- Haskell-Cafe - Haskell.org

RE: Type design question
by Hal Daume 31 Jul '03

31 Jul '03

Hi Konrad, > I am a Haskell newbie working on my first serious test case, > and I would like > some feedback from the experts to make sure I am not doing > anything stupid > ;-) Well, I may not exactly qualify, but I can give you a few suggestions, nonetheless... > data Floating a => Vector a = Vector !a !a !a > deriving (Eq, Show) Here. Do not use 'data C x =>'. It's essentially a useless part of the language that really should have gone away. What you want is just: | data Vector a = Vector !a !a !a deriving (Eq, Show) The Floating constraint can come in on the functions. (I say this feature is useless because it only does one thing: disallows you to construct Vectors whose elements aren't instances of Floating. However, the useful thing that it might do -- namely, allow you to *use* the fact that the elements are instances of Floating -- it doesn't.) > I would like to keep the numerical representation as general > as possible, in > particular to postpone decisions about precision (Float, > Double, eventually > arbitrary precision arithmetic) as much as possible. > Therefore the "Floating > a" constraint. Which leads to my first question: doing it the > way I did, I > find myself having to add a "Floating a" constraint to almost > every function > specification with a vector argument. I would prefer to > specify once and for > all that vector elements need to be "Floating", everywhere. > Is there a way of > doing that? Not really. > -- Periodic universe > data Floating a => OrthorhombicUniverse a = OrthorhombicUniverse a a a Again, get rid of the 'Floating a =>' part. > instance Floating a => Universe (OrthorhombicUniverse a) where > distanceVector (OrthorhombicUniverse a b c) > (Vector x1 y1 z1) (Vector x2 y2 z2) > = Vector (fmod (x2-x1) a) (fmod (y2-y1) b) (fmod (z2-z1) c) > where fmod x y = x - y*truncate (x/y) There are actually two problems. First, you want to write fmod as: | fmod x y = x - y * fromInteger (truncate (x/y)) otherwise the type is wrong. Secondly, since you're using truncate, you need to be in RealFrac, not in Floating. Changing the signature to: | instance RealFrac a => ... fixes this problem. However, once we fix this, we can see the real problem. Your Universe class has a method, distanceVector, of type: | distanceVector :: Universe u, Floating a => u -> Vector a -> Vector a -> Vector a And here's the problem. When 'u' is 'OrthorhombicUniverse x', it doesn't know that this 'x' is supposed to be the same as the 'a'. One way to fix this is to parameterize the Universe data type on the element, something like: class Universe u where distanceVector :: (RealFrac a, Floating a) => u a -> (Vector a) -> (Vector a) -> (Vector a) distance :: (RealFrac a, Floating a) => u a -> (Vector a) -> (Vector a) -> a distance u v1 v2 = norm (distanceVector u v1 v2) -- Infinite universe data InfiniteUniverse a = InfiniteUniverse instance Universe InfiniteUniverse where distanceVector u v1 v2 = v2 <-> v1 -- Periodic universe data OrthorhombicUniverse a = OrthorhombicUniverse a a a instance Universe OrthorhombicUniverse where distanceVector (OrthorhombicUniverse a b c) (Vector x1 y1 z1) (Vector x2 y2 z2) = Vector (fmod (x2-x1) a) (fmod (y2-y1) b) (fmod (z2-z1) c) where fmod x y = x - y*fromInteger (truncate (x/y)) This is beginning to get a little ugly, but I think that's largely because you're using type classes in too much of an OO style (this is what a lot of my first programs looked like and I think for reasons similar to this, I moved away from that). However, not knowing more about the domain of discourse, I don't know what to suggest. Perhaps using the module system would serve you better. Instead of having a universe class, simply put each universe in a separate module and provide the same functions in each. Then you just import them selectively, or qualified. HTH. - Hal

15 24

RE: Useful list output function
by Kevin S. Millikin 31 Jul '03

31 Jul '03

On Wednesday, July 30, 2003 1:52 PM, Shawn P. Garbett [SMTP:Shawn@Garbett.org] wrote: > > output :: (a -> IO b) -> [a] -> IO () > output f = (foldr (>>) (return ())).(map f) Study the Prelude :) > output :: (a -> IO b) -> [a] -> IO () > output = mapM_

