- Haskell-Cafe - Haskell.org

More help for the newb
by Matthew Morvant 21 Feb '04

21 Feb '04

I have gotten past the syntax errors and am now into type errors. Hopefully if I get some help with one I can fix the rest on my own. Here is the funct 36 code3 :: [b] -> [Char] 37 code3 = do{char '!'; digit; char 's'; many (do char 's'); return "Suc"} +++ do{char '%'; char 's'; digit; many (do digit): return "Suc"} ERROR "C:\Documents and Settings\mmorvant\Desktop\haskell.hs":37 - Type error in application *** Expression : many do {...} : return "Suc" *** Term : many do {...} *** Type : Parser [Char] *** Does not match : [Char]

1 0

Rational Haskell?
by Justin Walsh 21 Feb '04

21 Feb '04

Please bear with me. Point of departure: My goal is OWL (but replacing the "O" for Ontological with Noological) http://www.bright.net/~jclarke/kant/history.html 2. Noology (as in noumenon or "state of mind") i.e. rational http://en.wikipedia.org/wiki/Ontology Zoology (as in phenomenon or state of body) i.e. Empirical Declaration: None of these OO languages in their current context seem to work for me. Observations: I love AWK, I don't like C++, I Like plain vanilla Prolog and Smalltalk (as procedural and declarative) IDE's, but in there current form useless for my purposes. I'm unsure of Haskell but, feel it has untapped potential at a meta level. So far I have found, that Haskell is a great "what-if" language (the comparison with the spread sheet correct). I am very much a remnant of the "how-to" generation (one cannot get much lower than assembler). I found it difficult to take responsibility and let go of my "goto". But I do need to reach the "when-to" level: my project is Virtual Reality and SAI:(asIn Silicon Assisted Intelligence is gnot AI). Somehow though all of us early (infant) programmers moved pretty seamlessly form goto's and flowcharting to pseudo-code and structure programming techniques: because it works. However in 1989 I did BR and Semantic Modeling (Debenham and Montgomery UTS), realising immediately that it did not work. Since then I still see the Systems Designers equivalent of the flowchart and the goto i.e Entity Relationship Modeling http://www-106.ibm.com/developerworks/rational/library/319.html Why is it so hard for SysDesigners to make the same mind shift we did? Sorry if I sound rude but I'm stuck in an OO rut and I need to reach Process Orientation rather quickly. Any serious suggestions (apart from "just go away")? Justin PS: I don't like dialectical bickering over detail. Please spare me the flaming --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com) Version: 6.0.585 / Virus Database: 370 - Release Date: 2/11/04

1 0

Help a Newbie
by Matthew Morvant 21 Feb '04

21 Feb '04

I am trying (and have been for quite some time) to repair a broken Haskell script. I keep getting stuck on Last generator in do {...} must be an expression. I am betting that this is not uncommon. Can someone please help me understand? code5 :: Parser code5 = do dd <- code3 do ddd <- code3 do dddd <- code3 return "ce" code4 :: Parser code4 = do dd <- code3 do dde <- code3 digit ddd <- many (do digit) q <- code5 return (q : "ss") Thanks in advance for helping Matthew

3 2

RE: [Haskell-cafe] setPermissions bug?
by Simon Marlow 21 Feb '04

21 Feb '04

> On Wed, Feb 18, 2004 at 12:31:34PM -0000, Simon Marlow wrote: > > > I presume that this is a bug in setPermissions? > > > > Yes, it's a bug. However, since > getPermissions/setPermissions doesn't > > really give you enough control in Unix, I suggest you use > the functions > > available from System.Posix instead (fileStatus, fileMode, > setFileMode > > etc.). > > Well, in my case all I want to do is to copy permissions from > one file to > another, so get/setPermissions would have been fine. But yeah, I've > already switched to using setFileMode etc. Actually if you want to copy permissions from one file to another, then get/setPermissions is *not* what you want. It will only copy across the owner's permissions. The problem is that the Permissions type is fully defined and non-abstract; there's no place to store extra platform-specific permission bits. Cheers, Simon

