- Haskell-Cafe - Haskell.org

pattern matching vs if-then-else
by Maarten Faddegon 12 Aug '12

12 Aug '12

Hi there, I am writing a toy compiler in Haskell to further my skills in functional programming. One of the functions I wrote is to determine the iteration count of a loop. I have a number of different test that I want to do and I find myself now testing some of these using pattern matching and some properties using an if-then-else construction. I do not consider this very pretty. My question is: are there guidelines of when to use pattern matching and when to use if-then-else? Snippet of the function I mentioned: ---8<----------------------------------------------------------------------------------- itercount (ForLoop [ ( Assignment update_lcv (Op2 "+" (Content update_lcv') update_expr) ) ] [(Assignment init_lcv init_expr)] (TestStmt (Op2 "<" (Content test_lcv) test_expr)) bodyblock) = if -- All stmts use the same lcv test_lcv == init_lcv && test_lcv == update_lcv && test_lcv == update_lcv' -- And the lcv is not updated in the body && intersect [test_lcv] (blockkills bodyblock) == [] then Just $ simple_itercount init_expr test_expr update_expr else Nothing itercount _ = Nothing ---8<----------------------------------------------------------------------------------- Thanks, Maarten

2 1

Hackage is down?
by hanjoosten 12 Aug '12

12 Aug '12

Hi, Hackage seems to be down. Is there anyone out here who knows how to get it online again? Thanks! -- View this message in context: http://haskell.1045720.n5.nabble.com/Hackage-is-down-tp5715912.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

6 5

Fwd: hGetContents Illegal byte sequence / ghc-pkg
by Benjamin Edwards 11 Aug '12

11 Aug '12

Responding to the list.. ---------- Forwarded message ---------- From: Benjamin Edwards <edwards.benj(a)gmail.com> Date: 11 August 2012 17:37 Subject: Re: [Haskell-cafe] hGetContents Illegal byte sequence / ghc-pkg To: David McBride <toad3k(a)gmail.com> Thank you Gents, Most useful. One thing that had escaped me is that of course, even if hGetContents and friends are respecting *my* locale, I have a binary distro and the package database probably isn't in my locale. I will try and find that. Thanks, Ben On 11 August 2012 15:26, David McBride <toad3k(a)gmail.com> wrote: > I had this same problem a couple weeks ago when trying to install > virthualenv and I don't really understand it got into a bad state, but the > way I solved it was by fixing the locale settings on my gentoo machine so > that I'm using UTF8. That just involved a few changes in /etc and then the > problem went away. > > On Sat, Aug 11, 2012 at 7:13 AM, Benjamin Edwards <edwards.benj(a)gmail.com>wrote: > >> Hello café, >> >> I have a program that is crashing, and I have no idea why: >> >> module Main >> where >> >> import System.Process (readProcessWithExitCode) >> >> >> main :: IO () >> main = do _ <- readProcessWithExitCode "ghc-pkg" ["describe", "hoopl"] "" >> putStrLn "Should never get here" >> >> this is using the process package from hackage. The program crashes with >> >> minimal-test: fd:5: hGetContents: invalid argument (invalid byte sequence) >> minimal-test: thread blocked indefinitely in an MVar operation >> >> inspecting the source of readProcessWithExitCode yields an obvious >> explanation to the MVar problem, but I don't understand why hGetContents is >> so offended. >> >> For the lazy it is defined as follows: >> >> readProcessWithExitCode >> :: FilePath -- ^ command to run >> -> [String] -- ^ any arguments >> -> String -- ^ standard input >> -> IO (ExitCode,String,String) -- ^ exitcode, stdout, stderr >> readProcessWithExitCode cmd args input = do >> (Just inh, Just outh, Just errh, pid) <- >> createProcess (proc cmd args){ std_in = CreatePipe, >> std_out = CreatePipe, >> std_err = CreatePipe } >> >> outMVar <- newEmptyMVar >> >> -- fork off a thread to start consuming stdout >> out <- hGetContents outh >> _ <- forkIO $ C.evaluate (length out) >> putMVar outMVar () >> >> -- fork off a thread to start consuming stderr >> err <- hGetContents errh >> _ <- forkIO $ C.evaluate (length err) >> putMVar outMVar () >> >> -- now write and flush any input >> when (not (null input)) $ do hPutStr inh input; hFlush inh >> hClose inh -- done with stdin >> >> -- wait on the output >> takeMVar outMVar >> takeMVar outMVar >> hClose outh >> hClose errh >> >> -- wait on the process >> ex <- waitForProcess pid >> >> return (ex, out, err) >> >> Now having looked at the source of ghc-pkg it is dumping it's output >> using putStr and friends, so that should be using my local encoding on the >> system, right? and so should hGetContents in my program..? >> >> Now, for the curious: the reason I care is that this problem has >> effectively prevented me from using virthualenv. Sadness and woe. >> >> _______________________________________________ >> Haskell-Cafe mailing list >> Haskell-Cafe(a)haskell.org >> http://www.haskell.org/mailman/listinfo/haskell-cafe >> >> >

