March 2011 - Beginners

How to structure your Haskell installation?
by frode k 23 Mar '11

23 Mar '11

Hi. I'm installing the Haskell Platform om my Linux box and need some tips on how to structure the files on the OS. I need to install and use both GHC 6.12.3 and GHC 7.0.2. Hence I want to separate these installations. I've setup two directories, one for each platform (related to the GHC): # ls -d -1 $PWD/*.* /usr/haskell/2010.2.0.0 /usr/haskell/2011.2.0.0 I've installed GHC-6.12.3 in "2010.2.0.0" and will install GHC-7.0.2 in "2011.2.0.0". My question is more related to where I should install the packages. When I install a package using "runhaskell Setup ..." I assumed this to ONLY be installed for the version of GHC I'm using when installing it. Am I right about this? Having $PATH pointing to GHC-6.12.3 I've done this when installing packages, to add them to "2010.2.0.0": - - - - - - - cd /usr/src wget http://hackage.haskell.org/packages/archive/transformers/0.2.2.0/transforme… tar xzf transformers-0.2.2.0.tar.gz cd transformers-0.2.2.0 runhaskell Setup configure --prefix=/usr/haskell/2010.2.0.0/ runhaskell Setup build runhaskell Setup install - - - - - - - Later when I switch to GHC-7.0.2 I plan to install the packages into the "2011.2.0.0" directory in stead. Is this a good way to structure several Haskell installations on the same system? Is there a "best practice" on how to structure the files? I want to get this right from the start, to avoid having to go back and re-arrange everything. Regards, Frode K [k]

2 2

Re: [Haskell-beginners] Recursion in monad
by Adrian May 23 Mar '11

23 Mar '11

OK I did this: import System.Random walk :: Int -> IO Int walk i = randomRIO (0,1) >>= \r -> return (i+r*2-1) say :: Int -> IO () say i = putStrLn $ show i rep :: Int -> a -> (a -> IO a) -> (a -> IO ()) -> IO () rep n i w s | n<=0 = return () | otherwise = s i >> w i >>= \ii -> rep (n-1) ii w s main :: IO () main = rep 10 50 walk say Is that the easiest way? Adrian.

2 2

small expression evaluator
by Petr Novotnik 23 Mar '11

23 Mar '11

Hello, on my journey through Haskell I've hit a problem where I don't know any further and would greatly appreciate your help. I've got this code so far: data Value a e = VConst a | VFunc (e -> a) evalV :: Value a e -> e -> a evalV (VConst x) = const x evalV (VFunc f) = f liftV :: (a -> a -> t) -> (Value a e) -> (Value a e) -> e -> t liftV f x y e = (evalV x e) `f` (evalV y e) (.==.), (./=.) :: (Eq a) => (Value a e) -> (Value a e) -> e -> Bool (.==.) = liftV (==) (./=.) = liftV (/=) (.&&.), (.||.) .... some more operators This allows me to write client code like this: data Person = Person { personName :: String , personAge :: Int } deriving (Show) exampleExpr :: Bool exampleExpr = (VConst 99) .==. (VFunc personAge) $ Person "pete" 99 I was wondering, whether it'd be possible to enable defining expression without the Value data constructors, i.e. 99 .==. personAge $ Person "pete" 99 I've tried using a type class to have a function for lifting different types into Values. class ToValue a e | a -> e where valueLift :: a -> Value a e then I tried defining an instance for functions: instance ToValue ((->) Person a) Person where valueLift f = undefined in ghci: > :t valueLift personName valueLift personName :: Value (Person -> String) Person and suddenly I realized this will not work because I would need the type `Value String Person'. Now, I'm stuck and don't know which path to take in order to come closer to being able writing "99 .==. personAge" or "personAge .==. 99". Maybe I'm approaching it in a completely wrong way or it's simply not possible. Do you have any ideas, hints? Many thanks in advance for any kind of feedback, pete.

2 3

"cabal install wxcore --enable-shared" Error
by Paulo Pocinho 20 Mar '11

20 Mar '11

I'm trying to build an application using "ghc -dynamic mywxapp.hs". Everything not involving wx compiles with dll linking. However, I can't build the required dynamic libraries for wxcore. Using "cabal install wxcore --enable-shared" gives the following: collect2: ld returned 1 exit status Creating library file: dist\build\libHSwxcore-0.12.1.6-ghc7.0.2.dll.a cabal: Error: some packages failed to install: wxcore-0.12.1.6 failed during the building phase. The exception was: ExitFailure 1 Using "cabal install wxcore --enable-shared --enable-split-objs" gives the following: E:\Tools\Haskell Platform\2011.2.0.0\mingw\bin\ld.exe: BFD (GNU Binutils) 2.20.51.20100613 internal error, aborting at ../../binutils-2.20.51/bfd/coffcode.h line 954 in handle_COMDAT E:\Tools\Haskell Platform\2011.2.0.0\mingw\bin\ld.exe: Please report this bug. cabal: Error: some packages failed to install: wxcore-0.12.1.6 failed during the building phase. The exception was: ExitFailure 1 Is it possible to fix?

