re模块 Python语言下的正则表达式

•  re模块是什么意思呢？  Regular Expression
• match 模块  re.match(pattern,string,flags=0)  从初始位置开始匹配
  • pattern  匹配值

    • 表达式	描述
      普通字符	普通字符完全匹配
      []	匹配[]中列举的字符
      .	匹配任何单个字符，但‘\n’换行符除外
      \w	匹配单词字符：字母，数字，下划线
      \W	匹配任何非单词字符
      \b	字词与非字词之间的界限
      \s	匹配单个空格字符，即 空格 换行符 返回符 制表符
      \S	匹配任何非空格字符
      \t,\n,\r	制表符，换行符，退格符
      \d	十进制数【0-9】
      \D	匹配非数字
      ^	匹配字符串的开头
      $	匹配字符串的末尾
      \	转义字符
      匹配多个字符
      *	前一个字符出现0次或者任意次
      +	前一个字符至少出现一次
      ？	前一个字符要么出现一次要么不出现
      {m}	前一个字符出现m次
      {m,n}	前一个字符从m到n次；{,n}0到n从；{m,}m到无限次
      ()	括起来更好看,利于group的处理

    • 几个例子：

    • import re
      
      pattern1 = "senior"
      pattern2 = "sen.\S\w"
      pattern3 = "sen[.abcd123]\S\w"
      
      string1 = "senior sister is really pretty!"
      string2 = "senior sister is really pretty!"
      string3 = "senior sister is really pretty!"
      
      flags1 = re.I
      
      result1 = re.match(pattern=pattern1,string=string1)
      result2 = re.match(pattern=pattern2,string=string2)
      result3 = re.match(pattern=pattern3,string=string3)
      print("这是result1:" , result1)
      print("这是result2:" , result2)
      print("这是result3:" , result3)
      
      >>>这是result1: <re.Match object; span=(0, 6), match='senior'>
      >>>这是result2: <re.Match object; span=(0, 6), match='senior'>
      >>>这是result3: None
      

    • 一个综合例子

    • import re
      
      pattern1 = "(i)\s(\w*)\s(\w*)\s(\w*)\s(\w*)"
      
      string1 = "I miss my best friend cyx who now in Beijing engineering!"
      string2 = "I find that she resonate with me less and less times than before."
      string3 = "maybe she is not as powerful and passionate as before."
      
      flags1 = re.I
      
      result1 = re.match(pattern=pattern1,string=string1,flags=flags1)
      result2 = re.match(pattern=pattern1,string=string2,flags=flags1)
      result3 = re.match(pattern=pattern1,string=string3,flags=flags1)
      
      print("这是result1:" , result1)
      print("这是result2:" , result2)
      print("这是result3:" , result3)
      
      >>>这是result1: <re.Match object; span=(0, 21), match='I miss my best friend'>
      >>>这是result2: <re.Match object; span=(0, 24), match='I find that she resonate'>
      >>>这是result3: None
      

  • string  被匹配字符串
  • flags  标志位
    •  

      标志位	含义
      flags=re.I	忽略大小写
      flags=re.L	特殊字符集，取决的电脑的编译环境
      flags=re.M	多行模式
      flags=re.S	包括换行符在内的任意字符
      flags=re.U	特殊字符集
      flags=re.X	忽略空格和#后的注释

    • import re
      
      pattern1 = "Senior"
      string1 = "senior sister is really clever!"
      flags1 = re.I
      result1 = re.match(pattern=pattern1,string=string1,flags=flags1)
      print("这是result1:" , result1)
      
      >>>这是result1: <re.Match object; span=(0, 6), match='senior'>

  • 返回值的处理
    • 紧跟着上一个综合的例子

    • >>>result1.group()
      'I miss my best friend'
      
      >>>result1.group(1)
      'I'
      
      >>>result1.group(2)
      'miss'
      
      >>>result1.group(3)
      'my'
      
      >>>result1.group(4)
      'best'
      

    • 如果把pattern的括号去掉，问题就不一样了

    • pattern1 = "(i)\s\w*\s(\w*)\s(\w*)\s(\w*)"
      ...
      ...
      ...
      >>> result1.group()
      'I miss my best friend'
      >>> result1.group(1)
      'I'
      
