June 2011 - Haskell-Cafe

Re: [Haskell-cafe] Haskell-Cafe Digest, Vol 94, Issue 34
by Gregory Guthrie 15 Jun '11

15 Jun '11

------------------------------------------- > Subject: Re: [Haskell-cafe] Best platform for development with GHC? > On Wed, 15 Jun 2011, Dmitri O.Kondratiev wrote: > > Since I maintain the gnuplot binding for Haskell - what are the particular > problems with that package on Windows? > > ******************************************** [Guthrie] I installed gnuplot using cabal on Windows, and get this: :m + Graphics.Gnuplot.Simple plotFunc [] (linearScale 1000 (-10,10)) sin Microsoft Windows [Version 6.1.7601] Copyright (c) 2009 Microsoft Corporation. All rights reserved. C:\Users\guthrie>ghci GHCi, version 7.0.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package ffi-1.0 ... linking ... done. Prelude> :m + Graphics.Gnuplot.Simple Prelude Graphics.Gnuplot.Simple> plotFunc [] (linearScale 1000 (-10,10)) sin Loading package bytestring-0.9.1.10 ... linking ... done. Loading package Win32-2.2.0.1 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package time-1.2.0.3 ... linking ... done. Loading package filepath-1.2.0.0 ... linking ... done. Loading package old-time-1.0.0.6 ... linking ... done. Loading package directory-1.1.0.0 ... linking ... done. Loading package process-1.0.1.5 ... linking ... done. Loading package array-0.3.0.2 ... linking ... done. Loading package containers-0.4.0.0 ... linking ... done. Loading package monoid-transformer-0.0.2 ... linking ... done. Loading package utility-ht-0.0.7 ... linking ... done. Loading package gnuplot-0.4.2 ... linking ... done. *** Exception: gnuplot: createProcess: does not exist (No such file or directory)

2 1

Data import from octave / text file
by kaffeepause73 15 Jun '11

15 Jun '11

Dear all, I want to read numeric data in vector / matrix format generated by octave. As I haven't seen any octave specific libraries yet (only matlab), I'm tyrying the way via ascii - file. The folloing file log.txt contains three signal traces in the three columns (time, signal 1, signal 2) 0 30 9 0.1 30 9 0.2 30 9 0.3 30 9 0.4 30 9 0.5 30 9 0.6000000000000001 30 9 0.7000000000000001 30 9 0.8 30 9 0.9 30 9 1 30 9 I got the following bit of code compiled with ghc: import System.IO main = do text <- readFile "./log.txt" putStrLn text putStrLn "SecondLot" let m = read text :: [[Double]] putStrLn (show m) putStrLn "Finish" Reading text works, but not reading text into m. The Output: 0 30 9 0.1 30 9 0.2 30 9 0.3 30 9 0.4 30 9 0.5 30 9 0.6000000000000001 30 9 0.7000000000000001 30 9 0.8 30 9 0.9 30 9 1 30 9 SecondLot signalImport: Prelude.read: no parse Any suggestions ? - Thanks in advance. -- View this message in context: http://haskell.1045720.n5.nabble.com/Data-import-from-octave-text-file-tp44… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

5 6

Lazy lists vs generators
by oleg＠okmij.org 15 Jun '11

15 Jun '11

Yves Pare`s wrote: > I cannot help smiling when I hear Java/C#/Python developers telling about > the "wondeful features" of their languages ;) > (- We have generators! We can _yield_ values! [1] > - Yeah, so nice... We have simple lists... with lazy evaluation) Just for the record, lazy lists per se are sufficient only for simplest generators. Re-writing an accumulating traversal to accommodate suspension by hand is too ungainly and error prone. Still, if we just add exceptions (the Either monad), we can idiomatically write all the generator examples of Python or Icon, for example. If we do not need effects other than yielding or non-determinism, [Either e a], which is ErrorT e [] a, suffices. If we need IO or mutable state, we should replace [] with LogicT (which is the monad of lazy lists with effect). The following web page talks about lazy lists and generators http://okmij.org/ftp/continuations/generators.html in (perhaps too) great detail.

3 3

Re: [Haskell-cafe] Best platform for development with GHC?
by michael rice 15 Jun '11

15 Jun '11

I don't know about the other Linux distros, but Fedora seems to have excellent support for Haskell. Michael

1 0

YesodAuth documentation
by Никита Тимофеев 14 Jun '11

14 Jun '11

Does exist some simple examples using yesod authentication except code of Haskellers site? Does exist some examples using YesodAuthEmail? For example, can i use third-party mail server instead of Sendmail for YesodAuthEmail? Can i disable registration confirmation? Does passwords stored in a database in an unencrypted form? -- Timofeev N.D.

