在有多條欄位重複記錄的表中，經常需要查詢重複記錄中的最大，最小值。

1.直接使用max()函式，例子是獲取最大的id

測試表結構：

`id` int(10) unsigned NOT NULL AUTO_INCREMENT,
  `role_id` int(10) unsigned NOT NULL,
  `user_id` int(10) unsigned NOT NULL,
  PRIMARY KEY (`id`),

SELECT
					max(id)
				FROM
					test
				GROUP BY
					role_id

2.可以用表自連線實現

SELECT
	*
FROM
	test
WHERE
	id NOT IN (
		SELECT
			t.id
		FROM
			test t
		JOIN test e ON t.role_id = e.role_id
		WHERE
			t.id > e.id
		GROUP BY
			t.id
	)
 

DELETE
FROM
	test
WHERE
	id IN (
		SELECT
			m_id
		FROM
			(
				SELECT
					max(id) AS m_id
				FROM
					test
				GROUP BY
					role_id
			) b
	)