Input modes (devices other than 3270, SLU 2, or ISC subsystems)

MFS expects input message fields to be entered in the sequence in which they were defined to the MFS Language utility program. For devices other than SLU 2 and 3270, MFS application designers have a choice of how fields are defined and how MFS should scan those fields. You can select record mode or stream mode. Record mode is the default.

In record mode:

  • Input fields are defined as occurring within a specific record (a line or card from the 3770, or SLU 1; a transmission from the Finance or SLU P workstation) that is sent from the input device.
  • Fields must not be split across record boundaries.
  • Fields defined within a record must appear on that record to be considered by MFS.
  • When MFS locates the end of a record, the current field is terminated and any other fields defined for that record are processed with no device data (message fill).
  • If the record received by IMS TM contains more data fields than the number of fields defined for the record, the remaining data fields are not considered by MFS.

For input data from a Finance or SLU P workstation remote program, the input message header or //midname can be transmitted separately if the data fields for the first record do not fit in the same record. If no data follows the input message header or the //midname, MFS considers the next transmission received to be the first record of the input message.

In stream mode:

  • Fields are defined as a contiguous stream of data unaffected by record boundaries.
  • Fields can be split across input records and fields can be entered from any input record as long as they are entered in the defined sequence.