July 2013 - Haskell-Cafe

Workshop on Termination (deadline: July 22)
by Johannes Waldmann 15 Jul '13

15 Jul '13

Dear all. Still one week to go: submit your results and questions on all things related to termination and bounded resource consumption (of user programs, of type checkers, etc.) to the Workshop on Termination. We are specifically encouraging contributions from the Haskell (and Agda, and Idris) community. - Johannes. ################################################################# 13th International Workshop on Termination (WST) ( full CfP see http://www.imn.htwk-leipzig.de/WST2013/ ) Centro Residenziale Universitario di Bertinoro (near Bologna, Italy). submission: July 22, 2013 notification: July 25, 2013 final version: August 10, 2013 workshop: August 29 - 31, 2013 The Workshop on Termination traditionally brings together, in an informal setting, researchers interested in all aspects of termination, whether this interest be practical or theoretical, primary or derived. The workshop also provides a ground for cross-fertilisation of ideas from term rewriting and from the different programming language communities. The friendly atmosphere enables fruitful exchanges leading to joint research and subsequent publications.

1 0

OpenGL extensions survey results
by Brian Lewis 15 Jul '13

15 Jul '13

Hi, here are the results of the recent OpenGL extensions survey: https://github.com/bsl/opengl-extensions-survey If I did anything dumb, please send me mail or a pull request. Thanks to everyone who contributed!

1 0

Database conectivity in Haskell. HDBC-3 or something.
by Alexey Uimanov 15 Jul '13

15 Jul '13

Hello, haskellers. I has made some rework of HDBC, if you are interested follow the link http://s9gf4ult.github.io/hdbc/Database-HDBC.html So, why did I made this ? Because I did not found good enough (inside and outside) common interface for database connectivity. What I mean is flexible and simple interface, supporting arbitrary precision values and providing convenient and minimalistic interface for just query execution. In ''roadmap'' in the documentation you can see, that I think, that Haskell must have such interface which must become the conveient base for other high-level interfaces, such as Persistent and HaskellDB. Other languages, like Python goes this way, so why we must not ? I dont know will this patches accepted to main HDBC repository or not. I will fork HDBC under another name if necessary. The patches are: https://github.com/s9gf4ult/hdbc https://github.com/s9gf4ult/hdbc-postgresql There is just PostgreSQL driver for now, but I plan to write some other drivers as well. Any help and pull requests are welcome. My question is: what do you think about all of this? Do you think this is good idea or this is not necessary for now? I just want to notify the community about this and maybe find support. Discuss.

1 0

Do combinatorial algorithms have a matroid strucutre XOR non-matroid structure?
by KC 14 Jul '13

14 Jul '13

I ask this on this mailing list because there are quite a few mathematically oriented people here. Casey -- -- Regards, KC

2 1

Re: [Haskell-cafe] Deriving with generics without values
by Roman Cheplyaka 14 Jul '13