      >>> result1.group(2)
      'my'
      >>> result1.group(3)
      'best'
      >>> result1.groups()
      ('I', 'my', 'best', 'friend')
      

    • match 中的匹配与分组 

      • 字符	功能
        |	匹配左右任意一个表达式
        (ab)	括号中字符作为一个分组
        \num	引用分组num匹配到的字符串
        (?P<name>)	分组起别名
        (?P=name)	引用别名为name匹配到的字符串

      •  举例

      • import re
        
        pattern1 = "(.*)\s(?P<name>\w*)\s(\w*).(\w*)\s(\w*)"
        pattern2 = "(?P<name>.*)\s(\w*)\s(\w*)\s(\w*)\s(\w*)"
        pattern3 = "(Even)\s(\w*)\s(\w*)\s(\w*)\s(\w*)"
        
        string1 = "In the past,we overcome many hard times."
        string2 = "At times,I get a lot of courage from this closed friendship."
        string3 = "Even she admitted I ranked high in her few friends."
        
        flags1 = re.I
        
        result1 = re.match(pattern=pattern1,string=string1,flags=flags1)
        result2 = re.match(pattern=pattern2,string=string2,flags=flags1)
        result3 = re.match(pattern=pattern3,string=string3,flags=flags1)
        print("这是result1:" , result1)
        print("这是result2:" , result2)
        print("这是result3:" , result3)
        
        
        
        这是result1: <re.Match object; span=(0, 39), match='In the past,we overcome many hard times'>
        这是result2: <re.Match object; span=(0, 59), match='At times,I get a lot of courage from this closed >
        这是result3: <re.Match object; span=(0, 26), match='Even she admitted I ranked'>
        >>> result1.group("name")
        'many'
        >>> result1.group(2)
        'many'
        

      • 如果我们观察仔细的话，你会发现result1的group的确是有点用的
      • 如果我们观察仔细的话，你会发现 这一切都是因为Python语言自带的贪婪性，总是想匹配更多的字符

• re.search  函数
  • re.search 扫描整个字符串并返回第一个成功的匹配，或者None
  • 实例

  • import re
    
    pattern1 = "\d+"
    pattern2 = "\w*"
    pattern3 = "\w{4}"
    
    string1 = "From 2021,we become closed friends."
    string2 = "In 2022.10.1,I write this artical."
    string3 = "I have to face future and do more works in Physics."
    
    flags1 = re.I
    
    result1 = re.search(pattern=pattern1,string=string1,flags=flags1)
    result2 = re.search(pattern=pattern2,string=string2,flags=flags1)
    result3 = re.search(pattern=pattern3,string=string3,flags=flags1)
    
    print("这是result1:" , result1)
    print("这是result2:" , result2)
    print("这是result3:" , result3)
    
    这是result1: <re.Match object; span=(5, 9), match='2021'>
    这是result2: <re.Match object; span=(0, 2), match='In'>
    这是result3: <re.Match object; span=(2, 6), match='have'>
    

• re.findall  函数
  • 找到正则表达式所匹配的所有子串
  • 实例

  • import re
    
    pattern1 = "\d+"
    pattern2 = "\w*"
    pattern3 = "\w{4}"
    
    string1 = "From 2021,we become closed friends."
    string2 = "In 2022.10.1,I write this artical."
    string3 = "I have to face future and do more works in Physics."
    
    flags1 = re.I
    
    result1 = re.findall(pattern=pattern1,string=string1,flags=flags1)
    result2 = re.findall(pattern=pattern2,string=string2,flags=flags1)
    result3 = re.findall(pattern=pattern3,string=string3,flags=flags1)
    
    print("这是result1:" , result1)
    print("这是result2:" , result2)
    print("这是result3:" , result3)
    
    
    这是result1: ['2021']
    这是result2: ['In', '', '2022', '', '10', '', '1', '', 'I', '', 'write', '', 'this', '', 'artical', '', '']
    这是result3: ['have', 'face', 'futu', 'more', 'work', 'Phys']

•  re.finditer  函数
  • 与findall类似，并且作为一个迭代器返回

  • import re
    
    pattern1 = "\d+"
    pattern2 = "\w*"
    pattern3 = "\w{4}"
    
    string1 = "From 2021,we become closed friends."
    string2 = "In 2022.10.1,I write this artical."
    string3 = "I have to face future and do more works in Physics."
    
