前言
DECODE()函数,它将输入数值与函数中的参数列表相比较,根据输入值返回一个对应值。函数的参数列表是由若干数值及其对应结果值组成的若干序偶形式。当然,如果未能与任何一个实参序偶匹配成功,则函数也有默认的返回值。
语法结构如下:
decode (expr比较表达式和搜索字,如果匹配,返回结果;如果不匹配,返回default值;如果未定义default值,则返回空值。ession, search_1, result_1) decode (expr ession, search_1, result_1, search_2, result_2) decode (expr ession, search_1, result_1, search_2, result_2, ...., search_n, result_n) decode (expr ession, search_1, result_1, default) decode (expr ession, search_1, result_1, search_2, result_2, default) decode (expr ession, search_1, result_1, search_2, result_2, ...., search_n, result_n, default)
实例使用
1. 创建一个学生成绩表。栏位有: 学生名, 科目名和分数
create table student_score( name varchar2(30), subject varchar2(20), score number(4,1) );
2. 插入测试数据
insert into student_score (name,subject,score)values(‘zhang san‘,‘Chinese‘,90); insert into student_score (name,subject,score)values(‘zhang san‘,‘Mathematics‘,80); insert into student_score (name,subject,score)values(‘zhang san‘,‘English‘,79);
3.
测试一:
select name,subject,decode(subject, ‘Chinese‘,score,0) from student_score;结果如下:
如果是中文课程的话, 显示分数, 其他课程分数为零。
这条SQL 看上去使用意义不大。
测试二:
select name,sum(decode(subject, ‘Chinese‘,score,0)) as CHINESE from student_score group by name;
统计中文课程的分数。看上去有点意义。
总体看来, decode 的使用看上去和case when 有点类似。如果只是用作以上两种状况,看上去意义不大。
select name,sum(decode(subject, ‘Chinese‘,score,0)) as CHINESE from student_score group by name; select name,score as CHINESE from student_score;使用的两句使用后的效果一样,看上去使用decode 多此一举。
行转列-有意义的使用
往以上table 再插入一些其他学生的成绩:
insert into student_score (name,subject,score)values(‘li shi‘,‘Chinese‘,96); insert into student_score (name,subject,score)values(‘li shi‘,‘Mathematics‘,86); insert into student_score (name,subject,score)values(‘li shi‘,‘English‘,76); insert into student_score (name,subject,score)values(‘wang wu‘,‘Chinese‘,92); insert into student_score (name,subject,score)values(‘wang wu‘,‘Mathematics‘,82); insert into student_score (name,subject,score)values(‘wang wu‘,‘English‘,72);
使用以下SQL:
select name, sum(decode(subject, ‘Chinese‘, nvl(score, 0), 0)) "Chinese", sum(decode(subject, ‘Mathematics‘, nvl(score, 0), 0)) "Mathematics", sum(decode(subject, ‘English‘, nvl(score, 0), 0)) "English" from student_score group by name;
返回:
将行的数据转化为列, 是不是很有意义了。
使用case then 也可以达到相同的效果。
select name, sum(case when subject=‘Chinese‘ then nvl(score,0) else 0 end) "Chinese", sum(case when subject=‘Mathematics‘ then nvl(score,0) else 0 end) "Mathematics", sum(case when subject=‘English‘ then nvl(score,0) else 0 end) "English" from student_score group by name;