February 2011 - Beginners

Haskell Help
by Andrei M. 10 Feb '11

10 Feb '11

I need to write two functions for a coursework that will behave as follows... unction 1)removeTautologies :: Formula->Formula If in a clause, a literal and its negation are found, it means that the clause will be true, regardless of the value finally assigned to that propositional variable. Consider the following example: (A v B v -A) ^ (B v C v A) The first clause contains the literals A and -A. This means that the clause will always be true, in which case it can be simplify the whole set to simply (B v C v A) . I was tinking of using something like removeTautologies (f:fs)=filter rTf:removeTautologies fs where rT-is supposed to take the firs Literal from the clasue and search for a similar one,if one si found we compare the values if not the we go to the second literal. ->Function 2)pureLiteralDeletion :: Formula->Formula This is a little bit complicate but from What I get this function is suppose to implement a simplification step that assumes as true any atom in a formula that appears exclusively in a positive or negative form (not both). Consider the formula: (P v Q v R) ^ (P v Q v -R) ^ (-Q v R) Note that in this formula P is present but -P is not. Using Pure Literal Deletion it can be assumed that the value of P will be True thus simplifying the formula to (-Q v R). If the literal were false then the literal would simply be deleted from the clauses it appears in. In that case any satisfying model for the resulting formula would also be a satisfying model for the formula when we assume that the literal is true. removeTautologies :: Formula -> Formula removeTautologies = map rt.map head.group.sort where rt ((x, x') : (y, y') : rest) | x' == y' = rt rest | otherwise = (x, x') : rt ((y, y') : rest) pureLiteralDeletion :: Formula -> Formula pureLiteralDeletion = map lD.map head.group.sort.concat where lD ((x, x') : (y, y') : rest) | x' == y' && x/=y = lD rest | otherwise = (x, x') : lD ((y, y') : rest) this is what I've witten so far but they don't wok poperly....any sugestions?

1 0

Ffi, memory allocation and failure
by ch.gosch＠googlemail.com 09 Feb '11

09 Feb '11

Dear Haskellers, I have a question about the FFI and interfacing to C++ via C. I want to use funcionality that I have implemented a while ago using C++ classes, and I also see this as an exercise on how to create wrappers using the Haskell FFI. The class I want to wrap first is a matrix class. Here is the thing: In order to create a matrix object, I create a C function that calls the C++ new operator, creating a matrix object, and returning it as a void pointer. Deletion is handled analogously. Then in Haskell, I import the two functions, and I have a matrix type like this: newtype Matrix = Matrix (Ptr Matrix) From this I create a ForeignPtr with newForeignPtr, giving the deletion function (which calls the C++ destructor for a matrix) as finalizer. Testing this showed me that it does work. /But/: So far everything is happening in the IO monad. However, I want to be able to work with matrices outside IO. Now, of course memory allocation can fail, and I don't want to use something like unsafePerformIO when there can actually be a failure when calling the constructor for a matrix. Is there a recommended way to do this? Also, is there a way to get memory from the Haskell runtime system in a safer way and construct the object in that memory? I hope I could make myself clear enough, as usual thanks for any hints!! Christian

5 4

question concerning conversion from one type to another
by Michael Litchard 09 Feb '11

09 Feb '11

I'm using Network.Curl, and have to operate on values of type InfoValue. Now, an InfoValue looks exactly like a [String], and I would like to operate on it as if it were. Is the proper approach to convert it to a [String], or would I be using fmap?

2 1

Help me make sense of this error
by Michael Litchard 08 Feb '11

08 Feb '11

Here's the three main functions involved. Let me know if I am missing pertinent information. I'm having difficulty unraveling this error message. Help? > obtainCookies :: Curl -> String -> IO () > obtainCookies curl responseBody = do > return $ last $ liftM $ > mapM (flip (curlResp2 curl) resourceOpts) > screenScraping responseBody > screenScraping :: String -> [URLString] > screenScraping responseBody = > let collectedStrings = processHTML responseBody > collectedIDLists = createIDList collectedStrings > in constructedResourceURIs urlBase collectedIDLists > constructedResourceURIs :: String -> [String] -> [URLString] > constructedResourceURIs url resourceIDs = > let frontURI = url ++ "/launchWebForward.do?" > midURI = map (frontURI ++) resourceIDs > in map (++ "&policy=0&returnTo=%2FshowWebForwards.do") midURI HtmlParsing.lhs:81:22: Couldn't match expected type `[URLString]' against inferred type `String -> [URLString]' In the second argument of `mapM', namely `screenScraping' In the second argument of `($)', namely `mapM (flip (curlResp2 curl) resourceOpts) screenScraping responseBody' In the second argument of `($)', namely `liftM $ mapM (flip (curlResp2 curl) resourceOpts) screenScraping responseBody'

3 2

Sample code for Data-Vector-Unboxed
by C K Kashyap 08 Feb '11

08 Feb '11

Hi, Could someone please point me to sample code using Data-Vector-Unboxed? Regards, Kashyap

1 0

gtk2hs [GIMP ToolKit]
by Patrick Lynch 06 Feb '11

06 Feb '11

...I'm going to try to install gtk2hs again... ...I'll be using the link http://www.haskell.org/haskellwiki/Gtk2Hs ...I'd appreciate any help...

