Python执行Linux系统命令的4种方法
(1) os.system
仅仅在一个子终端运行系统命令，而不能获取命令执行后的返回信息
system(command) -> exit_status
Execute the command (a string) in a subshell.
如果再命令行下执行，结果直接打印出来
>>> os.system(\'ls\')
04101419778.CHM bash document media py-django video
11.wmv books downloads Pictures python
all-20061022 Desktop Examples project tools
(2) os.popen
该方法不但执行命令还返回执行后的信息对象
popen(command [, mode=\'r\' [, bufsize]]) -> pipe
Open a pipe to/from a command returning a file object.
例如：
>>>tmp = os.popen(\'ls *.py\').readlines()
>>>tmp
Out[21]:
[\'dump_db_pickle.py \',
\'dump_db_pickle_recs.py \',
\'dump_db_shelve.py \',
\'initdata.py \',
\'__init__.py \',
\'make_db_pickle.py \',
\'make_db_pickle_recs.py \',
\'make_db_shelve.py \',
\'peopleinteract_query.py \',
\'reader.py \',
\'testargv.py \',
\'teststreams.py \',
\'update_db_pickle.py \',
\'writer.py \']
好处在于：将返回的结果赋于一变量，便于程序的处理。
(3) 使用模块 subprocess
>>> import subprocess
>>> subprocess.call(["cmd", "arg1", "arg2"],shell=True)
获取返回和输出:
import subprocess
p = subprocess.Popen(\'ls\', shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
for line in p.stdout.readlines():
print line,
retval = p.wait()
(4) 使用模块 commands
>>> import commands
>>> dir(commands)
[\'__all__\', \'__builtins__\', \'__doc__\', \'__file__\', \'__name__\', \'getoutput\', \'getstatus\',\'getstatusoutput\', \'mk2arg\', \'mkarg\']
>>> commands.getoutput("date")
\'Wed Jun 10 19:39:57 CST 2009\'
>>>
>>> commands.getstatusoutput("date")
(0, \'Wed Jun 10 19:40:41 CST 2009\')
注意： 当执行命令的参数或者返回中包含了中文文字，那么建议使用subprocess，如果使用os.popen则会出现下面的错误:
Traceback (most recent call last):
File "./test1.py", line 56, inmain()
File "./test1.py", line 45, in main
fax.sendFax()
File "./mailfax/Fax.py", line 13, in sendFax
os.popen(cmd)
UnicodeEncodeError: \'ascii\' codec can\'t encode characters in position 46-52: ordinal not inrange(128)