1 0

hOp
by sebc＠macs.hw.ac.uk 20 Feb '04

20 Feb '04

The RTS of GHC is truly an amazing piece of software. By removing the system-specific bits, and just adding a few lines of C and assembly code, we were able to turn it into a standalone micro-kernel that can be extended in our favorite language. A screenshot is available here: http://etudiants.insia.org/~jbobbio/hOp-20040219.png Everything from the base hierarchical library not under the System subdirectory is available. This includes threads and communication primitives and the foreign interface, which means experimenting with writing drivers in Haskell should be reasonably easy. We hope to make a source release soon. -- Sébastien Carlier Jérémy Bobbio

1 1

[Q] What calls what
by Andre Betz 19 Feb '04

19 Feb '04

Hi, I have a question about the following function. I don't know how the definition after "where" are called. For example, in ( D x2 (-y1) , b ++ b2 ) there is x2 defined in the where-block. So does that mean that it calls the iteratin at line 7? But what is with b2, does it also calls the line 7? Who calls the definition in line 3 and who calls the def in line 7? Many thanks Andre 1 test :: ( D , [Int] ) -> ( D , [S] ) -> ( D , [Int] ) 2 test ( D x y , b ) ( D dx dy , st ) = ( D x2 (-y1) , b ++ b2 ) 3 where ( x1 , y1 ) = head $ dropWhile ( ( > 0 ) . snd ) $ iterate inc ( x , y ) 4 5 inc ( n , m ) = ( n+1 , m+1 ) 6 7 ( b2 , x2 ) = head $ do 8 S ( D x3 sty ) pl _ sb <- st 9 guard $ sty == -y1 10 return ( [1,2,3] ++ sb , x3 + sb) -- ********************************************************** ** * André Betz * * * ---------- * * * EMail: mail(a)andrebetz.de * WW * * Web: http://www.andrebetz.de * /__\ * * ICQ: 117759780 * | oo | _WWWWW_ * * Mobil: +49 (0)162 8847 391 * (|_()_|) / o o \ * * Office: +49 (0)9131 82611 27 * \__/ (| __O__ |) * * Home: +49 (0)911 6695 558 * /|\/|\ \ \___/ / * * Tel./Fax: +49 (0)89 1488 224189 * |||||||| /-------\ * * Snail: Hirschelgasse 34 * |||||||||=========| * * D-90475 Nürnberg * |||||||||=========| * ********************************************************** **

1 0

Re: [Haskell] Parsec question: attempted 'notMatching' combinator
by Graham Klyne 18 Feb '04

18 Feb '04

[To Haskell-cafe...] At 16:57 17/02/04 -0500, Andrew Pimlott wrote: >On Tue, Feb 17, 2004 at 07:48:52PM +0000, Graham Klyne wrote: > > Thanks! That got me going, though not with quite what you suggested. > > > > I ended up with this: > > [[ > > notMatching :: Show a => GenParser tok st a -> GenParser tok st () > > notMatching p = do { a <- try p ; unexpected (show a) } <|> return () > > ]] > > which does the required job for me. > >Oops, that does look better. I knew the try had to go somewhere. :-) > >The only remaining problem is when p succeeds but does not consume any >input, eg eof. In this case, the <|> return () cannot distinguish it >from p failing. > >I didn't realize at first what a dirty trick this function uses: It >distinguishes success or failure of p by whether any input from the >first part (before <|> return ()) was consumed. I don't think it is >possible (or desirable!) to get this approach 100% right: By "erasing" >the success of p, you lose the information you need. > >What about a more prosaic implementation: > > notFollowedBy' :: Show a => GenParser tok st a -> GenParser tok st () > notFollowedBy' p = do res <- do a <- try p; return $ Just a > <|> > return Nothing > case res of Just a -> unexpected (show a) > Nothing -> return () I don't see why that would work where the above case does not; i.e. when p consumes no input. Or indeed, your later, preferred version: [[ notFollowedBy' :: Show a => GenParser tok st a -> GenParser tok st () notFollowedBy' p = join $ do a <- try p; return (unexpected (show a)) <|> return (return ()) ]] >This works for the tests I've tried, but there's one little quirk with >error reporting: I never really cared so much about that bit! #g -- > aNoBC = do char 'a' > notFollowedBy' $ do char 'b'; char 'c' > > *Main> parseTest (aNoBC >> char 'e') "abe" > parse error at (line 1, column 2): > unexpected "e" > expecting "c" or "e" > >It seems that parsec both misreports which token is unexpected (should >be "b", and thinks that the failure to match "c" is a problem, even >though notFollowedBy' succeeded. > > > Using your version caused the notMatching parser to be equivalent to: > > return () > > presumably, because the failure was protected by the try combinator? > >Or perhaps more accurately, the success (of p) was protected by the try! >Ie, the unexpected and the try together undid the tell-tale token >consumption of p. > >Andrew ------------ Graham Klyne For email: http://www.ninebynine.org/#Contact

