October 2010 - Haskell-Cafe

Haskell Platform: failed to parse output of 'ghc-pkg dump'
by Tom Hawkins 05 Oct '10

05 Oct '10

I'm having trouble installing Haskell Platform on Windows. After the install, I run "cabal update", which appears to work: 00-index.tar.gz is deposited in C:/Documents and Settings/<user>/Application Data/cabal/packages/hackage.haskell.org. However, when I try to "cabal install" anything, I get: cabal.exe: failed to parse output of 'ghc-pkg dump' "ghc-pkg dump --global" appears to run just fine. Ideas? Suggestions? -Tom

1 1

ByteString missing rewrite RULES (zipWith' f a = pack . zipWith f a)
by Thomas DuBuisson 05 Oct '10

05 Oct '10

All, (I notice ByteString still isn't under l(a)h.o ownership, which is good because this way I can avoid the bureaucracy and e-mail the maintainers directly) The following is a Data.ByteString comment for the (non-exported) function zipWith' -- -- | (...) Rewrite rules -- are used to automatically covert zipWith into zipWith' when a pack is -- performed on the result of zipWith. -- This implies there should be a rule: {-# RULES "ByteString specialise zipWith'" forall (f :: Word8 -> Word8 -> Word8) p q . zipWith' f p q = pack (zipWith f p q) #-} But no such rule exists in the ByteString source (the inverse rule using 'unpack' does exist). 1) Is this an omission? Can we fix it? It's a rather important rule for crypto-api. 2) Can we export zipWith' so people can be explicit? If not, can we get the comment about the rule placed somewhere so it will make its way to the generated Haddock documentation for general users? 3) Very different issue: Could .Lazy export hGetN or have defaultChunkSize configurable by a CPP/compile time macro? If not, perhaps we could make "chunkOverhead = max 16 (2 * sizeOf (undefined ::Int))" so it will be the same on 64 and 32 bit systems (a 128 bit boundary, nice and fast for most modern cipher algorithms, sadly asking for it to match hash block sizes is a bit much). Cheers, Thomas

2 3

darcs hacking sprint 15-17 Oct, Orleans
by Eric Kow 05 Oct '10

05 Oct '10

Hi everybody, Just a quick reminder that the fifth Darcs hacking sprint is taking place in just ten days (Orleans, France, 15-17 Oct): http://wiki.darcs.net/Sprints/2010-10 It's still possible to attend, but you should contact Florent Becker (CC'ed) as soon as possible to confirm there are still spaces. Hope to see you there! -- Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> For a faster response, try +44 (0)1273 64 2905 or xmpp:kowey@jabber.fr (Jabber or Google Talk only)

1 0

Suggestions for improvement
by N. Raghavendra 05 Oct '10

05 Oct '10

I am reading the book `The Haskell Road to Math, Logic, ...". One of the exercises in the first chapter asks for a function that maps a string "abcd" to "abbcccdddd" and "bang!" to "baannngggg!!!!!". Since such a function f fixes the empty word, and maps wa to f(w)a^(length(w)+1) for any word w and any letter a, I came up with the following solution: -- Map "abcd" to "abbcccdddd" and "bang!" to "baannngggg!!!!!". blowup :: String -> String blowup [] = [] blowup x = blowup (allButLast x) ++ lastToTheLength x -- Map "abcd" to "abc". allButLast :: String -> String allButLast [] = [] allButLast [x] = [] allButLast (x : xs) = x : allButLast xs -- Map "abcd" to d^4 = "dddd". lastToTheLength :: String -> String lastToTheLength [] = [] lastToTheLength [x] = [x] lastToTheLength (_ : xs) = lastToTheLength xs ++ [last xs] One question I have is whether I can eliminate points in the above definition of blowup, and write something like blowup = (++) . (blowup . allButLast, lastToTheLength) thinking of (++) as a function String x String -> String. Also, I can't figure out whether it is possible to get a shorter solution using fold. I have tried Hlint on my file, but it gave no suggestions. I am sure there are better ways, and would like some pointers and any general suggestions for improvement. Thanks and regards, Raghavendra. -- N. Raghavendra <raghu(a)mri.ernet.in> | http://www.retrotexts.net/ Harish-Chandra Research Institute | http://www.mri.ernet.in/ See message headers for contact and OpenPGP information.

6 14

Network.HTTP, BasicAuth+Headers
by Tony Morris 04 Oct '10

04 Oct '10

I'm trying to send a simple request using Network.HTTP and getting a bit lost in all the possibilities. I've experimented with both Network.HTTP and Network.Browser but have been unable to come up with something specific to my needs, which are quite simple. I'd like to send a request equivalent to the following curl: curl --basic -u "user:pass" -H "Accept: application/xml" -H "Content-type: application/xml" "https://host/path" It seems with Network.Browser I cannot send headers, but with Network.HTTP I cannot see how to send the BasicAuth information. Any tips are appreciated. -- Tony Morris http://tmorris.net/

