Hello Mike & al,
A) I do notice that the following classic REXX functions are missing (not yet implemented ?) in NetRexx:
============================================================================
Bit string handling: BITAND BITOR BITXOR
file handling: CHARS, CHAROUT, LINES, LINEIN, LINEOUT, STREAM
interpreter: ERRORTEXT SOURCELINE SYMBOL VALUE TRACE(as function)
stack i/o: QUEUED
date & time: DATE TIME
general: DIGITS FORM RANDOM
===========================================================================
B) also, the following Netrexx functions are new as compared to classic REXX:
string handling: CHANGESTR COUNTSTREXISTS SEQUENCE SPACE UPPER
===========================================================================
Now my questions in sequence of importance:
1.) is something like SYMBOL and VALUE available in Netrexx ? at least for Rexx objcts ?
2.) how should we emulate 'commands' ? My proposal would be to 'call EXEC command parameters' , where the EXEC routine
executes the given command under the underlying operating system and sets the RC (return-code) variable.
Alternativley, I can generate rc=EXEC(command parameters): what would you prefer ?
-- Mike, as far as I have read the NetRexx manual, my understanding is that 'expressions' on instruction level are NOT
thrown to the underlying processor (like CMS or XEDIT) in NetRExx! Or did I miss something ?
3 .) has someone already written some of the routines mentioned above under A) in NetRexx ? If yes, could you please make the code available to me ? If not, I will have to write a meaningful emulation, especially for the basic file I/O routines which I heavily use
in my own Rexx routines.
4.) Ibm's OBJECT REXX does allow both notations for 'string' functions -- as functions and as method reference.
-- Mike, is my understanding correct that NetRexx does support only the method reference, i.e.
string.center(length,pad) (NetRexx) instead of center(string,length,pad) (classic Rexx and Object Rexx)
===========================================================================
Now the good news: work is progressing: the scanner is already adapted for REXX, and I am currently working on the parser
again. Actually, the scanner/parser are driven by a SYNTAX definition and a BUILTIN functions definiton file, hence work
progresses quite good. The basic standards (parse id's,id-lists,expressions, conditions, quoted literals, tokenizing,
and more complicated things) are hard coded, but the rest is syntax driven.
thanks in advance, Tom.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To unsubscribe from this mailing list ( ibm-netrexx ), please send a note to
[hidden email]
with the following message in the body of the note
unsubscribe ibm-netrexx <e-mail address>
Locked
