DISPLAY

       Indicates that each digit occupies one byte of storage. DISPLAY
       is the default if you do not include a USAGE clause.

    BYTE

       Indicates that field value is stored in binary format and that
       the value is stored in one byte of storage.

    WORD

       Indicates that field value is stored in binary format and that
       the value is stored in one word (two bytes) of storage.

    LONG

       Indicates that field value is stored in binary format and that
       the value is stored in one longword (four bytes) of storage.

    QUAD

       Indicates that field value is stored in binary format and that
       the value is stored in one quadword (eight bytes) of storage.

    COMP

       Indicates that the field value is stored in binary format.
       INTEGER is a synonym for COMP. The size of a COMP (or INTEGER)
       field depends on the number of digit positions specified in
       its PICTURE clause. You can avoid having both a USAGE IS COMP
       clause and a PICTURE clause by using the keywords WORD, LONG, and
       QUAD to specify the three types of storage allocation available
       with COMP. COMP Storage Allocation Types shows the COMP storage
       allocation types.

       Table 1-9 COMP Storage Allocation Types

                   Size of
       PIC Clause  Field      Alternate USAGE Type

       9(1) to     2 bytes    WORD
       9(4)
       9(5) to     4 bytes    LONG
       9(9)
       9(10) to    8 bytes    QUAD
       9(18)

    COMP-1

       Indicates that the field value is stored in single-precision
       real format. REAL is a synonym for COMP-1. COMP-1 fields are four
       bytes long and they correspond to the Oracle CDD/Repository F_
       floating-point data types.

    COMP-2

       Indicates that the field value is stored in double-precision real
       format. DOUBLE is a synonym for COMP-2. COMP-2 fields are eight
       bytes long and they correspond to the Oracle CDD/Repository D_
       floating-point data types.

    G_FLOATING

       Indicates that a field is an extended range 64-bit floating-point
       number with precision to approximately 15 decimal digits.

    H_FLOATING

       Indicates that a field is an extended range 128-bit floating-
       point number with precision to approximately 33 decimal digits.

    S_FLOATING

       Indicates that the field is a floating-point number accurate to
       approximately 7 decimal digits. An S_floating-point field is 4
       bytes long.

    T_FLOATING

       Indicates that the field is a floating-point number accurate to
       approximately 15 decimal digits. T_floating-point fields occupy 8
       contiguous bytes in memory.

    COMP-3

       Indicates that the field value is stored in packed-decimal
       format. PACKED is a synonym for COMP-3. The value is stored
       two digits per byte. The value of a COMP-3 field must contain
       a sign. The sign occupies the four low-ordered bits in the
       rightmost byte. The size of the field depends on the number of
       digit positions specified by the field's PICTURE clause:

                    size (in bytes) =  (digit-positions+1)__

                                                 2

       For example, a field with three digit positions is two bytes
       long. If the field contains an even number of digits, the size is
       rounded up. Thus, a 6-digit field is stored in four bytes.

    COMP-5

       Indicates that the field value is stored in signed decimal
       format. ZONED is a synonym for COMP-5. A value in a COMP-5 field
       is stored one digit per byte. Therefore, the size of a COMP-5
       field is the number of digit positions specified in its PICTURE
       clause.

       The sign of a COMP-5 value shares the rightmost byte with the
       lowest-valued digit of the value. The lowercase letters P through
       Y represent a negative sign for the values 0 through 9.

    DATE

       Indicates that the field is a date field.