January 2014 - Beginners

Just started working with Haskell. Need some help
by Pyro Crane 19 Jan '14

19 Jan '14

Hi, I just started working with Haskell Running it on a RHEL linux operating system I was able to download and install it, along with libraries and packages (or, at least, what I thought were the libraries and packages) But, when I try to run a simple script, I keep getting the error : "Could not find module XXXXXX".............. I understand what this error means ---- there is a package/library missing. I tried using the "Cabal Update" command, but this did not solve the problem. So, I am now fixing this problem the primitive way, as follows : (a) I use Google to first find out what the missing module is (b) then, based on what google says, I am able to determine the name of the actual package that needs to be installed (c) I use "Cabal Install" to install the package (d) I try to run the script again (e) I get the same error, but this time, it cannot find a different module (f) I go back to Google to locate the missing module etc, etc, etc, etc Obviously, this is as slow as it is ridiculous............... especially when one considers that there are probably hundreds of these modules/packages And, now, I've come up against a module/package, which I am unable to locate in Google. It's missing from my installation, and Google does not know what it is, or where I can find it Is there any way to simply install ALL required packages??? I thought I already did this, but obviously I missed something Thanks

5 5

Designing complex Haskell programs
by Courtney Robinson 15 Jan '14

15 Jan '14

I'm trying to take the training wheels of and moving more of my code base to Haskell from C++ but finding it increasingly tricky. I have a subset of a gossip protocol written in C++. When a server comes online it connects to 1 or more nodes already in the cluster and get data from them about other nodes they know of. The new node merges the information and keeps a copy of the merged view. Every so often it contacts the nodes it knows about and refreshes the merged view. It also must have the up to date view ready to be sent in response to a new node joining. I currently can't wrap my head around how to maintain this state. How would a more experienced Haskeller approach this problem? Code is OK if it demonstrates a particular point but I'm more interested in the line of thought that would go into designing a solution as I suspect that'll be more useful as I get further into the migration. As a gauge to you for my current level in Haskell. I read and understand most Haskell programs fine. I write some but currently heavily rely on hackage/hoogle docs for APIs, even some common ones. Thanks

4 10

Initialize a data type in (Writer (Endo a)) monoid
by Dmitriy Matrosov 14 Jan '14

14 Jan '14

Hi. I have the following data type > import Data.Monoid > import Control.Monad.Writer > data Graph = Graph {graphTitle :: String, graphPoints :: [Int]} > deriving (Show) > emptyGraph :: Graph > emptyGraph = Graph {graphTitle = "", graphPoints = []} Then i want to initialize it in Writer monad using (Dual (Endo Graph)) as monoid: > type GraphM t = WriterT (Dual (Endo Graph)) t > > myGraph :: Monad t => GraphM t [Int] > myGraph = do > tell (setTitle "ab") > tell (setTitle "ABCD") > tell (modifyPoints (1 :)) > tell (modifyPoints (2 :)) > return [1, 2] and then i can use it e.g. like > getGraph :: Monad t => (GraphM t a) -> t Graph > getGraph m = do > g <- execWriterT m > return (appEndo (getDual g) emptyGraph) > > printGraph :: IO () > printGraph = getGraph myGraph >>= print And to make this work i need two functions implemented for each record of Graph: get :: Graph -> a set :: a -> Graph -> Graph I can define per-record instances of them, like > setTitle :: String -> Dual (Endo Graph) > setTitle x = Dual . Endo $ \g -> g{graphTitle = x} > > modifyPoints :: ([Int] -> [Int]) -> Dual (Endo Graph) > modifyPoints f = Dual . Endo $ \g -> let xs = graphPoints g > in g{graphPoints = f xs} but if Graph has many records, for most of them these 'set' functions ('get' functions i'll have from records) will look very similar. Is there a way how i can define all of them "in one row", i.e. using some generic 'set' implementation to which i should only pass e.g. record name or smth else? I.e. so, that above myGraph will look like .. tell(set r x) .. where r is record name (or smth else referencing particulat record) and x is value.

