November 2006 - Haskell

ANNOUNCE: How to make a type-safe alloca: a new construction of a restricted IO monad
by Stefan O'Rear 28 Nov '06

28 Nov '06

Introducing RIO, an experimental library for extensible restricted IO in Haskell. Selling features: * Automatic permissions inference. If you define a function: fun fd r = do { str <- hGetContents fd ; modifyRIORef r (+1) ; return str } RIO can infer the requirements as part of the type signature: fun :: (Elem (ReadHandle a) o, Elem (ReadRef b) o, Elem (WriteRef b) o, Num c) => RHandle a -> RIORef b c -> RIO o String * Recursive context aquisition. If we have: foo ref1 = withContextRef $ \p -> do ref2 <- newRIORef p (undefined :: Int) -- do some stuff with ref1 and ref2 it works. * Potential for region allocation. When you use withContextRef, any variables allocated cannot be type- correctly accessed after withContextRef returns. Hence, they can be deallocated at that time. It is possible to write a function withNewRef that allocates a reference and does not grant the capability to allocate more in the same context; this could be implemented using *stack* allocation. * It is extremely extensible. It is possible to implement a new set of restricted actions in a separate module with very little code. See the file README.LONG for a detailed demonstration. http://members.cox.net/stefanor/RIO-0.1.tar.gz

