Lightning Talk on Timezones in Python.

Given to PghPy on May 28th 2014.

Timezones in python are hard. There are a lot of them, they’re constantly changing, and many follow different rules. This talk addressed some of the problems that make timezones hard, and some of the existing solutions for working with time-zone aware data.

The slides can be found here:

Timezones in Python

At the time of the presentation I was helping to build an open source package for making timezones simpler, called fleming, named after Sandford Fleming, father of the worldwide timezone system. This package includes a number of functions useful for doing timezone manipulations, with special care taken around daylight savings time boundaries.

I was also working on building a Django package for tracking recurring events in users local times. This package Django Localized Recurrence, used fleming to make conversions between users local timezones and UTC, the underlying data representation easier.

I still think there’s room for improvement in dealing with timezones using python. As evidenced by the huge number of packages on pypi for working with datetimes, timedeltas and with timezone data.

Additional Resources

Any questions, comments, suggestions about this post? Or anything else? Feel free to send me an email: