On Sat, Jun 6, 2009 at 1:46 AM, bill lam<[email protected]> wrote:
> ... Suppose a b c are verbs and you define a train
> f=: a [ b ] c
>
> when J interpreter execute this sentence, it will parse it to a syntax
> tree that is an internal representation of textual 'a [ b ] c' It
> will not deduce that f=: a. Later when it evaulate x f y, it will
> execute using the stack as described in DOJ
I view this process differently.
When the J interpreter executes this sentence, it will
parse it to an internal representation, using the stack as
described in DOJ.
If I diagram this tree, it would look something like this:
[
/ \
a ]
/ \
b c
If I diagram this tree with x and y for left and right arguments,
it would look something like this:
[
/ \
/ \
a \
/ \ ]
x y / \
/ \
b c
/ \ / \
x y x y
Or, if used without a left argument, you could use that
last diagram by eliminating the x parts from it.
FYI,
--
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
