February 2006 - Haskell-Cafe

Re: [Haskell] Re: Streams: the extensible I/O library
by Bulat Ziganshin 07 Feb '06

07 Feb '06

Hello Peter, Monday, February 06, 2006, 10:24:01 PM, you wrote: moving to nescafe :) PS> > You can find further information about the library at the PS> > page http://haskell.org/haskellwiki/Library/Streams and PS> > download it as http://freearc.narod.ru/Streams.tar.gz PS> Is there any chance of running this code on a non-Windows PS> system? I tried to compile the example programs, but failed PS> for lack of a System.Win32 module. I might be able to port PS> the code, but I figured it would be wise to ask first. of course, it should work under other systems! it's my fault and you can fix this just by adding #ifdef around this import: #if defined(mingw32_HOST_OS) || defined(__MINGW32__) || defined(_MSC_VER) import System.Win32 #endif btw, my "wc" has about the same speed as yours :) thank you for the first bug-report! i re-uploaded the fixed version just now :) -- Best regards, Bulat mailto:bulatz@HotPOP.com

2 2

main features of functional programming paradigm
by Abigail 06 Feb '06

06 Feb '06

Hi, I would like to ask about what are tha main features of functional programming? Additionally, about FAD: A Functional Analysis and Design Methodology What kind of software can i develop with this methodology? thanks for your answers Abigail. __________________________________________________ Correo Yahoo! Espacio para todos tus mensajes, antivirus y antispam ¡gratis! Regístrate ya - http://correo.espanol.yahoo.com/

3 2

Re: [Haskell-cafe] Why is $ right associative instead ofleftassociative?
by John Hughes 06 Feb '06

06 Feb '06

> Quoting Paul Hudak <paul.hudak(a)yale.edu>: > >> Actually, one of the main reasons that we chose (:) is that that's what >> Miranda used. So, at the time at least, it was not entirely clear what >> the "de facto universal inter-language standard" was. > Phil Wadler argued for the ML convention at the time, and wrote a document containing a fair amount of sample code to illustrate what it would look like. We noticed something surprising: instead of (x:xs) and the like, Phil had consistently written (x :: xs) -- note the extra spaces. Somehow, using the longer operator name led him to feel spaces were needed around it. That in turn made his lines longer, encouraged him to split definitions across lines, and so on. When I read the thing, I realised after a while that I was skipping all the code fragments -- because they just looked too big and complicated to take in during a quick reading. It was at least partly that experience that convinced us that using :: for cons would impose a small cost, but a real one, on readability. It may seem trivial, but the sum of many such decisions is significant. The story does illustrate the importance of actually trying out syntactic ideas and seeing how they play--one can be surprised by the result. >> I don't think Haskell Prime should be about changing the look and >> feel of the language. It's about consolidating the most important extensions into the standard, isn't it? Changes that break existing code should be very, very well motivated--if porting code to Haskell Prime is too difficult, people just won't do it. John

11 16

Haskell to call Microsoft COM (Dispatch)
by Gracjan Polak 05 Feb '06

05 Feb '06

Hi all, Is there any library to make Haskell call Microsoft COM functions using Dispatch? E.g I don't need the full COM binary functionality, scripting is enough. Google didn't seem to find anything interesting... beside rolling my own using FFI :) Thanks in advance! -- Gracjan

5 6

Re[2]: [Haskell-cafe] request for help
by Bulat Ziganshin 04 Feb '06

04 Feb '06

Hello Marco, Saturday, February 04, 2006, 4:39:23 PM, you wrote: MAFdA> Hello. MAFdA> English is not my native language (I'm Portuguese). MAFdA> The haskell-cafe mailing list is usually very active and subscribers MAFdA> do help each other, so I believe you will not have any trouble in MAFdA> finding "real" Englishmen who will read and correct the document ;-) thank you. J.P. Bernardy already corrected a part, so i think that Englishmens will help me MAFdA> However, if by any chance you don't, please let me know and I will do it MAFdA> (just tell me how, because the wiki is password-protected). you just need to register itself in this Wiki system. look at the "register" text on the page, i don't remember exact. or may be "login", where you will have chance to register -- Best regards, Bulat mailto:bulatz@HotPOP.com

1 0

Re: Haskell code for this example of flow control
by Dominic Steinitz 04 Feb '06

04 Feb '06

Here are some even older discussions on the subject. I don't know if anyone ever put them into a library or on the wiki. Dominic. http://haskell.org/pipermail/haskell-cafe/2005-May/009784.html http://www.haskell.org//pipermail/libraries/2005-February/003143.html

