SQL基础-->层次化查询(START BY ... CONNECT BY PRIOR)
--======================================================
--SQL基础-->层次化查询(START BY ... CONNECT BY PRIOR)
--======================================================
层次化查询,即树型结构查询,是SQL中经经常使用到的功能之中的一个,通常由根节点,父节点,子节点,叶节点组成,其语法例如以下:
SELECT [LEVEL] ,column,expression,...
FROM table_name
[WHERE where_clause]
[[START WITH start_condition] [CONNECT BY PRIOR prior_condition]];
LEVEL:为伪列,用于表示树的层次
start_condition：层次化查询的起始条件
prior_condition：定义父节点和子节点之间的关系
--使用start with ...connect by prior 从根节点開始遍历
SQL> select empno,mgr,ename,job from emp
2 start with empno = 7839
3 connect by prior empno = mgr;
EMPNO MGR ENAME JOB
---------- ---------- ---------- ---------
7839 KING PRESIDENT
7566 7839 JONES MANAGER
7788 7566 SCOTT ANALYST
7876 7788 ADAMS CLERK
7902 7566 FORD ANALYST
7369 7902 SMITH CLERK
7698 7839 BLAKE MANAGER
7499 7698 ALLEN SALESMAN
7521 7698 WARD SALESMAN
7654 7698 MARTIN SALESMAN
7844 7698 TURNER SALESMAN
EMPNO MGR ENAME JOB
---------- ---------- ---------- ---------
7900 7698 JAMES CLERK
7782 7839 CLARK MANAGER
7934 7782 MILLER CLERK
14 rows selected.
树型结构遍历过程(通过上面的查询来描写叙述)
1).从根节点開始(即where_clause中的条件,假设为非根节点则分根节点作为根节点開始遍历,如上例empno = 7839)
2).遍历根节点(得到empno = 7839记录的相关信息)
3).推断该节点是否存在由子节点，假设则訪问最左側未被訪问的子节点,转到),否则下一步
如上例中prior_condition为empno 的记录
4).当节点为叶节点，则訪问完成,否则,转到)
5).返回到该节点的父节点,转到)
--伪列level的使用
--注意connect by prior empno = mgr 的理解
--prior表示前一条记录,即下一条返回记录的mgr应当等于前一条记录的empno
SQL> select level,empno,mgr,ename,job from emp
2 start with ename = 'KING'
3 connect by prior empno = mgr
4 order by level;
LEVEL EMPNO MGR ENAME JOB
---------- ---------- ---------- ---------- ---------
1 7839 KING PRESIDENT
2 7566 7839 JONES MANAGER
2 7698 7839 BLAKE MANAGER
2 7782 7839 CLARK MANAGER
3 7902 7566 FORD ANALYST
3 7521 7698 WARD SALESMAN
3 7900 7698 JAMES CLERK
3 7934 7782 MILLER CLERK
3 7499 7698 ALLEN SALESMAN
3 7788 7566 SCOTT ANALYST
3 7654 7698 MARTIN SALESMAN
LEVEL EMPNO MGR ENAME JOB
---------- ---------- ---------- ---------- ---------
3 7844 7698 TURNER SALESMAN
4 7876 7788 ADAMS CLERK
4 7369 7902 SMITH CLERK
--获得层次数
SQL> select count(distinct level) "Level" from emp
2 start with ename = 'KING'
3 connect by prior empno = mgr;
Level
----------
4
--格式化层次查询结果(使用左填充* level - 1个空格)
SQL> col Ename for a30
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename",
3 job
4 from emp
5 start with ename = 'KING'
6 connect by prior empno = mgr;
LEVEL Ename JOB
---------- ------------------------------ ---------
1 KING PRESIDENT
2 JONES MANAGER
3 SCOTT ANALYST
4 ADAMS CLERK
3 FORD ANALYST
4 SMITH CLERK
2 BLAKE MANAGER
3 ALLEN SALESMAN
3 WARD SALESMAN
3 MARTIN SALESMAN
3 TURNER SALESMAN
LEVEL Ename JOB
---------- ------------------------------ ---------
3 JAMES CLERK
2 CLARK MANAGER
3 MILLER CLERK
14 rows selected.
