乐趣区

关于mysql:聊聊mysql的树形结构存储及查询

jiezi

序

本文次要钻研一下 mysql 的树形构造存储及查问

存储 parent

这种形式就是每个节点存储本人的 parent_id 信息

  • 建表及数据筹备

    CREATE TABLE `menu` (`id` int(11) NOT NULL AUTO_INCREMENT,
`name` varchar(50) NOT NULL,
`parent_id` int(11) NOT NULL DEFAULT '0',
PRIMARY KEY (`id`)
) ENGINE=InnoDB;

INSERT INTO `menu` (`id`, `name`, `parent_id`) VALUES
(1, 'level1a',  0),
(2, 'level1b', 0),
(3, 'level2a-1a',1),
(4, 'level2b-1a',1),
(5, 'level2a-1b', 2),
(6, 'level2b-1b', 2),
(7, 'level3-2a1a', 3),
(8, 'level3-2b1a', 4),
(9, 'level3-2a1b', 5),
(10, 'level3-2b1b', 6);

  • 查问

    -- 查问跟节点下的所有节点
SELECT t1.name AS lev1, t2.name as lev2, t3.name as lev3
FROM menu AS t1
LEFT JOIN menu AS t2 ON t2.parent_id = t1.id
LEFT JOIN menu AS t3 ON t3.parent_id = t2.id
WHERE t1.name = 'level1a';

+---------+------------+-------------+
| lev1    | lev2       | lev3        |
+---------+------------+-------------+
| level1a | level2a-1a | level3-2a1a |
| level1a | level2b-1a | level3-2b1a |
+---------+------------+-------------+

-- 查问叶子节点
SELECT t1.name FROM
menu AS t1 LEFT JOIN menu as t2
ON t1.id = t2.parent_id
WHERE t2.id IS NULL;

+-------------+
| name        |
+-------------+
| level3-2a1a |
| level3-2b1a |
| level3-2a1b |
| level3-2b1b |
+-------------+

    存储及批改上比拟不便，就是要在 sql 外头查问树比拟吃力，个别是加载到内存由利用本人结构

存储 path

这种形式在存储 parent 的根底上，额定存储 path，即从根节点到该节点的门路

  • 建表及数据筹备

    CREATE TABLE `menu_path` (`id` int(11) NOT NULL AUTO_INCREMENT,
`name` varchar(50) NOT NULL,
`parent_id` int(11) NOT NULL DEFAULT '0',
`path` varchar(255) NOT NULL DEFAULT '',
PRIMARY KEY (`id`)
) ENGINE=InnoDB;

INSERT INTO `menu_path` (`id`, `name`, `parent_id`, `path`) VALUES
(1, 'level1a', 0, '1/'),
(2, 'level1b', 0, '2/'),
(3, 'level2a-1a',1, '1/3'),
(4, 'level2b-1a',1, '1/4'),
(5, 'level2a-1b', 2, '2/5'),
(6, 'level2b-1b', 2, '2/6'),
(7, 'level3-2a1a', 3, '1/3/7'),
(8, 'level3-2b1a', 4, '1/4/8'),
(9, 'level3-2a1b', 5, '2/5/9'),
(10, 'level3-2b1b', 6, '2/6/10');

  • 查问

    -- 查问某个节点的所有子节点
select * from menu_path where path like '1/%'
+----+-------------+-----------+-------+
| id | name        | parent_id | path  |
+----+-------------+-----------+-------+
| 1  | level1a     | 0         | 1/    |
| 3  | level2a-1a  | 1         | 1/3   |
| 4  | level2b-1a  | 1         | 1/4   |
| 7  | level3-2a1a | 3         | 1/3/7 |
| 8  | level3-2b1a | 4         | 1/4/8 |
+----+-------------+-----------+-------+

    查找某个节点及其子节点比拟方面，就是批改比拟吃力，特地是节点挪动，所有子节点的 path 都得跟着批改

MPTT(Modified Preorder Tree Traversal)

