- Haskell-Cafe - Haskell.org

Re: [Haskell] Dynamic binding
by Andrew Ward 24 Jun '05

24 Jun '05

Jan-Willem Maessen wrote: > > On Jun 22, 2005, at 9:38 PM, Andrew Ward wrote: > >> Pal-Kristian Engstad wrote: >> >>> On Wednesday 22 June 2005 05:38 pm, Andrew Ward wrote: >>> >>>> What would be the normal way for a Haskell programmer to handle the >>>> typical shape example in beginner OO tutorials? >>>> >>> >>> By not doing OO. You have to ask yourself, what is the purpose >>> and/or benefit of using OO? In C++, OO is _useful_ because you >>> restrict the problems of mutable data (by enclosing it in C++ classes). >>> >>> ML type languages have other methods of doing things, and guess >>> what, OO is not that needed for these languages. Sum-types, >>> pattern-matching and data constructors make half of the need for OO >>> go away. Higher order functions and make it even less needed. For >>> the rest, there's always work-arounds. >>> >>> PKE. >>> >> To handle the problem of drawing all shapes, in c++, I would have a >> list of shape pointers: >> >> struct shape{ virtual void draw(...);}; >> struct circle : public shape {...}; >> struct square : public shape {...}; >> std::list<shape *> shapes; >> for(std::list<shape *>::iterator it = shapes.begin();it != >> shapes.end();++it) >> { (*it)->draw(...); } >> >> This general pattern of dynamic binding I use over and over again. >> Could you give me some example code of this type of thing handled in >> Haskell's way? Assuming that the number of classes deriving from >> shape might get quite large. > > > It seems to me that if this is the specific problem being addressed, a > list of higher-order functions is exactly the right solution in any > ML-like language, and this is why there have been several responses to > that effect. > > I do think it's fair to say "consider changing the way you think" to > OO programmers trying to learn Haskell. If we were on an OOP mailing > list, I could ask for days how to simulate pattern matching and > algebraic types---and get a nonsensical runaround involving the > visitor pattern and huge swaths of unreadable code. > > Ralf, I think it's incumbent on you, having said several times "that > isn't solving the problem", to more clearly explain what problem you > think exists and cannot be solved gracefully (I suspect it has to do > with extensible down-casting---which is indeed hard in Haskell, but > many reasonable people might consider irrelevant or even overtly bad). > > Andrew, if this isn't the problem you're actually trying to solve, can > you explain why a simple list of functions doesn't help you? > I think I will need to learn a little more Haskell before I can appreciate all the different responses. However I will try to answer your question. Assuming I have a list of shapes somewhere in my program. The list would be heterogeneous, or at least contain data constructed using various different type constructors. The list would also be changing all the time, for example in a drawing program. So, using the suggestion of a list of drawing functions, would this not mean I would have to alter the list of drawing functions each time I alter the list of shapes? It does not seem an extensible solution. When it comes to adding new types or new functionality for existing types would it not be easy to miss updating a part of the program source that needed updating? The example I have found the simplest is the one given by Lennart, however this to me seems equivalent to the way people use switch statements in C to mimic dynamic binding. As I said before, I will have to learn more haskell and investigate OOHaskell properly. Thanks for all your responses. > -Jan-Willem Maessen > > >> Andrew Ward. >> >> >> _______________________________________________ >> Haskell mailing list >> Haskell(a)haskell.org >> http://www.haskell.org/mailman/listinfo/haskell > > >

