- Haskell - Haskell.org

Exists any null instruction in Haskell?
by L. J. 26 Aug '06

26 Aug '06

Hi, I want to know if it exists a null instruction (a instruction that do anything, for stop the recursivity of a function, for exemple) in Haskell. Thanks you.

1 0

Re: [Haskell] Haskell / Full-fledge verified OS
by William DUCK 26 Aug '06

26 Aug '06

Hello all, Linus Torvalds said in 2000 : "Talk is cheap. Show me the code." (2000) I would want to say : "Talk is cheap. Show me the formal verification." (2006) Best Regards, Will

1 0

Haskell / Full-fledge verified OS
by William DUCK 26 Aug '06

26 Aug '06

Hello Misters, As you are iin functional programming, I am asking your help. I am attempting to build up a team. I believe it's time for a full-fledge verified OS. http://www.ertos.nicta.com.au/publications/papers/Tuch_KH_05.pdf Here are my guidelines : http://www.cse.ogi.edu/~hallgren/House/ *OS-design : http://www.ertos.nicta.com.au/publications/papers/Leslie_06.pdf ( for transition ) http://www.marcus-brinkmann.org/hurd-ng.pdf http://www.eecg.toronto.edu/~tornado/ ( scalability / hot swapping ) *kernel : http://os.inf.tu-dresden.de/L4/L4.Sec/ http://www.doclsf.de/papers/vstte06.pdf#search=%22formalising%20high%20perf… *network stack : http://www.cl.cam.ac.uk/~pes20/Netsem/ *programming : http://fling-l.seas.upenn.edu/~plclub/cgi-bin/poplmark/index.php?title=The_… http://www.informatik.uni-bonn.de/~loeh/GFP.html http://maude.cs.uiuc.edu/tools/scc/ *proving environment : http://isabelle.in.tum.de/ http://www.cl.cam.ac.uk/Research/HVG/HOL/ *compiler : http://pauillac.inria.fr/~xleroy/compcert-backend/ http://www.score.is.tsukuba.ac.jp/~okuma/vc/ Any suggestions ? Thank you for your answer, If you want to contact me, my mail is guillaume_dot_fortaine_at_wanadoo_dot_fr I will set up a mailing-list, a web server, a wiki and an IRC Best Regards, Best Regards, Guillaume FORTAINE

2 1

ANNOUNCE : HSPClientside 0.2
by Joel Björnson 25 Aug '06

25 Aug '06

I'm pleased to announce a new version of HSPClientside (0.2) ,developed as a GSoC project during this summer. HSPClientside is a Haskell Server Pages library for generating JavaScript code. Some features include. * An embedding of (typed) JavaScript language in Haskell. * Implementation of an interface to (parts of) XML/HTML DOM. * An embedded language (HJScript) , with combinatiors for producing blocks of JavaScript. * High-level interface to Ajax functionality. My desire is that it should be easily extendable with all sorts of more special purposed JavaScript functionality. Even though the GSoC is now officially over, I will continue working on the project as time occurs. I would also like to invite as many of you as possible to join ! At last, a big thank you to all (Google, Haskell.org, Niklas Broberg etc..) who made this project possible (and turned my summer into a Haskell coding session :-) ) . For more information see www.dtek.chalmers.se/~bjornson/soc/<http://www.dtek.chalmers.se/%7Ebjornson/soc/> Darcs repository is available at darcs.haskell.org/SoC/hsp.clientside/ /Joel

1 0

Graphics.SOE.Gtk
by Duncan Coutts 23 Aug '06

23 Aug '06

Due to popular demand, the new SOE implementation based on Gtk2Hs is now available separately: http://haskell.org/~duncan/soegtk-0.9.10.tar.gz http://haskell.org/~duncan/soegtk-0.9.10.zip or darcs get http://haskell.org/~duncan/soegtk/ It works with the latest released version of Gtk2Hs, ie version 0.9.10 It works at least on Linux and Windows. It's packaged with Cabal. The rendering quality is better than the original HGL version. Here's a side-by-side comparison: http://haskell.org/~duncan/gtk2hs/SOE-cairo.png and it works with GHC on Windows, which the original doesn't. Sadly it doesn't work with Hugs98 (but I hear that the HGL version does work with Hugs98). There may be some slight differences in the semantics from the original. Please test it with any SOE examples you've got. If you find any mismatches, then report them as bugs here: http://hackage.haskell.org/trac/gtk2hs/newticket Enjoy. Duncan

1 0

ANNOUNCE: The experimental GHCi debugger
by pepe 22 Aug '06

22 Aug '06

Dear all It is that time of the year again. Google SoC has come to an end and a flurry of students announce their results to the corresponding community. Yesterday Jun Mukai reported on the status of HaskellNet and now it is the turn of the GHCi debugger project. The contributions of this project are mainly two: - Intermediate value viewer: The new ':print' and ':sprint' in GHCi allow to explore semievaluated values without forcing them, accessing the internals and recovering the types. - Dynamic breakpoints: The new ':breakpoint' command integrates seamlessly in GHCi and allows to set breakpoints in your code Which add to previously existing work polished as part of this project too: - The breakpoint and breakpointCond functions in GHC.Base allow to manually install (conditional) breakpoints in your code, which will be activated when you run your program in GHCi DETAILS Though it needs an update, you can check the wiki dump page at: http://haskell.org/haskellwiki/GHC/GHCiDebugger In it you will find a lot of outdated stuff, but also some interesting things: * some examples of the closure viewer * how to get the patches and build it * A link to the Users Guide documentation page for GHCi that now includes a section for the debugger (I know it does not render well in all the browsers out there, but this is just a snapshot) * A video! Watch me debug a Haskell Universal Machine (I know I know, it's not so impressing, and sorry for the poor audio. Macbooks don't seem to like cheap mics so I had to use a phone bluetooth hands-free) FUTURE WORK Overall, this version is a basic, "the simplest thing that could possibly work" implementation. Call stack traces have unhappily not made it into this version, but there are some proposals already and the GHC people seem very interested in getting themselves a feature complete debugger, so you should not lose hope. Future work should include exploring more advanced realizations of dynamic breakpoints. ACKNOWLEDGEMENTS Big thanks to Google for making this project possible. Big thanks to my mentors too, David Himmelstrup and Simon Marlow, for accepting this role. I know how busy they are and I hope I have not been too big of a burden during these months. As a colofon to the project, odds are that the GHC people will include my patches in the 6.6 Release Candidate at the end of the week. Needless to say, I will be most interested in hearing from any issues you may find, especially if you try it before the Release Candidate date. Please let me know of any additional feature requests you have too, or better add them to the wiki where other people will see them too. So please, if you feel like doing it, give it a go before the Release Candidate and let me know if you found any issue! pepe

1 0

status report of HaskellNet
by Jun Mukai 22 Aug '06

22 Aug '06

Dear all, HaskellNet is develped under GSoC program, and I'd like to report the current status of HaskellNet. HaskellNet will be a collection library for networking. Now, it contains, - SMTP - POP3 - IMAP - HTTP - (FTP) - URI - Mime parser and, I wrote a Stream like type class of ByteString. I also wrote the example code to connect these libraries with hsgnutls. Because FTP client is came from missingh, it is licensed under LGPL (Thanks to John Goerzen). Others includeing HTTP (came from Bjorn Bringert's code) are licensed under BSD3. You can obtain the codes from http://darcs.haskell.org/SoC/haskellnet (add .lgpl to get with FTP client). The period of SoC will end soon, but HaskellNet is now under development. Although the current codes have (maybe) no problems, they can not be tested fully. In addition to it, I want to add other protocols and other ideas. Actually, There are discussions between me and Alexander Jacobson about importing Protocol.DNS from HAppS. But, I did not have the time to do it (Although it can be done shortly). I mean, I'd like to keep the development of HaskellNet. Any comments are welcome. Best Regards, Jun Mukai mukai(a)jmuk.org

