July 2016 - Beginners

search function by type
by OxFord 09 Jul '16

09 Jul '16

Hello, is it possible to search for all functions that have concrete type? For example I want to convert [Char] to [Word], but I don't know whether function for that exist. Or I wanna find contructor for SomeSpecialType and I don't know it's name. Can you search those function in GHCI via some command? E.G. GHCi, version 8.0.1: http://www.haskell.org/ghc/ :? for help [}{] :searchtype / :st [Char] -> [Word] Found X match(es): foo1 :: (Char -> Word) -> [Char] -> [Word] foo2 :: [Char] -> [Word] ... [}{] :st SomeSpecialType Found 1 match(es): foo3 :: ... -> .. -> SomeSpecialType King Regards, Ford

3 2

Case vs Guards. I still don't know what is the difference
by Semih Masat 07 Jul '16

07 Jul '16

Hello, I am new to Haskell and trying to learn it with learnyouahaskell.com and Pluralsight Haskell course. And i have a very noob question. I understand that *if .. else* is just a syntactic sugar over *case. *But what about guards then ? Are guards also *case *in different syntax ? Or vice versa ? Like with an example. anyEven nums | (length (removeOdd nums)) > 0 = True | otherwise = False anyEven' nums = case (removeOdd nums) of [] -> False (x:xs) -> True I can do the same thing with both of them. As i understand the only different thing is, with *case *i can manipulate the parameter (like here in the example i used removeOdd) and can use the manipulated parameter to decide what to do after that. So i will not need to use removeOdd function inside the case. ( maybe i will need to use in every guard definition if i choose to use guards ) Is this it? Is this the only difference between them ? And if it is, why haskell needed do implement both of them. Can't we use function like removeOdd before using it on case or guard functions ? Thanks, and sorry if my english is bad. Semih Masat

3 4

Polymorphism ad hoc
by OxFord 07 Jul '16

07 Jul '16

Hello, why does haskell support only parametric polymorphism? Is it because ad hoc p. || multiple dispatch is bad in general? King Regards, Ford

2 1

Preorder function application
by OxFord 04 Jul '16

04 Jul '16

Hello, why Haskell doesn't apply functions in preorder? e.g. f x = x max 1 f 2 > 2 max f 1 f 2 > 2 max max f 1 f 2 f 3 > 3 f f f f f f f f f f f 1 > 1 Thus you would need to put the arguments into brackets only when you want to partially apply that function. Is the current method more readable or error prone? King regards, Ford

2 1

Howto reverse a Data.Array
by Timothy Washington 04 Jul '16

04 Jul '16

I'm still trying to get an intuitive understanding of Haskell's Data.Array <http://hackage.haskell.org/package/array-0.5.1.1/docs/Data-Array.html#g:5>, in contrast to Data.List or Data.Vector. I very much want a nested array (a matrix), where the parent list (or rows) are reversed. But neither *A.array* nor *A.istArray* allow indicies to be reversed in their constructors, nor the list comprehensions that generate the elements The only reason I'm using an array, is for the *A.//* function (operating on a matrix). Otherwise, I'd use Data.Vector <https://hackage.haskell.org/package/vector-0.11.0.0/candidate/docs/Data-Vec…> which does have a reverse function, but a less powerful *V.//* , that doesn't accept coordinates in a matrix. Can I reverse a Data.Array? If not, then why. Thanks Tim

2 2

Problem with numeric types
by Russ Abbott 02 Jul '16

02 Jul '16

I'm trying to define primes using just list comprehension. The following works. Prelude> odds = [3, 5..] Prelude> primes = 2 : [p | p <- odds, null [d | d <- take (p `div` 4) odds, p `mod` d == 0]] But when I replace "take (p `div` 4)" with takeWhile (<=(sqrt p)) as in Prelude> primes' = 2 : [p | p <- odds, null [d | d <- (takeWhile (<=(sqrt p)) odds), p `mod` d == 0]] I get no error message on entering the statement, but when I run Prelude> take 8 primes' I get an error message that I can't understand. <interactive>:23:1: error: • Ambiguous type variable ‘a0’ arising from a use of ‘it’ prevents the constraint ‘(Floating a0)’ from being solved. Probable fix: use a type annotation to specify what ‘a0’ should be. These potential instances exist: instance Floating Double -- Defined in ‘GHC.Float’ instance Floating Float -- Defined in ‘GHC.Float’ • In the first argument of ‘print’, namely ‘it’ In a stmt of an interactive GHCi command: print it I tried using (ceiling (sqrt p)), and I tried to explicitly declare it an Int, but neither of those helped. I'd appreciate some help. Thanks. P.S. The following runs fine. Prelude> takeWhile (<= (sqrt 99)) odds So I'm completely stumped.

2 1

Problem with installing GHCi 8.0.1
by Russ Abbott 02 Jul '16

02 Jul '16

I downloaded GHCi 8.0.1 for Windows. Since I don't have administrator privileges I repeatedly clicked no when asked if I wanted to stop and restart with elevated privileges. Eventually it installed (in a folder on the desktop). The next step tells me to change some lines in the cabal user-config file. I followed the instructions to find that file, but all I got was an html file, which didn't seem right. I looked manually in multiple directories, but nothing seemed like the right file. Would someone help me find the file that needs to be changed. Also, WinGHCi does not appear in my start menu. Should it appear there by this point? Thanks.

2 2