Danger Will Robinson! Warning! Warning! Warning! (What Python warnings are, why you should turn them on and how to create your own)

Python’s warnings mechanism is an important part of creating robust, reliable, production grade code. Unfortunately, it’s also one of the least known and one of the most under-used standard libraries. Help find the bugs and errors with warnings! What bugs are lurking in your code? Ignore them at your peril! Suitable for beginner to intermediate. This talk will cover the basics of the warning mechanism, system-generated and user-generated warnings, how to turn them on and off , and how to filter and log warnings.

• Give me all the warnings! Why you want the warnings - what can go wrong if you don’t know about them, or if you ignore them.
• The history of the warnings module, PEP 230 and more.
• The warnings mechanism - how it works. Turning them on and off.
• The kinds of warnings
  • examples of system warnings.
  • deprecation warnings (if you learn nothing else, turn this on).
  • all about the bytes
  • py3k warnings.
  • How to generate your own warnings.
  • Other kinds of warnings.
• Controlling warnings with the environment variable.
• Changes to warnings in different versions (2.7, 3.2, 3.6, changes coming in 3.7)
• More advanced warnings control - filtering warnings, creating your own filters, formatting warnings, understanding the stack, testing that your warnings work, the context manager.
• Adding warnings into the logging framework.
• How warnings are used in conjunction with testing, logging, asserts and exceptions.