2 1

RE: Re[4]: [Haskell] Dynamic binding
by Ralf Lammel 24 Jun '05

24 Jun '05

Bulat, a) building (i) a list of data is fundamentally different from building (ii) a list of anticipated results of each datum. I would be surprised to hear that this counts as a valid technique. BTW, you can do the *same* in a lazy OO language. (No lazy OO language at hand -- well lazyness can be simulated.) Anyway, even if people end up coding as you propose, it won't work in general. Think of mutation methods that change the state but preserve the type. Then your list will still be heterogonous. NO? b) You wrote: > this state is just don't need to appear in interface definition :) > circle x y r You are not talking about state but constructor arguments. Please have a look here: http://www.angelfire.com/tx4/cus/shapes/cpp.html Circle Interface (Circle.h) class Circle: public Shape { public: Circle(int newx, int newy, int newradius); int getRadius(); void setRadius(int newradius); void draw(); private: int radius; }; [See getRadius and setRadius) In OO, mutable state tends to leak to the interface, perhaps not as public fields, perhaps not even as public proxy properties (but in the shapes example, it *does*), but then still as sublcass-specific interface member. Anyway, I am happy to add your and Lennart's proposals as appendices to the OOHaskell saga. And finally, Bulat, I agree with your point that > original exercise was about OO way to solve some problem. i want > to say that in Haskell it's better in most cases to use another, > functional way Thanks, Ralf (returning to his VB problem) > -----Original Message----- > From: Bulat Ziganshin [mailto:bulatz@HotPOP.com] > Sent: Thursday, June 23, 2005 4:55 AM > To: Ralf Lammel > Cc: Andrew Ward; Pal-Kristian Engstad; haskell-cafe(a)haskell.org > Subject: Re[4]: [Haskell] Dynamic binding > > Hello Ralf, > > Thursday, June 23, 2005, 11:36:20 AM, you wrote: > > >> just create list of draw functions itself: > >> > >> [drawCircle (10,10) 5, drawSquare (20,20) 10] > > RL> No! the exercise is about lists of shapes > RL> not lists of results of drawing shapes. > RL> This is clearly a major difference. > > in cases where you need to call only one function on created objects, > you can just insert in list calls to this functions (not their results! i > suppose > that drawXXX functions has "... -> IO ()" type) > > in cases where you need to call several functions for this object, you > can insert in list tuple or structure for each object, as i do in next > example. original exercise was about OO way to solve some problem. i want > to say that in Haskell it's better in most cases to use another, > functional way > > RL> Bulat wrote: > > >> for more complex tasks - declare interface as a structure: > >> > >> data ShapeInterface = Shape { draw :: IO (), > >> moveTo :: Point -> IO (), > >> calcArea :: Float > >> } > > RL> No! You miss the point that the different shapes > RL> differ regarding state types. > RL> You don't have a chance when you use one datatype. > > this state is just don't need to appear in interface definition :) > > see for example: > > data ShapeInterface = Shape { draw :: IO (), > calcArea :: Float > } > > circle x y r = Shape { draw = drawCircle x y r, > calcArea = pi*r*r > } > > square x y size = Shape { draw = drawSquare x y size, > calcArea = size*szie > } > > figures = [circle 1 2 3, square 4 5 6, circle 7 8 9] > > > > > > if you need to maintain mutable state, this is also not a problem: > > data ShapeInterface = Shape { draw :: IO (), > moveTo :: (Int,Int) -> IO (), > calcArea :: Float > } > circle x y r = do > center <- ref (x,y) > return Shape { draw = val center >>= drawCircle r > , moveTo = (center=:) > , calcArea = pi*r*r > } > main = do > figures <- sequence [circle 1 2 3, square 4 5 6, circle 7 8 9] > mapM_ draw figures > mapM_ (moveTo (0,0)) figures > mapM_ draw figures > > ref=newIORef > val=readIORef > (=:)=writeIORef > > > RL> haskell-cafe? > > as you wish :) > > -- > Best regards, > Bulat mailto:bulatz@HotPOP.com > >