1 0

Useful list output function
by Shawn P. Garbett 31 Jul '03

31 Jul '03

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've got a little function which I derived from Bird's book "Intro to Functional Programming using Haskell". Basically I had several instances of lists that needed outputing with all kinds of text decoration. This function was used repeatedly throughout the code, by removing all the cut and paste instances, this was the common function: output :: (a -> IO b) -> [a] -> IO () output f = (foldr (>>) (return ())).(map f) So given a function that transforms a value of type 'a' into an output of 'IO' and a list of 'a', all IO is performed across the list. Example usage: output f [1,2,3] where f a = putStr ("List contains "++(show a)++"\n")) - -- List contains 1 List contains 2 List contains 3 Probably not very exciting for the old hats at Haskell, but pretty handy to have around at my level of expertise. Shawn -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iEYEARECAAYFAj8oIdAACgkQDtpPjAQxZ6AC0ACaA8boOZc8Hjmp9rTAyZ00lPLX QwkAniPsBPHUvBfUcsZlU6TvF1pzkTox =3aEG -----END PGP SIGNATURE-----

1 0

RE: Parsing CSV files
by Simon Peyton-Jones 30 Jul '03

30 Jul '03

| > Hey, I wrote a CSV parser too. seeing as how this seems like a common | > thing (having 3 independent implementations at least), perhaps it | > belongs in libraries somewhere? some area dedicated to useful little | > grammers would be handy. csv, c header files, .x (rpcgen), various | > preference file formats, etc... | | What haskell needs is something along the lines of CPAN. Contributed modules | of various utility. Not part of the base libraries, but never-the-less useful | and maybe needed for your project. There's a lot of activity on this front on the libraries mailing list (libraries(a)haskell.org) http://haskell.org/hawiki/LibraryInfrastructure Simon

1 0

RE: functional graph
by Hal Daume 30 Jul '03

30 Jul '03

> But it looks the Graph class works on types a and b > where a is the Node type where b is the Edge type, or > just the other way around :), I don't have the code > right now. Yes, 'Graph n e' is a graph whose nodes are labelled with elements of type n and whose edges are labelled with elements of type e. > The above rules out my way of representing the data, > or doesn't it (i.e. is there some fancy, magical way > of doing this) ? Yes. (I.e., it rules it out.) > The second intuitive solution is to take a datatype > like > data GraphNode|Teacher etc > |Subject etc > |Student etc > > Should this be the way to do it? Yes. > And then yet another question: I need some kind of way > to remember, which possibilities I already had. > > In example in my C++ program I used a couple of > array's with in each array pointers to the > StudentObjects, TeacherObjects. You could use the same here, or use a FiniteMap. If i'm understanding you correctly you miss the fact that there's no way to look up a node given it's label. This is because the graph module doesn't assume that node labels are unique (otherwise it wouldn't need to generate the Node labels itself). I generally bundle a 'Graph n e' together with a 'FiniteMap n Node' so that I can look up Node identifiers based on node labels. > But then I yet have another problem. How can I > guarantee, to do a complete path traversal, each time > in exactly the same way? I don't understand what you mean. In your case, what is a "complete path traversal." Regardless, if you have the same graph and the same functions for traversing it, since Haskell is referentially transparent you should also "traverse it in the same way..." - Hal

2 1

Parsing CSV files
by Shawn P. Garbett 30 Jul '03

30 Jul '03

I did a small search for parsing a comma seperated file in Haskell and didn't find anything-- so I put together some code to do this. It doesn't handle whitespace very well, this would be a nice addition if someone has an idea out there. Also the rows method, I had trouble just using two do loops without brackets, I couldn't prevent it from being ambiguous. Anyway here's the code: ------------------------- import Parsec import System.IO -- Code to parse a comma seperated file content :: Parser String content = many1 (noneOf "\"") cell :: Parser String cell = do char '\"' w <- content; char '\"' <?> "end of cell" return w; <|> return "" -- Empty cell separator :: Parser () separator = skipMany1 (char ',') cells :: Parser [String] cells = sepBy1 cell separator contents :: Parser [String] contents = sepBy1 content separator; rows :: Parser [[String]] rows = do c <- cells newline do { cs <- rows; return (c:cs); } <|> return [c] csv :: Parser [[String]] csv = do r <- rows eof return r -- Main routine main :: IO () main = do result <- parseFromFile csv "simple.txt" case (result) of Left err -> print err Right sb -> print sb ------------------- Shawn Garbett

