python—收集系统信息
1.1hashlib模块使用获取文件的MD5值，和shell下的MD5sum一样
方法一：先实例化一个对象，再使用update做校验，最后十六进制查看hexdigest
In : import hashlib
In : md5 = hashlib.md5()
In : md5.update("a")
In : md5.hexdigest()
Out: '0cc175b9c0f1b6a831c399e269772661'
In : md5.update("b\n")#叠加，实际为ab\n的值
In : md5.hexdigest()
Out: 'daa8075d6ac5ff8d0c6d4650adb4ef29'
http://blog.51cto.com/huangzp/C:/Users/test/AppData/Local/YNote/data/hzpzc68@126.com/387a555b88e548afa5f107010ce0486d/clipboard.png
说明：
对比shell，哈希值一样
# echo "ab"|md5sum
daa8075d6ac5ff8d0c6d4650adb4ef29-
与shell下一致，-n #表示不加换行符
# echo "a"|md5sum
60b725f10c9c85c70d97880dfe8191b3-
# echo -n "a"|md5sum
0cc175b9c0f1b6a831c399e269772661-
方法二：字符串短时，直接调用
In : hashlib.md5("hello").hexdigest()
Out: '5d41402abc4b2a76b9719d911017c592'
脚本：
#!/usr/bin/env python
import sys
import hashlib
import codecs
def getMd5(f):
md5 = hashlib.md5()
with codecs.open(f) as fd:
while True:
data = fd.read(4096)
if data:
md5.update(data) #每读一次，获取一次md5值，MD5值累加
else:
break
return md5.hexdigest()
if __name__ == "__main__":
try:
print getMd5(sys.argv)
except IndexError:
print ("%s follow a argument" % __file__)
运行结果：
# python 01.py 1.py
e796a8f418fa90d7e4f0a162119f114a
# cat 1.py |md5sum
e796a8f418fa90d7e4f0a162119f114a-
1.2 os.walk模块使用
os.walk()：迭代目录里文件，返回一个列表，分别是：路径、路径下的目录、路径下的文件
In : walk = os.walk("/root/python/a")
In : for i,j,k in walk:print i,j,k
/root/python/a ['b'] ['a.txt']
/root/python/a/b ['c'] ['b.txt']
/root/python/a/b/c [] ['c.txt']
脚本：
#!/usr/bin/env python
import sys
import hashlib
import codecs
import os
def getMd5(f):
md5 = hashlib.md5()
with codecs.open(f) as fd:
while True:
data = fd.read(4096)
if data:
md5.update(data)
else:
break
return md5.hexdigest()
walk = os.walk(sys.argv)
s = ""
for i,j,k in walk:
for fil in k:
fa = os.path.join(i,fil)
result = getMd5(fa)
s += result +" " +fa +"\n"
print s,
运行结果：
# python 02.py /root/python/
b9f6f7a02766b3f9bd28369c6c331218 /root/python/20.py
e796a8f418fa90d7e4f0a162119f114a /root/python/1.py
01b1b4f4b2eb7155187c27057e9a4de2 /root/python/1.pyc
1.3生成器yield
想捕获函数的值，但是又不想退出程序。yield，记住当前函数的值，用于下次调用。调用时用next()方法或者for循环遍历
修改如上脚本：
#!/usr/bin/env python
import sys
import hashlib
import os
import codecs
def getMd5(f):
md5 = hashlib.md5()
with codecs.open(f) as fd:
while True:
data = fd.read(4096)
if data:
md5.update(data)
else:
break
return md5.hexdigest()
def fileMd5(topdir):
walk = os.walk(topdir)
for i,j,k in walk:
for fil in k:
fa = os.path.join(i,fil)
result = getMd5(fa)
yield "%s %s" % (result,fa)
if __name__ == "__main__":
topdir = sys.argv
md5 = fileMd5(topdir)
for lis in md5:
print lis
1.4 文件md5值的校验
找出目录中内容相同的文件
脚本：
#!/usr/bin/env python
import sys
import hashlib
import os
import codecs
def getMd5(f):
md5 = hashlib.md5()
with codecs.open(f) as fd:
while True:
data = fd.read(4096)
if data:
md5.update(data)
else:
break
return md5.hexdigest()
def getDir(topdir):
dic = {}
walk = os.walk(topdir)
for i,j,k in walk:
