The data set described in Table 1 is optional. You must allocate it only if an application uses file-tailoring services.

File-tailoring output can be written to a temporary sequential data set provided by ISPF. The temporary data set is allocated automatically, so there is no need for the dialog to allocate a data set. The fully qualified name of the temporary data set is available in system variable ZTEMPF. The ddname is available in ZTEMPN. This temporary data set always uses 80-character fixed-length records.

If the temporary data set is not used, file-tailoring output can be written to either a partitioned or a sequential data set. Both fixed-length and variable-length records are permitted. The maximum logical record length is 255 bytes. A data set must be allocated to ddname ISPFILE before starting file-tailoring services. The dialog can dynamically allocate the output library, and can free it upon completion.

For a sequential data set, ISPFILE must be allocated with DISP=OLD. For a partitioned data set, ISPFILE can be allocated with DISP=SHR. ISPFILE must not specify a concatenated sequence of data sets.