- Haskell-Cafe - Haskell.org

RE: [Haskell-cafe] Numerical methods in Haskell
by Simon Peyton-Jones 20 Feb '06

20 Feb '06

| Between google searching and looking through the activity | report, I take it that no one has really developed serious | libraries for matrix manipulations, diff eqs, etc. | | Are there any practical reasons for this or is it just a | matter of the haskell community being small and there not | being many people interested in something so specialized? The latter I think, but it's just the sort of thing that a functional language should be good at. Two other difficulties (a) It's hard to compete with existing libraries. The obvious thing is not to compete; instead, just call them. But somehow that doesn't seem to be as motivating. Perhaps some bindings exist though? (b) A concern about efficiency, because numerical computation is typically an area where people really care about how many instructions you take. It's a legitimate concern, but I don't think that it'll turn out to be justified. With unboxed arrays, and/or calling external libraries for the inner loops -- and the potential for aggressive fusion and/or parallelism, there is plenty of upward potential. I also want to work on nested data parallelism (a la NESL, and NEPAL) which fits right in here. I'd love to see a little community of matrix manipulators spinning up. Simon

5 5

Shapes Compiling Error
by Chatzianastassiou Achilleas 19 Feb '06

19 Feb '06

Hi everybody, I have written some code for several shapes, and i get a compiling error: parse error on input `:' though I can't find where the error is... data Shape = Ellipse Radius Radius | Polygon [Vertex] deriving Show type Radius = Float type Vertex = (Float, Float) area :: Shape -> Float area (Rectangle s1 s2) = s1 * s2 area (RtTriangle s1 s1) = s1 *s2 /2 area (Ellipse r1 r2) = pi * r1 * r2 area (Polygon(v1:v2:v3:vs)) = triArea v1 v2 v3 + area(Polygon(v1:v3:vs)) area(Polygon _ ) = 0 area (Polygon(v1:vs')) = polyArea vs where polyArea :: [Vertex] -> Float polyArea(v2:v3:vs') = triArea v1 v2 v3 + polyArea(v3:vs') polyArea _ = 0 triArea :: Vertex -> Vertex -> Vertex -> Float triArea v1 v2 v3 = let a = distBeetween v1 v2 b = distBeetween v2 v3 c = distBeetween v3 v1 s = 0.5 * (a + b + c) in sqrt (s * (s - a) * (s - b) * (s - c)) distBeetween :: Vertex -> Vertex -> Float distBeetween (x1, y1) (x2, y2) = sqrt ((x1 - x2) ^ 2 + (y1 - y2) ^ 2) Any suggestions? Thanks in advance, Skag55 _________________________________________________________________ On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement

3 3

GADTs and bar :: Foo t1 -> Foo t2
by Daniel McAllansmith 19 Feb '06

19 Feb '06

Hello, I have a recursive type > data Foo = A | B [Foo] | C [Foo] that I would like to restrict so that a C can only contain Bs, and a B can only contain As. If I use a GADT as follows the bar function, understandably, will not type check. > data AType > data BType > data CType > data Foo a where > A :: Foo AType > B :: [Foo AType] -> Foo BType > C :: [Foo BType] -> Foo CType > --get the successor of the Foo > bar c@(C [] ) = c > bar (C (b:_)) = b > bar b@(B [] ) = b > bar (B (a:_)) = a > bar a@A = a How do I achieve this restriction? Do I need to have some sort of successor class with dependent types? Ta Daniel

1 1

Infinite loop?
by rgo 19 Feb '06

19 Feb '06

Hi all, my program probably goes into infinite loop... But i cannot understand where and why. code: import System.Directory data MyFile = MyDir { dir_name :: String, dir_files :: [MyFile] } | MyFile { file_name :: String } read_dir_entries :: [FilePath] -> IO [MyFile] read_dir_entries [] = return [] read_dir_entries (name:names) = do isdir <- doesDirectoryExist name; entry <- if isdir then read_dir name else return (MyFile {file_name = name}); entries <- read_dir_entries names; return (entry:entries) read_dir :: FilePath -> IO MyFile read_dir name = do content <- getDirectoryContents name; files <- read_dir_entries content; return MyDir { dir_name = name, dir_files = files } instance Show MyFile where show (MyDir {dir_name = name, dir_files = files}) = "\n" ++ name ++ ":\n" ++ show files show (MyFile {file_name = name}) = "\t" ++ name ++ "\n" main = do cwd <- getCurrentDirectory; dir <- read_dir cwd; print dir;

7 8

Re: [Haskell-cafe] Infinite loop?
by Donn Cave 19 Feb '06

19 Feb '06

Quoth Barbier de Reuille Pierre <pierre.barbier(a)cirad.fr>: | On Fri, 17 Feb 2006 17:26:08 +0000 | Jon Fairbairn <jon.fairbairn(a)cl.cam.ac.uk> wrote: | | > On 2006-02-17 at 09:22PST Jared Updike wrote: | > > Yep. change one line to: | > > | > > entry <- if isdir && name /= "." && name /= ".." | > > | > > and it does in fact work. | > | > Only if no-one has been tricky with symbolic links. | > | | And how can you, in Haskell, detect symbolic links ? I refer you to the documentation in System.Posix.Files, especially getFileStatus :: FilePath -> IO FileStatus, and isSymbolicLink :: FileStatus -> Bool. I don't think I've used this function, but it looks promising. Donn Cave, donn(a)drizzle.com

1 0

classes inheritance seems not to work.
by asker 19 Feb '06

19 Feb '06

Hi,   I'm trying to make three haskell classes and an instance this way:  ------------------------------------------------------------------------------ class (Eq a) => Graph a where vert :: [a] ady :: a -> [a] class (Graph a) => Paths a where gps :: a -> a -> [[a]] class (Paths a) => Minimum a where mgps :: a -> a -> [a] instance Minimum Char where vert = ['a' .. 'f'] -- This is the line of the error message. ady 'a' = ['b', 'c'] ady 'b' = ['d', 'c', 'e'] ady 'd' = ['f'] gps = bepgr -- bepgr is defined and implemented in other code's part mgps = head gps ------------------------------------------------------------------------------   When I try to run this I get: "ERROR: filepath:line -No member vert in class "Minimum"" I don't know why does this happen, am I doing something wrong?   Thank you. -- View this message in context: http://www.nabble.com/classes-inheritance-seems-not-to-work.-t1148006.html#… Sent from the Haskell - Haskell-Cafe forum at Nabble.com.

2 2

libreadline dependency
by J. Scott Thayer, M.D. 19 Feb '06

19 Feb '06

Running Suse 9.3 64bit on Acer Ferrarri laptop. Haskell wants libreadline 4 while I have libreadline 5. Is there a definitive fix for this? The answers in archived messages were 'sub-optimal'.

