1、拼接字符串
1)可以使用“||”来拼接字符串
select '拼接'||'字符串' as str from dual
2)通过concat()函数实现
select concat('拼接', '字符串') as str from dual
注:oracle的concat函数只支持两个参数的方法,即只能拼接两个参数,如要拼接多个参数则嵌套使用concat可实现,如:
select concat(concat('拼接', '多个'), '字符串') from dual
2、截取字符串
SUBSTR(string,start_position,[length]) 求子字符串,返回字符串
解释:string 源字符串
start_position 开始位置(从0开始)
length 可选项,子字符串的个数
select substr(to_char(sysdate, 'yyyy-mm-dd HH:mi:ss'), 12, 5) as time from dual
substr("ABCDEFG", 0); //返回:ABCDEFG,截取所有字符
substr("ABCDEFG",2); //返回:CDEFG,截取从C开始之后所有字符
substr("ABCDEFG",0, 3); //返回:ABC,截取从A开始3个字符
substr("ABCDEFG",0, 100); //返回:ABCDEFG,100虽然超出预处理的字符串最长度,但不会影响返回结果,系统按预处理字符串最大数量返回。
substr("ABCDEFG",-3); //返回:EFG,注意参数-3,为负值时表示从尾部开始算起,字符串排列位置不变。
字符串 "AAA-BBB" 截取"AAA" "BBB"
select substr('AAA-BBB',1,instr('AAA-BBB','-',-1)-1) 值 from dual;
select substr('AAA-BBB',instr('AAA-BBB','-',-1)+1) 值 from dual;
3、查找字符串
INSTR(string,subString,position,ocurrence)查找字符串位置
解释:string:源字符串
subString:要查找的子字符串
position:查找的开始位置
ocurrence:源字符串中第几次出现的子字符串
select INSTR('CORPORATE FLOOR','OR', 3, 2) as loc from dual
4、替换字符串
replace(strSource, str1, str2) 将strSource中的str1替换成str2
解析:strSource:源字符串
str1: 要替换的字符串
str2: 替换后的字符串
select '替换字符串' as oldStr, replace('替换字符串', '替换', '修改') as newStr from dual
5、长度
length(nvl(字段,''))=1
6、判断
SELECTSBXH, SFZJHM,case
when length(nvl(XM,''))=1 thenXMwhen length(nvl(XM,''))=2 then substr(XM, 0, 1)||'*'
when length(nvl(XM,''))=3 then substr(XM, 0, 1)||'**'
when length(nvl(XM,''))=4 then substr(XM, 0, 1)||'***'
else substr(XM, 0, 1)||'****'
endXM,
DJXH, SDNF, DWMC, YNSEFROMZHZS_SEWSR TWHERE YNSE > 120000.000000 AND DWMC IS NOT NULL AND SFZJHM LIKE '36%'
ORDER BY YNSE DESC