PyConf Hyderabad 2017

Your scripts will eventually grow into a package. Your package needs to be user-friendly to get more people to use it. Your package might grow big in order to address everyone’s needs. entry_points make your package more accessible. entry_points help you add functionality without adding almost any extra code.

After this talk, you will be able to make your package accessible from the command line and extend your package’s functionality using entry_points. You’ll also have an understanding of how entry_points work.

This talk is intended for audience with a basic understanding of functions and packages in Python.

Outline

A lot of popular Python packages use entry points to make it easier to use them and to make it easier to extend their functionality. A very popular examples is pip.

Broadly speaking, entry points can be used to make your package easily accessible and extensible. Specifically, entry_points is a keyword argument to the setup function, the one in every setup.py file.

The console_scripts and gui_scripts entry points can be used to make your package accessible from the command line. Popular packages that makes use of this are pip and flask.

You can define your own entry point which others can contribute functionality to. A popular example is sphinx.html_themes, which the Sphinx package listens to. The entry point allows the Sphinx maintainers to prevent the number of default themes from blowing up while still allowing the community to contribute themes.

Finally, I will talk about how entry points are discovered and used. We’ll look at the files in *.egg-info and how the pkg_resources package lets you use entry points.

Requirements

N.A.

Speaker bio

I am a Scientific Software Developer at Enthought India. I graduated from IIT Madras with a B.S. & M.S. in Physics.

I have been involved with the Python community in Pune over the last year, during which I gave talks on Using Cython and on Using virtualenvs. I conducted two workshops at SciPy India 2016 on Using Git & GitHub and on Automated testing using the unittest and mock libraries. Finally, I was invited to IIT Madras, my almamater, to give a conduct a workshop on Scientific Computing using the Numpy, Pandas and Cython libraries.

I hope that answers both the question of whether or not I am competent in Python and it’s ecosystem of packages and the question of whether or not I am good at public speaking.

Links

• https://github.com/rahulporuri/talks/tree/master/py_entry_points - contains an initial draft of the presentation and code related to the presentation.
• Slides from other talks/workshops I organised can be found at - https://github.com/rahulporuri/talks

Slides

https://github.com/rahulporuri/talks/blob/master/py_entry_points/py_entry_points.pdf