Assume that a join logical file is specified as in Basic concepts of joining two physical files (example 1), and there are four records in PF1 and five records in PF2, as shown in the following two tables.
Employee number | Name |
---|---|
235 | Anne |
440 | Doug |
500 | Mark |
729 | Sue |
Employee number | Salary |
---|---|
235 | 1700.00 |
235 | 1500.00 |
440 | 950.50 |
500 | 2100.00 |
729 | 1400.90 |
In PF2, the record for 235 is duplicated.
The program gets five records.
Employee number | Name | Salary |
---|---|---|
235 | Anne | 1700.00 |
235 | Anne | 1500.00 |
440 | Doug | 950.50 |
500 | Mark | 0000.00 |
729 | Sue | 1400.90 |
In the join records, the record for 235 is duplicated. The order of the records received for the duplicated record is unpredictable unless the JDUPSEQ keyword is used.