Raul wrote:
> Let us imagine you could pick an arbitrary word to
> mark the end of a strand.
This feature is already available in my script (which is generalized, so it
covers this & other cases). For example, to emulate your conjunction
using my script, you could write:
end =: s ]
] 'a' 2 'b' 3 'c' end
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
Here, the end-of-strand marker can be any noun or verb (or proverb, even
the uassigned kind):
end =: s begin
begin 'a' 2 'b' 3 'c' end
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
Note that the script I posted permits the user to select his own markers.
If he doesn't want to, I've included a utility the hard-codes the marker
to __ . This is fine so long as he doesn't want to include __ itself
in the strand:
__ 'a' 2 'b' 3 'c' t
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
I personally wouldn't use t , because I find balanced markers more
aesthetically pleasing:
| 'a' 2 'b' 3 'c' s|
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
# 'a' 2 'b' 3 'c' s#
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
[ 'a' 2 'b' 3 'c' fs]
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
< 'a' 2 'b' 3 'c' fs>
+-+-+-+-+-+
|a|2|b|3|c|
+-+-+-+-+-+
and I find it easy to ignore the wart of s (or fs ). Note that fs is
a little special; it permits balanced markers (which precludes use of
either marker within the strand).
> I am going to reject ideas like getting rid of the current
> definitions of tokens, because that does not work in
> any current version of J -- I hope you do not mind.
I am not sure what exactly you're rejecting, so I guess I'm ok with it by
default :)
Kidding aside, I hope it's clear I don't propose changing the J interpreter
in any way to support strand notation; my scripts work entirely within the
context of J's existing grammar and definition. I would support J
permitting Unicode identifiers, for many reasons -- the ability to
simulate extensions of J's syntax among them.
-Dan
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
