Hi there,
*totally new topic* (sorry).
For historical reasons, I did use a *lot of classic Rexx Stems*, in
their notation stem.index1.index2 ...
in my own soft, developed over years.
Most of them, but not all, are actually using *integer indices*, but
some of them, of course,
did use the power of classic rexx Stems, actually implementing
*Associative Arrays* ;-)
Now, the issue of this topic is as follows:
Many of those are so called (in Computer Science, at least) *sparse
arrays*, i.e. *most entries are
not filled with a content*, at all.
As classic Rexx had no way to define a *structure* (as PL/I does), the
only way to introduce
a collection of structures in *Classic Rexx* has been to introduce:
-- either some blank delimited wordlists, and handle alls those strucs
by the proper functions
-- *or* to implement them as STEMS, e.g.: item_name., item_class., ...
item_xxxx, as I did.
Now, the funny thing, when converting all of my soft to NetRexx is, as
follows:
I cannot *easily* change a *sequence* (set) of associated STEMS to a
Java COLLECTION,
as, unlike PL/I, NetRexx has the brackets for arrays, and parenthesis to
other means.
What I shall really want is to have a *very easy*, and human,
methodology to switch
between former SDtems (now called Indexed Strings), Arrays, and Sets
(part of Collections)!
What do you say? Except that I'm *Crazy Uncle Tom*, of course!
Thomas.
--
Thomas Schneider, IT Consulting;
http://www.thsitc.com; Vienna, Austria,
Europe
_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive :
http://ibm-netrexx.215625.n3.nabble.com/
Thomas Schneider, Vienna, Austria (Europe) :-)
www.thsitc.com
www.db-123.com