March 2009 - Haskell-Cafe

Re: [Haskell-beginners] laziness and optimization
by Adrian Neumann 21 Mar '09

21 Mar '09

You should not rely on the compiler to spot such things. As far as I know GHC doesn't do automatic caching (in many cases that would hurt performance, I think). Have a look at http://haskell.org/haskellwiki/ Memoization perhaps. Am 21.03.2009 um 14:02 schrieb Michael Mossey: > I understand a bit about the concept of "lazy evaluation." I think > of that as saying an imperative language would always make one > evaluation, whereas Haskell might make 0 evaluations. I have > another similar situation where an imperative language would make N > evaluations of the same expression, and I would like Haskell to > make only 1. > > This is the situation: the graphical score editor displays > "LayoutItems." A LayoutItem can be a single displayed entity, like > a round notehead, or it can be composed of several entities. > > A common situation in my code is the need to determine the size and > shape of a LayoutItem. For a fundamental item, this can be looked > up in a table or read from the font properties. For a composite > item, some computation is required: the code must determine the > positions of each sub-item and compute the bounds of a shape > containing all of them. > > It's this latter computation, finding the bounds of a composite > item, which might come up multiple times. Consider that I ask for > the bounds of a composite-composite item (a composite item composed > of composite items). It will run the computation associated with > each composite sub-item, even though it is very likely I already > make that computation when I first constructed and placed that sub- > item. > > In an imperative language, one might cache values for later lookup. > This raises the problem of keeping the cache current to the current > state. > > So I'm wondering to what extent the haskell compiler recognizes > computations it's done before. In a purely functional language this > should be pretty easy, right? If it sees the same expression, it > knows it will have the same value. That's my understanding, so far. > > Thanks, > Mike > _______________________________________________ > Beginners mailing list > Beginners(a)haskell.org > http://www.haskell.org/mailman/listinfo/beginners

2 1

Re: [Haskell-beginners] appropriateness of haskell for GUIs
by Adrian Neumann 21 Mar '09

21 Mar '09

Am 21.03.2009 um 13:30 schrieb Michael Mossey: > > > Thomas Davie wrote: >> On 21 Mar 2009, at 00:16, Michael P Mossey wrote: >>> Hello, I'm totally new to Haskell. I'm thinking of using it for a >>> personal project, which is a gui-based musical score editor. >> The rough situation of GUI programming on Haskell is that it works >> just as well as in any imperative programming language. This is >> rather disappointing, simply because so many other things are >> massively easier in Haskell, and this isn't true of GUI >> programming (yet). > > Hi Bob, > > I can imagine that GUI programming is no easier (yet). It is > inherently very "stateful." GUI's have modes, such as which screens > are displayed, which dialogs are displayed, which options within > those dialogs are valid given the other state of the program, etc. > When I write GUIs, I often diagram them as state machines to get a > handle on what's going on. > > So, I'm not familiar with GUI programming on Haskell, but would you > say the statefulness of GUIs (in their typical implementations) is > the reason they are no easier on Haskell? > > I strongly prefer to use qtHaskell because I'm familiar with Qt, > and Qt is extremely capable. For example, it can draw text and > shapes with antialiasing, which will be great for a music score > editor. Music scores have lots of small shapes to fit on the > screen, and antialiasing will provide ease of reading. I don't know > how much of Qt is implemented in qtHaskell, or whether the latest > version of Qt (4.4) is implemented. > > Thanks, > Mike The main problem is, as far as I know, the complete lack of any usable GUI designer. You have to type everything yourself. That's very annoying. It's a lot easier in other languages because your tools take away the cumbersome twiddling with widgets. However, I haven't googled. Maybe the situation has changed since I last looked. Regards, Adrian

