June 2009 - Haskell-Cafe

Haskell Weekly News: Issue 122 - June 21, 2009
by Brent Yorgey 21 Jun '09

21 Jun '09

--------------------------------------------------------------------------- Haskell Weekly News http://sequence.complete.org/hwn/20090621 Issue 122 - June 21, 2009 --------------------------------------------------------------------------- Welcome to issue 122 of HWN, a newsletter covering developments in the [1]Haskell community. Are you ready for the [2]12th Annual ICFP programming contest? It begins this Friday, don't miss it! Let's reclaim Haskell's rightful place as the programming language of choice for discriminating hackers. Announcements Haskell protocol-buffers version 1.5.0. Chris Kuklewicz [3]announced version 1.5.0 of the [4]protocol-buffers, [5]protocol-buffers-descriptor, and [6]hprotoc packages to Hackage. This catches up to Google's version 2.1.0: support for "repeated" fields for primitive types; fields can now be marked deprecated; the type name resolver will no longer resolve type names to fields; and more. 12th Annual ICFP Contest. Mark Huntington Snyder [7]announced the 12th Annual [8]ICFP Programming Contest, hosted by the University of Kansas Computer Systems Design Laboratory at the Information and Telecommunication Technology Center. The contest will be held on the weekend of June 26-29. The contest task will be released sixteen seconds after 13:00 Central Daylight Time (US) on Friday, and entries will be accepted until 13:00:16 CDT on Monday. There is no preregistration required, and participation is free and open to all. Teams may participate from any location, and may use any programming language(s). Read the [9]contest blog or subscribe to the [10]RSS feed to receive timely updates before and during the contest. clock 0.1 released. Cetin Sert [11]announced the release of [12]clock, a package for convenient access to high-resolution clock and timer functions of different operating systems. It is planned to consist of two layers; the lower layer will provide direct access to OS-specific clock and timer functions like clock_gettime of Posix or GetTickCount of Windows, and its upper layer shall then provide a common API for all supported systems. Currently only the lower level is being developed. Turbinado V0.7. Alson Kemp [13]announced version 0.7 of [14]Turbinado, a Ruby-On-Rails-like web server and web framework for Haskell. It is designed to make creating web application using Haskell both easy and joyful. The primary additions in version 0.7 are FastCGI support and a new templating system (which includes HAML and HTML support). Additional details can be found [15]here. haskeline-class. Antoine Latter [16]announced [17]haskeline-class, a small library providing a newtyped MonadState instance for haskeline which lifts the class operations to an inner monad (as opposed to its existing instance). hyena. Johan Tibell [18]announced the first release of [19]hyena, a library for building web servers, based on the work on [20]iteratee style I/O by Oleg Kiselyov. The library allows you to create web servers that consume their input incrementally, without resorting to lazy I/O. This should lead to more predictable resource usage. Haskell-based iPhone development. Conal Elliott [21]announced a [22]collaboration wiki page for anyone working with Haskell to make iPhone apps. Fwd: Boston Haskell June 23rd meeting: openings for Lightning Talks. Ravi Nanavati [23]announced that there are several available slots for "lightning" (5 minute) talks at the June 23 meeting of the [24]Boston Area Haskell Users' Group. haskell-src-exts 1.0.0 rc1. Niklas Broberg [25]announced a series of release candidates for haskell-src-exts-1.0.0 (as of this writing, the most recent release candidate is version 0.5.6). This version is intended to fully support parsing of almost all Haskell extensions. Please help with testing! BostonHaskell: Next meeting - June 23rd at MIT CSAIL Reading Room (32-G882). Ravi Nanavati [26]announced the second meeting of the [27]Boston Area Haskell Users' Group, scheduled for