Proclib is a high level wrapper/abstraction around the standard library subprocess module, written in Python, with proper piping support which aims to simplify the usage of Unix utilities right from Python and help the developer focus on the commands and not the code which calls the commands.

proclib.api.spawn(cmd)

Given a string or list making up commands cmd, return a Response object which is the result of piping the commands, i.e. they are run in parallel. The data parameter can be used to configure the data passed in to the initial process. Usage example:

>>> from proclib.api import spawn
>>> r = spawn('yes | head')
>>> r.stdout.read()
'y\ny\ny\ny\ny\ny\ny\ny\ny\ny\n'
>>> r.close()
>>> r.history[0].explain_signal()
{'action': 'kill',
 'description': 'write on a pipe with no readers',
 'id': 13,
 'signal': 'SIGPIPE'}

Streaming support is built-in- that is that the stdout of any process can be streamed lazily instead of read and stored in memory all in one go. Also, any kind of iterable can be piped to the process:

def gen():
    yield 'hi\n'
    yield 'ho\n'

r = spawn('cat', data=gen())
assert r.out.split() == ['hi', 'ho']