在写python脚本的时候,经常需要调用系统命令,常用的python调用系统命令的方法主要有subprocess.call和os.popen。默认情况下subprocess.call的方法结果是返回值,即1或0,而os.popen则是命令运行的结果,可以用readlines(读取所有行,返回数组)或者read(读读取所有行,返回str)来读取。
subprocess类总主要的方法有:
subprocess.call:开启子进程,开启子进程,运行命令,默认结果是返回值,不能try
subprocess.check_call:运行命令,默认结果是返回值,可以try
subprocess.check_out(2.7中才有这个方法) 开启子进程,运行命令,可以获取命令结果,可以try
subprocess.Popen 开启子进程,运行命令,没有返回值,不能try,可以获取命令结果
subprocess.PIPE 初始化stdin,stdout,stderr,表示与子进程通信的标准流
Popen.poll 检查子进程是否结束,并返回returncode
Popen.wait等待子进程是否结束,并返回retrurncode
比如check_call的sample:
import subprocess
import traceback
cmd='hadoop fs -ls hdfs://xxxxx'
try:
e=subprocess.check_call(cmd,shell=True,stdout=subprocess.PIPE)
print "return code is: %s"%(str(e))
#print stdout.read()
except Exception,re:
print "message is:%s" %(str(re))
traceback.print_exc()
分析subprocess的源码:
class CalledProcessError(Exception): #首先定义了一个exception,用来在check_call和 check_output中raise exception
def __init__(self, returncode, cmd, output=None):
self.returncode = returncode
self.cmd = cmd
self.output = output
def __str__(self):
return "Command '%s' returned non-zero exit status %d" % (self.cmd, self.returncode)
..........
def call(*popenargs, **kwargs):
return Popen(*popenargs, **kwargs).wait() #call方法调用wait
def check_call(*popenargs, **kwargs):
retcode = call(*popenargs, **kwargs) #调用call,返回返回值
if retcode:
cmd = kwargs.get("args")
if cmd is None:
cmd = popenargs[0]
raise CalledProcessError(retcode, cmd) #可以抛出异常
return 0
def check_output(*popenargs, **kwargs):
if 'stdout' in kwargs:
raise ValueError('stdout argument not allowed, it will be overridden.')
process = Popen(stdout=PIPE, *popenargs, **kwargs)
output, unused_err = municate() #获取标准输出和标准错误输出
retcode = process.poll() #检查子进程是否结束,并返回returncode
if retcode:
cmd = kwargs.get("args")
if cmd is None:
cmd = popenargs[0]
raise CalledProcessError(retcode, cmd, output=output)
return output
有时候我们需要在运行命令时可以获取返回值,获取结果,并且能够try。
可以对上面的代码进行组合
# -*- coding: utf8 -*-
import exceptions
import subprocess
import traceback
class CalledCommandError(Exception):
def __init__(self, returncode, cmd, errorlog,output):
self.returncode = returncode
self.cmd = cmd
self.output = output
self.errorlog = errorlog
def __str__(self):
return "命令运行错误:'%s',返回值: %s,错误信息: %s" % (self.cmd, str(self.returncode) ,self.errorlog)
def run_command_all(*popenargs, **kwargs):
allresult = {}
cmd = popenargs[0]
if 'stdout' in kwargs or 'stderr' in kwargs :
raise ValueError('标准输出和标准错误输出已经定义,不需设置。')
process = subprocess.Popen(stdout=subprocess.PIPE,shell=True,stderr = subprocess.PIPE,*popenargs, **kwargs)
output, unused_err = municate()
retcode = process.poll()
if retcode:
#print retcode,cmd,unused_err,output
raise CalledCommandError(cmd,retcode,errorlog=unused_err,output=output)
allresult['cmd'] = cmd
allresult['returncode'] = retcode
allresult['errorlog'] = unused_err
allresult['outdata'] = output
return allresult
if __name__ == '__main__':
cmd = 'hadoop fs -ls xxxx|wc -l'
try:
e=run_command_all(cmd)
print "ok"
except Exception,re:
print (str(re))
print "failed"
traceback.print_exc()
©著作权归作者所有:来自51CTO博客作者菜菜光的原创作品,如需转载,请注明出处,否则将追究法律责任
调用python shell命令dev