(Sorry, Rafael, the first e-mail I forgot to "reply all") On Thu, Jul 17, 2008 at 11:42 AM, Felipe Lessa wrote:

specilized to (->) we get

(***) :: (b -> c) -> (b -> c') -> (b -> (c, c'))

That should read (&&&) :: (b -> c) -> (b -> c') -> (b -> (c, c')) :) -- Felipe.

Marco Túlio Gontijo e Silva wrote:

Hi.

Em Qui, 2008-07-17 às 11:42 -0300, Felipe Lessa escreveu:

...

f = sum &&& foldr max 0 >>> uncurry (-)

You can replace

...

foldr max 0 with simply foldr1 max or even maximum

If it's important that the function return 0 when given an empty list, then you don't want to use foldr1 or maximum. But, if you don't need it to, then it may be sounder to have it fail on an empty list, because it's not particularly well-defined what it should return then. -Isaac

Is there any good tutorial on arrows? Thanks On Thu, Jul 17, 2008 at 13:08, Rafael Gustavo da Cunha Pereira Pinto < rafaelgcpp@gmail.com> wrote:

I will use foldr max 0, because I want 0 for the empty list.

On Thu, Jul 17, 2008 at 12:09, Isaac Dupree wrote:

...

Marco Túlio Gontijo e Silva wrote:

...

Hi.

Em Qui, 2008-07-17 às 11:42 -0300, Felipe Lessa escreveu:

...

f = sum &&& foldr max 0 >>> uncurry (-)

You can replace

...

foldr max 0

with simply

...

foldr1 max

or even

...

maximum

If it's important that the function return 0 when given an empty list, then you don't want to use foldr1 or maximum. But, if you don't need it to, then it may be sounder to have it fail on an empty list, because it's not particularly well-defined what it should return then.

-Isaac

_______________________________________________ Beginners mailing list Beginners@haskell.org http://www.haskell.org/mailman/listinfo/beginners

-- Rafael Gustavo da Cunha Pereira Pinto Electronic Engineer, MSc.

-- Rafael Gustavo da Cunha Pereira Pinto Electronic Engineer, MSc.

...

Is there any good tutorial on arrows? Ross Paterson. Arrows and Computation. http://www.soi.city.ac.uk/~ross/papers/fop.htmlhttp://www.soi.city.ac.uk/%7Eross/papers/fop.html

If you're new to Haskell, I'm guessing you're looking at the instance of Arrow for ordinary functions, so that you can write in the point-free style. instance Arrow (->) where arr f = f f >>> g = g . f first f = f *** id second f = id *** f (***) f g ~(x,y) = (f x, g y) If so, Ross Paterson's paper may scare you off. I'm not aware of any good tutorial for ordinary functions - maybe because using arrows this way is sort of a point-free programming gimmick. Anyway, here's my attempt at a quick tutorial of the Arrow combinators for ordinary functions. -- Pass input to one function, then its output to another (f >>> g) x == g (f x) -- Pass two inputs to two functions (f *** g) ~(x,y) == (f x, g y) -- Pass one input to both functions (f &&& g) x == (f x, g x) -- Apply function to only first of two inputs (first f) (x,y) == (f x, y) -- Apply function to only second of two inputs (second f) (x,y) == (x, f y) The pictures on this page are helpful: http://www.haskell.org/arrows/ Arrows are more interesting when composing more abstract things where it doesn't make sense to give the user access to inputs and outputs, such as modeling circuits where the inputs are wires. Lots of good examples in the paper apfelmus recommended. -Greg On Fri, Jul 18, 2008 at 1:33 AM, apfelmus wrote:

Rafael Gustavo da Cunha Pereira Pinto wrote:

...

Is there any good tutorial on arrows?

I found

Ross Paterson. Arrows and Computation. http://www.soi.city.ac.uk/~ross/papers/fop.htmlhttp://www.soi.city.ac.uk/%7Eross/papers/fop.html

to be pretty good.

Regards, apfelmus

_______________________________________________ Beginners mailing list Beginners@haskell.org http://www.haskell.org/mailman/listinfo/beginners