- Haskell - Haskell.org

ANN: monadLib 2.0
by Iavor Diatchki 20 Mar '06

20 Mar '06

Hello everyone, There is now a new version of 'monadLib': a library of monad transformers for Haskell. 'monadLib' is a descendent of 'mtl', the monad template library that is distributed with most Haskell implementations. The library web page is at: http://www.csee.ogi.edu/~diatchki/monadLib Comments and feedback are very welcome! -Iavor

2 1

Compile error in GHC HEAD revision - nobody responding to bug report
by Asfand Yar Qazi 20 Mar '06

20 Mar '06

Hi, I filed a bug report for a compile error in the latest GHC build, but nobody has responded to it for 4 days. I might as well try to fix it myself, but I'm just a functional programming and Haskell beginner. Here's the bug I filed: http://hackage.haskell.org/trac/ghc/ticket/715 Here's what the output of GHC says: ------------------------------------------------------------------------ ==fptools== make all -wr -f Makefile; in /home/ayqazi/src/packages/ghc/ghc-build/libraries/OpenAL ------------------------------------------------------------------------ ../../ghc/compiler/ghc-inplace -H32m -O -W -fno-warn-unused-matches -fwarn-unused-imports -Wall -fffi -Iinclude '-#include "HsOpenAL.h"' -cpp -DCALLCONV=ccall -ignore-package OpenAL -package base -package OpenGL -fgenerics -split-objs -c Sound/OpenAL/ALC/Capture.hs -o Sound/OpenAL/ALC/Capture.o -ohi Sound/OpenAL/ALC/Capture.hi Sound/OpenAL/ALC/Capture.hs:85:3: Couldn't match expected type `NumSamples' against inferred type `Sound.OpenAL.Config.ALCint' Expected type: GettableStateVar NumSamples Inferred type: GettableStateVar Sound.OpenAL.Config.ALCint In the expression: makeGettableStateVar $ (getInteger (Just device) CaptureSamples) In the definition of `captureNumSamples': captureNumSamples device = makeGettableStateVar $ (getInteger (Just device) CaptureSamples) make[2]: *** [Sound/OpenAL/ALC/Capture.o] Error 1 make[1]: *** [all] Error 1 make[1]: Leaving directory `/home/ayqazi/src/packages/ghc/ghc-build/libraries' make: *** [build] Error 1 make: Leaving directory `/home/ayqazi/src/packages/ghc/ghc-build' Here is where (I think) the problem lies (I added the line numbers): File Capture.hs: 42: type NumSamples = ALCsizei 83: captureNumSamples :: Device -> GettableStateVar NumSamples 84: captureNumSamples device = 85: makeGettableStateVar $ getInteger (Just device) CaptureSamples ALCsizei is defined in a file Config.hs two directories below this point: 108: -- | Non-negatitve 32-bit binary integer size 109: type ALCsizei = HTYPE_ALCSIZEI A C header file has the entry '#define HTYPE_ALCSIZEI Word32'. Know what's happening? Thanks

