On 18 Aug 2006 05:16:53 -0700, in bit.listserv.ibm-main you wrote: >In <[EMAIL PROTECTED]>, on 08/15/2006 > at 09:18 PM, Clark F Morris <[EMAIL PROTECTED]> said: > >>While I like the idea, plan on major effort to convert JES exits, >>IEFUJI and IEFUJV among other things. I would prefer we start >>looking a command language be it REXX, one of the Unix shells or >>something else that would allow longer data set names, mixed case, >>less convoluted syntax etc. Obviously JCL will have to be supported >>for the next decade but the limitations associated with it are >>probably so deep in the guts of z/OS that we would bleed if they were >>relaxed. > >The deeply buried limitations of JCL are not in things like case and >record length, they are in things like field sizes. Handling long >record, RECFM=VB and mixed case would be a piece pf cake; allowing, >e.g., DSN longer than 44 bytes would be a nightmare. > And these things are passed from JCL to all of the necessary components of z/OS. However, case folding or whatever solution is found for allowing entry of lower case has to be accepted and handled by all recipients of the data depending on the solution. If it is handled by the JCL interpreters, there are fewer places but still the number of places could be non-trivial. Record length also might impact in strange places.
---------------------------------------------------------------------- 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
