3

我知道这种问题已经在这里发布过很多次了，例如：Java way

id我在标准树模式（ ，，， ）parent_id中有大量数据（150k+ some_data）

问题：如何获取给定 node_id 的叶子？

表结构：

CREATE TABLE `DATA_TREE` (
  `ID` int(11) NOT NULL,
  `PARENT_ID` int(11) NOT NULL,
  `DATA` varchar(45) DEFAULT NULL,
  PRIMARY KEY (`ID`),
  UNIQUE KEY `ID_UNIQUE` (`ID`),
  KEY `fk_DATA_TREE_1_idx` (`PARENT_ID`),
  CONSTRAINT `fk_DATA_TREE_1` FOREIGN KEY (`PARENT_ID`) REFERENCES `DATA_TREE` (`ID`) ON DELETE NO ACTION ON UPDATE NO ACTION
) ENGINE=InnoDB DEFAULT CHARSET=utf

数据库：MySQL 5.1.61

4

1 回答 1

7

没有办法在单个查询中执行此操作。即使有，也可能效率很低。

我们可以使用存储过程和循环来做到这一点。使用您添加的索引，它也应该很快。这使用两个表从输入表 (A) 中选择节点并将节点及其子节点插入 (B)。然后它将 B 交换为 A，并重复直到 A 中不再存在非叶节点。好消息是循环迭代只会与输入节点和最后一个叶节点之间的级别一样多，在大多数情况下是可能没那么深。这个存储过程会比在代码外部执行它更快。

仅供参考，我在安装处理临时表时遇到了困难，如果您收到“错误 2”，请删除临时关键字。

delimiter $$
drop procedure if exists GetLeafNodes $$
create procedure GetLeafNodes(nodeid int)
begin
declare N int default 1;

-- create two working sets of IDs, we'll go back and forth between these two sets
drop temporary table if exists A;
drop temporary table if exists B;
create temporary table A(node int, child int);
create temporary table B(node int, child int);

-- insert our single input node into the working set
insert into A values (null, nodeid);

while (N>0) do
  -- keep selecting child nodes for each node we are now tracking
  -- leaf nodes will end up with the child set to null
  insert into B
  select ifnull(A.child,A.node), tree.ID
    from A
    left outer join DATA_TREE as tree on A.child=tree.parent_id;

  -- now swap A and B
  rename table A to temp, B to A, temp to B;

  -- remove non-leaf nodes from table B
  delete from B;

  -- exit when there are no longer any non-leaf nodes in A
  set N=(select count(*) from A where child is not null);
end while;

-- now output our list of leaf nodes
select node from A;

drop temporary table A;
drop temporary table B;
end $$
DELIMITER ;
call GetLeafNodes(4);

我使用以下样本集进行测试：

CREATE TABLE `DATA_TREE` (
  `ID` int(11) NOT NULL,
  `PARENT_ID` int(11) NOT NULL,
  PRIMARY KEY (`ID`),
  UNIQUE KEY `ID_UNIQUE` (`ID`),
  KEY `fk_DATA_TREE_1_idx` (`PARENT_ID`)
) ENGINE=InnoDB
;

insert into DATA_TREE values
(1,0),(2,1),(3,1),(4,1),(5,3),(6,3),(7,4),(8,4),(9,4),(10,6),(11,6),(12,7),(13,9),(14,9),(15,12),(16,12),(17,12),(18,14);
于 2012-10-09T10:04:16.217 回答