6 6

Haskell Weekly News: March 20, 2006
by dons＠cse.unsw.edu.au 19 Mar '06

19 Mar '06

Haskell Weekly News: March 20, 2006 Greetings, and thanks for reading issue 29 of HWN, a weekly newsletter covering developments in the Haskell community. Each Monday, new editions are posted to [1]the Haskell mailing list and to [2]The Haskell Sequence. [3]RSS is also available. 1. http://www.haskell.org/mailman/listinfo/haskell 2. http://sequence.complete.org/ 3. http://sequence.complete.org/node/feed Announcements * lhs2TeX version 1.11. Andres Loeh [4]announced lhs2TeX version 1.11, a preprocessor to generate LaTeX code from literate Haskell sources. lhs2TeX includes the following features: + Highly customized output. + Liberal parser -- no restriction to Haskell 98. + Generate multiple versions of a program or document from a single source. + Active documents: call Haskell to generate parts of the document (useful for papers on Haskell). + A manual explaining all the important aspects of lhs2TeX. 4. http://article.gmane.org/gmane.comp.lang.haskell.general/13414 Haskell' This section covers activity on [5]Haskell'. * [6]Dropping implicit universal quantification * [7]Refine overlap handling for instance declarations * [8]Ranges and the Enum class * [9]Strict tuples * [10]Time library * [11]Associated types 5. http://hackage.haskell.org/trac/haskell-prime 6. http://thread.gmane.org/gmane.comp.lang.haskell.prime/914 7. http://thread.gmane.org/gmane.comp.lang.haskell.prime/918 8. http://thread.gmane.org/gmane.comp.lang.haskell.prime/937 9. http://thread.gmane.org/gmane.comp.lang.haskell.prime/948 10. http://thread.gmane.org/gmane.comp.lang.haskell.prime/949 11. http://article.gmane.org/gmane.comp.lang.haskell.prime/944 Discussion * Deep Functors. Oleg Kiselyov [12]described an fmap over arbitrarily deep `collections': lists of maybes of maps of IOs, etc. -- arbitrarily nested fmappable things. 12. http://article.gmane.org/gmane.comp.lang.haskell.cafe/11951 * GHC 6.4.2. Simon Marlow [13]put out a heads up for the forthcoming 6.4.2 release of GHC. The rough timescale is to go into release candidate testing in about a week, and have two weeks of release candidates before the final release. 13. http://article.gmane.org/gmane.comp.lang.haskell.cvs.ghc/13935 * Hexdump. Dominic Steinitz [14]mentioned a "hexdump" function he'd written, posing a question about where it would live in the module hierarchy.. 14. http://news.gmane.org/gmane.comp.lang.haskell.libraries Quote of the Week ihope :: My factorial function uses GADTs. Contributing to HWN You can help us create new editions of this newsletter. Please see the [15]contributing information, send stories to dons -at- cse.unsw.edu.au. The darcs repository is available at darcs get http://www.cse.unsw.edu.au/~dons/code/hwn 15. http://haskell.org/haskellwiki/HWN

