Python学习笔记（四）——模式匹配与正则表达式

正则

创建正则表达对象

>>> import re

查找段落中的电话号码 ‘xxx-xxx-xxxx’

\d

表示一个数字字符

re.compile()

传递原始字符串

phoneNumRegex = re.compile(r'\d{3}-\d{3}-\d{4}')

匹配Regex对象

>>> import re
>>> phoneNumRegex = re.compile(r'\d{3}-\d{3}-\d{4}')
>>> mo = phoneNumRegex.search('My number is 123-456-7890')
>>> mo.group()
'123-456-7890'

>>> print(mo)
<re.Match object; span=(13, 25), match='123-456-7890'>

re.compile(r'\d{3}-\d{3}-\d{4}').search('My number is 123-456-7890').group()

正则表达式的复习

用

import re

将正则模块导入

用

re.complie()

函数创建一个Regex对象（记得是使用原始字符串）

向Regex对象的

serach（）

方法传入想查找的字符串

调用Match对象的

group()

，反会实际匹配文本的字符串

用正则表达式匹配更多模式

利用括号分组

>>> phoneNumRegex = re.compile(r'(\d{3})-(\d{3}-\d{4})')
>>> mo = phoneNumRegex.search('My number is 123-456-7890')
>>> mo.group()
'123-456-7890'
>>> mo.group(1)
'123'
>>> mo.group(2)
'456-7890'
>>> mo.group(0)
'123-456-7890'
>>> mo.groups()
('123', '456-7890')

当使用括号的时候

\(

和

\)

来表示

>>> phoneNumRegex = re.compile(r'(\(\d{3}\)) (\d{3}-\d{4})')
>>> mo = phoneNumRegex.search('My number is (123) 456-7890')
>>> mo
<re.Match object; span=(13, 27), match='(123) 456-7890'>
>>> mo.groups()
('(123)', '456-7890')

利用管道技术分组

>>> heroRegex = re.compile(r'Tom|Job')
>>> mo = heroRegex.search('Tom is better Job')
>>> mo1 = heroRegex.search('Job is better Tom')
>>> mo
<re.Match object; span=(0, 3), match='Tom'>
>>> mo.group()
'Tom'
>>> mo1.group()
'Job'

用 “?” 实现可选分配

(wo)?

表示匹配时出现零次或者一次

>>> batRegex = re.compile(r'Bar(wo)?man')
>>> temp1 = batRegex.search('Barman')
>>> temp2 = batRegex.search('Barwoman')
>>> temp1.group()
'Barman'
>>> temp2.group()
'Barwoman'
>>> temp1
<re.Match object; span=(0, 6), match='Barman'>
>>> temp2
<re.Match object; span=(0, 8), match='Barwoman'>

用“*”匹配零次或者多次

>>> batRegex = re.compile(r'Bar(wo)*man')
>>> temp1 = batRegex.search('Barman')
>>> temp2 = batRegex.search('Barwoman')
>>> temp3 = batRegex.search('Barwowowowoman')
>>> print(temp1,temp2,temp3)
<re.Match object; span=(0, 6), match='Barman'>
<re.Match object; span=(0, 8), match='Barwoman'>
<re.Match object; span=(0, 14), match='Barwowowowoman'>
>>> print(temp1.group(),temp2.group(),temp3.group())
Barman Barwoman Barwowow
4000
owoman

用“+”表示一次或者多次

>>> batRegex = re.compile(r'Bar(wo)+man')
>>> temp1 = batRegex.search('Barman')
>>> temp2 = batRegex.search('Barwoman')
>>> temp3 = batRegex.search('Barwowowowoman')
>>> print(temp1,temp2,temp3)
None
#因为temp1中对应字符串中没有匹配项，所以为 None
<re.Match object; span=(0, 8), match='Barwoman'>
<re.Match object; span=(0, 14), match='Barwowowowoman'>
>>> print(temp2.group(),temp3.group())
Barwoman Barwowowowoman

使用“ { }”匹配特定次数

#(Ha){3}
#(Ha)(Ha)(Ha)

