- Libraries - Haskell.org

Restructured version of the Streams library
by Takano Akio 15 Jun '06

15 Jun '06

Hi, I wrote an IO library, inspired by, and partly based on, Bulat Ziganshin's Streams library. The main aim of the library is to experiment with different design choices, especially emphasizing the following: * Many small classes * Iconv-friendly interface The source code and documentation can be found at: http://yogimo.sakura.ne.jp/ssc/ == Type structure The library tries to assign a different class for a different concept. For example, it distinguishes read-only streams from write-only ones, and seekable streams from non-seekable ones. As a result, the library provides a small and clean interface, so it ensures more type safety, and type signatures of stream-operating functions are able to carry more information. == Iconv friendliness Although Unicode is quickly becoming popular, there are still many widely-used, character-set-specific encodings. Therefore i18n-aware applications may have to handle a locale-specific encoding, which is not statically known. Then, using an external conversion routine, such as iconv, will be virtually the only possible way to go. The library's interface intaracts well with iconv and other c runtines. This also means that adding zlib functionality can be easy, for example. == Performance The library is currently not very fast. For example, typical repitition of putStr/getLine is 1.5x slower than the corresponding standard Handle operations. When you omit locking, the library performs roughly equally well to the standard Handle. == Any comments are welcome. Regards, Takano Akio

3 6

Partial Relax NG Compact support for HXT
by Samuel Bronson 14 Jun '06

14 Jun '06

Attached is partial support for Relax NG Compact for HXT. I haven't been working on this much in the last week and a half or so, so I thought I better send it before I forget about it entirely. (After all, unlike many of my other projects, this one actually seems to do most of what it is intended to do!) Among other things, it is missing: * Include support (couldn't figure out how to get this to work without changing the simplifier, which I am not comfortable doing without a repository) * Full annotation support -- I got frustrated with qualified names... I should really rework that part. * Nice code: the code is very ugly, especially that which relates to names It is not very tested, partly because HXT's RelaxNG validator doesn't seem to support the XML Schema Datatypes, which are very important to Relax NG Compact. (They have their own default namespace prefix xsd.) I might have done a better job if I'd looked at the section of the happy manual that relates to monadic parsers before going very far, but oh well... Comments are welcome and I'd really like to see someone try to do better -- which shouldn't be too hard. The formal grammar at the end of the spec actually looks an awful lot like a computer program... unfortunatey, it uses a slightly different flavour of XML library :-(. Another neat thing that uses Relax NG Compact is James Clark's nxml-mode and rng-validate-mode for emacs, which validates your XML as you write it, tells you what is missing as well as what you can insert. All in Emacs Lisp.

1 0

newbie question: suppressing dos window
by Maarten 13 Jun '06

13 Jun '06

Dear all, Is it somehow possible to suppress the dos window (in windows), for example to run tasks in the background? Couldn't find it anywhere in the documentation. Any suggestions appreciated. Thanks. Regards, Maarten

3 3

FW: [GHC] #794: System.Random: StdGen's genRange doesn't match its next
by Simon Peyton-Jones 13 Jun '06

13 Jun '06

Would someone who knows and cares about random numbers like to fix this bug in the Random library? Many thanks Simon -----Original Message----- From: glasgow-haskell-bugs-bounces(a)haskell.org [mailto:glasgow-haskell-bugs-bounces@haskell.org] On Behalf Of GHC Sent: 12 June 2006 23:28 To: glasgow-haskell-bugs(a)haskell.org Subject: [GHC] #794: System.Random: StdGen's genRange doesn't match its next #794: System.Random: StdGen's genRange doesn't match its next -------------------------------------+-------------------------------------- Reporter: rturk(a)science.uva.nl | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: libraries/base | Version: 6.4.2 Severity: normal | Keywords: Os: Unknown | Difficulty: Easy (1 hr) Architecture: Unknown | -------------------------------------+-------------------------------------- {{{ > g <- getStdGen > length $ filter (<0) $ take (10^6) $ unfoldr (Just . next) g 0 > genRange g (-2147483648,2147483647) }}} Apparently, {{{StdGen}}}'s {{{next}}} only returns non-negative values. Also, {{{StdGen}}} doesn't override de default method {{{genRange}}}. However, the System.Random docs (http://haskell.org/ghc/docs/latest/html/libraries/base/System- Random.html#v%3Anext) promise: ''The next operation returns an Int that is uniformly distributed in the range returned by genRange (including both end points), and a new generator.'' Thus, {{{StdGen}}}'s {{{next}}} violates the uniform distribution requirement. -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/794> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

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 08 Jun '06

08 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