Table 12.11 Arithmetic Operators
The usual arithmetic operators are available. The result is determined according to the following rules:
If both operands are integers and any of them are unsigned, the result is an unsigned integer. For subtraction, if the
NO_UNSIGNED_SUBTRACTIONSQL mode is enabled, the result is signed even if any operand is unsigned.
In division performed with
/, the scale of the result when using two exact-value operands is the scale of the first operand plus the value of the
div_precision_incrementsystem variable (which is 4 by default). For example, the result of the expression
5.05 / 0.014has a scale of six decimal places (
These rules are applied for each operation, such that nested
calculations imply the precision of each component. Hence,
(14620 / 9432456) / (24250 / 9432456),
resolves first to
(0.0014) / (0.0026), with
the final result having 8 decimal places
Because of these rules and the way they are applied, care should be taken to ensure that components and subcomponents of a calculation use the appropriate level of precision. See Section 12.10, “Cast Functions and Operators”.
For information about handling of overflow in numeric expression evaluation, see Section 11.2.6, “Out-of-Range and Overflow Handling”.
Arithmetic operators apply to numbers. For other types of
values, alternative operations may be available. For example, to
add date values, use
see Section 12.7, “Date and Time Functions”.
SELECT 3+5;-> 8
SELECT 3-5;-> -2
Unary minus. This operator changes the sign of the operand.
SELECT - 2;-> -2
SELECT 3*5;-> 15 mysql>
SELECT 18014398509481984*18014398509481984.0;-> 324518553658426726783156020576256.0 mysql>
SELECT 18014398509481984*18014398509481984;-> out-of-range error
SELECT 3/5;-> 0.60
Division by zero produces a
SELECT 102/(1-1);-> NULL
A division is calculated with
BIGINTarithmetic only if performed in a context where its result is converted to an integer.
Integer division. Discards from the division result any fractional part to the right of the decimal point.
If either operand has a noninteger type, the operands are converted to
DECIMALand divided using
DECIMALarithmetic before converting the result to
BIGINT. If the result exceeds
BIGINTrange, an error occurs.
SELECT 5 DIV 2, -5 DIV 2, 5 DIV -2, -5 DIV -2;-> 2, -2, -2, 2