General definition statements

Type specifiers

Arithmetic operator types

The meaning of the various operators might not conform to your experience with other programming languages.

Operator	Comparison is True If
%	Integer division. The first term is divided by the second term. Example 17 % 3 yields 5. Any remainder is ignored. Division by zero yields zero.
*	Multiplication. Signs are handled algebraically. Example 5 * -3 yields -15; -5 * -3 yields 15.
+	Addition.
-	Subtraction. The second term is subtracted from the first term.
/	Rounded division. The first term is divided by the second term and the result is rounded using the round half away from zero or commercial rounding method. Example 17 / 3 yields 6; -5 / 3 yields -2; ‑5 / -3 yields 2. Division by zero yields zero.
//	Modulus or remainder on division of the first term by the second term. Example 17 // 3 yields 2, the remainder when 17 is divided by 3. If the second term is zero, the result is zero.

Comparison relationship types

Operator	Comparison is True If
=	First value is equal to second value.
==	First value is equal to second value.
EQ	First value is equal to second value.
>	First value is greater than second value.
GT	First value is greater than second value.
>=	First value is greater than or equal to second value.
GE	First value is greater than or equal to second value.
<	First value is less than second value.
LT	First value is less than second value.
<=	First value is less than or equal to second value.
LE	First value is less than or equal to second value.
!=	First value is not equal to second value.
<>	First value is not equal to second value.
NE	First value is not equal to second value.

Float types

float_type	Description
Short	A hexadecimal (traditional S/390) floating point number occupying four bytes.
Long	A hexadecimal (traditional S/390) floating point number occupying eight bytes.
Ext Extended	A hexadecimal (traditional S/390) floating point number occupying sixteen bytes.

Integer types

A number of the format and condition specifications require the specification of an integer type or int_type. The following are the integer types:

int_type	Description
Char	A signed 8-bit integer occupying one byte.
Double or LongLong	A signed 64-bit integer occupying eight bytes.
Full or Long	A signed 32-bit integer occupying four bytes.
Half or Short	A signed 16-bit integer occupying two bytes.
UChar	An unsigned 8-bit integer occupying one byte.
UCharUChar	A pair of unsigned 8-bit integers to be formatted as mmm.nnn and occupying two bytes.
UDouble or ULongLong	An unsigned 64-bit integer occupying eight bytes.
UFull or ULong	An unsigned 32-bit integer occupying four bytes.
UHalf or UShort	An unsigned 16-bit integer occupying one byte.
UInt24	An unsigned integer occupying three bytes.
UNiblUNibl	A pair of 4-bit unsigned integers to be formatted as mm.nn and occupying one byte.

Time types

time_type	Description
ChangeMan	Local time of day in the format used in ChangeMan events, character YYYY/MM/DD HH:MM:SS
Compuware	Time of day YYYYMMDDHHMMSS as written by Compuware Application Audit.
CompuwareMS	Time of day YYYYMMDDHHMMSSmmm as written by Compuware Abend-AID.
DMicro	Duration in microseconds and occupying four bytes. The low-order bit represents 1/1000000 of a second or 1 microsecond.
DMicro64	Duration in microseconds and occupying eight bytes. The low-order bit represents 1/1000000 of a second or 1 microsecond.
DMilli64	Duration in milliseconds and occupying eight bytes. The low-order bit represents 1/1000 of a second or 1 millisecond.
DMilliPack	Duration in packed milliseconds occupying four bytes mmmmmmmF.
DSec100	Duration in seconds * 100 and occupying four bytes. The low-order bit represents 1/100 of a second or 10 milliseconds.
DSec1000	Duration in seconds * 1000 and occupying four bytes. The low-order bit represents 1/1000 of a second or 1 millisecond.
DSec1024	Duration in units of 1.024 milliseconds and occupying four bytes.
DSec1048	Duration in units of 1.04876 seconds and occupying four bytes.
DSec62500	Duration in seconds * 62500, sometimes phrased as units of 16 microseconds, and occupying four bytes.
DRMF	Duration in the format of SMF7xINT as written by RMF: Four bytes of packed time 0hhmmssF.
DSTCK	Duration in z/Architecture STCK format; in other words, in units of microseconds times 4096 and occupying eight bytes.
Hex3	Date-only expressed as three hex (binary, really) bytes yydddd as in the JFCB date fields.
IMS	IMS timestamps: packed yyyy dddF hh mm ss uuuuuu occupying eight bytes.
RMF	Date and time in the format of SMF7xIST and SMF7xDAT as written by RMF: Four bytes of packed time 0hhmmssF and four bytes of packed date 0cyydddF.
SMF	Time of day in SMF format: seconds*100 and 0cyydddF packed date, local time, used by most SMF timestamps and occupying eight bytes
SMFdate	0cyydddF with no time occupying four bytes.
SMFdt,	0cyydddF tttttttt seconds * 100 (reversed order relative to SMF) and occupying eight bytes.
SMFUTC	Time of day in SMF format: seconds*100 and 0cyydddF packed date, UTC time, used by Telnet SNA Init and Terminate, and occupying eight bytes.
STCK	Time of day in z/Architecture STCK format, occupying eight bytes. The value is windowed as suggested by IBM. Values indicating times before 11-May-1971 11:56:54 are assumed to refer instead to times after 17-Sep-2042 23:53:46. New applications should consider using STCKE instead.
STCKE	Time of day in z/Architecture STCKE format, occupying sixteen bytes. Only the leftmost 64 bits are examined, yielding a resolution finer than one microsecond.
STCKDB2	DB2's shifted z/Architecture STCK value, occupying eight bytes.
STCKwLSO	Time of day in z/Architecture STCK format, occupying eight bytes, and already adjusted for any Leap Second Offset (LSO). The value is windowed as suggested by IBM. Values indicating times before 11-May-1971 11:56:54 are assumed to refer instead to times after 17-Sep-2042 23:53:46.
TSec100	Time of Day in seconds * 100 and occupying four bytes. The low-order bit represents 1/100 of a second or 10 milliseconds.

This section contains the following topics: