I did now some conversations with some of you, and would *like*
to repeat:
The current behaviour of the reference compiler for NetRexx is incorrect.
Let me tell you the shortest example I can make:
class Scan USES ReyMsg, ReyFile
method abc
ix=ic1 -- where ic1 is a public property of ReyMsg,
-- and ic1 is, thru the USES clause, actually
-- takenn from the class ReyMsg ... ok, so far
ic1 = ic1 +14
Here, I get my problem (not a real problem, as I did id already resolve
in my code)
Let me repeat it:
ic1= ic1 +14
The right hand side of the usage of 'ic1' doese refer to ReyMsg,
where ic1 is a public property of, (correctly, as specified=
but the left hand side of the assignment does introduce a *new*
local variable !!!
Thus, we have the interesting point that the *same variable*
on the left hand side and the right hand side of the same statement
do have a *different* meaning ...
I'm, while re-re-reading the NetRexx 2 language definition,
still thinking it is a *BUG* in the reference implementation....
At least when looking at the defnition of a so called *NetRexx TERM*.
greetings from Vienna,
Tom.
===========================================================
--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Tom. (ths@db-123.com)