- Libraries - Haskell.org

time bounds for BinaryRandList (was ANN: Edison 1.2RC4)
by Okasaki, C. DR EECS 02 Jun '06

02 Jun '06

Jim Apple wrote: > the source for BinaryRandList at > http://www.eecs.tufts.edu/~rdocki01/edison/_darcs/current/edison-core/sr c/Data/Edison/Seq/BinaryRandList.hs > , I don't think the constant time bounds in the documentation are > correct for this sequence type. Ross Patterson replied: > Indeed not. If you have a sequence of exactly 2^n elements and then > alternate between tail and cons, each operation will cost O(log n). Yes, for this implementation, head and tail cost O(log n). However, using amortization, you can still count cons as O(1), using the usual arguments to show that incrementing a binary number takes O(1) amortized time. -- Chris

1 0

xml in fptools?
by Isaac Jones 02 Jun '06

02 Jun '06

Greetings, I'm wondering if we've ever discussed putting an XML parser into fptools. I'm asking because cabal-install would like to use xml-rpc, but it drags in a bunch of dependencies that aren't in fptools. What do other languages do: * Python seems to have XML and XML-RPC as standard libraries * Same with ruby * Not java (shocking): http://java.sun.com/j2se/1.3/docs/api/ I guess that HaXml probably can't get into fptools because of license incompatibilities, but hxml and HXmlToolbox both have MIT licenses, and so could probably be put into fptools: http://www.flightlab.com/~joe/hxml/ http://www.fh-wedel.de/~si/HXmlToolbox/ What do others think? peace, isaac

13 28

Cabal 1.1.4 + Haddock 0.7 + GADT = Headache
by Robert Dockins 29 May '06

29 May '06

Hello all, I've got a problem with a package I maintain. I have a Cabal package which contains one module which uses GADTs. This module is hidden. Now, Haddock barfs if you try to generate Haddock docs from a module containing GADTs. With Cabal 1.1.3, hidden modules (apparently...) are not parsed by Haddock at all, so it has no opportunity to barf on my GADTs, and I could get useable docs out. However, with 1.1.4, Cabal now tells Haddock to parse this file, but also gives it the -- hide option. Haddock subsequently refuses to generate any documentation. I can remove the hidden modules from the "other modules" section, but then the object code doesn't end up in the libraries and programs won't build. Aggh! Is there any way to fix this problem? Rob Dockins Speak softly and drive a Sherman tank. Laugh hard; it's a long way to the bank. -- TMBG

3 3

Parsing Packets with House
by Dominic Steinitz 29 May '06

29 May '06

Now that I have the Haskell libpcap binding working, I want to be able to parse the packets that I capture. How feasible would it be to use the modules in House (programatica/hOp/kernel/Net) standalone? For example, here is a capture of a packet showing the 14-byte ethernet header followed by the IP packet which in this case is an ICMP echo request. I assume I can somehow use Net.Ethernet, Net.IPv4 and Net.ICMP to parse the packet? Dominic. [0,16,220,204,93,240,0,48,5,177,121,178,8,0, 69,0,0,84,104,184,0,0,255,1,9,54,192,168,100,55,192,168,100,50, 8,0,136,151,164,247,0,0, 68,120,161,81, 0,0,249,163,8,9,10,11,12,13,14,15,16,17,18,19, 20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35, 36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51, 52,53,54,55]

3 4

Experimental Support to build OpenAL package separately from GHC.
by shelarcy 25 May '06

25 May '06

Hello. I know Ross Paterson added Setup file to Win32 paclage, and added files that is used by configure for ALUT and OpenAL by these messages. http://www.haskell.org//pipermail/cvs-libraries/2006-May/005006.html http://www.haskell.org//pipermail/cvs-libraries/2006-May/005029.html http://www.haskell.org//pipermail/cvs-libraries/2006-May/005032.html So I try to add Setup file and build OpenAL package, separately from GHC. Of course, autoreconf command success alone package, and can build OpenAL package. I send patch to build OpenAL package separately from Compliter and Interpreter. This patch includes Mac OS X platform support - I sent previous mail - but I changed two points from this mail. http://www.haskell.org//pipermail/libraries/2006-March/005102.html 1. Darcs HEAD OpenAL package's Sound.OpenAL.ALC.Capture can be bult under Mac OS X platform. So, I remove unnecessary #ifdef. 2. This patch add [AC_REQUIRE([AC_CANONICAL_TARGET]) same as other packages - ALUT, OpenGL, GLUT. But OpenAL didn't use this, so forgot that configure.ac's AC_SUBST([CALLCONV]) doesn't set CALLCONV=ccall on Windows - if configure.ac works correctly. And break this convention, by previouse patch. So I commented out *-mingw32) case for that OpenAL uses ccall, even on Windows. http://www.haskell.org//pipermail/libraries/2005-October/004488.html If you applied this patch by darcs apply comannd - like this; $ darcs apply openal-separate-build.patch then you can build and install OpenAL package separately, by below command. $ autoreconf $ runhaskell Setup.lhs configure $ runhaskell Setup.lhs build $ runhaskell Setup.lhs install I tested to build on Windows and Mac OS X platform. And If you copy Setup.lhs for ALUT package, then you can also build and install by that. I think OpenGL and GLUT, too. But ALUT also require to use ccall, I think. If you build ALUT without any chage, to build example cause undefined reference error. -- shelarcy <shelarcy capella.freemail.ne.jp> http://page.freett.com/shelarcy/

