- Haskell - Haskell.org

Fwd: Re: Haskell help!
by Marc Ziegert 26 Mar '03

26 Mar '03

---------- Weitergeleitete Nachricht ---------- Subject: Re: Haskell help! Date: Wed, 26 Mar 2003 23:57:42 +0100 From: Marc Ziegert <coeus(a)gmx.de> To: "Weix, Rachel Lynn" <weixrl(a)uwec.edu> i'm just programming the solution. imagine a matrix / a rectangle with a grid: the first word at the top side, the second at the right. each entry takes the best solutions of your sequencing problem using just a part of the words. in the top right there is the entry for the two empty words: (0,[ ("","") ]) the value is zero, the list has just one solution: ("","") below this entry: (-1,[ ("~","s") ]) at the bottom right: (-7,[ ("~~~~~~~","Masters") ]) all this was the rightmost column. the whole matrix is a list of columns - the head is the leftmost column. the head of one column is the entry at the bottom. the solution of the whole problem is at the bottom left - the head of the head of the matrix. - marc Am Mittwoch, 26. März 2003 23:32 schrieben Sie: > P.S. The example given is for the set of sequences/strings (The,Masters) > > -----Original Message----- > From: Weix, Rachel Lynn > Sent: Wed 3/26/2003 4:30 PM > To: coeus(a)gmx.de > Cc: haskell(a)haskell.org > Subject: RE: Haskell help! > > > Currently I'm having problems with type checking due to Haskell being a > strongly typed language. In order to return all optimal solutions, my > professor suggested I create a list of tuples if they all have the same > score, as indicated in my new maxSeq method (see attachment). However, > this means that the maxSeq method would return type [[(Char,Char)]]. Since > maxSeq and getSeq must return the same type, I end up changing all my > signatures and end up with lots of problems. In my caseX methods, I can't > append [(Char)] to [[(Char,Char)]], and it also messes up trying to get my > score of each tuple, etc. I've been trying to solve the problem in Scheme > first but even then I'm having problems, and Scheme is only loosely typed. > Any suggestions? The format should be something like the following > (partial example): > > [ [(-,M)] [(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)] [(-,s)(h,t)(e,e)(-,r)(-,s)]]]. > > Once I get back my list, I need to pair everything up in order to return a > list of optimal solutions. Taking the previous example, it would be the > following when paired up correctly: > > [ [(-,M)(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)], > [(-,M)(T,a)(-,s)(h,t)(e,e)(-,r)(-,s)] ] > > From there, I find which sequence has the maximum score. I then walk > through the list again, making a list of all the sequences that have that > score. These two steps seem fairly trivial, it's all the above stuff which > I'm struggling on. > > Rachel > > -----Original Message----- > From: Marc Ziegert [mailto:coeus@gmx.de] > Sent: Tue 3/25/2003 3:13 PM > To: Weix, Rachel Lynn > Cc: > Subject: Re: Haskell help! > > > > maxSeq had one mistake: || instead of && > > i will think about the problem itself, before rewriting all. > > this is the file a little bit more in haskell style. > > Am Dienstag, 25. März 2003 20:16 schrieben Sie: > > I found my two mistakes, plus I fixed the method my Professor said was > > incorrect. Now I just have to be able to find ALL optimal solutions, > > instead of just one. Hooray! > > > > Rachel -- Ein Unseliger, der nur kreist um sich selbst, im Leben wird er dem Ruhm nachsehen und doppelt sterbend untergehen; im gemeinen Staub, aus dem er entsprungen, unbeweint, ungeehrt und unbesungen. [Sir Walter Scott] ------------------------------------------------------- -- Ein Unseliger, der nur kreist um sich selbst, im Leben wird er dem Ruhm nachsehen und doppelt sterbend untergehen; im gemeinen Staub, aus dem er entsprungen, unbeweint, ungeehrt und unbesungen. [Sir Walter Scott]

