On Tue, Jul 12, 2011 at 11:52 AM, David Vaughan
<[email protected]> wrote:
> You misunderstood Raul, the second line is simply "in C" the programming
> language. I just meant I wanted the first line's equivalent in J.
Ok... except that's not valid C, it's a fragment of C. And, worse,
it's a fragment which has a purpose which relates to C's structure
where best practice in J is to use different kinds of structures.
For example,
A=:1 2 3 2 1
B=: 3 2 3 2 3
N=:2 3 5 7 11
A ~: B
1 0 0 0 1
I would argue that this serves the same role in J that the if
statement served in C. It performs the stated comparison. But your
dangling "if" statement suggested that something else was intended.
But what?
Let's pick out some values from N where the condition is true:
N * A ~: B
2 0 0 0 11
Or let's say that we do not want to maintain N's structure, we just
want the selected values:
N#~A~:B
2 11
Or let's say that we want to negate values in N where A does not match B:
N * _1 ^ A ~: B
_2 3 5 7 _11
... and ... so on...
The point I am trying to make is that you can do a lot of things -- in
essence, any J expression which allows for numeric values might be
appropriate. But that's a lot of ground to cover.
Meanwhile, in C, you have micro management control over what happens
after an if. And while you can do that in J
(http://www.jsoftware.com/help/dictionary/ctrl.htm) we often have
better alternatives. You can construct arbitrary problems where there
is no better alternative, but you can also often change problems to
allow for better alternatives.
--
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
