Question

python正则表达式匹配字符串的问题？

import re
txt ="bn','dd_ff');xm_a([1,'zhangming',0,2,'yuwenjige','lishijige',1,0,3,1,0,0.999,1,1,0,0]);xm_a([2,'wangmeng',0,2,'shuxuejige','dilijige',1,0,3,1,0,0.999,1,1,1,0]);xm_a([3,'wangli',0,2,'shuxuejige','dilijige',1,0,2,0,0,0.999,1,1,1,0]);zuobi(6,3,'4:5');fg_gh('xxx','vb_4');xm_a([4,'dashan',0,2,'huaxuejige','yingyujige',1,0,3,1,1,0.999,0,1,0,0]);</"
jige = re.findall("???怎么写？？？", txt)
print jige
>>>["zhangming", "wangmeng", "dashan"]

txt是有规律的，每句以xm_a([1开始，分号结束。用正则匹配xm_a([3,'wangli'里面的姓名，但xm_a([3,'wangli...;后面紧跟着这种特征的字符串zuobi(6,3,'4:5');则不匹配前面的wangli，正则应该怎么写？

Answer 1

完整代码： #!/usr/bin/python# -*- coding: utf-8 -*-
"""
Function:
python正则表达式匹配字符串的问题？

Author: Crifan LiVersion: 2012-12-08
Contact: admin at crifan dot com
"""

import re;
txt ="bn','dd_ff');xm_a([1,'zhangming',0,2,'yuwenjige','lishijige',1,0,3,1,0,0.999,1,1,0,0]);xm_a([2,'wangmeng',0,2,'shuxuejige','dilijige',1,0,3,1,0,0.999,1,1,1,0]);xm_a([3,'wangli',0,2,'shuxuejige','dilijige',1,0,2,0,0,0.999,1,1,1,0]);zuobi(6,3,'4:5');fg_gh('xxx','vb_4');xm_a([4,'dashan',0,2,'huaxuejige','yingyujige',1,0,3,1,1,0.999,0,1,0,0]);</";
#注：此处用到了 前向否定匹配#不了解的，可以参考：
#【教程】详解Python正则表达式之： (?!…) negative lookahead assertion 前向否定匹配 /前向否定断言
jige = re.findall("xm_a\(\[\d+,'(\w+)',[^\(\)\[\]]+\]\);(?!zuobi)", txt);
print jige; #['zhangming', 'wangmeng', 'dashan']

另外，感兴趣的，可以去看我写的系列的教程：
【教程】详解Python正则表达式

(此处不给贴地址，请自己用google搜标题，即可找到帖子地址)

Answer 2

jige=[]
allmatch=findall("(?isu)xm_a\(\[\d+,'([^']+)?'),txt)
zuobi=findall("(?isu)xm_a\(\[\d+,'([^']+)?'.*?\);zuobi),txt)
if allmatch:
for name in allmatch:
if zuobi and name in zuobi:continue
jige.append(name)
print jige

追问

zuobi这句正则有问题吧？我试了一下结果不对
...
allmatch=re.findall("(?isu)xm_a\(\[\d+,'([^']+)?'",txt)
zuobi=re.findall("(?isu)xm_a\(\[\d+,'([^']+)?'.*?\);zuobi",txt)
if allmatch:
for name in allmatch:
if zuobi and name in zuobi:continue
jige.append(name)
print jige
>>>['wangmeng', 'wangli', 'dashan']

应该是>>>["zhangming", "wangmeng", "dashan"]

追答

#python2.7
import re
txt ="bn','dd_ff');xm_a([1,'zhangming',0,2,'yuwenjige','lishijige',1,0,3,1,0,0.999,1,1,0,0]);"+\
"xm_a([2,'wangmeng',0,2,'shuxuejige','dilijige',1,0,3,1,0,0.999,1,1,1,0]);"+\
"xm_a([3,'wangli',0,2,'shuxuejige','dilijige',1,0,2,0,0,0.999,1,1,1,0]);zuobi(6,3,'4:5');fg_gh('xxx','vb_4');"+\
"xm_a([4,'dashan',0,2,'huaxuejige','yingyujige',1,0,3,1,1,0.999,0,1,0,0]);</";
jige=[]
allmatch=re.findall("(?isu)xm_a\(\[\d+,'([^']+)?'",txt)
zuobi=re.findall("(?isu)xm_a\(\[\d+,'([^']+)?'[^\)]*?\);zuobi",txt)
if allmatch:
for name in allmatch:
print name
if zuobi and name in zuobi:continue
jige.append(name)
print jige

刚刚调试了一下。

追问

能用了，不过有提前写好的我给确认了，还是谢谢你