Question

If I generate multiple subprocess.Popen(['commands', 'that', 'I', 'called']) and for each I do stdin.write(..) or p.communicate(...)to interact with the commands, is it guarantee to be independent and will come back to the each process (stdout from the called command)?

Answer 1

If you do this:

proc = subprocess.Popen(
    cmd, stdin=subprocess.PIPE, stdout=subprocess.PIPE)
stdout, _ = proc.communicate()

You will get a separate set of pipes for each process. The value subprocess.PIPE is just a special flag to tell subprocess.Popen() to create a new pipe — so the above command creates two new pipes: one pipe for stdin and a separate pipe for stdout.

If you do this:

proc = subprocess.Popen(cmd)
proc.wait()

The new process will share stdin, stdout, and stderr with your process (well, basically — the kernel side will be shared).