To me, this is a very weird design. The programmer has some sort of "ad hoc" program which he runs for himself. Others have seen the output and like it. So they want to be able to do the same sort of "ad hoc" query. But, some of them being non-z people, they have _no_ concept of "records" of a specific length. They only think of "records" as in Windows text files. And so they apparently cannot get their minds around needing to make a "structured record" on Windows to upload to a "fixed length" file on the mainframe. The can get the concept of "put the policy number in column 10". So the programmer wants to make that the entire interface. Put anything else anywhere in the record and let it be of any length, but put the policy number starting in a specific column. The only other thing the user needs to specify is the mainframe data set name that they uploaded. This is "user friendly". But I'm old and am definitely "user hostile".
On Mon, Sep 9, 2013 at 12:26 PM, John Gilmore <[email protected]> wrote: > Do you need Peter Farley's experimental approach? You do if even the > job submitter will not know the input-file RECFM. > > If someone will know it, and, say, PARM='RECFM=FB|VB' can be supplied, > a simple COBOL program that dumps largely canned JCL into the internal > reader will solve your problem. The function of the internal reader > is to cope with generated JCL; and the generation required here is > trivial. > > John Gilmore, Ashland, MA 01721 - USA > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- As of next week, passwords will be entered in Morse code. Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
