[Haskell-beginners] The Missing Arrow Function Strikes Back

[Brent Yorgey]

Ah. Well, you can implement split and unsplit just as they are written
on that page.  You can even make split nicer by implementing it as

  split = id &&& id

which avoids the use of 'arr'.  However, I don't really see much
practical point to 'split' (though it is nice theoretically).
Usually, when you split something you follow it up by applying more
arrows to both components of the tuple.  But in that case you might as
well just write (f &&& g) in the first place, instead of split >>> (f
*** g).

-Brent

On Wed, Jun 05, 2013 at 10:46:10PM +0800, Adrian May wrote:
> Well I just read this:
> 
> http://www.haskell.org/haskellwiki/Arrow_tutorial
> 
> Adrian.
> 
> 
> 
> On 5 June 2013 21:58, Brent Yorgey <byorgey at seas.upenn.edu> wrote:
> 
> > What are the types of 'split' and 'unsplit'?  It is hard to guess what
> > you want just from their names.
> >
> > -Brent
> >
> > On Wed, Jun 05, 2013 at 01:02:40PM +0800, Adrian May wrote:
> > > Thanks Ertugrul. In the meantime I noticed that split and unsplit are
> > also
> > > missing. Is there a similar replacement for them?
> > >
> > > Adrian.
> > >  On 5 Jun 2013 12:57, "Ertugrul Söylemez" <es at ertes.de> wrote:
> > >
> > > > Adrian May <adrian.alexander.may at gmail.com> wrote:
> > > >
> > > > > I just banged up against this problem:
> > > > >
> > > > >
> > > >
> > http://haskell.1045720.n5.nabble.com/The-case-of-the-missing-Arrow-function-td3125388.html
> > > > >
> > > > > Was liftA2 (not the applicative one) a bad idea, or is there another
> > > > > way to do it, or what?
> > > >
> > > > That liftA2 (let me call it liftA2') likely has this type signature:
> > > >
> > > >     liftA2' :: (Arrow cat)
> > > >                => (b -> c -> d)
> > > >                -> cat a b
> > > >                -> cat a c
> > > >                -> cat a d
> > > >
> > > > Does this sound familiar?  You can write this function in terms of the
> > > > arrow combinators:
> > > >
> > > >     liftA2' f c d = arr (uncurry f) . (c &&& d)
> > > >
> > > > However, if your arrow is also a family of applicative functors
> > > > (i.e. pretty much always),
> > > >
> > > >     instance Applicative (MyArrow a)
> > > >
> > > > then it's probably a bad idea, because you really want to use the
> > > > cleaner liftA2 instead:
> > > >
> > > >     liftA2 :: (Applicative f)
> > > >               => (a -> b -> c)
> > > >               -> f a
> > > >               -> f b
> > > >               -> f c
> > > >
> > > >
> > > > Greets,
> > > > Ertugrul
> > > >
> > > > --
> > > > Not to be or to be and (not to be or to be and (not to be or to be and
> > > > (not to be or to be and ... that is the list monad.
> > > >
> > > > _______________________________________________
> > > > Beginners mailing list
> > > > Beginners at haskell.org
> > > > http://www.haskell.org/mailman/listinfo/beginners
> > > >
> > > >
> >
> > > _______________________________________________
> > > Beginners mailing list
> > > Beginners at haskell.org
> > > http://www.haskell.org/mailman/listinfo/beginners
> >
> >
> > _______________________________________________
> > Beginners mailing list
> > Beginners at haskell.org
> > http://www.haskell.org/mailman/listinfo/beginners
> >

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