Data formats used by SDF
The SDF input or output can have either of the following data formats:
- File
- A hierarchical file system (HFS) or zFS file is a data stream with one or more FIN messages
which are separated by a specific record delimiter character or records delimiters. SDF supports the
following formats:
- SWIFT RJE
- In an RJE stream, FIN messages are encoded in ASCII and separated by a dollar ($) character. A carriage-return line-feed (CRLF) character sequence in a FIN message is represented by a carriage-return/newline character sequence (X'0D0A').
- Records delimited by CRLF
- SDF supports data input streams in which individual FIN records are delimited using a carriage-return line-feed (CRLF) (X'0D0A') character sequence.
- No record delimiter
- This format assumes that SWIFT FIN messages are stored without any record delimiter in the input stream. This means that the Sequential Data Facility will try to interpret the FIN block structures and will assume a new record if a new (not nested) block 1 structure is detected.
- PS data set or PDS member
-
A physical sequential (PS) data set or a partitioned data set (PDS) member with a variable blocked record structure (DSORG=VB). Typically, each FIN message in the data set corresponds to one data set record. If the maximum record length is smaller than the largest FIN message, all FIN messages in this data set must be segmented as defined by IBM® MERVA. In a data set containing segmented FIN messages, each record starts with one of the following segmenting identifier characters:
- 0 (X'F0')
- Only segment of a message.
- 1 (X'F1')
- First segment of a message.
- 2 (X'F2')
- Last segment of a message.
- 3 (X'F3')
- Middle segment of a message.