1 0

Haskell Weekly News: March 13, 2006
by dons＠cse.unsw.edu.au 18 Mar '06

18 Mar '06

Haskell Weekly News: March 13, 2006 Greetings, and thanks for reading issue 28 of HWN, a weekly newsletter covering developments in the Haskell community. Each Monday, new editions are posted to [1]the Haskell mailing list and to [2]The Haskell Sequence. [3]RSS is also available. 1. http://www.haskell.org/mailman/listinfo/haskell 2. http://sequence.complete.org/ 3. http://sequence.complete.org/node/feed Announcements * Alternative to Text.Regex. Chris Kuklewicz [4]announced an alternative to Text.Regex. While working on the [5]language shootout, Chris implemented a new efficient regex engine, using parsec. It contructs a parser from a string representation of a regular expression. 4. http://article.gmane.org/gmane.comp.lang.haskell.cafe/11825 5. http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=all * pass.net. S. Alexander Jacobson [6]launched Pass.net. Written in Haskell, using HAppS, Pass.net lets websites replace registration, confirmation mails, and multiple passwords with a single login, authenticating via their email domain. 6. http://article.gmane.org/gmane.comp.lang.haskell.cafe/11824 Haskell' This section covers activity on [7]Haskell'. * [8]Partial application syntax * [9]Extending the `...` notation * [10]The dreaded offside rule * [11]Strictness standardization 7. http://hackage.haskell.org/trac/haskell-prime 8. http://thread.gmane.org/gmane.comp.lang.haskell.prime/874 9. http://thread.gmane.org/gmane.comp.lang.haskell.prime/881 10. http://thread.gmane.org/gmane.comp.lang.haskell.prime/883 11. http://thread.gmane.org/gmane.comp.lang.haskell.prime/901 Discussion * Non-trivial markup transformations. Further on last week's article on encoding markup in Haskell, Oleg Kiselyov [12]demonstrates non-trivial transformations of marked-up data, markup transformations by successive rewriting (aka, `higher-order tags') and the easy definition of new tags. 12. http://article.gmane.org/gmane.comp.lang.haskell.general/13393 * Popular libraries and tools. John Hughes [13]posted (and [14]here) some interesting figures on the most important libraries and tools, based on the results of his survey of users earlier this year. 13. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/11829 14. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/11875 * haskell-prime fun. Just for fun, Ross Paterson [15]posted, some thought-provoking [16]statistics on haskell-prime traffic. 15. http://article.gmane.org/gmane.comp.lang.haskell.cafe/11831 16. http://www.soi.city.ac.uk/~ross/haskell-prime-stats/ * New collections package. Jean-Philippe Bernardy [17]hinted that his new collections package is almost done. 17. http://article.gmane.org/gmane.comp.lang.haskell.cvs.ghc/13880 * Is notMember not member? John Meacham [18]sparked a bit of a discussion on whether negated boolean functions are useful with a patch adding Data.Set and Data.Map.notMember. 18. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/4411 * Namespace games. In a similar vein, Don Stewart [19]triggered discussion on how to sort the hierarchical namespace, when proposing alternatives to the longish Text.ParserCombinators module name. 19. http://thread.gmane.org/gmane.comp.lang.haskell.libraries/4383 Darcs Corner * Darcs-server. Unsatisified with the current techniques for centralised development with darcs, Daan Leijen went ahead and [20]wrote darcs-server. With darcs-server you can: + push changes remotely via a CGI script + or push changes via a single SSH account that serves many users + use cryptographic verification and authorization of users for reading and writing + use gpg encryption (for CGI) + use non-public repositories that can only be accessed by authorized users. 20. http://thread.gmane.org/gmane.comp.version-control.darcs.user/9686 * darcsweb 0.15, by Alberto Bertogli, has been [21]released. 21. http://article.gmane.org/gmane.comp.version-control.darcs.user/9664 Contributing to HWN You can help us create new editions of this newsletter. Please see the [22]contributing information, send stories to dons -at- cse.unsw.edu.au. The darcs repository is available at darcs get http://www.cse.unsw.edu.au/~dons/code/hwn 22. http://haskell.org/haskellwiki/HWN

9 19

Re: Haskell Digest, Vol 31, Issue 15
by Paul Johnson 17 Mar '06

17 Mar '06

"minh thu" <noteed(a)gmail.com> wrote: >2006/3/15, Duncan Coutts <duncan.coutts(a)worc.ox.ac.uk>: > > >>On Wed, 2006-03-15 at 17:21 +0100, Sebastian Sylvan wrote: >> >> >>>You can also use laziness (untested!): >>> >>>data DLink a = (DLink a) a (DLink a) | Nil >>> >>>test = d1 >>> where d1 = Nil 5 d2 >>> d2 = d1 6 Nil >>> >>>test here is a linked list containing 5 and the next node containing >>>6. Both nodes have references to the next and previous links (wich is >>>Nil at the ends). The magic here is laziness. You reference d2 in the >>>definition for d1 and d1 in the definition for d2. It gets sort of >>>clumsy to work with, though. You're probably better off using STRefs >>>(for example) if you really need that type of data structures... >>> >>> >>I wrote a talk once on this topic: >>http://web.comlab.ox.ac.uk/oucl/work/duncan.coutts/papers/recursive_data_st… >> >>Duncan >> >> >thanks, but I cannot construct the whole sturcture in one time (i need >incremental update). > > You might also go back and think about why you needed that double linked list in the first place. Many things that require complicated list structures in C are better represented as compositions of functions. The simplest example is StringS in the Standard Prelude, which avoids O(n^2) behaviour in list concatenation by composing functions instead. Similarly you might find that by representing your data structure as the composition of the functions needed to build it you can defer creation of the actual structure to the point at which it gets read. Then all the closures get evaluated, but the evaluated stuff stays around and acts as the foundation for subsequent updates. It all depends on what you want to do. I suggest reading Chris Okasaki (sp?) on the subject. Paul.

