文章出處

Python第十二天收集主機信息正則表達式正則表達式無名分組有名分組

收集主機信息
1. 主機名：hostname
2. IP地址：ip
3. 操作系統版本：osver
4. 服務器廠商：vendor
5. 服務器型號：product
6. 服務器序列號：sn
7. cpu型號：cpu_model
8. cpu核數：cpu_num
9. 內存大小：memory

python中，platform模塊給我們提供了很多方法去獲取操作系統的信息

示例1
#!/usr/bin/env python

from subprocess import Popen, PIPE

p = Popen(['dmidecode'], stdout=PIPE)
data = p.stdout # 從標準輸出讀取 返回一個文件對象
lines = []
dmi = {}
a = True
while a:
line = data.readline()
if line.startswith('System Information'):
while True:
line = data.readline()
if line == '\n': # 如果讀到空行，那么結束讀取，整行就有一個換行符就是空行\n
a = False # 退出外層循環
break
else:
lines.append(line)
dmi_dic = dict([i.strip().split(':') for i in lines]) # 將列表變為字典
dmi['Manufacturer'] = dmi_dic['Manufacturer'].strip()
dmi['Product'] = dmi_dic['Product Name'].strip()
dmi['Serial'] = dmi_dic['Serial Number'].strip()
print dmi

示例2
#!/usr/bin/env python

from subprocess import Popen, PIPE

def getDmi():
p = Popen(['dmidecode'], stdout=PIPE)
data = p.stdout.read()
return data

def parseDmi(data):
lines = []
line_in = False
dmi_list = [i for i in data.split('\n') if i]
for line in dmi_list:
if line.startswith('System Information'):
line_in = True
continue
if line_in:
if not line[0].strip():
lines.append(line)
else:
break
return lines

def dmiDic():
dmi_dic = {}
data = getDmi()
lines = parseDmi(data)
dic = dict([i.strip().split(': ') for i in lines])
dmi_dic['vendor'] = dic['Manufacturer']
dmi_dic['product'] = dic['Product Name']
dmi_dic['sn'] = dic['Serial Number']
return dmi_dic

if __name__ == '__main__':
print dmiDic()

在Python里，以下這些對象相當于布爾值的False
空列表([] )
空元組(() )
空字典({} )
空字符串('' )
零值(0 )
特殊對象None
對象False

獲取網卡信息示例1
#!/usr/bin/env python

from subprocess import Popen, PIPE

def getIfconfig():
p = Popen(['ifconfig'], stdout=PIPE)
data = p.stdout.read().split('\n\n')
return [i for i in data if i and not i.startswith('lo')]

def parseIfconfig(data):
dic = {}
for lines in data:
line_list = lines.split('\n')
devname = line_list[0].split()[0]
macaddr = line_list[0].split()[-1]
ipaddr = line_list[1].split()[1].split(':')[1]
dic[devname] = [ipaddr, macaddr]
return dic

if __name__ == '__main__':
data = getIfconfig()
print parseIfconfig(data)

獲取網卡信息示例2
#!/usr/bin/env python

from subprocess import Popen, PIPE

def getIP():
p = Popen(['ifconfig'], stdout=PIPE, stderr=PIPE)
stdout, stderr = p.communicate()
data = [i for i in stdout.split('\n') if i]
return data

def genIP(data):
new_line = ''
lines = []
for line in data:
if line[0].strip():
lines.append(new_line)
new_line = line + '\n'
else:
new_line += line + '\n'
lines.append(new_line)
return [i for i in lines if i and not i.startswith('lo')]

def parseIfconfig(data):
dic = {}
for lines in data:
line_list = lines.split('\n')
devname = line_list[0].split()[0]
macaddr = line_list[0].split()[-1]
ipaddr = line_list[1].split()[1].split(':')[1]
dic[devname] = [ipaddr, macaddr]
return dic

if __name__ == '__main__':
data = getIP()
data_list = genIP(data)
print parseIfconfig(data_list)

#!/usr/bin/env python

import re
from subprocess import Popen, PIPE

def getIfconfig():
p = Popen(['ifconfig'], stdout=PIPE)
data = p.stdout.read().split('\n\n')
return [i for i in data if i and not i.startswith('lo')]

def parseIfconfig(data):
re_devname = re.compile(r'(br|eth|em|virbr|lo|bond)[\d:]+',re.M)
re_mac = re.compile(r'HWaddr ([0-9A-F:]{17})', re.M)
re_ip = re.compile(r'inet addr:([\d\.]{7,15})', re.M)
devname = re_devname.search(data)
if devname:
devname = devname.group()
else:
devname = ''
mac = re_mac.search(data)
if mac:
mac = mac.group(1)
else:
mac = ''
ip = re_ip.search(data)
if ip:
ip = ip.group(1)
else:
ip = ''
return {devname: [ip, mac]}

if __name__ == '__main__':
dic = {}
data = getIfconfig()
for i in data:
dic.update(parseIfconfig(i))
print dic

