python自动化运维-基础篇之差异对比

difflib模块

该模块对字符串文件等的差异性对比，并提供HtmlDiff类来增加阅读性。

字符串比较

#!/usr/bin/env python
#author: mark
#descri:对比text1和text2的内容差别
import difflib

text1 = '''text1:
woshi shui
wobushini
hade'''
text1_lines = text1.split('\n')

text2 = '''text2:
woshi shui
wobushini
haodi
haha'''
text2_lines = text2.splitlines()

d = difflib.HtmlDiff()
print(d.make_file(text1_lines, text2_lines))

最后terminal执行 python xx.py >diff.html
结果如下：

nginx配置文件对比脚本：

#!/usr/bin/env python
#author: mark
#descri:nginx配置文件
import difflib, sys
print('--------------注意：请将文件放入到和%s一致的目录或使用绝对路径比较----------------'%sys.argv[0])
try:
    file1_name = sys.argv[1]
    file2_name = sys.argv[2]
except Exception as e:
    print('Erorr:%s'%e)
    print('Usage:python %s file1_name file2name'%sys.argv[0])
    sys.exit()

def readfile(filename):
    try:
        with open(filename, 'rb') as file_obj:
            contents = ''
            while True:
                content = file_obj.read(60).decode(encoding='utf-8') #decode解码：bytes to str.
                if not content:
                    break
                contents += content
            file_lines = contents.splitlines()
            return file_lines
    except Exception as e:
        print('Erorr:%s'%e)


def main():
    text1_file = readfile(file1_name)
    text2_file = readfile(file2_name)
    d = difflib.HtmlDiff()
    print(d.make_file(text1_file, text2_file))

if __name__ == '__main__':
    main()

运行命令：python diff_nginx.py nginx1.txt nginx2.txt > diff_nginx.html
结果如下：

文件与目录比较

>>> filecmp.cmp("/root/testFileDir/file1", "/root/testFileDir/file2") #单文件比较，默认shallow为True,shallow-->os.stat
True
>>> filecmp.cmp("/root/testFileDir/file1", "/root/testFileDir/file2", shallow=False)  #会比较文件内容
True
>>> a = filecmp.cmpfiles('/root/testFileDir/dir1', '/root/testFileDir/dir2', ['file1','file2','file3','file4']) #多文件比较，返回元祖([匹配：两目录文件内容相同]， [不匹配：匹配反之]， [错误：文件不存在])
>>> '匹配成功文件：%s; 匹配失败文件：%s; 错误文件：%s'%(a[0], a[1], a[2])
"匹配成功文件：['file1']; 匹配失败文件：['file2']; 错误文件：['file3', 'file4']"

>>> a = "/root/testFileDir/dir1"
>>> b = "/root/testFileDir/dir2"
>>> dirobj = filecmp.dircmp(a, b, ['test.py']) #忽略test.py
>>> dirobj.report() #目录的内容
diff /root/testFileDir/dir1 /root/testFileDir/dir2
Only in /root/testFileDir/dir2 : ['b']
Identical files : ['file1']
Differing files : ['file2']
Common subdirectories : ['a']

>>> dirobj.report_partial_closure() #当前目录以及一级子目录
diff /root/testFileDir/dir1 /root/testFileDir/dir2
Only in /root/testFileDir/dir2 : ['b']
Identical files : ['file1']
Differing files : ['file2']
Common subdirectories : ['a']

diff /root/testFileDir/dir1/a /root/testFileDir/dir2/a
Identical files : ['a1', 'a2', 'a3']

>>> dirobj.report_full_closure()  #当前目录以及递归
diff /root/testFileDir/dir1 /root/testFileDir/dir2
Only in /root/testFileDir/dir2 : ['b']
Identical files : ['file1']
Differing files : ['file2']
Common subdirectories : ['a']

diff /root/testFileDir/dir1/a /root/testFileDir/dir2/a
Identical files : ['a1', 'a2', 'a3']

>>> dirobj.left
'/root/testFileDir/dir1'            
>>> dirobj.right
'/root/testFileDir/dir2'
>>> 
>>> dirobj.common #当前目录共同的文件和目录
['a', 'file1', 'file2']
>>> dirobj.common_dirs
['a']
>>> dirobj.common_files
['file1', 'file2']
>>> dirobj.common_funny
[]
>>> dirobj.same_files #当前目录的匹配的文件
['file1']
>>> dirobj.diff_files
['file2']
>>> dirobj.funny_files #都存在但无法比较
[]

案例脚本