实例详解python正则表达式的用法

- 普通字符  
   - 元字符
     . ^ $ * + ? {} [] | ()

    # 匹配普通字符
    >>> import re
    >>> pattern = r"ab"
    >>> re.findall(pattern, "123abc")
    ['ab']
 

  - 在字符序列中选择一个
  - 常用来指定一个字符集: [abc], [0-9], [a-zA-Z]
  - 元字符在字符集中当做普通字符处理: [abc$]
  - 补集 : [^a-z]

    >>> import re
    # 字符集
    >>> pattern = "[a-z]"
    >>> re.findall(pattern, "abc")
    ['a', 'b', 'c']
    # 补集
    >>> pattern = "[^a-z]"
    >>> re.findall(pattern, "abc")
    []
    # 特殊字符
    >>> pattern = "[a^$]"
    >>> re.findall(pattern, "abc^$")
    ['a', '^', '$']

  - 匹配行首
    >>> pattern = "^a"
    >>> re.findall(pattern, "baaa")
    []
    >>> re.findall(pattern, "abbb")
    ['a'] 

  - 匹配行尾

    >>> pattern = "a$"
    >>> re.findall(pattern, "aaab")
    []
    >>> re.findall(pattern, "bbba")
    ['a']

- 取消元字符的特殊含义, 将其当成普通字符处理
  - 特殊含义
    - d  <==> [0-9] , 匹配十进制数, decimal
    - D  <==> [^0-9], 匹配非数字字符
    - s  <==> [tnrfv] , 匹配空白字符
    - S  <==> [^tnrfv]
    - w  <==> [a-zA-Z0-9_], 匹配 字母 数字 下划线
    - W  <==> [^a-zA-Z0-9_] 

- 重复次数: [0, +无穷)

    >>> pattern = r"ab*"
    >>> re.findall(pattern, "a")
    ['a']
    >>> re.findall(pattern, "ab")
    ['ab']
    >>> re.findall(pattern, "abb")
    ['abb']
    >>> re.findall(pattern, "abbbbbbbbbb")
    ['abbbbbbbbbb']

   - 重复次数: [1, +无穷)

    >>> pattern = r"ab+"
    >>> re.findall(pattern, "a")
    []
    >>> re.findall(pattern, "ab")
    ['ab']
    >>> re.findall(pattern, "abbbbbb")
    ['abbbbbb']

  - 重复次数: [0, 1] , 即 有 或 没有

    >>> pattern = r"ab?"
    >>> re.findall(pattern, "a")
    ['a']
    >>> re.findall(pattern, "ab")
    ['ab']
    >>> re.findall(pattern, "abbbbb")
    ['ab']

  - {m,n} 重复次数: [m, n]
    - {m}   重复次数: m
    - {m,}  重复次数: [m, +无穷)
    - m缺省值为0

    >>> pattern = r"d{1,3}"
    >>> re.findall(pattern, "1234")
    ['123', '4']
    >>> pattern = r"d{1,}"
    >>> re.findall(pattern, "1234")
    ['1234']
    >>> pattern = r"d{1}"
    >>> re.findall(pattern, "1234")
    ['1', '2', '3', '4']

 - re模块 提供了 一个正则表达式引擎接口,
   可以将 REstring 编译成对象

    >>> import re
    >>> telPatternString = r"d{3}"
    >>> telPattern = re.compile(telPatternString)
    >>> telPattern
    <_sre.SRE_Pattern object at 0x01806170>
    >>> telPattern.findall("1")
    []
    >>> telPattern.findall("123")
    ['123']
    >>> telPattern.findall("1234")
    ['123']
 

- 忽略大小写

    >>> import re
    >>> namePatternString = r"[a-z]{3}"  
    >>> namePattern = re.compile( namePatternString, re.IGNORECASE )
    >>> namePattern.findall("abc")
    ['abc']
    >>> namePattern.findall("abC")
    ['abC']

  - 字符串前加"r", 反斜杠就不会被任何特殊方式处理

    >>> pattern = r""
    >>> re.findall(pattern, "c:dirA")
    ['']
    >>> pattern = ""
    >>> re.findall(pattern, "c:dirA")
    Traceback (most recent call last):
      File "<stdin>", line 1, in <module>
      File "D:Pythoninstall-2.7libre.py", line 177, in findall
        return _compile(pattern, flags).findall(string)
      File "D:Pythoninstall-2.7libre.py", line 242, in _compile
        raise error, v # invalid expression
    sre_constants.error: bogus escape (end of line)

>>> import re
    >>> pattern = r"a"
    >>> re.match(pattern, "abc")
    <_sre.SRE_Match object at 0x01407FA8>
    >>> re.match(pattern, "ba")
    >>> re.match(pattern, "bac")
    >>>