6 30

Re: [Haskell-beginners] lazy IO by example
by Tim Baumgartner 05 Feb '11

05 Feb '11

2011/2/5 Dean Herington <heringtonlacey(a)mindspring.com> > In your simplified example, it suffices to wrap unsafeInterleaveIO around > (mapM content ds). In a more realistic case, you'd need to add some > `unsafeInterleaveIO`s in `files` and `directories`, too. > Thanks a lot! Now it was easy to write a lazy findFiles. Would one have to write down a lot in order to see "on paper" why the original example was not lazy? If I coud do this, I might run into fewer problems in the future. > Dean > > P.S. You might also want to look into iteratees, which accomplish a similar > goal in a more robust way. > Yes, I already looked over the article in Monad reader #16. But then I thaught I will first gain more understanding of the traditional way and then return to iteratees. Regards Tim

1 0

New to Haskell: constructing Data Types
by Tyler Hayes 04 Feb '11

04 Feb '11

As an exercise, I'm attempting to create an abstract syntax tree for a very small language: data Expression = Variable String | Integer | Expression BinOp Expression deriving Show data Statement = Assignment Variable Expression --error! | If Expression Statement Statement | While Expression Statement | Compound Statement deriving Show data BinOp = Add | Mult | LessThan | GreaterThan deriving Show However, there is a problem with having "Variable" as an argument to the "Assignment" constructor of the "Statement" data type: Prelude> :l test.hs [1 of 1] Compiling Test ( test.hs, interpreted ) test.hs:8:28: Not in scope: type constructor or class `Variable' Failed, modules loaded: none. An 'assignment' is a variable followed by an expression in this example, so what can I do to fix this? Thanks, Tyler

3 2

lazy IO by example
by Tim Baumgartner 04 Feb '11

04 Feb '11

Hi community, In the following code, the main function only wants to print the first 5 numbers of a huge list. But since the computation is not lazy, this doesn't work in a satisfactory way. content :: Int -> IO [Int] content i = do fs <- files i ds <- directories i fss <- mapM content ds return $ fs ++ concat fss files i = return [1..i] directories i = return [1..i-1] main = content 1000000 >>= print . take 5 Now I'd like to know if it's possible to make this "IO" lazy, using unsafeInterleaveIO. I tried to do it, failed. Perhaps someone can help. I guess another weakness of this code is the bad performance of "dfs ++ concat fss", but currently I don't care about that. Thanks for any help Tim

2 1

Informal proof of bijective mapping between Naturals and Natural pairs
by Lyndon Maydwell 03 Feb '11

03 Feb '11

Hi Beginners. I was musing over weather Naturals were isomorphic to Natural pairs and wrote the following "proof". I don't have any formal mathematical training beyond calculus, and was wondering if this was an okay approach to take, if there were any better techniques available, if my terminology makes sense, and if my use of Haskell could be improved. Any pointers would be greatly appreciated! The code follows, and is also available as a gist: https://gist.github.com/804248 ---------- import Control.Monad.Logic import Data.Maybe import Data.List import Data.Numbers.Primes import Test.QuickCheck -- This program demonstrates a mapping between the pairs of natural numbers and a subset of co-primes. -- The property should hold for all sized lists, not just pairs. Ordering is preserved. -- Question: Are we able to compress the range to create a bijection? -- Answer: Yes! We can use the breadth-wise indces of the products, rather than the products themselves. -- Pair to Number x_p = (evens !!) y_p = (odds !!) xy_p x y = x_p x * y_p y xy_c x = ns_nc . xy_p x -- Number to Pair p_xy n = (x,y) where x = fromJust $ findIndex (`divides` n) evens y = fromJust $ findIndex (`divides` n) odds c_xy = p_xy . nc_ns -- Sparse Number to Compact Number numbers = odds >>- \x -> evens >>- \y -> return (x*y) -- I don't really understand LogicT... :-( ns_nc n = fromJust $ findIndex (==n) numbers -- Compact Numner to Sparse Number nc_ns = (numbers !!) -- Helpers evens = map (primes !!) [0,2..] odds = map (primes !!) [1,3..] x `divides` y = y `mod` x == 0 -- Id referencing properties prop_p1 = forAll (elements [(x,y) | x <- [1..6], y <- [1..7]]) f where f (x,y) = (x,y) == c_xy (xy_c x y) prop_p2 = forAll (elements [1..100]) f where f c = c == uncurry xy_c (c_xy c)

2 2