3 2

Re: [Haskell-cafe] Problems instancing a class
by Benjamin Franksen 18 Feb '06

18 Feb '06

On Friday 17 February 2006 21:03, Juan Carlos Arevalo Baeza wrote: > What really got me is that if I don't use a type synonym: > > instance MyClass (HT.HashTable String String) where > htLookup h var = > do result <- HT.lookup h var > case result of > Nothing -> return "" > Just s -> return s > > it still complains: > > Illegal instance declaration for `MyClass (HashTable String String)' > (The instance type must be of form (T a b c) > where T is not a synonym, and a,b,c are distinct type variables) > In the instance declaration for `MyClass (HashTable String String)' > > even though the documentation in GHC states that HashTable is > declared as "data". (Guessing) Maybe because 'String' is not a type variable? > How exactly is type undecidable if this is allowed? I found here: > > http://hackage.haskell.org/trac/haskell-prime/wiki/FlexibleInstances > > an example of a non-terminating instance: > > instance C b b => C (Maybe a) b > > I don't see how/where/when this would be non-terminating. I > mean... if you now define an instance "C Int Int", then you actually > also have "C (Maybe Int) Int", C (Maybe (Maybe Int)) Int" and so on, > I can see that. But how exactly is that a problem? Can you show me a > use of C that cannot work with that instance? What is (or better: might be) non-terminating is the process (algorithm) of /type inference/. This: http://www.haskell.org//pipermail/haskell-prime/2006-February/000609.html message contains a detailed explanation. Ben

2 1

Re: (Newbie Question) How to get the value of an "IO String" Monad
by Peter 17 Feb '06

17 Feb '06

Hello All, Thanks Cale, Udo, Daniel, Lemmih, Ketil, and Marc for your very helpful posts, and Haskell resources. Every post really helped a lot. Thanks for all the help, Peter

1 0

(Newbie Question) How to get the value of an "IO String" Monad?
by Peter 17 Feb '06

17 Feb '06

Hello, For the purpose of familiarizing myself with Haskell, and also because I love Haskell :), I am trying to re-make a script that I made in Python that sends a request to a server and extracts the list of email addresses from a Mailman web-page by using an XML Parser on the page's HTML that has been converted to XHTML by "HTML Tidy". However, I cannot seem to figure out a way to get the state of a Monad; Specifically I cannot get the value of an "IO String" Monad. I have read some tutorials on Monads but I guess I must have missed something. I have read that the >>= operator is the only way to extract the state of an action as a string, and pipe it to a function. So far so good. But, That does not seem to work, because as I understand the >>= operator, it expects the function on the right hand side to return an IO Monad, which completely defeats the purpose here. So, How am I supposed to get the value of an IO Monad, such as "IO String", without returning an IO Monad? If this is of any help, here is the function I am stuck on: recv_headers' :: Socket.Socket -> String -> IO [[String]] recv_headers' sock bulk | received == "" = error "Connection died unexpectedly." | received == "\n" && endswith bulk "\r\n\r" = return [["foo", "bar"]] | otherwise = recv_headers' sock (bulk ++ received) where received = (Socket.recv sock 1) --- End code --- And here is the (expected) error I get from trying to compare "IO String" to "String": MemberBackup.hs:29:18: Couldn't match `IO String' against `[Char]' Expected type: IO String Inferred type: [Char] In the second argument of `(==)', namely `""' In a pattern guard for the definition of `recv_headers'': received == "" Failed, modules loaded: none. --- End error --- Thanks for the help, Peter

7 6

Problems instancing a class
by Juan Carlos Arevalo Baeza 17 Feb '06

17 Feb '06

So, consider this code: import Data.HashTable as HT class MyClass a where htLookup :: a -> String -> IO String type Context = HT.HashTable String String instance MyClass Context where htLookup h var = do result <- HT.lookup h var case result of Nothing -> return "" Just s -> return s This doesn't compile. GHC says: Illegal instance declaration for `MyClass Context' (The instance type must be of form (T a b c) where T is not a synonym, and a,b,c are distinct type variables) In the instance declaration for `MyClass Context' If I use "data" instead of "type", it works: data Context = C (HT.HashTable String String) instance MyClass Context where htLookup (C h) var = do result <- HT.lookup h var case result of Nothing -> return "" Just s -> return s Why? What's going on here? JCAB

4 4

Badly designed Parsec combinators?
by Juan Carlos Arevalo Baeza 17 Feb '06

17 Feb '06

So... I see no reason why someone can't just do it themselves, but... I was playing around with Parsec (as included in GHC 6.4.1), and I found two functions that are... not quite what I believe they should be. optional :: GenParser tok st a -> GenParser tok st () optional p = do{ p; return ()} <|> return () Now, this completely loses the result of the optional parser. Better would be: optional :: GenParser tok st a -> GenParser tok st (Maybe a) optional p = do{ x <- p; return (Just x) } <|> return Nothing Same thing with manyTill: manyTill :: GenParser tok st a -> GenParser tok st end -> GenParser tok st [a] manyTill p end = scan where scan = do{ end; return [] } <|> do{ x <- p; xs <- scan; return (x:xs) } Better: manyTill :: GenParser tok st a -> GenParser tok st end -> GenParser tok st ([a], end) manyTill p end = scan where scan = do{ endr <- end; return ([], endr) } <|> do{ x <- p; (xs, endr) <- scan; return (x:xs, endr) } Is there any reason I can't see why they are the way they are? JCAB

8 15

Re: HaXml: ampersand in attribute value
by Koen.Roelandt＠mineco.fgov.be 17 Feb '06

17 Feb '06

> But speaking of HaXml bugs, I'm pretty sure HaXml doesn't handle > % correctly. It seem to treat % specially everywhere, but I think > it is only special inside DTDs. I have many XML files produced by > other tools that the HaXml parser fails to process because of this. I had a similar problem where the parser choked on % signs in attribute values in the XML file. I solved it by playing around in Lex.hs. I could check for the (ugly) solution... > Yes, it is expected. An ampersand indicates the start of a reference, > e.g. < or  If there is no semicolon to indicate the end of the > reference, then it is a parse error. The XML specification is quite > clear that neither & nor < are valid standalone characters in an > attribute value. Which is exactly the problem. The & is part of a reference, namely ë. Regards, Koen.

1 0

RE: [Haskell-cafe] library sort
by Bayley, Alistair 17 Feb '06

17 Feb '06

> From: haskell-cafe-bounces(a)haskell.org > [mailto:haskell-cafe-bounces@haskell.org] On Behalf Of Radu Grigore > > Is there a sort function in the libraries that come with GHC > (6.4)? My search at > http://www.haskell.org/ghc/docs/latest/html/libraries/index.html > has failed, but I can't believe there is none. Data.List.sort (and sortBy): http://www.haskell.org/ghc/docs/latest/html/libraries/base/Data-List.htm l#v%3Asort ***************************************************************** Confidentiality Note: The information contained in this message, and any attachments, may contain confidential and/or privileged material. It is intended solely for the person(s) or entity to which it is addressed. Any review, retransmission, dissemination, or taking of any action in reliance upon this information by persons or entities other than the intended recipient(s) is prohibited. If you received this in error, please contact the sender and delete the material from any computer. *****************************************************************

