- Haskell-Cafe - Haskell.org

Mission: To take args from an n-tuple ... generally
by Martin Hofmann 31 Jan '08

31 Jan '08

Dear Community. I have recently read Joel Koerwer's posting how to evaluate a function of type (a->a->...->a->a), taking the arguments from a list (http://haskell.org/pipermail/haskell-cafe/2006-October/018658.html) Therefore, he introduced a function multApply: multApply :: (a->a->...->a->a) -> [a] -> a I wondered, why not take an n-tuple of arguments s.t. multApply' :: (a1->a2->...->an->o) -> (a1,(a2,(...(an,o)...))) -> o I naively tried to modify Joel's code, but in vain. As far as I can deduce from the error messages, GHC fails to fix the accordant types. However, in my understanding the type solely depends on the type of the passed function. Or am I wrong? Maybe there is a reason nobody has tried it this way? Following the code. First Joel's original multApply and then my attempt to define multApply'. ------------------------ MultApply.hs ------------------------------ {-# OPTIONS -fglasgow-exts -fallow-overlapping-instances #-} module MultApply where class MultApply func arg | func -> arg where multApply :: func -> [arg] -> arg instance MultApply (a->a) a where multApply f (x:xs) = f x multApply f _ = error "MultApply: one too few args" instance MultApply cascade a => MultApply (a -> cascade) a where multApply f (x:xs) = multApply (f x) xs multApply f _ = error "MultApply: n too few args" -- some random examples oneArg = multApply sqrt [25..] twoArg = multApply (+) [1..] fiveArg = multApply (\a b c d e -> sqrt ((a+b)^2+(d-e)^2)-5*c) [13..] class MultApply' f arg out | f -> arg out where multApply' :: f -> arg -> out instance MultApply' (a->b) a b where multApply' f x = f x multApply' f _ = error "MultApply: one too few args" instance MultApply' cascade a2 b => MultApply' (a1 -> cascade) (a1,a2) b where multApply' f (x,xs) = multApply' (f x) xs multApply' f _ = error "MultApply: n too few args" -------------------End File ------------------------ Hope I didn't stumble in a stupid newbie question. Thanks a lot to everybody who could give me some explanation and hints. Cheers, Martin

2 1

Re: [Haskell] Why functional programming matters
by Don Stewart 31 Jan '08

31 Jan '08

stephan.friedrichs: > Simon Peyton-Jones wrote: > >[...] > > > >2. War stories from real life. eg "In company X in 2004 they rewrote > >their application in Haskell/Caml with result Y". Again, for my purpose > >I can't tell very long stories; but your message can give a bit more > >detail than one might actually give in a presentation. The more concrete > >and specific, the better. E.g. what, exactly, about using a functional > >language made it a win for you? > > > > We [1] implemented an ad-hoc chat system in Haskell in the SEP [2] at > the TU-Braunschweig. Wow! Looks like lots of great code in there, http://sep07.mroot.net/documentation/barracuda/ Text.ParserCombinators.Parsec.XML Codec.Encryption.PKCS1 Network.GnuTLS Data.Heap Will we see this packaged for hackage.haskell.org soon? Some of those modules look very reusable. -- Don

3 2

trying to use unix's openFD ..
by Galchin Vasili 31 Jan '08

31 Jan '08

Hello, The signature for openFD is openFd: FilePath -> OpenMode -> Maybe FileMode -> OpenFileFlags -> IO Fd I am currently reading http://www.haskell.org/ghc/docs/latest/html/libraries/unix/System-Posix-Fil… and diving down into links but can't find the definition of data type FileMode. ??? Basically I want to open in O_RDONLY filemode. Kind regards, Vasili

2 5

Improving 'project'
by Justin Bailey 30 Jan '08

30 Jan '08

All, project doesn't easily let me add a few columns to an existing query (or take a few columns away). Instead, each use of project requires me to build the entire list of columns I'd like to pass on by hand. Before I go further, if there is a way to do that, please let me know. An example of what I'd like to do is below. Imagine I want to add a "selected" boolean column to my query. select_customers = do cust <- table customers project ( (selectedField <<- False) # cust) So I get all the columns in my customers table, plus the "selected" column. Unfortunately the code above doesn't work. cust has type "Rel r", and project wants a "Record r", so I originally tried something like: select_customers = do let toRec :: Rel r -> Record r toRec _ = undefined ... project (unRel cust) But project needs to recover the "Expr" values associated with 'r' and this fails. I then came up with this solution, which lets me add columns at the head of a query: class RecoverExpr a b where toExpr :: a -> b instance RecoverExpr RecNil RecNil where toExpr rec = RecNil instance (FieldTag f, RecoverExpr r r) => RecoverExpr (RecCons f (Expr b) r) (RecCons f (Expr b) r) where toExpr ~rec@(RecCons _ r) = RecCons (attribute . fieldName $ recField $ rec) (toExpr r) recField :: RecCons f a b -> f recField _ = undefined rest :: (RecoverExpr r r) => Rel r -> r rest qry = let unRel :: Rel r -> r unRel _ = undefined in toExpr . unRel $ qry Then I can write: select_customers = do cust <- table customers_tbl ... project (selectField <<- False # (rest cust)) And it actually works. However, it seems over-engineered. My questions are: 1) Is there a way to make project behave like I want without these tricks? 2) If not 1, would it make more sense for the Query monad to pack around the original Expr values? 3) If not 2, how can the solution above be better? I'd like to extend it to adding/removing arbitrary columns and more control over column order. Thanks for any and all thoughts! Justin