3 2

Haskell Weekly News: Issue 110 - March 21, 2009
by Brent Yorgey 21 Mar '09

21 Mar '09

--------------------------------------------------------------------------- Haskell Weekly News http://sequence.complete.org/hwn/20090321 Issue 110 - March 21, 2009 --------------------------------------------------------------------------- Welcome to issue 110 of HWN, a newsletter covering developments in the [1]Haskell community. [2]Facebook apps with Happstack, [3]Sudoku with Cryptol, what next? Tic-tac-toe with darcs? Anyway, lots of neat stuff this week, including new releases of [4]GHC, [5]jhc, and the [6]Monad.Reader, some [7]fun [8]visualizations, and more. Also, students: apply to work on a [9]Haskell project for the Google Summer of Code! Announcements GHC 6.10.2 Release Candidate 1. Ian Lynagh [10]announced the [11]first release candidate for GHC 6.10.2. Please test as much as possible; bugs are much cheaper if we find them before the release! jhc 0.6.0 Haskell Compiler. John Meacham [12]announced the release of [13]jhc 0.6.0. Safe Lazy IO in Haskell. Nicolas Pouillard [14]announced the [15]safe-lazy-io package that provides special types and combinators for performing safe lazy I/O. game-tree - a library for searching game trees. Colin Paul Adams [16]announced [17]game-tree 0.1.0.0, which provides a class for dynamic game trees, and purely functional algorithms for searching them. random-shuffle package. Manlio Perillo [18]announced the availability of the [19]random-shuffle package, which is based on [20]Oleg's description. random-stream package. Manlio Perillo [21]announced the [22]random-stream package, which provides a portable interface for the operating system source of pseudo random data. Supported sources are Unix /dev/urandom, Win32 CryptGenRandom and OpenSSL pseudo random numbers generator. language-python. Bernie Pope [23]announced the [24]language-python package, which provides a parser (and lexer) for Python, written in Haskell. Currently it only supports version 3 of Python (the most recent version), but it will support version 2 in the future. Google Summer of Code. Malcolm Wallace [25]announced that haskell.org has once again been accepted as a mentoring organisation for the 2009 Google Summer of Code. Student applications open on Monday (23rd March) at 1900 UTC, for a period of 12 days (until Fri 3rd April, also at 1900 UTC). Students applicants are encouraged to interact with the community via mailing lists, prior, during, and after the submission of their ideas for projects. Because (sadly) the darcs community did not get accepted as a separate organisation this year, haskell.org will be willing to accept proposals relating to darcs. regex-tdfa-1.1.0. ChrisK [26]announced the release of [27]regex-tdfa-1.1.0. This version is a small performance update to the old regex-tdfa-1.0.0 version. Previously all text (e.g. ByteString) being search was converted to String and sent through a single engine; the new version uses a type class and SPECIALIZE pragmas to avoid converting to String. This should make adding support for searching other Char containers easy to do. Haskell on your system? Information wanted!. Don Stewart [28]announced that haskell.org now features links to wiki pages explaining how to obtain Haskell on windows, mac osx and linux and bsd. If you're a distro maintainer for these systems, please consider adding relevant pointers to the pages, so that users of these systems can find all the info they need. libffi 0.1 released. Remi Turk [29]announced the release of [30]libffi 0.1, bindings to the C library libffi, allowing C functions to be called whose types are not known before run-time. Haskell Logo Voting has started!. Eelco Lempsink [31]announced that voting has begun to choose the new Haskell logo. All subscribed to haskell-cafe should have received a ballot; if you are not directly subscribed, you can still send ballot requests until the end of the competition (March 24, 12:00 UTC). Make sure the message contains 'haskell logo voting ballot request' in the subject. A long discussion of what color to paint the bike shed and why this particular bike shed will not do for storing bikes ensued. The Monad.Reader (13). Wouter Swierstra [32]announced that a new issue of [33]The Monad.Reader, a quarterly magazine about functional programming, is now available. Issue 13 consists of the following four articles: "Rapid Prototyping in TEX" by Stephen Hicks; "The