1、作用：

用于存在父子，祖孙，上下级等层级关系的数据表进行层级查询。

2、语法

SELECT ... 
FROM ....
START WITH cond1 
CONNECT BY cond2
WHERE cond3;

2.1、说明

start with: 指定起始节点的条件

connect by: 指定父子行的条件关系

prior: 查询父行的限定符，格式: prior column1 = column2 or column1 = prior column2 and … ，

nocycle: 若数据表中存在循环行，那么不添加此关键字会报错，添加关键字后，便不会报错，但循环的两行只会显示其中的第一条

循环行: 该行只有一个子行，而且子行又是该行的祖先行

connect_by_iscycle: 前置条件:在使用了nocycle之后才能使用此关键字，用于表示是否是循环行，0表示否，1 表示是

connect_by_isleaf: 是否是叶子节点，0表示否，1 表示是

level: level伪列,表示层级，值越小层级越高，level=1为层级最高节点

3、构造数据

-- 创建表
create table employee(emp_id number(18),lead_id number(18),emp_name varchar2(200),salary number(10,2),dept_no varchar2(8)
);-- 添加数据
insert into employee values('1',0,'king','1000000.00','001');
insert into employee values('2',1,'jack','50500.00','002');
insert into employee values('3',1,'arise','60000.00','003');
insert into employee values('4',2,'scott','30000.00','002');
insert into employee values('5',2,'tiger','25000.00','002');
insert into employee values('6',3,'wudde','23000.00','003');
insert into employee values('7',3,'joker','21000.00','003');

4、查询jack下的所有子节点

select * from employee start with emp_name='jack' connect by prior emp_id=lead_id;emp_id | lead_id | emp_name | salary | dept_no 
--------+---------+----------+--------+---------2 |       1 | jack     |  50500 | 0024 |       2 | scott    |  30000 | 0025 |       2 | tiger    |  25000 | 002
(3 rows)

5、查询jack和arise下的所有子节点

select * from employee start with emp_name in ('jack', 'arise') connect by prior emp_id=lead_id;emp_id | lead_id | emp_name | salary | dept_no 
--------+---------+----------+--------+---------2 |       1 | jack     |  50500 | 0024 |       2 | scott    |  30000 | 0025 |       2 | tiger    |  25000 | 0023 |       1 | arise    |  60000 | 0036 |       3 | wudde    |  23000 | 0037 |       3 | joker    |  21000 | 003
(6 rows)

6、查询jack的祖先节点

select * from employee start with emp_name='jack' connect by prior lead_id=emp_id;emp_id | lead_id | emp_name | salary  | dept_no 
--------+---------+----------+---------+---------2 |       1 | jack     |   50500 | 0021 |       0 | king     | 1000000 | 001
(2 rows)

7、查询一个节点的叔叔伯父节点

--查看emp_id为6的节点的叔叔伯父节点
with temp as (select  employee.*,prior emp_name,level lefrom employee start with lead_id = 0connect by lead_id=prior emp_id
)
select *
from temp t
left join temp tt on tt.emp_id=6 --此处需要限定
where t.le = (tt.le-1)and t.emp_id not in (tt.lead_id);emp_id | lead_id | emp_name | salary | dept_no | ?column? | le | emp_id | lead_id | emp_name | salary | dept_no | ?column? | le 
--------+---------+----------+--------+---------+----------+----+--------+---------+----------+--------+---------+----------+----2 |       1 | jack     |  50500 | 002     | king     |  2 |      6 |       3 | wudde    |  23000 | 003     | arise    |  3
(1 row)

8、查询族兄

--查看employee id是6的节点的族兄节点
with temp as (select employee.*,prior emp_name,level lefrom employee start with lead_id=0connect by lead_id= prior emp_id
)select t.*
from temp  t
left outer join temp tton tt.emp_id=6 --此处需要条件限制
where t.le=tt.le and t.emp_id<>6; --此处需要条件限制 emp_id | lead_id | emp_name | salary | dept_no | ?column? | le 
--------+---------+----------+--------+---------+----------+----4 |       2 | scott    |  30000 | 002     | jack     |  35 |       2 | tiger    |  25000 | 002     | jack     |  37 |       3 | joker    |  21000 | 003     | arise    |  3
(3 rows)