---------------------------------------------------------------

正則表達式

re模塊（標準庫里面）
分析日志、提取信息
普通字符
元字符： .、^、$、*、+、?、{}、[]、\ 轉義、|、()

\d: 匹配任何十進制數，相當于[0-9]
\D: 匹配任何非數字字符，相當于[^0-9]
\s: 匹配任何空白字符，相當于[\t\n\r\f\v]
\S: 匹配任何非空白字符，相當于[^\t\n\r\f\v]
\w: 匹配任何字母數字字符，相當于[a-zA-Z0-9]，包括下劃線等特殊符號
\W: 匹配任何非字母數字字符，相當于[^a-zA-Z0-9]

findall(pattern, string, flags=0)
flags：指定匹配的模式，比如re.S匹配空字符
re.findall(r'.','ab\nss',re.S)
['a', 'b', '\n', 's', 's']

正則方法
re.findall()方法：找到re匹配的所有子串，并把它作為一個列表返回，如果用到()元字符只返回()里的內容這時候要用search的group方法
re.match()方法：返回一個對象，只匹配開始的位置
text = ‘hello world ’
m = re.match(r’(\w+)\s’, text)
m.group() 等價于 m.group(0) 整行數據
m.group(1) 打印第一組
按照括號分的，第一個括號是group(1)，第二個括號是group(2)
re.sub()方法：替換
re.sub('he' ,'HE' ,'hellohe',count=2)
re.sub(old ,'new ,字符串,替換的次數)

re.search()方法:從左到右掃描字符串，直到找到第一個re匹配的字符串就停止，不會再匹配后面的即使后面也有匹配字符串
re.match與re.search的區別：re.match只匹配字符串的開始，如果成功的話，返回MatchObject實例，
如果字符串開始不符合正則表達式，則匹配失敗，函數返回None；
而re.search匹配整個字符串，直到找到一個匹配，如果成功的話，返回MatchObject實例

compile()：將正則表達式編譯成對象，并用他們來進行匹配，compile(正則表達式)
p = re.compile(r'ab*')
p = re.compile(r'ab*',re.I|re.S|re.M)
p.findall('G\nd')
>>> import re
>>> p = re.compile(r'ab*') #正則表達式
>>> p
<_sre.SRE_Pattern object at 0x7fa2c0deb960>
s = ‘abc’
p.findall(s) # 匹配ab* 正則
或：
reg = r’ab*’
re.findall(reg, s)

標志flags
re.I 對應2
re.IGNORECASE 忽略大小寫
re.L 對應4
re.LOCALE
re.M 對應8
re.MULTILINE 多行匹配
re.S 對應16 使.匹配包括換行符在內的所有字符
re.Scanner 包括換行符
re.T 對應1
re.TEMPLATE 模板匹配
re.U 對應32
re.UNICODE 包含unicode字符
re.X 對應64 正則可以使用多行來寫

re.S和re.I
s = 'god GOD G\nd'
re.findall(r'g.d',s ,re.S|re.I)

re.M
data = open('/etc/passwd').read()
ftp = re.compile(r'(^ftp)',re.M)
ftp.findall(data)
a = ftp.search(data)
a.group()

示例1
>>> smail = "root@123.com"
>>> rmail = r'[\w_]+@\w+\.[a-zA-Z]{2,3}‘
>>> re.findall(rmail, smail)

>>> rmail = r"""[0-9a-zA-Z_]+
... @
... [0-9a-zA-Z]+
... \.
... [a-zA-Z]{2,3}
... ""“
>>> re.findall(rmail,smail,re.X)
re.X：正則可以使用多行來寫

greedy模式 greedy貪婪貪婪匹配
?：抑制貪婪
>>> ss = '<h1> hello </h1>‘
>>> re.findall(r'<.*>', ss)
['<h1> hello </h1>']

>>> re.findall(r'<.*?>', ss)
['<h1>', '</h1>']

syslog reg
reg_syslog = re.compile(r'\w+ \d+ [\d:]+ [\w\d.]+ \w+(\[\d+\])?: .*')
ss = 'Nov 21 14:52:23 localhost dhclient[1005]: bound to 192.168.3.163 -- renewal in 2843 seconds.'
s = reg_syslog.search(ss) # 不能用findall，因為用到了()元字符
s.group()

無名分組：沒有定義別名的分組，訪問的時候使用數字來訪問，s.group(1)

有名分組：
(?P<name>...)
REG_LOG = re.compile(r'(?P<logtime>\w+ \d+ [\d:]+) (?P<hostname>[\w\d.]+) (?P<programe>\w+(\[\w+\])?:) (?P<msg>.*)')
s = REG_LOG.search(ss)
s.groupdict() #返回一個字典

/var/log/message有四個部分
1、時間戳
2、主機名，由哪臺主機產生的日志，有些主機名是用FQDN的形式
3、進程名，一般包含pid
4、具體日志信息

