October 2013 - Haskell-Cafe

building a FFI library with cabal
by Tad Doxsee 05 Oct '13

05 Oct '13

Hi, I'm trying to create an FFI library to CHOLMOD ( http://www.cise.ufl.edu/research/sparse/cholmod/) but am having problems. My project is here: github.com/tdox/hcholmod. I can link an executable with the build script in the examples directory (line 4). But line 7 does not work (I get undefineds for functions that are in cholmod.a. Also, I can't build example1 with cabal. (Uncommenting the -- lines in hcholmod.cabal results is similar undefineds.) I would appreciate any help or hints. Thanks, Tad

1 1

Seeking Control.Lens Combinator
by Charlie Paul 05 Oct '13

05 Oct '13

Hello, I'm looking for a combinator along the lines of (&&&) :: Lens' a b -> Lens' a b' -> Lens' a (b,b') I can see how it could lead to lenses that don't follow the laws, but for Lenses which are somehow independent (like _1 and _2), it works perfectly well. Is there a way in lens to specify this "independence"? Thank you, Charles Paul

3 2

New command line parsing package
by Simon bergot 05 Oct '13

05 Oct '13

Hello all, I have recently uploaded argparser<http://hackage.haskell.org/package/argparser-0.3.2>, a command line parser library. I made it because of 3 reasons: - I am writing more and more scripts in haskell instead of python - I did not want to use the ones presented here<http://www.haskell.org/haskellwiki/Command_line_option_parsers>. I found that they were either lacking features I wanted, or using overkill extensions such as TemplateHaskell. - I wanted to learn how to create a package. So this is a beginner's work. I have uploaded it anyway because I feel it has very few dependencies and a satisfying set of features. Any feedback is welcome (especially about the documentation). -- Simon