1 1

Interfacing to OLE/COM: HaskellDirect the only game in town?
by Alistair Bayley 25 May '06

25 May '06

I want to use some MS Windows OLE/COM libraries from Haskell, and this page: http://www.haskell.org/haskellwiki/Libraries_and_tools/Interfacing_other_la… and Google suggest that HaskellDirect is the only game in town. Is this the case? And, does anyone have recent experience of it? Is it still usable, or has it bit-rotted? Alistair

3 2

System.Time
by Ashley Yakeley 25 May '06

25 May '06

Now that the "time" package is in current, can I get rid of System.Time, or should I deprecate it in some way? -- Ashley Yakeley, Seattle WA

3 2

(no subject)
by Bjorn Bringert 22 May '06

22 May '06

I'm moving this thread over to the libraries list. For those new to the thread, it is about replacing the current Network.CGI module with the NewCGI library. On May 22, 2006, at 8:06 AM, Lemmih wrote: > On 5/22/06, Simon Marlow <simonmar(a)gmail.com> wrote: >> On 5/22/06, Lemmih <lemmih(a)gmail.com> wrote: >> >> > NewCGI already supports POST data. >> > http://www.cs.chalmers.se/~bringert/darcs/haskell-cgi/doc/ >> >> Slightly off-topic, but maybe NewCGI should take over as Network.CGI? >> The current implementation isn't maintained, AFAIK. > > I volunteer to do the actual bone work if consensus is reached. I'd also be happy to do whatever work is needed for this. I'm still actively maintaining NewCGI. Here is a list of current features: - Access to CGI parameters (e.g. form input) from both GET and POST requests. - Access to CGI environment variables. - Ability to set arbitrary response headers. - Support for HTTP cookies. - An efficient implementation of multipart/form-data using Data.ByteString. This allows for efficient handling of file uploads. - Wrapper functions for compatibility with the existing Network.CGI module. - A CGI monad transformer. - Basic exception handling and logging (these should be improved) - Low-level run functions that allow using NewCGI programs with protocols other than CGI, for example FastCGI (http:// www.cs.chalmers.se/~bringert/darcs/haskell-fastcgi/doc/) /Björn

1 1

Announce: HGAL 1.2
by Jean-Philippe Bernardy 21 May '06

21 May '06

Hello folks, This is an announcement for HGAL 1.2. http://users.skynet.be/jyp/HGAL/ HGAL (Haskell Graph Automorphism Library) is a haskell implementation of Brendan McKay's algorithm for graph canonic labeling and automorphism group. (aka Nauty) Here are the main improvements over the previous release: * The algorithm is now more completely implemented and therefore a lot faster; * The library is now a cabal package. I'm also proud to proclaim that zero bug has been reported in the 3 years of existence of the library. (That either being due to a high quality of the library or a low number of users -- I'm not sure :) Cheers, JP.

1 0

Types vs. Classes
by Ashley Yakeley 20 May '06

20 May '06

Occasionally in library proposals one comes across classes of this form: class Thingy a where foo :: a -> this bar :: a -> that spong :: a -> theotherthing Such classes can be replaced by data-types, which turn out to also be more general: data Thingy = MkThingy { foo :: this, bar :: that, spong :: theotherthing } My question: is there a reason not to use types? I'm particularly interested in two cases that I've come across, references and streams. Here's the class version: class (Monad m) => Ref m r | r -> m, m -> r where newRef :: a -> m (r a) readRef :: r a -> m a writeRef :: r a -> a -> m () class (Monad m) => Stream m h where vPutStrLn :: h -> String -> m () vGetContents :: h -> m String vIsEOF :: h -> m Bool vClose :: h -> m () -- etc. And here's the type version: data Ref m a = MkRef { readRef :: m a, writeRef :: a -> m () } class (Monad m) => HasRefs m where newRef :: a -> m (Ref m a) data Stream m = MkStream { vPutStrLn :: String -> m (), vGetContents :: m String, vIsEOF :: m Bool, vClose :: m (), -- etc. } Is there any reason not to prefer using types? -- Ashley Yakeley, Seattle WA WWEWDD? http://www.cs.utexas.edu/users/EWD/

3 2