1 0

Haskell Weekly News: November 28, 2006
by dons＠cse.unsw.edu.au 28 Nov '06

28 Nov '06

--------------------------------------------------------------------------- Haskell Weekly News http://haskell.org/haskellwiki/HWN Issue 51 - November 28, 2006 --------------------------------------------------------------------------- Welcome to issue 51 of HWN, a weekly newsletter covering developments in the Haskell community. Automated testing fever strikes the camp, with three new QuickCheck-related libraries and tools released. Announcements * QuickCheck 2 development version. Bjorn Bringert [1]announced that the development version of QuickCheck 2 is now available in a public darcs repository. Highlights of the new QuickCheck version include: shrinking of failing test cases; supports testing monadic code; handles exceptions gracefully; coarbitrary has moved to a separate class; type-level modifiers for changing test data generation (e.g. NonNegative); function table printing; and user-defined actions when properties fail. The source is [2]available via darcs. 1. http://article.gmane.org/gmane.comp.lang.haskell.general/14511 2. http://www.cs.chalmers.se/~bringert/darcs/QuickCheck/ * PQC: QuickCheck in the Age of Concurrency. Don Stewart [3]announced PQC: Parallel QuickCheck. [4]PQC provides a single module: [5]Test.QuickCheck.Parallel. This is a QuickCheck driver that runs property lists as jobs in parallel, and will utilise as many cores as you wish, with the SMP parallel GHC 6.6 runtime. It is simple, scalable replacement for Test.QuickCheck.Batch. 3. http://article.gmane.org/gmane.comp.lang.haskell.general/14503 4. http://www.cse.unsw.edu.au/~dons/pqc.html 5. http://www.cse.unsw.edu.au/~dons/pqc/Test-QuickCheck-Parallel.html * cabal-test: automatic testing for Cabal projects. David Himmelstrup [6]announced cabal-test, the automatic tester for Cabal projects. The cabal-test tool is capable of testing embedded QuickCheck properties in any and all cabalized projects. The tests are currently executed in parallel with PQC. QuickCheck properties can reside anywhere in the code and don't have to be exported. The [7]darcs repo is available. 6. http://article.gmane.org/gmane.comp.lang.haskell.general/14519 7. http://darcs.haskell.org/~lemmih/cabal-test * Streams 0.1.7. Bulat Ziganshin [8]announced Streams version 0.1.7, a fast extensible [9]I/O and serialization library. Changes include: GHC 6.6 support, support for files larger than 4G on Windows, haddock documentation. 8. http://article.gmane.org/gmane.comp.lang.haskell.general/14504 9. http://haskell.org/haskellwiki/Library/Streams * Ranged Sets 0.0.3. Paul Johnson [10]announced the 0.0.3 release of [11]Ranged Sets. Ranged sets allow programming with sets of values described by a list of ranges. A value is a member of the set if it lies within one of the ranges. The ranges in a set are ordered and non-overlapping, so the standard set operations can be implemented by merge algorithms in O(n) time. 10. http://article.gmane.org/gmane.comp.lang.haskell.general/14492/ 11. http://ranged-sets.sourceforge.net/Ranged/ * Type-class overloaded functions. Oleg Kiselyov [14]presented functions polymorphic over classes of types. Each instance of such (2-polymorphic) function uses ordinary 1-polymorphic methods, to generically process values of many types, members of that 2-instance type class. The typeclass constraints are thus manipulated as first-class entities. We also show how to write typeclass instances with back-tracking: if one instance does not apply, the typechecker will chose the `next' instance -- in the precise meaning of `next'. 14. http://thread.gmane.org/gmane.comp.lang.haskell.general/14482/focus=14483 * Cabal mode for emacs. Matthew Danish [15]released a small (and developing) major mode for editing Cabal files in emacs. 15. http://mapcar.org/haskell/cabal-mode/ * YCR2JS Programmers Guide Draft. Dimitry Golubovsky [16]announced the draft of low-level [17]programming guide for Yhc Core to Javascript converter. Everyone interested in future use of this tool is encouraged to read and review the Guide. Its purpose is to give some ideas about interaction of Haskell programs converted into Javascript with a web browser on the lowest possible level, without application frameworks and support libraries (just because these haven't been developed). 16. http://article.gmane.org/gmane.comp.lang.haskell.cafe/16764 17. http://haskell.org/haskellwiki/Yhc/Javascript/Programmers_guide * NeHe Tutorials in Haskell. Jason Dagit [18]announced the availability of the (somewhat) famous NeHe tutorials for OpenGL have been ported to HOpenGL. A [19]darcs repository is provided. 18. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16870/focus=16870 19. http://codersbase.com/index.php/Nehe-tuts Haskell' This section covers the [20]Haskell' standardisation process. * [21]Standard (core) libraries initiative: rationale 20. http://hackage.haskell.org/trac/haskell-prime 21. http://thread.gmane.org/gmane.comp.lang.haskell.prime/1919/focus=1919 Discussion * The Future of MissingH. John Goerzen [22]opened a discussion on the future development of MissingH, in order to improve usability and adoption by the community. It was felt that the current monolithic structure, and nondescript naming raises the barrier to use. 22. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16819/focus=16819 * Controlling backtracking with the list monad. Isto [23]posed an interesting question about whether the backtracking in a list monad could be controlled, producing a nice way to build branch and bounding algorithms 23. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16776/focus=16776 * Lifting conditionals. Dougal Stanton [24]asked about lifting if/then/else into the IO monad, revealing some interesting subtleties in the implementation 24. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16779/focus=16779 * Cheap commits for improving the library documentation. Don Stewart [25]created a new page, for non-core contributors to submit suggestions on improving library documentation. Rather than require the user to produce a darcs patch, than can instead add their change to the wiki page. Ideally, a wiki-editable version of the entire haddock docs would be desired 25. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16848/focus=16848 * Optimising a hash function. Ivan Tomac [26]asked about optimising low-level bit shifting code to run close to C speed. 26. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/16849/focus=16849 * Common subexpression elimination. Christian Maeder [27]asked about common subexpression elimination in GHC. 27. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/11318/focus=11… * Passing arrays between Haskell and C without overhead. Brian Hulley. [28]asked for advice on how best to bind to a C library, from Haskell, such that data can be passed with zero copying. 28. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/11313/focus=11… * Creating Loadable Dynamic libraries Libraries in OSX. Alfonso Acosta. [29]described how to create dynamic loadable libraries (.so) with GHC 6.6 under Linux X86 using some C glue code and linking with -optl -shared. Advice was sought on porting this to OSX. 29. http://thread.gmane.org/gmane.comp.lang.haskell.glasgow.user/11297/focus=11… Conference roundup [30]Fun in the Afternoon 30. http://sneezy.cs.nott.ac.uk/fun/nov-06/ * Links: web programming without tiers. Phil Wadler. * Mechanized Metatheory Model-Checking. James Cheney. Syntactic techniques based on operational semantics and type systems are extremely useful for studying the metatheory of logics and programming languages. However, proving properties of such systems is often an onerous task, because there are often a large number of 'standard' or straightforward cases. It is tempting to cut corners and do careful proofs only for cases that seem interesting, but this can easily lead to a counterexample being missed. It therefore seems attractive to try to formalize such proofs in order to ensure their validity. However, this cure is often worse than the disease, since theorem proving tools tend to have a high learning curve. Moreover, theorem proving generally addresses the uncommon case of formalizing an informal proof of a well-understood system which has stood the test of time, rather than the common case of finding a bug in a poorly-understood system that may still be under development. * A Principled Approach to Version Control. Wouter Swierstra. Version control systems are essential for managing the distributed development of large software pro jects. We present a formal model for reasoning about version control. In particular, we give a general definition of patch. Patches abstract over the data on which they operate, making our framework equally suited for version control of everything from highly-structured XML files to blobs of bits. We model repositories as a multiset of patches. The mathematical definitions of patches and repositories enable us to reason about complicated issues such as conflicts and conflict resolution. * C# is a functional programming language. Andrew Kennedy. Polymorphic functions, parameterized types, lambda expressions, type