This entry is used to match the records of one file with those of another or to sequence check match fields within one file. Match fields can be specified only for fields in primary and secondary files.
Binary, float, integer, character varying, graphic varying, UCS-2, and unsigned fields cannot be used as match fields.
Match fields within a record are designated by an M1 through M9 code entered in positions 65 and 66 of the appropriate field description specification line. A maximum of nine match fields can be specified.
The match field codes M1 through M9 can be assigned in any sequence. For example, M3 can be defined on the line before M1, or M1 need not be defined at all.
When more than one match field code is used for a record, all fields can be considered as one large field. M1 or the lowest code used is the rightmost or low-order position of the field. M9 or the highest code used is the leftmost or high-order position of the field.
The ALTSEQ (alternate collating sequence) and FTRANS (file translation) keywords on the control specification can be used to alter the collating sequence for match fields.
If match fields are specified for only a single sequential file (input, update, or combined), match fields within the file are sequence checked. The MR indicator is not set on and cannot be used in the program. An out-of-sequence record causes the RPG IV exception/error handling routine to be given control.
In addition to sequence checking, match fields are used to match records from the primary file with those from secondary files.