Note that types can help. Eg. on http://www.haskell.org/hoogle/ you would
search for something like "[a] -> (a, [a])".  That is, you would be
searching by semantics and not by syntax.

Cheers,

Mike


On Fri, Mar 28, 2014 at 6:24 AM, Raul Miller <[email protected]> wrote:

> Note however:
>
>    split ;:'this is an example'
> ┌──────┬───────────────┐
> │┌────┐│┌──┬──┬───────┐│
> ││this│││is│an│example││
> │└────┘│└──┴──┴───────┘│
> └──────┴───────────────┘
>    13 :' ({.y);}.y' ;:'this is an example'
> ┌──────┬──┬──┬───────┐
> │┌────┐│is│an│example│
> ││this││  │  │       │
> │└────┘│  │  │       │
> └──────┴──┴──┴───────┘
>
> Do you see the difference?
>
> Do you see why the difference occurs?
>
> Hint:
> http://www.jsoftware.com/help/dictionary/d330.htm
>
> (search for the word "boxed")
>
> Thanks
>
> --
> Raul
>
>
> On Fri, Mar 28, 2014 at 6:32 AM, Linda Alvord <[email protected]
> >wrote:
>
> > Nice example.
> >
> >    A=:'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
> >    f=: 13 :'  ({.y);}.y'
> >    f A
> > ┌─┬─────────────────────────┐
> > │A│BCDEFGHIJKLMNOPQRSTUVWXYZ│
> > └─┴─────────────────────────┘
> >    7 f A
> > ┌───────┬───────────────────┐
> > │ABCDEFG│HIJKLMNOPQRSTUVWXYZ│
> > └───────┴───────────────────┘
> >
> >    f
> > {. ; }.
> >
> > -----Original Message-----
> > From: [email protected] [mailto:
> > [email protected]] On Behalf Of Raul Miller
> > Sent: Thursday, March 27, 2014 11:46 PM
> > To: Programming forum
> > Subject: [Jprogramming] lost art
> >
> > I am continually struck by how difficult it is to find relevant
> > information.
> >
> > Once upon a time, I had hopes that search engines would help resolve such
> > issues, but more and more I'm seeing their limitations (and the
> limitations
> > of the groups of people running them). And, similarly, for educational
> > institutions. And, similarly, for books. Each of these have limitations -
> > working together they can achieve a lot, but ultimately it's up to the
> > individual to have the drive and interest to piece things together.
> >
> > Anyways, here's an example: split
> >
> >    split 1 2 3
> > ┌─┬───┐
> > │1│2 3│
> > └─┴───┘
> >
> > In other words: given a list, return the pair (first; rest). Or, more
> > precisely:
> >
> > {. ,&< }.
> >
> > This is kind of a fun concept (though, frankly, it's often better to work
> > with flat data - trying to introduce habitual abstractions often gets in
> > the way of getting to where you want to go).
> >
> > Still, I remember times when I wanted something like split and I could
> not
> > think of how to find it. So I had to reinvent it, or something similar.
> I'm
> > not sure why this bothers me, but it does.
> >
> > Also, this is a nice example of monadic and dyadic definitions having a
> > nice simularity:
> >
> >    4 split 'abcdefghi'
> > ┌────┬─────┐
> > │abcd│efghi│
> > └────┴─────┘
> >
> > It's just a minor thing, but it's still food for thought.
> >
> > Thanks,
> >
> > --
> > Raul
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to