--从非根节点開始遍历(仅仅需改动start with 中的条件就可以)
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename",
3 job
4 from emp
5 start with ename = 'SCOTT'
6 connect by prior empno = mgr;
LEVEL Ename JOB
---------- ------------------------------ ---------
1 SCOTT ANALYST
2 ADAMS CLERK
--从下向上遍历(交换connect by prior中的条件就可以,使用mgr = empno)
--注意connect by prior mgr = empno 的理解
--prior表示前一条记录,即下一条返回记录的empno应当等于前一条记录的mgr
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename",
3 job
4 from emp
5 start with ename = 'SCOTT'
6 connect by prior mgr = empno;
LEVEL Ename JOB
---------- ------------------------------ ---------
1 SCOTT ANALYST
2 JONES MANAGER
3 KING PRESIDENT
--从下向上遍历(也能够将prior置于等号右边,得到同样的结果)
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename",
3 job
4 from emp
5 start with ename = 'SCOTT'
6 connect by empno = prior mgr;
LEVEL Ename JOB
---------- ------------------------------ ---------
1 SCOTT ANALYST
2 JONES MANAGER
3 KING PRESIDENT
--从层次查询中删除节点和分支
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename"
3 ,job
4 from emp
5 where ename != 'SCOTT' --通过where子句来过滤SCOTT用户，但SCOTT的下属ADAMS并没有过滤掉
6 start with empno = 7839
7 connect by prior empno = mgr;
LEVEL Ename JOB
---------- -------------------- ---------
1 KING PRESIDENT
2 JONES MANAGER
4 ADAMS CLERK
3 FORD ANALYST
4 SMITH CLERK
2 BLAKE MANAGER
3 ALLEN SALESMAN
3 WARD SALESMAN
3 MARTIN SALESMAN
3 TURNER SALESMAN
3 JAMES CLERK
LEVEL Ename JOB
---------- -------------------- ---------
2 CLARK MANAGER
3 MILLER CLERK
13 rows selected.
--通过将过滤条件由where 子句的内容移动到connect by prior 子句中过滤掉SCOTT及其下属
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename"
3 ,job
4 from emp
5 start with empno = 7839
6 connect by prior empno = mgr and ename != 'SCOTT';
LEVEL Ename JOB
---------- -------------------- ---------
1 KING PRESIDENT
2 JONES MANAGER
3 FORD ANALYST
4 SMITH CLERK
2 BLAKE MANAGER
3 ALLEN SALESMAN
3 WARD SALESMAN
3 MARTIN SALESMAN
3 TURNER SALESMAN
3 JAMES CLERK
2 CLARK MANAGER
LEVEL Ename JOB
---------- -------------------- ---------
3 MILLER CLERK
12 rows selected.
--在层次化查询中添加过滤条件或使用子查询
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename"
3 ,job
4 from emp
5 where sal > 2500
6 start with empno = 7839
7 connect by prior empno = mgr
8 ;
LEVEL Ename JOB
---------- -------------------- ---------
1 KING PRESIDENT
2 JONES MANAGER
3 SCOTT ANALYST
3 FORD ANALYST
2 BLAKE MANAGER
SQL> select level,
2 lpad(' ',2 * level - 1) || ename as "Ename"
3 ,job
4 from emp
5 where sal > (select avg(sal) from emp)
6 start with empno = 7839
7 connect by prior empno = mgr ;
LEVEL Ename JOB
---------- -------------------- ---------
1 KING PRESIDENT
2 JONES MANAGER
3 SCOTT ANALYST
3 FORD ANALYST
2 BLAKE MANAGER
2 CLARK MANAGER
6 rows selected.
很多其它參考：
使用OEM,SQL*Plus,iSQL*Plus 管理Oracle实例
SQL基础-->层次化查询(START BY ... CONNECT BY PRIOR)的更多相关文章
- SQL基础-->层次化查询(START BY ... CONNECT BY PRIOR)[转]
--====================================================== --SQL基础-->层次化查询(START BY ... CONNECT BY ...