# 收集主機信息
# collect_info.py


#!/usr/bin/env python

import urllib, urllib2
from subprocess import Popen, PIPE
import pickle

def getIfconfig():
p = Popen(['ifconfig'], stdout=PIPE)
data = p.stdout.read()
return data

def getDmi():
p = Popen(['dmidecode'], stdout=PIPE)
data = p.stdout.read()
return data

def parseData(data):
parsed_data = []
new_line = ''
data = [i for i in data.split('\n') if i]
for line in data:
if line[0].strip():
parsed_data.append(new_line)
new_line = line+'\n'
else:
new_line += line+'\n'
parsed_data.append(new_line)
return [i for i in parsed_data if i]

def parseIfconfig(parsed_data):
dic = {}
parsed_data = [i for i in parsed_data if not i.startswith('lo')]
for lines in parsed_data:
line_list = lines.split('\n')
devname = line_list[0].split()[0]
macaddr = line_list[0].split()[-1]
ipaddr = line_list[1].split()[1].split(':')[1]
break
dic['ip'] = ipaddr
return dic

def parseDmi(parsed_data):
dic = {}
parsed_data = [i for i in parsed_data if i.startswith('System Information')]
parsed_data = [i for i in parsed_data[0].split('\n')[1:] if i]
dmi_dic = dict([i.strip().split(':') for i in parsed_data])
dic['vendor'] = dmi_dic['Manufacturer'].strip()
dic['product'] = dmi_dic['Product Name'].strip()
dic['sn'] = dmi_dic['Serial Number'].strip()[:15]
return dic

def getHostname(f):
with open(f) as fd:
for line in fd:
if line.startswith('HOSTNAME'):
hostname = line.split('=')[1].strip()
break
return {'hostname':hostname} # 返回一個字典

def getOSver(f):
with open(f) as fd:
for line in fd:
osver = line.strip()
break
return {'osver':osver}

def getCpu(f):
num = 0
with open(f) as fd:
for line in fd:
if line.startswith('processor'):
num += 1
if line.startswith('model name'):
cpu_model = line.split(':')[1].split()
cpu_model = cpu_model[0]+' '+cpu_model[-1]
return {'cpu_num':num, 'cpu_model':cpu_model}

def getMemory(f):
with open(f) as fd:
for line in fd:
if line.startswith('MemTotal'):
mem = int(line.split()[1].strip())
break
mem = "%s" % int(mem/1024.0)+'M'
return {'memory':mem}

if __name__ == '__main__':
dic = {}
data_ip = getIfconfig()
parsed_data_ip = parseData(data_ip)
ip = parseIfconfig(parsed_data_ip)
data_dmi = getDmi()
parsed_data_dmi = parseData(data_dmi)
dmi = parseDmi(parsed_data_dmi)
hostname = getHostname('/etc/sysconfig/network')
osver = getOSver('/etc/issue')
cpu = getCpu('/proc/cpuinfo')
mem = getMemory('/proc/meminfo')
dic.update(ip)
dic.update(dmi)
dic.update(hostname)
dic.update(osver)
dic.update(cpu)
dic.update(mem)
print dic
#d = urllib.urlencode(dic)
d = pickle.dumps(dic)
req = urllib2.urlopen('http://192.168.1.5:8000/hostinfo/collect/',d)
print req.read()

 

 

 

# 收集主機信息 正則表達式版本
#!/usr/bin/env python

import re
from subprocess import Popen, PIPE

def getIfconfig():
p = Popen(['ifconfig'], stdout=PIPE)
data = p.stdout.read().split('\n\n')
return [i for i in data if i and not i.startswith('lo')]

def parseIfconfig(data):
re_devname = re.compile(r'(br|eth|em|virbr|lo|bond)[\d:]+',re.M)
re_mac = re.compile(r'HWaddr ([0-9A-F:]{17})', re.M)
re_ip = re.compile(r'inet addr:([\d\.]{7,15})', re.M)
devname = re_devname.search(data)
if devname:
devname = devname.group()
else:
devname = ''
mac = re_mac.search(data)
if mac:
mac = mac.group(1)
else:
mac = ''
ip = re_ip.search(data)
if ip:
ip = ip.group(1)
else:
ip = ''
return {devname: [ip, mac]}

if __name__ == '__main__':
dic = {}
data = getIfconfig()
for i in data:
dic.update(parseIfconfig(i))
print dic

# group示例
# 按照括號分的，第一個括號是group(1),第二括號是group(2)，以此類推。
group(0)與group()相同，表示匹配到的內容。
In [35]: m = re.match(r'(\w+)\s(\w+)\s', 'hello world abc')

In [36]: m.group()
Out[36]: 'hello world '

In [37]: m.group(0)
Out[37]: 'hello world '

In [38]: m.group(1)
Out[38]: 'hello'

In [39]: m.group(2)
Out[39]: 'world'

文章列表

不含病毒。www.avast.com

AutoPoster

互聯網 - 大數據

AutoPoster 發表在痞客邦留言(0) 人氣(13)

互聯網 - 大數據

本部落格為互聯網熱門頭條訊息管理中心