1 1

Re: [Haskell-cafe] Simple network client
by Gary Bickford 30 Jan '08

30 Jan '08

One rather funky but effective solution might be to use the tftp protocol. No security, but simple, flexible and efficient. I think there are C libraries that implement it. This would take care of handshaking binary data. I have no idea if anyone has ever used it in Haskell. GB On Wed, 2008-01-30 at 12:50 -0500, haskell-cafe-request(a)haskell.org wrote: > robably not all of them, but some of them, definitely. > > If you want to transmit an arbitrary bytestring then I'm pretty sure > that transmitting a length word first is the way to go. An arbitrary > bytestring can have any value in it, so there are no values left to > act > as delimiters :) You'd have to have some kind of escaping mechanism, > like show, which is expensive. -- "Isn't it funny how the Global Village includes everybody but the villagers?" (http://wiki.laptop.org/go/OLPC_Publications)

1 0

ffi and pass by value structs
by Federico Squartini 30 Jan '08

30 Jan '08

Is there a way in the foreign function interface to pass to a function a struct by value? I explain it better. There are two ways in c to pass structs, as in the following example: /* begin c code */ struct couple{ int a; int b; }; //by value int print_couple (struct couple cpl) { printf("%d\n", cpl.a); printf("%d\n", cpl.b); return(0); } //by address int print_couple_ptr (struct couple *cpl) { printf("%d\n", cpl->a); printf("%d\n", cpl->b); return(0); } /* end c code */ Apparently in the ffi it is possible to import the second version, as: foreign import ccall "print_couple_ptr" printCouple::Ptr Couple->IO Int But not the first, as a struct it's not an atomic type. So how does one solve this issue? Cheers, Federico

2 1

NDP
by Stephan Friedrichs 30 Jan '08

30 Jan '08

Hi all, is someone familiar with compiling ndp (nested data parallel Haskell), "Speed with less convenience"-version? I followed the guide at http://www.haskell.org/haskellwiki/Data_Parallel_Haskell/PackageNDP but executing "make" in the examples directory issues the following error: (.text+0x8d5): undefined reference to `__stginit_parallelzm1zi0zi0zi0_ControlziParallelziStrategies_' manually compiling the dotp example on the link mentioned above doesn't work either: [...] (.data+0x5c): undefined reference to `ndp_DataziArrayziParallelziUnliftedziFlatziUArr_zdf14_closure' DotP.o: In function `sz3_srt': (.data+0x88): undefined reference to `ndp_DataziArrayziParallelziUnliftedziFlatziBasics_toU_closure' DotP.o: In function `sz3_srt': (.data+0x8c): undefined reference to `ndp_DataziArrayziParallelziUnliftedziFlatziUArr_zdf14_closure' /home/sfriedrichs/ghc/rts/libHSrts.a(Main.o): In function `real_main': Main.c:(.text+0x12): undefined reference to `__stginit_ZCMain' Main.c:(.text+0x2c): undefined reference to `ZCMain_main_closure' Does someone know, what's wrong? I'm using ghc-6.9.20080124 as source base (which compiled and works just fine). //Stephan -- Früher hieß es ja: Ich denke, also bin ich. Heute weiß man: Es geht auch so. - Dieter Nuhr

4 7

Stacking monads - beginner design question
by Adam Smyczek 30 Jan '08

30 Jan '08

Hi, My application has to manage a data set. I assume the state monad is designed for this. The state changes in functions that: a. perform IO actions and b. return execution status and execution trace (right now I'm using WriteT for this). Is the best solution: 1. to build a monad stack (for example State -> Writer -> IO) or 2. to use IORef for the data set or 3. something else? Are monad stacks with 3 and more monads common? How could an example implementation look like? What I have for now is: -- Status data Status = OK | FAILED deriving (Show, Read, Enum) -- Example data set manages by state type Config = [String] -- WriterT transformer type OutputWriter = WriterT [String] IO Status -- example execute function execute :: [String] -> OutputWriter execute fs = do rs <- liftIO loadData fs tell $ map show rs return OK -- run it inside e.g. main (s, os) <- runWriterT $ execute files How do I bring a state into this, for example for: execute fs = do ?? conf <- get ?? -- get Config from state rs <- liftIO loadData conf fs ?? set conf ?? -- change Config and set to state tell "new state:" tell $ show conf return OK Do I have to use and how do I use StateT in this context: data DataState = StateT Config OutputWriter ?? and how do I run it runStateT . runWriterT? Thanks for help, Adam

4 4

Deleting list of elements from Data.Set
by Gracjan Polak 30 Jan '08

30 Jan '08

My strictness analyser in my brain hurts. Which one (foldl,foldl',foldr) is the best way? Prelude Data.Set Data.List> let s = fromList [1,2,3,4,5] Loading package array-0.1.0.0 ... linking ... done. Loading package containers-0.1.0.0 ... linking ... done. Prelude Data.Set Data.List> foldl (.) id (Data.List.map Data.Set.delete [1,3,5]) s fromList [2,4] Prelude Data.Set Data.List> foldl' (.) id (Data.List.map Data.Set.delete [1,3,5]) s fromList [2,4] Prelude Data.Set Data.List> foldr (.) id (Data.List.map Data.Set.delete [1,3,5]) s fromList [2,4] Which one is best? -- Gracjan

4 4

Security Notice: Buffer overflow fixed in PCRE library
by ChrisK 30 Jan '08

30 Jan '08

The PCRE library has just fixed a buffer overflow (related to UTF-8 mode). There are several haskell wrappers for the pcre library. If you use a wrapper for the PCRE library (libpcre) then you may want to upgrade the underlying library. http://pcre.org/news.txt states: > News about PCRE releases > ------------------------ > > Release 7.6 28-Jan-08 > --------------------- > > The main reason for having this release so soon after 7.5 is because it fixes a > potential buffer overflow problem in pcre_compile() when run in UTF-8 mode. In > addition, the CMake configuration files have been brought up to date. Cheers, Chris

1 0