- Haskell-Cafe - Haskell.org

Re: 2 points on language
by D. Tweed 24 Aug '01

24 Aug '01

On Fri, 24 Aug 2001, S.D.Mechveliani wrote: > D. Tweed <tweed(a)compsci.bristol.ac.uk> writes > > On a purely pragmatic note ( :-) ) these optimisations can't necessarily > > be safely used in three of the four most common cases, namely Int, Float > > and Double because of the restricted range of intermediates. > Int, Float, Double > are not supposed to be provided with the CommutativeRing instance. > And most common cases are, to my mind, different: > Integer, Rational, (Integer, Rational), Polynomial Integer ... Ah, I must admit I'd rather missed the point that you were only intending them to apply to things that were of the algebraic class `Commutative Ring'. I was thinking that it'd be something where a particular funciton over a (possibly polymorphic) type could be declared associative, commutative, etc and then those more elementary classes came together in the particular example of the CommutativeRing. For example, the operator ++ on lists is associative, there are useful computational savings to be made utilising this (IIRC Lennart Augustsson had this specal cased in the hbc compiler), but it's difficult to see how you gain anything by trying to pin down any of the other properties of lists that would figure in the mathematically standard classes (there's no inverse as far as I can see so we don't even make it as far as a group). I think from the point of view of lobbying compiler implementors to put in the spade work to make this happen you're more likely to get support if the elegant algebraic classes come as a neat combination of more low-level, `unstructured' features. Don't get me wrong, I like the idea of extending the power of the Haskell compilers, I was just pointing out that the obvious thought that associative- and commutative-rewrite based optimisation `will also be very helpful for programs that are doing "everyday arithmetic" operations (say running a Kalman filter over a sequence of data represented using Doubles)' isn't necessarily true, and it certainly wasn't obvious to me: it was only happening upon the reference in the MetaFont book that made me realise it. ___cheers,_dave________________________________________________________ www.cs.bris.ac.uk/~tweed/pi.htm |tweed's law: however many computers email: tweed(a)cs.bris.ac.uk | you have, half your time is spent work tel: (0117) 954-5250 | waiting for compilations to finish.

1 0

Braces vs. Whitespace
by Ashley Yakeley 22 Aug '01

22 Aug '01

At 2001-08-21 20:31, Dean Herington wrote (on the Haskell list): >Now realizing that the "where" keyword is optional when there are no >accompanying declarations, I think it would be preferable to omit the >"where" keyword to indicate that the absence of declarations is >intentional. Personally I prefer the explicit braces/semicolons syntax rather than worrying about whitespace: module DeepSeq where { class DeepSeq a where { deepSeq :: a -> b -> b; deepSeq = seq; -- default, for simple cases }; ... instance DeepSeq Ordering; instance DeepSeq Integer; instance DeepSeq Int; instance DeepSeq Float; instance DeepSeq Double; } ...or if you prefer... module DeepSeq where { class DeepSeq a where { deepSeq :: a -> b -> b; deepSeq = seq; -- default, for simple cases }; ... instance DeepSeq Ordering; instance DeepSeq Integer; instance DeepSeq Int; instance DeepSeq Float; instance DeepSeq Double; } etc. Basically you put a braced block after every 'do', 'in' and 'where', and semicolons after every declaration. When I was first learning Haskell, I got massively confused by the whitespace rules, especially with mixed tabs/spaces etc. I reckon the braces style is more readable to anyone used to C/C++/Java etc. -- Ashley Yakeley, Seattle WA

1 0

Re: Subtyping
by Fritz K Ruehr 21 Aug '01

21 Aug '01

[sorry about that, let's try again...] Ross Paterson wrote (in regards variations on factorial): > Another version: > > fac = foldr (*) 1 . unfoldr (\n -> guard (n > 0) >> return (n, n-1)) Thanks for forwarding this; may I quote it on the web site? (with full attribution, of course) If so, you are welcome to give it a name and a pithy little "slogan" (also for the one below). > Also, the combinatory programmer would regard defining y by recursion > as cheating, and would probably prefer one of these versions: > > newtype SelfApp a = SA (SelfApp a -> a) > > selfApply :: SelfApp a -> a > selfApply (SA f) = f (SA f) > > yCurry :: (a -> a) -> a > yCurry f = selfApply (SA (\x -> f (selfApply x))) > > yTuring :: (a -> a) -> a > yTuring = selfApply (SA (\x f -> f (selfApply x f))) Yes, this is definitely the right way to go here. I comment on the site about the fact that we should really *define* Y (and I felt a bit guilty about it :) ), but I didn't try hard enough to get around the typing issue. You have done so quite elegantly. Of course, there is also a trade-off at some point between the impact of the various examples and their sheer number (I added an environment-model based interpreter yesterday that started to push this boundary). Still, more monads (arrows, even?) are needed. -- Fritz Ruehr PS: for those who missed it, the site is at: <http://www.willamette.edu/~fruehr/haskell/evolution.html>