14 Jul '13

Hi, (Redirecting this back to cafe to keep it discoverable — hope you don't mind.) * JP Moresmau <jpmoresmau(a)gmail.com> [2013-07-14 16:02:56+0200] > Hello, sorry to bother you after you've been kind enough to answer me on > the list! I've looked a the smallcheck code but I don't see how to apply it > to my issue. First of all, I can't find the definition of >>- that you use > for example in <~>. I suppose I'm doing something silly... That comes from LogicT. >>- is almost the same as >>=, and <~> is almost the same as <*>. Byt that's not relevant to the generic aspect of the code. > But then, if I understand the code, the instances for GSerial rebuild > a type definition that mimicks the one passed as type argument, using > the final values (under the K1s) to generate a series. But in my case, > I want to be able to extract information from the original type, I'm > not just interested in the final value. What I want to be able is to > extract for example the constructor name and do something with it. I > don't see how to achieve that with your system. Yes, it doesn't do everything that you want to do, but it shows the idea. The information you need is all there — for example, to get the constructor name, you need to get hold of the C1 type (which is a synonym for M1 C), and then call the conName method. Lest you get lost in all this, it is useful to visualize the generic representation by running your code through -ddump-deriv. Here's an example: % ghci -XDeriveGeneric -ddump-deriv -dsuppress-module-prefixes Prelude GHC.Generics> data T a = A { a :: a } | B | C deriving Generic ==================== Derived instances ==================== Derived instances: instance Generic (T a_ao7) where from (A g1_aph) = M1 (L1 (M1 (M1 (K1 g1_aph)))) from B = M1 (R1 (L1 (M1 U1))) from C = M1 (R1 (R1 (M1 U1))) to (M1 (L1 (M1 (M1 (K1 g1_api))))) = A g1_api to (M1 (R1 (L1 (M1 U1)))) = B to (M1 (R1 (R1 (M1 U1)))) = C instance Datatype D1T where datatypeName _ = "T" moduleName _ = ":Interactive" instance Constructor C1_0T where conName _ = "A" conIsRecord _ = True instance Constructor C1_1T where conName _ = "B" instance Constructor C1_2T where conName _ = "C" instance Selector S1_0_0T where selName _ = "a" Generic representation: Generated datatypes for meta-information: D1T C1_0T C1_1T C1_2T S1_0_0T Representation types: type Rep (T a_ao7) = D1 D1T (C1 C1_0T (S1 S1_0_0T (Rec0 a_ao7)) :+: (C1 C1_1T U1 :+: C1 C1_2T U1)) This should give you an idea about how the structure you're dealing with looks like, and where the important information resides. > I've also look at Aeson with the generic JSON parsing, but that's > using SYB which is again different, and SYB doesn't seem to provide > the types of the constructor fields, which I would need too... It provides the types of the constructor fields through the Typeable class, which may or may not be sufficient for your needs... Anyway, you're right in that it is a completely different approach. Roman > On Fri, Jul 12, 2013 at 10:57 AM, Roman Cheplyaka <roma(a)ro-che.info> wrote: > > > Well, in your case, you need not 'from', but 'to', in order to convert > > from a generic representation to yours. > > > > Take a look at how a similar task is done in SmallCheck: > > > > https://github.com/feuerbach/smallcheck/blob/master/Test/SmallCheck/Series.… > > > > https://github.com/feuerbach/smallcheck/blob/master/Test/SmallCheck/Series.… > > > > Roman > > > > * JP Moresmau <jpmoresmau(a)gmail.com> [2013-07-12 10:45:39+0200] > > > Hello all, > > > My problem is the following: I have my own data types, and I'd like to > > > derive automatically instances of some type class from them. I've started > > > looking at GHC.Generics, which offer tools to do exactly that. However, > > > some functions of my typeclass do not take my data type as a parameter, > > but > > > as a result. Basically: > > > class MyClass where > > > fromString :: String -> a > > > > > > data MyData=MkMyData { > > > myField ::Int > > > } deriving (Generic) > > > > > > and I want to automatically generate the instance instance MyClass > > MyData, > > > using default methods, etc. > > > The GHC Generic class does say that it uses a from function that convert > > > from the datatype to its representation: from :: a -> > > > Rep< > > http://www.haskell.org/ghc/docs/7.4.1/html/libraries/ghc-prim-0.2.0.0/GHC-G… > > > > > > a > > > xfrom :: a -> Rep< > > http://www.haskell.org/ghc/docs/7.4.1/html/libraries/ghc-prim-0.2.0.0/GHC-G… > > > > > > a > > > x > > > But I don't have a "a" to start from! I see from the related papers that > > > the automatically generated code from from actually does pattern matches > > on > > > constructors, so I need a value, undefined won't work. However I see the > > > GHC.Generics also provide :+: (Sums: encode choice between constructors). > > > If I have to provide an value, then the choice between constructor has > > been > > > done! The examples about generics on > > > http://www.haskell.org/haskellwiki/GHC.Generics do provide an example of > > > defining the instance for :+: but I don't understand how we can get > > there. > > > If I have a class method that takes a value as a parameter, and I pass > > > undefined to it, the code will crash, since it can't pattern match on > > > undefined. > > > > > > Can somebody shed some light on this? Am I using the wrong tool for the > > > job? How can I achieve what I want? I want the full type representation > > > with sums, but without a value to start from. > > > > > > Thanks a million! > > > > > > JP > > > -- > > > JP Moresmau > > > http://jpmoresmau.blogspot.com/ > > > > > _______________________________________________ > > > Haskell-Cafe mailing list > > > Haskell-Cafe(a)haskell.org > > > http://www.haskell.org/mailman/listinfo/haskell-cafe > > > > > > > -- > JP Moresmau > http://jpmoresmau.blogspot.com/

1 1

OS X ghci problem
by Brian Lewis 14 Jul '13

14 Jul '13

Here's a problem variations of which have been plaguing the Haskell community for as long as I can remember. To see it for yourself: 1.) Be running OS X 2.) install GLFW-b-1.0.0 (you may need to cabal update) [1] 3.) ghci -package GLFW-b 4.) import Graphics.UI.GLFW as GLFW 5.) GLFW.init Security camera footage of the crime in progress: https://gist.github.com/dagit/5980438/raw/2e92a63135008921040478566e56ac7c0… One of the first things GLFW.init does on OS X is ... = [[NSAutoreleasePool alloc] init]; That causes -[NSAutoreleasePool init]: unrecognized selector sent to instance ... which seems to mean that NSAutoreleasePool doesn't understand the "init" message/method. But [[NSAutoreleasePool alloc] init] is really basic stuff. What is going on? [1]: http://hackage.haskell.org/package/GLFW-b

