule is accepted, the author is awarded 1 ECU. 5. - a rule that says that all rules must have a name. 6. - a rule that says that democracy is abolished. God save the King, Player #1! 7. - a rule that says that a rule mustn't suppress another. As you can see, the rules are highly self-referencing! some rules, like #1, describes a protocol, involving IO (players voting). some rules, like #7, could be implemented with a test over new rules (that rule could test proposed rules against rule suppression). some rules, like #5, must inspect the code of other rules. A proposed rule must pass all active rule's criterias to become active. Once active, a rule should be applyed immediatly to all active rules, and to all futur rules, until it is repealed. Any comments? Cheers, Corentin

5 7

Haskell web dev - ASP .NET analogue?
by Christopher Done 15 Apr '10

15 Apr '10

The state of Haskell web development is exciting and messy. We don't have a de facto way to do it like Rails or Django or ASP .NET, but we do have many imitations and ideas popping up. You can easily judge whether the community is happy with the state of web development by the number of new web frameworks coming out all the time. I'm wondering if there is a ASP .NET analogue for Haskell. ASP .NET manages compiled languages like C# well by compiling a given page when you run it, and also allowing you to build pages into DLLs and then link them in your project. I have used this development process for a year in a commercial business and quite like it. It's just really easy to develop. Often you make little mistakes on a given page, so you just edit one bit of code -- you don't want to have to run a command to reload this code yourself. I think that, with Hint, one could write a Haskell equivalent. The nice thing is you can test your Haskell code in GHCi and then save the page and load it in the browser and have it auto-recompile. As to how pages are generated I guess is up to whoever is using it. You have a kind of standard separation in ASP .NET: backend.vb: Sub submit(sender As Object, e As EventArgs) lbl1.Text="Your name is " & txt1.Text End Sub frontend.asp: <form runat="server"> Enter your name: <asp:TextBox id="txt1" runat="server" /> <asp:Button OnClick="submit" Text="Submit" runat="server" /> <p><asp:Label id="lbl1" runat="server" /></p> </form> One could blatantly copy this model. And it's not a bad model. To start with anyway. It has the benefit that it's familiar to many web programmers, and, importantly, business practice/industry. My coworkers who only deal with HTML/CSS/JavaScript wouldn't care if backend.vb became backend.hs. I wrote an ecommerce site (http://productsforhair.co.uk/) a year ago in Haskell and used the Text.XHtml.Strict and hated it to be honest. It looks like a complete mess. Look at these two examples of using EDSLs to produce content. Messy and unmanageable. http://paste.lisp.org/display/97840 http://paste.lisp.org/system-server/show/lisppaste/web-server As much as I love EDSLs like html combinators, as soon as you get a nontrivial project with big complex elements it soon gets out of hand. BlazeHtml makes this a little nicer with its `do' syntax but it's still a problem.