3 2

haskell question
by IBRAHIM MOSHAYA 17 Feb '06

17 Feb '06

please can you help me with this function: Significant Figures I need to write a function to return a given number (1st argument) to a given number of significant figures (2nd argument). The function should be called sigFig and have a type signature of the form: sigFig :: (???) => a -> Int -> a For instance: *Main> sigFig 123.456 2 120.0 *Main> sigFig 123.456 4 123.5 *Main> sigFig 0.987 2 0.99 *Main> sigFig 0.432 2 0.43 The type variable a will need to be constrained, but you should aim for the most general implementation you can (replacing ??? above with the appropriate constraint(s)). You may make free use of Prelude functions. Referring to [Jon03] you may wish to consider the use of functions / operators including: /, div, mod, ^, ^^, **, round, fromIntegral, realToFrac thanks alot

2 1

Re: [Haskell-cafe] Decidable type systems? (WAS: Associated TypeSynonyms question)
by Robin Green 17 Feb '06

17 Feb '06

Miles Sabin <miles(a)milessabin.com> wrote: > Can someone explain to me why decidability is of any practical > interest at all? procedure which might never terminate and one which might take > 1,000,000 years to terminate? Actually, why push it out to 1,000,000 > years: in the context of a compiler for a practical programming > language, a decision procedure which might take an hour to terminate > might as well be undecidable But are there any decidable type checking algorithms that have been seriously proposed or used which would take far too long to terminate for real code? If not, then decidability is the only thing that matters. -- Robin

3 2

library sort
by Radu Grigore 16 Feb '06

16 Feb '06

Is there a sort function in the libraries that come with GHC (6.4)? My search at http://www.haskell.org/ghc/docs/latest/html/libraries/index.html has failed, but I can't believe there is none. -- regards, radu http://rgrig.blogspot.com/

4 3

Cyclical Type Synonyms
by Tom Hawkins 16 Feb '06

16 Feb '06

I want to define a type for a function that returns its own type... type F a = a -> (F a,a) But the compiler gives me an error: "Cycle in type synonym declaration: ...". Why is this a restriction? Of course I can create a user defined type, but then I need an extra mechanism to call the embedded function: data F a = F (a -> (F a ,a)) call :: F a -> a -> (F a, a) call (F f) a = f a -Tom

2 1

Haskell' Status
by Isaac Jones 16 Feb '06

16 Feb '06

I'll try to occasionally post an announcement of the the status of Haskell'[1], the next Haskell standard, so that you can all be aware of my thinking, and our current place in the timeline. There is a list of proposals and a "strawman" categorization of them on the wiki[2]. The categorization reflects someone's current thinking about what's in and out, but it's really just for discussion at this phase. The timeline for this effort is also on the wiki[4]. You'll notice that it's very aggressive; we plan to announce something at the next Haskell Workshop in September. Our current spot in the timeline is that we're meant to be brainstorming, discussing, and refining proposals. I've just posted some leading questions [3] in order to focus the discussion a bit. I can't help but notice that there is a lot of excitement (and support?) for a larger standard; one that would have a bigger impact. I think that part of the Haskell' effort should be to make a plan for moving forward. At the very least we should have a set of recommendations for what the community needs to work on between this and the next standard; what are the most promising proposals and what needs to happen to make them a reality. I think that the wiki will be a great resource for any future standard, and we should work to make it as nice as possible. Please reply to the Haskell' mailing list, and email me if you have any questions. peace, isaac [1] Haskell' Wiki: http://hackage.haskell.org/trac/haskell-prime [2] list of proposals and strawman categorization: http://hackage.haskell.org/trac/haskell-prime/report/9 [3] plan for moving discussion forward http://www.haskell.org//pipermail/haskell-prime/2006-February/000582.html [4] The TimeLine is here. Ascii text of the timeline follows (might not make sense unless you use a proportional font: http://hackage.haskell.org/trac/haskell-prime/wiki/TimeLine Write Report----Review---- | | Edit | Discuss / Refine--|---------- | | | | | | | brainstorm----------- | | | | | | | | | | | setup | | | | | | | | | | | | | ------------------------------------------------------------------- 10 | 11 | 12 | 01 | 02 | 03 | 04 | 05 | 06 | 07 | 08 | 09 | 10 | 11 2005 2006 ------------------------------------------------------------------- | | | | | | Face-To-Face? | Strawman Trial Decision Announce @HW

1 0

Re[2]: strict Haskell dialect
by Bulat Ziganshin 15 Feb '06

15 Feb '06

Hello Wolfgang, Friday, February 03, 2006, 1:46:56 AM, you wrote: >> i had one idea, what is somewhat corresponding to this discussion: >> >> make a strict Haskell dialect. implement it by translating all >> expressions of form "f x" into "f $! x" and then going to the standard >> (lazy) haskell translator. the same for data fields - add to all field >> definitions "!" in translation process. then add to this strict >> Haskell language ability to _explicitly_ specify lazy fields and lazy >> evaluation, for example using this "~" sign >> >> what it will give? ability to use Haskell as powerful strict language, >> what is especially interesting for "real-world" programmers. i have >> found myself permanently fighting against the lazyness once i starting to >> optimize my programs. for the newcomers, it just will reduce learning >> path - they don't need to know anything about lazyness WJ> Since laziness often allows you to solve problems so elegantly, I'm really WJ> scared of the idea of a "Strict Haskell"! :-( Is laziness really so "unreal" WJ> that real-world programmers have to see it as an enemy which they have to WJ> fight against? WJ> In fact, I was kind of shocked as I read in Simon Peyton Jones' presentation WJ> "Wearing the hair shirt" [1] that in his opinion "Lazyness doesn't really WJ> matter". i suggest you to write some large program like darcs and try to make it as efficient as C++ ones. i'm doing sort of it, and i selected Haskell primarily because it gives unprecedented combination of power and safety due to its strong but expressive type system, higher-order functions and so on. i also use benefits of lazyness from time to time, and may be even don't recognize each occasion of using lazyness. but when i'm going to optimize my program, when i'm asking myself "why it is slower than C counterparts?", the answer is almost exclusively "because of lazyness". for example, i now wrote I/O library. are you think that i much need lazyness here? no, but that i really need is the highest possible speed, so now i'm fighting against lazyness even more than usual :) well, 80% of any program don't need optimization at all. but when i write remaining 20% or even 5%, i don't want to fight against something that can be easily fixed in systematic way. all other widespread languages have _optional_, explicitly stated lazyness in form of callable blocks, even the Omega goes in this way. and i'm interested in playing with such Haskell dialect in order to see how my programming will change if i need to explicitly specify lazyness when i need it, but have strictness implicitly. i think that newcomers from other languages who wants to implement real projects instead of experimenting will also prefer strict Haskell you may hear that last days Haskell become one of fastest language in the Shootout. why? only because all those programs was rewritten to be strict. it was slow and hard process. and adding preprocessor that makes all code strict automagically will allow to write efficient Haskell programs without reading fat manuals each laguage feature has its time. 15 years ago i could substantially speed up C program by rewriting it in asm. Now the C compilers in most cases generate better code than i can. moreover, strict FP languages now are ready to compete with gcc. But lazy languages are still not compiled so efficient that they can be used for time-critical code. so, if we don't want to wait another 10 years, we should implement easier ways to create strict programs. if you think that lazy programming is great, you can show this in shootout or by showing me the way to optimize code of my real programs. i'm open to new knowledge :) -- Best regards, Bulat mailto:bulatz@HotPOP.com