Tuesday, June 23rd from 6:30pm - 8:30pm. It will be held in the MIT CSAIL Reading Room (32-G882, i.e. a room on the 8th floor of the Gates Tower of the MIT's Stata Center at 32 Vassar St in Cambridge, MA). Talks include "Automagic Font Conversion with Haskell Typeclasses" by Frank Berthold, and "Intermediate Language Representations via GADTs" by Nirav Dave. traversal transformations. Sjoerd Visscher [28]exhibited some code for Church-encoded container structures using their Foldable instance, and later [29]announced the [30]fmlist package based on the same code, along with a surprising example of a lazy 'middle-infinite' list (where elements can be taken from the beginning or the end!). hledger 0.6 released. Simon Michael [31]announced the release of [32]hledger 0.6. See the announcement for a list of the new features and other information. Discussion Adding swap to Data.Tuple. roconnor [33]proposed adding swap and swap' functions to Data.Tuple. Revamping the module hierarchy. Johan Tibell began an interesting [34]discussion about package names, module names, and the module hierarchy. Confusion on the third monad law when using lambda abstractions. Jon Strait [35]asked about the third monad law, leading to some clarification on what precisely the law says, and some interesting discussion on idiomatic use of the (<=<) (Kleisli composition) operator. Need some help with an infinite list. Gunther Schmidt [36]asked for some help generating a particular infinite list, and got a number of interesting suggestions. Blog noise [37]Haskell news from the [38]blogosphere. Blog posts from people new to the Haskell community are marked with >>>, be sure to welcome them! * Thomas ten Cate: [39]Cosmetics. Nice-looking icons for EclipseFP! * Niklas Broberg: [40]GSoC status report, week 4. More release candidates for haskell-src-exts 1.0.0. * >>> Uwe Hoffmann: [41]publishing nike runs, part 4: string templates. Real-world example of using HStringTemplate. * Andy Gill: [42]Call for Participation in the 12th Annual ICFP Programming Contest!. June 26-29! * Sebastian Fischer: [43]Reinventing Haskell Backtracking. * Remco Niemeijer: [44]Programming Praxis - Monte Carlo factorization. Remco implements Pollard's factorization algorithm in 9 lines of Haskell. * >>> Lee Duhem: [45]Understanding Functions Which Use 'instance Monad []' by Equational Reasoning. * Alex McLean: [46]Patterns in Haskell. A Haskell music generation EDSL. * David Amos: [47]Group generators for graph symmetries. * >>> adam: [48]Experience writing a ray tracer in Haskell. Adam's final project in a Haskell class taught by Mark Jones and Tim Sheard. * Petr Rockai: [49]soc progress 4. * Yaakov Nemoy: [50]Haskell Bindings to C from Start to Finish. Yaakov outlines his experience getting c2hs and the FFI to work. * Alex McLean: [51]Patterns in Haskell. Representing rhythmic patterns in Haskell. * >>> Abhishek Tiwari: [52]Haskell for Bioinformatics. * Roman Cheplyaka: [53]Shootout. A hilarious comic featuring sound advice on Haskell optimization. * Ketil Malde: [54]Dephd updates. * Neil Mitchell: [55]Draft paper on Derive, comments wanted. * Remco Niemeijer: [56]Programming Praxis - Who Owns The Zebra?. * Erik de Castro Lopo: [57]Two More for the Debian New Queue.. * David Amos: [58]Graph symmetries. * Alson Kemp: [59]Announce: Turbinado V0.7. * Gergely Patai: [60]You can draw your own graphs now!. * >>> Jens Petersen: [61]Haskell cabal-install rocks . Quotes of the Week * Botje: <Cheery> oh man. de bruijn again kicked me to groin <Botje> the easy fix is to label your groin as (-1) :) * Pseudonym: Telling dons that something has been added to the shootout is the new telling Oleg that it can't be done in the type system. About the Haskell Weekly News New editions are posted to [62]the Haskell mailing list as well as to [63]the Haskell Sequence and [64]Planet Haskell. [65]RSS is also available, and headlines appear on [66]haskell.org. To help create new editions of this newsletter, please see the information on [67]how to contribute. Send stories to byorgey at cis dot upenn