1 0

Re: [Haskell-beginners] Lazy vs Strict ponderings...
by sean＠objitsu.com 20 Mar '11

20 Mar '11

>> 2) Don't worry about memory leaks until they actually appear. If they >> do, use a profiling tool to find out what's going on. Most likely, one >> of the two things above happened in an interesting way. At the risk of sounding stupid in Haskell mode, how would I be aware of this other than obvious messages or seg-faults about memory getting low ? > > It is likely that, if you develop a medium to large sized application for > a customer, memory leaks will appear at the customers site. It will often > be very difficult to reproduce the situation and find the leak. That scares me enough to not want to use Haskell for serious application development. With 26 years in the trade I have had too many times when a fault cannot be reproduced and despite an 'obvious fault in the source' being fixed and the problem never coming back, without being able to reproduce and thus confirm that you have eliminated the problem, you can't ever relax on a Saturday night! LOL I hate the word 'random' when clients describe problems too! > If you > want to produce quality software, it is best to be sure what you are doing > during the whole development process. There should be a set of guidelines, > on how to prevent space leaks. > I guess "being sure..." comes with experience of the language, like any other. I can 'think' in quite a few languages with complete confidence, knowing that things will just work. I guess Haskell is a bigger elephant sandwich! There is something beautiful about its syntax and the fact that everything is (or can be) so precise and concise, sometimes a little too concise for me to read. It *makes* me want to be in that club just so I can have that level of understaning about my job. I learned BASIC and UCSD-PASCAL at age 11 and I have never stopped being fascinated about *what* computing is all about. Sometimes I think that a CPU *is* a perfectly enlightened consciousness always in the *now*. Time for my medication and music therapy..... > See also "On the reliability of programs", > http://www.cs.utexas.edu/users/EWD/transcriptions/EWD03xx/EWD303.html Will do, thanks. Sean

4 6

SQLite3 Row <-> Data models -- Is this a typical Haskell pattern ?
by Stef T 19 Mar '11

19 Mar '11

Hey Everyone, Greetings, I am a total newb to haskell, so, I have a strange question (perhaps). Consider I have a data model which looks like this; data Brand = Brand { id :: Int, name :: String, created_at :: String, updated_at :: String } Now, I have a sqlite3 function such as checkout :: Int -> IO (Either String [[Row Value]]) checkout a = do handle <- openConnection "/Users/stef/haskell/db/development.sqlite3" execStatement handle $ "SELECT * from brands where id = " ++ show a From this, I get a Row/Tuple. So far, so good, but, the question _I_ have is, how do you go about mapping a returned row to the data model ? Is this even a design pattern that is used in FP languages ? I admit, I have spent numerous years in the MVC world, so, perhaps this is simply "not done". It would seem to be a much nicer thing to then do ; name myBrandModel Thanks for reading this far, feel free to complain about my design :D Regards S.

2 3

emacs mode?
by Rob Nikander 19 Mar '11

19 Mar '11

Hi all, Where can I get emacs mode for Haskell? I tried the following page but the links from it are hanging forever or dead. http://www.haskell.org/haskellwiki/Haskell_mode_for_Emacs Rob

2 1

Lazy vs Strict ponderings...
by Sean Charles 19 Mar '11

19 Mar '11

I *think* I understand about lazy evaluation and its effects on I/O and that it can internally create thunks, promises, continuations etc. whatever you want to call them, and then at some point there will be a sudden spike in memory and CPU loads as something triggers a whole chain of promises to be fulfilled but what I don't have a feel for is just how serious a problem that might be today in say, a simple desktop application. For example, let's just say I write a simple application in 'C' and a comparable one in Haskell both running on the same hardware, but not concurrently! Would the Haskell one ever find itself running out of memory given the same input sequence as the 'C' program because of the way it works ? I know that the 'C' one could also run out of memory but I hope that the sentiment of my question is clear; what are the ramifications, gotchas and other considerations that using a purely functional lazy language like Haskell you need to be aware of when coming from an imperative / scripted background ? Notwithstanding the fact that there are strict flavours of functions that can be used.... but only if you are truly aware of the reasons for their existence in the first place, is this a real design issue to consider when coding an application in Haskell or is it only for certain 'groups' of applications. I guess the nature of the application would be the governing factor. Is it something one needs to worry about at all or should one just code away and write the application and then worry!? I think that having a clearer understanding of what 'types' of problems and their implementations have on CPU/RAM would be a good one to have! :) Thanks Sean Charles.

