If you know the dataset name structure, here's how I would break it down: In option 3.4 enter the datasets wild card name, such as
PEOPLE.THAT.**.NEED.SOME.HELP Save the list that is produced. It will be in your SPF LIST dataset. So PF3 until the option appears KEEP, ALLOCATE NEW. Note the LIST dataset name (SPF4.LIST or whatever). Back to ISPF, edit the list dataset to get a list of dataset names . Write yourself a JCL PROC that has all the bits you need and have DSNIN as a symbolic. Now write a REXX program that reads the list and puts out the JCL steps. You should end up with something like. //SNNN EXEC MYPROC,DSNIN=file from list Your REXX can count up to 250 and add a job card after each 250 steps. On Fri, Oct 9, 2020 at 1:40 PM Joseph Reichman <[email protected]> wrote: > Oh please I have enough problems > > > > > On Oct 8, 2020, at 10:34 PM, Clark Morris <[email protected]> wrote: > > > > [Default] On 8 Oct 2020 17:42:10 -0700, in bit.listserv.ibm-main > > [email protected] (Joseph Reichman) wrote: > > > >> I have no idea who the sysorog is > >> > >> Honestly I’m told to do something I try my best to accomplish my > superiors just tell me to ask different people > >> > >> I have come to the point no one wants to help so I take the attitude > I’ll do what ever tell me what ever happens happens s > >> > > This task should be simple and require no authorized code. COBOL, PL1, > > and C/C== have the capability of writing JCL by reading the catalog > > (at worst an IDCAMS listing) emitting the appropriate JOB and EXEC > > images and using the data set names to create the input DD cards. The > > actual selection and extraction program should be written in the > > higher level (COBOL, PL1, C/C++) language for which record > > descriptions for the data on the files exist (I would hope in 2020 > > that it isn't assembler only). Having written programs to deal with > > the SMF30 records with variable position keys and extract CALL and > > COPY information from COBOL and DYL280 (now Vision Results) programs, > > I find it hard to believe that you are faced with a difficult task. If > > the language is COBOL, I have written code in COBOL to handle bit > > switches (snitched from a posting on comp.software.year-2000) and > > reference modification is simple. > > > > Clark Morris > > > > > > Clark Morris > >> > >>>> On Oct 8, 2020, at 8:36 PM, Jeremy Nicoll < > [email protected]> wrote: > >>> > >>> ?On Thu, 8 Oct 2020, at 22:17, Joseph Reichman wrote: > >>>> There are number if restrictions in size of concatenation > >>> > >>> Yes, but suppose you can concatenate n datasets per dd? > >>> > >>> You read the list of datasetnames and then, n at a time, > >>> generate the right JCL. > >>> > >>> The point is, whatever the restrictions are, it cannot be hard > >>> to do it. > >>> > >>> > >>>> in addition there is CPU time step limit > >>> > >>> I don't understand that. > >>> > >>> If you have a legitimate business need to process all this data your > >>> systems programmers should provide you with a suitable job class. > >>> > >>> Indeed didn't someone else who IS an IRS sysprog reply a bit earlier > >>> saying that such classes DO exist? > >>> > >>> > >>>> I not working for a software co where I can go APF authorized and do > >>>> what ever I want > >>> > >>> Things may have changed but I don't recall APF authorisation having > >>> anything to do with time limits. > >>> > >>> Speak to your sysprogs. > >>> > >>> > >>> > >>> -- > >>> Jeremy Nicoll - my opinions are my own. > >>> > >>> ---------------------------------------------------------------------- > >>> For IBM-MAIN subscribe / signoff / archive access instructions, > >>> send email to [email protected] with the message: INFO IBM-MAIN > >> > >> ---------------------------------------------------------------------- > >> For IBM-MAIN subscribe / signoff / archive access instructions, > >> send email to [email protected] with the message: INFO IBM-MAIN > > > > ---------------------------------------------------------------------- > > For IBM-MAIN subscribe / signoff / archive access instructions, > > send email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- Wayne V. Bickerdike ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
