I would like to execute multiple commands in a row:
i.e. (just to illustrate my need):
cmd (the shell)
then
cd dir
and
ls
and read the result of the ls.
Any idea with subprocess module?
Update:
cd dir and ls are just an example. I need to run complex commands (following a particular order, without any pipelining). In fact, I would like one subprocess shell and the ability to launch many commands on it.
To do that, you would have to:
shell=True argument in the subprocess.Popen call, and; if running under a *nix shell (bash, ash, sh, ksh, csh, tcsh, zsh etc)& if running under the cmd.exe of Windows
There is an easy way to execute a sequence of commands.
Use the following in subprocess.Popen
"command1; command2; command3"
Or, if you're stuck with windows, you have several choices.
Create a temporary ".BAT" file, and provide this to subprocess.Popen
Create a sequence of commands with "\n" separators in a single long string.
Use """s, like this.
"""
command1
command2
command3
"""
Or, if you must do things piecemeal, you have to do something like this.
class Command( object ):
def __init__( self, text ):
self.text = text
def execute( self ):
self.proc= subprocess.Popen( ... self.text ... )
self.proc.wait()
class CommandSequence( Command ):
def __init__( self, *steps ):
self.steps = steps
def execute( self ):
for s in self.steps:
s.execute()
That will allow you to build a sequence of commands.
subprocess.Popen("ls") works. However, subprocess.Popen("ls; ls") fails for me. Error: Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/lib64/python2.6/subprocess.py", line 639, in __init__ errread, errwrite) File "/usr/lib64/python2.6/subprocess.py", line 1228, in _execute_child raise child_exception OSError: [Errno 2] No such file or directory
s = """my script with line break""" and running it with subprocess.run([s], shell=True) worked well for me.Finding 'bar' in every file whose name contains 'foo':
from subprocess import Popen, PIPE
find_process = Popen(['find', '-iname', '*foo*'], stdout=PIPE)
grep_process = Popen(['xargs', 'grep', 'bar'], stdin=find_process.stdout, stdout=PIPE)
out, err = grep_process.communicate()
'out' and 'err' are string objects containing the standard output and, eventually, the error output.
Yes, the subprocess.Popen() function supports a cwd keyword argument, with which you can set the directory it runs the process in.
I guess the first step, the shell, is not needed, if all you want is to run ls, there's no need to run it through a shell.
Of course, you could also just pass the desired directory as an argument to ls.
Update: it might be worth noting that for typical shells, cd is implemented in the shell itself, it is not an external command on disk. This is because it needs to change the process' current directory, which must be done from within the process. Since commands run as child processed, spawned by the shell, they cannot do this.
To be more simple, just type all the commands in a single list/tuple. Then just iterate over the list/tuple and pass it as an argument to the system() funtion of os module.
from os import system
cmds = ['cmd1', 'cmd2']
for cmd in cmds:
system(cmd)
below python script have 3 function what you went just excute:
import sys
import subprocess
def cd(self,line):
proc1 = subprocess.Popen(['cd'],stdin=subprocess.PIPE)
proc1.communicate()
def ls(self,line):
proc2 = subprocess.Popen(['ls','-l'],stdin=subprocess.PIPE)
proc2.communicate()
def dir(silf,line):
proc3 = subprocess.Popen(['cd',args],stdin=subprocess.PIPE)
proc3.communicate(sys.argv[1])
cd has no effect on the parent and it is likely that OP wants builtin into the shell command (Popen() doesn't run the shell unless you ask explicitly -- though it won't help in this case).