inference, streams, GADTs, and combinator libraries: these are features usually associated with functional languages such as Haskell and ML. In this talk I will show how the design of C# 3.0, launching next year, was strongly influenced by ideas from functional programming, and supports all these features. I will use classic examples from the functional programming literature to illustrate the talk. Blog noise [31]Haskell news from the blogosphere. * [32]Scoping, and my Scheme Interpreter * [33]XP Day Benelux: the report * [34]From L�b's Theorem to Spreadsheet Evaluation * [35]A search engine written in Haskell: part two * [36]OCaml - First Impressions * [37]House: an Operating System Written in Haskell * [38]Reverse Engineering Machines with the Yoneda Lemma * [39]Haskell and functional programming * [40]Query for readers: Interested in Haskell? * [41]Why Haskell? * [42]What rocks * [43]Why pointers and references are a bad thing * [44]Reasons to learn C * [45]Some Thoughts on Haskell * [46]Simon Peyton Jones and Tim Harris explain STM [video] 31. http://planet.haskell.org/ 32. http://computativestylings.blogspot.com/2006/11/scoping-and-my-scheme-inter… 33. http://matteo.vaccari.name/blog/?p=59 34. http://sigfpe.blogspot.com/2006/11/from-l-theorem-to-spreadsheet.html 35. http://haskell-web.blogspot.com/2006/11/search-engine-written-in-haskell-pa… 36. http://triple.aeoth.net/2006/11/24/ocaml-first-impressions/ 37. http://www.osnews.com/story.php?news_id=16554 38. http://sigfpe.blogspot.com/2006/11/yoneda-lemma.html 39. http://chanson.livejournal.com/156835.html 40. http://scienceblogs.com/goodmath/2006/11/query_for_readers_interested_i.php 41. http://scienceblogs.com/goodmath/2006/11/why_haskell.php 42. http://n3dst4.livejournal.com/50393.html 43. http://kawagner.blogspot.com/2006/11/why-pointers-and-references-are-bad.ht… 44. http://www.nekomancer.net/blog/archives/reasons-to-learn-c 45. http://geeklikemetoo.blogspot.com/2006/11/some-haskell-basics.html 46. http://channel9.msdn.com/Showpost.aspx?postid=231495 Quotes of the Week * Cale: GADT pattern match in non-rigid context. There are prescriptions available for that sort of thing. * Bulat: Base lib includes everything, from 'const' to starship control * Cale: I should actually think before coding, but the type system is so good. * Syzygy-: Why do I get the feeling that the arrow code was written during 'Speak like a pirate day'? * SamB: [on the subject of fromJust] thinks that unJust would be a more fun name for that * apfelmus: Ah, yet another UndeadArray necromancer exhausting his stack of bones. May the forces of light suggest to structure the incantation of darkness? * dons: Ah, it *is* like Haskell. I see: Jiffy Pop is the family fun treat. No other brand of popcorn offers a self-contained popping pan. It can even be used outdoors. * thetallguy: Using and advocating Haskell is like being Calvin (and Hobbes). To you, it's alive, real, a true delight. To those who know better, it's a stuffed tiger. Code Watch * Mon Nov 27 05:06:02 PST 2006. Wolfgang Thaller. [47]Initial support for x86_64-darwin. Basic -fvia-C code generation is there, not much testing. 47. http://article.gmane.org/gmane.comp.lang.haskell.cvs.ghc/18000 * Fri Nov 24 15:05:48 PST 2006. simonpj. [48]Fix constraint handling for lazy patterns. Lazy patterns are quite tricky! Consider f ~(C x) = 3. Can the Num constraint from the 3 be discharged by a Num dictionary bound by the pattern? Definitely not! See Note 'Hopping the LIE in lazy patterns' in TcPat. The type checker wasn't ensuring this, and that was causing all manner of strange things to happen. It actually manifested as a strictness bug reported by Sven Panne. 48. http://article.gmane.org/gmane.comp.lang.haskell.cvs.ghc/17969 About the Haskell Weekly News Each week, new editions are posted to [49]the Haskell mailing list as well as to [50]the Haskell Sequence and [51]Planet Haskell. [52]RSS is also available, and headlines appear on [53]haskell.org. Headlines are available as PDF. The Haskell Weekly News is also [54]available in Spanish translation. To help create new editions of this newsletter, please see the [55]contributing information. Send stories to dons at cse.unsw.edu.au. The darcs repository is available at darcs get [56]http://www.cse.unsw.edu.au/~dons/code/hwn 49. http://www.haskell.org/mailman/listinfo/haskell 50. http://sequence.complete.org/ 51. http://planet.haskell.org/ 52. http://sequence.complete.org/node/feed 53. http://haskell.org/ 54. http://haskell.org/haskellwiki/HWN/es 55. http://haskell.org/haskellwiki/HWN 56. http://www.cse.unsw.edu.au/~dons/code/hwn