1 0

Re: [Haskell-cafe] Data.Data and OverlappingInstances
by oleg＠okmij.org 11 Aug '12

11 Aug '12

Timo von Holtz wrote: > class Test a where > foo :: Monad m => m a > > instance Num a => Test a where > foo = return 1 > > instance Test Int where > foo = return 2 > > test constr = fromConstrM foo constr I'm afraid the type checker is right. From the type of fromConstrM fromConstrM :: forall m a. (Monad m, Data a) => (forall d. Data d => m d) -> Constr -> m a we see its first argument has the type (forall d. Data d => m d) If instead it had the type (forall d. Test d => m d) we would have no problem. As it is, when you pass 'foo' of the type (Test d, Monad m) => m d as the first argument of fromConstrM, which only assures the Data d constraint on 'd' and _nothing_ else, the compiler checks if it can get rid of (discharge) the constraint Test d. That is, the compiler is forced to choose an instance for Test. But there is not information to do that. Overlapping here is irrelevant. If you had non-overlapping instances > class Test a where > foo :: Monad m => m a > > instance Num a => Test [a] where > foo = return [1] > > instance Test Int where > foo = return 2 > > test constr = fromConstrM foo constr 'test' still causes the problem. The error message now describes the real problem: Could not deduce (Test d) arising from a use of `foo' from the context (Monad m, Data a) bound by the inferred type of test :: (Monad m, Data a) => Constr -> m a at /tmp/d.hs:16:1-36 or from (Data d) bound by a type expected by the context: Data d => m d at /tmp/d.hs:16:15-36 Possible fix: add (Test d) to the context of a type expected by the context: Data d => m d or the inferred type of test :: (Monad m, Data a) => Constr -> m a In the first argument of `fromConstrM', namely `foo' and it recommends the right fix: change the type of fromConstrM to be fromConstrM :: forall m a. (Monad m, Data a) => ( forall d. (Test d, Data d) => m d) -> Constr -> m a That will solve the problem. Alas, fromConstrM is a library function and we are not at liberty to change its type. > Right now I use a "case (typeOf x) of" kind of construct That is precisely the right way to use Data. SYB provides good combinators for building functions (generic producers) of that sort. But you never need unSafeCoerce: gcast is sufficient.

1 0

Re: [Haskell-cafe] CRIP: the Curiously Reoccuring Instance Pattern
by oleg＠okmij.org 11 Aug '12

11 Aug '12

Anthony Clayden wrote: > So three questions in light of the approach of abandoning FunDeps and > therefore not getting interference with overlapping: > A. Does TTypeable need to be so complicated? > B. Is TTypeable needed at all? > C. Does the 'simplistic' version of type equality testing suffer possible > IncoherentInstances? It is important to keep in mind that Type Families (and Data Families) are _strictly_ more expressive than Functional dependencies. For example, there does not seem to be a way to achieve the injectivity of Leibniz equality http://okmij.org/ftp/Computation/extra-polymorphism.html#injectivity without type families (and relaying instead on functional dependencies, implemented with TypeCast or directly). I'd like to be able to write data Foo a = Foo (SeqT a) where SeqT Bool = Integer SeqT a = [a] otherwise (A sequence of Booleans is often better represented as an Integer). A more practical example is http://okmij.org/ftp/Haskell/GMapSpec.hs http://okmij.org/ftp/Haskell/TTypeable/GMapSpecT.hs It is possible to sort of combine overlapping with associated types, but is quite ungainly. It is not possible to have overlapping associated types _at present_. Therefore, at present, TTYpeable is necessary and it has to be implemented as it is. You point out New Axioms. They will change things. I have to repeat my position however, which I have already stated several times. TTypeable needs no new features from the language and it is available now. There is no problem of figuring out how TTypeable interacts with existing features of Haskell since TTypeable is implemented with what we already have. New Axioms add to the burden of checking how this new feature interacts with others. There have been many examples when one feature, excellent by itself, badly interacts with others. (I recall GADT and irrefutable pattern matching.)

