In a test run, the database is not updated. Since the payroll results from the previous period are used as the basis for calculating the results of the following period, the results of the actual payroll run would differ from those of the test run, if this test run were executed over several periods.

The use of the buffer enables trouble-free access to the required results for the previous period.

What is required for exporting/importing data to/from the PCLn files using the buffer?

• The following includes contain the data definition for the buffer. They must be included in the report that writes the data to or reads the data from the database.

RPPPXD00

RPPPXD10

Include RPPPXM00, which contains the buffer management routines, is also required.

The macros for importing and exporting data must comply with the following naming convention:

RP-aaa-Cn-xy

where aaa = IMP / EXP, n=1 for PCL1, 2 for PCL2, 3 for PCL3, 4 or PCL4
and xy = cluster name.

This guarantees consistency between the export and import of data and also ensures that all exported objects are imported again.

When macros are used for exporting, records are written to a main memory buffer and not directly to the database. When the program run has been completed, the records in the buffer are stored in the appropriate PCLn database.

When the macros are used to import data, the data records are not read directly from file PCLn. Instead, the system checks the buffer directory to see whether the main memory already contains a record with the same key. If this is not the case, the record is read from PCLn to the buffer and then retrieved from the buffer for the report.