1 0

Re: base libraries
by Paul Johnson 27 Nov '06

27 Nov '06

Duncan Coutts wrote: > Cabal can be upgraded using Cabal. This is important. I think there are two things here. First, its important that it can be upgraded. Most distributions already have a package management system that can handle this. I just got GHC, Haddock and a bunch of other stuff onto my new FC6 installation via Pirut, the FC6 graphical installer (which is coded in Python and would probably be much faster and better if it was in Haskell, but thats another story). If Cabal and its supporting libraries are split out from the main distribution into separate packages then I can get an upgraded Cabal the same way without having to get those libraries as well. > Unlike well specified libraries like ByteString, Cabal tackles an > ill-specified problem and we're still learning how best to solve it, so > the api and features change much more frequently. So if I become a Cabal developer rather than just a user, or I want the source for some other reason, then of course I can do a cabal-get cabal and all the relevant libraries get installed for me. In short, I think that Cabal should work like any other application: you can just install the binary and go, or you can install the source (via Cabal) and hack it if you want to. Binary installs should be handled by the distribution (Windows users can have a .msi, Mac users can have whatever the Mac equivalent is). Compiler packages should generally include the Cabal binaries along with the compiler binary, Haddock binary and so on. Paul.

1 0

polymorphism and existential types
by Louis-Julien Guillemette 27 Nov '06

27 Nov '06

Supposing a polymorphic value (of type, say, forall a . ExpT a t) is stored inside an existential package (of type, say, forall a . Exp a), I wonder how to recover a polymorphic value when eliminating the existential. The ``natural way'' to write this doesn't work: {-# OPTIONS -fglasgow-exts #-} data ExpT a t data Exp a = forall t . Exp (ExpT a t) f :: (forall a . ExpT a t) -> () f e = () g :: (forall a . ExpT a t) -> () g e = let e1 :: forall a . Exp a e1 = Exp e in case e1 of Exp e' -> f e' {- Test.hs:18:17: Inferred type is less polymorphic than expected Quantified type variable `a' is mentioned in the environment: e' :: ExpT a t (bound at Test.hs:18:11) In the first argument of `f', namely `e'' In the expression: f e' In a case alternative: Exp e' -> f e' Failed, modules loaded: none. -} In particular, the solution of pushing the forall inside doesn't work for me (it breaks other parts of my code...) : data Exp' = forall t . Exp' (forall a . ExpT a t) g' :: (forall a . ExpT a t) -> () g' e = let e1 :: Exp' e1 = Exp' e in case e1 of Exp' e' -> f e' Any help welcome! Louis-Julien

1 0

Midlands Graduate School 2007
by Graham Hutton 27 Nov '06

27 Nov '06

+--------------------------------------------------------------------+ Midlands Graduate School in the Foundations of Computing Science MGS 2007 16-20 April 2007 University of Nottingham, UK The Midlands Graduate School (MGS) in the Foundations of Computing Science was established in 1999, and organises an annual Spring School on the mathematical foundations of computing. It is primarily aimed at PhD students, but everyone is welcome to attend, including computer scientists, mathematicians, and participants from outside of the UK. PROGRAMME The programme will consist of ten short courses given over a period of five days. The speakers are as follows: Roy Crole Operational Semantics Neil Ghani Category Theory Paul Levy Typed Lambda Calculus Thorsten Altenkirch Dependently Typed Programming Roland Backhouse Algorithmic Problem Solving Martin Escardo Domain Theory and Denotational Semantics Achim Jung Stone Duality Henrik Nilsson Advanced Functional Programming Colin Stirling Concurrency Theory Steve Vickers Topology and Logic LOCATION The school will be held on the Jubilee Campus of the University of Nottingham, which has won many awards for its design and environmental features. Nottingham is centrally located in the UK, and is easily reachable by road, rail and air (Nottingham East Midlands Airport.) REGISTRATION The registration fee will be around 350 pounds, including accommodation. We also hope to have a number of free places for PhD students. To register your interest in MGS 2007, and be informed when registration is open (the number of places will be limited), please contact the organiser, Henrik Nilsson <nhn(a)cs.nott.ac.uk>. FURTHER DETAILS Google search - MGS 2007 Web page - http://www.cs.nott.ac.uk/~nhn/MGS2007/ Xmas seminars - http://www.dcs.shef.ac.uk/~georg/mgs_xmas06.html +--------------------------------------------------------------------+ This message has been checked for viruses but the contents of an attachment may still contain software viruses, which could damage your computer system: you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

1 0

Announcement: cabal-test
by Lemmih 26 Nov '06

26 Nov '06

Greetings fellow hackers, It's my great pleasure to announce cabal-test, the automatic tester for Cabal projects. The cabal-test tool is capable of testing embedded quickcheck properties in any and all cabalized projects. The tests are currently executed in parallel with pqc. QC properties can reside anywhere in the code and don't have to be exported. Requirements: * GHC-6.6 * pqc (http://www.cse.unsw.edu.au/~dons/code/pqc/) Future goals: * Use Haddock information when printing test names. * Ability to test multiple targets simultaneously. Example usage: > $ ./cabal-test > Testing cabal-test-0.1 > No library > Testing executable: cabal-test > 2: Tests.prop_sort2 : OK, 100 tests. > 2: Tests.prop_sort3 : OK, 100 tests. > 2: Tests.prop_sort4 : OK, 100 tests. > 2: Tests.prop_sort6 : OK, 100 tests. > 1: Tests.prop_sort1 : OK, 100 tests. > 2: Tests.prop_sort5 : OK, 100 tests. > $ ./cabal-test --help > Usage: cabal-test [OPTION] [PATH] > > Options: > -h, -? --help Show this help text > -v[n] --verbose[=n] Control verbosity (n is 0-5, normal verbosity level is 1, -v alone is equivalent to -v3) > -w path --ghc-path=path Use this command as ghc > -n n --threads=n Use 'n' threads. Default is 2 > -t n --tests=n Depth of test. Default 100 > -p prop --proptag=prop Tests must contain this tag. Default 'prop_'. The darcs repository is available at: http://darcs.haskell.org/~lemmih/cabal-test -- Cheers, Lemmih PS: I'm so glad we're not using Makefiles anymore.

2 2

ANNOUNCE: QuickCheck 2 development version
by Björn Bringert 24 Nov '06

24 Nov '06

This is just a quick announcement that the development version of QuickCheck 2 is now available in a public darcs repository. Some highlights: - Shrinks failing test cases. - Supports testing monadic code. - Handles exceptions gracefully. - coarbitrary has moved to a separate class, to make it easier to write simple instances of Arbitrary. - Type-level modifiers for changing test data generation (e.g. NonNegative). - Magic function table printing. - User-defined actions when properties fail. You can get it with: darcs get http://www.cs.chalmers.se/~bringert/darcs/QuickCheck/ This is a development version, and the API is not necessarily stable yet. It uses the same module names as QuickCheck 1, Test.QuickCheck.*, but it breaks backwards compatibility in many cases. Use explicit package versioning if you need to use it alongside QuickCheck 1. You can build the code and API documentation with Cabal, see README for instructions. Happy hacking! Koen & Björn

3 2

base libraries
by Bulat Ziganshin 24 Nov '06

24 Nov '06

Hello libraries, current situation when Base lib includes everything, from 'const' to starship control, creates serious problems for upgrading to next GHC version. this problem arise because every GHC version arrives with its own, non-upgradable version of Base library and even minor API changes here results in total loss of ability to use new ghc version without #ifdef workarounds it seems strange, but even in this situation some developers prefer to include their modules in base lib instead of providing them as separate libs that can be easily upgraded using Cabal so that any library version can be used with any version of ghc after analyzing the situation, i concluded that source of problem, at least partially, is GHC HQ's policy of packaging libraries with GHC. The list of such libraries is *closed* and odd at first look. why, for example, it includes 2 of 4 regexp libs? at last end, we can learn that this is the list of libs required to build ghc itself! unfortunately, ghc building process don't require FilePath or ByteString support. this results in situation when authors of these libs, looking for a way to include their libs in every ghc distro, don't find other solution except for including them in Base lib. moreover, while Simon & Ian argued against inclusion of these very useful libs in base libs set, they don't have anything against inclusion of the same code in Base lib i'm all for docracy, but not in this case. having the experience of patching the program after each ghc upgrade, and having a bunch of conditionally compiled code in my libs, i can easily imagine ghc version hell that these new modules will provide for us so, i think, it is time to revise ghc libraries policy. i propose to split libs into the 3 "rings": core, base and other * core libs includes ghc version dependent libs. currently it's base, stm and th * base libs are libs, which is guaranteed to be shipped with *every* haskell distribution. on windows, it should be a part of monolithic installer while on unixes these libs may be a separate packages, but porters should ensure that they ported all these libs too this set should include only small and widely used libs, what is mainly datastructure processing ones. my current list includes regexps, bytestring, filepath, time, mtl, fgl, newbinary and so on * other libs are either too specific or too large, such as gui, database and xml ones. these may be included in some distros or installed separately these rules don't say anything about ghc self-building process. it will be great to maintain the list of libraries, required to do it, separately and while we are here - splitting of Base library should be very easy task. its ghc version specific part is in GHC.* modules. these modules should be moved into the new Core library. plus a few modules from Data.* or System.* hierarchy they imports. plus a few modules imported by Hugs/NHC Prelude. the rest of Base library should be pretty portable, at least between various GHC versions. and independent on this work, we can always look into moving pieces of Base into independent libs. as a rule of thumb, everything not used in Base lib may be moved outside of it. of course, these new libs should be included in base libs set in particular, my hottest hope is that ghc 6.6.1 will be shipped with fps 0.8 as separate library that will provide both backward compatibility with 6.6 and will allow to upgrade fps without recompiling ghc itself :D -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

