By default, Enterprise Server uses the ASCII character set. If you wish to use EBCDIC, you can override this default for an individual enterprise server.

To select ASCII or EBCDIC for an enterprise server, you use the environment variable MF_CHARSET. You specify it via ES Admin. On the Edit page for the enterprise server, on the General tab, you make the following entry in the [ES-Environment] section of the Configuration Information field:

MF_CHARSET=E

or:

MF_CHARSET=A

You can enter these as:

MF_CHARSET=EBCDIC

or:

MF_CHARSET=ASCII

for clarity if you wish. However, only the first character (E or A) is significant.

Your programs should be compiled with the CHARSET compiler directive that matches the character set of the enterprise server. The system will handle user program data appropriately for programs invoked as a result of an EXEC PGM= JCL statement, the RUN subcommand of the DSN command, or the TSO CALL command.

You should not mix ASCII and EBCDIC programs. If you do so, you are responsible for any conversion required. The compiler directive is:

CHARSET(EBCDIC)

or:

CHARSET(ASCII)

The character set attribute of all datasets is set to the current system character set. Based on their file’s character set, LSEQ datasets and SYSIN in-stream datasets are automatically converted to suit the program.