August 2006 - Haskell

Problem with types.
by Szymon Ząbkiewicz 16 Aug '06

16 Aug '06

Hi! I'm new to Haskell and FP so i've got some problems. I wrote this code : getNr :: Int -> Int getNr x = do putStrLn ("enter " ++ x ++ " number:") nr <- getLine return nr And when I try to load it to GHCi I get an error (there is also plenty of others but they look the same): couldn't match Int against t t1 Help needed.

3 2

CUFP Final Program and Early Registration Deadline Reminder
by Kathleen Fisher 16 Aug '06

16 Aug '06

Hi all! Reminder! The early registration deadline for Commercial Users of Functional Programing (CUFP) is this Friday, August 18. Registration is available from the ICFP registration page: http://regmaster2.com/conf/icfp2006.html CUFP will take place on Thursday, September 21st, the day after ICFP. The final program for CUFP is now available: 8:30 Haskell and the Linspire Toolchain. Clifford Beshers, Linspire Inc. 9:00 Automated Testing of Reactive Systems with SML. Steve Sims, Reactive Systems, Inc. 9:30 Scheme in Space. Richard Cleis, Air Force Research Lab. 10:00 B R E A K 10:30 Mission Critical OCaml: Experiences using FP in Finance. Yaron Minsky, Jane Street Capital. abstract 11:00 Why Functional Programming Matters to Credit Suisse. Gabriele Keller and Lennart Augustsson, Credit Suisse First Boston. abstract 11:30 Group Discussion: Mixed Mode Approaches and/or Finding Employees. 12:00 L U N C H 2:00 Analysis of Microprocessor Components with a Functional Language-based Formal Verification Toolbox. Roope Kaivola, Intel Corporation. 2:30 Teaching Old Dogs New Tricks: Selling Bluespec to Hardware Designers. Rishiyir Nihkil, Bluespec, Inc. 3:00 Group Discussion: Adoption of New Technology: Making the Case. 3:30 B R E A K 4:00 Artifically Intelligent Haskell. J. Garret Morris, Aetion Inc. 4:30 Betting on Functional Programming and Winning. Erik Stenman, Kreditor. 5:00 Wrap-Up.

1 0

Call for participation: Workshop on Generic Programming 2006
by Ralf Hinze 15 Aug '06

15 Aug '06

Dear all, the Workshop on Generic Programming early registration deadline is only a few days away: August 18, 2006 (http://regmaster2.com/conf/icfp2006.html) ==> We have reserved 30 minutes for *lightning talks*. If you plan to ==> attend and if you would like to give a short talk (about half-baked, ==> exciting, new stuff) please drop me a short note. Slots will be ==> reserved on a first-come-first-serve basis. Looking forward to seeing you in Portland, Ralf Hinze ============================================================================ CALL FOR PARTICIPATION Workshop on Generic Programming 2006 Portland, Oregon, 16th September 2006 The Workshop on Generic Programming is sponsored by ACM SIGPLAN and forms part of ICFP 2006. Previous Workshops on Generic Programming have been held in Marstrand (affiliated with MPC), Ponte de Lima (affiliated with MPC), Nottingham (informal workshop), Dagstuhl (IFIP WG2.1 Working Conference), Oxford (informal workshop), and Utrecht (informal workshop). http://www.informatik.uni-bonn.de/~ralf/wgp2006.{html,pdf,ps,txt} ============================================================================ Preliminary program ------------------- 9.oo - 1o.3o, session chair: Ralf Hinze (Universität Bonn) Welcome Design Patterns as Higher-Order Datatype-Generic Programs Jeremy Gibbons (Oxford University) Type Theoretic Design Patterns Ondrej Rypacek, Roland Backhouse, Henrik Nilsson (University of Nottingham) Generating Generic Functions Johan Jeuring, Alexey Rodriguez, Gideon Smeding (Utrecht University) 11.oo - 12.3o, session chair: Peter Dybjer (Chalmers University of Technology) Good Advice for Type-Directed Programming Geoffrey Washburn, Stephanie Weirich (University of Pennsylvania) Context-Parametric Polykinded Types Pablo Nogueira (University of Nottingham) Modular Generic Programming with Extensible Superclasses Martin Sulzmann, Meng Wang (University of Singapore) 14.3o - 16.oo, session chair: Jeremy Gibbons (Oxford University) Report from the program chair Ralf Hinze (Universität Bonn) Scrap++: Scrap Your Boilerplate in C++ Gustav Munkby, Andreas Priesnitz, Sibylle Schupp, Marcin Zalewski (Chalmers University of Technology) A Technique for Generic Iteration and Its Optimization Stephen Watt (University of Western Ontario) Lightning talks: to be announced 16.3o - 18.oo, session chair: Jeremy Siek (Rice University) Towards An Automatic Complexity Analysis for Generic Programs Kyle Ross (Indiana University) An Object-Oriented Approach to Datatype Generic Programming Adriaan Moors, Frank Piessen, Wouter Joosen (Katholieke Universiteit Leuven) Discussion ============================================================================

1 0

is there a library for GIS formats?
by Martin Tomko 15 Aug '06

15 Aug '06

Hi list, I was wondering if there is a library for accessing data from GIS formats (ESRI shp files and similar), in Haskell. I would like to test some of my algorithms with "real" data... Thanks Martin

1 0

Problem with comboBox
by L. J. 14 Aug '06

14 Aug '06

I have send this mail to the gtkhs(a)haskell.org mailing list. I do not know if this list is active, so I reply the mail in this one. If the other list is active, I'm sorry. Hi, i'm making my first graphic interface program with Haskell. I use Glade and Gtk+2.0. I create an empty comboBox with Glade and I want to add items (strings) in run time (by means of the haskell code). I tried with the functions 'comboBoxAppendText' and 'comboBoxInsertText' (http://haskell.org/gtk2hs/docs/current/Graphics-UI-Gtk-MenuComboToolbar-Com…) but I can not do anything for display the information I want. This functions require to work with a comboBox creates with the constructor comboBoxNewText (and I think this constructor is not the used by Glade). I can create it from Glade? I can do I want by another method? Thanks.

3 4

Haskell Weekly News - August 14, 2006
by dons＠cse.unsw.edu.au 14 Aug '06

14 Aug '06

--------------------------------------------------------------------------- Haskell Weekly News http://haskell.org/haskellwiki/HWN Issue 40 - August 14, 2006 --------------------------------------------------------------------------- Welcome to issue 40 of HWN, a weekly newsletter covering developments in the Haskell community. Each week, new editions are posted to [1]the Haskell mailing list as well as to [2]the Haskell Sequence and [3]Planet Haskell. [4]RSS is also available, and headlines appear on [5]haskell.org. A mega super bumper issue, for the 1st birthday of the Haskell Weekly News 1. http://www.haskell.org/mailman/listinfo/haskell 2. http://sequence.complete.org/ 3. http://planet.haskell.org/ 4. http://sequence.complete.org/node/feed 5. http://haskell.org/ Announcements * The Haskell Workshop . Andres Loeh [6]announced the preliminary schedule of the Haskell Workshop 2006, part of the 2006 International Conference on Functional Programming (ICFP) 6. http://article.gmane.org/gmane.comp.lang.haskell.general/14104 * dbus haskell bindings . Evan Martin [7]announced preliminary D-Bus Haskell bindings. D-Bus is a message bus system, a simple way for applications to talk to one another. [8]More 7. http://article.gmane.org/gmane.comp.lang.haskell.cafe/13771 8. http://neugierig.org/software/hdbus/ * The GHC typechecker is Turing-complete . Robert Dockins was able to [9]show how that the GHC typechecker with multi-parameter typeclasses, functional dependencies, and undecidable instances is Turing-complete. 9. http://article.gmane.org/gmane.comp.lang.haskell.general/14088 * Haskell Program Coverage . Colin Runciman [10]announced the first release of hpc, a new tool for Haskell developers. Hpc records and displays Haskell program coverage. It provides coverage information of two kinds: source coverage and boolean-control coverage. [11]More here 10. http://article.gmane.org/gmane.comp.lang.haskell.general/14087 11. http://www.galois.com/~andy/hpc-intro.html * Smash your boiler-plate without class and Typeable . Oleg Kiselyov [12]described a new generic programming technique, expressive enough to traverse a term and return another term of a different type, determined by the original term's type/structure. [13]More details 12. http://article.gmane.org/gmane.comp.lang.haskell.general/14086 13. http://okmij.org/ftp/Haskell/syb4.hs * Paper: Software Extension and Integration with Type Classes . Ralf Laemmel and Klaus Ostermann [14]invite comments towards the final version of their paper [15]Software Extension and Integration with Type Classes 14. http://article.gmane.org/gmane.comp.lang.haskell.general/14040 15. http://homepages.cwi.nl/~ralf/gpce06/ * HSP.Clientside 0.01 . Joel Bj�rnson [16]announced a release of his Summer of Code project HSP.Clientside 0.01. Present features include an embedding of (typed) JavaScript language in Haskell, a small combinator library for generating JavaScript code, and high-level interface to Ajax functionality. 