June 2006 - Libraries

Cabal package feature request [was: Streams 0.1e released]
by Robert Dockins 08 Jun '06

08 Jun '06

[thread hijacked and moved to libraries] On Jun 8, 2006, at 11:00 AM, Bulat Ziganshin wrote: > Hello > > I released Streams library version 0.1e. Changes are: > > - Fixed bug: "openFD name WriteMode" don't truncated files on unixes > * Full library now released under BSD3 license, thanks to John Goerzen > + Now cabalized, thanks to Jeremy Shaw > > Download: http://freearc.narod.ru/Streams.tar.gz > Docs: http://haskell.org/haskellwiki/Library/Streams > > > btw, my cabal file contain line: > > Build-Depends: base, Win32, template-haskell > > if i correctly understand, this will not work on unix systems, while > without Win32 package program can't be compiled on my windows box. > what i can do here? only provide two separate cabal files - one for > unix, one for windows and give to user text instructions about > manual renaming of appropriate file? This brings up a feature I have sometimes wanted from Cabal. I'd like to be able to say the following: "Cabal, find and use package X if it exists. Furthermore, when CPPing source code, set a #define HAS_PACKAGE_X so I can do conditional compilation based on the (non-) avaliablility of package X". Is this possible? Difficult? It seems like a nice way to handle a number of problems, including platform- dependencies (as here) and optional features with external dependencies. Rob Dockins Speak softly and drive a Sherman tank. Laugh hard; it's a long way to the bank. -- TMBG

2 1

error vs Maybe in Collections package
by Jean-Philippe Bernardy 07 Jun '06

07 Jun '06

Good idea. I added a ticket. http://cvs.haskell.org/trac/ghc/ticket/792 If there is no taker I'll try to do it this weekend. Cheers, JP. On 6/6/06, Robert Dockins <robdockins(a)fastmail.fm> wrote: > > On Jun 6, 2006, at 9:57 AM, David F. Place wrote: > > > It bothers me that the functions findMin, findMax, deleteFindMin > > and deleteFindMax are error when applied to the empty set in > > various types of collections. Shouldn't they return Maybe? > > > FYI, Edison has operations that generalize 'deleteFind*' (named > viewMin/viewMax) which call 'fail' in an arbitrary monad when applied > to an empty collection; obviously, these can be instantiated at type > 'Maybe'. > > http://www.eecs.tufts.edu/~rdocki01/docs/edison/Data-Edison- > Coll.html#v%3AminView > > > > -------------------------------- > > David F. Place > > mailto:d@vidplace.com > > > Rob Dockins > > Speak softly and drive a Sherman tank. > Laugh hard; it's a long way to the bank. > -- TMBG > > _______________________________________________ > Libraries mailing list > Libraries(a)haskell.org > http://www.haskell.org/mailman/listinfo/libraries >

2 1

error vs Maybe in Collections package
by David F. Place 06 Jun '06

06 Jun '06

It bothers me that the functions findMin, findMax, deleteFindMin and deleteFindMax are error when applied to the empty set in various types of collections. Shouldn't they return Maybe? -------------------------------- David F. Place mailto:d@vidplace.com

2 1

Posix IPC?
by Ashley Yakeley 06 Jun '06

06 Jun '06

Is there currently any functionality in the std libs for Posix IPC such as messages, shared memory and so forth? I looked, but didn't see anything. It's not hard to write the FFI bindings oneself, of course, but I wanted to check. -- Ashley Yakeley

2 1

My summer of code project: HaskellNet
by Jun Mukai 04 Jun '06

04 Jun '06

Hello Haskellers, I am Jun Mukai, and my project for Google Summer of Code 2006 is HaskellNet: network related libraries for haskell. my blog for SoC can be read at http://sequence.complete.org/blog/111. The repository will be set up at darcs.haskell.org soon. I prepare another repository at http://www.city5.org/haskellprog/haskellnet/ tentatively. By now, it have only SMTP and POP3 library. Any comments are welcome. Best Regards, -- Jun Mukai mukai(a)jmuk.org

1 0

ANN: Edison 1.2RC4
by Robert Dockins 03 Jun '06

03 Jun '06

Fellow Haskellers! I am pleased to announce the 4th (and with any luck, final) release candidate for Edison 1.2. Edison is a library of efficient data structures for Haskell. Changes from RC3 include: * introduce strict/strictWith operations for all APIs * add Ord* instances for PatriciaLoMap and TernaryTrie * add David F. Place's EnumSet implementation * complete the FiniteMap unit test coverage and fix a bunch of bugs in finite map implementations * add 'symmetricDifference' to Collection and Associated Collection APIs * add Ord instances for data structures * add Monoid instances for data structures * Edison now requires Cabal 1.1.4 to build The project homepage is located at: http://www.eecs.tufts.edu/~rdocki01/edison.html API docs are located at: http://www.eecs.tufts.edu/~rdocki01/docs/edison/index.html And the darcs repository is located at: http://www.eecs.tufts.edu/~rdocki01/edison/ As before, comments are welcome, particularly comments relating to the API. However, barring major disasters I may have overlooked, I hope to release RC4 as 1.2, essentially unchanged. Thanks, Rob Dockins

4 8

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 01 Jun '06

01 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