On Sun, 20 Aug 2006 15:57:04 -0300, Shmuel Metz (Seymour J.) <[EMAIL
PROTECTED]> wrote:
>
> >How? One can come close by coding a function package in some other
> >language, but if I understand correctly even that can't precisely
> >entirely copy a compound from one stem to another.
>
> Why not? You go through the list of variable names and return a list
> of the names beginning with the stem. The caller can then use the list
> of names to copy all of the compound variables.
>
No. I believe if I say:
X. = 'Cosmos'
Null = ''
X.Null = 'Chaos'
drop X.3.14159
that the EXECVARS interface will not return any indication of the
dropped status of X.3.14159, nor any distinction between the stem
with no tail and the stem with a tail which is the null string.
Symbol( 'X.' ) and symbol( 'X.Null' ) will correctly supply information
to resolve the latter ambiguity, but there's no way to identify members
of a stem dropped after a universal assignment.
Unfortunately, an optimization to Rexx causes incorrect behavior, making
it impossible to test this. IBM is working on an APAR for me, but taking
their time.
> Admittedly the equivalent in OREXX is cleaner, but it can be done.
>
Will the OREXX facility correctly identify a single member or several
members DROPped after a universal assignment?
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