1 0

RE: Haskell help!
by Weix, Rachel Lynn 26 Mar '03

26 Mar '03

P.S. The example given is for the set of sequences/strings (The,Masters) -----Original Message----- From: Weix, Rachel Lynn Sent: Wed 3/26/2003 4:30 PM To: coeus(a)gmx.de Cc: haskell(a)haskell.org Subject: RE: Haskell help! Currently I'm having problems with type checking due to Haskell being a strongly typed language. In order to return all optimal solutions, my professor suggested I create a list of tuples if they all have the same score, as indicated in my new maxSeq method (see attachment). However, this means that the maxSeq method would return type [[(Char,Char)]]. Since maxSeq and getSeq must return the same type, I end up changing all my signatures and end up with lots of problems. In my caseX methods, I can't append [(Char)] to [[(Char,Char)]], and it also messes up trying to get my score of each tuple, etc. I've been trying to solve the problem in Scheme first but even then I'm having problems, and Scheme is only loosely typed. Any suggestions? The format should be something like the following (partial example): [ [(-,M)] [(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)] [(-,s)(h,t)(e,e)(-,r)(-,s)]]]. Once I get back my list, I need to pair everything up in order to return a list of optimal solutions. Taking the previous example, it would be the following when paired up correctly: [ [(-,M)(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)], [(-,M)(T,a)(-,s)(h,t)(e,e)(-,r)(-,s)] ] From there, I find which sequence has the maximum score. I then walk through the list again, making a list of all the sequences that have that score. These two steps seem fairly trivial, it's all the above stuff which I'm struggling on. Rachel -----Original Message----- From: Marc Ziegert [mailto:coeus@gmx.de] Sent: Tue 3/25/2003 3:13 PM To: Weix, Rachel Lynn Cc: Subject: Re: Haskell help! maxSeq had one mistake: || instead of && i will think about the problem itself, before rewriting all. this is the file a little bit more in haskell style. Am Dienstag, 25. März 2003 20:16 schrieben Sie: > I found my two mistakes, plus I fixed the method my Professor said was > incorrect. Now I just have to be able to find ALL optimal solutions, > instead of just one. Hooray! > > Rachel

2 1

RE: Haskell help!
by Weix, Rachel Lynn 26 Mar '03

26 Mar '03

Currently I'm having problems with type checking due to Haskell being a strongly typed language. In order to return all optimal solutions, my professor suggested I create a list of tuples if they all have the same score, as indicated in my new maxSeq method (see attachment). However, this means that the maxSeq method would return type [[(Char,Char)]]. Since maxSeq and getSeq must return the same type, I end up changing all my signatures and end up with lots of problems. In my caseX methods, I can't append [(Char)] to [[(Char,Char)]], and it also messes up trying to get my score of each tuple, etc. I've been trying to solve the problem in Scheme first but even then I'm having problems, and Scheme is only loosely typed. Any suggestions? The format should be something like the following (partial example): [ [(-,M)] [(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)] [(-,s)(h,t)(e,e)(-,r)(-,s)]]]. Once I get back my list, I need to pair everything up in order to return a list of optimal solutions. Taking the previous example, it would be the following when paired up correctly: [ [(-,M)(T,a)(h,s)(-,t)(e,e)(-,r)(-,s)], [(-,M)(T,a)(-,s)(h,t)(e,e)(-,r)(-,s)] ] >From there, I find which sequence has the maximum score. I then walk through the list again, making a list of all the sequences that have that score. These two steps seem fairly trivial, it's all the above stuff which I'm struggling on. Rachel -----Original Message----- From: Marc Ziegert [mailto:coeus@gmx.de] Sent: Tue 3/25/2003 3:13 PM To: Weix, Rachel Lynn Cc: Subject: Re: Haskell help! maxSeq had one mistake: || instead of && i will think about the problem itself, before rewriting all. this is the file a little bit more in haskell style. Am Dienstag, 25. März 2003 20:16 schrieben Sie: > I found my two mistakes, plus I fixed the method my Professor said was > incorrect. Now I just have to be able to find ALL optimal solutions, > instead of just one. Hooray! > > Rachel

