June 2008 - Haskell-Cafe

ieee 754
by Sean McLaughlin 22 Jun '08

22 Jun '08

Hello, I'm considering using Haskell for a numerical application. However, I need to rely on IEEE 754 standards being implemented correctly. What is the current state of 754 in Haskell? The definition has this paragraph, which makes me suspect Haskell is not appropriate for this application: "The default floating point operations defined by the Haskell Prelude do not conform to current language independent arithmetic (LIA) standards. These standards require considerably more complexity in the numeric structure and have thus been relegated to a library. Some, but not all, aspects of the IEEE floating point standard have been accounted for in Prelude class RealFloat." Thank you, Sean

3 2

Testing
by Haskell Hall 22 Jun '08

22 Jun '08

Testing Testing _________________________________________________________________ Need to know now? Get instant answers with Windows Live Messenger. http://www.windowslive.com/messenger/connect_your_way.html?ocid=TXT_TAGLM_W…

1 0

Scripting OpenOffice with Haskell
by Henning Thielemann 22 Jun '08

22 Jun '08

Has someone already tried to communicate with OpenOffice via Haskell? An interface seems to exist: http://framework.openoffice.org/scripting/index.html For instance, I like to mark some cells in oocalc, preprocess them and write them to a database.

1 0

message passing style like in Haskell?
by jinjing 22 Jun '08

22 Jun '08

Hi guys, This is my second attempt to learn Haskell :) Any way here's the code: module Dot where import Prelude hiding ( (.) ) (.) :: a -> (a -> b) -> b a . f = f a infixl 9 . So for example, 99 questions: Problem 10 (*) Run-length encoding of a list. comparing: encode xs = map (\x -> (length x,head x)) (group xs) to encode xs = xs.group.map token where token x = (x.length, x.head) I found starting with data and working my way to a solution seems to be easier to think with, or maybe it's just me ... What is your thought? Jinjing

10 11

Pretty little definitions of left and right folds
by George Kangas 21 Jun '08

21 Jun '08

Hi, Cafe, For no practical purpose, I wrote new definitions of list folds as chains of partially applied functions, joined by composition. I took the liberty of rearranging the type signature. > foldright :: (a -> b -> b)) -> [a] -> b -> b > foldright f = chain where > chain (a:as) = (f a).(chain as) > chain _ = id > foldleft :: (a -> b -> b) -> [a] -> b -> b > foldleft f = chain where > chain (a:as) = (chain as).(f a) > chain _ = id These definitions are point free, with respect to the "initializer" argument (which is now the last argument). Also, you can see how similar they are to each other, with the difference boiling down to the order of the composition, e.g.: foldright (+) [1, 2, 3] 0 == ( (1 +).(2 +).(3 +).id ) 0 foldleft (+) [1, 2, 3] 0 == ( id.(3 +).(2 +).(1 +) ) 0 We can also see the following identities: foldright f as == foldright (.) (map f as) id foldleft f as == foldright (flip (.)) (map f as) id I like that second one, after trying to read another definition of left fold in terms of right fold (in the web book "Real World Haskell"). The type signature, which could be written (a -> (b -> b)) -> ([a] -> (b -> b)), suggests generalization to another type constructor C: (a -> (b -> b)) -> (C a -> (b -> b)). Would a "foldable" typeclass make any sense? Okay, it goes without saying that this is useless dabbling, but have I entertained anyone? Or have I just wasted your time? I eagerly await comments on this, my first posting. very truly yours, George Kangas

5 8

an example from YAHT doesn't run
by trifod＠gmail.com 21 Jun '08

21 Jun '08

Hello, On pdf page 43 of YAHT it provides a code example. I typed this into a file named "Guess.hs" and tried unsuccessfully to load it into ghci. I am running ghci version 6.8.2 This is the output it gave me: " [1 of 1] Compiling Main ( Guess.hs, interpreted ) Guess.hs:19:12: parse error on input `doGuessing' Failed, modules loaded: none. " I've attached the code module. As far as I can tell it is identical to the example in YAHT. Did I type it wrong, is the tutorial wrong, or something else? thanks again, Jared Langson