Typeclassopedia" by Brent Yorgey; a Real World Haskell book review by Chris Eidhof and Eelco Lempsink; and "Calculating Monads with Category Theory" by Derek Elkins. dzen-utils 0.1. Felipe Lessa [34]announced the release of [35]dzen-utils 0.1, which contains various utilities for creating dzen input strings in a type-safe way using some combinators, including the ability to apply colors locally (instead of applying for everything beyond some point). It can also emulate dbar and gdbar, do automatic padding, and more. Discussion transformers versus mtl. Ganesh Sittampalam began a [36]discussion on the relative status of the 'transformers' and 'mtl' packages. least fixed points above something. Jens Blanck [37]asked about a function to compute fixed points starting from a seed value (as opposed to computing the least defined fixed point). Type equality proof. Martijn van Steenbergen [38]requested feedback on a proposed module collecting utilities for working with type equality proofs. What unsafeInterleaveIO is unsafe. Yusaku Hashimoto began a [39]discussion by asking why unsafeInterleaveIO is considered unsafe, or under what circumstances its use can be considered safe. Jobs How do students learn Haskell? Postgraduate project at University of Kent. S.J.Thompson [40]announced that funding is available for a [41]postgraduate project to study how students learn Haskell, based on the wealth of data collected through the instrumented version of the [42]Helium system for Haskell. The project will be supervised by Simon Thompson and Sally Fincher, in collaboration with Jurriaan Hage, Utrecht University. Blog noise [43]Haskell news from the [44]blogosphere. * Neil Mitchell: [45]Concise Generic Queries. Neil compares solving a query problem with several different generic programming libraries. * Jeff Heard: [46]Preview: Data Waves. A neat new type of data visualization, built using Hieroglyph. * >>> necrobious: [47]A fun example of Haskell's newtype. * GHC / OpenSPARC Project: [48]Peak issue rate is 18.64 Gig instrs/sec. * happstack.com: [49]Jeremy Shaw creates first Facebook App with Happstack. * GHC / OpenSPARC Project: [50]Peak performance. * Galois, Inc: [51]Solving Sudoku Using Cryptol. * Xmonad: [52]xmonad on ubuntu. A tutorial on getting started with xmonad on Ubuntu. * Don Stewart (dons): [53]Visualising the Haskell Universe. Pretty dependency graphs of ten thousand Haskell modules! * >>> Sean Chapel: [54]Haskell. * >>> Dean Berris: [55]The Haskell Experiment: HaskellDB, HTTP, and Monads. * Holumbus: [56]Hayoo! Update. The Hayoo! package index has been updated to include everything currently on Hackage! * >>> John Wiegley: [57]Journey into Haskell, Part 1. John ventures down the rabbit hole. * Bjorn Buckwalter: [58]Blogging with Pandoc, literate Haskell, and a bug. * Bjorn Buckwalter: [59]Extended sessions with the Haskell Curl bindings. * Brent Yorgey: [60]Monad.Reader #13 is out!. * Manuel M T Chakravarty: [61]Final version of "GPU Kernels as Data-Parallel Array Computations in Haskell".. * Xmonad: [62]Visualising xmonad. * Darcs: [63]darcs weekly news #21. * mightybyte: [64]Transactional Integrity Problem. * >>> Dean Berris: [65]The Haskell Experiment: Learning a New Programming Language. * >>> John Wiegley: [66]Hello Haskell, Goodbye Lisp. Quotes of the Week * ray: three dimensional zippers make my scalp hurt when i get my hair caught in them * dolio: [regarding a paypal spam message on #haskell] Take that, Harrop! Does OCaml have illegal cracking utilities? * lament: I think I speak for everyone in this channel when I say haskell is absolutely horrible and nobody would ever want to use it * MiguelMitrofanov: The first glimpse of this [logo] vote scared me so much that I've closed the page, stopped the browser, and shut my computer down. * osfameron: <ImInYourMonad> can I store gtk2hs-Buttons in a datastructure? <osfameron> ImInYourMonad: I think you have to sew them on with gtk2hs-Thread * chrisdone: I think you mean Peyton `Simon` Jones. About the Haskell Weekly News New editions are posted to [67]the Haskell mailing list as well as to [68]the Haskell Sequence and [69]Planet Haskell. [70]RSS is also available, and headlines appear on [71]haskell.org. To help create new editions of this newsletter, please see the information on [72]how to contribute. Send stories to byorgey at cis dot upenn dot edu. The darcs repository is available at darcs get [73]http://code.haskell.org/~byorgey/code/hwn/ . References 1. http://haskell.org/ 2. http://blog.happstack.com/2009/03/18/jeremy-shaw-creates-first-facebook-app… 3. http://www.galois.com/blog/2009/03/18/solving-sudoku-using-cryptol/ 4. http://article.gmane.org/gmane.comp.lang.haskell.glasgow.user/16506 5. http://www.haskell.org//pipermail/haskell/2009-March/021116.html 6. http://www.haskell.org/haskellwiki/The_Monad.Reader 7. http://vis.renci.org/jeff/2009/03/20/preview-data-waves/ 8. http://donsbot.wordpress.com/2009/03/16/visualising-the-haskell-universe/ 9. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55109 10. http://article.gmane.org/gmane.comp.lang.haskell.glasgow.user/16506 11. http://www.haskell.org/ghc/dist/6.10.2-rc1/ 12. http://www.haskell.org//pipermail/haskell/2009-March/021116.html 13. http://repetae.net/computer/jhc/ 14. http://article.gmane.org/gmane.comp.lang.haskell.general/16977 15. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/safe%2Dlazy%2Dio 16. http://article.gmane.org/gmane.comp.lang.haskell.general/16976 17. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/game%2Dtree 18. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55159 19. http://haskell.mperillo.ath.cx/random-shuffle-0.0.2.tar.gz 20. http://okmij.org/ftp/Haskell/perfect-shuffle.txt 21. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55143 22. http://haskell.mperillo.ath.cx/random-stream-0.0.1.tar.gz 23. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55116 24. http://projects.haskell.org/language-python/ 25. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55109 26. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55073 27. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/regex%2Dtdfa 28. http://article.gmane.org/gmane.comp.lang.haskell.cafe/55036 29. http://article.gmane.org/gmane.comp.lang.haskell.cafe/54959 30. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/libffi 31. http://article.gmane.org/gmane.comp.lang.haskell.cafe/54958 32. http://article.gmane.org/gmane.comp.lang.haskell.cafe/54880 33. http://www.haskell.org/haskellwiki/The_Monad.Reader 34. http://article.gmane.org/gmane.comp.lang.haskell.cafe/54835 35. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/dzen-utils 36. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/10807 37. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/55132 38. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/54946 39. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/54840 40. http://article.gmane.org/gmane.comp.lang.haskell.general/16967 41. http://www.cs.kent.ac.uk/research/pg/ 42. http://www.cs.uu.nl/wiki/Helium 43. http://planet.haskell.org/ 44. http://haskell.org/haskellwiki/Blog_articles 45. http://neilmitchell.blogspot.com/2009/03/concise-generic-queries.html 46. http://vis.renci.org/jeff/2009/03/20/preview-data-waves/ 47. http://necrobious.blogspot.com/2009/03/fun-example-of-haskells-newtype.html 48. http://ghcsparc.blogspot.com/2009/03/peak-issue-rate-is-1864-gig-instrssec.… 49. http://blog.happstack.com/2009/03/18/jeremy-shaw-creates-first-facebook-app… 50. http://ghcsparc.blogspot.com/2009/03/peak-performance.html 51. http://www.galois.com/blog/2009/03/18/solving-sudoku-using-cryptol/ 52. http://xmonad.wordpress.com/2009/03/18/xmonad-on-ubuntu/ 53. http://donsbot.wordpress.com/2009/03/16/visualising-the-haskell-universe/ 54. http://seanchapel.blogspot.com/2009/03/haskell.html 55. http://www.deanberris.com/mental-blabberings/2009/3/17/the-haskell-experime… 56. http://holumbus.fh-wedel.de/blog/?p=19 57. http://www.newartisans.com/2009/03/journey-into-haskell-part-1.html 58. http://flygdynamikern.blogspot.com/2009/03/blogging-with-pandoc-literate-ha… 59. http://flygdynamikern.blogspot.com/2009/03/extended-sessions-with-haskell-c… 60. http://byorgey.wordpress.com/2009/03/16/monadreader-13-is-out/ 61. http://justtesting.org/post/86905420 62. http://xmonad.wordpress.com/2009/03/16/visualising-xmonad/ 63. http://blog.darcs.net/2009/03/darcs-weekly-news-21.html 64. http://softwaresimply.blogspot.com/2008/02/transactional-integrity-problem.… 65. http://www.deanberris.com/mental-blabberings/2009/3/14/the-haskell-experime… 66. http://www.newartisans.com/2009/03/hello-haskell-goodbye-lisp.html 67. http://www.haskell.org/mailman/listinfo/haskell 68. http://sequence.complete.org/ 69. http://planet.haskell.org/ 70. http://sequence.complete.org/node/feed 71. http://haskell.org/ 72. http://haskell.org/haskellwiki/HWN 73. http://code.haskell.org/~byorgey/code/hwn/