for lis in k:
fil = os.path.join(i,lis)
md5 = getMd5(fil)
if dic.has_key(md5):
dic.append(fil)
else:
dic = #dic = ，字典的key为MD5，value值为fn
return dic
if __name__ == "__main__":
result = getDir(sys.argv)
for i,j in result.items():
if len(j)>1:
print i,j
运行结果：
# python 04.py /root/python
b026324c6904b2a9cb4b88d6d61c81d1 ['./test/a/1.txt', './test/b/2.txt', './test.bak/test/a/1.txt', './test.bak/test/b/2.txt']
1.5 字典排序
字典是无序的
sorted（可迭代对象，cmp，key=根据什么排序，reverse=是否倒序），返回一个排序的列表
说明：key表示关键字；operator.itemgetter(0)：根据字典的key值排序，(1)表示根据字典的value排序；reverse=True倒序
x.iteritems()返回一个序列，items()与iteritems()的区别和range()与xrange()的区别一样
找出占用空间大的文件
脚本：
#/usr/bin/env python
import sys
import os
import operator
def getDic(topdir):
dic = {}
walk = os.walk(topdir)
for i,j,k in walk:
for lis in k:
fs = os.path.join(i,lis)
sz = os.path.getsize(fs)
dic = sz
return dic
if __name__ == "__main__":
dic =getDic(sys.argv)
sorted_dic = sorted(dic.iteritems(),key=operator.itemgetter(1),reverse=True)
for k,v in sorted_dic[:10]:
if v>1024:
K = int(v)/1024
if K>1024:
M = K/1024
if M>1024:
v = str(M/1024) +"G"
else:
v = str(M) +"M"
else:
v = str(K) +"K"
print k,"------>",v
运行结果：
# python 05.py /data/program/
/data/program/mongo/data/data27018/local.2 ------> 511M
/data/program/mongo/data/data27018/local.1 ------> 511M
/data/program/mongo/data/data27017/local.1 ------> 511M
/data/program/mongo/data/data27017/local.2 ------> 511M
/data/program/mongo/logs/mongodb-27018.log ------> 488M
/data/program/mysql/lib/libmysqld.a ------> 235M
/data/program/mongo/data/data27018/journal/j._7 ------> 128M
/data/program/mysql/bin/mysqld ------> 95M
/data/program/mysql/bin/mysql_client_test_embedded ------> 87M
/data/program/mysql/bin/mysql_embedded ------> 87M
1.6 python调用外部命令
1）os.system()：输出在终端上，捕获不到
In : os.system("ls")
001.sh 10.sh 1.py
2）os.popen()：只能捕捉到标准输出，捕捉不到标准错误输出
In : a = os.popen("ls")
In : a.read
a.read a.readline
a.readintoa.readlines
3）os.popen2()：返回2个对象，一个标准输入、一个标准输出
In : sdtin,stdout = os.popen2("sort")
/usr/bin/ipython:1: DeprecationWarning: os.popen2 is deprecated.Use the subprocess module.
#!/usr/bin/python2
4）os.popen3()：返回3个对象，标准输入、标准输出、标准错误输出
In : stdin,stdout,stderr = os.popen3("ls a")
/usr/bin/ipython:1: DeprecationWarning: os.popen3 is deprecated.Use the subprocess module.
#!/usr/bin/python2
5）os.popen4：返回2个对象，pipe_in和pipe_out_err
6）subprosess
subpeocess.call('ls -l --color /root',shell=True)
说明：输出不能捕捉到，类似于os.system；不加-l，就不用shell
subprocess.check_call(['mkdir','/tmp/aaa'])
说明：会抛出python异常
In : subprocess.call("ls -l",shell=True)
total 580868
-rw-r--r--1 root root 103 Nov 22 16:23 001.sh
-rw-r--r--1 root root 375 Nov1 03:00 100.sh
-rw-r--r--1 root root47624178 Nov1 19:35 100.txt
捕捉异常
脚本：
#/usr/bin/env python
import subprocess
try:
subprocess.check_call("exit 1",shell=True)#check_call会抛出python异常
except subprocess.CalledProcessError:
pass
print "hello world"
运行结果：
# python 06.py
hello world
页:
[1]