2 1

Getting a variables type as a string at runtime
by Gareth Morgan 13 Jan '14

13 Jan '14

Bit of a simple question but I'm struggling to find information on this via Google. I have a sum type, when the input is type A I want to process it, when it is type B I want to return MyError "Expected type A but actually got type B". However I also have C,D,E,F, etc which should also return similar errors so don't want to hardcode this string. Is there a function I can call on a variable to get the name of the data constructor that created it? I could create a typeName function but I want to avoid boiler plate if I can. For reference what I'm actually doing is parsing Java class files. The constant pool I've represented as a map of a sum ConstantPoolEntry type (the Java constant pool for some reason skips an index on some types so can't be a simple list). Certain Java class file entries have a reference to the constant pool that must be of a specific constant type so I want to report badly formed entries to the user. Thanks, Gareth

4 5

combine expressions in do block
by Renah Scarowsky 09 Jan '14

09 Jan '14

Is it possible to bind two expressions together within a case inside a do block? Given the following code running in the writer monad: f a = do case g a of Nothing -> return () Just b -> h a (field1 b) (field2 b) case g a of Nothing -> return () Just b -> i a (j a) (field1 b) (field2 b) How can I combine these two cases together (or otherwise simplify the code)? Thanks, Renah Scarowsky Suite Solutions Create>Manage>Deploy <http://www.suite-sol.com/> http://www.suite-sol.com

3 2

Picking apart getLine
by Angus Comber 08 Jan '14

08 Jan '14

Before looking at getLine, I can understand this: getnumber :: IO Int getnumber = do x <- getChar if isDigit x then return (ord x - ord '0') else return 0 OK, it is not a very useful function but at least I understand it. return is required so that the function returns an IO Int. But I don't understand this: getLine' :: IO String getLine' = do x <- getChar if x == '\n' then return [] else do xs <- getLine' return (x:xs) I can understand what will happen if a user enters a newline (only). return [] brings the empty list into the monadic world. But what is happening if x is not a newline? xs <- getLine' will recursively call getChar and retrieve another character from the input stream. But it will do this BEFORE the return (x:xs) - so what is happening to all the head elements in the list - the x element? It is difficult to picture in my mind how this is working. I understand recursion but this looks tricky. Can someone help me work this out?

2 3

Why WinGHCi does not complain with no import?
by Angus Comber 08 Jan '14

08 Jan '14

I created my own version of getLine like this: getLine' :: IO String getLine' = do x <- getChar if x == '\n' then return [] else do xs <- getLine' return (x:xs) Basically I created a new file called test.hs and copied above code into it. I opened WinGHCi afresh and :load test.hs I can run getLine' in WinGHCi with no problem. But how does WinGHCi know about getChar and IO? I assumed I would have to import like this: import System.IO Is it because WinGHCi does this import for me? If I write this program and invoke using runghci: getLine' :: IO String getLine' = do x <- getChar if x == '\n' then return [] else do xs <- getLine' return (x:xs) main = do getLine' It also works without complaining. Is there a list of built-in libraries? Is that platform dependent?

3 2

Re: [Haskell-beginners] How to run Parser in chapter 8 Graham Hutton Haskell book
by Angus Comber 08 Jan '14

