On Jan 13, 2008, at 9:34 PM, Andy Wood wrote:
On Sun, 13 Jan 2008 18:27:56 -0600, Paul Gilmartin
<[EMAIL PROTECTED]> wrote:
. . .
Name/token sounds like a better choice, even though it's innovative.
The question of where to stash something arose in this thread when
it was
suggested that storage could be acquired once, and not every time the
function was called. In that context name/token has a complication
- the last
time I looked, simply calling the name/token services needed some
writeable
storage to create the parameter list, which is a bit tricky if you
don't yet
know where your working storage is.
I can think of ways to do it, but none that I would be really happy
with.
Not that that makes TCBUSER a better choice. I think I would stick
to simply
obtaining and releasing the storage each time the function was called.
Andy,
It might be in a backup plan to use tcbuser field but be aware that
you need APF authorization to alter it and anytime you muck around in
the area you need to be dead on correct with what you alter. Myself I
would think APF authorization would be a nono unless there is *NO*
other way to do what is being done. That probably means a user SVC
and all that entails (think security issues).
Ed
----------------------------------------------------------------------
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
