2014-09-02 04:46:28 +08:00
|
|
|
|
#!/usr/bin/env python
|
|
|
|
|
|
# -*- encoding: utf-8 -*-
|
|
|
|
|
|
"""
|
2014-10-18 10:17:00 +08:00
|
|
|
|
Topic: 简单的正则式搜索，包括分组捕获
|
|
|
|
|
|
Desc : 正则式例子
|
2014-09-02 04:46:28 +08:00
|
|
|
|
建议阅读Jeffrey E. F. Friedl编写的《精通正则表达式》
|
|
|
|
|
|
（Mastering Regular Expression）
|
|
|
|
|
|
"""
|
|
|
|
|
|
import re
|
|
|
|
|
|
|
|
|
|
|
|
__author__ = 'Xiong Neng'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def my_re():
|
|
|
|
|
|
# group示例
|
|
|
|
|
|
data = 'Thu'
|
|
|
|
|
|
patt1 = r'^(\w{3})'
|
|
|
|
|
|
m = re.match(patt1, data)
|
|
|
|
|
|
print(m.group(1))
|
|
|
|
|
|
patt2 = r'^(\w){3}'
|
|
|
|
|
|
m = re.match(patt2, data)
|
|
|
|
|
|
print(m.group(1))
|
|
|
|
|
|
|
|
|
|
|
|
# 贪婪匹配
|
|
|
|
|
|
data = "Sat Mar 21 09:20:57 2009::spiepu@ovwdmrnuw.com::1237598457-6-9"
|
|
|
|
|
|
# 获取最后的那三个连字符连起来的三个数，
|
|
|
|
|
|
# 搜索比匹配更合适，因为不在开头
|
|
|
|
|
|
patt = r'\d+-\d+-\d+'
|
|
|
|
|
|
print(re.search(patt, data).group()) # 打印出 1237598457-6-9
|
|
|
|
|
|
# 使用匹配，必须用到group
|
|
|
|
|
|
patt = r'.+(\d+-\d+-\d+)'
|
|
|
|
|
|
print(re.match(patt, data).group(1)) # 打印出 7-6-9，知道贪婪的厉害了吧。哈哈
|
|
|
|
|
|
# 接下来使用非贪婪操作符?
|
|
|
|
|
|
patt = r'.+?(\d+-\d+-\d+)'
|
|
|
|
|
|
print(re.match(patt, data).group(1)) # 打印出 1237598457-6-9
|
|
|
|
|
|
# 只获取三个数的中间那个数字：
|
|
|
|
|
|
patt = r'-(\d+)-'
|
|
|
|
|
|
print(re.search(patt, data).group()) # 打印-6-
|
|
|
|
|
|
print(re.search(patt, data).group(1)) # 打印6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def my_pattern():
|
|
|
|
|
|
# (?ims) i表示忽略大小写，m表示多行模式，s表示.可以匹配所有字符，包括换行符
|
|
|
|
|
|
s = "This is a book, And Hello World!! Hello you World?"
|
|
|
|
|
|
print(1, re.match(r'.*Hello', s).group())
|
|
|
|
|
|
# 匹配圆括号中的正则式，但丢弃匹配的子字符串
|
|
|
|
|
|
print(2, re.match(r'.*(?:Hello)', s).group())
|
|
|
|
|
|
# 分组名为haha
|
|
|
|
|
|
print(3, re.match(r'.*(?P<haha>Hello)', s).group())
|
|
|
|
|
|
# 注释，括号中内容被忽略
|
|
|
|
|
|
print(4, re.match(r'.*(?#Hello)', s).group())
|
|
|
|
|
|
# 只有在括号中的模式匹配时，才匹配前面的表达式
|
|
|
|
|
|
print(5, re.match(r'.*Hello (?=World)', s).group())
|
|
|
|
|
|
# 只有在括号中的模式不匹配时，才匹配前面的表达式
|
|
|
|
|
|
print(6, re.match(r'.*Hello (?!World)', s).group())
|
|
|
|
|
|
# 只有在括号中的模式匹配时，才匹配后面的表达式
|
|
|
|
|
|
print(7, re.match(r'.*(?<=Hello )World', s).group())
|
|
|
|
|
|
# 只有在括号中的模式不匹配时，才匹配后面的表达式
|
|
|
|
|
|
print(8, re.match(r'.*(?<!Hello )World', s).group())
|
|
|
|
|
|
# (?(id|name)ypat|npat) 检查id或name组是否存在，如果存在匹配ypat，否则npat
|
|
|
|
|
|
print(9, re.match(r'.*(Hello )?(?(1)World|Howdy)', s).group())
|
|
|
|
|
|
print(9.1, re.search(r'(Hello )?(?(1)World|Howdy)', s).group())
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def my_sample():
|
|
|
|
|
|
"""一个详细的字符串搜索、提取、替换示例"""
|
|
|
|
|
|
text = "Guido will be out of the office from 12/15/2012 - 1/3/2013"
|
|
|
|
|
|
|
|
|
|
|
|
# 日期的正则式
|
|
|
|
|
|
datepat = re.compile(r'(\d+)/(\d+)/(\d+)')
|
|
|
|
|
|
|
|
|
|
|
|
# 找到并打印所有日期
|
|
|
|
|
|
for m in datepat.finditer(text):
|
|
|
|
|
|
print(m.group())
|
|
|
|
|
|
|
|
|
|
|
|
# z找到所有日期，但是以不同格式打印
|
|
|
|
|
|
monthnames = [None, 'Jan', 'Feb', 'Mar', 'Apr', 'May',
|
|
|
|
|
|
'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
|
|
|
|
|
|
for m in datepat.finditer(text):
|
|
|
|
|
|
print("%s %s, %s" % (monthnames[int(m.group(1))], m.group(2), m.group(3)))
|
|
|
|
|
|
|
|
|
|
|
|
# 将所有日期替换为欧洲日期格式(日/月/年)
|
|
|
|
|
|
def fix_date(m):
|
|
|
|
|
|
return "%s/%s/%s" % (m.group(2), m.group(1), m.group(3))
|
|
|
|
|
|
newtext = datepat.sub(fix_date, text)
|
|
|
|
|
|
print(newtext)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
|
|
my_sample()
|
