Amazing Python Package Showcase (1) – virtualenv

Virtualenv is a powerful Python tool that enables the creation of isolated Python environments. This isolation ensures that dependencies from different projects do not interfere with each other, making it an essential tool for Python developers.

What is virtualenv?

Virtualenv is a tool to create isolated Python environments. It helps manage project-specific dependencies without affecting the system Python installation or other projects. You can find it at https://virtualenv.pypa.io/en/latest/index.html

Installing virtualenv

To install virtualenv, use pip:

$ pip install virtualenv
Defaulting to user installation because normal site-packages is not writeable
Collecting virtualenv
  Downloading virtualenv-20.26.3-py3-none-any.whl (5.7 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.7/5.7 MB 3.6 MB/s eta 0:00:00
Collecting distlib<1,>=0.3.7
  Downloading distlib-0.3.8-py2.py3-none-any.whl (468 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 468.9/468.9 KB 3.6 MB/s eta 0:00:00
Collecting filelock<4,>=3.12.2
  Downloading filelock-3.15.4-py3-none-any.whl (16 kB)
Collecting platformdirs<5,>=3.9.1
  Downloading platformdirs-4.2.2-py3-none-any.whl (18 kB)
Installing collected packages: distlib, platformdirs, filelock, virtualenv
Successfully installed distlib-0.3.8 filelock-3.15.4 platformdirs-4.2.2 virtualenv-20.26.3

Creating a Virtual Environment

To create a virtual environment, navigate to your project directory and run:

$ virtualenv myenv
created virtual environment CPython3.10.12.final.0-64 in 927ms
  creator CPython3Posix(dest=/home/dynotes/myenv, clear=False, no_vcs_ignore=False, global=False)
  seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=/home/dynotes/.local/share/virtualenv)
    added seed packages: pip==24.1, setuptools==70.1.0, wheel==0.43.0
  activators BashActivator,CShellActivator,FishActivator,NushellActivator,PowerShellActivator,PythonActivator

This command creates a directory named myenv containing a standalone Python installation.

Activating and Deactivating the Virtual Environment

Activate the environment:

On Windows

myenv\Scripts\activate

On macOS and Linux

source myenv/bin/activate

Deactivate the environment

dynotes@P2021:~$ source myenv/bin/activate
(myenv) dynotes@WIN-P2021:~$ deactivate
dynotes@P2021:~$

Installing Packages in the Virtual Environment

Once the environment is activated, you can install packages using pip install package_name

(myenv) dynotes@P2021:~/projects/python/numpy$ pip install numpy
Collecting numpy
  Downloading numpy-2.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 60.9/60.9 kB 1.4 MB/s eta 0:00:00
Downloading numpy-2.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (19.3 MB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.3/19.3 MB 3.6 MB/s eta 0:00:00
Installing collected packages: numpy
Successfully installed numpy-2.0.0

The packages installed will only be available in the virtual environment.

Example Script

Here’s an example script that you can use within your virtual environment

import numpy as np

a = np.arange(15).reshape(3, 5)
print(a)
print(a.shape)
print(a.ndim)
print(a.dtype.name)
print(a.itemsize)
print(a.size)

To run the script, in the activated environment execute

(myenv) dynotes@P2021:~/projects/python/numpy$ python example1.py
[[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]]
(3, 5)
2
int64
8
15

Different Tools for Creating Isolated Python Environments

virtualenv vs Conda

Virtualenv

• Purpose: Primarily for creating isolated Python environments.
• Package Management: Uses pip to install Python packages.
• Language Support: Only Python.
• Environment Management: Simple, lightweight, and quick to set up.
• Dependency Management: Relies on requirements.txt for managing dependencies.
• Activation: Requires manual activation using commands like source venv/bin/activate (Linux/Mac) or venv\Scripts\activate (Windows).

Conda

• Purpose: Manages environments and packages for multiple languages (Python, R, Ruby, etc.).
• Package Management: Uses conda to install packages, which can include non-Python dependencies.
• Language Support: Multi-language support.
• Environment Management: More comprehensive, can handle complex dependencies.
• Dependency Management: Uses environment.yml to manage dependencies, which can include non-Python packages.
• Activation: Uses conda activate <env_name> for environment activation.

Key Differences

• Scope: Virtualenv is focused on Python, while Conda supports multiple languages and broader environment management.
• Package Sources: Virtualenv relies on PyPI, whereas Conda has its own package repositories that include a wide range of software, including system libraries and tools.
• Complexity: Conda is more complex but offers greater flexibility, especially for data science and scientific computing where multiple types of dependencies are common.

virtualenv and python -m venv

virtualenv

• Purpose: A third-party tool for creating isolated Python environments.
• Compatibility: Works with Python 2 and 3.
• Features: More features and flexibility (e.g., can create environments with different versions of Python).
• Installation: Requires installation via pip (pip install virtualenv).

python -m venv

• Purpose: A built-in module in Python 3.3 and later for creating isolated environments.
• Compatibility: Only works with Python 3.3+.
• Features: Simplified and part of the Python standard library.

Key Differences

• virtualenv offering more flexibility and
• python -m venv being more straightforward and part of the standard library for Python 3.3+.

Note: all command lines are run in WLS 2,. If you are interested in WLS 2, please refer to https://templespark.com/master-wsl-2-a-concise-guide-for-windows-subsystem-for-linux/