1 0

Newclasses
by Wvv 05 Oct '13

05 Oct '13

"Newclasses" are not a new vision of classes! Not at all! Newclasses could elegant solve several instance problems! 1) we want to have "partly applied instances", like Parent2Child: Parent a => Child a like instance Applicative m => Monad m where return = pure -- we won't define here (>>=) 2) we want to have superclass' instances, like Child2Parent: Child a => Parent a like instance Monad m => Applicative m where pure = return (<*>) = ap 3) we want to have default instances outside of class and as many as possible, not the only one. like class Foo a where foo :: ... default foo :: ... foo = ... 4) we want to have multi-class instances to separate (or unite) classes like type Stringy a = (Show a, Read a) instance Stringy SomeData where read = ... show = ... (4)th problem we could solve separately, but maybe it isn't easy enough to do such de-sugaring, and it could much easier to add them in newclasses. (3)rd problem is solved partly, but not in universal way, non-flexible and a bit ugly. (2)nd problem is solved, but it is mostly impossible to use them and it is not recommend to use it for overlapping and incoherent issues. (1)st problem is unsolved at all (partly, it is possible to make depended classes, but checker don't check if we implement parent classes). This is a compose proposal. Newclasses solve these problems at once! As newtype is a data looking like a type, same newclass is looking like a class, but is mostly an instance! Mostly, but not full. To best understanding what "newclasses" is, let's look at (1)st problem: -- we wish to write -- instance Applicative m => Monad m where -- return = pure -- we write "newclass" instead of "instance" -- add "=>" and giving a name to newclass like a class -- this is not instance, so newclass can't overlap with any instance newclass (Applicative m) => Monad m => ApMonad m where return = pure data D a .... instance Applicative D where pure = ... (<*>) = ... -- creating instances from newclass is intuitive -- we implement here Monad class, not "ApMonad"; ApMonad is a just newclass instance ApMonad D where -- we already have 'return = pure', so we define only (>>=) (>>=) = ... What do we see here? Newclass looks like class, but it's mostly an instance! Newclass: Grammar: newclass constraint => Parent a => NewClassName a where ... As class, newclass has a name, which is unique and can't conflict with any other newclass or class names. As class, methods of newclass could be empty or implemented. As class, methods of newclass are not use in function inference. As class we can make an instance of newclasses and overwrite any of his functions! But, instance of newclass IS an instance of the parent (!)class! So, newclasses is like a de-sugaring. As instance, newclass contains only parents methods! On contrary to instancess, we don't use newclasse directly, it only help us to create instances. If we allow for newclass to be as Parent not only classes, but newclasses, then newclass can't be recursive: neither of his (Grand)Parent could be he by himself. If we allow for newclass muliple Parents, we solve (4)th problem too. Examples: We have: class Functor f where fmap :: (a -> b) -> f a -> f b class Applicative f where -- without "Functor f =>", this is a misfeature pure :: a -> f a (<*>) :: f (a -> b) -> f a -> f b class Monad m where return :: a -> m a (>>=) :: m a -> (a -> m b) -> m b class Ord' a where -- without "Eq a =>", this is a misfeature ... compare a b -- not as in Prelude | a < b = LT | a > b = GT | otherwise = EQ (1)st, Parent2Child: Common pattern to write newclass like: newclass Parent a => Child a => ParChild a where ... Examples: newclass (Applicative m) => Monad m => ApMonad m where return = pure data C1 a ... instance Applicative C1 where pure = ... (<*>) = ... instance ApMonad C1 where -- return is already defined (>>=) = ... -- newclass (Ord' a) => Eq a => OEq a where a == b = case compare a b of EQ -> True _ -> False data C2 ... instance Ord' C2 where (>) = ... (<) = ... instance OEq C2 -- empty instance for Eq (2)nd, Child2Parrent: Common pattern to write newclass like: newclass Child a => Parent a => ChParent a where ... Examples: newclass (Eq a) => Ord' a => Ord a where compare a b | a == b = EQ | a >= b = GT | otherwise = LT data C3 ... instance Eq C3 where (==) = ... instance Ord C3 where -- Ord with much effective `compare`, than Ord' (>) = ... (<) = ... -- newclass Monad m => Applicative m => MApplicative m where pure = return (<*>) = ap data C4 a ... instance Monad C4 where return = ... (>>=) = ... instance MApplicative C4 -- empty, everything is already defined ! -- newclass Monad m => Functor m => MFunctor m where fmap = liftM newclass Applicative f => Functor f => ApFunctor f where fmap f x = pure f <*> x data C5 a ... instance Applicative C5 where pure = ... (<*>) = ... instance ApMonad C5 where (>>=) = ... -- without return instance MFunctor C5 --(!) or instance ApFunctor C5 --(!!!) but not both (3)rd, default instances: Common pattern to write newclass like: newclass GenericConstraint a => Class a => GNewclass a where ... Examples: class ToJSON a where toJSON :: a -> Value newclass (Generic a, GToJSON (Rep a)) => ToJSON a => GenToJSON a where toJSON = genericToJSON defaultOptions newclass (Data a) => ToJSON a => DataToJSON a where toJSON = dataToJSON defaultOptions data C6 ... deriving (Generic, Data) -- don't forget to create an empty instance! instance GenToJSON C6 --(!) or instance DataToJSON C6 --(!!!) but not both (4)th, multi-class instances Common pattern to write newclass: newclass (Class1 a, Class2 a, Class3 a)=> Newclass a where ... Examples: newclass (Read a, Show a) => Stringy a -- without where data C7 ... instance Stringy C7 where read = ... show = ... -- class MinBounded a where minBound :: a class MaxBounded a where maxBound :: a newclass (MinBounded a, MaxBounded a) => Bounded a -- class Additive a where (+) :: a -> a -> a class Additive a => AdditiveZero a where zero :: a class Mulipicative a where (*) :: a -> a -> a class Mulipicative a => MulipicativeOne a where one :: a newclass (Additive a, Mulipicative a, Substravive a, FromInteger a) => Num a -- without where newclass (AdditiveZero a, MulipicativeOne a, Substravive a, FromInteger a) => NumFull a -- Benefits: - "newclass" is a very powerful tool - it is universal solution without extra assumptions! - solve 4 big problems with instances - it is a Huge step - it is a step forward, not aside - it is developing function's muscles to Haskell Disadvantages: - new reserved word "newclass" - with Child2Parrent problem it is needed to write empty instances - main difficulty - it is not easy to implement this extension! Newclasses could also help to reorganize some Prelude classes, like Ord, Applicative, Functor, Bound, Num, ... What do you think of this proposal? You opinion is important and significant! Let's help to develop Haskell together! Do you like "newclasses"? Do you want to use them? Do you think newclasses are appropriate solution for written problems? Could newclasses help to resolve some other problems? Are they useless? Do you see huge difficulties of implementation? Do you have more elegant ideas, which correspondent with newclasses? If it is not clear enough, I write more clearly. Any feedback is welcome! -- View this message in context: http://haskell.1045720.n5.nabble.com/Newclasses-tp5737596.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

3 9

haskell platform broken in ubuntu
by Rustom Mody 05 Oct '13

05 Oct '13

I just upgraded my ubuntu laptop to 13.04 and haskell platform is gone!! http://askubuntu.com/questions/286764/how-to-install-haskell-platform-for-u… What is the current status on this? Is 13.10 going to correct this?