2 1

Automating injections: adding a type parameter
by Romulus 15 Apr '10

15 Apr '10

Hello everyone, It's my very first post to this mailing list, though i read it for several months. I'm very interested in Swiertra's automatic injection technique [1]. http://www.cse.chalmers.se/~wouter//Publications/DataTypesALaCarte.pdf<http://www.cse.chalmers.se/%7Ewouter//Publications/DataTypesALaCarte.pdf> http://www.cse.chalmers.se/~wouter//Misc/LaCarte.hs<http://www.cse.chalmers.se/%7Ewouter//Misc/LaCarte.hs> As a first exercise, i've tried to write an interpreter/transformer for propositional logic, in the spirit of [2]. http://www.haskell.org/sitewiki/images/6/6a/TMR-Issue11.pdf data LProp r = LProp String instance Functor LProp where fmap _ (LProp x) = LProp x data LAnd r = LAnd r r instance Functor LAnd where fmap f (LAnd x y) = LAnd (f x) (f y) Automatic injections are cool things to write helpers that work with any type for formulae extending LProp with coproducts: -- not very expressive, but it's not the point type ConjLogicF = LProp :+: LAnd type ConjLogic = Mu ConjLogicF lprop :: (LProp :<: f) => String -> Mu f lprop = \s -> inject (LProp s) land :: (LAnd :<: f) => Mu f -> Mu f -> Mu f land = \x y -> inject (LAnd x y) However, i'd like to parameterize the type of propositionnal variables for the logic, that is, not only strings, but a given type parameter p. For instance (that should make LProp' and LAnd' bifunctors i guess): data LProp' p r = LProp' p instance Functor (LProp' p) where fmap _ (LProp' x) = LProp' x data LAnd' p r = LAnd' r r instance Functor (LAnd' p) where fmap f (LAnd' x y) = LAnd' (f x) (f y) type ConjLogicF' p = (LProp' p) :+: (LAnd' p) type ConjLogic' p = Mu (ConjLogicF' p) lprop' :: ((LProp' p) :<: (f p)) => p -> Mu (f p) lprop' = \x -> inject (LProp' x) I'm stuck with the definition of the helper for LAnd'. I expect : land' :: ((LAnd' p) :<: (f p)) => Mu (f p) -> Mu (f p) -> Mu (f p) land' = \x y -> inject (LAnd' x y) ... but ghci 6.10.4 does not really like this definition... Could not deduce (LAnd' p :<: f p1) from the context (LAnd' p1 :<: f p1) arising from a use of `inject' at PropSample.hs:128:16-33 Possible fix: add (LAnd' p :<: f p1) to the context of the type signature for `land'' or add an instance declaration for (LAnd' p :<: f p1) In the expression: inject (LAnd' x y) In the expression: \ x y -> inject (LAnd' x y) In the definition of `land'': land' = \ x y -> inject (LAnd' x y) Failed, modules loaded: none. Does anybody have a clue for this problem ? I don't really understand where is the trouble actually. Cheers, PS: haskellers rulez ;) [1] Swierstra, W. Data types à la carte J. Funct. Program. Cambridge University Press, 2008, 18, 423-436 [2] Knowles, K. First-Order Logic à la Carte The Monad.Reader, 2008, issue 11 -- Romuald THION Docteur level 1.0 - Great master access control, +3 against half-dead "la vie, c'est comme un jeu mal foutu dont on ne connait pas les règles et où il n'y a pas de sauvegardes"

2 2

ANN: haskell-src-exts 1.9.0
by Niklas Broberg 15 Apr '10

15 Apr '10

Fellow Haskelleers, I'm pleased to announce the release of haskell-src-exts-1.9.0! * On hackage: http://hackage.haskell.org/package/haskell-src-exts * Via cabal: cabal install haskell-src-exts * Darcs repo: http://code.haskell.org/haskell-src-exts 1.8.2 --> 1.9.0 =============== * OptionPragma is renamed to the more descriptive ModulePragma, and adds a constructor AnnModulePragma for handling ANN pragmas preceding module header. * Add instances for Eq/Ord/Data/Typeable for Fixity. * Add 'parseFileWithComments' and 'parseFileContentsWithComments' to L.H.Exts . * More informative error messages when HSX tags are mismatched (thanks Gracjan Polak for the patch!) Cheers, /Niklas

4 7

What is the consensus about -fwarn-unused-do-bind ?
by Ivan Lazar Miljenovic 15 Apr '10