2 1

Theme One Project
by Jon Awbrey 26 Mar '03

26 Mar '03

o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o My name is Jon Awbrey. I have returned to university in my 50's to work on a doctorate in systems engineering and also to "capstone" a few old projects that I did not get to finish up in my last millennium. In the process I have dug up one of my old unfinished programming symphonies that I thought I might offer to the Haskell community as a potentially interesting exercise or project, since one of the things that I was trying to do there was to transpose Pascal into a functional style. The main idea of the program is to integrate an "empiricist" learning algorithm (working on 2-level formal languages) with a "rationalist" reasoning algorithm (working at the level of propositional calculus). It implements (what I used to think of as) some pretty radical graph-theoretic data structures, extending some ideas of C.S. Peirce on graphical syntaxes for propositional logic. (I had started this somewhat before Conceptual Graphs hit the streets.) I have begun documenting this work in a detailed way at the following list site, dedicated to this and related purposes: http://stderr.org/pipermail/inquiry/ http://stderr.org/cgi-bin/mailman/listinfo/inquiry You can also get the postings in newsgroup format from Gmane: http://gmane.org/ http://news.gmane.org/ http://news.gmane.org/thread.php?group=gmane.comp.inquiry The Pascal source code is here: 01. http://stderr.org/pipermail/inquiry/2003-March/000115.html 02. http://stderr.org/pipermail/inquiry/2003-March/000116.html 03. http://stderr.org/pipermail/inquiry/2003-March/000117.html 04. http://stderr.org/pipermail/inquiry/2003-March/000118.html 05. http://stderr.org/pipermail/inquiry/2003-March/000119.html And I will send the Exe file to anybody who wants it. Jon Awbrey o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o

1 0

ANNOUNCE: hat-2.02
by Malcolm Wallace 26 Mar '03

26 Mar '03

hat-2.02 -------- http://www.haskell.org/hat/ We are pleased to announce a new release of Hat, the Haskell Tracer. Hat is a very useful tool for understanding and debugging programs. Hat is compiler independent, and can be used on any Unix-like platform. This new release, hat-2.02, can be used with the compilers ghc-5.04.2, nhc98-1.16, upwards. Particular features of note in this release are: * Support for hierarchical module namespaces. (The Hat distribution also includes a tracing version of a subset of the base package of standard hierarchical libraries.) * Support for common type-system extensions (if your compiler also supports them. These include: multi-parameter type classes, functional dependencies, and existential quantification of datatype declarations. * A significant speed improvement (20-40%) of traced programs over the previous release. (Plus, if you build Hat with ghc -O, you can get another 20-40% performance boost.) * We believe Hat now works under Windows (provided you have Cygwin, and some of the X toolset - please check the requirements listed on the website carefully). It continues to work under MacOS-X as well (provided you have Apple's X toolset). There are also lots of more minor improvements in the viewing tools. For instance: * The viewers now support qualified name syntax. * Named field constructions and updates are now displayed correctly. * List sugaring is fixed. * Large expressions that scroll off the display are now fixed. * Lots of other small bugfixes. See the website http://www.haskell.org/hat/ for full details. There is a dedicated mailing list for bug-reports and other discussion of Hat: http://www.haskell.org/mailman/listinfo/hat Regards, The Hat team at York [ The Hat team consists of Malcolm Wallace, Olaf Chitil, and Colin Runciman, with thanks to many other students, visitors, collaborators and reviewers. The development of Hat was funded by EPSRC grant number GR/M81953, and the latest release was partially supported by the University of York and Microsoft Research (Cambridge). ]

1 0

CFP: JFP Special Issue on Parallel Functional Programming
by Phil Trinder 26 Mar '03

