Ciro Santilli

SUM

Sum of a column.

It is possible to use any function of the input row such as c0 * c1.

It is possible to make two aggregate function queries on the same SELECT as in (1), but making a non-aggregate function query with an aggregate function query only shows the first non aggregate output as in (2), which is probably not what you want.

WHERE is applied before aggregate functions, and selects which rows will be used for the calculation of the aggregate:

CREATE TABLE t (c0 INT, c1 INT);
INSERT INTO t VALUES (1, 1), (2, 4), (3, 9);
SELECT SUM(c0) FROM t;
SELECT SUM(c1) FROM t;
SELECT SUM(c0 * c1) FROM t;
SELECT SUM(c0 + c1) AS name FROM t;
SELECT MAX(c0), SUM(c0) FROM t;     #(1)
SELECT MAX(c0) + SUM(c0) FROM t;
SELECT c0, SUM(c0) FROM t;          #(2)
SELECT SUM(c0) FROM t WHERE c0 > 1;
DROP TABLE t;

Output:

SUM(c0)
6

SUM(c1)
14

SUM(c0 * c1)
36

name
20

MAX(c0) SUM(c0)
3       6

MAX(c0) + SUM(c0)
9

c0  SUM(c0)
1   6

c0  SUM(c0)
1   5

NULL is treated as TODO: is giving 1 in MySQL 5.5. But I have seen many places say it could break up, like return NULL.

CREATE TABLE t (c INT);
INSERT INTO t VALUES (1), (NULL);
SELECT SUM(c) FROM t;
DROP TABLE t;