5 6

The Cell processor and Haskell
by Mads Lindstrøm 16 Mar '06

16 Mar '06

Hi all It is often claimed that Haskell's referential transparency would make it an excellent choice for a high-level parallel language. Furthermore, IBM, Sony, and Toshiba's Cell processor is in great need of a high-level parallel language. Thus, they seem to be a perfect match. See http://www.research.ibm.com/cell/ and http://en.wikipedia.org/wiki/Cell_processor for more about the Cell processor. My knowledge of parallel programming is limited, but I would still like to know, what the wiser amongst of us, thinks of making a Haskell compiler for the Cell processor. It is not that I consider implementing one. It is just that I am curious. Is the Cell processor better suited, as a target for a parallel Haskell compiler than Intel, AMD, and Sun's multicore approaches? What I am asking here is, if it is easier to make use of the Cell's closely coupled but limited SPEs or it is easier to make use of the loosely couple but full-featured parallelism of multicore approaches. Is it realistic to think that a Haskell compiler targeting the Cell processor, could make use of it's parallelism without requiring huge changes to the source code of ones already existing meant-for-GHC programs? /Mads Lindstrøm

3 2

strange ghc behaviour?
by Christopher Brown 15 Mar '06

15 Mar '06

Hi, I use ghc 6.4.1 on Mac OS X 10.4.5 and recently I have been getting some very strange errors. It has happened when I have been trying to compile DrIFT, and also today when I was compiling Haddock. The errors seem to change each time, and they always complain about an undefined symbol. For example, when compiling Haddock I got: /tmp/ghc24547.s:22969:Invalid mnemonic 'mtcts' /tmp/ghc24547.s:103127:Invalid mnemonic 'mtcur' /tmp/ghc24547.s:103430:Invalid mnemonic 'mtcur' /tmp/ghc24547.s:108482:Invalid mnemonic 'mtcts' I flushed out /tmp and ran make again, and this time I get: /tmp/ghc24804.s:unknown:Undefined local symbol L___DISCARE__$stub make[3]: *** [HsSyn.o] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all] Error 1 make: *** [build] Error 1 Eventually Haddock will compile if I keep typing make every time ghc fails, but is there any reason why ghc is doing this? and how can I correct it? Thanks for your help. Chris Brown. Christopher Brown Ph.D. Student University of Kent. UK. http://www.cs.kent.ac.uk/people/rpg/cmb21/index.html

2 5

implementing pointers-based data structure
by minh thu 15 Mar '06

15 Mar '06

hi everybody, i have to implement some data structure which is usually implemented with pointers in imperative languages (can think of it as a double linked list). i'd like to know how i have to do that. is there any standard way of converting pointer-based data structure into an inductively-defined data type (like standard haskell list) ? is-it possible ? or would it be good to define the data structure in c and write a haskell wrapper around the c code ? thank you a lot, bye, vo minh thu

4 5

Trying to get a Composite design pattern to work
by Asfand Yar Qazi 13 Mar '06

13 Mar '06

I'm trying to implement hierarchical states in Haskell, following on from my work at doing them in C++. Here's what I've got so far: data StateNode a = CompositeState [ StateNode a ] | State a stateslist :: StateNode a -> [a] stateslist(State x) = [x] stateslist(CompositeState xs) = {- return list of objects of type a -} The following give errors (as they should) -- stateslist(CompositeState xs) = [ stateslist(x) | x <- xs ] -- stateslist(CompositeState xs) = map stateslist xs You see what I'm trying to do? This is how I want it to behave: sm1 = CompositeState [ State 1, State 2, State 3 ] stateslist(sm1) => [1, 2, 3] How can I get this to work? My mind has just gone all wobbly! Thanks

3 3

debuggers
by Ernesto De Santis 11 Mar '06

11 Mar '06

Hi everybody I'm searching debuggers for Haskell, and I have this list: nhc tracer hood QuickCheck Auburn Hugs interpreter ghci (GHC interpreter) Some body know another debugger? Thanks a lot! Ernesto.

5 4