The records to be stored using the BINILOAD utility routine must be readied in an input file.
The records of the input file must all have the same structure. The same input file can, however, be used to store database records of varying structure if different items of the input record are selected.
An input file record may contain, in addition to items for the database record, items with the following information:
User information. Ignored by BINILOAD.
Key values which are interpreted by BINILOAD in order to determine the correct set occurrence.
Information interpreted by BINILOAD in order to determine the sequence of member records.
Information interpreted by BINILOAD in order to indicate the insertion or non-insertion of member records.
Figure 18: Example of the structure of an input file record
BINILOAD statements are used to specify the position and length of the items to be transferred to the database record (see table 23 RECORD DISPL statement, "Statements for BINILOAD").
The item contents are stored in the database in the format in which they appear in the records of the input file. This means that the item contents specified are not converted to the type defined for these items in the Schema DDL (see "Defining an alphanumeric item of fixed length" in the "Design and Definition" manual).
The input file may be stored on disk or tape. In both cases the input file can be a SAM or ISAM file (EDT format) containing records of the same length in either fixed or variable record format (RECFORM=V or F). (See table 21 USER RECORD LENGTH statement, "Statements for BINILOAD".)
If the input comes from an uncataloged tape file, STATE=FOREIGN must also be specified in the /CREATE-FILE command for the input file.
The name of the input file is specified in the INPUT-FILE statement, its record length in the
USER RECORD LENGTH statement.
At the end of this chapter there is an example for the BINILOAD command sequence, followed by an example of a BINILOAD input file.