26 Mar '03

Journal of Functional Programming Special Issue on Functional Approaches to High-Performance Parallel Programming Guest editor: Phil Trinder Submission deadine: 26th September 2003 Web Page: http://www.cee.hw.ac.uk/~trinder/jfpCFP.htm High-performance parallel programs are hard to write. For a long time it has been recognised that ideas and approaches drawn from functional programming may be particularly applicable in this area because * Concurrent stateless computations are much easier to coordinate. * High-level coordination reduces programming effort. * Declarative notations are amenable to reasoning, i.e. to transformation, derivation and analysis. After a long gestation, this potential is now finally being realised in practice. Declarative techniques are being used to construct significant parallel and high-throughput systems, e.g. real-time image analysis and high-end telephone exchanges. Many production systems use conventional imperative technologies in a declarative way, e.g. coordinating legacy Fortran code. Furthermore, the recent emergence of computational Grids pose new challenges that the functional paradigm is well-placed to address: grids offer enormous amounts of computing power, but require sophisticated and dynamic management that is hard to provide in a low-level paradigm. A special issue of the Journal of Functional Programming will be devoted to functional approaches to high-performance parallel programming. Full-length, archival-quality submissions are solicited on topics including but not limited to the following. * Skeletons and higher-order parallel program construction * Parallel Program development methodologies, including derivation and optimisation. * Parallel program modelling, including static analyses to predict properties (e.g. cost models) and dynamic analyses to profile execution properties. * Parallel functional languages, including design, semantics, implementation and performance. * High-performance functional and skeleton applications. * Adaption of high-performance functional technologies for emerging architectures such as the Grid. Papers covering a fusion of functional and more conventional technologies are encouraged, in addition to papers on purely functional technologies. Submissions should be sent to the guest editor Phil Trinder (trinder(a)macs.hw.ac.uk) with a copy to Jenny Parker (jparker(a)cambridge.org) Submitted articles should be sent in PDF or Postscript format, preferably gzipped. In addition please send as plain text: title, abstract, and contact information. The submission deadline is 26th September 2003, and for other submission details, please consult an issue of the Journal of Functional Programming or see the Journal's web pages. -------------------------------------------------- Phil Trinder School of Mathematical and Computer Sciences Heriot-Watt University Edinburgh, EH14 4AS E-mail: trinder(a)macs.hw.ac.uk Teleph: +44 (0)131 451 3435 Depart: +44 (0)131 451 3328 Fasmly: +44 (0)131 451 3327 Intrnt: http://www.macs.hw.ac.uk/~trinder

1 0

Arbitrary precision reals?
by Niall Dalton 26 Mar '03

26 Mar '03

Hi, Its been a while since I've been using Haskell seriously, so I might simply have overlooked the answer.. Is there an arbitrary precision real number library available for Haskell? IIRC, atleast GHC uses the GMP library, but only for integers? Thanks, niall ------------------------------ This e-mail is intended for the named addressee only. It may contain confidential and/or privileged information. If you have received this message in error, please let us know and then delete this message from your system. You should not copy the message, use it for any purpose or disclose its contents to anyone. ________________________________________________________________________ This e-mail has been scanned for all viruses by Star Internet. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ________________________________________________________________________

5 4

RE: Arbitrary precision reals?
by Simon Peyton-Jones 26 Mar '03

26 Mar '03

David Lester, who I saw today by coincidence, has a very nice arbitrary-precision real arithmetic package, which he is planning to finish up and offer to us Haskellites. I'll leave him to say when, and what it does, but it's a very cool thing and I hope we can persuade him to give it to us soon. Simon | -----Original Message----- | From: Tom Pledger [mailto:Tom.Pledger@peace.com] | Sent: 24 March 2003 21:14 | To: haskell(a)haskell.org | Subject: Arbitrary precision reals? | | | Niall Dalton writes: | | Hi, | | | | Its been a while since I've been using Haskell seriously, | so I might simply | have overlooked the answer.. | | | Is there an arbitrary precision real number library | available for Haskell? | IIRC, atleast GHC uses the GMP | library, but only for integers? |

