how to execute a bash command in a python script

Question

How can I launch a bash command with multiple args (for example "sudo apt update") from a python script?

Answer 1

@milne's answer works, but subprocess.call() gives you little feedback.

I prefer to use subprocess.check_output() so you can analyse what was printed to stdout:

 import subprocess res = subprocess.check_output(["sudo", "apt", "update"]) for line in res.splitlines(): # process the output line by line 

check_output throws an error on on-zero exit of the invoked command

Please note that this doesn't invoke bash or another shell if you don't specify the shell keyword argument to the function (the same is true for subprocess.call(), and you shouldn't if not necessary as it imposes a security hazard), it directly invokes the command.

If you find yourself doing a lot of (different) command invocations from Python, you might want to look at plumbum. With that you can do the (IMO) more readable:

from plumbum.cmd import sudo, apt, echo, cut res = sudo[apt["update"]]() chain = echo["hello"] | cut["-c", "2-"] chain() 

Answer 2

It is possible you use the bash as a program, with the parameter -c for execute the commands:

Example:

bashCommand = "sudo apt update" output = subprocess.check_output(['bash','-c', bashCommand]) 

Answer 3

The subprocess module is designed to do this:

import subprocess subprocess.call(["sudo", "apt", "update"]) 

If you would like the script to terminate if the command fails, you might consider using check_call() instead of parsing the return code yourself:

subprocess.check_call(["sudo", "apt", "update"]) 

Answer 4

Also you can use 'os.popen'.

Example:

import os command = os.popen('ls -al') print(command.read()) print(command.close()) 

Output:

total 16 drwxr-xr-x 2 root root 4096 ago 13 21:53 . drwxr-xr-x 4 root root 4096 ago 13 01:50 .. -rw-r--r-- 1 root root 1278 ago 13 21:12 bot.py -rw-r--r-- 1 root root 77 ago 13 21:53 test.py None 

Answer 5

For python 3.5 and more you can use:

import subprocess try: result = subprocess.run("sudo apt update", check=True, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) except subprocess.CalledProcessError as err: raise Exception(str(err.stderr.decode("utf-8"))) except Exception as err: raise Exception(err) else: return result.stdout.decode("utf-8") 

Answer 6

use subprocess module

import subprocess command = 'sudo apt update' subprocess.check_call(command.split()) 

Answer 7

import os os.system("sudo apt update") 

Answer 8

I like @ricardo130's os.popen() approach because it can handle pipes, redirects, and incremental (realtime) output:

import os with os.popen("sudo apt update") as stream: for line in stream.readlines(): print(line) 

And you can do whatever you like with the output before displaying the output. Or use tqdm for a progress bar. To create a table of the largest files in your git history from this answer:

from tqdm import tqdm import json cmd = """ git rev-list --objects --all \ | git cat-file --batch-check='%(objecttype) %(objectname) %(objectsize) %(rest)' \ | sed -n 's/^blob //p' \ | sort --numeric-sort --key=2 \ | cut -c 1-12,41- \ | $(command -v gnumfmt || echo numfmt) --field=2 --to=iec-i --suffix=B --padding=7 --round=nearest """ with os.popen(cmd) as stream: table = [] for line in tqdm(stream.readlines()): row = [line[12:20].strip(), line[20:].strip()] table.append(row) print(json.dumps(table, indent=2)) 

Which outputs something like this:

... [ "51MiB", ".nlpia2-data/fake-news-dataset-true.csv" ], [ "60MiB", ".nlpia2-data/fake-news-dataset-fake.csv" ] ]