3 2

Re: [Haskell-beginners] Unit conversion
by Tom Murphy 19 Mar '11

19 Mar '11

On Fri, Mar 18, 2011 at 5:02 AM, Magnus Therning <magnus(a)therning.org> wrote: > On Thu, Mar 17, 2011 at 21:17, Tom Murphy <amindfv(a)gmail.com> wrote: > > Hi All, > > Is there a good way to easily convert between units? > > For example, let's say I have a data type: > > data Volumes = Teaspoon | Tablespoon | Slice | FluidOunces > > ... and I want to define an infix function '<+>', which adds > together > > amounts of food: > > (1, Slice, cake) <+> (1, Volume Tablespoon, sugar) <+> (32, Volume > > FluidOunces, almondMilk) > > Which would return: > > (3200, Teaspoons) > > What is the most efficient way to define equivalency/conversion > between > > these measures? > > I remember an interesting method for celsius-farenheit conversion in > > Higher-Order Perl, using function composition, but that was between only > 2 > > units. I'd like something where I don't have to provide n^2 definitions. > > I wrote two blog posts on something that sounds related: > http://therning.org/magnus/archives/354 > http://therning.org/magnus/archives/355 > > Maybe they could help. > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus(a)therning.org jabber: magnus(a)therning.org > twitter: magthe http://therning.org/magnus So given data Measure = M Volume Float deriving (Show, Eq) data Volume = Slice | FluidOunce | Tablespoon | Teaspoon deriving (Show, Eq) Method one and two would be to convert all units to the smallest unit: toTsp :: Measure -> Measure [...] toTsp (M Slice a) = M Teaspoon (a * 350) toTsp (M FluidOunce a) = M Teaspoon (a * 34) [...] Perform the addition, then convert it to a more appropriate measure (back to Cake Slices...). It seems that there's a potential loss of precision, if I have to for example convert Megaliter measures to Teaspoons. Method three involves using typeclasses to define common "meeting points" for the units. So, we say that if we're adding Slices and FluidOunces, both measures should be converted to FluidOunces before the addition. The problem there is that, given n measurement units, we have to declare n^2 typeclass instances by hand. Another way that I can see to do this is to define a "chain" of conversions. I'd then "convert down" only to the smallest-common unit. So I could define (warning: extreme kludge ahead) a series of "step-down"s and "step-up"s: Given a largest-to-smallest ordering of Slice, FluidOunce, Tablespoon, Teaspoon, I can define: stepDown :: Measure -> Measure stepDown (M Slice a) = M FluidOunce (a * 120) stepDown (M FluidOunce a) = M Tablespoon (a * 2) stepUp (M FluidOunce a) = M Slice (a / 120) stepUp (M Tablespoon a) = M FluidOunce (a / 2) [...] and then a function to "step" as far down as needed: convertDown :: Measure -> Volume -> Measure convertDown (M unit measure) goalUnit | unit == goalUnit = M unit measure | otherwise = fun5 (stepDown (M unit measure)) goalUnit And then if I wanted to add a Slice of cake to a Tablespoon of almond milk, I'd have to find the smaller unit, convert down, perform the math and then do a "convert up" operation. An added benefit to this method is that defining a unit smaller than the current smallest wouldn't involve revising the "base unit" measures. This seems like a pretty common C.S. problem; I'm surprised I haven't been able to find more resources about it. Thanks for your time! Tom P.S. Thank you, Magnus, for those resources!

2 2

haskell install on low memory server?
by Rob Nikander 19 Mar '11

19 Mar '11

Hi, I'd like to install Haskell on a linux virtual server with about 600 MB of RAM. (Maybe that is doomed to failure?) The goal is to try "happstack" for web apps, behind nginx. Compiling the whole platform is asking for OpenGL... so I thought I would get cabal and install pieces that I want. I downloaded the 'generic ghc binary' from here: http://hackage.haskell.org/platform/linux.html and that seems to work. But when I download the cabal-install tarball and try to run it's 'bootstrap.sh' I run out of memory. Is there a way to tell it to use less memory? ------- shell commands ------- $ cd cabal-install-0.10.2 $ sh bootstrap.sh Checking installed packages for ghc-7.0.2... Cabal is already installed and the version is ok. transformers-0.2.2.0 will be downloaded and installed. mtl-2.0.1.0 will be downloaded and installed. parsec-3.1.1 will be downloaded and installed. network-2.3.0.2 will be downloaded and installed. time is already installed and the version is ok. HTTP-4000.1.1 will be downloaded and installed. zlib-0.5.3.1 will be downloaded and installed. Downloading transformers-0.2.2.0... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 17148 100 17148 0 0 27246 0 --:--:-- --:--:-- --:--:-- 39420 [1 of 1] Compiling Main ( Setup.hs, Setup.o ) Linking Setup ... collect2: ld terminated with signal 9 [Killed] Error during cabal-install bootstrap: Compiling the Setup script failed ------- end of shell commands ------- It seems strange that `ld` would need to use ~450 MB, which is what it's doing to get killed by the kernel. Any ideas? thanks, Rob

3 5