6 7

Using exceptions in imperative code
by Bayley, Alistair 29 Jul '03

29 Jul '03

I've got resource acquisition (and release) code which looks like this at present: > connect = do > envHandle <- envCreate > errHandle <- handleAlloc oci_HTYPE_ERROR envHandle > serverHandle <- handleAlloc oci_HTYPE_SERVER envHandle > connection <- dbLogon "user" "password" "database" envHandle errHandle > session <- getSession errHandle connection > return (envHandle, errHandle, serverHandle, connection, session) > disconnect envHandle errHandle serverHandle connection session = do > sessionEnd errHandle connection session > serverDetach errHandle serverHandle > handleFree oci_HTYPE_SERVER serverHandle > handleFree oci_HTYPE_ERROR errHandle > handleFree oci_HTYPE_ENV envHandle > terminate How should I structure it to handle exceptions raised by the various functions (envCreate, handleAlloc, dbLogon, etc)? I've started to restructure it like this: getEnv = do catchDyn ( do envHandle <- envCreate getErr envHandle ) (putStrLn "envCreate failed") getErr envHandle = do catchDyn ( do errHandle <- handleAlloc oci_HTYPE_ERROR envHandle getServer envHandle errHandle ) (do reportOCIException envHandle handleFree oci_HTYPE_ENV envHandle ) getServer envHandle errHandle = do catchDyn ( do serverHandle <- handleAlloc oci_HTYPE_SERVER envHandle getServer envHandle ) (do reportOCIException envHandle handleFree oci_HTYPE_ERROR errHandle handleFree oci_HTYPE_ENV envHandle ) ... but this is so awkward. How should I structure it? ***************************************************************** The information in this email and in any attachments is confidential and intended solely for the attention and use of the named addressee(s). This information may be subject to legal professional or other privilege or may otherwise be protected by work product immunity or other legal rules. It must not be disclosed to any person without our authority. If you are not the intended recipient, or a person responsible for delivering it to the intended recipient, you are not authorised to and must not disclose, copy, distribute, or retain this message or any part of it. *****************************************************************

2 1

XMLParser
by Ron de Bruijn 29 Jul '03

29 Jul '03

Hi there, I used the PARSEC parser combinator library to create an following XML-Parser. It's not exactly XML, but it's a usefull format for saving various data. The big difference there is, is that I don't have attributes, but those could easily be added, although I don't need them. Below is the complete code (20! lines of code for the core). I was excited when I saw it also worked like I had imagined, but then I tested it sometimes with increasingly big data. Using my test function below I noticed that there's some exponentional behaviour. I don't see a reason for this. Although I use the try combinator, it should at max double the time, because the parser uses the other parser when there occurs a parse error in the one. My definition of xmlParser looks also a bit odd, but without at least one of the functions of(subEnTop,topEnSub) it will fail parsing. I also don't understand why. Then I question myself: why does my program work with the downloaded version, and not with the PARSEC library that ships with GHC 5.0.4.2, I atleast thought that Parsec hasn't changed for a long time? Does anyone have an explanation for the above problems? Greets Ron module XMLCreator where --this module can parse XML import Parsec import GHC.Show import System.Time --datatype om XML te representeren: XML bestaat altijd uit een topelement en een aantal nestingen data XML a = TopElement String [XML a] | SubElement String a deriving Show type Level = Int --not finished Pretty Printer ppXML::XML a->Level->String ppXML (TopElement name (xml:xmls)) level = concat(take level (repeat " "))++openTag name ppXML _ _ = "hallo" --ppXML (NestedElement name info) level = openTag::String->String openTag s = "<"++s++">" --run :: Show a => Parser a -> String -> IO () run p input= do printTijd case (parse p "" input) of Left err -> do{ putStr "parse error at "; print err } Right x -> do print x printTijd --prints time(used above for printing the start time and the endtime of the algoritm printTijd = do time<-getClockTime (putStrLn.show) time test n = run xmlParser (concat(replicate n str)) --test n = parseTest xmlParser (concat(replicate n str)) str::String str = "<TopicList><TopicItem><Readed>1</Readed><ForumTitle>forumTitle</ForumTitle><TopicTitle>topicTitle</TopicTitle><TopicID>100100</TopicID><TopicStarter>someOne</TopicStarter><LastReplier>someOtherOne</LastReplier><ReplierID>1234</ReplierID><MessageID>messageID</MessageID><SubForumTitle>PW</SubForumTitle><SubForumID>14</SubForumID><Replies>2</Replies><ReplyTime>1050767736</ReplyTime><StartTime>1050762676</StartTime></TopicItem><FileInfo><Version>5</Version><TimeCreated>1059155053</TimeCreated><Application>SomeProgram</Application><About>aboutText</About></FileInfo></TopicList> "; --determine which sign can not be in the elementname or in the informationpiece signs = noneOf ['<','>'] --parses a piece of xml xmlParser::Parser ([XML String]) xmlParser = choice [try (many1 sub),try topEnSub,try subEnTop, many1 top] --xmlParser = choice[try(many1 sub),many1 top] getName::Parser String getName = do teken '<' name<-many1 signs teken '>' return name where teken t = do skipMany space char t skipMany space end::String->Parser String end name = do skipMany space string ("</"++name++">") sub::Parser (XML String) sub = do name<-getName info<-many1 signs end name return (SubElement name info) top::Parser (XML String) top = do name<-getName moreXml<-xmlParser end name return (TopElement name moreXml) subEnTop::Parser [(XML String)] subEnTop = do x<-many1 sub y<-many1 top return (x++y) topEnSub::Parser [(XML String)] topEnSub = do x<-many1 top y<-many1 sub return (x++y) __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

