The answer on how the aliases are maintained is unknown but this is what the
man page says (below). I did test it and using -X and -I works in both
directions. There are no additional files in the USS directory so perhaps they
are including the z/OS PDS directory info along with the load module in the USS
file so that they can 'restore' it properly since there is meaningful info in
the directory which would have the alias info, among other things.
What it doesn't do is tag the file in USS as binary (imho it should).
======================================================================
-X
Specifies that the data to be copied is an executable. Cannot be used
with -F, -T, or -B.
-I (UNIX to MVS only)
When the specified file has an alias and the file is copied from UNIX
to MVS, the alias information is also copied.
Restriction: The -I option can only be used with the -X option.
Attention: If the -I option is specified when the data set that is
copied from MVS to z/OS UNIX does not have an alias, that option is
ignored.
=======================================================================
Lionel B. Dyck <sdg><
Website: https://www.lbdsoftware.com
"Worry more about your character than your reputation. Character is what you
are, reputation merely what others think you are." - John Wooden
-----Original Message-----
From: IBM Mainframe Discussion List <[email protected]> On Behalf Of
Paul Gilmartin
Sent: Thursday, May 7, 2020 11:19 AM
To: [email protected]
Subject: Re: OMVS cp -X and -I ???
On Thu, 7 May 2020 09:25:55 -0500, Lionel B Dyck wrote:
>With z/OS 2.2/2.3/2.4 I have the -X option which supports the use of cp
>for load modules (executables) between z/OS and USS and back. I only
>have the -I (support to retain executable aliases) on the 2.4 system.
>
How is that implemented, given that such aliases are utterly alien to UNIX?
Do they actually work? The UNIX practice has long been to switch based on
argv[0].
>Does anyone know if the -I option has been made available via PTF for
>2.2 and 2.3?
>
>And - does anyone know if OGET/OPUT will be updated with these options?
>
It's regrettable that:
o cp -r doesn't deal with PDS(E)s.
o IEBCOPY doesn't deal with zFS.
o pax doesn't deal with PDS(E)s
All in all, too many incompatible commands to deal with otherwise similar
things.
Long ago, on MVS-OE, WJS mused about making the MVS catalog hierarchy a VFS.
(He was provoked.)
Imagine, even, a filesystem containing program objects, NFS-mounted from a z/OS
PDSE.
Imagine having a zFS directory in STEPLIB, even as it works in SYSLIB.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to
[email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