不存储 parent_id，改为存储 lft,rgt，它们的值由树的先序遍历程序决定

  • 建表及数据筹备

    CREATE TABLE `menu_preorder` (`id` int(11) NOT NULL,
`name` varchar(50) NOT NULL,
`lft` int(11) NOT NULL DEFAULT '0',
`rgt` int(11) NOT NULL DEFAULT '0',
PRIMARY KEY (`id`)
) ENGINE=InnoDB;

                 1(level1a)14
       2(level2a)7                8(level2b)13
3(level3a-2a)4 5(level3b-2a)6 9(level3c-2b)10 11(level3d-2b)12

INSERT INTO `menu_preorder` (`id`, `name`, `lft`, `rgt`) VALUES
(1, 'level1a', 1, 14),
(2, 'level2a',2, 7),
(3, 'level2b',8, 13),
(4, 'level3a-2a', 3, 4),
(5, 'level3b-2a', 5, 6),
(6, 'level3c-2b', 9, 10),
(7, 'level3d-2b', 11, 12);

select * from menu_preorder
+----+------------+-----+-----+
| id | name       | lft | rgt |
+----+------------+-----+-----+
| 1  | level1a    | 1   | 14  |
| 2  | level2a    | 2   | 7   |
| 3  | level2b    | 8   | 13  |
| 4  | level3a-2a | 3   | 4   |
| 5  | level3b-2a | 5   | 6   |
| 6  | level3c-2b | 9   | 10  |
| 7  | level3d-2b | 11  | 12  |
+----+------------+-----+-----+

  • 查问

    -- 查问某个节点及其子节点，比方 level2b
select * from menu_preorder where lft between 8 and 13
+----+------------+-----+-----+
| id | name       | lft | rgt |
+----+------------+-----+-----+
| 3  | level2b    | 8   | 13  |
| 6  | level3c-2b | 9   | 10  |
| 7  | level3d-2b | 11  | 12  |
+----+------------+-----+-----+

-- 查问所有叶子节点
SELECT name
FROM menu_preorder
WHERE rgt = lft + 1;

+------------+
| name       |
+------------+
| level3a-2a |
| level3b-2a |
| level3c-2b |
| level3d-2b |
+------------+

-- 查问某个节点及其父节点
SELECT parent.*
FROM menu_preorder AS node,
menu_preorder AS parent
WHERE node.lft BETWEEN parent.lft AND parent.rgt
AND node.name = 'level2b'
ORDER BY parent.lft;

+----+---------+-----+-----+
| id | name    | lft | rgt |
+----+---------+-----+-----+
| 1  | level1a | 1   | 14  |
| 3  | level2b | 8   | 13  |
+----+---------+-----+-----+

-- 树形构造展现
SELECT CONCAT(REPEAT(' ', COUNT(parent.name) - 1), node.name) AS name
FROM menu_preorder AS node,
menu_preorder AS parent
WHERE node.lft BETWEEN parent.lft AND parent.rgt
GROUP BY node.name
ORDER BY node.lft;

+--------------+
| name         |
+--------------+
| level1a      |
|  level2a     |
|   level3a-2a |
|   level3b-2a |
|  level2b     |
|   level3c-2b |
|   level3d-2b |
+--------------+

益处是通过 lft 进行范畴 (该节点的 lft,rgt 作为范畴) 查找就能够，毛病就是增删节点导致很多节点的 lft 及 rgt 都要批改

小结

  • 存储 parent 的形式最为场景，个别树形构造数据量不大的话，间接在应用层内存结构树形构造和搜寻
  • 存储 path 的益处是能够借助 path 来查找节点及其子节点，毛病就是挪动 node 须要级联所有子节点的 path，比拟吃力
  • MPTT 的形式益处是通过 lft 进行范畴 (该节点的 lft,rgt 作为范畴) 查找就能够，毛病就是增删节点导致很多节点的 lft 及 rgt 都要批改

doc

  • Managing Hierarchical Data in MySQL
  • hierarchical-data-database
  • hierarchical-data-database-2
  • hierarchical-data-database-3
退出移动版