    flags1 = re.I
    
    result1 = re.finditer(pattern=pattern1,string=string1,flags=flags1)
    result2 = re.finditer(pattern=pattern2,string=string2,flags=flags1)
    result3 = re.finditer(pattern=pattern3,string=string3,flags=flags1)
    
    print("这是result1:" , result1)
    print("这是result2:" , result2)
    print("这是result3:" , result3)
    
    这是result1: <callable_iterator object at 0x0000019B7873CAC0>
    这是result2: <callable_iterator object at 0x0000019B7873C610>
    这是result3: <callable_iterator object at 0x0000019B787D26B0>
    
    >>> next(result2)
    <re.Match object; span=(0, 2), match='In'>
    >>> next(result2)
    <re.Match object; span=(2, 2), match=''>
    >>> next(result2)
    <re.Match object; span=(3, 7), match='2022'>
    >>> next(result2)
    <re.Match object; span=(7, 7), match=''>
    >>> next(result2)
    <re.Match object; span=(8, 10), match='10'>
    

• re.compile 函数
  • 生成一个正则表达式对象，包括match和search两个方法
  • 实例

• import re
  
  pattern1 = "(.*)\s(?P<name>\w*)\s(\w*).(\w*)\s(\w*)"
  pattern2 = "(?P<name>.*)\s(\w*)\s(\w*)\s(\w*)\s(\w*)"
  pattern3 = "(.*)\s(\w*)\s(\w*)\s(\w*)\s(\w*)"
  
  string1 = "I know that maybe today or tomorrow I will say goodbye and go and for always.All is over and long gone."
  string2 = "We seperate and no longer reunit.I leave a message here that I did really want do everything for her."
  string3 = "And this time,what I should do is to forsake her and never disturb her."
  
  flags1 = re.I
  
  prog1 = re.compile(pattern1,flags=flags1)
  prog2 = re.compile(pattern2,flags=flags1)
  prog3 = re.compile(pattern3,flags=flags1)
  
  result1 = prog1.match(string=string1)
  result2 = prog2.match(string=string2)
  result3 = prog3.match(string=string3)
  
  print("这是result1:" , result1)
  print("这是result2:" , result2)
  print("这是result3:" , result3)
  
  
  >>>这是result1: <re.Match object; span=(0, 102), match='I know that maybe today or tomorrow I will say go>
  >>>这是result2: <re.Match object; span=(0, 100), match='We seperate and no longer reunit.I leave a messag>
  >>>这是result3: <re.Match object; span=(0, 70), match='And this time,what I should do is to forsake her >
  

•  re.sub  函数
  • sub -> substitute 替换
  • repl -> replace
  • re.sub(pattern,repl,string,count)
• re.subn  函数
  • 返回一个元组（替换后的字符串，替换次数）
• re.split   函数
  • 切割字符串
  • re.split（pattern,string,maxsplit=0,flags=0）
• 实例

  • import re
    
    pattern1 = "\d+"
    pattern2 = "\w{4}"
    pattern3 = "\w{4}"
    
    string1 = "Ok.I do Physics from 2021 9,but like computer science.Even Feynman love striptease."
    string2 = "I hope I can do something and be someone in the future."
    string3 = "However,this may be impossible.I will tell you next time."
    
    flags1 = re.I
    
    result1 = re.sub(pattern=pattern1,repl="being admitted into SCU",string=string1,flags=flags1)
    result2 = re.subn(pattern=pattern2,repl="GAUSS",string=string2,flags=flags1)
    result3 = re.split(pattern=pattern3,string=string3,maxsplit=3,flags=flags1)
    
    print("这是result1:" , result1)
    print("这是result2:" , result2)
    print("这是result3:" , result3)
    
    这是result1: Ok.I do Physics from being admitted into SCU being admitted into SCU,but like computer science.Even Feynman love striptease.
    这是result2: ('I GAUSS I can do GAUSSGAUSSg and be GAUSSone in the GAUSSre.', 5)
    这是result3: ['', 'ver,', ' may be ', 'ssible.I will tell you next time.']