2 1

Re[4]: [Haskell] Dynamic binding
by Bulat Ziganshin 23 Jun '05

23 Jun '05

Hello Ralf, Thursday, June 23, 2005, 11:36:20 AM, you wrote: >> just create list of draw functions itself: >> >> [drawCircle (10,10) 5, drawSquare (20,20) 10] RL> No! the exercise is about lists of shapes RL> not lists of results of drawing shapes. RL> This is clearly a major difference. in cases where you need to call only one function on created objects, you can just insert in list calls to this functions (not their results! i suppose that drawXXX functions has "... -> IO ()" type) in cases where you need to call several functions for this object, you can insert in list tuple or structure for each object, as i do in next example. original exercise was about OO way to solve some problem. i want to say that in Haskell it's better in most cases to use another, functional way RL> Bulat wrote: >> for more complex tasks - declare interface as a structure: >> >> data ShapeInterface = Shape { draw :: IO (), >> moveTo :: Point -> IO (), >> calcArea :: Float >> } RL> No! You miss the point that the different shapes RL> differ regarding state types. RL> You don't have a chance when you use one datatype. this state is just don't need to appear in interface definition :) see for example: data ShapeInterface = Shape { draw :: IO (), calcArea :: Float } circle x y r = Shape { draw = drawCircle x y r, calcArea = pi*r*r } square x y size = Shape { draw = drawSquare x y size, calcArea = size*szie } figures = [circle 1 2 3, square 4 5 6, circle 7 8 9] if you need to maintain mutable state, this is also not a problem: data ShapeInterface = Shape { draw :: IO (), moveTo :: (Int,Int) -> IO (), calcArea :: Float } circle x y r = do center <- ref (x,y) return Shape { draw = val center >>= drawCircle r , moveTo = (center=:) , calcArea = pi*r*r } main = do figures <- sequence [circle 1 2 3, square 4 5 6, circle 7 8 9] mapM_ draw figures mapM_ (moveTo (0,0)) figures mapM_ draw figures ref=newIORef val=readIORef (=:)=writeIORef RL> haskell-cafe? as you wish :) -- Best regards, Bulat mailto:bulatz@HotPOP.com

2 1

RE: [Haskell-cafe] Re: Using type classes for polymorphism of dataconstructors
by Ralf Lammel 23 Jun '05

23 Jun '05

Thomas Sutton wrote: > Main> :t (Impl (Prop "p") (Poss (Prop "p"))) -- "p -> <>p" > Impl (Prop "p") (Poss (Prop "p")) :: PC So the type says that this is a formula in the predicate calculus? (Even though you combine two formulae of modal logics.) Are you happy with that? Just wondering? Also, are you fine to take the conjunction of a PC and Modal formula, which is admitted by your types? (If not, you may want to use phantom types.) So it seems like the type doesn't carry a lot of information. Then we could also use different types. In particular, we could do without existentials. The types we get will (additionally) record the arity of the outermost constructor. (In your model, the type records the kind of logic of the outermost constructor.) class (Show f) => Formula f data PC0 = Prop String data Formula x => PC1 x = Neg x data (Formula x, Formula y) => PC2 x y = Conj x y | Disj x y | Impl x y instance Formula PC0 instance Formula x => Formula (PC1 x) instance (Formula x, Formula y) => Formula (PC2 x y) instance Show PC0 where show (Prop s) = s instance Formula x => Show (PC1 x) where show (Neg x) = "~" ++ show x instance (Formula x, Formula y) => Show (PC2 x y) where show (Conj a b) = (show a) ++ " /\\ " ++ (show b) show (Disj a b) = (show a) ++ " \\/ " ++ (show b) show (Impl a b) = (show a) ++ " -> " ++ (show b) data Formula x => Modal1 x = Poss x | Necc x instance Formula x => Formula (Modal1 x) instance Formula x => Show (Modal1 x) where show (Poss a) = "<>" ++ (show a) show (Necc a) = "[]" ++ (show a) main = do print (Prop "p") print (Poss (Prop "p")) print (Impl (Prop "p") (Poss (Prop "p"))) (Again, just wondering whether this might be an Ok solution.) Thomas Sutton wrote: > > In Java (C#, Python, etc) I'd do this ... Please post your Java code. I might want to add it to the OOHaskell demo suite. Ralf > -----Original Message----- > From: haskell-cafe-bounces(a)haskell.org [mailto:haskell-cafe- > bounces(a)haskell.org] On Behalf Of Thomas Sutton > Sent: Thursday, June 23, 2005 3:27 AM > To: haskell-cafe(a)haskell.org > Subject: [Haskell-cafe] Re: Using type classes for polymorphism of > dataconstructors > > On 11/06/2005, at 11:18 PM, Thomas Sutton wrote: > > In Java (C#, Python, etc) I'd do this by writing an interface > > Formula and have a bunch of abstract classes (PropositionalFormula, > > ModalFormula, PredicateFormula, etc) implement this interface, then > > extend them into the connective classes Conjunction, Disjunction, > > etc. The constructors for these connective classes would take a > > number of Formula values (as appropriate for their arity). > > > > I've tried to implement this sort of polymorphism in Haskell using > > a type class, but I have not been able to get it to work and have > > begun to work on implementing this "composition" of logics in the > > DSL compiler, rather than the generated Haskell code. As solutions > > go, this is far from optimal. > > > > Can anyone set me on the right path to getting this type of > > polymorphism working in Haskell? Ought I be looking at dependant > > types? > > I've finally managed to find a way to get this to work using > existentially quantified type variables and am posting it here for > the benefit of the archives (and those novices like myself who will > look to them in the future). My solution looks something like the > following: > > A type class over which the constructors ought to be polymorphic: > > > class (Show f) => Formula f where > > language :: f -> String > > A type exhibiting such polymorphism: > > > data PC > > = Prop String > > | forall a. (Formula a) => Neg a > > | forall a b. (Formula a, Formula b) => Conj a b > > | forall a b. (Formula a, Formula b) => Disj a b > > | forall a b. (Formula a, Formula b) => Impl a b > > instance Formula PC where > > language _ = "Propositional Calculus" > > instance Show PC where > > show (Prop s) = s > > show (Neg s) = "~" ++ (show s) > > show (Conj a b) = (show a) ++ " /\\ " ++ (show b) > > show (Disj a b) = (show a) ++ " \\/ " ++ (show b) > > show (Impl a b) = (show a) ++ " -> " ++ (show b) > > Another such type: > > > data Modal > > = forall a. (Formula a) => Poss a > > | forall b. (Formula b) => Necc b > > instance Formula Modal where > > language _ = "Modal Logic" > > instance Show Modal where > > show (Poss a) = "<>" ++ (show a) > > show (Necc a) = "[]" ++ (show a) > > Some example values of those types: > > Main> :t (Prop "p") -- "p" > Prop "p" :: PC > > Main> :t (Poss (Prop "p")) -- "<>p" > Poss (Prop "p") :: Modal > > Main> :t (Impl (Prop "p") (Poss (Prop "p"))) -- "p -> <>p" > Impl (Prop "p") (Poss (Prop "p")) :: PC > > I also have a sneaking suspicion I'd also be able to solve this > problem using dependant types, but I have not investigated that > approach. > > Cheers, > Thomas Sutton > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe

1 0

Using type classes for polymorphism of data constructors
by Thomas Sutton 23 Jun '05

23 Jun '05

Hi all, I've just started working on a theorem prover (labelled tableaux in case anyone cares) in Haskell. In preparation, I've been attempting to define some data types to represent logical formulae. As one of the requirements of my project is generality (i.e. it must be easily extendible to support additional logics), I've been attempting to build these data types modularly. The end goal in all of this is that the user (perhaps a logician rather than a computer scientist) will describe the calculus they wish to use in a simple DSL. This DSL will then be translated into Haskell and linked against some infrastructure implementing general tableaux bits and pieces. These logic implementations ought to be composable such that we can define modal logic to be propositional calculus with the addition of [] and <>. In Java (C#, Python, etc) I'd do this by writing an interface Formula and have a bunch of abstract classes (PropositionalFormula, ModalFormula, PredicateFormula, etc) implement this interface, then extend them into the connective classes Conjunction, Disjunction, etc. The constructors for these connective classes would take a number of Formula values (as appropriate for their arity). I've tried to implement this sort of polymorphism in Haskell using a type class, but I have not been able to get it to work and have begun to work on implementing this "composition" of logics in the DSL compiler, rather than the generated Haskell code. As solutions go, this is far from optimal. Can anyone set me on the right path to getting this type of polymorphism working in Haskell? Ought I be looking at dependant types? Thanks in advance, Thomas Sutton

2 3

Compiling ghc with openal
by Ron 22 Jun '05

22 Jun '05

Hi, I am configuring ghc from source(no Ports(since that is broken for this purpose too)) on freebsd 5.4: CPPFLAGS="-I/usr/X11R6/include/ -I/usr/local/include/ -L/usr/local/lib/ " ./configure I had issues with opengl, which were resolved by adding the first dir to CPPFLAGS, but it seems impossible to get ghc to recongize OpenAL, while I have it in /usr/local/lib libogg.a libogg.so libopenal.a libopenal.so.0 liborbit-c-backend.a libogg.la libogg.so.5 libopenal.so libopenal.so.0.0.8 In /usr/local/include/AL, I have al.h alc.h alctypes.h alext.h alexttypes.h altypes.h alut.h aluttypes.h I get this: checking for library containing alGenSources... no configure: WARNING: no OpenAL library found, so this package will not be built So, basically the question is: how do I tell ghc to "find" OpenAL? Regards, Ron