1 0

request for help
by Bulat Ziganshin 04 Feb '06

04 Feb '06

Hello haskell-cafe, i'm wrote new general i/o library called Streams. it's so great that i hope it will eventually replace using Handles. i plan to present it in Haskell list on Monday. in order to do it i wrote overview of library facilities. the problem is that i'm not native english speaker and i feel that my text is awkward, in best case. when i wrote last time docs about Arrays, Cale Gubbard fixed lots of my misspellings on wiki page :) this time i try to go different way - i published my version of report at the http://haskell.org/haskellwiki/Library/Streams page and ask your help in correcting it. it will allow to include in presentation the proper English text that is much more pleasing for readers than my scribbles :) so, please correct text at this wiki according to English rules. even if you correct paragraph or two, that will forward the work further -- Best regards, Bulat mailto:bulatz@HotPOP.com

1 0

Pythonic monads
by Graham Klyne 03 Feb '06

03 Feb '06

Constructing some code today in Python, using some functional-style coding idioms, I found myself wondering if there would be any real benefit to using a monad-based implementation (i.e. other than to demonstrate that it can be done). The application that sparked this line of thought was a simple filter to trim comments and whitespace out of an XML document. I ended up with a simple state-machine driven filter, thus: [[ # Strip comments from XML strings def stripXmlComments(x): # State table for stripping XML comments stStripXmlComments = \ { 0: match1('<',(1,stateFilterHold),(0,stateFilterPass)) , 1: match1('!',(2,stateFilterHold),(0,stateFilterPass)) , 2: match1('-',(3,stateFilterHold),(0,stateFilterPass)) , 3: match1('-',(4,stateFilterDrop),(0,stateFilterPass)) , 4: match1('-',(5,stateFilterDrop),(4,stateFilterDrop)) , 5: match1('-',(6,stateFilterDrop),(4,stateFilterDrop)) , 6: match1('>',(0,stateFilterDrop),(4,stateFilterDrop)) } return stateFilter(stStripXmlComments,x) # Simple state machine driven filter # # The state table is a dictionary indexed by state values, where the # initial state is 0, and each entry is a function that accepts a next # symbol and returns a pair of (next state, action), where action is # one of 'stateFilterPass', 'stateFilterDrop', 'stateFilterHold'. # stateFilterHold means that the disposition will be determined later. # # The result is an iterator that returns elements from the filtered # subsequence of the supplied sequence. # def stateFilter(stable,seq): queue = [] state = 0 for symbol in seq: (state,action) = stable[state](symbol) (queue,emit) = action(queue,symbol) for e in emit: yield e return def stateFilterPass(q,n): return ([],q+[n]) def stateFilterDrop(q,n): return ([],[]) def stateFilterHold(q,n): return (q+[n],[]) # State transition function to match the specified symbol and return # 'eqval' if matched, otherwise 'neval' def match1(sym,eqval,neval): def m(sym,eqval,neval,next): if next==sym: return eqval return neval return curry(m,sym,eqval,neval) def curry(func, *args): """ Curry multiple arguments: See: http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/229472 """ def curried(*args2): args2 = args + args2 return func(*args2) return curried ]] and a test case: [[ def testFilter02(self): fullstr = "<!-abc->  <!- -ghi-->" trimstr = list(stripXmlComments(fullstr)) expstr = list("<!-abc-> <!- -ghi-->") assert trimstr==expstr, \ "stripSpaces, expected:\n"+expstr+"\nFound:\n"+trimstr ]] In thinking about this implementation, it seemed to me that this employed patterns characteristic of a monadic type: each entry in the state table (in this case, an instance of match1, a curried function) is like a step in a monadic computation, updating the monadic value and also returning some value. What I can't quite visualize is if the code in Python would actually look any better if it were implemented with a monadic type, as one might readily choose for a Haskell implementation. Or would there be no real benefit? I have noticed that, while I like to use functional idioms in some of my Python code, and the Python language is easily able to support these (even some lazy evaluation, courtesy of generators), that the code doesn't always look as clean as its Haskell equivalent. In Haskell, composition and currying are fundamental patterns and are directly supported by the syntax. In Python, one has to work harder to achieve these (e.g. the "curry" function above seems rather convoluted to me, for such a fundamental notion). Thoughts? Comments? #g -- Graham Klyne For email: http://www.ninebynine.org/#Contact

2 1

Haskell code for this example of flow control
by Maurício 03 Feb '06

03 Feb '06

Hi, I would like to know what options I have in Haskell to do something similar to this C++ code: double a = 1000; while (a>1) a/=2; I'm able to do that with lists, but I would like to know how to do that with monads and variables with state. Thanks, Maurício

7 9

HUnit
by Matt Roberts 03 Feb '06

03 Feb '06

I am in love with HUnit and QuickCheck. However, I am missing one important functionality I want to call a command line program that will search all the .hs files I have in a particular directory, pull out all HUnit test (based on what they are called, or perhaps their type signature) and run them all. I have done a bit of a search and can't find anything like this, am I going to have to write it myself? Matt

3 2