2 1

HDBC converting a date sql value to UTCTime
by George Moschovitis 21 Jun '08

21 Jun '08

Hello, I am HDBC with the Postgres driver. I am trying to read a simple date column from the DB: stmt <- DB.prepare conn "SELECT date FROM my_table ORDER BY date DESC LIMIT 1" DB.execute stmt [] DB.commit conn rows <- fetchAllRows' stmt let ((d:_):_) = rows the d variable is of type: SqlEpochTime 1213736400 I am wondering how to convert this value to an UTCTime value. Any help appreciated (I am a newbie). -g. -- http://gmosx.me.gr http://joy.gr http://cull.gr http://nitroproject.org http://phidz.com http://joyerz.com

2 2

What's wrong with the classes/insances?
by Pieter Laeremans 20 Jun '08

20 Jun '08

HI, What 's wrong with this: type Id = String class Catalog a where listItems :: a -> IO [String] getItem :: a -> Id -> IO (Maybe String) class Item a where getCatalog :: Catalog catalog => a -> catalog data Catalog c => Content c = Content {auteur :: String, inhoud:: String, catalog::c} instance Catalog c => Item (Content c) where getCatalog (Content _ _ c) = c I get this as error from ghci: Couldn't match expected type `catalog' against inferred type `c' `catalog' is a rigid type variable bound by the type signature for `getCatalog' at ../Sites/liberaleswebsite/www.liberales.be/cgi-bin/Test.hs:16:26 `c' is a rigid type variable bound by the instance declaration at ../Sites/liberaleswebsite/www.liberales.be/cgi-bin/Test.hs:20:17 In the expression: c In the definition of `getCatalog': getCatalog (Content _ _ c) = c In the definition for method `getCatalog' Failed, modules loaded: none. thanks in advance, P -- Pieter Laeremans <pieter(a)laeremans.org> "The future is here. It's just not evenly distributed yet." W. Gibson

6 6

Wrapping FTGL in FFI calls
by Jefferson Heard 20 Jun '08

20 Jun '08

I've been looking for awhile now for a simple way to get truetype fonts into my visualizations so I can abandon the hideous GLUT fonts and make things that look like they were developed in the 1990s instead of back in the days of TRON. I found FTGL, but I'm mostly a Haskell developer these days, and resent having to go back to C just to write a simple application. So I was wondering if anyone had ever wrapped the FTGL library in Haskel FFI or whether those out there who are experts on the FFI think at first glance it should be readily wrappable by a rank amateur at FFI such as myself. http://ftgl.sourceforge.net/docs/html/ -- Jeff

4 7

Haddock compilation problem
by Ronald Guida 20 Jun '08

20 Jun '08

I just upgraded to ghc-6.8.3, using a linux binary, and I am having a problem compiling Haddock. Haddock 2.1.0 and Haddock 2.0.0.0 both fail to build under ghc-6.8.3, but they both build successfully with ghc-6.8.2. I don't know if this is a Haddock problem, or a GHC problem, or perhaps something else entirely? Here is the error I'm getting. It is the same error for either version of Haddock. [15 of 24] Compiling Haddock.GHC.Typecheck ( src/Haddock/GHC/Typecheck.hs, dist/build/haddock/haddock-tmp/Haddock/GHC/Typecheck.o ) src/Haddock/GHC/Typecheck.hs:82:4: Constructor `HsModule' should have 7 arguments, but has been given 8 In the pattern: HsModule _ _ _ _ _ mbOpts _ _ In a pattern binding: HsModule _ _ _ _ _ mbOpts _ _ = unLoc parsed In the definition of `mkGhcModule': mkGhcModule (mod, file, checkedMod) dynflags = GhcModule {ghcModule = mod, ghcFilename = file, ghcMbDocOpts = mbOpts, ghcHaddockModInfo = info, ghcMbDoc = mbDoc, ghcGroup = group, ghcMbExports = mbExports, ghcExportedNames = modInfoExports modInfo, ghcNamesInScope = fromJust $ modInfoTopLevelScope modInfo, ghcInstances = modInfoInstances modInfo} where HsModule _ _ _ _ _ mbOpts _ _ = unLoc parsed (group, _, mbExports, mbDoc, info) = renamed (parsed, renamed, _, modInfo) = checkedMod

3 4