dot edu. The darcs repository is available at darcs get [68]http://code.haskell.org/~byorgey/code/hwn/ . References 1. http://haskell.org/ 2. http://icfpcontest.org/ 3. http://article.gmane.org/gmane.comp.lang.haskell.libraries/11331 4. http://hackage.haskell.org/package/protocol-buffers 5. http://hackage.haskell.org/package/protocol-buffers-descriptor 6. http://hackage.haskell.org/package/hprotoc 7. http://article.gmane.org/gmane.comp.lang.haskell.general/17298 8. http://icfpcontest.org/ 9. http://www.icfpcontest.org/wordpress 10. http://www.icfpcontest.org/wordpress/?feed=rss2 11. http://article.gmane.org/gmane.comp.lang.haskell.general/17279 12. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/clock 13. http://article.gmane.org/gmane.comp.lang.haskell.general/17277 14. http://wiki.github.com/turbinado/turbinado 15. http://www.alsonkemp.com/turbinado/announce-turbinado-v07/ 16. http://article.gmane.org/gmane.comp.lang.haskell.cafe/60246 17. http://hackage.haskell.org/package/haskeline-class 18. http://article.gmane.org/gmane.comp.lang.haskell.cafe/60117 19. http://hackage.haskell.org/package/hyena 20. http://hackage.haskell.org/package/iteratee 21. http://article.gmane.org/gmane.comp.lang.haskell.cafe/60104 22. http://haskell.org/haskellwiki/IPhone 23. http://article.gmane.org/gmane.comp.lang.haskell.cafe/60091 24. http://www.haskell.org/haskellwiki/Boston_Area_Haskell_Users%27_Group 25. http://article.gmane.org/gmane.comp.lang.haskell.cafe/59993 26. http://article.gmane.org/gmane.comp.lang.haskell.cafe/59991 27. http://www.haskell.org/haskellwiki/Boston_Area_Haskell_Users%27_Group 28. http://article.gmane.org/gmane.comp.lang.haskell.cafe/59894 29. http://article.gmane.org/gmane.comp.lang.haskell.cafe/60120 30. http://hackage.haskell.org/cgi-bin/hackage-scripts/package/fmlist 31. http://www.haskell.org/pipermail/haskell-cafe/2009-June/062787.html 32. http://hledger.org/ 33. http://article.gmane.org/gmane.comp.lang.haskell.libraries/11326 34. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/11325 35. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/60105 36. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/59999 37. http://planet.haskell.org/ 38. http://haskell.org/haskellwiki/Blog_articles 39. http://eclipsefp.wordpress.com/2009/06/20/cosmetics/ 40. http://nibrofun.blogspot.com/2009/06/gsoc-status-report-week-4.html 41. http://www.codemanic.com/uwe/2009/06/publishing-nike-runs-part-4-string-tem… 42. http://blog.unsafeperformio.com/?p=53 43. http://www-ps.informatik.uni-kiel.de/~sebf/pub/atps09.html 44. http://bonsaicode.wordpress.com/2009/06/19/programming-praxis-monte-carlo-f… 45. http://leeduhem.wordpress.com/2009/06/19/understanding-functions-which-use-… 46. http://yaxu.org/patterns-in-haskell/ 47. http://haskellformaths.blogspot.com/2009/06/group-generators-for-graph-symm… 48. http://blog.finiteimprobability.com/2009/06/18/experience-writing-a-ray-tra… 49. http://web.mornfall.net/blog/soc_progress_4.html 50. http://loupgaroublond.blogspot.com/2009/06/haskell-bindings-to-c-from-start… 51. http://yaxu.org/patterns-in-haskell/ 52. http://www.abhishek-tiwari.com/2009/06/haskell-for-bioinformatics.html 53. http://ro-che.blogspot.com/2009/06/shootout.html 54. http://blog.malde.org/index.php/2009/06/16/dephd-updates/ 55. http://neilmitchell.blogspot.com/2009/06/draft-paper-on-derive-comments-wan… 56. http://bonsaicode.wordpress.com/2009/06/16/programming-praxis-who-owns-the-… 57. http://www.mega-nerd.com/erikd/Blog/CodeHacking/Debian/safe_uniplate.html 58. http://haskellformaths.blogspot.com/2009/06/graph-symmetries.html 59. http://www.alsonkemp.com/turbinado/announce-turbinado-v07/ 60. http://just-bottom.blogspot.com/2009/06/you-can-draw-your-own-graphs-now.ht… 61. http://juhp.blogspot.com/2009/06/haskell-cabal-install-rocks.html 62. http://www.haskell.org/mailman/listinfo/haskell 63. http://sequence.complete.org/ 64. http://planet.haskell.org/ 65. http://sequence.complete.org/node/feed 66. http://haskell.org/ 67. http://haskell.org/haskellwiki/HWN 68. http://code.haskell.org/~byorgey/code/hwn/