5 5

Library versioning policy
by Bulat Ziganshin 24 Nov '06

24 Nov '06

Hello haskell, We all know what is a "DLL hell" - if some program written against version 2.0 of library, then using version 1.0 or 3.0 of the same library when program compiled may call all sorts of devil there is a great solution of this problem - Eternal Compatibility Theory. unfortunately, it's so great that seldom used on practice because it requires significant efforts from library developer i propose another solution for this problem which don't require to write anything and just impose some policies both for library developer and user library developer should just explicitly declare and obey some policy of incrementing library versions which will allow library user to select proper range of versions which are guaranteed to work with his code. i propose the following as standard policy: * major version change (2.2->3.0) means total incompatibility. all sorts of changes may come * middle version change (2.1.3->2.2) means addition of new facilities (functions, constructors, classes...). in most cases, such semi-compatible version can be used if that's done with caution * minor change (2.1.1->2.1.2) means only internal changes and bugfixes other policies may be used as well, library developer should just declare his policy at the library birth and use it consistently this will allow library users to declare in their cabal files range of library versions they are accepted and be sure that this range will include latest bugfixes but don't include any incompatible changes. then Cabal (which now can handle many versions of the same lib) should select latest acceptable version of library requested and make it available for compilation say, if i wrote my program using Streams 1.2.1, i can specify Build-Depends: Streams 1.2.* or, if i brave/accurate enough: Build-Depends: Streams 1.* and Cabal should select the best library available at build time -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0

Streams 0.1.7 library announce
by Bulat Ziganshin 24 Nov '06

24 Nov '06

Hello haskell, I'm glad to present Streams library version 0.1.7. If you don't yet know, Streams is a fast extensible I/O and serialization library ( http://haskell.org/haskellwiki/Library/Streams ). Main changes against previous version: * true support for GHC 6.6 * support of files larger than 4 gb on windows (see FD5gb.hs example) * files are now open in shared mode on all systems The most dramatic change, though, is that Pupeno <pupeno(a)pupeno.com> has taken a work of polishing, unix testing, packaging and distributing the library, so now it contains: * haddock'ized internal docs * ready to be included in any unix packaging system * semi-automatic self-testing system ;) To get Streams 0.1.7, you can download one of http://pupeno.com/software/streams/Streams-0.1.7.tar.bz2 http://pupeno.com/software/streams/Streams-0.1.7.tar.gz or you can get it from its repository by running: darcs get --tag=0.1.7 http://software.pupeno.com/Streams-0.1 Streams-0.1.7 You can also download and keep track of the 0.1 branch, which is supposed to remain stable and only get bug-fixes by running darcs get http://software.pupeno.com/Streams-0.1/ and then run 'darcs pull' inside it to get further changes. To get the latest unstable and fluctuating version, the development version, run: darcs get http://software.pupeno.com/Streams/ Note: as of this moment, while the project is being darcsified you are not going to find anything useful there, but we expect that to change. Preferably, you should send patches to code to Bulat.Ziganshin(a)gmail.com and to other parts of library to pupeno(a)pupeno.com . Documentation may be edited right at the project homepage, which remains http://haskell.org/haskellwiki/Library/Streams -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0