Five types of sequential file dataviews are supported under VSE‑disk, standard label tape, non‑labeled tape, printer, and punch. Each type has its own set of one or more entries in the CA Ideal File Table. (See the Administration Guide for more details on file table entries IDSYSFT).
Before a program using a sequential dataview can be successfully compiled, you must enter the definition for the dataview, either in Datadictionary (a modeled dataview) or in CA Ideal (an unmodeled dataview) and then cataloged in CA Ideal. The catalog step puts the device type, record size, block size, file name and whether the file is labeled into the dataview object code used at compile time.
The display of the cataloged dataview shows the information needed to code the batch JCL. You can override some of this information with the ALTER PROGRAM and ASSIGN DATAVIEW commands. For dataview definition and for ALTER and ASSIGN commands, see the Administration Guide. With the file table entry, this provides the information needed to access the file.
Information as to whether a sequential dataview is disk, standard label tape, non‑labeled tape, printer, or punch. You can override this option with an ALTER PROGRAM or ASSIGN DATAVIEW command before the run. At run time, the device type specifies to CA Ideal what type of file table entry to use‑DISK, SLTAPE, NLTAPE, PRT, or PUNCH.
The DLBL for disk files and TLBL for standard labeled tape files.
For standard label tape, non‑labeled tape, printer, and punch files, the default assignment comes from the file table entry. You can override it with an ASSIGN DATAVIEW command before the RUN command or an ASSIGN statement executed before the first FOR construct referencing the dataview. The logical assignments for disk files are made completely through the JCL.
Under VSE, two additional dataview status codes are used for error conditions. I7 is used if the block size specified for the file is larger than the maximum defined in the VPE File Table entry. I8 is used if multiple sequential files of the same type are referenced in the same run and there are not enough file table entries defined for that type. For more information about how to define the file table to handle these conditions, see the Administration Guide.
If multiple tape, print, or punch files are referenced from the same run, there can be some ambiguity as to the logical unit assignment that is used. This is because multiple file table entries of the same type are allocated on a first come, first serve basis. That is, the first dataview of a type is allocated the first file table entry for that type; the second file of the same type gets the second, and so on. You can override the logical unit assignment defined in the file table with an ASSIGN COMMAND or ASSIGN statement.
Consequently, for runs that access multiple dataviews of the same type, assign the logical unit of each dataview with either an ASSIGN command before the run or an ASSIGN statement as the first thing done in the program. This prevents any JCL mix‑ups.
This is not required for dataviews where there is only one of its type in the run; the logical unit used is the one for the first file table entry of that type. This is not required for disk dataviews since logical unit assignments are completely handled in the JCL.
For logical unit assignments defined in the file table, see the CA Ideal Administrator Guide.
You must add the following to batch run JCL to reference a sequential file dataview.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|