15 Apr '10

As of 6.12.1, the new -fwarn-unused-do-bind warning is activated with -Wall. This is based off a bug report by Neil Mitchell: http://hackage.haskell.org/trac/ghc/ticket/3263 . However, does it make sense for this to be turned on with -Wall? For starters, why should this warning apply only to do blocks and not to explicit usage of >>, etc.? That is, the following code (as specified in the above bug report) generates an error: do doesFileExist "foo" return 1 yet this doesn't: doesFileExist "foo" >> return 1 If monadic code is going to return an error, then shouldn't _all_ monadic code do so, and not just those in do blocks? Secondly, a fair number of packages seem to be disabling this globally; the packages I know of (from mentions on mailing lists and grepping on /srv/code/*/*.cabal on code.haskell.org) that have the -fno-warn-unused-do-bind option being passed to GHC in their .cabal file include: * HsOpenCL * leksah-server * xmonad (including xmonad-contrib) * xmobar * pandoc My reason for bringing this up is that I'm soon about to release a new version of my graphviz library, and am debating what to do. Note that most of these errors are being caused by usage of a monadic-style of parsing (before anyone tells me I should be using an Applicative style instead, polyparse doesn't support Applicative, so I can't) and as such the "return value" is being evaluated anyway. The way I see it, I have 4 options: 1. Do as the error suggests and preface usage of these parser combinators with "_ <-". 2. Use some function of type "(Monad m) => m a -> m ()" instead of doing "_ <-". 3. Duplicate the parser combinators in question so that I have one version that returns a value and another that does the main parser and then returns (); then use this second combinator in do blocks where I don't care about the returned value. 4. Put "-fno-warn-unused-do-bind" in the .cabal file. The first two options don't appeal to me as being excessive usage of boilerplate; the third involves too much code duplication. However, I am loath to just go and disable a warning globally. What does the Haskell community think? Is -fwarn-unused-do-bind a worthwhile warning (and code should be updated so as not to cause it to find anything to warn about)? Or is it more of a hindrance to be disabled? -- Ivan Lazar Miljenovic Ivan.Miljenovic(a)gmail.com IvanMiljenovic.wordpress.com

13 14

Can't login to GHC trac
by Erik de Castro Lopo 15 Apr '10

15 Apr '10

Hi all, I'm having trouble logging into GHC's Trac bug tracker at: http://hackage.haskell.org/trac/ghc/ I'm using Iceweasel 3.5.8 (Debian's rebranded Firefox) and I've also tried Galeon. The message I get is: Redirect Loop Redirection limit for this URL exceeded. Unable to load the requested page. This may be caused by cookies that are blocked. The browser has stopped trying to retrieve the requested item. The site is redirecting the request in a way that will never complete. * Have you disabled or blocked cookies required by this site? * NOTE: If accepting the site's cookies does not resolve the problem, it is likely a server configuration issue and not your computer. I do not have cookies disabled. I have hwoever tried deleting the cookies for hackage.haskell.org and trying again, but that didn't help either. Anyone else have the same problem? Cheers, Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

3 4

Can't install Criterion package on ghc ..
by Mozhgan kabiri 15 Apr '10

15 Apr '10

Hi, I am trying to install Criterion package, but I keep getting an error and I can't figure it out why it is like this !! mozhgan@Mozhgan-KCh:~$ cabal install Criterion Resolving dependencies... Configuring vector-algorithms-0.3... Preprocessing library vector-algorithms-0.3... Building vector-algorithms-0.3... [1 of 9] Compiling Data.Vector.Algorithms.Common ( Data/Vector/Algorithms/Common.hs, dist/build/Data/Vector/Algorithms/Common.o ) [2 of 9] Compiling Data.Vector.Algorithms.Search ( Data/Vector/Algorithms/Search.hs, dist/build/Data/Vector/Algorithms/Search.o ) [3 of 9] Compiling Data.Vector.Algorithms.Radix ( Data/Vector/Algorithms/Radix.hs, dist/build/Data/Vector/Algorithms/Radix.o ) [4 of 9] Compiling Data.Vector.Algorithms.Optimal ( Data/Vector/Algorithms/Optimal.hs, dist/build/Data/Vector/Algorithms/Optimal.o ) [5 of 9] Compiling Data.Vector.Algorithms.Insertion ( Data/Vector/Algorithms/Insertion.hs, dist/build/Data/Vector/Algorithms/Insertion.o ) [6 of 9] Compiling Data.Vector.Algorithms.Merge ( Data/Vector/Algorithms/Merge.hs, dist/build/Data/Vector/Algorithms/Merge.o ) [7 of 9] Compiling Data.Vector.Algorithms.TriHeap ( Data/Vector/Algorithms/TriHeap.hs, dist/build/Data/Vector/Algorithms/TriHeap.o ) [8 of 9] Compiling Data.Vector.Algorithms.Intro ( Data/Vector/Algorithms/Intro.hs, dist/build/Data/Vector/Algorithms/Intro.o ) ghc: panic! (the 'impossible' happened) (GHC version 6.10.4 for i386-unknown-linux): idInfo co{v a9WB} [tv] Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug cabal: Error: some packages failed to install: criterion-0.5.0.0 depends on vector-algorithms-0.3 which failed to install. statistics-0.5.1.0 depends on vector-algorithms-0.3 which failed to install. vector-algorithms-0.3 failed during the building phase. The exception was: exit: ExitFailure 1 I tried to install Statistics and vector-algorithms package separately before installing the Criterion package itself .. but it wasn't successful. Thanks, Mozhgan

