python3 利用re模块实现日志数据的统计

import re

def count_something(log_file, patt):
    ori_dict = {}
    with open(log_file) as  fobj:
        for line in fobj:
            m = patt.search(line)
            if m:
                key = m.group()
                ori_dict[key] = ori_dict.get(key, 0) + 1  #字典值等于get方法，默认为0
    return ori_dict



if __name__ == '__main__':
    apatt = re.compile('(\d+\.){3}\d+')     #预编辑
    bpatt = re.compile('Firefox|MSIE')      #预编辑
    log_file = '/var/log/httpd/access_log'
    print(count_something(log_file, apatt))
    print(count_something(log_file, bpatt))

面向过程的写法

 

 

import re

class CountSomething:
    def __init__(self, file, patt):
        self.file = file
        self.patt = patt

    def __call__(self):
        ori_dict = {}
        with open(self.file) as fobj:
            for line in fobj:
                m = self.patt.search(line)
                if m:
                    key = m.group()
                    ori_dict[key] = ori_dict.get(key, 0) + 1
        return ori_dict

if __name__ == '__main__':
    log_file = '/var/log/httpd/access_log'
    apatt = re.compile('(\d+\.){3}\d+')
    bpatt = re.compile('Firefox|MSIE')
    c = CountSomething(log_file, apatt)
    print(c())
    d = CountSomething(log_file, bpatt)
    print(d())

面向对象的写法

 

 

import re
from collections import Counter

def count_thing(file, patt):
    c = Counter()
    with open(file) as fobj:
        for line in fobj:
            m = patt.search(line)
            if m:
                key = m.group()
                c.update([key])
    return c


if __name__ == '__main__':
    file = '/var/log/httpd/access_log'
    patt = re.compile('(\d+\.){3}\d+')
    print(count_thing(file, patt))

利用Counter模块，进行统计。