常用的字符串函数：
SQL> select instr(\'Hello World!\',\'o\',1,1),instr(\'Hello World!\',\'o\',1,2),instr(\'Hello World!\',\'o\',6,2) from dual;
INSTR(\'HELLOWORLD!\',\'O\',1,1) INSTR(\'HELLOWORLD!\',\'O\',1,2) INSTR(\'HELLOWORLD!\',\'O\',6,2)
---------------------------- ---------------------------- ----------------------------
5 8 0
SQL> select substr(\'Hello World!\',1,4),substr(\'Hello World!\',3,4) from dual;
SUBSTR(\'HELLOWORLD!\',1,4) SUBSTR(\'HELLOWORLD!\',3,4)
------------------------- -------------------------
Hell llo
SQL> SELECT TRANSLATE(\'as f as f as bbdd\', \' afb\', \'*?@\') FROM DUAL;
TRANSLATE(\'ASFASFASBBDD\',\'AFB\'
------------------------------PS:每个字符匹配替换
?s*@*?s*@*?s*dd
SQL> SELECT REPLACE(\'as f af af as\', \'af\', \'t\') FROM DUAL;
REPLACE(\'ASFAFAFAS\',\'AF\',\'T\')
-----------------------------PS:字符串匹配替换
as f t t as
SQL> select lpad(\'aaa\',10,\'#\') from dual;
LPAD(\'AAA\',10,\'#\')(默认为空格)
------------------
#######aaa
SQL> select rpad(\'aaa\',10,\'#\') from dual;
RPAD(\'AAA\',10,\'#\')(默认为空格)
------------------
aaa#######
ltrim(字符串1,字符串2) 和rtrim(字符串1,字符串2)
PS:从字符串1的左端开始扫描,逐一去掉在字符串2中出现的字符,当遇到不是字符串2中的字符终止,返回结果集.
SQL> select ltrim(\'morning\',\'m\'),ltrim(\'morning\',\'or\') from dual;
LTRIM(\'MORNING\',\'M\') LTRIM(\'MORNING\',\'OR\')
-------------------- ---------------------
orning morning
SQL> select rtrim(\'morning\',\'m\'),rtrim(\'morning\',\'ng\') from dual;
RTRIM(\'MORNING\',\'M\') RTRIM(\'MORNING\',\'NG\')
-------------------- ---------------------
morning morni
SQL> select trim(\'a\' from \'abcaa\') from dual;
TRIM(\'A\'FROM\'ABCAA\')
--------------------PS:截取的字符集(蓝色部分)只能为一个字符
bc
SQL> select initcap(\'huang-kai-bin\') from dual;
INITCAP(\'HUANG-KAI-BIN\')
------------------------
Huang-Kai-Bin
SQL> select length(\'hello world\') from dual;
LENGTH(\'HELLOWORLD\')
--------------------
11
SQL> select upper(\'Hello World\') from dual;
UPPER(\'HELLOWORLD\')
-------------------
HELLO WORLD
SQL> select lower(\'Hello World\') from dual;
LOWER(\'HELLOWORLD\')
-------------------
hello world
SQL> select concat(\'Hello\',\' World\') from dual;
CONCAT(\'HELLO\',\'WORLD\')
-----------------------ps:效果和字符串连接符 || 相同
Hello World
写了一个简单的过程，把一个分段记录插入表中：
在这个过程中，用到了两个字符串的函数：
1.SUBSTR(目标字符串，输出的子串的起点，子串的长度)
如：select substr(ssn,1,3)|| \',\'||substr(ssn,4,2)||\',\'||substr(ssn,6,4);
输出: ssn ssn
300541117 变成 300-54-1117
2.INSTR(目标字符串，匹配的内容，搜索起点，第几个满足条件)
如：select lastname,instr(lastname,\'o\',2,1) from characters;
输出: lastname instr(lastname,\'o\',2,1)
purvis 0
kong 2
create or replace procedure p_hyj_hkb_test is
v_in_string varchar2(10000);
v_post number(4);
v_out_string varchar2(5000);
v_cn number(1);
create_table varchar2(500);
v_name varchar2(20);
v_age number(2);
v_address varchar2(100);
i number(2);
v_in_string2 varchar2(10000);
--v_post2 number(4);
begin
begin
select count(*)
into v_cn
from user_tables a
where a.table_name = \'USER_INFO\';
if v_cn = 0 then
--表不存在,建立
create_table := \'create table user_info(name varchar2(15),
age number(6),
address varchar2(100))\';
execute immediate create_table;
else
--表存在
null;
end if;
end;
begin
--把串\'黄开彬,25,福建省福州市闽清,\'写到一个表里，字段要按user_name,age,address这三个字段来保存
v_in_string := \'黄开彬,25,福建省福州市闽清县;黄依江,30,福建省福州市;小黄,35,福建省垃圾的;\';
while instr(v_in_string, \';\', 1) > 0 loop
if v_in_string is null then
goto end_loop;
end if;
v_post := instr(v_in_string, \';\', 1);
v_in_string2 := substr(v_in_string, 1, v_post - 1)||\',\';
v_in_string := substr(v_in_string, v_post + 1);
i := 1;
while instr(v_in_string2, \',\', 1) > 0 loop
if v_in_string2 is null then
goto end_loop;
end if;
v_post := instr(v_in_string2, \',\', 1);
v_out_string := substr(v_in_string2, 1, v_post - 1);
v_in_string2 := substr(v_in_string2, v_post + 1);
if i = 1 then
v_name := v_out_string;
elsif i = 2 then
v_age := v_out_string;
elsif i = 3 then
v_address := v_out_string;
end if;
i := i + 1;
end loop;
--插入表
begin
-- insert into user_info values (v_name, v_age, v_address);
create_table := \'insert into user_info values(\' || \'\'\'\' || v_name || \'\'\'\' || \',\' || \'\'\'\' ||
v_age || \'\'\'\' || \',\' || \'\'\'\' || v_address || \'\'\'\' || \')\';
execute immediate create_table;
end;
end loop;
end;
<<end_loop>>
null;
end p_hyj_hkb_test;