(Ha){3,5}
((Ha)(Ha)(Ha))|(Ha)(Ha)(Ha)(Ha)|(Ha)(Ha)(Ha)(Ha)(Ha)

>>> haRegex = re.compile(r'(ha){3}')
>>> mo = haRegex.search('hahaha')
>>> mo.group()
'hahaha'
>>> mo.group(0)
'hahaha'
>>> mo.group(1)
'ha'

贪心和非贪心匹配

>>> import re
>>> greedyHaRegex = re.compile(r'(Ha){3,5}')
>>> mo1 = greedyHaRegex.search('HaHaHaHaHa')
>>> mo1.group()
'HaHaHaHaHa'
>>> greedyHaRegex = re.compile(r'(Ha){3,5}?')
>>> mo2 = greedyHaRegex.search('HaHaHaHaHa')
>>> mo2.group()
'HaHaHa'

“findall()”方法

查找全部符合项

>>> phoneNumRegex = re.compile(r'\d{3}-\d{3}-\d{4}')
>>> mo = phoneNumRegex.findall('Cell:415-555-9999 work: 212-555-1000')
>>> mo
['415-555-9999', '212-555-1000']
>>> phoneNumRegex = re.compile(r'(\d{3})-(\d{3})-(\d{4})')
>>> mo = phoneNumRegex.findall('Cell:415-555-9999 work: 212-555-1000')
>>> mo
[('415', '555', '9999'), ('212', '555', '1000')]

字符串分类

缩写字符类型	表示
\d	[0-9]
\D	[^0-9]
\w	任何字母数字或者下划线
\W	除字母数字或者下划线的任何字符
\s	空格、制表符、换行符
\S	除空格、制表符、换行符的任何字符

>>> xmasRegex = re.compile(r'\d+\s\w+')
>>> mo = xmasRegex.findall('12 dru, 11 pip, 10 lords, 5 rings')
>>> mo
['12 dru', '11 pip', '10 lords', '5 rings']

>>> vowelRegex = re.compile(r'[123abcefg]')
>>> mo = vowelRegex.findall('Hello Word 2018!')
>>> mo
['e', '2', '1']
-------------------------------------------------------
>>> vowelRegex = re.compile(r'[1-5.]')
>>> mo = vowelRegex.findall('Hello Word 2018.')
>>> mo
['2', '1', '.']
-------------------------------------------------------
>>> vowelRegex = re.compile(r'[^123abcefg]')
>>> mo = vowelRegex.findall('Hello Word 2018.')
>>> mo
['H', 'l', 'l', 'o', ' ', 'W', 'o', 'r', 'd', ' ', '0', '8', '.']

建立自己的字符分类

>>> vowelRegex = re.compile(r'[123abcefg]')
>>> mo = vowelRegex.findall('Hello Word 2018!')
>>> mo
['e', '2', '1']
>>> vowelRegex = re.compile(r'[1-5.]')
>>> mo = vowelRegex.findall('Hello Word 2018.')
>>> mo
['2', '1', '.']

在

[]

的前端加上^,代表取反，除[]内字符都可以匹配

>>> vowelRegex = re.compile(r'[^123abcefg]')
>>> mo = vowelRegex.findall('Hello Word 2018.')
>>> mo
['H', 'l', 'l', 'o', ' ', 'W', 'o', 'r', 'd', ' ', '0', '8', '.']

插入字符和美元字符

与取反不同，这里的

(r'^xxxx')

是代表匹配的字符串需要以特定的字符串开头

>>> begRegex = re.compile(r'^hello')
>>> mo = begRegex.search('hello world')
>>> mo
<re.Match object; span=(0, 5), match='hello'>
>>> mo.group()
'hello'

如果没有匹配成功的话，返回的是

None

>>> begRegex = re.compile(r'^hello')
>>> mo = begRegex.search('hi world')
>>> mo == None
True

当匹配为

(r'world$')

是代表匹配的字符串需要以特定的字符串结尾

>>> begRegex = re.compile(r'world$')
>>> mo = begRegex.search('hello world')
>>> mo.group()
'world'

当

(r'^\d+$')

