KICKS
A
transaction processing system for
TSO
What's New? - 2009
End of December 2009
- Implemented KSMT (master terminal) transaction
- Implemented major improvements to transaction dump including trace table
"translation"
- Implemented parallel xmit and iebcopy distribution/update packaging
- Tested KICKS in z/OS.
- Revamped these web pages, including live kicks sessions.
End of November 2009
- Posted V0R9M2 (bug fixes, trial of IEBCOPY distro for V1R0M0) to Yahoo
forum,
- Continued weekly KooKbooK recipes with articles on LINK/XCTL, AICA
abends, and BLL cells.
- Implemented "SIT" overrides.
End of October 2009
- Posted preview of MVS version of mapgen to Yahoo forum..
- Continued testing examples against real CICS.
- Initiated weekly KICKS KooKbooK
recipes, posted articles on bms mappnig and cobol programming.
- Implemented "spin off" of transaction dump sysouts.
- Decided to postpone development environment upgrade.
End of September 2009
- Not much feedback on V0R9M0 so assuming it's relatively good.
- Almost finished MVS version of mapgen, expect to post in a few days.
- Began work comparing KICKS examples to same pgms in real CICS. So far
still getting them compiled, need to finish that, setup testing scripts and
compare results.
End of August 2009
- Posted V0R9M0
- Created yahoo group
kicksfortheturnkey for registered KICKS users
- Posted list of known issues for V0R9M0
- Posted goals for V1R0M0 (December?)
- Bug fixes
- MVS based mapgen
- Test example programs on real CICS system, confirm similar operation
- On vacation
End of July 2009
- Nearly ready to post V0R9M0 of KICKS.
- Posted some draft material
End of June 2009
- Testing is going well. In addition to the file control issues I expected
to find I'm also finding bugs elsewhere in the code. For example I recently
had to spend a couple days tracking down what turned out to be a bug in
RECEIVE MAP, where the field length was being returned in the following
field's length. I thought that code was "good" 6 months ago. Oh Well...
End of May 2009
- Vsam file support (including file sharing) is finally "working" but my
testing to this point has been superficial, and this is complex enough it
will definitely need some real testing. So "working" doesn't mean "done".
- Got involved in an interesting thread on the tur(n)key forum and
demonstrated the possibility of connecting thousands of tn3270 sessions to hercules.
End of April 2009
- Still plugging away on vsam file support. Obviously I really
underestimated this task...
- Obtained cobol manuals via interlibrary loan. Bound, can't scan. And
less useful than I hoped.
- With CBT posting of SPLA stuff (http://www.cbttape.org/spla.htm)
the CICS share tape appeared.
End of March 2009
- Added page describing how to set up the Tur(n)key for
lots of 3270's.
- Finished converting the Cobol pre-processor from Perl to C.
- Wrote several Perl scripts to partially automate conversion of programs
written in Cobol2 (aka Cobol 85) to run with the Tur(n)key ANSI Cobol. Lots
of issues - "while", "inspect", implied length, "set", hex values, ... This
will never be a complete conversion (with my tools anyway) but at least it
relieves much of the manual effort involved.
- Fixed a bunch of pre-processor issues surfaced by another "example"
application system one of you sent me.
- Dealt with a variety of issues related to HANDLE CONDITION - mostly
trying to align KICKS abend, RESP, and RESP2 returns with what real CICS
returns. Also added IGNORE CONDITION and NOHANDLE.
- Installed and worked with
OpenCobol
a bit (on the PC) to help convert some application data. Pretty nice. May
tack porting it to GCCMVS on the end of my KICKS project (assuming I ever
get to the end of it) - would certainly help with running more modern COBOL
programs on the Tur(n)key.
- Color KSGM screen.
End of February 2009
- Added support for HANDLE AID and HANDLE CONDITION.
- Wrote several Perl scripts to automate conversion of programs written
for the VS Cobol compiler ( programs with COPY and Librarian includes ) to
the Tur(n)key ANSI Cobol. Somewhat problematic due to the way ANSI Cobol
does COPY's.
- Started converting Cobol pre-processor from Perl to C so it can deploy
on MVS as well as on the PC.
- Posted progress and messaged the general tur(n)key forum.
End of January 2009
- Put up this web page to share with
hercules-os380
( home of gccmvs ) folks
Copyright © Mike Noel, 2008-2010;
last updated 9/5/2010