1 0

Re: Can't compile hmake
by Thomas Davie 29 Jul '03

29 Jul '03

On Tuesday, Jul 29, 2003, at 11:16 Europe/London, Keith Wansbrough wrote: >> I was wondering if idiocy was my problem there, I don't see the >> directory though: >> >> [nimbus2:project/hmake-3.08/targets] tatd100% ls -la >> total 0 >> drwxr-xr-x 3 tatd100 staff 102 Jul 29 10:42 . >> drwxr-xr-x 14 tatd100 staff 476 Jul 29 10:43 .. >> drwxr-xr-x 7 tatd100 staff 238 Jul 29 10:44 powerpc-Darwin6 >> >> Am I being incredibly stupid here? > > 'Fraid so ;-/ `harch` means "powerpc-Darwin6" in your case. The > current host architecture name. I suspected as much. In which case I have the problem that the file doesn't exist :/ [nimbus2:hmake-3.08/targets/powerpc-Darwin6] tatd100% ls -l total 32 -rw-r--r-- 1 tatd100 staff 4 Jul 29 10:43 buildwith -rw-r--r-- 1 tatd100 staff 428 Jul 29 10:54 config.cache -rw-r--r-- 1 tatd100 staff 3 Jul 29 10:43 ghcsym -rw-r--r-- 1 tatd100 staff 721 Jul 29 10:43 hmake3.config drwxr-xr-x 3 tatd100 staff 102 Jul 29 10:44 obj Thanks for putting up with my idiocy, I'm not used to the unix build tools and where it puts things. Tom Davie

1 0

Re: Can't compile hmake
by Thomas Davie 29 Jul '03

29 Jul '03

On Tuesday, Jul 29, 2003, at 11:03 Europe/London, Keith Wansbrough wrote: >>> Yes, after configuring, try editing the generated config file as >>> follows: >>> In targets/`harch`/config, change the line >>> READLINE="" >>> to >>> READLINE="-DUSE_READLINE=1 -L/sw/lib/ -lreadline" >>> then proceed to 'make' as usual. > [..] > >> to configure, then editing the config.cache file - I have tried the >> exact text you specified and also this: >> READLINE="-DUSE_READLINE=1 -L/sw/lib/ -lreadline -I/sw/include/ >> -include readline/readline.h" >> But still with no joy. I'm currently getting this error: > > He said to edit targets/`harch`/config, not config.cache. > config.cache is just information for the next run of ./configure; it > doesn't affect the build process. I was wondering if idiocy was my problem there, I don't see the directory though: [nimbus2:project/hmake-3.08/targets] tatd100% ls -la total 0 drwxr-xr-x 3 tatd100 staff 102 Jul 29 10:42 . drwxr-xr-x 14 tatd100 staff 476 Jul 29 10:43 .. drwxr-xr-x 7 tatd100 staff 238 Jul 29 10:44 powerpc-Darwin6 Am I being incredibly stupid here? Thanks Tom Davie

1 0