2 1

ANN: SmallCheck 0.0 another lightweight testing tool
by Colin Runciman 22 Aug '06

22 Aug '06

I have written a prototype tool that is similar in spirit, and in some of its workings, to QuickCheck, but based on exhaustive testing in a bounded space of test values. Sales pitch: wouldn't you like to ... * write test generators for your own types more easily? * be sure that any counter-examples found are minimal? (##) * write properties using existentials as well as universals? * establish complete coverage of a defined test-space? * display counter-examples of functional type? (##) * guarantee repeatable test results? (##) For more details and a Haskell 98 module SmallCheck download the small tar file at: http://www.cs.york.ac.uk/fp/smallcheck0.0.tar If you try it, do let me know how you get on. Comments & suggestions welcome. Colin R -------- (##) There have been versions of QuickCheck with extras that address these issues in other ways.

1 0

[ANNOUNCE] Frisby: composable, linear time parser for arbitrary PEG grammers
by John Meacham 22 Aug '06

22 Aug '06

Frisby is an implementation of the 'packrat' parsing algorithm, which parse PEG grammars and have a number of very useful qualities, they are a generalization of regexes in a sense that can parse everything in LL(k), LR(k), and more, including things that require unlimited lookahead, all in guarenteed linear time. There is no need for a separate lexing pass when using a packrat parser due to the subsumpion of regular expressions, and the unlimited lookahead obviating the need to tokenize the input. The work is still alpha, but has had some very promising test results, parsing (subsuming lexing) megabytes of test input in seconds that traditional backtracking parsers completely fail at. Although the basic mechanism works, the API is still in flux, at the moment, parsers are specified almost exactly in their native grammar form. here is an example of a simple parser and evaluator for numeric expressions. additive = mdo additive <- mkRule $ multitive <> char '+' ->> additive ## uncurry (+) // multitive multitive <- mkRule $ primary <> char '*' ->> multitive ## uncurry (*) // primary primary <- mkRule $ char '(' ->> additive <<- char ')' // decimal decimal <- mkRule $ many1 (oneOf ['0' .. '9']) ## read return additive // is choice, it attempts its left argument, and if it fails, it trys its right one. <> is concatination <<- and ->> are concatination that discard the results of one side or the other and ## is jus a fancy way of writing 'fmap' to process the results. notice the use of 'mdo' to implement monadic recursion. the home page and more information is here: http://repetae.net/computer/frisby/ I would greatly welcome feedback on the API. I have included routines that take standard regular expression syntax and translate them into an equivalent frisby parser. John -- John Meacham - ⑆repetae.net⑆john⑈

1 0

ANNOUNCE: GHC's new support engineer: Ian Lynagh
by Simon Marlow 21 Aug '06

21 Aug '06

We are delighted to announce that GHC now has a full-time support engineer. He is Ian Lynagh (aka Igloo on IRC), and is well known to many of you on the Haskell and GHC mailing lists. He'll be helping with all aspects of GHC, especially release management, bug diagnosis and tracking, documentation, packaging, and supporting other GHC hackers. Welcome Ian! We were absolutely delighted by both the number and quality of the people who applied for the post, and we want to thank them all very much for doing so. We love you all! Andrew Herbert, the lab director of Microsoft Research Cambridge lab director, also deserves a big thank-you, for supporting this post (no strings attached!). It'll make a huge difference. Looking forward to seeing many of you soon, at the GHC Hackathon [1] (25 people registered), and/or at ICFP in Portland. Simon Peyton Jones and Simon Marlow [1] http://hackage.haskell.org/trac/ghc/wiki/Hackathon

1 0