Creating Dataviews for Sequential Files › Runtime Considerations for Sequential File Dataviews › CICS

CICS

Setting up the environment for writing to a sequential file from one or more CICS CA Ideal applications is described in this section. When more than one CICS user writes to the same sequential file, the records are interleaved. This makes it possible, for example, to log records from various users to a common log file.

Note: In CICS, CA Ideal can write to sequential files but cannot read sequential files. Until CICS closes a sequential file, it cannot be read in batch.

The steps for defining the sequential file are as follows:

1. Define the sequential file and a corresponding dataview and define it as a resource for the CA Ideal programs that uses it (see Defining Modeled Sequential Files and Defining Unmodeled Sequential File Dataviews earlier in this chapter).

   For dataviews modeled in the dictionary, make sure that the ACCESS‑METHOD is defined as QSAM in the FILE entity‑type, and that the MONITOR‑NAME has a value in the first four characters, which corresponds to a CICS DCT DESTID. Make sure that the DATAVIEW has an UPDATE‑INTENT of Y (yes).

   For unmodeled dataviews, make sure that in the dataview parameter fill‑in the access method is defined as QSAM and that the filename has a value in the first four characters that corresponds to a CICS DCT DESTID. For example, a file called PAYFILE is defined with a MONITOR‑NAME of PAYR and a corresponding DATAVIEW is defined with a name of PAYLOG. The dataview name PAYLOG is used in the CA Ideal program to add records to it as follows:

   FOR NEW PAYLOG
   MOVE ... TO PAYLOG.field
   ENDFOR
   

2. Define the file to CICS as a Transient Data Queue using RDO.

   For a dataview modeled in the dictionary, the queue name must match the MONITOR-NAME attribute, and for unmodeled dataviews, it must match the first four characters of the filename.

You must add a DD statement (z/OS) or DLBL or TLBL statements (VSE) for the file to the CICS startup deck.

z/OS Batch

Add a DD statement to the CA Ideal batch JCL for the file using the specified MONITOR‑NAME or filename.

The z/OS IDSYSFT contains 16 entries of the form SYSUT01 through SYSUT16. CA Ideal uses these entries for sequential files, actually modifying the file name dynamically. For this reason, no changes to the z/OS IDSYSFT (batch file table) are required.

The maximum number of active sequential files at any point in time is 15.

VSE Batch

For a standard label tape file, add a TLBL statement to the batch CA Ideal JCL using the specified MONITOR‑NAME or filename. For a disk file, add a DLBL statement using the specified MONITOR‑NAME or filename. Otherwise, no JCL changes are required.

Changes might be required to the VSE IDSYSFT (batch file table). Five types of sequential file dataviews are supported under VSE: Disk, standard‑label tape, non‑labeled tape, printer, and punch. One entry exists in the CA Ideal batch file table (IDSYSFT) for each type as shown in the following table:

You must add more entries to the file table if a program is run in batch with more than one of the same type of file. To add additional entries, change the name by incrementing the digit at the end of the file name (from 1 to G) without leaving gaps. Each type has its own set of one or more ROSFD entries.

The following are the installed ROSFD entries for sequential file dataviews:

 IDISK1 ROSFD ACCMETH=SEQ,   for disk        X
   DTFTYPE=DTFSD,                   X
   DTFNAME=IDISK1, overridden at open        X
   DEVADDR=SYS000, overridden in JCL         X
   RECFM=FB,                     X
   DEVICE=nnnn,   installation option        X
   LRECL=4096,   overridden at open        X
   IBLKSZ=4096,   max overridden at open      X
   OBLKSZ=4104,   max overridden at open      X
   OPSYS=VSE
 ISLTAPE1 ROSFD ACCMETH=SEQ, for standard‑label tape  X
   DTFTYPE=DTFMT,                   X
   DTFNAME=SLTAPE, overridden at open        X
   DEVADDR=SYSnnn, opt overridden at open      X
   FILABL=STD,                    X
   RECFM=FB,                     X
   LRECL=4096,   overridden at open        X
   BLKSIZE=4096,  max overridden at open      X
   OPSYS=VSE
 INLTAPE1 ROSFD ACCMETH=SEQ, for no‑label tape     X
   DTFTYPE=DTFMT,                   X
   DTFNAME=NLTAPE, ignored              X
   DEVADDR=SYSnnn, opt overridden at open      X
   FILABL=NO,                     X
   RECFM=FB,                     X
   LRECL=4096,   overridden at open        X
   BLKSIZE=4096,  max overridden at open      X
   OPSYS=VSE

 IPRT1 ROSFD ACCMETH=SEQ,    for a printer     X
   DTFTYPE=DTFPR,                   X
   DEVICE=1403,                    X
   DEVADDR=SYSnnn, opt overridden at open      X
   CTLCHR=ASA,                    X
   RECFM=F,                      X
   BLKSIZE=133,   max overridden at open      X
   OPSYS=VSE
 IPUNCH1 ROSFD ACCMETH=SEQ,   for a punch      X
   DTFTYPE=DTFCDP,                  X
   DEVICE=2540,                    X
   DEVADDR=SYSPCH, opt overridden at open      X
   RECFM=F,                      X
   BLKSIZE=80,   max overridden at open      X
   OPSYS=VSE

Removing entities from the Batch CA Ideal File Table that your site never uses saves memory use. These entries provide an outline of the information needed at run time.

Before the Batch CA Ideal File Table is opened, additional or overriding information is stored in the DTF. Information can be extracted from the dictionary or the dataview definition. You can override some of this information by using the ALTER PROGRAM and ASSIGN DATAVIEW commands described in the Command Reference Guide. This information, with information from the CA Ideal File Table entry, provides the information needed to access the file.

Device Type

Indicates whether a sequential file dataview is disk (DISK), standard‑label tape (SLTAPE), non‑labeled tape (NLTAPE), printer (PRT), or punch (PUNCH).

For dataviews modeled in the dictionary, this is obtained from the dictionary when the dataview is cataloged.

For unmodeled dataviews, this is obtained from the dataview parameters definition.

At run time, the device type specifies to CA Ideal what type of file table entry to use: DISK, SLTAPE, NLTAPE, PRT, or PUNCH.

You can override this entry with an ALTER PROGRAM or ASSIGN DATAVIEW command before the run. For a description of the commands, see the Command Reference Guide.

Record Size

Indicates logical record length.

For dataviews modeled in the dictionary, this information is obtained from the dictionary when the dataview is cataloged.

For unmodeled dataviews, this information is obtained from the dataview parameters definition.

It is stored into the ROSFD entry before the file is opened.

The Administration Guide describes how to update the IDSYSFT.

Block Size

Indicates the physical block size.

For dataviews modeled in the dictionary, this information is obtained from the dictionary when the dataview is cataloged.

For unmodeled dataviews, this information is obtained from the dataview parameters definition.

Filename

Indicates the DLBL name for disk files or the DESTID for use in CICS.

For modeled dataviews, this information comes from the MONITOR‑NAME attribute of the dictionary FILE entity at the time a dataview is cataloged.

For unmodeled dataviews, this information is obtained from the dataview parameters definition.

Logical Unit Assignment

For standard‑label tape, non‑labeled tape, printer, and punch files, the default assignment comes from the CA Ideal batch file table (IDSYSFT) entry. For more information, see the Administration Guide.

You can override the logical unit assignment with an ASSIGN DATAVIEW command before the RUN command or an ASSIGN statement executed before the first FOR construct that references the dataview. For more information, see the Programming Reference Guide.

The JCL does the complete logical assignments for disk files.