9、level伪列的使用，格式化层级

select lpad(' ',level*2,' ')||emp_name as name,emp_id,lead_id,salary,level
from employee
start with lead_id=0
connect by prior emp_id=lead_id; -- level数值越低级别越高name     | emp_id | lead_id | salary  | level 
-------------+--------+---------+---------+-------king      |      1 |       0 | 1000000 |     1jack    |      2 |       1 |   50500 |     2scott |      4 |       2 |   30000 |     3tiger |      5 |       2 |   25000 |     3arise   |      3 |       1 |   60000 |     2wudde |      6 |       3 |   23000 |     3joker |      7 |       3 |   21000 |     3
(7 rows)

10、connect_by_root 查找根节点

select connect_by_root emp_name,emp_name,lead_id,salary
from employee
start with lead_id=1
connect by prior emp_id = lead_id;connect_by_root | emp_name | lead_id | salary 
-----------------+----------+---------+--------jack            | jack     |       1 |  50500jack            | scott    |       2 |  30000jack            | tiger    |       2 |  25000arise           | arise    |       1 |  60000arise           | wudde    |       3 |  23000arise           | joker    |       3 |  21000
(6 rows)
-- 注意： connect_by_root关键字后面跟着字段，表示根节点对应记录的某一字段的值，-- 如 connect_by_root  emp_name表示根节点的员工名，connect_by_root salary表示根节点的工资

11、connect_by_isleaf 是否是叶子节点

select emp_id,emp_name,lead_id,salary,connect_by_isleaf
from employee
start with lead_id=0
connect by nocycle prior emp_id=lead_id;emp_id | emp_name | lead_id | salary  | connect_by_isleaf 
--------+----------+---------+---------+-------------------1 | king     |       0 | 1000000 |                 02 | jack     |       1 |   50500 |                 04 | scott    |       2 |   30000 |                 15 | tiger    |       2 |   25000 |                 13 | arise    |       1 |   60000 |                 06 | wudde    |       3 |   23000 |                 17 | joker    |       3 |   21000 |                 1
(7 rows)
-- 叶节点指的是没有子节点的节点，那些是既是父节点又是子节点的节点不属于叶节点

12、使用connect by rownum生成序列

ROWNUM是一个伪列，即先查到结果集之后再加上去的一个列，它的取值从1开始排依次递增。ROWNUM其实是oracle数据库从数据文件或缓冲区中读取数据的顺序。取得第一条记录则rownum值为1，第二条为2，依次类推。
connect by rownum是通过递归迭代第一行生成一个序列。格式如下：

select ***
from dual
connect by rownum<=n;

举例：

12.1 生成1-5之间的一个序列

select rownum 
from dual 
connect by rownum<=5;rownum 
--------12345
(5 rows)

12.2 生成10个60到100之间的随机整数

select rownum No,ROUND(DBMS_RANDOM.VALUE(60,100),0) Value
from dual
connect by rownum<=10;no | value 
----+-------1 |    772 |    643 |    674 |    865 |    826 |    777 |    948 |    899 |    6110 |    93
(10 rows)

12.3 生成连续的日期值

select rownum No,sysdate+rownum MyDate
from dual
connect by rownum<=10;no |       mydate        
----+---------------------1 | 2023-10-12 15:45:392 | 2023-10-13 15:45:393 | 2023-10-14 15:45:394 | 2023-10-15 15:45:395 | 2023-10-16 15:45:396 | 2023-10-17 15:45:397 | 2023-10-18 15:45:398 | 2023-10-19 15:45:399 | 2023-10-20 15:45:3910 | 2023-10-21 15:45:39
(10 rows)