正则表达式python
一、re的正则方法简介
import re """ 小结: re.match() # 默认从行首开始匹配 re.search() # 从任意位置开始匹配。不局限于行首 re.findall() # 匹配多次,并返回匹配值的列表 re.split() # 按正则regx分割字串,并返回分割后的字串列表 re.sub() # 按正则regx找到多个子字串值a1、a2、an,并用指定字串b统一替换a1、a2、an,并返回替换后的整体新字串 说明:ret是re.match()、re.search()的返回结果 ret.group() # 获取整体匹配上的结果 ret.group(1) # 获取第1个分组的结果。就是正则regx中第1个小括号中匹配上的值 ret.group(2) # 获取第2个分组的结果。就是正则regx中第2个小括号中匹配上的值 ret.group(3) # 获取第3个分组的结果。就是正则regx中第3个小括号中匹配上的值,如果超过了分组数量,就会报错。 """ # 一、re.match() 能够匹配出以xxx(正则表达式)开头的字符串 # 1、不分组匹配。没有小括号 ret = re.match(r"H", "Hello Python") print(ret.group()) # 2、分组匹配。有小括号,一个小括号就是1个组,小组编号从1开始算。 # re.match()、re.search()都支持小括号的分组。 ret = re.match(r"([^-]*)-(\d+)", "010-12345678") print(ret.group()) # 010-12345678 print(ret.group(1)) # 010 print(ret.group(2)) # 12345678 # 非贪婪模式。在"*","?","+","{m,n}"后面加上?,使贪婪变成非贪婪。 s = "aa2343ddd" r = re.match(r"aa(\d+?)", s) print(r.group()) # 'aa2' print(r.group(1)) # '2' # 通过引用分组中匹配到的数据即可,但是要注意是元字符串,即类似 r""这种格式 ret = re.match(r"<([a-zA-Z]*)>\w*</\1>", "<html>hh</html>") print(ret.group()) # <html>hh</html> print(ret.group(1)) # html # print(ret.group(2)) # 报错。因为没有第2组的小括号,越界了。 # 二、 re.search() 不仅匹配开头 ret = re.search(r"\d+", "阅读次数为 9999") print(ret.group()) # 9999 # 三、re.findall()匹配多个值,并返回一个列表 ret_list = re.findall(r"\d+", "python = 9999, c = 7890, c++ = 12345") print(ret_list) # ['9999', '7890', '12345'] # 四、re.split()按指定字符进行分割。匹配并切割 ret_list = re.split(r":| ", "info:xiaoZhang 33 shandong") print(ret_list) # ['info', 'xiaoZhang', '33', 'shandong'] # 五、re.sub()将匹配到的值替换。匹配并替换多个值,并返回替换后的整体结果 ret_new_str = re.sub(r"\d+", '998', "python = 997 python = 997") print(ret_new_str) # python = 998 python = 998 print('ret_new_str=====: %s' % ret_new_str) print('=' * 50) print('案例:域名相关的正则匹配') # 匹配:'http://www.freebuf.com', # url = 'http://www.freebuf.com' url = 'https://freebuf.com/articles/es/123%e7%b1%b3%e9%9b%aa%e5%84%bf' url = 'http://www.freebuf.com/author/%e7%b1%b3%e9%9b%aa%e5%84%bf' # url = 'http://www.freebuf.com/157843sdf.html' # 匹配域名:https://www.freebuf.com ret = re.search(r'https?://(\w+?\.)+\w+\/?', url) print(ret.group()) # 'http://www.freebuf.com/' print(ret.group(1)) # 'freebuf.' print(ret.group()) if ret != None else print('ret = None') # 匹配文件夹:https://www.freebuf.com/articles/es url = 'https://www.freebuf.com/articles/es' ret = re.search(r'https?://(\w+?\.)+\w+(\/\w+)*(\/\w+\/?)', url) print(ret.group()) # 'https://www.freebuf.com/articles/es' print(ret.group(1)) # 'freebuf.' print(ret.group(2)) # '/articles' print(ret.group(3)) # '/es' if ret != None: print(ret.group()) # 'https://www.freebuf.com/articles/es' else: print('ret = None') # 匹配文件:http://www.freebuf.com/news/157843.html url = 'http://www.freebuf.com/news/157843.html' ret = re.search(r'https?://(\w+?\.)+\w+(\/\w+)*(\/\w+\.\w+)', url) print(ret.group()) # 'http://www.freebuf.com/news/157843.html' print(ret.group(1)) # 'freebuf.' print(ret.group(2)) # '/news' print(ret.group(3)) # '/157843.html' print(ret.group()) if ret_list != None else print('ret_list = None')
二、re的预编译方式,结合前面的正则方式使用
1、先编译一次,后面就不编译了
import re str = 'https://www.freebuf.com/page/357' pattern = re.compile(r'(p|s)\:\/\/www\.(.+?\..+?)\/+?', re.DOTALL) match = pattern.findall(str) print(type(match)) # <class 'list'> print(match) # [('s', 'freebuf.com')]
2、忽略大小写
在正则表达式前面加(?i)
import re str1 = """ 201.158.69.116 - - [03/Jan/2013:21:17:20 -0600] fwf[-] tip[-] 127.0.0.1:9000 0.007 0.007 MX pythontab.com GET /html/test.html http/1.1 " 200" 2426 "http://a.com" "es-ES,es;q=0.8" "Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.11 (KHTML, like Gecko) Chrome/23.0.1271.97 Safari/537.11" 172.16.119.8 - admin [15/Aug/2011:18:17:50 +0800] "PROPFIND /svn/EAGLE HTtP/1.1" 201 649 172.16.119.8 - admin [15/Aug/2011:18:17:50 +0800] "PROPFIND /svn/EAGLE/!svn/vcc/default HTTP/1.1" 207 401 172.16.119.8 - admin [15/Aug/2011:18:17:50 +0800] "PROPFIND /svn/EAGLE/!svn/bln/31 HTTP/1.1" 207 454 172.16.119.8 - admin [15/Aug/2011:18:17:50 +0800] "PROPFIND /svn/EAGLE HTTP/1.1" 207 649 172.16.119.8 - admin [15/Aug/2011:18:17:50 +0800] "PROPFIND /svn/EAGLE/!svn/vcc/default HTTP/1.1" 207 454 """ pattern = re.compile(r'(?i)HTTP/.+?\b[^\d]+?([2345]\d{2})', re.DOTALL) ret = pattern.findall(str1) print(ret) # ['200', '201', '207', '207', '207', '207']
3、点号默认不匹配换行,要想匹配换行,需要设置re.DOTALL
pattern = re.compile(r'(?i)HTTP/.+?\b[^\d]+?([2345]\d{2})', re.DOTALL)