Gerhard, Thanks for the nice explanation.
Do you know if it is possible for a running job to read the values of PROC and/or SET JCL variables in effect for its (current) jobstep? It appears that those are all resolved/washed away in the system control blocks, but I wonder if they are lost after CI processing as well. On 10/25/07, Gerhard Postpischil <[EMAIL PROTECTED]> wrote: > > Bass, Walter W wrote: > > I would like to enhance this program to avoid to necessity of passing > > the symbolic values through the parms. I am certain I can successfully > > get the information I need by parsing the JESJCL. What I have not been > > able to find out so far is where the JESJCL data is? Is this in an MVS > > data area? Is it in a dataset allocated to the job? > > You don't say whether you're using JES2 or JES3. I've only > worked with JES2, for which the processing is as follows: JCL is > read by a reader (with an optional exit) and saved to the spool > in the JESJCL pseudo data set. In-line data cards are extracted > and written separately. The job is passed to the > Converter/Interpreter, where JCL is read, passed to the IEFUJV > exit, and processed into OS JobQueue format, passed through an > optional text exit (not intended for user access), and written > as another pseudo data set. When the job executes, one or more > times, the JobQueue records are read and converted to system > control blocks (JCT, ACT, SCTs, SIOTs, etc.) in virtual memory. > > If you wish to change the JCL, you need to do so in the JES > reader exit, or in IEFUJV. It is possible to read the JESJCL > data set at a later time (for code look at the Q command at > cbttape.org), but any changes you make will be ignored. The big > gotcha is the annoying little detail that the reader and > converter need not run on the system the job will execute on. To > make the changes at execution time, you could use the IEFUJI > (job initiation exit), and read and possibly alter the text > records in JobQueue format. > > If you look at the group archives, you'll find a slew of > messages regarding the resolution of symbolic parameters, which > seems to be what you're driving at. If you only have a single > system, the easiest place to make substitutions would in IEFUJV > or the JES reader exit, and forget about the JESJCL data set. > > > Gerhard Postpischil > Bradford, VT > > new e-mail address: gerhardp (at) charter (dot) net > > ---------------------------------------------------------------------- > 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 > ---------------------------------------------------------------------- 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
