How to use UNIX files in Mainframe Batch job

HFS files are designed to run in a UNIX environment, you can still access them through a batch job. The JCL you use is a little different, though, than for standard data sets. Zo/s provides some additional parameters for the DD statement that allow you to create, access, and delete HFS files.

A pathname must…

  • Begin with a slash.
  • Include all directory and subdirectory names that lead to the file itself.
  • Be from 1 to 255 characters long.
  • Be enclosed in single quotes if it contains any lowercase letters.
  • Be enclosed in single quotes if it is continued on more than one line.

A PATH parameter that’s used to access a master file in HFS format

//DD1 DD PATH='/usr/data1/r1/ab0020/MASTER.file'

A PATH parameter with a long file name

//DD1 DD PATH='/USR/DATA1/R1/CA0020/ACCOUNTING/REP
ORTS/FEB01/INVENTORY.RPT'

Do not code the PATH parameter on any of the following DD statements

  • JOBCAT
  • STEPCAT
  • SYSABEND
  • SYSUDUMP
  • JOBLIB
  • STEPLIB
  • SYSMDUMP

Additional points

  • The PATH parameter on a DD statement is used in place of a DSN or DSNAME parameter to identify a file in HFS format.
  • Symbolic parameters can be used within the PATH parameter.
  • Backward references to a DD statement with a PATH parameter are not allowed and are treated as errors. In some cases, you also need to include the PATHOPTS parameter in order to successfully access an HFS file.

Author: Srini

Experienced software developer. Skills in Development, Coding, Testing and Debugging. Good Data analytic skills (Data Warehousing and BI). Also skills in Mainframe.