1 0

Buddha and GHC 6.4
by Jim Apple 21 Jun '05

21 Jun '05

Has anyone gotten Buddha (now called "plargleflarp" http://www.cs.mu.oz.au/~bjpop/buddha/) do work with GHC 6.4? I'm getting ghc-6.4: unknown package: buddha Jim P.S. What happened to the old name?

4 6

Announcing cross-platform implementation of Pan
by Sean Seefried 19 Jun '05

19 Jun '05

Hi all, I'd like to announce a new cross-platform implementation of Conal Elliott's image synthesis language - Pan. For more information see: http://www.cse.unsw.edu.au/~sseefried/pan/ This is very much an alpha release of the software. I have only built and tested it on Mac OS X and Linux, although it has been tested on Windows by another. Features: * Uses Don Stewart's hs-plugins library to dynamically load and display effects. * Cross-platform. Uses Daan Leijen's wxHaskell and Sven Panne's HOpenGL libraries. * Uses the Haskell Cabal libraries. Builds with version released in GHC 6.4. Drawbacks: * Not as fully featured as the original. * Not as fast. Implemented directly in Haskell instead of generating C code. Sean

1 0

Functional Programming lecture on video
by Carsten Otto 17 Jun '05

17 Jun '05

Hello! I am student at the RWTH Aachen (Germany) and this semester professor Giesl is lecturing "Functional Programming". In the first part of this lecture Haskell is presented and all basic concepts are explained (recursion, pattern matching, lambda expressions, user declared data types, ...). Later in this lecture the theoretical background is the main topic (semantics, lambda calculus, ...). If you are interested, you can watch this lecture on video: http://video.s-inf.de/#FP.2005-SS-Giesl.(COt).HD_Videoaufzeichnung The lecture of course is being held in English. The video is encoded with XviD 1250 KBit and 64 KBit Mono-MP3. The download-server is connected with about 100 MBit, so your download should be quite fast. Have fun, -- Carsten Otto carsten.otto(a)gmail.com www.c-otto.de

1 0

Hugs parse error in library file
by james pentland 16 Jun '05

16 Jun '05

i am working on ch 15, School of Expression. unfortunately, import Memo causes an Hugs error (see below). also, memoB in ch 15 calls memo1, but there is no memo1 in h:\Program Files\Hugs98\libraries\Hugs\Memo.hs this module contains memoN :: Int -> (a -> b) -> (a -> b) memoN = mkMemo eql hash of course, it should be easy enough to write memo1 :: (a -> b) -> (a -> b) memo1 = memoN 1 Memo.hs imports ST, which is where Hugs reports a parse error. h:\Program Files\Hugs98\libraries\Hugs\ST.hs contains at line 50 newtype ST s a = ST (forall r. (a -> r) -> r) what is the workaround? pentland@thetis /ntd/lo $ hugs Hugs session for: H:\Program Files\Hugs98\libraries\Hugs\Prelude.hs H:\Program Files\Hugs98\libraries\Prelude.hs H:\Program Files\Hugs98\libraries\Hugs\Base.hs Type :? for help Hugs.Base> :v -- Hugs Version 20050113 Hugs.Base> :load xva ERROR "H:\Program Files\Hugs98\libraries\Hugs\ST.hs":50 - Syntax error in type expression (unexpected `.') Animation> pentland@thetis /ntd/lo $ cat xva.hs import Animation (picToGraphic) import Shape import Picture import Memo import SOEGraphics hiding (Region, Event) import qualified SOEGraphics as G (Region, Event) import Draw (xWin,yWin,intToFloat) import Win32Misc (timeGetTime) -- import Word (word32ToInt) import Channel type Time = Float type UserAction = G.Event newtype Behavior a = Behavior (([Maybe UserAction],[Time]) -> [a]) newtype Event a = Event (([Maybe UserAction],[Time]) -> [Maybe a]) time :: Behavior Time time = Behavior (\(_,ts) -> ts) __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1