How do I check the versions of Python modules?

Miscellaneous ways to check version details

There are plenty of ways to probe the version details of Python packages and modules. Here, we'll explore them at the command line level to Python's built-in facilities.

Communicating with Python from the terminal

Whether it's Linux, MacOS, or Windows, your command-line interface has got you covered:

• On Linux/Mac:

# To check for a specific package version
# Salvation:
pip freeze | grep <package_name>

# Or, to upgrade pip itself because hey, pip needs love too!
pip install --upgrade pip


• On Windows:

# For specific package version checking:
# Sherlock Holmes mode
pip freeze | findstr <package_name>

# To upgrade pip because updates are good (usually!)
python -m pip install --upgrade pip


Digging deeper into Python

Sometimes, you've got to go deeper:

# Digging deeper in Python when `__version__` attribute is off the grid
import pkg_resources
pkg_resources.get_distribution('package_name').version


And starting Python 3.8, you can even go meta:

from importlib.metadata import version
version('package_name') 


Harnessing the power of virtual environments

For project isolation, use tools such as virtualenvwrapper:

# Run this magic command inside a virtual environment
pip freeze # BAM! Get all the packages and their versions


From easy_install to pip: an evolution story

Transitioning from easy_install to pip is like upgrading from a bicycle to a sports car. With pip, you unlock commands like:

# Show package details including the version. Mmm, details...
pip show <package_name>


Remember to be precise with package names, as the exact name you input should correspond to the PyPI listing.

What if there is no __version__?

Some modules got attitude, they don't have a __version__:

try:
    import module
    print(module.__version__)
except AttributeError:
    print("This module thinks it's too cool for a __version__.")


Plugging into importlib metadata

As Python development evolves, the importlib metadata library is becoming the new norm for retrieving package versions:

from importlib import metadata
metadata.version('package_name') # Ain't nobody got time for __version__


Using virtual environments for peace of mind

Virtual environments back up your project environment. This way, you can replicate environments with ease using requirements.txt files generated by pip freeze.