需求：对一个包含数字的字符串排序
search后参考了 http://www.cnblogs.com/m-cnblogs/archive/2012/03/30/2425938.html
截屏：
（用到了to_number(str,split 1,...) instr() replace() substr()）
TO_NUMBER(x [, format], [ nls_language ])
converts x to a NUMBER.
- x is the string that will be converted to a number.
- format, optional, is the format that will be used to convert x to a number.
- nls_language, optional, is the nls language used to convert x to a number
INSTR(string, pattern-to-find)
INSTR returns the location (beginning) of a pattern in a given string. Its simple form is:
The general syntax of INSTR is:
INSTR (string to search, search pattern [, start [,occurrence]])
The arguments within brackets ([]) are optional.
在一个字符串中查找指定的字符,返回被查找到的指定的字符的位置。
如果start的值为负数，则代表从右往左进行查找，但是位置数据仍然从左向右计算。
SUBSTR retrieves a portion of the string. The general format for this function is:
SUBSTR(string, start_at_position[, number_of_characters_to_retrieve])
从给定的字符表达式返回一个子字符串。
REPLACE(source-string, pattern-to-find, pattern-to-replace-by)
（to）‘2011-10-11’ ‘2011/10/11’ 转换
update 表1 t set t.列1=replace((select 列1from 表1 a where a.主键列=t.主键列) , '/' , '-' )
例：select replace ('111222333444','222','888') from dual;
输出为 '111888333444'
最后我的代码
pn
to_number(str,’-’)
str:
replace(source ,’-’)
source:
substr(pn,instr(pn,’-’)+1, length(pn)-1-instr(pn,’-’))