On 11/7/2013 12:41 PM, Richard Pinion wrote:
And to throw another twist to this thread, some people say the LRECL and RECFM should not be coded in the JCL. That way when a change is made to the program source, that affects LRECL and/or RECFM, the corresponding JCL doesn't have to be updated. What are some opinions about that methodology?
DCB parameters should never be required for input files. An output block size should be used only in special situations (e.g., transfer to another system). For output files, I prefer to leave the attributes to the DCB exit, allowing me to route print to a PDS without destroying it (and the program/subroutine handles padding). What I detest are programs where nothing is specified, and there are no defaults, requiring explicit DCB parameters on each statement.
I'm unhappy about the utilities, and sloppy CoBOL, forcing the DCB parameters thus requiring an extra step to copy the output someplace else with appropriate padding and carriage control conversion. I wish that IBM had provided a single output routine for the utilities, and made it generally available as a subroutine. (Ditto with writing messages)
Gerhard Postpischil Bradford, Vermont ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