16 40

Layout processing
by Maurício 15 Feb '06

15 Feb '06

Hi, Is it possible to ask GHC or other Haskell compiler to generate a Haskell source file just after the processing of layout? I.e., I would like to see this: main = do a <- some_function Transformed into this: main = do {a <- some_function; I thought that would be usefull to understand better how layout works. Thanks, Maurício

1 0

Haskell XSLT interpreter?
by S. Alexander Jacobson 15 Feb '06

15 Feb '06

Has anyone written a pure haskell xslt interpreter? If not, how difficult would it be to do so? -Alex- ______________________________________________________________ S. Alexander Jacobson tel:917-770-6565 http://alexjacobson.com

7 9

foreign import unsupported in GHCi on amd64?
by Daniel 15 Feb '06

15 Feb '06

Hi, is foreign import expected to work when using GHCi (ghc 6.4.1) on an amd64 linux platform? I'm just (attempting) to play around with some code (Oleg et. al.'s Zipper FS) that has the following lines in it: ZFS/ZFS.hs:522:foreign import ccall "unistd.h select" c_select ZFS/ZFS.hs:573:foreign import ccall "fcntl.h fcntl" fcntl and I get: Compiling ZFS ( ZFS.hs, interpreted ) ghc-6.4.1: foreign import is not implemented for GHCi on this platform. Ticket 290 ( http://cvs.haskell.org/trac/ghc/ticket/290 ) seems to imply the foreign import wrappers work in ghc 6.4. Cheers Daniel

1 1

List-to-outline
by Steve Schafer 14 Feb '06

14 Feb '06

I have some lists of integers; e.g., [0,1,2,2,3,3,3,2,1,2,3,3,1] Think of each integer value as representing the indentation level in a hierarchical outline: e.g., 0 1 2 2 3 3 3 2 1 2 3 3 1 I want to convert the list into a structure that better represents the hierarchy. So, I first define a datatype to represent each node of the new structure: data Node = Nd Int [Node] That is, a node consists of an Int representing the value of the node, followed by a list of its immediate child nodes. (In principle, I can deduce the value of a node simply from the nesting level, of course, but in the real problem I'm trying to solve, each node contains other information that I need to preserve as well.) Next, I define some functions to perform the transformation isChild :: Int -> Node -> Bool isChild i (Nd j _) = (j > i) isChild _ _ = False prepend :: Int -> [Node] -> [Node] prepend i [] = [Nd i []] prepend i ns = (Nd i f):s where (f,s) = span (isChild i) ns unflatten :: [Int] -> [Node] unflatten ns = foldr prepend [] ns Finally, I add some code to display the result in an aesthetically pleasing way: showsNodeTail :: [Node] -> String -> String showsNodeTail [] = showChar '}' showsNodeTail (n:ns) = showChar ' '.shows n.showsNodeTail ns showsNodeList :: [Node] -> String -> String showsNodeList [] = showString "" showsNodeList (n:ns) = showChar '{'.shows n.showsNodeTail ns showsNode :: Node -> String -> String showsNode (Nd i ns) = shows i.showsNodeList ns instance Show Node where showsPrec n = showsNode This all works just fine, and when I enter unflatten [0,1,2,2,3,3,3,2,1,2,3,3,1] I get [0{1{2 2{3 3 3} 2} 1{2{3 3}} 1}] as expected. The reason I'm posting this here is that I have a gnawing suspicion that the unflatten/prepend/isChild functions, and possibly the Node data type as well, are not the most elegant way to go about solving the problem, and that I'm missing another more obvious way to do it. Any suggestions? Thanks, Steve Schafer Fenestra Technologies Corp. http://www.fenestra.com/

4 3

Overlapping instance problem
by Jeff.Harper＠handheld.com 13 Feb '06

13 Feb '06

Hi, I've posted a couple messages to the Haskell Cafe in the last few months. I'm new to Haskell. But, I've set out to implement my own vectors, matrices, complex numbers, etc. One goal I have, is to overload operators to work with my new types. The pursuit of this goal, has pushed me to learn a lot about the Haskell type system. When I get stuck from time-to-time, the kind folks on this list have pointed me in the right direction. I'm stuck now. One thing I want to avoid is adding new multiplication operators to handle multiplication of dissimilar types. For instance, I'd like to be able to have an expression like k * m where k is a Double and m is a Matrix. This doesn't work with the prelude's (*) operator because the prelude's (*) has signature: (*) :: (Num a) => a -> a -> a. To get around this, I wrote my own versions of a Multiply class that allows dissimilar types to be multiplied. You can see my Multiply class in the module at the end of this Email. At the bottom of the module, I've attempted to implement multiplication of the forms: scalar * matrix matrix * scalar matrix * matrix The problem is that when I try to do matrix * matrix at the interpreter, I get an error message from Glaskgow: *My_matrix> m1 * m2 <interactive>:1:3: Overlapping instances for Multiply (Matrix Double) (Matrix Double) (Matrix c) arising from use of `*' at <interactive>:1:3 Matching instances: My_matrix.hs:63:0: instance (Multiply a b c, Add c c c, Num a, Num b, Num c) => Multiply (Matrix a) (Matrix b) (Matrix c) My_matrix.hs:57:0: instance (Multiply a b c, Num a, Num b, Num c) => Multiply (Matrix a) b (Matrix c) My_matrix.hs:51:0: instance (Multiply a b c, Num a, Num b, Num c) => Multiply a (Matrix b) (Matrix c) In the definition of `it': it = m1 * m2 I don't understand how m1 * m2 can match the scalar multiplication instances. For instance, the scalar * matrix instance has signature: instance (Multiply a b c, Num a, Num b, Num c) => Multiply a (Matrix b) (Matrix c) where m1 in my expression would correspond to the 'a' type variable. But, 'a' is constrained to be a Num. However, I never made my Matrix type an instance of Num. Is there a work around for this? In my first implementation, I did not have the Num constraints in the matrix Multiply instances. I added the Num constraints specifically, to remove the ambiguity of the overlapping instance. Why didn't this work? Thanks, Jeff Harper >> Begining of code for My_matrix.hs ------------------------------------------ {-# OPTIONS -fglasgow-exts #-} module My_matrix where import qualified Prelude as P import Prelude hiding ( (*), (+), (-), negate) default ( ) class Add a b c | a b -> c where (+) :: a -> b -> c class Multiply a b c | a b -> c where (*) :: a -> b -> c class Coerce a b where coerce :: a -> b infixl 7 * infixl 6 + instance Coerce Float Float where { coerce x = x } instance Coerce Float Double where { coerce x = realToFrac x } instance Coerce Double Double where { coerce x = x } instance Add Float Float Float where { (+) x y = ( x) P.+ ( y) } instance Add Float Double Double where { (+) x y = (coerce x) P.+ ( y) } instance Add Double Float Double where { (+) x y = ( x) P.+ (coerce y) } instance Add Double Double Double where { (+) x y = ( x) P.+ ( y) } instance Multiply Float Float Float where { (*) x y = ( x) P.* ( y) } instance Multiply Float Double Double where { (*) x y = (coerce x) P.* ( y) } instance Multiply Double Float Double where { (*) x y = ( x) P.* (coerce y) } instance Multiply Double Double Double where { (*) x y = ( x) P.* ( y) } -- Matrices are stored in a list of list. For now, I can create a -- matrix of Float, or Double. Later, I'd like to extend this and -- make it possible to create a matrix of other number types. For -- instance, it might be possible to have a matrix of complex or -- imaginary numbers. data Matrix a = Matrix [[a]] deriving Show -- For simplicity, the instances below omit the implementation for (*). -- This instance of Multiply is for doing multiplication of the form -- k * m where k is a scalar and m is a matrix. instance (Multiply a b c, Num a, Num b, Num c) => Multiply a (Matrix b) (Matrix c) where (*) x y = Matrix [[]] -- This instance of Multiply is for doing multiplication of the form -- m * k where k is a scalar and m is a matrix. instance (Multiply a b c, Num a, Num b, Num c) => Multiply (Matrix a) b (Matrix c) where (*) x y = Matrix [[]] -- This instance of Multiply is for doing multiplication of the form -- m1 * m2 where m1 and m2 are both matrices instance (Multiply a b c, Add c c c, Num a, Num b, Num c) => Multiply (Matrix a) (Matrix b) (Matrix c) where (*) x y = Matrix [[]] -- Some test variables to use in the interpreter k = (3.0::Double) m1 = Matrix [[1.0::Double]] m2 = Matrix [[2.0::Double]]