1 0

Re: [Haskell-cafe] 3 level hierarchy of Haskell objects
by Patrick Browne 11 Aug '12

11 Aug '12

4 3

resource post on Concurrent and Parallel Haskell
by Obscaenvs 10 Aug '12

10 Aug '12

Hi! I have recently gone through some tutorials on the subject of Concurrent and Parallel Haskell, most notably chapter 24 in Real World Haskell. In doing so I have collected some thoughts and resources on the subject in a post on my blog: http://monoid.se/haskell/a-resource-post-on-concurrent-and-parallel-haskell/ Is there some other resource that should really be included in such a list, or is there in your opinion anything else that should be pointed out for somebody starting out in this particular corner of the Haskell ecosystem? I would appreciate pointers and suggestions, even pertaining to English grammar :) /F

2 1

unix package licensing
by Oleg Sidorkin 10 Aug '12

10 Aug '12

System/Posix/DynamicLinker/Module/ByteString.hsc and System/Posix/DynamicLinker/Prim.hsc sources in unix-2.5.1.0 package contains the following reference to GPL-2 c2hs package: -- Derived from GModule.chs by M.Weber & M.Chakravarty which is part of c2hs -- I left the API more or less the same, mostly the flags are different. Does it mean that GPL license should be applied to unix package and all dependent packages? unix package is included into Haskel Platform so does it mean that Haskell Platform should be GPL also?

1 0

MIN_VERSION_base
by Kazu Yamamoto 10 Aug '12

10 Aug '12

Hello, The stable branch of the network library includes the following code: #if !(MIN_VERSION_base(4,6,0)) I cannot compile the stable branch with HP 2011.4.0.0 on Linux 2: Preprocessing library network-2.3.0.14... BSD.hsc:115:23: error: missing binary operator before token "(" BSD.hsc:115:23: error: missing binary operator before token "(" BSD.hsc:115:23: error: missing binary operator before token "(" compiling dist/build/Network/BSDhsc_make.c failed (exit code 1) command was: /usr/bin/gcc -c dist/build/Network/BSD_hsc_make.c -o dist/build/Network/BSD_hsc_make.o -fno-stack-protector -fno-stack-protector -DGLASGOW_HASKELL_=700 -Dlinux_BUILD_OS -Dlinux_HOST_OS -Di386_BUILD_ARCH -Di386_HOST_ARCH -Iinclude -DCALLCONV=ccall -I/ghc7.0.4/lib/ghc-7.0.4/unix-2.4.2.0/include -I/ghc7.0.4/lib/ghc-7.0.4/bytestring-0.9.1.10/include -I/ghc7.0.4/lib/ghc-7.0.4/base-4.3.1.0/include -I/ghc7.0.4/lib/ghc-7.0.4/include -I/ghc7.0.4/lib/ghc-7.0.4/include -I/ghc7.0.4/lib/ghc-7.0.4/include/ "dist/build/autogen/cabal_macros.h" defines MIN_VERSION_base but "dist/build/Network/BSD_hsc_make.c" does not include it. If you guys know why this happens, please let me know. For more information, please refer to: https://github.com/haskell/network/issues/43 --Kazu

3 2

[Wadler 89] Philip Wadler. "Theorems for free!": Can't get the same results in Haskell
by Stayvoid 10 Aug '12

10 Aug '12

Hi there, I was reading "Real World Haskell" and found this paper. [1] On the first page of the paper there is an example: inc* (odds_Int [1,2,3]) = [2,4] /= [4] = odds_Int (inc* [1,2,3]) I tried to implement it in Haskell: (I'm a newbie. I guess it's possible to write a better version.) module Param where import Prelude odds :: [Int] -> [Int] odds [] = [] odds [x] = if odd x then [x] else [] odds (x:xs) = if odds [x] == [] then odds xs else [x] ++ odds xs inc :: [Int] -> [Int] inc [] = error "Empty list" inc [x] = [succ x] inc (x:xs) = inc [x] ++ inc xs Looks fine: *Param> odds [1,2,3] [1,3] *Param> inc [1,2,3] [2,3,4] But my results differ from the paper's: *Param> inc (odds [1,2,3]) [2,4] *Param> odds (inc [1,2,3]) [3] I doubt that there is an error in the paper. So it seems that something is wrong with my code. But I can't find the error. Could you help me? (Or there is no error at all and I misunderstood something.) [1] http://ttic.uchicago.edu/~dreyer/course/papers/wadler.pdf Thanks

2 1