08 Jan '14

For anyone else benefit, this is how I fixed issue. 1. download Parsing.lhs from author's website. 2. Also download parser.lhs and paste contents into my prog_haskell.hs file, removing text comments and > symbols. 3. At top of my prog_haskell.hs file you need import Parsing You can then call the functions in WinGHCi, eg eval "3+5*2" For some reason I was not able to work out how to import both Parsing.lhs and parser.lhs together. Will have to read up on modules later. On 7 January 2014 11:43, Christian Maeder <Christian.Maeder(a)dfki.de> wrote: > Hi Angus, > > did you get an answer for your question? (Just pressing "reply" only > answers to the list, as happened to my earlier answer to your initial > question.) > > I don't know if WinGHCi works like ghci under linux (and I don't know > which files Parsing.lhs and parser.lhs you tried to load.) > > At the ghci Prompt you can type ":browse" to see which names are in scope. > Maybe parser.lhs could not be loaded for some reason that I cannot > reproduce. (Maybe eval is explicitly not exported?) > > Cheers Christian > > > Am 03.01.2014 18:43, schrieb Angus Comber: > >> I found some notes on the authors website explaining that the book code >> would no longer work. So I downloaded two files: Parsing.lhs and >> parser.lhs. In WinGHCi I then issued :load on both files. >> >> parser.lhs contains a function, eval, as follows: >> >> <other stuff above> >> > eval :: String -> Int >> > eval xs = case (parse expr xs) of >> > [(n,[])] -> n >> > [(_,out)] -> error ("unused input >> " ++ out) >> > [] -> error "invalid input" >> >> But in WinGHCi if I run: eval "2*3+4" I get error := Not in scope: `eval' >> >> I assume this is some sort of namespaces feature of Haskell that I have >> not read about? Or I somehow have to state what library modules to >> use??? Can anyone help? >> >> >> >> >> >> >> >> On 3 January 2014 15:58, Angus Comber <anguscomber(a)gmail.com >> <mailto:anguscomber@gmail.com>> wrote: >> >> I am reading Chapter 8 of Programming Haskell by Graham Hutton and >> trying to run the code in the book. >> >> It seems that there are already defined library implementations of >> Parser so I used Parser' and generally ' on the end of each >> identifier to attempt to eliminate this problem. >> >> So the code I copied from the book is: >> >> type Parser' a = String -> [(a, String)] >> >> return' :: a -> Parser' a >> return' v = \x -> [(v, x)] >> >> failure' :: Parser' a >> failure' = \inp -> [] >> >> -- item doesn't seem to conflict with anything so no ' >> item :: Parser' Char >> item = \inp -> case inp of >> [] -> [] >> (x:xs) -> [(x,xs)] >> >> >> parse' :: Parser' a -> String -> [(a, String)] >> parse' p inp = p inp >> >> >> p :: Parser' (Char, Char) >> p = do x <- item >> item >> y <- item >> return' (x,y) >> >> When run from WinGHCi I get error: >> >> prog_haskell.hs:458:9: >> Couldn't match type `[(Char, String)]' with `Char' >> Expected type: String -> [((Char, Char), String)] >> Actual type: Parser' ([(Char, String)], [(Char, String)]) >> In the return type of a call of return' >> In a stmt of a 'do' block: return' (x, y) >> In the expression: >> do { x <- item; >> item; >> y <- item; >> return' (x, y) } >> >> 458.9 is line at end with return' (x,y) >> >> >> Also in the chapter was the definition of >>= sequencing. as in: >> >> (>>=) :: Parser a -> (a -> Parser b) -> Parser b >> p >>= f = \inp -> case parse p inp of >> [] -> [] >> [(v,out)] -> parse (f v) out >> >> But I assumed this would already be in standard Haskell library and >> so I didn't need to define. But even when I did add, still get same >> error. >> >> How do I fix this? >> >> >> >> >> _______________________________________________ >> Beginners mailing list >> Beginners(a)haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> >

1 0

Using IORef
by Courtney Robinson 07 Jan '14

07 Jan '14

I've found this example. counter :: IO Int counter = unsafePerformIO $ newIORef 0 getUnique :: IO Int getUnique = atomicModifyIORef counter $ \x -> let y = x + 1 in (y, y) Which I ran, and I know works in as much as calling getUnique multiple times results in an incremented value. I have two things I'm failing to see about it. 1) Why doesn't counter return a newIORef with value 0, every time getUnique is called there by producing a value of 1 no matter how often getUnique is called? Is this this because it returns an IORef so counter just becomes bound to the value it first returns and it's not evaluated again and the IORef is just returned, or something else? 2) Why does the lambda return a tuple and why does the example return the same value in both positions i.e. (y,y)? The only thing I could think of is because the name has "atomic" that the values in the tuple is being used for some sort of compare and swap,CAS. P.S I didn't find http://hackage.haskell.org/package/base-4.6.0.1/docs/Data-IORef.html enlightening on the matter and I know that example is not great because the atomicModifyIORef is lazy so could lead to stackoverflow as that doc. page says... Thanks

3 5

"Segmentation fault/access violation" when working with FFI
by Leza Morais Lutonda 06 Jan '14

06 Jan '14

Hi, I has trying to use the portaudio [1] package to play some sound, and running the examples [2] results in "Segmentation fault/access violation" error. How can I solve this problem? Thanks. [1] http://hackage.haskell.org/package/portaudio [2] https://github.com/sw17ch/portaudio/blob/master/examples/Example1.h

2 2