3 2

Small syntax question
by Maurício 13 Feb '06

13 Feb '06

Hi, This is valid code: module Main where main = do let a = 3 return () Why isn't this one? module Main where main = do { let a = 3; return (); }; Thanks for your help, Maurício

5 4

Sorting Integers
by JimpsEd 13 Feb '06

13 Feb '06

I have 3 integers, a b c that I want to pass to a function, and I want the function to return the 3 integers sorted, from largest to smallest - any idea how to do this? Thanks; Jamie -- View this message in context: http://www.nabble.com/Sorting-Integers-t1113340.html#a2908719 Sent from the Haskell - Haskell-Cafe forum at Nabble.com.

3 4

How to deal with a translator???
by Huong Nguyen 13 Feb '06

13 Feb '06

Hello all, Now I am trying on a simple translator module. This module needs to translate an input string( this string represent a formula) as follows: (x^2 - 5x + 4=0) /\ (x^3 - 5>0) -> x>3 And I want the output string represent a formula like: (x^2 - 5x + 4=0) /\ (x^3 - 5>0) IMPLIES x>3, i.e, the translator attempt to translate (->) into IMPLIES. If you have ay suggesstions for that kind of problems, please share with me. I appreciate for that. Thanks all for your time. Nguyen,

2 1

round function
by Chatzianastassiou Achilleas 12 Feb '06

12 Feb '06

Hi all, I am trying to implement a function that finds the significant figure of a number to a specified point i.e. 2.55555 3 = 2.556. I have implemented something like: sig :: (RealFrac a, Integral b) => a -> Int -> a sig x y = round y However this doesn't work. Moreover at the Prelude when i type round 2.3333 2 i get an error. Any suggestions? Thanks, Skag55 _________________________________________________________________ Dont just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/

3 9

Why is $ right associative instead of left associative?
by Brian Hulley 12 Feb '06

12 Feb '06

Hi - In the Haskell98 report section 4.4.2 $ is specified as being right associative. This means that f $ a0 a1 $ b0 b1 would parse as f (a0 a1 (b0 b1)) which seems rather strange to me. Surely it would be much more useful if $ were defined as left associative so that it could be used to separate the args to f? Does anyone know why this strange associativity was chosen? Thanks, Brian. (The reason I'm asking is that I'm working on the syntax of a language similar to Haskell but which uses layout to allow expressions like: f #$ -- can be followed by an explicit block or layout block a0 a1 b0 b1 which is sugar for (f $ a0 a1) $ b0 b1 ie f (a0 a1) (b0 b1) ) and I was surprised to discover that the parentheses are needed for the most obvious reading)

16 40

How to check for memory leak.
by zbo 12 Feb '06

12 Feb '06

I was using the code from http://www.haskell.org/hawiki/HaskellNewbie_2fTcpIpAndNetworking It seems to be losing a few Kbytes every call (i use 'top' on linux to check the RES field value). I dont know whether this is a memory leak or just the GC not collecting yet. I was just wondering What is the general method to check this. BTW, I have tried using -prof and -P, but not really able to see anything suspicious. Cheers, ed

2 1

building ghc on win with gcc of cygwin, without cygwin1.dll (-mno-cygwin flag) ?
by Marc Weber 12 Feb '06

12 Feb '06

Hi.. I've found the nice documentation about building ghc on haskell.org/ghc -> documentation. There is one chapter (#1) about C compilers and environments to use: Either MSYS or cywin and gcc of MinGW because gcc of cygwin will link to cygwin1.dll by default which may change and therefore can brake your apps. Googling around I've found that there exists the -mno-cygwin flag which you can use to not include this lib.. So would it might be possible to get a ghc build not using cygwin.dll with just cygwin ? I haven't read anything about that in those docs. Or does this flag simply include cygwin.dll within your apps (statically linked) ? #1 :http://www.haskell.org/ghc/docs/latest/html/building/platforms.html#ghc-cygwin Marc

1 0

Digrams
by Dominic Steinitz 11 Feb '06

11 Feb '06

I've quickly put this together to measure frequencies of pairs of letters (e.g. 1st and 2nd) in words. It works fine on a small test data sets but I have a feeling that it will perform poorly as it spends a lot of time updating a 26*26 array. Before I throw a dictionary at it, does anyone have any suggestions? Thanks, Dominic.