- oracle递归层级查询 start with connect by prior
递归层级查询:start with connect by prior 以部门表作为解析 表结构:dept{id:'主键',name:'部门名称',parent_id:'父亲id'} select * ...
- SQL基础学习_02_查询
SELECT语句 1. SELECT语句查询列(字段): SELECT <列名> FROM <表名>; 该语句使用了两个SQL子句,SELECT子句列举了 ...
- sql基础语法-联接查询
交叉联接 1.不带where条件的,将返回两个表的 行乘积 select c.*, e.* from Sales.Customers c cross join hr.Employees e 2.带wh ...
- SQL基础--> 约束(CONSTRAINT)
--============================= --SQL基础--> 约束(CONSTRAINT) --============================= 一.几类数据完 ...
- oracle 层次化查询(生成菜单树等)
1.简介:Oracle层次化查询是Oracle特有的功能实现,主要用于返回一个数据集,这个数据集存在树的关系(数据集中存在一个Pid记录着当前数据集某一条记录的Id). 2.层次化查询主要包含两个子句 ...
- （二十）sql基础
sql基础 --单表查询 select * from student; select * from score; --投影查询 select * from student; --条件查询 select ...
- Oracle学习（一）SQL基础
一.认识SQL SQL是什么? SQL,结构化查询语言,全称是 Structured Query Language. SQL 是一门 ANSI(American National Standards ...
- 【SQL】CONNECT BY 层次化查询
层次化查询,顾名思义就是把查询结果有层次的呈现出来.层次化查询结果类似于树状结构,最顶端的是“根节点”,下面是“父节点”,没有子节点的是“叶节点”. 为了让一个或多个表具有层次关系,必须使用相关的字段 ...
随机推荐
- （原）剑指offer之栈和队列
题目描述 用两个栈来实现一个队列,完成队列的Push和Pop操作. 队列中的元素为int类型. 设两个栈为stack1,stack2: 1:首先想到最简单的方法:是入栈时将数据压入stack1,出栈时 ...
- PAT Basic 1021
1021 个位数统计 给定一个k位整数N = d~k-1~*10^k-1^ + ... + d~1~*10^1^ + d~0~ (0<=d~i~<=9, i=0,...,k-1, d~k- ...
- 算法学习记录-图——应用之拓扑排序（Topological Sort）
这一篇写有向无环图及其它的应用: 清楚概念: 有向无环图(DAG):一个无环的有向图.通俗的讲就是从一个点沿着有向边出发,无论怎么遍历都不会回到出发点上. 有向无环图是描述一项工程或者系统的进行过程的 ...
- Oracle中Restore和Recovery的区别
一.参考解释一 在Oracle的备份与恢复的知识点中,经常会出现Restore 和 Recovery两个词. 由于这两个词在字典中的解释很接近,困扰了我很久.直到我在Oracle的官方文档中看到了以下 ...
- cf898d Alarm Clock
区间上有 \(n\) 个点,问你为达到目的:长度为 \(m\) 的区间内点的个数都 $ < k$需要去掉多少个点. 贪心.每个区间我们总是去掉最后的,也就是说除非万不得已我们是不会去掉点的. 队 ...
- CI - Set CSRF Hash and Cookie
/** * Set CSRF Hash and Cookie * * @return string */ protected function _csrf_set_hash() { if ($this ...
- UVA - 10591 Happy Number
Happy Number UVA - 10591 Let the sum of the square of the digits of a positive integer S0 be represe ...
- iOS控件-3级城市列表-plist版
@import url(http://i.cnblogs.com/Load.ashx?type=style&file=SyntaxHighlighter.css);@import url(/c ...
- lua学习随笔
1.1 Chunks 1.2 全局变量 访问一个没有初始化的全局变量也不会出错,只不过的到的结果是nil 如果想删除一个全局变量,只需要将变量赋值为nil 1.3 词法约定 标识符 保留字不能作为 ...
- php 学习随笔
---恢复内容开始--- round进行格式化数值(进位规则遵守“四舍六入五双”,即前一位是奇数,则进一,前一位是偶数则舍入,因此,rount(1.5)=2,round(2.5)=2,round(0. ...