November 2007 - Haskell-Cafe

Is there a module for multivariate linear regression?
by Lihn, Steve 06 Nov '07

06 Nov '07

I am looking for a Haskell module that will do multivariate linear regression. Does someone know which module will do it? That is, the equivalent of Perl's Statistics::Regression.pm. http://search.cpan.org/~itub/PerlMol-0.35_00.ppm/lib/Statistics/Regressi on.pm Thanks, Steve ------------------------------------------------------------------------------ Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. ------------------------------------------------------------------------------

3 5

speeding up fibonacci with memoizing
by Thomas Hartman 06 Nov '07

06 Nov '07

I just thought this was interesting, so I would share it. Thanks to whoever it was on #haskell who helped me, sorry I can't remember who. -- horribly slow. try slow_fibs 30, not too much higher than that and it hangs slow_fibs = map slow_fib [1..] slow_fib 1 = 1 slow_fib 2 = 1 slow_fib n = ( slow_fib (n-2) ) + (slow_fib(n-1)) -- versus, try memoized_fibs !! 10000 memoized_fibs = map memoized_fib [1..] memoized_fib = ((map fib' [0 ..]) !!) where fib' 0 = 0 fib' 1 = 1 fib' n = memoized_fib (n - 1) + memoized_fib (n - 2)

14 16

Strange Type Inference
by C.M.Brown 06 Nov '07

06 Nov '07

Hi, I was given a quandary this evening, suppose I have the following code: module Test1 where import qualified Data.Map as Map testFunction :: Ord a => Map.Map a b -> Map.Map a b -> a -> (Maybe b, Maybe b) testFunction m0 m1 k = (lookup0 k, lookup1 k) where lookup0 x = Map.lookup x m0 lookup1 x = Map.lookup x m1 This compiles and type checks fine. However, the only way I could add type signatures to lookup0 and lookup1 was to do something along the lines of this: testFunction :: Ord a => Map.Map a b -> Map.Map a b -> a -> (Maybe b, Maybe b) testFunction m0 m1 k = (lookup0 k m0, lookup1 k m1) where lookup0 :: (Monad m, Ord a) => a -> Map.Map a b -> m b lookup0 x m0 = Map.lookup x m0 lookup1 :: (Monad m, Ord a) => a -> Map.Map a b -> m b lookup1 x m1 = Map.lookup x m1 Is there a way to give lookup0 and lookup1 explicit type signatures without passing in m0 and m1 as parameters? (So their definitions are the same as in the first example) If ghc can infer the type, surely it must be possible? Regards, Chris.

4 5

Web Application Sessions
by Lars Oppermann 06 Nov '07

06 Nov '07

Hello, I'm currently looking into web application programming with Haskell. Though I am aware of HAppS, a lighttpd/fastcgi based approach is what I'm currently considering. It seems rather straightforward to implement a session manager, that would maintian some server side state which is passed to the function handling the request along with the request information and that gets updated as a result of the request processing. Just like in Java servlets (maybe call it funclets ;). In fact, this seems so straight forward to me, that I was convinced that somebody would already have implemented a library for that, that takes care of associating requests with session-ID cookies - however, I haven't found anything. Does anyone know whether something like this has been done? Maybe I can't find anything because my state-infested notion of session-data is totally inappropriate in a functional programming environment. In that case I would like to know about that (and how to better myself)... Cheers, Lars

1 0

package maintainers: updating your packages to work with GHC 6.8.1
by Duncan Coutts 05 Nov '07

05 Nov '07

If you maintain a Haskell package this is for you. So now that GHC 6.8.1 is out you'll want to test your package with it. We'd especially like maintainers of packages that are distributed on hackage.haskell.org to test their packages and update them as necessary. However we would appreciate it if your packages will still work with GHC 6.6.x, so read below for details on how to do that. You'll most likely find that your package fails to build with GHC 6.8.1 with an error message like this: Could not find module `Data.Map': it is a member of package containers-0.1.0.0, which is hidden. This is the first symptom of the change in the base library in GHC 6.8.1. Several parts of the old base library got split out into separate packages. For example Data.Map is now in the containers package. Unfortunately this means most packages need updating. Of course you'd also like to make your packages continue to work with GHC 6.6.x (and perhaps older). If you go and just add containers to your package's build-depends then it will no longer work with GHC 6.6.x or older. So the solution at the moment is to use Cabal configurations like so: flag splitBase description: Choose the new smaller, split-up base package. library if flag(splitBase) build-depends: base >= 3, containers else build-depends: base < 3 Since this uses a new Cabal feature you'll have to make your packages require Cabal-1.2 or later: cabal-version: >=1.2 This is ok since Cabal-1.2.x comes with ghc-6.8 and can also be installed from hackage for older ghc versions. For a bigger example see Cabal's own .cabal file: http://hackage.haskell.org/packages/archive/Cabal/1.2.2.0/Cabal.cabal The new Cabal syntax is explained in the Cabal user guide: http://haskell.org/cabal/release/rc/doc/users-guide/x30.html#configurations This is all explained in more detail on the wiki: http://haskell.org/haskellwiki/Upgrading_packages And the details of other api changes in the libraries are in the GHC 6.8.1 release notes: http://haskell.org/ghc/docs/6.8.1/html/users_guide/release-6-8-1.html A list of packages on hackage that currently fail to build with GHC 6.8.1 is here: http://www.haskell.org/pipermail/libraries/2007-November/008529.html We realise that this change is rather disruptive, requiring changes in each package. We've started a wiki page with ideas on how to avoid this in future when the base package is shrunk again. http://hackage.haskell.org/trac/ghc/wiki/PackageCompatibility Duncan (Cabal release manager)

6 10

Compiling GHC on Scientific Linux V5 [http://ftp.scientificlinux.org/linux/scientific/50/iso/i386/SL-5.0-050407-i386-DVD.iso, Build on RHEL5]
by Animesh Sharma 05 Nov '07

05 Nov '07

For a long time I wanted to go 'strictly functional' and what better way then to start with the beast, ghc. So about a couple of months back, I downloaded version 6.6.1 [ http://www.haskell.org/ghc/dist/6.6.1/rpm/rhel5/i386/ghc661-6.6.1-2.el5.i38…] and started reading YAHT [ http://darcs.haskell.org/yaht/yaht.pdf ]. Then the 'laziness' took over and my 'idea of being strictly functional' took a back-seat. The news of release of ghc Version 6.8.1 (3 November 2007) reignited the desire. This time I thought of compiling the ghc from source [ Main - http://www.haskell.org/ghc/dist/6.8.1/ghc-6.8.1-src.tar.bz2 , Extra Libs - http://www.haskell.org/ghc/dist/6.8.1/ghc-6.8.1-src-extralibs.tar.bz2 ]. Make sure you have 4-5 hours and 1-1.5 GB of disc space! Become 'root' and extract the above two files: tar xvjf ghc-6.8.1-src.tar.bz2 tar xvjf ghc-6.8.1-src-extralibs.tar.bz2 Both of them will extract in the directory ./ghc-6.8.1. For compiling there are some dependencies to be taken care of: 1. ghc [I already had v 6.6.1, http://www.haskell.org/ghc/dist/6.6.1/rpm/rhel5/i386/ghc661-6.6.1-2.el5.i38…] , it is interesting as the compilers compiles its next version [ http://en.wikipedia.org/wiki/Bootstrapping_(compilers) ]: rpm -ivhf ghc661-6.6.1-2.el5.i386.rpm 2. cabal, Common Architecture for Building Applications and Libraries, which can be downloaded from http://www.haskell.org/cabal/release/rc/cabal-1.2.2.0.tar.gz and installation instructions need to be followed [ http://haskell.org/haskellwiki/Cabal/How_to_install_a_Cabal_package ]: tar xzf cabal-1.2.2.0.tar.gz cd cabal-1.2.2 runhaskell Setup.hs configure runhaskell Setup.hs build runhaskell Setup.hs install 3. readline-devel, readline library which in turn has terminfo as dependence, so better to use yum for installation: yum install readline-devel 4. alex, A lexical analyser generator for Haskell, whose rpm can be downloaded from http://www.haskell.org/alex/dist/2.1.0/RH9/alex-2.1.0-1.i386.rpm and installed easily: rpm -ivhf alex-2.1.0-1.i386.rpm 5. happy, the yacc of haskell [I read on http://tritium.livejournal.com/ to compile happy you need old version of happy]. But for me simply getting the source http://www.haskell.org/happy/dist/1.17/happy-1.17.tar.gz and installation worked fine: tar xvzf happy-1.17-html.tar.gz cd happy-1.17 runhaskell Setup.hs configure runhaskell Setup.hs build runhaskell Setup.hs install We are all set for the ghc compilation: cd ghc-6.8.1 ./configure make make install Clean up ~300MB from the older ghc (v 6.6.1) installation, I went the brute force way: find /usr -name ghc-6.6.1 > <list> rm -rf <list> Now I have to dig deeper into the release note (http://www.haskell.org/ghc/docs/6.8.1/html/users_guide/release-6-8-1.html ), one thing which has really got me interested is the introduction of debugger to GHCi. Any tutorials on how to go about it? Regards, Ani ---------The Answer lies in the Genome------------ fuzzylife.org

2 1

The question of ByteString
by Andrew Coppin 05 Nov '07

05 Nov '07

Somewhat related to the discussions about Haskell's performance... String. ByteString. Do we really need both? Can one replace the other? Why is one faster? Can't we make *all* lists this fast? [insert further variations here] Thoughts?

7 11

Efficient and type safe flag sets
by Henning Thielemann 04 Nov '07

04 Nov '07

We have http://www.haskell.org/haskellwiki/EnumSet_EnumMap Is there also an efficient implementation for bit sets that fit into a machine word? This would be useful for foreign function interfaces. E.g. where C defines #define SND_SEQ_PORT_CAP_READ (1<<0) /**< readable from this port */ #define SND_SEQ_PORT_CAP_WRITE (1<<1) /**< writable to this port */ #define SND_SEQ_PORT_CAP_SYNC_READ (1<<2) /**< allow read subscriptions */ #define SND_SEQ_PORT_CAP_SYNC_WRITE (1<<3) /**< allow write subscriptions */ In Haskell we could define module SndSeq data PortCap = Read | Write | SyncRead | SyncWrite bit32 :: Enum a => a -> FlagSet32 a flags32 :: Enum a => [a] -> FlagSet32 a In C we have to write foo (SND_SEQ_PORT_CAP_SYNC_READ | SND_SEQ_PORT_CAP_SYNC_READ) which is efficient but unsafe, because flags of the wrong sort or plain numbers could be passed accidentally. In Haskell we would write foo (flags32 [Read, SyncRead]) which is both efficient and safe, because foo :: FlagSet32 PortCap -> IO () (Given that constant expression evaluation works ...)

2 1

Adding GLUT package to WinHugs
by Peter Verswyvelen 04 Nov '07

04 Nov '07

The precompiled WinHugs Sep 2006 does not seem to come with the GLUT package. Although I'm able to add packages to GHC (using Cabal), I failed doing the same for WinHugs. How can I add the GLUT package to WinHugs? I tried starting MSYS, typed configure, which worked fine, but then "make" fails. Also typing runhugs Setup.hs configure fails with runhugs: Error occurred ERROR "c:\program files\winhugs\packages\base\Text\ParserCombinators\ReadP.hs":156 - Syntax error in type expression (unexpected `.') Thanks for any help, Peter Verswyvelen

2 4

space leak?
by Massimiliano Gubinelli 04 Nov '07

04 Nov '07

> ( these two lines are just to fool the gmane post algorithm which > complains for top-posting....) Hi, i'm learning Haskell and trying to use the HPDF 1.2 library I've come across some large memory consumption for which I do not understand the origin. I've tried heap profiling but without much success. This is my code > module Main where > import Control.Monad.State > import Graphics.PDF > data Opcodes = Rect | Ship deriving (Show) > doPage (Rect:ops) = do > stroke $! Rectangle 10.0 10.0 10.0 10.0 > doPage ops > doPage l = return l > doOps [] = return () > doOps (Ship:ops) = {-# SCC "OPSHIP" #-} do > p <- addPage Nothing > ops' <- drawWithPage p $! do > strokeColor red > applyMatrix $ (translate 72.0 72.0) > doPage ops > doOps ops' > doOps (op:_) = error ("unexpected " ++ show op) > testpdf = do > let ops = concat $ replicate 100 (Ship : (replicate 1000 Rect )) > pageRect = PDFRect 0 0 > (floor $ 21.0/2.54*72.0) (floor $ 29.7/2.54*72.0) > runPdf "test1.pdf" (standardDocInfo { author=toPDFString "mgubi", > compressed = False}) > pageRect $ doOps ops > testpdf' = do > let pageRect = PDFRect 0 0 > (floor $ 21.0/2.54*72.0) (floor $ 29.7/2.54*72.0) > runPdf "full.pdf" (standardDocInfo { author=toPDFString "mgubi", > compressed = False}) > pageRect $ sequence_ $ foldM f [] $ replicate 100 $ > (\p -> sequence_ $ replicate 1000 $ > drawWithPage p $ stroke $ > Rectangle 0.0 0.0 10.0 10.0) > where f ps acts = do > p <- addPage Nothing > acts p > return $ p:ps > main = testpdf now, if I run testpdf' then memory profile is very low and everything is as expected while if I run testpdf then the profile grows up to 80MB and more. This is the stripped down version of the original program (which is a DVI interpreter) so there I will have also some StateT and more complicated opcodes. I would like to know what is wrong with the above code. Could someone help me? thanks, Massimiliano Gubinelli

2 2