1 0

Re: [Haskell-cafe] Slightly off-topic: Lambda calculus
by Keith Sheppard 21 Jun '09

21 Jun '09

forgot to cc the cafe :-) On Sun, Jun 21, 2009 at 1:08 PM, Keith Sheppard<keithshep(a)gmail.com> wrote: > hmm, it's been a while but... > > i think this "infinite loop" with a free variable would cause collision > > (\a . a a) (\b . b b d) > > On Sun, Jun 21, 2009 at 12:53 PM, Andrew > Coppin<andrewcoppin(a)btinternet.com> wrote: >> OK, so I'm guessing there might be one or two (!) people around here who >> know something about the Lambda calculus. >> >> I've written a simple interpretter that takes any valid Lambda expression >> and performs as many beta reductions as possible. When the input is first >> received, all the variables are renamed to be unique. >> >> Question: Does this guarantee that the reduction sequence will never contain >> name collisions? >> >> I have a sinking feeling that it does not. However, I can find no >> counter-example as yet. If somebody here can provide either a proof or a >> counter-example, that would be helpful. >> >> _______________________________________________ >> Haskell-Cafe mailing list >> Haskell-Cafe(a)haskell.org >> http://www.haskell.org/mailman/listinfo/haskell-cafe >> > > > > -- > keithsheppard.name > -- keithsheppard.name

1 1

Re: [Haskell-cafe] Obscure weirdness
by Raynor Vliegendhart 21 Jun '09

21 Jun '09

On 6/20/09, Andrew Coppin <andrewcoppin(a)btinternet.com> wrote: > > Is this a known bug in GHC 6.10.1? Will upgrading fix it? (Obviously, it's > quite a lot of work to change GHC.) Suffice it to say that my program is > quite big and complicated; it worked fine when it was still small and > simple. ;-) There is a bug in ghci 6.10.1 that seems to be fixed in 6.10.3 (not sure whether it's fixed in 6.10.3). Certain non-terminating expressions causes ghci to crash immediately and go back to the prompt: ---------------------------------------- C:\>ghci GHCi, version 6.10.1: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. Prelude> let loop = loop in loop C:\> ---------------------------------------- Compiling a module with a non-terminating expression first with ghc and then loading it in ghci does not seem to result in a crash.

2 2

Edinburgh Haskell users meeting this afternoon
by Dougal Stanton 21 Jun '09

21 Jun '09

Hello Haskell users If you're in the Edinburgh (Scotland!) area and interested in Haskell we may have just the thing for you. I will be meeting one other Haskell Cafe list member at Doctor's pub at 4pm this afternoon (Sunday 21 June). This could be the start of something bigger, so if you're interested in Haskell please drop in and swell the numbers. The conversation will probably be about some of the basics - what is currying? what are typeclasses? what does the "deriving" keyword do? - so don't feel nervous about being out of your depth! If I can get a table, I will try to display a copy of Hudak's _Haskell School of Expression_ prominently. Please come up and say hello, or fire me back a note if you can't make it but are still interested in making this a regular meeting. Don't be shy about joining us if you're available though! Thanks everyone! Dougal -- Dougal Stanton dougal(a)dougalstanton.net // http://www.dougalstanton.net

1 0

Need some help with an infinite list
by Günther Schmidt 21 Jun '09

21 Jun '09

Hi guys, I'd like to generate an infinite list, like ["a", "b", "c" .. "z", "aa", "ab", "ac" .. "az", "ba", "bb", "bc" .. "bz", "ca" ...] When I had set out to do this I thought, oh yeah no prob, in a heartbeat. Uhm. Help, pls! Günther PS: I know this should be a no-brainer, sry

16 30

ANN: haskeline-class
by Antoine Latter 20 Jun '09

20 Jun '09

I'd like to announce a small library newly on hackage: haskeline-class http://hackage.haskell.org/package/haskeline-class Haskeline is an easy to use library, with a reasonable interface. The one quirk that got in my way was that its 'InputT' monad transformer defined its own instance of MonadState which allowed the user of the library to peer into the guts of Haskeline. What I really wanted at the time was an instance of MonadState which lifted the class operations to an inner monad. So I wrote a small newtype wrapper and a support class to work around this. If anyone else has the same issue, the code is now on Hackage. I've heard that this won't even be needed in the next major release of Haskeline, but I'll likely keep this package up-to-date if anyone finds the MonadHaskeline class useful. Antoine

