Question

我有一个具有以下结构的表：

+----+------+------+----------------+---------------------+
| id | val1 | val2 |     email      |     lastUpdated     |
+----+------+------+----------------+---------------------+
|  1 |    0 |    3 | test@gmail.com | 2015-02-07 13:21:55 |
|  2 |    5 |    0 | test@gmail.com | 2015-02-07 09:12:27 |
+----+------+------+----------------+---------------------+

由于我所做的查询中存在错误，因此有多行具有相同的“电子邮件”值。我想为所有重复项合并val1，val2和lastUpdated（更大的值）。它会给出这样的结果：

+----+------+------+----------------+---------------------+
| id | val1 | val2 |     email      |     lastUpdated     |
+----+------+------+----------------+---------------------+
|  1 |    5 |    3 | test@gmail.com | 2015-02-07 13:21:55 |
+----+------+------+----------------+---------------------+

Answer 1

假设除了一行外，所有valX列都为零，

RENAME TABLE originalTable TO messedUpTable
;
CREATE TABLE originalTable
SELECT
  id,
  SUM(val1) AS val1,
  SUM(val2) AS val2,
  email,
  MAX(lastUpdated) AS lastUpdated
FROM messedUpTable
GROUP BY email
;
-- check originalTable to be sure it is okay
;
DROP TABLE messedUpTable

如果您在originalTable上有索引，则必须再次定义它们。

带有索引示例的

SQLFiddle。

Answer 2

select max(val1), max(val2), email, max(lastUpdated)
from your_table
group by email

如何组合具有匹配列的行？

2 个答案: