Specifying an External Format for a Numeric Field

For any of the fields in Table 31, specify one of the following valid external numeric formats:

B

Binary

F

Float

I

Integer

L

Left sign

P

Packed decimal

R

Right sign

S

Zoned decimal

U

Unsigned

The default external format for float numeric data is called the external display representation. The format for 4-byte float data is:

   +n.nnnnnnnE+ee,
where + represents the sign (+ or -)
      n represents digits in the mantissa
      e represents digits in the exponent

The format for 8-byte float data is:

   +n.nnnnnnnnnnnnnnnE+eee

Note that a 4-byte float value occupies 14 positions and an 8-byte float value occupies 23 positions.

For numeric data other than float, the default external format is zoned decimal. The external format for compile-time arrays and tables must be zoned-decimal, left-sign or right-sign.

For float compile-time arrays and tables, the compile-time data is specified as either a numeric literal or a float literal. Each element of a 4-byte float array requires 14 positions in the source record; each element of an 8-byte float array requires 23 positions.

Non-float numeric fields defined on input specifications, calculation specifications, or output specifications with no corresponding definition on a definition specification are stored internally in packed-decimal format.