1 0

CCC MyType
by Lafras Uys 04 Oct '10

04 Oct '10

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 List, I'm trying to wrap my head around Haskell and Category Theory---very new to both and in keeping with character, I've jumped into the deep-end. The categories and category-extras packages have helped to illustrate some concepts. However, I remain confused and hope someone can point me in the right direction. The ultimate goal is to construct my own instance of a CCC. Here is the class definition for CCC from the categories package: > class ( Cartesian (<=) > , Symmetric (<=) (Product (<=)) > , Monoidal (<=) (Product (<=)) > ) => CCC (<=) where > type Exp (<=) :: * -> * -> * > apply :: (Product (<=) (Exp (<=) a b) a) <= b > curry :: ((Product (<=) a b) <= c) -> a <= Exp (<=) b c > uncurry :: (a <= (Exp (<=) b c)) -> (Product (<=>) a b <= c) I have the following: > {-# LANGUAGE TypeFamilies, > MultiParamTypeClasses, > UndecidableInstances #-} > import Control.Category.Braided > import Control.Category.Associative > import Control.Category.Cartesian > import Control.Category.Monoidal > import Control.Category.Cartesian.Closed > data MyType a b = MyType {f::(a -> b)} > data SomeType a b = SomeType {g::(a,b)} > instance (Symmetric MyType (Product MyType), > Monoidal MyType (Product MyType), > PreCartesian MyType) => CCC MyType where > type Exp MyType = SomeType It type checks in ghci and gives the appropriate warnings that apply, curry and uncurry have not been defined---I can't see how to define these. Is it actually possible with the two data types I have? I would really appreciate some help. Lafras -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkyaGWYACgkQKUpCd+bV+kqgWgCfY7B7pUttB0xfeOAN1V3NDqRL fgQAniK/EJsV9jS7XWxmxElCVD6AW0as =l0x/ -----END PGP SIGNATURE-----

2 5

Haskell Platform, Hackage and Cabal : The 2nd Year : Status Report
by Don Stewart 04 Oct '10

04 Oct '10

The Haskell Implementors Workshop was held in Baltimore, today. Duncan Coutts from Well-Typed and I presented a status report on the Haskell distribution infrastructure: Hackage , Haskell Platform and Cabal. The slides are here: http://donsbot.wordpress.com/2010/10/01/hackage-cabal-and-the-haskell-platf…

9 10

Template Haskell: hiding declarations
by Jonas Almström Duregård 04 Oct '10

04 Oct '10

Hi Café, I'm doing some code generation with Template Haskell that results in few hundred top level declaration, of which only 10 or so should actually be exposed to the user (the rest are only used by generated code). Since I cant splice stuff into the module header (i.e. into the export list), I cant think of a good way to hide the internal declarations. One way would be to put all the declarations into one or several where-clauses, but this doesn't work very well when several functions share the hidden declarations. Also I might want to hide a few data types. Does anyone have a clever solution? Has anyone had a similar problem (maybe TH needs to be extended with support for hidden declarations)? /J

3 2

Problem with monad transformer stack
by Michael Vanier 04 Oct '10

04 Oct '10

I'm having a problem with a simple monad transformer stack that has me stumped. Here's the sample code: {-# LANGUAGE GeneralizedNewtypeDeriving #-} import Control.Monad.Error import Control.Monad.State import Data.Typeable data SomeError = Error1 | Error2 | ErrorFail deriving (Eq, Show, Typeable) data MyData a = MyData [a] instance Error SomeError where noMsg = ErrorFail {- This works: -} {- newtype StateError e s a = StateError ((StateT s (Either e) a)) deriving (Monad, MonadState s, MonadError e, Typeable) type MyMonad a = StateError SomeError (MyData a) a -} {- This doesn't work: -} newtype MyMonad a = MyMonad ((StateT (MyData a) (Either SomeError) a)) deriving (Monad, MonadState (MyData a), MonadError SomeError, Typeable) ---------- Basically, the more abstracted (commented-out) version works, but the more specific one gives this error: Weird.hs:33:12: Can't make a derived instance of `Monad MyMonad' (even with cunning newtype deriving): cannot eta-reduce the representation type enough In the newtype declaration for `MyMonad' Weird.hs:34:12: Cannot eta-reduce to an instance of form instance (...) => MonadState (MyData a) MyMonad In the newtype declaration for `MyMonad' Weird.hs:35:12: Can't make a derived instance of `MonadError SomeError MyMonad' (even with cunning newtype deriving): cannot eta-reduce the representation type enough In the newtype declaration for `MyMonad' These error messages mean nothing to me. What's going on? Can the more specific code be made to work? This is with ghc 6.12.3. Thanks, Mike

3 4

A parsec question
by Ben Franksen 03 Oct '10

03 Oct '10

I have a question about Parsec. The following program > import Control.Applicative ((*>),(<*)) > import Text.Parsec > import Text.Parsec.Char > block p = char '{' *> p <* char '}' > parser = block (many digit) > main = parseTest parser "{123a}" gives the output parse error at (line 1, column 5): unexpected "a" expecting "}" Note the last line mentions only '}'. I would rather like to see expecting "}" or digit since the parser could very well accept another digit here. (1) What is the reason for this behaviour? (2) Is there another combinator that behaves as I would like? (3) Otherwise, how do I write one myself? BTW, I am using parsec-3.1.0 and ghc-6.12.3. Cheers Ben

5 10