Hi all,

Many of you sent me mail about the problems related to the CI echoing to
$stdlist commands that are essentially being skipped due to a
previous expression evaluating false.  The most common problem is
false alarms where operators reading JCL output think something bad
happened even though the 'alarm' message appears in a block of skipped
commands.

I have made an enhancement to the HPMSGFENCE variable that I would like
a quick review on.  I've split up HPMSGFENCE into 3 bit fields and each
of these fields acts like a msg "fence".  Here are the bits:

  bits 29..31   - controls CI error/warn msgs, like today:
                  0 = show all CI error and warn messages
                  1 = show only CI error msgs (suppress warnings msgs)
                  2 = suppress CI error and warn messages.

  bits 26..28   - new.  controls echoing of "skipped" cmds and the "*** RESUME
                  EXECUTION..." or "*** COMMANDS SKIPPED UNTIL..." msgs:
                  0 = show all skipped cmds and show the above "***" msgs
                  1 = don't echo skipped cmds, but show the "***" msgs
                  2 = don't echo skipped cmds nor the "***" msgs

  bits 0..25    - reserved (enforced).


This gives the following HPMSGFENCE values:

  0    - show "***" msgs, skipped cmds, CI warns, CI error msgs
  1    - show "***" msgs, skipped cmds, CI error msgs, no warns
  2    - show "***" msgs, skipped cmds, no errors/warns
  8    - show "***" msgs, no skipped cmds. show CI warns/errors
  9    - show "***" msgs, no skipped cmds, show CI errors, no warns
  10   - show "***" msgs, no skipped cmds, no errors/warns
  16   - no "***" msgs, no skipped cmds, show CI errors/wanrs
  17   - no "***" msgs, no skipped cmds, no warns, show errors
  18   - no "***" msgs, no skipped cmds, no warns, no errors

Any other values for hpmsgfence are an error.
New fields can be added later as needed.

I have coded and tested the result and it works fine.  It is backwards
compatible.  hpcmdtrace overrides hpmsgfence cmd echo suppression, that
is, hpcmdtrace has precedence over hpmsgfence (like today).  hpmsgfence
overrides option list, meaning an option list script with hpmsgfence =
8..10,16..18 will suppress the echoing to stdlist of all 'skipped' cmds.
Of course option nolist works as today.

I believe that with Gopi's help we can get this submitted to the next
main release.  We ***may*** also get this into the about-to-be-released
LISTF,access patch... I don't know yet...

Please let me know if this has been designed/implementd to your statisfaction.

regards,
Jeff Vance, CSY


--