1.查询语句的使用
使用 select语句和子查询(subquery)可以从一个或多个表,视图,实体试图中返回数据.
1.1相关子查询
可以将子查询(as subquery)或in或exists当成where的一个条件的一部分,这样的查询称为子查询
.where中可以包含一个select语句的子查询
.where中可以包含in,exists语句
.最多可以嵌套16层
.层次过多会影响性能
[例]简单子查询实例
查询是否有的专家既以研究所的名义来申请基金项目,又以大学系为单位申请项目
(按规定只能以一个单位来申请)
SQL> create table univ_subject
2(
3 name varchar2(12) not null,
4 per_idnumber not null,
5dept_name varchar2(20)
6);
SQL> insert into univ_subjectvalues('gaoqianjing',1001,'信息工程系');
SQL> insert into univ_subjectvalues('wangbing',1002,'物理系');
SQL> insert into univ_subjectvalues('liming',1003,'化学系');
===============
SQL> create tablecolle_subject
2 (
3colle_namevarchar2(20),
4per_idnumber
5 );
SQL> insert into colle_subject values('电子研究所',1001);
SQL>insert into colle_subject values('物理研究所',1005);
================
SQL> select name,per_id,dept_name from univ_subject where per_id in
2(select per_id from colle_subject);
NAMEPER_ID DEPT_NAME
--------------------- --------------------
gaoqianjing1001 信息工程系
1.2外连接
[例]外连接实例
招生中所有学生的信息放在students表中,而部分有特长的学生在另一个表中stuent_skill中同样有该学生的信息。现在要全部列出所有学生,如果某个学生在表student_skill中就有其特长信息,并显示特长信息,如果某个学生没有特长就显示特长问空.
SQL>create table students
2(
3 st_idvarchar2(20),
4 namevarchar2(10),
5 agenumber(2),
6 tol_score number(3)
7) ;
SQL> insert into students values('973231','wangbindu',22,501);
SQL> insert into students values('973232','zhuzhijing',21,538);
SQL>insert into students values('973233','gaojing',21,576);
===================
SQL>create table student_skill
2 (
3st_idvarchar2(20),
4skillvarchar2(20)
5);
SQL>insert into student_skill values('973231','篮球');
SQL>insert into student_skill(st_id) values('973232');
SQL>insert into student_skill values('973233','足球');
===================
SQL> select a.* , b.skill from students a,student_skill b where a.st_id=b.st_id(+)
order by a.st_id;
ST_IDNAME AGE TOL_SCORE SKILL
-------------------- ---------- --------- --------- ----------
973231 wangbindu22 501篮球
973232 zhuzhijing 21 538
973233 gaojing21 576足球
1.3自我连接
自我连接是在同一个表或视图内进行条件连接.
[例]自我连接实例
查询每个雇员的名字和该雇员的经理的名字:
SQL> select e1.ename||' work for '||e2.ename "Employees and their Managers"
2fromscott.emp e1,scott.emp e2 where e1.mgr=e2.empno;
Employees and their Managers
-------------------------------------------------
SMITHwork for FORD
ALLEN work for BLAKE
WARDwork for BLAKE
JONES work for KING
MARTIN work for BLAKE
BLAKE work for KING
CLARK work for KING
SCOTTwork for JONES
TURNERwork for BLAKE
ADAMS work for SCOTT
JAMESwork for BLAKE
FORDwork for JONES
MILLER work for CLARK
1.4UNION , INTERSECT及 MINUS
UNION:可以将两个以上的表的相类似的查询结果放在一起 (union all则表示返回所有的行)
具体语法:
select ...
union[all]
select...
==========
INTERSECT:返回两个表中相同的信息
具体语法:
select ...
intersect
select...
==========
MINUS:返回一个表中出现的信息
具体语法:
select ...
minus
select...
[例1]UNION操作实例
SQL> selectst_idfrom students
2union
3selectst_idfrom student_skill;
ST_ID
--------------------
973231
973232
973233
[例2]INTERSECT操作实例
列出有特长的学生的学号
SQL> select st_id from students
2intersect
3select st_id from student_skill;
ST_ID
--------------------
973231
973233
[例3]MINUS操作实例
列出没有特长学生的学号
select st_id from students
minus
select st_id from student_skill;
ST_ID
--------------------
973232
2.创建复杂的视图
许多应用系统有统计等功能,建议最好把这些复杂语句写成视图.下面是几个常用的视图.
2.1分组视图
[例1]简单的分组视图
SQL> create or replace view dept_tot as
2select a.dname dept,sum(b.sal) total_sal from scott.dept a,scott.emp b
3where a.deptno=b.deptno group by a.dname;
查看已建立。
SQL> select * from dept_tot;
DEPTTOTAL_SAL
-----------------------
ACCOUNTING 8750
RESEARCH 10875
SALES9400
[例2]带复杂函数视图
SQL> create or replace view itemtot as
2select persion,sum(amount) itemtot from ledger
3where actiondate between
4to_date('01-MAR-1901','dd-mon-yyyy') and
5to_date('31-MAR-1901','dd-mon-yyyy')
6and action in('bought','raid') group by persion;
2.2合计视图
[例]合计函数视图实例
SQL> create or replace view emp_no1as
2select deptno,sum(sal) 工资和,sum(comm) 总和
3from scott.emp group by deptno;
SQL> select * from emp_no1;
DEPTNO工资和总和
-------- --------- ---------
108750
20 10875
3094002200
2.3组合视图
[例]带组合函数的视图
SQL> create or replace view byitem as
2select l.persion persion.item, amount, 100*amount/item bypersion,100*amount/total bytotal
3from ledgent l,itemtotal i,total where l.persion=i.persion where l.persion=i.persion
4and actiondate between
5to_date('01-MAR-1901','dd-mon-yyyy') and
6to_date('31-MAR-1901','dd-mon-yyyy')
7 and action in('bought','raid') ;
3.家族树
语法:
select column from table_name start with column=value
connect by prior 父主键=子主键
3.1排除单一性和分枝
以ORACLE中的EMP表为例
[例]从顶到底列出各雇员的信息
SQL> select lpad(' ',4*(level-1))||ename name,empno,mgr from emp start with mgr is null
2connect by prior empno=mgr;
NAME EMPNO MGR
---------------------------
KING 7839
JONES75667839
SCOTT77887566
ADAMS78767788
3.2遍历至根
[例1]现在要从某个雇员开始向他的上级列出该雇员的层次结构
SQL> col ename for a30;
SQL> select lpad(' ',4*(level-1))||ename ename,mgr,empno from scott.emp
2start with mgr=7788 connect by prior mgr=empno;
ENAMEMGR EMPNO
--------------------------------------------
ADAMS 77887876
SCOTT 75667788
JONES 78397566
KING7839
[例2]列出所有雇员的层次结构
SQL> select lpad(' ',4*(level-1))||ename ename,empno,mgr from scott.emp
2start with mgr is not null
3connect by empno=prior mgr;
ENAMEEMPNO MGR
------------------ --------- ---------
SMITH 7369 7902
FORD7902 7566
JONES 7566 7839
KING7839
ALLEN 7499 7698
BLAKE 7698 7839
KING7839
WARD7521 7698
BLAKE 7698 7839
KING7839
JONES 7566 7839
KING7839
MARTIN7654 7698
BLAKE 7698 7839
KING7839
BLAKE 7698 7839
KING7839
CLARK 7782 7839
KING7839
SCOTT 77