5 7

build failure on Hackage
by Brian Lewis 13 Jul '13

13 Jul '13

I maintain a library that, on Linux, needs libXxf86vm to build. The server where Hackage runs doesn't have that library, so the build fails. I think this is reasonable -- that box can't possibly have all the libraries various packages might need. But the build failure seems to cause the Haddock docs to not be generated. Why? My understanding is, if someone specifies my library as a build dependency, my library's build failure will also cause the docs for their package not to be built. I think my library will kill the Hackage Haddock docs for any other package that depends on it, due to a build error that's not even meaningful. What can I do?

2 4

getting haddock to cooperate with cpp
by Evan Laforge 13 Jul '13

13 Jul '13

So haddock ignores {-# LANGUAGE CPP #-}, which makes it crash on any file that uses it. But if you pass --optghc=-cpp, it runs CPP on everything, which makes it crash on any file that uses string gaps, or happens to contain a /*. /* is rare and easily fixed, but not string gaps. It looks like a workaround would be to manually inspect the files for LANGUAGE CPP and run two haddock passes, but then I would have to get the two passes to cooperate creating a single TOC and index. Isn't there some way to run haddock on files that use CPP? In the broader scheme, it seems perverse to be using CPP in the first place. I use it to configure imports and exports, e.g. to swap out a driver backend on different OSes, and to export more symbols when testing. Would it make sense to have a haskell version of CPP that provides only these features (e.g. just #ifdef, #else, #endif, and #define) and leaves out the problematic C comments and backslash expectations?

4 7

ANNOUNCE: MFlow 3.0
by Alberto G. Corona 13 Jul '13

13 Jul '13

The third version of MFlow is out. http://hackage.haskell.org/package/MFlow MFlow is an all-heterodox web application framework, but very haskellish. Now MFlow support restful URLs. It is the first stateful web framework to my knowledge that supports it. The type safe routes are implicitly expressed as normal monadic code within a navigation monad. The application look as a normal imperative console application, but the navigation monad goes back and forth to match the path of the URL. The user has control of the state, that can roll-back or not when the navigation goes back depending on the application needs. The state is in the form of normal Haskell variables In a monadic computation, with the weird addition of backtracking. The menu of the application below is implemented as an imperative-like syntax, but the application navigate forward and backward to synchronize with the requests of the web browser: http://mflowdemo.herokuapp.com/ This version support in-page flows. What is that? look at this example: http://mflowdemo.herokuapp.com/noscript/fviewmonad These flows are implemented as formlets with a monad instance, and callbacks which change the look. I call them "widgets": http://haskell-web.blogspot.com.es/2013/06/the-promising-land-of-monadic-fo… Each page may have many of these active widgets, each one running their own flow. These widgets refresh themselves trough Ajax if they are enclosed in the primitive "autoRefresh". If there is no Ajax or JavaScript available, they gracefully degrade by refreshing the entire page: http://mflowdemo.herokuapp.com/noscript/combination http://haskell-web.blogspot.com.es/2013/06/and-finally-widget-auto-refreshi… The page flows and the multiflow idea was inspired in Seaside<http://www.seaside.st/>, a great Smalltalk web framework and adapted to the pure recursive nature of Haskell and the formlets. It also support some JQuery widgets integrated: modal and not modal dialogs, datePicker and other active widgets that handle other widgets. It also support the older features: persistent state, WAI, blaze-html and others integration, server process timeouts, Ajax, requirements, content management, caching of widget rendering and all the other previous stuff. I wish to thank some people for their feedback. Specially Adrian May for his feedback and interest -- Alberto.

3 6

Quick-check: how to generate arbitrary complex data?
by martin 13 Jul '13

13 Jul '13

Hello all, I have a type ("Mail") which consists of hash and a list, where the hash keeps some redundant data of the list for faster access. I can add and remove elements to values of this type using custom functions, called push and pop. Now I wanted to write some quick checks, but I have no clue how to generate arbitrary values of this type. It will certainly no suffice to write arbitrary instances for the underlying types (Int and Char), because the hash and the list need to be synchronized. Currently "Mail" it is only a type synonym. I suppose as a prerequisite I need to wrap it into a type constructor. But then what?

3 4