2 2

Compiling hdirect on windows with COM support
by Marc Weber 11 Feb '06

11 Feb '06

On Fri, Feb 03, 2006 at 07:55:30AM +0100, Gracjan Polak wrote: > Hi, > I would be iterested in seeing what you have done. And maybe helping > in getting it to work. > I did not find the examples :) Many links seem to be broken on HDirect > page. Hi. I have finished. It took some more time as expected ;-) Just install cygwin, and run this script. It will download parts of fptools via cvs (so you need cvs and GNU make from cygwin) Greetings, Marc ---------------------------- compilehdirect.sh --- #!/bin/sh # author: Marc Weber # marco-oweber(a)gmx.de # configure fptools WORKDIR=${WORKDIR:-myfptoolsdirectory} # you can use "" here because cvs will checkout the folder "fptools" everything will be done in that.. CREATEZIPSTODEBUGTHISSCRIPT="" # "" = no "yes"=yes, for debugging purposes. # hdirect lib info (will replace old file due to packaging info changes) hdirectlibpkginfo="name: \"hdirect\" \n\ import-dirs: \"\${hd_libdir}\" \n\ library-dirs: \"\${hd_libdir}\" \n\ hs-libraries: \"HShdirect\" \n\ include-dirs: \"\${hd_libdir}\" \n\ depends: \"base\", \"haskell98\" \n\ exposed:True \n\ exposed-modules:\"HDirect\",\"Pointer\" " #hdirect comlib cabal pkg info: hdirectcomlibpkginfo="name: com \n\ import-dirs: \"\${hd_imp}\" \n\ library-dirs: \"\${hd_lib}\" \n\ hs-libraries: \"HScom\" \n\ extra-libraries: \"kernel32\", \n\ \"user32\", \n\ \"ole32\", \n\ \"oleaut32\", \n\ \"advapi32\", \n\ \"HScom\" \n\ include-dirs: \"\${hd_inc}\" \n\ depends: \"base\", \n\ \"haskell98\" \n\ exposed:True \n\ exposed-modules: \"Com\", \n\ \"TypeLib\", \n\ \"ComPrim\", \n\ \"Automation\", \n\ \"AutoPrim\", \n\ \"WideString\", \n\ \"StdTypes\", \n\ \"StdDispatch\", \n\ \"ComServ\", \n\ \"Connection\", \n\ \"SafeArray\" " confirm() { echo -e $1 "\n" "press return to proceed, Strg-C to exit"; read; } info() { echo; echo $1; echo $1 | sed -e 's/./-/g' ; } die() { echo $1 ; exit; } question() { echo -e $1 "\n [y]es, [n]o"; read d; if [[ "$d"="y" ]]; then True ; else False; fi; } dozip() { [ -z CREATEZIPSTODEBUGTHISSCRIPT ] || zip -r ../$1.zip . || die "couldn't create zip" ; } doziphd() { [ -z CREATEZIPSTODEBUGTHISSCRIPT ] || zip -r ../../$1.zip . || die "couldn't create zip"; } confirm "this script isn't well tested yet .. run it .. Read the code first because you are running it on your risk" echo " you will need cvs, GNU make from cygwin. I've used ghc-6.4.1 from Win Installer" confirm "Be sure to have your ghc in win PATH environment variable, else everything will work till compiling hdirect ;-) if you can run ghc.exe from cygwin shell everything should be fine." [ -d "$WORKDIR" ] || mkdir "$WORKDIR" || die "couldn't create working directory" cd "$WORKDIR" || die "couldn't cd to working directory" if [[ ! -d fptools ]]; then #if :; then question "Shall I dwonload happy (say yes and add --enable-src-tree-happy to configure if you don't have happy)" && coHappy=1 question "Shall I dwonload alex (say yes and add --enable-src-tree-alex to configure if you don't have alex)" && coAlex=1 info "checking out fptools top directory" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co -l fptools || die "cvs checkout of fptools top directory failed" info "checking out fptools/glafp-utils (used by hdirect)" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/glafp-utils || die "cvs checkout of fptools top directory failed" info "checking out ghc" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/ghc || die "cvs checkout of fptools top directory failed" info "checking out hdirect" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/hdirect || die "cvs check of hdirect failed" info "checking out fptools/mk (needed by hdirect make)" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co -l fptools/mk || die "cvs checkout of fptools top directory failed" info "checking out ghc/mk (else hdirect/make will complaing about ghc/mk/paths.mk" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/ghc/mk || die "cvs check of hdirect failed" if [ ! -z coHappy ]; then info "checking out happy (else hdirect/make will complaing about ghc/mk/paths.mk" remhappy="add --enable-src-tree-happy" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/happy || die "cvs check of hdirect failed" fi if [ ! -z coAlex ]; then info "checking out alex (else hdirect/make will complaing about ghc/mk/paths.mk" remalex="add --enable-src-tree-alex" cvs -d:pserver:anoncvs@cvs.haskell.org:/cvs co fptools/alex || die "cvs check of hdirect failed" fi cd fptools dozip AfterCO cd .. else info "omitting cvs co because fptools already exists" fi cd fptools if [ -f configure ]; then info "omitting autoreconf (configure already exists)"; else info "calling autoreconf"; autoreconf ||die "autoreconf failed" ; fi if [ -f config.status ]; then info "I omit configuring because fptools/conf exists. Remove it to configure again" else if [ -f configure ]; then info "omitting autoreconf (configure already exists)"; else info "calling autoreconf"; autoreconf ||die "autoreconf failed" fi echo "Now you have to configure fptools, you can add --with-ghc=ARG to specify the ghc directory and --with-gcc to specify to point to the gcc/bin directory (because you want to compile native win apps you don't want to use gcc from cygwin!)" confirm "running bash now, please call configure" echo "Your command might look like this: ./configure --with-gcc='c:/MinGWStudio/MinGW/bin/gcc' --with-ghc='c:/ghc/ghc-6.4.1/bin/ghc.exe'" $remhappy $remalex "; exit" echo "Press Ctrl-d when finished or enter exit to continue this script" bash dozip afterConfiguring fi info "***** starting compiling " if [ -d happy ]; then if [ ! -f happy/src/happy.exe ]; then info "happy in source tree. So I think you don't have it, yet. I'm running make" cd happy || die "couldn't cd to happy" make boot || die "couldn't make make boot" make || die "making happy failed" cd .. else echo "happy already built" fi fi if [ ! -z "$remalex" ]; then echo "I think alex is used to make configure not complain.. ;-)" fi if [ -f glafp-utils/.depend ]; then info "omitting glafp-utils because glasp-utils/.depend exists" else info "Making glafp-utils now" cd glafp-utils make boot || die "couln't make boot (glafp-utils) but this doesn't seem to be serious, just restart script" make || die "couln't make hdirect^ (glafp-utils) but this doesn't seem to be serious, just restart script" cd .. fi cd hdirect [ -f src/.depend ] || doziphd beforeBuildingHdirect info "I'm trying to build hdirect now!" if [ ! -f src/ihc.exe ]; then info "running make boot" make boot || die "couln't make boot (hdirect)" doziphd afterMakeBOOT fi if [ ! -f comlib/Automation.o ]; then info "running make (will build comlib and rebuild ihc.exe" make || die "couln't make hdirect^ (hdirect)" doziphd afterMake1 fi # registering hdirect: if ! ghc-pkg.exe describe hdirect; then cd lib # check wether hdirect.pkg has been replaced, yet due to changes in pkg info: if grep 'Package {' hdirect.pkg; then info "replacing hdirect.pkg due to changees in pkg info." echo "old pkg info (hdirect.pkg)" cat hdirect.pkg echo -e $hdirectlibpkginfo > hdirect.pkg echo "new pkg info (hdirect.pkg)" cat hdirect.pkg fi info "registering hdirect" #export hd_libdir="d:\\\\haskell\\\\fptools\\\\fptools\\\\hdirect\\\\lib" hd_libdir=$(cygpath -w $(pwd)) export hd_libdir=${hd_libdir//\\/\\\\} # double "\" becaus of escaping echo $hd_libdir ghc-pkg.exe -u < hdirect.pkg || die "couldn't register hdirect" cd .. fi #registering comlib of hdirect if ! ghc-pkg.exe describe comlib; then cd comlib # check wether com.pkg has been replaced, yet due to changes in pkg info: if grep 'Package{' com.pkg; then info "replacing com.pkg due to changees in pkg info." echo "old pkg info (com.pkg)" cat com.pkg echo -e $hdirectcomlibpkginfo > com.pkg echo "new pkg info (com.pkg)" cat com.pkg fi info "registering comlib" export hd_lib="" export hd_inc="" export hd_imp="" hd_lib=$(cygpath -w $(pwd)) export hd_inc=${hd_lib//\\/\\\\} export hd_imp=${hd_lib//\\/\\\\} export hd_lib=${hd_lib//\\/\\\\} echo hd_lib $hd_lib echo hd_int $hd_int echo hd_inc $hd_inc ghc-pkg.exe -u < com.pkg || die "couldn't register comlib" cd .. fi # Now it's getting really hot: trying to build ihc.exe with support Typelibs: export SUPPORT_TYPELIBS=YES info "running make boot with exported SUPPORT_TYPELIBS=YES" if [ ! -f src/ihc.exe ]; then info "running make boot" make boot || die "couln't make boot (hdirect)" doziphd afterMakeALL fi cd examples info "running bash so that you can try some examples type exit then to exit script" info "notice: you are now in $(pwd)" bash