3 6

Calling Python from Haskell
by Manuel Gómez 04 Oct '13

04 Oct '13

Hi list, What’s the preferred way of calling into Python from Haskell? I’ve found MissingPy[0], but it seems to be somewhat bitrotten and a couple of experiments yielded segfaults. There’s also the cpython package[1], but that seems to require Python 3.3, and I’m trying to call into code written for 2.7. Are there any other alternatives, apart from direct execution of a Python interpreter with forkProcess and such? [0]: <http://hackage.haskell.org/package/MissingPy> [1]: <http://hackage.haskell.org/package/cpython>

2 1

Machine Learning Startup Seeks Software Architect
by Charles Weitzer 04 Oct '13

04 Oct '13

My name is Charles Weitzer. I have a client with a startup quantitative hedge fund located in Northern California. The Fund currently manages a very healthy amount of capital. The founders have PhD's in Computer Science from Stanford and in Statistics from Berkeley. They have made unpublished discoveries in the field of Machine Learning and in other areas as well. This is a small, nimble company that is going to be the next big thing in terms of quantitative strategies and systems. The team is looking for an extremely talented software architect to design and build a new trading system and to rebuild their current system. The position is very challenging intellectually and involves architecting and implementing a diverse set of core infrastructures, including new production systems, scientific-computing environments, and modern data stores. From the company's point of view, this is a distributed computing problem. The position is definitely high-impact and the scope of work mostly greenfield. There is quite a bit of freedom in the choice of software tooling (e.g. it doesn't have to be C++ or Java and they are very open to Erlang, Scala, Haskell, and others). Qualifications include a proven track record of writing correct, well-designed software, solving hard problems and delivering complex projects on time. You should preferably have experience with high assurance, distributed, fault-tolerant systems. Experience with functional programming as well as real-time, low-latency, cache-friendly systems is a very big bonus. They are getting big fast, so a willingness to take initiative and a gritty determination to produce is essential. The title, scope, and level of seniority is very flexible. As well, given their returns, this is a chance to become extremely wealthy. Join a team that includes faculty at premier universities and PhD's from top-tier schools, led by the founder and CEO of a successful Internet infrastructure startup. You will have high-impact, and can expect frequent interaction with the researchers, officers, and founders. Compensation and benefits are highly competitive. If you are interested or just curious, let me know the best way to start a dialog. If you know of someone who you would highly recommend, I would appreciate it if you could forward this on to them. Talk soon, Charles Weitzer CEO\Senior Recruiter Charles Weitzer and Associates, Inc. Global Financial Recruiting Services <mailto:Charles@CharlesWeitzer.com> Charles(a)CharlesWeitzer.com Voice: USA (510) 558-9182

1 0

Store type-class polymorphic values generically
by Christopher Done 04 Oct '13

04 Oct '13

It's very easy to state this problem without enough details and mislead people into providing a solution for a different problem, so I'll try to include all the information and use-case. I need a function that can store a value in a concrete opaque type. I know the type of the value when I store it, and I know the type of the value when I extract it, insofar as I know the type inclusive of a class constraint context. But the container must store many different things at once. Given that I know the type of the thing when I store it and when I extract it, there ought to be a “safe’ way to do this, in the same way that Data.Dynamic is “safe”. The use-case is I want to store a table of IO-like actions in an opaque like [(Name,Dynamic)] and I write into it statically, e.g. in GHCi. The reason is due to this approach: go_ref = newIORef go_ go_ = putStrLn "Hello, World!" go = do go' <- liftIO $ readIORef go_ref go' main = forkIO $ forever $ go then in GHCi I writeIORef go_ (putStrLn "What's up?") then go will now start printing "What's up?" This works now, I can do this presently. However, I want to write this as a core-to-core translation as a ghc-plugin. I want the definition go = putStrLn "Hello World!" to be translated to what I wrote above. Core cannot generate new names to be exported from a module, so go_ is now gone. So how does the `go' function read and write the IORef? With a table: table = unsafePerformIO $ newIORef [("go",Dynamic go_ref)] Hurrah! I can now put table in some module like DynamicUpdate.table and then read/write to it from Core in the generated definition of go. Done. But how do I update it from the GHCi REPL? What follows is where I'm stuck. So with Data.Dynamic, I have: toDyn :: Typeable a => a -> Dynamic With toDyn I can store concrete values and functions and then extract them: λ> fmap ($ 12) (fromDynamic (toDyn ((*2) :: Int -> Int)) :: Maybe (Int -> Int)) Just 24 But the problem with toDyn is that it can only store concrete values due to the Typeable constraint. So then I turn to existentials and unsafeCoerce: λ> data Anything = forall a. Anything a λ> let x = Anything id λ> case x of Anything (unsafeCoerce -> (id' :: Int -> Int)) -> id' 123 123 Great, I was able to store a value which contained non-concrete types. But now consider the case of a value with type-class polymorphic type in it: y = Anything (print :: Show a => a -> IO ()) Which cannot be type checked, because the Show instance is ambiguous. So you might propose to use rank-N types, like this: data Anything = Anything (Show a => a -> IO ()) Now I can store print in there happily. But now the type is not Anything, it's not generic anymore. Maybe I want to store `print', maybe I want to store `readLn'. Dead end. I don't want to have to generate existential wrappers for all functions I might possibly want to store. The support of Constraints in this page http://blog.omega-prime.co.uk/?p=127 makes me think that it's still possible. It would be cool to somehow store a tuple of the dictionary of the constraint and the value itself and then I'd later be able to extract it. But I'm finding it difficult developing anything even simple. Any ideas/help appreciated!