1 0

Numeric literals
by Lauri Oksanen 21 Mar '09

21 Mar '09

Hi, Is there some way to tell ghc, how to interpret numeric literals? I would like it to interpret 1 as 1 :: Integer not 1 as fromInteger (1 :: Integer) I have been playing with the following (rather ugly) code. {-# OPTIONS -XFunctionalDependencies -XMultiParamTypeClasses -XTypeSynonymInstances #-} module Test where import Prelude(Integer, Double) import qualified Prelude as P default(Integer, Double) type Z = Integer type R = Double class Plus a b c | a b -> c where (+) :: a -> b -> c instance Plus Z Z Z where j + k = j P.+ k instance Plus R R R where x + y = x P.+ y instance Plus R Z R where x + j = x P.+ P.fromInteger j instance Plus Z R R where j + x = P.fromInteger j P.+ x x = (1 :: Z) + (1.0 :: R) --y = 1 + 1.0 The commented line can't be compiled (because of ambiguous types, I think). - Lauri

3 5

random shuffle and random list partition
by Manlio Perillo 20 Mar '09

20 Mar '09

Hi. For my Netflix Prize project I have implemented two reusable modules. The first module implements a random shuffle on immutable lists. It uses http://okmij.org/ftp/Haskell/perfect-shuffle.txt, with an additional "wrapper" function, having a more friendly interface. The second module implements a function used to partition a list into n sublists of random length. I have pasted the modules here: http://hpaste.org/fastcgi/hpaste.fcgi/view?id=2483 http://hpaste.org/fastcgi/hpaste.fcgi/view?id=2485 If someone is interested (and if Oleg give me permission), I can release them as a package on Hackage. I need to improve documentation, however. In future I can add an implementation of the random shuffle algorithm on mutable arrays in the ST monad. Manlio

2 3

least fixed points above something
by Jens Blanck 20 Mar '09

20 Mar '09

Hi, I found myself writing the following leastFixedPoint :: (Eq a) => (a -> a) -> a -> a leastFixedPoint f x = fst . head . dropWhile (uncurry (/=)) $ zip l (tail l) where l = iterate f x and was a bit surprised that I couldn't get any matches on hoogle for the type above. The closest one is fix :: (a -> a) -> a but that sort of assumes that we're starting the fixed point search from the bottom element (undefined). Anyway, is there a nicer way of doing the above? Jens

9 15

Grokking zippers
by Nathan Bloomfield 20 Mar '09

20 Mar '09

Hello all- I'm trying to understand the categorical guts underlying zippers. In the Haskell wikibook (and other places) I've seen zippers described roughly as the derivatives of functors. However, I haven't been able to find any references that develop this idea rigorously. For instance, what exactly is the "derivative" of a functor at an object, in the direction of some arrow? I'm interested in studying this concept in more depth, but I can't find a definition to start with. Any pointers to good books or papers would be greatly appreciated. :) Thank you all- Nathan Bloomfield

3 2

Re: [Haskell-cafe] Re: categories and monoids
by Kalman Noel 20 Mar '09

20 Mar '09

Wolfgang Jeltsch schrieb: > Okay. Well, a monoid with many objects isn’t a monoid anymore since a monoid > has only one object. It’s the same as with: “A ring is a field whose > multiplication has no inverse.” One usually knows what is meant with this but > it’s actually wrong. Wrong for two reasons: First, because the multiplication > of a field has an inverse. Second, because the multiplication of a ring is > not forced to have no inverse but may have one. “A ring is like a field, but without a multiplicative inverse” is, in my eyes, an acceptable formulation. We just have to agree that “without” here refers to the definition, rather than to the definitum.

6 7

what does "atomically#" mean?
by Daryoush Mehrtash 20 Mar '09

20 Mar '09

Any idea was the atomically# mean in the following code? atomically :: STM a -> IO a atomically (STM m) = IO (\s -> (*atomically# *m) s ) Code is from GHC.Conc module http://www.haskell.org/ghc/docs/6.6/html/libraries/base/GHC-Conc.html daryoush

4 5

Re: [Haskell-cafe] Ease of Haskell development on OS X?
by Jason Dusek 20 Mar '09

20 Mar '09

2009/03/20 Mark Spezzano <mark.spezzano(a)chariot.net.au>: > Question: Does the iMac have good support for Haskell > development? Yes. I generally use the Mac binary of GHC and then install C libs with MacPorts. > Question: What environment setups do people commonly use (e.g. > Eclipse Xcode etc)? I use screen and vim. In particular, I use MacVim, which provides both a vim console application and a Mac native GUI. > Question: Are there any caveats I should be aware of before > changing systems (i.e. unpleasant surprises). Mac laptops get very warm. You'll have a powerful urge to buy an iPhone and pimp Macs to your friends. There is really no 'home' or 'end' key. To move back and forth on a line in the Terminal application, use fn-shift-leftarrow and fn-shift-rightarrow (annoyed me for days). To move back and forth on a line in grapical application -- for example, the search bar in Firefox -- use cmd-leftarrow and cmd-rightarrow. MacPorts is not as robust as comparable systems on other operating systems: Gentoo's Portage and FreeBSD's Ports. It corrupts its registry and terrible things happen. The remedy is usually a total reinstall of all packages. This has happened to me only once on one Mac in a year and a half of administering several Macs; but it's not something I've ever dealt with on other source-based systems. -- Jason Dusek

1 0