16. http://article.gmane.org/gmane.comp.lang.haskell.general/14023 * Monadic probabilistic functional programing . Stefan Karrmann [17]announced that he had extended Martin Erwig's PFP library to support abstract monads, cabal and darcs 17. http://article.gmane.org/gmane.comp.lang.haskell.general/14012 * hdbc-odbc 1.0.0.1 . John Goerzen [18]released DBC-odbc, the ODBC backend driver for HDBC, version 1.0.0.1. 18. http://article.gmane.org/gmane.comp.lang.haskell.general/13998 * Few Digits 0.5.0 . Russell O'Connor This year, Few Digits competed in the [19]More Digits contest. To celebrate, version 0.5.0 of Few Digits is available. Few Digits 0.5.0 is now ten times faster and three times more complicated. Few Digits has been Cabalized for your convenience. [20]More info 19. http://rnc7.loria.fr/competition.html 20. http://r6.ca/FewDigits/ * System.FilePath 0.9 . Neil Mitchell [21]announced System.FilePath 0.9 21. http://article.gmane.org/gmane.comp.lang.haskell.general/13985 * The History of Haskell . Phil Wadler, John Hughes, Paul Hudak and Simon Peyton Jones [22]have been writing a paper, The History of Haskell, for the History Of Programming Languages conference (HOPL'07), and they invite feedback. Wiki page [23]here. 22. http://article.gmane.org/gmane.comp.lang.haskell.general/13983 23. http://haskell.org/haskellwiki/History_of_Haskell * AngloHaskell . Lemmih [24]mentioned that AngloHaskell will be held at Cambridge in August. The agenda includes beer, unicycles, hacking and other fun. [25]More info 24. http://article.gmane.org/gmane.comp.lang.haskell.general/13979 25. http://haskell.org/haskellwiki/AngloHaskell * Haskell XML Toolbox Version 6.0, 6.1 . Uwe Schmidt [26]announced [27]two new versions of the Haskell XML Toolbox. New features include ghc 6.4.2 support, better XPath integration, separate documentation for filter API and an arrow API 26. http://article.gmane.org/gmane.comp.lang.haskell.general/13924 27. http://article.gmane.org/gmane.comp.lang.haskell.general/13949 * Down the rabbit hole . Bulat Ziganshin [28]announced the availability of a new tutorial directed toward comprehensive explanation of the IO monad, and it's use in complex programs 28. http://article.gmane.org/gmane.comp.lang.haskell.general/13914 * ldap-haskell, arch2darcs and darcs-buildpackage . John Goerzen [29]posted new versions of these packages 29. http://article.gmane.org/gmane.comp.lang.haskell.general/13912 * Internships on GHC and Haskell at MSR Cambridge . Simon Peyton-Jones [30]announced that MSR Cambridge is taking interns year-round, not just in the summer months. GHC HQ are keen to attract motivated and well-qualified folk to work on improving or developing GHC. [31]More details 30. http://article.gmane.org/gmane.comp.lang.haskell.glasgow.user/10219 31. http://hackage.haskell.org/trac/ghc/wiki/Internships * FGL . Martin Erwig [32]announced a new release of his well known Functional Graph Library (FGL). 32. http://article.gmane.org/gmane.comp.lang.haskell.libraries/4948 * Takusen . Alistair Bayley and Oleg Kiselyov [33]released a new version of Takusen, a library for accessing DBMSs. The most significant code change is a new internal design, giving better separation of concerns like statement preparation, binding, and result-set processing. Takusen is now held in darcs, and [34]hosted at haskell.org 33. http://article.gmane.org/gmane.comp.lang.haskell.libraries/4887 34. http://darcs.haskell.org/takusen * Text.Regex.Lazy 0.44, 0.56, 0.66 and 0.70 . Chris Kuklewicz [35]announced [36]Text.Regex.Lazy 0.44-0.70, with many enhancements. Multiple backends are supported, in addition to the "full lazy" and the DFA backends. Text.Regex.Lazy is a replacement and enhancement for Text.Regex. More details [37]here 35. http://article.gmane.org/gmane.comp.lang.haskell.libraries/4977 36. http://sourceforge.net/projects/lazy-regex 37. http://article.gmane.org/gmane.comp.lang.haskell.libraries/5028 * Streams 0.2.1 beta . Bulat Ziganshin [38]released Streams 0.2.1 beta, featuring various bug fixes and improvements to the streams library 38. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/4865/focus=4865 Haskell' This section covers the [39]Haskell' standardisation process. * [40]Numeric Class reworking 39. http://hackage.haskell.org/trac/haskell-prime 40. http://thread.gmane.org/gmane.comp.lang.haskell.prime/1568/focus=1568 Discussion * Package "mounting" . Sven Hallberg [41]proposed drop the assumption that Haskell modules are closed entities, and rather always consider them to be seen in the context of a particular package. 41. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/4900/focus=4900 * Cabal and DLLs . Jason Dagit [42]described how he was able to use Cabal with Visual Haskell to build Dlls in Windows. 42. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/4952/focus=4952 * Replacing GMP . Peter Tanski [43]sparked a long discussion on the merits and difficulties of replacing the GMP numerical library used by GHC and nhc98 with a library using a less restrictive license. More [44]here, [45]here and [46]here 43. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/10287/focus=10… 44. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/10324/focus=10… 45. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/10393/focus=10… 46. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/10408/focus=10… * The cost of Integer . Serge Mechveliani [47]asked about the runtime cost of Integer, triggering an interesting discussion 47. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/10313/focus=10… * Thread-local variables . Frederik Eaton [48]started a large discussion on the merits or otherwise of thread-local variables. 48. http://thread.gmane.org/gmane.comp.lang.haskell.general/14033/focus=14033 * Type level programming made simple . Oleg Kiselyov [49]composed a useful comparison between Prolog and type level programming in Haskell 49. http://article.gmane.org/gmane.comp.lang.haskell.cafe/13785 * IRC channel statistics . Don Stewart [50]created some graphs and analysis of activity and growth of the [51]#haskell IRC channel 50. http://article.gmane.org/gmane.comp.lang.haskell.cafe/13789 51. http://www.haskell.org/haskellwiki/IRC_channel Quotes of the Week * Dan Piponi : "Writing introductions to monads seems to have developed into an industry" * John Meacham : "I liken learning Haskell to tipping over a vending machine. You can't just push it, you gotta rock it back and forth a few times building up momentum until bam! suddenly the flash of insight hits and it all makes sense." * Andrew Bromage : "What was considered 100 milli-Olegs of type hackery five years ago is standard operating procedure th ese days" * Tim Toady : "Learning Haskell itself is easy -- I've done it several times already" * Adam : "Lisp is like Ruby with an ugly syntax, impossible stdlib, and macros that don't compensate for above shortcomings" * Lennart A : "Hbc still has some unique features, like views. That I've never used.." * Edward K : "I've been playing with type level 2s complement arithmetic" * Edwin B : "I think I'd panic if I had to write something that wasn't a compiler" * Ozone : "Our C++ guru at work is getting a bit sick of me saying: Ah, so that's like in Haskell, but not quite as elegant..." * roconnor : "Life lesson from today: 1 closure is nice, 500 000 closures sucks" * scsibug : "I could have sworn it was incorrect until it type-checked" * stepcut : "I wrote 1 + 1 in php once, and got 11 -- wasn't quite what I was hoping for" Contributing to HWN To help create new editions of this newsletter, please see the [52]contributing information. Send stories to dons at cse.unsw.edu.au . The darcs repository is available at darcs get http://www.cse.unsw.edu.au/~dons/code/hwn 52. http://haskell.org/haskellwiki/HWN