1 0

Obscure weirdness
by Andrew Coppin 20 Jun '09

20 Jun '09

OK, so here's an interesting problem... I've been coding away all day, but now my program is doing something slightly weird. For a specific input, it summarily terminates. The registered exception handler does not fire. There is no output to stdout or stderr indicating what the problem is. It just *stops* half way through the printout. Weirder: If I run it in GHCi, then GHCi itself terminates. (I didn't think you could *do* that!) It's not as if my program is anything unusual. There are no unsafe functions. No FFI. Nothing. Just regular high-level Haskell. Is this a known bug in GHC 6.10.1? Will upgrading fix it? (Obviously, it's quite a lot of work to change GHC.) Suffice it to say that my program is quite big and complicated; it worked fine when it was still small and simple. ;-)

5 6

unique identity and name shadowing during type inference
by Geoffrey Irving 20 Jun '09

20 Jun '09

Hello, I am designing a type inference algorithm for a language with arbitrary function overloading. For various reasons (beyond the scope of this email), it's impossible to know the full type of an overloaded function, so each function is assigned a unique primitive type and the inference algorithm gradually learns more information about the primitive. For example, if we declare an identity function f x = x the algorithm will create a primitive type F, and record f :: F. If we use the function a few times, f 1 f "blah" the algorithm will infer F Int = Int F String = String My question is: what's the best way to represent these unique primitive types in Haskell? A new type primitive needs to be created whenever we process a function declaration. Nested function declarations produce a different primitive each time the parent is invoked with different argument types. These separate primitives can escape if local functions are returned, so the inference algorithm must be able to keep them separate and learn more about them after their parent function is forgotten. Here are a few ways I know of: 1. Thread a uniqueness generator monad through the whole algorithm. I'd prefer to avoid this extra plumbing if possible. 2. Label primitives with the full context of how they were created. If function f declares a nested function g, and f is called with Int and Char, the primitives for g would be labeled with "f Int" and "f Char" to keep them separate. This is similar to lambda lifting. 3. Scary hacks involving makeStableName and unsafePerformIO. Some sort of context would have to be thrown around here to make sure GHC doesn't merge the different makeStableName calls. Unfortunately, method (2) is complicated by the fact that variable names are not unique even in the internal representation (I'm using the trick from [1]), so I'm not sure what the minimal unique "context" would be. Does anyone know other methods outside of (1), (2), or (3), or clean ways of structuring (2) or (3)? Thanks! Geoffrey [1]: http://www.haskell.org/~simonmar/bib/ghcinliner02_abstract.html

2 1

Re: Wiki user accounts
by Gwern Branwen 20 Jun '09

20 Jun '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On Mon, Jun 15, 2009 at 1:18 PM, Ashley Yakeley wrote: > For requesting accounts: who would receive the email, and which person > would create the account? > Why not just list everyone's email and let the requester pick who to send the request to? - -- gwern -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREKAAYFAko2ik8ACgkQvpDo5Pfl1oK4fwCeJvNQpc0ubJl0AmfDebmudxUq UrEAn3rwVLmLRl1RIcJLpZET6SjrssYD =Mhng -----END PGP SIGNATURE-----

5 12

Tree Semantics and efficiency
by Rouan van Dalen 20 Jun '09

20 Jun '09

Hi everyone. I would like to confirm the following semantics. I want a tree which can be traversed both downwards and upwards. To do this i need to store the following for each node: o) a reference to the parent node (so we can traverse up the tree). This must be a reference as i dont want to copy the parent node each time) o) a list of child nodes that belong to this node. It is important to store only a reference to the parent and not a copy of the entire parent for efficiency. How would one write this in Haskell so that it has the above mentioned semantics. Or does GHC do that automatically for me so I don't need to do it explicitly? I am thinking of writing something like this (naive implementation) : Tree = TreeNode String Tree [Tree] -- TreeNode <data> <ParentReference> <ChildNodes> OR (implementation using IORef for parentNode reference) Tree = TreeNode String (IORef Tree) [Tree] -- TreeNode <data> <ParentReference> <ChildNodes> Thanks in advance Rouan.

5 4