4 3

Simple game: a monad for each player
by Yves Parès 14 Apr '10

14 Apr '10

Hello Cafe, I have a question about program design. Let's say I have a simple sequential game (a TicTacToe for instance, but with more than 2 players). I have a Player datatype which is like: data Player m = Player { plName :: String, -- unique for each player plTurn :: GameGrid -> m Move -- called whenever the player must play } As you may guess, the 'm' type variable is intended to be a Monad. The goal is that every player has his own monad. For instance : - a human player needs to interact with the program, so its monad would be IO, or an instance of MonadIO. - a network player, which transparently sends the game state and receives moves through network, must also have access to IO to play. - an AI doesn't need to interact with the outside of the program, so its monad can be the one we want (e.g. Identity). First, do you think it is a good way to design the program? I want the game to be totally independent of the players who are currently playing. They can be humans, AIs, AIs and network players and humans, and so on. But when running the game, the program cannot "switch" from a player's monad to another. If we want every player to run in his own monad, I think we have to stack every players' monad, and lift their actions each time they have to play. This is not a problem, the sole condition now is that every player has type: (Monad m, MonadTrans m) => Player m. But I think it's a little bit of overkill, and it may be a little bit complicated to implement such a thing. What I try to avoid is having every player running in IO monad. Do you have any suggestion? ----- Yves Parès Live long and prosper -- View this message in context: http://old.nabble.com/Simple-game%3A-a-monad-for-each-player-tp28183930p281… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

8 13

problem getting cabal-install to work on CentOS 5.2
by Chris Dornan 14 Apr '10

14 Apr '10

Hi, I am trying to install cabal-install 0.8.2 with GHC 6.12.1 on a CentOS 5.2 system and am running up against a 'ExitFailure 127' error. I can download the package and run 'cabal install' inside the package directory and all will be fine, but if I let cabal download and install the package cabal runs into problems as soon as it tries to configure. For example, if I try 'cabal install permutation --verbose=3' it finishes like this: link: done *** Deleting temp files: Deleting: /tmp/ghc1324_0/ghc1324_0.lpp *** Deleting temp dirs: Deleting: /tmp/ghc1324_0 /tmp/permutation-0.4.132758/permutation-0.4.1/dist/setup/setup configure --verbose=3 --ghc --prefix=/home/chris/.cabal --user --constraint=base ==3.0.3.2 --constraint=ghc-prim ==0.2.0.0 cabal: Error: some packages failed to install: permutation-0.4.1 failed during the configure step. The exception was: ExitFailure 127 I have tried the same exercise with GHC 6.10.4 with cabal-install 0.8.0 the same results. I have downloaded the Cabal sources to try and add some instrumentation to find out what is going wrong I have yet to succeed. Does anybody have any suggestions? Is there an easier way of getting cabal-install working on CentOS? Any help much appreciated! Chris

4 5