2 1

RE: [Haskell-cafe] setPermissions bug?
by Simon Marlow 18 Feb '04

18 Feb '04

> I've run into a problem using Directory.setPermissions (or > equivalently System.Directory.setPermissions). > > Basically the problem is that > > getPermissions f >>= setPermissions f > > is not a noop. Instead it is equivalent to > > chmod o-r f > chmod g-r f > etc > > I presume that this is a bug in setPermissions? Yes, it's a bug. However, since getPermissions/setPermissions doesn't really give you enough control in Unix, I suggest you use the functions available from System.Posix instead (fileStatus, fileMode, setFileMode etc.). Cheers, Simon

2 1

setPermissions bug?
by David Roundy 18 Feb '04

18 Feb '04

I've run into a problem using Directory.setPermissions (or equivalently System.Directory.setPermissions). Basically the problem is that getPermissions f >>= setPermissions f is not a noop. Instead it is equivalent to chmod o-r f chmod g-r f etc I presume that this is a bug in setPermissions? The documentation claims that Note that to change some, but not all permissions, a construct on the following lines must be used. makeReadable f = do p <- getPermissions f setPermissions f (p {readable = True}) Which certainly implies that it is possible to change only some permissions, contrary to what happens in the actual implementation. It also seems that the Permissions data type could include in it additional (hidden, since it isn't portable perhaps) information in order to make the above code work. Then the only major question would be whether setPermissions f (p {writeable = True}) should make the file world-writeable or just user-writeable. I don't know about that. But at a minimum, it should preserve world-readability. -- David Roundy http://www.abridgegame.org

1 0

Re: [Haskell] updating graphs non-destructively
by Graham Klyne 18 Feb '04

18 Feb '04

[to Haskell-cafe] At 19:29 17/02/04 -0500, S. Alexander Jacobson wrote: >Thank you for the link to FGL. I also looked at >the boilerplate stuff. If *feels* like there >should be a way to embed the graph stuff in the >boilerplate stuff to allow non-destructive update >of arbitrary object graphs without handcoding the >mapping of the datastructure to an object graph? > >Is this possible? Has anyone does this? Any >reason to believe its a bad idea? > >Alternatively, I could decide up front to >represent my data as RDF triples (amenable to a >graph system), but I'd rather take advantage of >Haskell's type system. Hmmm (speaking as an RDF advocate) ... that seems like an odd reason for using RDF. Much of the value of RDF is, I would say, as means to exchange information *between* programs rather than for internal structuring of data. An RDF-graph-based interface to data provides for some uniformity of access and processing that can be useful. In my own work, I have at least two different internal representations for RDF data (neither of which have been designed for efficiency). #g ------------ Graham Klyne For email: http://www.ninebynine.org/#Contact

1 0