3 3

Re: [Haskell-cafe] Acquiring a random set of a specific size (w/o dups) from a range of Ints
by Patrick Perry 14 Jun '11

14 Jun '11

There are time/space tradeoffs in sampling without replacement. The version in monte-carlo takes space O(n) and time O(k), for all k. I chose this algorithm instead of a streaming algorithm, with takes space O(k) and time O(n). If k is much less than n, you can improve a bit. Here is a version taking space O(k) and time O(k^2), *provided k << n*; when k is large relative to n, the time increases to roughly O(n log n). import Control.Monad( liftM ) import Control.Monad.MC import Data.List( nub ) sampleSmallIntSubset :: MonadMC m => Int -> Int -> m [Int] sampleSmallIntSubset n k = (take k . nub) `liftM` repeatMC (sampleInt n) sampleSmallSubset :: MonadMC m => [a] -> Int -> m [a] sampleSmallSubset xs k = (map (xs !!)) `liftM` sampleSmallIntSubset (length xs) k -- strict version of the above sampleSmallSubset' :: MonadMC m => [a] -> Int -> m [a] sampleSmallSubset' xs k = do ys <- sampleSmallSubset xs k (length ys) `seq` return ys -- by forcing the length, we make sure we are done with the RNG when we return If you're wondering about the strict version, the following example should be instructive: Let's note the first three numbers generated by the RNG with seed 0: ghci> (replicateM 3 $ uniform 0 1) `evalMC` mt19937 0 [0.999741748906672,0.16290987539105117,0.28261780529282987] Now, let's look at the first number we get after a call to sampleSmallSubset, if we don't evaluate the subset: ghci> sampleSmallSubset [1..10000] 2 >> uniform 0 1) `evalMC` mt19937 0 0.999741748906672 Apparently, the call to sampleSmallSubset doesn't consume any random numbers. Now, let's try the strict version: ghci> (sampleSmallSubset' [1..10000] 2 >> uniform 0 1) `evalMC` mt19937 0 0.28261780529282987 This time, the call to sampleSmallSubset' consumes two random numbers, even if we throw away the result. Since you're just starting out, I recommend using the strict versions of the functions. They are less space-efficient than the lazy versions, but they are guaranteed to be safe. > From: Jonas Almström Duregård <jonas.duregard at chalmers.se> > Subject: Re: [Haskell-cafe] Acquiring a random set of a specific size (w/o dups) from a range of Ints > To: "michael rice" < > nowgate at yahoo.com > > > Cc: "Felipe Almeida Lessa" < > felipe.lessa at gmail.com>, haskell-cafe at haskell.org > > Date: Tuesday, June 14, 2011, 5:17 AM > > > > Shuffle [1..20], then take 5? > > > > Yes, so simple, I'm embarrassed I didn't think of it. > > > That works well for small numbers, but I'm guessing it will evaluate the entire list so it should not be used for large inputs. If you have a large interval and use a relatively small part of it, the following function should be significantly faster (it builds a random permutation lazily):

1 1

YesodAuth documentation
by Тимофеев Никита Дмитриевич 14 Jun '11

14 Jun '11

Does exist some simple examples using yesod authentication except code of Haskellers site? Does exist some examples using YesodAuthEmail? For example, can i use third-party mail server instead of Sendmail for YesodAuthEmail? Can i disable registration confirmation? Does passwords stored in a database in an unencrypted form? -- Timofeev N.D.

3 4

Dead else branch does influence runtime?
by Johannes Waldmann 14 Jun '11

14 Jun '11

Dear all, I am very puzzled by a program that contains an "else" branch that is never executed, but still seems to slow down the program. (When I replace it by "undefined", the resulting program runs much faster.) http://hackage.haskell.org/trac/ghc/ticket/5256 I thought it may be a type issue (the type of the else branch forces the type of the "then" branch to be more general, thus some optimization might not fire) but the types of the branches look identical. (They are generic, but the specializer should take care of that.) I am sure GHC headquarters will look at this when they find the time but perhaps there's some additional knowledge on this mailing list that might help. J.W.

3 7

Re: [Haskell-cafe] Acquiring a random set of a specific size (w/o dups) from a range of Ints
by michael rice 14 Jun '11

14 Jun '11

Thanks, all. It seemed like something like this should exist in a prob/stat package, and if so, didn't want to reinvent the wheel. Shuffle [1..20], then take 5? Yes, so simple, I'm embarrassed I didn't think of it. Michael --- On Mon, 6/13/11, Felipe Almeida Lessa <felipe.lessa(a)gmail.com> wrote: From: Felipe Almeida Lessa <felipe.lessa(a)gmail.com> Subject: Re: [Haskell-cafe] Acquiring a random set of a specific size (w/o dups) from a range of Ints To: "michael rice" <nowgate(a)yahoo.com> Cc: haskell-cafe(a)haskell.org Date: Monday, June 13, 2011, 9:38 PM On Mon, Jun 13, 2011 at 8:56 PM, michael rice <nowgate(a)yahoo.com> wrote: > Is there an (existing) way to select 5 Ints randomly (no duplicates) from a population, say 1-20 (inclusive)? Yes, already implemented in the monte-carlo package as sampleSubset [1], sampleSubset :: MonadMC m => [a] -> Int -> m [a] Complete example code for your example: evalMC (sampleSubset [1..20] 5) (mt19937 0) Cheers! [1] http://hackage.haskell.org/packages/archive/monte-carlo/0.4.1/doc/html/Cont… -- Felipe.

3 5

GHC7 build problem
by Scott Lawrence 14 Jun '11

14 Jun '11

Trying to compile GHC7 from source (as the ubuntu repository is still on GHC6), I came across the following error in the final phase: libraries/base/GHC/ST.lhs:78:1: You cannot SPECIALISE `forever' because its definition has no INLINE/INLINABLE pragma (or its defining module `Control.Monad' was compiled without -O) This was after: git clone [...] ./sync-all get perl boot ./configure make Is there some problem with trying to build GHC7 on a machine with GHC6, or did I just get the build procedure wrong (the last line of the error looks like a hint that I did). -- Scott Lawrence

3 4