4 5

Using lenses
by Simon Peyton-Jones 03 Oct '13

03 Oct '13

(I sent this to 'libraries' but Kim-Ee suggested adding Café, where so many smart people hang out.) Friends Some of you will know that I've promised to give a talk about Edward's lens library<http://hackage.haskell.org/package/lens> at the Haskell Exchange<http://skillsmatter.com/event/scala/haskell-exchange> in London next Wednesday (9th). I did this to give everyone (including me) a break from GHC hackery, and also to force me to learn about this lens voodoo that everyone is twittering about. Edward generously gave me quite a bit of one-to-one attention last week (my hair is still standing on end), but this message is to ask your help too. Specifically, I'd like to give some compelling use-cases. If you are using the lens library yourself, could you spare a few minutes to tell me how you are using it? I expect to cover Lens and Traversal but not Prism. The use-case everyone starts with is nested records, but I'd like to go beyond that. The next levels seem to be: · Lenses as views of data that isn't "really there" e.g. regarding a record with rectangular coordinates as having polar coordinates too. · Lenses and Traversals that focus on elements of finite maps (Control.Lens.At) What else? I'm sure you are using them in all sorts of cool ways that I would never think of, and I'd love to know. Please don't tell me anything secret! To give everyone the benefit I may just concatenate all the replies and send to you all, so please say if you don't want me to do that with yours. And don't burn too many cycles on this...I don't want to waste your time, and I can always get back to you if I can't understand what you say. Sooner is better than later...Weds is coming. Simon "Edward's prophet" PJ

9 8

ANN: Nomyx 0.3 beta, the game where you can change the rules
by Corentin Dupont 03 Oct '13

03 Oct '13

Hello everybody! I released the third beta of Nomyx <http://www.nomyx.net>, the only game where You can change the rules! The second beta was a success. Great players (byorgey, nomeata, Toxaris...) proposed amazing rules, effectively building a nice universe. For example, a banking system in ecu was proposed, with cash monies and even a way to buy pies! Other rules like different democratic systems were proposed. You can check those rules in this thread: http://www.nomyx.net/forum/viewtopic.php?f=4&t=1520 Let's start a new game! I propose to set it in the medieval era. Dear rulers, I'll let your imagination speak! Please login to the game here: www. nomyx.net:8000/Nomyx. The corresponding forum thread is here: http://www.nomyx.net/forum/viewtopic.php?f=4&t=1523 Some background: this is an implementation of a Nomic [2] game in Haskell(I believe the first complete implementation of a Nomic game on a computer). In a Nomyx game you can change the rules of the game itself while playing it. The players can submit new rules or modify existing ones, thus completely changing the behaviour of the game through time. The rules are managed and interpreted by the computer. They must be written in the Nomyx language <http://hackage.haskell.org/package/Nomyx-Language>, which is a subset of Haskell. At the beginning, the initial rules are describing: - how to add new rules and change existing ones. For example a unanimity vote is necessary to have a new rule accepted. - how to win the game. For example you win the game if you have 5 rules accepted. But of course even that can be changed! Here is a video introduction and first tutorial of the game: http://vimeo.com/58265498 The example file gives a lot of examples of rules that you can submit: www. nomyx.net:8000/src/Language/Nomyx/Examples.hs changes in V0.3: - creation of admin role - inputs made by the rules now include checkboxes, buttons, textarea - outputs made by the rules are now managed (create, update, delete) - refactored the GUI - bumped to GHC 7.6 - compilation errors are displayed in context - bug fixes Cheers, Corentin [1] www.nomic.net

1 1