1 2

question about type lambda and decidability of typechecking
by Robert Dockins 10 Feb '06

10 Feb '06

OK. I've been doing a little thinking about type lambda in Haskell. Now, I understand the prevailing wisdom is that adding type lambda and/or partially applied type synonyms to the haskell type system would make type checking/inference undecidable. The reason given is that higher-order unification is undecidable. I have to admit that I don't fully understand this reason. Setting aside typeclasses for now, it seems to me that type expressions together with the kind system are just the simply-typed lambda calculus with unit, which is well known to be strong normalizing. So any type with kind * has a normal form with (by definition) no internal redexes. I think this is sufficient to guarantee that all type lambdas are removed. Now you can proceed using first-order unification, which is decidable. Of course, all valid expressions have kind * (ignoring unboxing and other trickiness for now). So where have I gone wrong? Do typeclasses complicate the matter? Or have I missed something more basic? Thanks, Rob Dockins Speak softly and drive a Sherman tank. Laugh hard; it's a long way to the bank. -- TMBG

1 1

Numerical methods in Haskell
by Rob Tougher 10 Feb '06

10 Feb '06

I'm currently working on a mathematics library in Haskell. I'm not a mathematician -- I'm basically writing this library as an excuse to learn mathematics and Haskell at the same time. Feel free to check out the code: http://robtougher.com/ My main focus is financial mathematics (options pricing, etc), but I took a detour last week and implemented basic matrix stuff. I might work on some Linear Programming algorithms (simplex, interior point) before returning to finance. - Rob

2 1

Matching constructors
by Creighton Hogg 10 Feb '06

10 Feb '06

Hi, If I have something like data Patootie = Pa Int | Tootie Int and I want to pull out the indices of all elements of a list that have type constructor Tootie, how would I do that? I thought I might be able to use findIndices, but I don't know how to express the predicate.

7 6

multiple defentions of .. fst defined here errors.
by Marc Weber 10 Feb '06

10 Feb '06

Why do these error occur? After hiding the packages com (beeing in comlib) and hdirect (beeing in lib) and adding -package com -package hdirect it works fine. I haven't tried without hiding but adding -package options yet. ghc --make -H16m -O -fglasgow-exts -syslib com -fno-warn-missing-methods daotest.hs Chasing modules from: daotest.hs Skipping DAO ( ./DAO.hs, ./DAO.o ) Skipping DB ( ./DB.hs, ./DB.o ) Skipping Main ( daotest.hs, daotest.o ) Linking ... c:\haskell\myfptoolsdirectory\fptools\hdirect\comlib/libHScom.a(PointerPrim.o)(.text+0x696):ghc2648.hc: multiple definition of `__stginit_PointerPrim_' c:\haskell\myfptoolsdirectory\fptools\hdirect\lib/libHShdirect.a(PointerPrim__12.o)(.text+0x0):ghc2036.hc: first defined here c:\haskell\myfptoolsdirectory\fptools\hdirect\comlib/libHScom.a(PointerPrim.o)(.text+0x6da):ghc2648.hc: multiple definition of `__stginit_PointerPrim' c:\haskell\myfptoolsdirectory\fptools\hdirect\lib/libHShdirect.a(PointerPrim__12.o)(.text+0x44):ghc2036.hc: first defined here Marc

1 0

Combinatory logic interpreter?
by Colin Paul Adams 10 Feb '06

10 Feb '06

Is there any Haskell code around that can interpret combinatory logic expressions? -- Colin Adams Preston Lancashire

3 2

Language shootout (reloaded)
by dons＠cse.unsw.edu.au 10 Feb '06

10 Feb '06

Just so we can feel that we're doing the right things :) On the great language shootout, as of last night, we're: * Ranked overall number 1, by a good margin: * Ranked number 1 on lines of code * Ranked number 2 on speed. http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=all Thanks to everyone who's contributed to the entries over the last few weeks, on the wiki, and on this list. Cheers, Don P.S. I remember having a discussion on #haskell 2 weeks ago where we all agreed that Haskell placing #1 was pretty much impossible. Did we have an inferiority complex?

9 8

RE: [Haskell-cafe] Language shootout (reloaded)
by Simon Peyton-Jones 10 Feb '06

10 Feb '06

Congratulations, guys! An amazing, and (maybe it should have been!) to me surprising, achievement. Simon | -----Original Message----- | From: haskell-cafe-bounces(a)haskell.org [mailto:haskell-cafe-bounces@haskell.org] On Behalf Of | Donald Bruce Stewart | Sent: 10 February 2006 01:35 | To: haskell-cafe(a)haskell.org | Subject: [Haskell-cafe] Language shootout (reloaded) | | Just so we can feel that we're doing the right things :) | | On the great language shootout, as of last night, we're: | | * Ranked overall number 1, by a good margin: | * Ranked number 1 on lines of code | * Ranked number 2 on speed. | | http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=all | | Thanks to everyone who's contributed to the entries over the last few | weeks, on the wiki, and on this list. | | Cheers, | Don | | P.S. I remember having a discussion on #haskell 2 weeks ago where we all | agreed that Haskell placing #1 was pretty much impossible. Did we have | an inferiority complex? | _______________________________________________ | Haskell-Cafe mailing list | Haskell-Cafe(a)haskell.org | http://www.haskell.org/mailman/listinfo/haskell-cafe

1 0

Re: Betreft: Re: [Haskell-cafe] Elementary HaXml question
by Malcolm Wallace 10 Feb '06

10 Feb '06

Koen.Roelandt(a)mineco.fgov.be writes: > foldXml (txt ?> keep :> > fontSize4 /> tag "em" ?> mkSubtitle :> > children) > now transforms > <font size="4"><em>Hello World</em></font> > into > <h1 class="subtitle"><em><Hello World</em></font> > > I still have a problem with the other example: > <em><font size="4">Hello World</font></em> Ah, I see. In the first case, you simply change the <font> tag into <h1>, keeping the <em> tag exactly the same. But in the second case, you not only want to change <font> into <h1>, but also to move the <em> tag from outside the <font> to inside. I think you need something like this: mkSubtitle `o` (tag "em" /> fontSize4 /> mkElem "em") |>| children i.e. build an <em> around the children of the <font> tag, and then build the <h1> subtitle around the outside of that. Regards, Malcolm

2 1

Numerical methods in Haskell
by Creighton Hogg 10 Feb '06

10 Feb '06

Hi, Between google searching and looking through the activity report, I take it that no one has really developed serious libraries for matrix manipulations, diff eqs, etc. Are there any practical reasons for this or is it just a matter of the haskell community being small and there not being many people interested in something so specialized?