    >>> import re
    >>> pattern = r"a"
    >>> re.search(pattern, "abc")
    <_sre.SRE_Match object at 0x01419330>
    >>> re.search(pattern, "bac")
    <_sre.SRE_Match object at 0x01407FA8>
    >>> re.search(pattern, "bca")
    <_sre.SRE_Match object at 0x01419330>
    >>>
 

  >>> import re
    >>> pattern = r"a"
    >>> re.findall(pattern, "abacad")
    ['a', 'a', 'a']

    >>> import re
    >>> pattern = r"[0-9]"
    >>> re.finditer(pattern, "1234")
    <callable-iterator object at 0x017FE990>
    >>> for x in re.finditer(pattern, "1234") :
    ...     print x
    ...
    <_sre.SRE_Match object at 0x01407FA8>
    <_sre.SRE_Match object at 0x01419330>
    <_sre.SRE_Match object at 0x01407FA8>
    <_sre.SRE_Match object at 0x01419330>
    >>>

- subn(pattern, repl, string, count=0, flags=0)

    >>> re.sub(r"a", "x", "abca")
    'xbcx'
    >>> re.subn(r"a", "x", "abca")
    ('xbcx', 2)

  - split(pattern, string, maxsplit=0, flags=0)

    >>> re.split("[^d]", "1999-09/19 23:34:59")
    ['1999', '09', '19', '23', '34', '59']
    >>> re.split("[^d ]", "1 + 2 + 3 - 4 * 5")
    ['1 ', ' 2 ', ' 3 ', ' 4 ', ' 5']

- 编译标识 flags
   - DOTALL/S       使匹配包括换行在内的所有字符
   - IGNORECASE/I   忽略大小写
   - LOCALE/L       本地化匹配
   - MULTILINE/M    多行匹配, 影响 ^$
   - VERBOSE/X      去除"""编写正则时的换行符
   
    # re.S
    >>> re.findall(r"a.b", "anb")
    []
    >>> re.findall(r"a.b", "anb", re.S)
    ['anb']
   
    # re.M
    >>> s = """
    ... line1: a1
    ... line2: a2
    ... line3: a3
    ... """
    >>> s
    'nline1: a1nline2: a2nline3: a3n'
    >>> re.findall(r"^line[0-9]", s)
    []
    >>> re.findall(r"^line[0-9]", s, re.M)
    ['line1', 'line2', 'line3']

    # re.X
    >>> telPatternStr = r"""
    ... d{3,4}
    ... -?
    ... d{7}
    ... """
    >>> telPatternStr
    'nd{3,4}n-?nd{7}n'
    >>> re.findall(telPatternStr, "011-1234567")
    []
    >>> re.findall(telPatternStr, "011-1234567", re.X)
    ['011-1234567']

   # 爬网址
>>> s = """
... <a href="www.baidu.com">baidu</a>
... <a href="www.sina.com.cn">sina</a>
... """
>>> print s

<a href="www.baidu.com">baidu</a>
<a href="www.sina.com.cn">sina</a>

>>> re.findall( r"<a href=".+">.+</a>", s )
['<a href="www.baidu.com">baidu</a>', '<a href="www.sina.com.cn">sina</a>']
>>> re.findall( r"<a href="(.+)">.+</a>", s )
['www.baidu.com', 'www.sina.com.cn']
>>>

    - 下载 贴吧或QQ空间中 所有图片
    - GrapPicture.py
'''''
Created on 2013-10-4
 
@site: www.jb200.com
''' 
 
import re 
import urllib 
 
# url : web site 
# return : get src code from the URL 
def getHtml(url) : 
    page = urllib.urlopen(url)  # connect to the url 
    html = page.read()          # read it 
    return html                 # return src code 
 
# html : html src code 
# return : a list of jpg URLs 
def getImg(html) : 
    reg = r'src="(http://[^s]*.jpg)" width' 
    imgRe = re.compile(reg) 
    imgUrlList = re.findall(imgRe, html) 
    return imgUrlList 
 
# url : download by this url 
# name : saved by this name in current dir  
def downByUrl(url, name) : 
    urllib.urlretrieve(url, name) 
 
################################################ 
if __name__ == "__main__" :    
    html = getHtml("http://tieba.baidu.com/p/2306540022") 
 
    imgUrlList = getImg(html) 
       
    count = 1 
    stopNum = 10 
    for imgUrl in imgUrlList : 
        print "download....", imgUrl 
        pictureName = "E:desktoppythonpy_srcjpg%s.jpg" % count 
        downByUrl(imgUrl, pictureName) 
        count+=1 
        if count > stopNum : 
            break; 
    
    print "the number of pictures =", count-1