2 1

"class []" proposal Re: [Haskell-cafe] One thought: Num to 0 as ? to list?
by Bulat Ziganshin 14 Aug '06

14 Aug '06

Hello Marc, Sunday, August 13, 2006, 10:36:39 PM, you wrote: > In other words: why not overload (:) ? i have such proposal, more or less complete: 1) define [] as type class and [] and ':' as operations of this class: class [] c where [] :: c a -- creates empty container (:) :: a -> c a -> c a -- prepends new element to the head of container head :: c a -> a -- returns first container's element tail :: c a -> c a -- returns remaining container's elements 2) allow to use type classes in type declarations like the types itself. for example, allow the following: f :: Num a => a -> Int write as f :: Num -> Int and following: sequence :: Monad m => [m a] -> m [a] write as sequence :: [Monad a] -> Monad [a] these two changes together will change the meaning of existing list functions declarations, making them polymorphic functions that are ready to work with any container which is an instance of [] class. For example, the following declaration: foldr :: (a -> b -> b) -> b -> [a] -> b now will be treated as: foldr :: ([] c) => (a -> b -> b) -> b -> c a -> b 3) translate pattern matching on the left side of function equations to the equivalent pattern guards: foldr k z [] = z foldr k z (x:xs) = x `k` foldr k z xs turns into: foldr k z xx | xx==[] = z | x<-head xx, xs<-tail xx = x `k` foldrL k z xs 4) improve Haskell defaulting mechanism to allow declaring default container type used when there is not enough context: default [] => List main = print (head [1,2,3]) Here, [1,2,3] desugared into 1:2:3:[]. Because there is no information which allow to decide which container should be used, we defaults to use List instance: type List a = Cons a (List a) | Nil instance [] List where ..... Some Haskeller once said that lists are widely used in Haskell primarily because they are supported much better than any other data structure. All these changes together will allow to use any other container with the same easy as lists. i have attached a file which implements the idea without language changes. at the last end, it computes length and sum of two containers - list and array slice - using the same polymorphic functions: lengthL xs = foldrL (\_->(+1)) 0 xs sumL xs = foldrL (+) 0 xs Of course, this idea is not yet completely developed. Two problems that i see just now is dealing with pattern matching failures (f.e. trying to call 'head' for empty list) and building an hierarchy of classes representing various list features. at least, array slices don't support efficient ':' implementation. -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0

Haskell Workshop 2006 Call for participation
by Andres Loeh 14 Aug '06

14 Aug '06

Please note that the early registration deadline is August 18, 2006. Cheers, Andres

1 0

2 programs VERY URGENT
by Mohammad Mehdi Bezyan 14 Aug '06

14 Aug '06

Hello All There are 2 questions for my project and I need to aswer them as soon as possible. The answers need to be written in Haskell using recursive methods. The questions are the following: Number 1: The set of n-combinations of a set S is the set of subsets of S of size n. Write a function combs :: Int -> [a] -> [[a]] which, representing sets as lists, calculates n-combinations of a given list. For example, combs 3 [1,3,5,6,9] is [[1,3,5],[1,3,6],[1,3,9],[1,5,6],[1,5,9], [1,6,9],[3,5,6],[3,5,9],[3,6,9],[5,6,9]] Within each combination, the n values should appear in the same order as they appear in the original list (so, for example, [1,5,3] should not appear instead of [1,3,5] in the result for combs 3 [1,3,5,6,9]). But the list of combinations may appear in any order (ie, [1,3,5] need not appear first in the result of combs 3 [1,3,5,6,9]). You may assume that the list argument to combs does not contain duplicate members. Number 2: Given a list of lists, a list is in the Cartesian product of them if it contains one member from each of them. Write a function cart :: [[a]] -> [[a]] which calculates this. For example, cart [[1,2], [4], [7,8,9]] is [[1,4,7],[1,4,8],[1,4,9],[2,4,7],[2,4,8],[2,4,9]] Each member of the Cartesian product must appear in the appropriate order: eg, in [1,4,7] above, 1 appears first, then 4, then 7, because 1 is from the first input list [1,2], 4 is from the second input list [4], and 7 is from the third input list [7]. The list of members of the Cartesian product may appear in any order (ie, [1,4,7] need not appear first in the list). It would be very much appreciated if you could help me or provide me with some code, please. I need them very soon. Thanks! Send instant messages to your online friends http://au.messenger.yahoo.com

4 4

haskell.org account cleanup / administration
by John Peterson 14 Aug '06

14 Aug '06

I'm going to try and be more organized about administering haskell.org. I've started a wiki page, http://haskell.org/haskellwiki/Haskell.org, that will allow me to track the accounts I've given out and tell everyone what haskell.org is for and who to contact for various matters. Would everyone with an account on haskell.org please add themselves to the list on this page? We'll be shutting down accounts that have become inactive so this is the best way to let us know you're still around. Anyone that administers something on haskell.org should also place themselves in the contact list so that people will know who to turn to if something isn't working. It would be nice if the other haskell machines (like cvs.haskell.org) added something to this page (or created a new page). Thanks!!! John

1 0