是代表整个字符串都是数字

>>> begRegex = re.compile(r'^\d+$')
>>> mo = begRegex.search('123hello world')
>>> mo
>>> mo == None
True

通配字符

.(句点)

匹配换行之外的任何字符

>>> atRegex = re.compile(r'.at')
>>> atRegex.findall('The cat in the hat sat on the flat mat.')
['cat', 'hat', 'sat', 'lat', 'mat']

使用“.*”匹配所有的字符

>>> nameRegex = re.compile(r'First Name:(.*) Last Name:(.*)')
>>> mo = nameRegex.search('First Name:AL Last Name:Sweigart')
>>> mo.groups()
('AL', 'Sweigart')
>>> mo
<re.Match object; span=(0, 32), match='First Name:AL Last Name:Sweigart'>

使用句点字符匹配换行“re.DOTALL”

>>> newLineREgex = re.compile(r'.*',re.DOTALL)
>>> mo = newLineREgex.search("Hi:\nYes,I can do It.\nThank you.")
>>> print(mo.group())

Hi:
Yes,I can do It.
Thank you.

>>> newLineREgex = re.compile(r'.*')
>>> mo = newLineREgex.search("Hi:\nYes,I can do It.\nThank you.")
>>> print(mo.group())

Hi:

不区分大小写的匹配”re.I or re.IGNORECASE”

>>> robocop = re.compile(r'robocop',re.I)
>>> mo = robocop.findall('RobocoproBOCOPRRROBOCOP')
>>> print(mo)
['Robocop', 'roBOCOP', 'ROBOCOP']

使用sub()方法替换字符串

>>> namesRegex = re.compile(r'Agent \w+')
>>> namesRegex.sub('CENSORED','Agent Alice gave the secret documents to Agent Bob')
'CENSORED gave the secret documents to CENSORED'

保留前一个字符

>>> namesRegex = re.compile(r'Agent (\w)\w+')
>>> namesRegex.sub(r'\1****','Agent Alice told Agent Carol that Agent ')
'A**** told C**** that Agent '

如果我想保留三个，我可以这么做

>>> namesRegex = re.compile(r'Agent (\w{3})\w+')
>>> namesRegex.sub(r'\1****','Agent Alice told Agent Carol that Agent ')
'Ali**** told Car**** that Agent '

>>> namesRegex = re.compile(r'(\d{3})\d{4}(\d{4})')
>>> namesRegex.findall(' tell:15735184252;call:13835213493')
[('157', '4252'), ('138', '3493')]
>
b997
;>> namesRegex.sub(r'\1****\2',' tell:15735184252;call:13835213493')
' tell:157****4252;call:138****3493'

管理复杂的正则表达式

>>> namesRegex = re.compile(r'''
(\d{3}|\(\d{3}\))?       #area code
(\s|-|\.)?               #separator
\d{4}                    #fist 4 digits
(\s|-|\.)?               #separator
\d{4}                    #end 4 digits)''',re.VERBOSE)

组合使用

re.DOTALL

、

re.VERBOSE

、

re.I

利用管道技术，解决只有一个值作为第二参数

>>> someRegex = re.compile('foo',re.IGNORECASE|re.VERBOSE|re.DOTALL)

小项目——电话号码和email提取程序

#! python3
import pyperclip,re
mailRegex = re.compile(r'''(
[a-zA-Z0-9_%+-]+   #username
@                 #
[a-zA-Z0-9.-]+    #domain name
(\.[a-zA-Z]{2,4})    #
)''',re.VERBOSE)
phoneRegex = re.compile(r'''
\(?(\d{3})\)?
(\s|-|\.)?
(\d{3})
(\s|-|\.)
(\d{4})''',re.VERBOSE)

text = str(pyperclip.paste())
matches = []
#123-456-1234
for group in phoneRegex.findall(text):
phoneNum = '-'.join([group[0],group[2],group[4]])
matches.append(phoneNum)
for group in mailRegex.findall(text):
matches.append(group[0])

if(len(matches) > 0):
pyperclip.copy('\n'.join(matches))
print('Copied to clipboard:')
print('\n'.join(matches))
else:
print("No found")