Difference between revisions of "PythonVirtualEnv"

From UMIACS
Jump to navigation Jump to search
Line 1: Line 1:
A virtual environment is an isolated working copy of Python, which allows you to work on specific projects without affecting others.  It creates an environment that has its own installation directories and that doesn’t share libraries with other virtualenv environments (and optionally doesn’t access the globally-installed libraries either).
+
A virtual environment is an isolated working copy of Python, which allows you to work on specific projects without affecting others.  It creates an environment that has its own installation directories and that does not share libraries with other virtualenv environments (and optionally doesn’t access the globally-installed libraries either).
 
<br/><br/>
 
<br/><br/>
virtualenv can be found in the following locations:
+
While virtualenv should be installed on <I>most</I> UMIACS supported machines, the source can be downloaded from the projects GitHub:
*'''RHEL5'''
 
:<code>/usr/local/stow/virtualenv-1.5.1/virtualenv.py</code>
 
*'''RHEL6'''
 
:<code>/opt/local/stow/virtualenv-1.9.1/virtualenv.py</code>
 
*'''RHEL7'''
 
:<code>/usr/lib/python2.7/site-packages/virtualenv.py</code>
 
*'''SOURCE''' (most up-to-date)
 
:<code>https://virtualenv.pypa.io/en/latest/</code>
 
 
:<code>https://github.com/pypa/virtualenv</code>
 
:<code>https://github.com/pypa/virtualenv</code>
 +
<br/><br/>
  
 
==Basic Usage==
 
==Basic Usage==

Revision as of 13:35, 15 May 2017

A virtual environment is an isolated working copy of Python, which allows you to work on specific projects without affecting others. It creates an environment that has its own installation directories and that does not share libraries with other virtualenv environments (and optionally doesn’t access the globally-installed libraries either).

While virtualenv should be installed on most UMIACS supported machines, the source can be downloaded from the projects GitHub:

https://github.com/pypa/virtualenv



Basic Usage

The following steps outline how to create a virtual environment using the system Python. Please note the that following examples were done using RHEL5. Please adjust the commands to reflect the OS you are on.

Creating the virtual environment (example)

-bash-3.2$ git clone https://github.com/pypa/virtualenv.git
-bash-3.2$ python virtualenv/virtualenv.py env
New python executable in env/bin/python
Installing setuptools.............done.

-bash-3.2$ source env/bin/activate

(env)-bash-3.2$ which python
~/env/bin/python

You will notice that once you have created your virtual environment, you will need to use the 'source' command to load it into your environment. In bash and tcsh, the environment can be deactivated by typing deactivate

Installing Python Modules

Once you have created your virtual environment and sourced it, you can install additional modules using the 'pip' command.

(env)-bash-3.2$ pip install nose
Downloading/unpacking nose
  Downloading nose-1.3.3.tar.gz (274Kb): 274Kb downloaded
  Running setup.py egg_info for package nose

...output omitted for brevity...

Successfully installed nose
Cleaning up...
(env)-bash-3.2$ 

Listing installed Python Modules

(env)-bash-3.2$ pip freeze
nose==1.3.3

Uninstalling Python Modules

'pip' can also be used to remove a module from the environment.

(env)-bash-3.2$ pip uninstall nose
Uninstalling nose:
  /chimerahomes/sabobbin/env/bin/nosetests
  /chimerahomes/sabobbin/env/bin/nosetests-2.4
  /chimerahomes/sabobbin/env/lib/python2.4/site-packages/nose
  /chimerahomes/sabobbin/env/lib/python2.4/site-packages/nose-1.3.3-py2.4.egg-info
  /chimerahomes/sabobbin/env/man/man1/nosetests.1
Proceed (y/n)? y
  Successfully uninstalled nose

Switching between virtual environments

To switch between different environments, simply deactivate your current virtual environment, and source the other.

(env)-bash-3.2$ which python
~/env/bin/python

(env)-bash-3.2$ deactivate 

-bash-3.2$ source env2.7/bin/activate

(env2.7)-bash-3.2$ which python
~/env2.7/bin/python

Using a different python version

To create a virtual environment that uses a version of python that is different then the system default, simply create the virtualenv with your target version of Python. This could be a Python build we provide via Modules, or one you've built yourself. Virtualenv will pick up the first python version it finds in your $PATH, or you can direct it to a specific location with the "--python" flag.

-bash-3.2$ module load Python

-bash-3.2$ which python
/usr/local/stow/python-2.7.2/bin/python

Important: virtualenv will include any modules listed in your PYTHONPATH when initializing the virtual environment. To ensure a vanilla environment, it might be a good idea to verify your PYTHONPATH is empty.

-bash-3.2$ echo $PYTHONPATH
/usr/local/stow/python-commonmodules-2.7.2.0/lib/python2.7/site-packages

-bash-3.2$ PYTHONPATH=''
-bash-3.2$ /usr/local/stow/virtualenv-1.5.1/virtualenv.py env2.7
New python executable in env2.7/bin/python
Installing setuptools....................done.

-bash-3.2$ source env2.7/bin/activate
(env2.7)-bash-3.2$ which python
~/env2.7/bin/python