Apr. 24, 2013

How to use sh in Python

What is sh?

sh is a unique subprocess wrapper that maps your system programs to Python
functions dynamically.  sh helps you write shell scripts in Python by giving you
the good features of Bash (easy command calling, easy piping) with all the power
and flexibility of Python. [source]

Starting with sh

sh is a full-fledged subprocess interface for Python that allows you to call any
program as if it were a function. sh lets you call just about anything that you
could run from a login shell much more neatly than you can with subprocess.Popen,
and more importantly lets you capture and parse the output much more readily.


The installation of sh is done through the pip command
$> pip install sh


The easiest way to get up and running is to import sh directly or import your
program from sh. Every command you want to run is imported like any other module.  

That command is then usable just like a Python statement.  

Arguments are passed per usual, and output can be captured and worked with in a
like fashion.
# get interface information
import sh
print sh.ifconfig("eth0")

from sh import ifconfig
print ifconfig("eth0")

# print the contents of this directory
print ls("-l")

# substitute the dash for an underscore for commands that have dashes in their names

Executing Commands with sh

Commands are called just like functions. 

"Note that these aren't Python functions, these are running the binary commands
on your system dynamically by resolving your PATH, much like Bash does. In this
way, all the programs on your system are easily available in Python."
Many programs have their own command subsets, like git (branch, checkout). 

sh handles subcommands through attribute access.
from sh import git

# resolves to "git branch -v"

print(git("branch", "-v")) # the same command

Keyword Arguments

Keyword arguments also work like you'd expect: they get replaced with the
long-form and short-form commandline option. [source]
# Resolves to "curl http://duckduckgo.com/ -o page.html --silent"
sh.curl("http://duckduckgo.com/", o="page.html", silent=True)

# If you prefer not to use keyword arguments, this does the same thing
sh.curl("http://duckduckgo.com/", "-o", "page.html", "--silent")

# Resolves to "adduser amoffat --system --shell=/bin/bash --no-create-home"
sh.adduser("amoffat", system=True, shell="/bin/bash", no_create_home=True)

# or
sh.adduser("amoffat", "--system", "--shell", "/bin/bash", "--no-create-home")

Finding Commands

"Which" finds the full path of a program, or returns None if it doesn't exist.

This command is one of the few commands implemented as a Python function, and
therefore doesn't rely on the "which" program actually existing.
print sh.which("python")     # "/usr/bin/python"
print sh.which("ls")         # "/bin/ls"

if not sh.which("supervisorctl"): sh.apt_get("install", "supervisor", "-y")
There are many more features that you can use with sh, and you can find them all
in the official documentation. 


sh is capable of "baking" arguments into commands. 
# The idea here is that now every call to ls will have the “-la” arguments already specified.
from sh import ls

ls = ls.bake("-la")
print(ls) # "/usr/bin/ls -la"

# resolves to "ls -la /"
Baked ssh command
Calling "bake" on a command creates a callable object that automatically passes
along all of the arguments passed into "bake".
# Without baking, calling uptime on a server would be a lot to type out:
serverX = ssh("myserver.com", "-p 1393", "whoami")

# To bake the common parameters into the ssh command
myserver = sh.ssh.bake("myserver.com", p=1393)

print(myserver) # "/usr/bin/ssh myserver.com -p 1393"
Now that the “myserver” callable represents a baked ssh command, you can call
anything on the server easily:
# resolves to "/usr/bin/ssh myserver.com -p 1393 tail /var/log/dumb_daemon.log -n 100"
print(myserver.tail("/var/log/dumb_daemon.log", n=100))

# check the uptime
print myserver.uptime()
 15:09:03 up 61 days, 22:56,  0 users,  load average: 0.12, 0.13, 0.05
For more advanced features, please see the official documentation.

Share this article

Recommended Python Training – DataCamp

For Python training, our top recommendation is DataCamp.

Datacamp provides online interactive courses that combine interactive coding challenges with videos from top instructors in the field.

Datacamp has beginner to advanced Python training that programmers of all levels benefit from.


Download Our Free Guide To Learning Python

* indicates required

Read more about:
Disclosure of Material Connection: Some of the links in the post above are “affiliate links.” This means if you click on the link and purchase the item, I will receive an affiliate commission. Regardless, PythonForBeginners.com only recommend products or services that we try personally and believe will add value to our readers.