I have two tables, t1
and t2
with two columns each - id_user
and age
.
t1.age
to the greatest of t1.age
and t2.age
for matching ID's and leave t1.age
unchanged if there is no matching ID in t2
.
Before update:
t1 +-------+---+ |id_user|age| +-------+---+ | 1| 5| +-------+---+ | 2| 10| +-------+---+ | 3| 10| +-------+---+ t2 +-------+---+ |id_user|age| +-------+---+ | 2| 12| +-------+---+ | 3| 8| +-------+---+ | 4| 20| +-------+开发者_如何学Python---+
After update:
t1 +-------+---+ |id_user|age| +-------+---+ | 1| 5| +-------+---+ | 2| 12| +-------+---+ | 3| 10| +-------+---+
You may want to try:
UPDATE t1
JOIN t2 ON (t2.id_user = t1.id_user)
SET t1.age = t2.age
WHERE t2.age > t1.age;
Test Case:
CREATE TABLE t1 (id_user int, age int);
CREATE TABLE t2 (id_user int, age int);
INSERT INTO t1 VALUES (1, 5);
INSERT INTO t1 VALUES (2, 10);
INSERT INTO t1 VALUES (3, 10);
INSERT INTO t2 VALUES (2, 12);
INSERT INTO t2 VALUES (3, 8);
INSERT INTO t2 VALUES (4, 20);
Result:
SELECT * FROM t1;
+---------+------+
| id_user | age |
+---------+------+
| 1 | 5 |
| 2 | 12 |
| 3 | 10 |
+---------+------+
3 rows in set (0.00 sec)
UPDATE t1
SET age = T2.age
FROM t1
INNER JOIN t2
ON t2.id_user = t1.id_user
WHERE t2.age > t1.age
精彩评论