1 0

Re: Subtyping
by Fritz K Ruehr 21 Aug '01

21 Aug '01

Ross Paterson wrote (in regards variations on factorial):

1 0

Re: GHC Mandrake problem
by Pixel 21 Aug '01

21 Aug '01

"phaywood(a)mail.iprimus.com.au" <phaywood(a)pop.iprimus.com.au> writes: > Hello, I tried to install version ghc 5.00.2 & prof from mandrake (vers 8.0 > with current updates). It installed fine, but i get this message when i try to > run ghci: > > > Loading package std ... ghc-5.00.2: can't find .o or .so/.DLL for: m (lib.so: > cannot open shared object file: No such file or directory) booh :'-( it tries to find m.o then tries lib.so > > Can anyone tell me what's missing, or how to fix this?? i don't know either. It used to work. I'm currently trying a rebuild...

1 0

Re: Subtyping
by Fritz K Ruehr 21 Aug '01

21 Aug '01

Frank Atanassow wrote: > This is a situation where the explicit fixpoint datatype comes in handy: > > data Fix f = In (f (Fix f)) > > ... > > BTW, this is not a beginner, or even intermediate, Haskell programmer's > technique. For those who are interested in what exactly is or isn't a beginner or intermediate Haskell programmer's technique, I humbly offer the following <http://www.willamette.edu/~fruehr/haskell/evolution.html> which includes examples of the afore-mentioned approach, along with many others, from accumulating parameters to comonads. Fair warning: tongue held firmly in cheek ... -- Fritz Ruehr fruehr(a)willamette.edu

2 1

Modular arithmetic
by Dylan Thurston 21 Aug '01

21 Aug '01

They said it couldn't be done! They said Haskell didn't have dependent types! Here's a way to mimic dependent types using existential types, illustrated by an implementation of modular arithmetic. To try it out, load modulus.hs and try something like Modulus> inModulus (mkModulus (1234567890123::Integer)) (^98765432198765) 2 1160454313058 to compute 2 to the 98765432198765'th power modulo 1234567890123. The key is the definitions at the top of TypeVal.hs: > class TypeVal a t | t -> a where > -- typeToVal should ignore its argument. > typeToVal :: t -> a > > data Wrapper a = forall t . (TypeVal a t) => Wrapper t > > class ValToType a where > valToType :: a -> Wrapper a `valToType' takes a value `x' and returns a (wrapped version of a) fake value in a new type; from the new type, `x' can be recovered by applying typeToVal. This code works under ghc. It uses existentially quantified data constructors, scoped type variables, and explicit universal quantification. Enjoy, Dylan Thurston

1 1

Singular Type Constructor
by Ashley Yakeley 21 Aug '01

21 Aug '01

Consider: -- data T a = T -- Is there anything in the Prelude or a standard library that looks like this? Should there be? What should it be called? 'Singleton'? 'Reified'? 'T'? I looked but couldn't find anything. Such a simple type constructor is in fact very useful when you need to pass types around to disambiguate class instances. For instance: -- class HasTypeName t where getTypeName :: T t -> String instance HasTypeName Int where getTypeName T = "Int" instance HasTypeName () where getTypeName T = "()" intName = getTypeName (T :: T Int) -- -- Ashley Yakeley, Seattle WA

3 2

deriving
by Mark Carroll 20 Aug '01

20 Aug '01

Does anyone know of any Haskell texts that introduce 'deriving' fairly comprehensively? All I've been able to find out from what I have is that your type can derive Eq, Ord, Show etc. and then they'll magically work as if you'd written the instance declarations yourself. I've no idea how you can make your own type classes that types can 'derive' from, etc. IYSWIM, or really what's going on at all. -- Mark

2 2

Re: Subtyping
by Mansour Al-Mutairi 20 Aug '01

20 Aug '01

Hi, I am a newbie but I think that what you are proposing can be simulated in haskell. It is like subclassing in OOP. You can create a new type that is an extension of the older one. Unlike OOP overloading, you will need to create a new function that can work on the new type. >data SomeObject = SomeInt Int | SomeString String | _ will become: data SomeObject = SomeInt Int | SomeString String >...later... > >data SomeObject |= SomeReal Double will become: data SomeObject2 = SO SomeObject | SomeReal Double >same :: SomeObject -> SomeObject -> Bool >same (SomeInt a) (SomeInt b) = a == b >same (SomeString a) (SomeString b) = a == b >same _ _ = False > >...later... > >same (SomeReal a) (SomeReal b) = a == b will become: same :: SomeObject -> SomeObject -> Bool same (SomeInt a) (SomeInt b) = a == b same (SomeString a) (SomeString b) = a == b ...later... same2 :: SomeObject2 -> SomeObject2 -> Bool same2 (SomeReal a) (SomeReal b) = a == b same2 x y = same x y tell me if I am wrong. thanks. Mansour Al-Mutairi :) _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

2 1