Launch a subprocess in a pseudo terminal (pty), and interact with both the process and its pty.

Sometimes, piping stdin and stdout is not enough. There might be a password prompt that doesn’t read from stdin, output that changes when it’s going to a pipe rather than a terminal, or curses-style interfaces that rely on a terminal. If you need to automate these things, running the process in a pseudo terminal (pty) is the answer.


p = PtyProcessUnicode.spawn(['python'])


What is a pty?

A pty is a kernel-level object which processes can write data to and read data from, a bit like a pipe.

Unlike a pipe, data moves through a single pty in both directions. When you use a program in a shell pipeline, or with subprocess.Popen in Python, up to three pipes are created for the process’s standard streams (stdin, stdout and stderr). When you run a program using ptyprocess, all three of its standard streams are connected to a single pty:


A pty also does more than a pipe. It keeps track of the window size (rows and columns of characters) and notifies child processes (with a SIGWINCH signal) when it changes. In cooked mode, it does some processing of data sent from the parent process, so for instance the byte 03 (entered as Ctrl-C) will cause SIGINT to be sent to the child process.

Many command line programs behave differently if they detect that stdin or stdout is connected to a terminal instead of a pipe (using isatty()), because this normally means that they’re being used interactively by a human user. They may format output differently (e.g. ls lists files in columns) or prompt the user to confirm actions. When you run these programs in ptyprocess, they will exhibit their ‘interactive’ behaviour, instead of the ‘pipe’ behaviour you’ll see using Popen().

See also

The TTY demystified
Detailed article by Linus Akesson

Indices and tables