2 1

FFI Wiki Page (was: FFI question)
by Dimitry Golubovsky 09 Feb '06

09 Feb '06

Bulat wrote: > we need to establish FFI > page on the wiki and give at least links to all this packages and > small info about using FFI with ghc/hugs There is such a page (to be more precise, a section of a page): http://haskell.org/hawiki/LibrariesAndTools section "FFI Preprocessors" Also there are http://haskell.org/hawiki/FfiCookbook and http://haskell.org/hawiki/FfiTutorial pages. -- Dimitry Golubovsky Anywhere on the Web

1 0

FFI question
by Frederico Franzosi 09 Feb '06

09 Feb '06

I'll try to make it short. I'm developping a package wich imports C functions. the fact is that when I try to compile if I call the compiler in the usual way, using -package and -llib it gives an undefined reference error... For example if I use: $ghc -package PKGname-PKGversion -fffi -o main Main.hs -llib it gives an error message (undefined references of objects from PKG to functions in lib) but if I use: $ghc -fffi -o main Main.hs -llib /path/to/PKGlib.a it compiles correctly. and also when I do: $ghc -package PKGname-PKGversion -fffi -o main Main.hs /path/to/lib.a it compiles correctly. I know that it is surely my mistake but I didn't find anything helpful in docummentation or in FFI list. I know that I should send this message to ffi mailing list but it is kind of stopped and I think I'll get my answers here... Thank you for the atention.

7 8

Elementary HaXml question
by Koen.Roelandt＠mineco.fgov.be 09 Feb '06

09 Feb '06

Hi all, I'm new to Haskell and HaXml and I'm playing around with the latter to clean some (well-formed) 'legacy' html. This works fine except for the following cases. Some of the elements to be cleaned are: <font size="4"><i>Hello World</i></font> <i><font size="4">Hello World</font></i> This should become: <h1 class="subtitle">Hello World</h1> I can build filters to find <font>, <font size="4"> or <i>, but the combination does not seem to work. >From what I could gather from the documentation, it should be something like: foldXml (txt ?> keep :> (attrval("size",AttValue[Left "4"]) `o` tag "font") /> tag "i" ?> replaceTag "h1" :> children) This doesn't work. I am clearly missing something elementary here, so any hints are welcome. Cheers, K.

2 2

Type inference
by Fred Hosch 09 Feb '06

09 Feb '06

Is type inferencing in Haskell essentially the same as in SML? Thanks. ---Fred Hosch

4 9

Fast Mutable Variables for the IO and ST monads
by Bulat Ziganshin 09 Feb '06

09 Feb '06

Hello haskell-cafe, only for the souls interesting in writing efficient programs :) i included in this letter my own module used for fast variables - "DataVariables.hs". and "wc.hs" contains example of using these vars. DataVariables module contains many interesting beasts but for optimization purposes newVar/readVar/writeVar is especially interesting. these primitives mimics IORef interface, supports Int/Word/Bool variables, works in IO and ST monad and nevertheless they are compiled to the most-effective GHC primitives. moreover, they can be used outside GHC (of course, on other compilers just IORef/STRer are used to implement this) it was impossible to add Ptr's support with the same interface, so Ptrs supported by another functions - newPtr/readPtr/writePtr and don't work in ST monad. in other aspects fast Ptr variables are the same as fast general variables if you need to include these variables in data structures, use the types IntVar/WordVar/BoolVar/"MutPtr a" for the IO monad, and "MutInt s", "MutWord s", "MutBool s" for the ST monad another interesting beasts are unboxed Int and Bool values. although GHC can automatically unbox vars in some cases, using this module will guarantee unboxing. moreover, some operations such as bit shifts, are ineffective without explicit unboxing because their definitions don't use unboxed operations directly, but adds additional checks and conversions. using types FastInt and FastBool and associated operations guarantee unboxing on GHC and works on other compilers (where just boxed values are used to implement these types). one place from my lib where i used them to speed up program: putWord32be :: (Stream m h, Integral int, Bits int) => h -> int -> m () putWord32be h w = do let n = iUnbox (fromIntegral w) vPutByte h $! iBox ((n >># _ILIT 24) ) vPutByte h $! iBox ((n >># _ILIT 16) &# _ILIT 0xff) vPutByte h $! iBox ((n >># _ILIT 8) &# _ILIT 0xff) vPutByte h $! iBox ((n ) &# _ILIT 0xff) -- Best regards, Bulat mailto:bulatz@HotPOP.com

4 9

Re: Re[2]: Tuple-like constructors
by Robert Dockins 08 Feb '06

08 Feb '06

[Moved to cafe; time to stop bothering the Haskell' committee...] On Feb 8, 2006, at 1:19 PM, Malcolm Wallace wrote: > Robert Dockins <robdockins(a)fastmail.fm> writes: > >> instance (Bin a,Bin b,Bin c,Bin d) => Bin (a,b,c,d) >> >> See the problem? Sooner or later (probably sooner) I'll get tired of >> typing. I have to write down an 'instance' declaration for each >> value of n. Clearly this can't generalize to all n. > > There has been a suggestion that the 'deriving' mechanism be de- > coupled > from the datatype declaration. Together with a generic default > definition, that means you could write something like > > deriving Bin for (,,,,,,,,,,,,,,,,,,,,) > > and hence not need to write the tedious instance header yourself, > since the compiler can easily infer it. Humm. That is nice, and it would help keep my fingers from cramping, but it doesn't solve the root objection. Consider machine-generated code using largish tuples. Should the generated code include the 'deriving' clause? If no, someone from outside has to supply it. If yes, I have to deal with overlapping instance from other generated code. If maybe, I have to know when to generate it and when not. Solvable? yes, but kind of ugly. Acceptable? probably. Now, (speculatively combining a whole host of language extensions) if I could write down: deriving Bin for (_::*) Which would mean "derive (as necessary) Bin for all types at kind * where the suitable preconditions can be found/derived" I'd be completely happy. Humph. That's more or less adding an axiom schema to the constraint solving system of the typehecker which could be realized at link-time via C++ style template thingies or at compile-time by whole-program analysis. I am simultaneously appalled by and attracted to this idea..... Rob Dockins Speak softly and drive a Sherman tank. Laugh hard; it's a long way to the bank. -- TMBG

2 1

Spotted in Haskell Weekly News: February 06, 2006
by Graham Klyne 08 Feb '06

08 Feb '06

Donald Bruce Stewart wrote: > Haskell Weekly News: February 06, 2006 : > Oleg says, "The implementation of RSA on the type level is left > for future work". Methinks this gives a whole new meaning to "type security". :) #g -- Graham Klyne For email: http://www.ninebynine.org/#Contact

1 0