1 0

RE: Haskell help!
by Weix, Rachel Lynn 25 Mar '03

25 Mar '03

I have since been able to figure out my problems - the algorithm for one of my methods was wrong and I found two "typos" so to speak. I now need to figure out how to return all optimal solutions, and I have an idea as to how to start. If I have any more problems I may send out another email. Rachel P.S. I need to solve the problem using functional programming, thus I cannot use looping structures or mutation. -----Original Message----- From: Iavor S. Diatchki [mailto:diatchki@cse.ogi.edu] Sent: Tue 3/25/2003 2:45 PM To: Weix, Rachel Lynn Cc: haskell(a)haskell.org Subject: Re: Haskell help! hi, i think you might find the following paper relevant: "Algebrainc Dynamic Prorgamming" by Robert Gigerich and Carsten Meyer http://link.springer-ny.com/link/service/series/0558/papers/2422/24220349.p… bye iavor Weix, Rachel Lynn wrote: > Hi, > > I'm a college student trying to write a Haskell program, and I'm having > some problems getting the correct output. I need to write a program > which will return a set of optimally aligned sequences, with "optimal" > being defined as such: mismatch or space (represented by a "-") = -1; > else 1 for every pair (see below for what a pair is) > > An example set of sequences with an alignment of -3 is as follows: > > sc-h-e-me > icecream- > > (s,i) are a pair, (c, c), etc. > > For right now, I'm only returning a potential solution (vs. a set of > solutions), and I'm having problems with any set of sequences that > aren't the same length. My results are listed below. I don't want to > include the code for now since it's about a page long (printed out) but > was hoping maybe someone had an idea as to why I'm getting such odd > results? If code will help, I can exchange back and forth with whoever > thinks they might be able to help me out. Your help is much > appreciated! > > Rachel > > > -- Each word/sequence you see has been predefined in my Haskell code > -- This first example appears to work just fine > > Main> printSeq icecream scheme > > ("icecream", "scheme--") > > > > Main> printSeq hate hatter > > ("hat-e", "hatte") > > > > Main> printSeq scheme saturn > > ("scheme", "saatur") > > > > Main> printSeq saturn scheme > > ("saaturn", "scheme-") > > > > Main> printSeq saturn hatter > > ("saaturn", "hatter-") > > > > Main> printSeq hatter saturn > > ("hatter", "saatur") > > > > Main> printSeq mad saturn > > ("mad", "saa") > > > > Main> printSeq hate hatter > > ("hat-e", "hatte") > > > > Main> printSeq snowball icecream > > ("snowb-all", "icecream-") > > > > Main> printSeq mad computer > > ("--mad", "compu") > > > > Main> printSeq mad snowball > > ("mad", "sno") > > -- ================================================== | Iavor S. Diatchki, Ph.D. student | | Department of Computer Science and Engineering | | School of OGI at OHSU | | http://www.cse.ogi.edu/~diatchki | ==================================================

2 1

Haskell help!
by Weix, Rachel Lynn 25 Mar '03

25 Mar '03

Hi, I'm a college student trying to write a Haskell program, and I'm having some problems getting the correct output. I need to write a program which will return a set of optimally aligned sequences, with "optimal" being defined as such: mismatch or space (represented by a "-") = -1; else 1 for every pair (see below for what a pair is) An example set of sequences with an alignment of -3 is as follows: sc-h-e-me icecream- (s,i) are a pair, (c, c), etc. For right now, I'm only returning a potential solution (vs. a set of solutions), and I'm having problems with any set of sequences that aren't the same length. My results are listed below. I don't want to include the code for now since it's about a page long (printed out) but was hoping maybe someone had an idea as to why I'm getting such odd results? If code will help, I can exchange back and forth with whoever thinks they might be able to help me out. Your help is much appreciated! Rachel -- Each word/sequence you see has been predefined in my Haskell code -- This first example appears to work just fine Main> printSeq icecream scheme ("icecream", "scheme--") Main> printSeq hate hatter ("hat-e", "hatte") Main> printSeq scheme saturn ("scheme", "saatur") Main> printSeq saturn scheme ("saaturn", "scheme-") Main> printSeq saturn hatter ("saaturn", "hatter-") Main> printSeq hatter saturn ("hatter", "saatur") Main> printSeq mad saturn ("mad", "saa") Main> printSeq hate hatter ("hat-e", "hatte") Main> printSeq snowball icecream ("snowb-all", "icecream-") Main> printSeq mad computer ("--mad", "compu") Main> printSeq mad snowball ("mad", "sno")

3 2