In a recent note, Steve Grimes said:
> Date: Fri, 30 Sep 2005 13:23:22 -0500
>
> Here's the doc from the free one (member $$DOCJCL). It's called JCLSCAN.
> The CBT references are at the bottom.
>
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> DOCUMENTATION FOR JCLSCAN TSO COMMAND PROCESSOR
>
> This highly convenient program calls MVS's JCL converter
> to report JCL errors in a jobstream on your TSO tube. It is the
> same program which checks the errors when you run a real jobstream
> with TYPRUN=SCAN. The output has been massaged to make it more
> convenient to display on the terminal. All syntax errors are found,
>
... which is sheer nonsense. The following DD statement passes
TYPRUN=SCAN with no exception reported:
4 //SYSUT1 DD DISP=(MOD,DELETE),
// DSN=X2345678.X2345678.X2345678.X2345678.X2345678.X2345678
... whereas with the TYPRUN omitted it properly gets:
13.33.35 JOB08407 IEF452I SCAN - JOB NOT RUN - JCL ERROR
13.33.35 JOB08407 $HASP396 SCAN TERMINATED
...
4 //SYSUT1 DD DISP=(MOD,DELETE),
// DSN=X2345678.X2345678.X2345678.X2345678.X2345678.X2345678
STMT NO. MESSAGE
4 IEF642I EXCESSIVE PARAMETER LENGTH IN THE DSNAME FIELD
Clearly, since the IEF452I message says the job was not run,
this is a syntax error, not a semantic error, but TYPRUN=SCAN
fails to report it.
TYPRUN=SCAN is worse than worthless; it's a WOMBAT for anyone
who relies on its output. A sensibly designed SCAN would process
up to the point at which the decision is made to issue or not
issue the IEF452I message, and bypass execution if SCAN is
specified and IEF452I is not issued.
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
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
