Python Truthiness and Truth Table

I thought I might be able to find more fuel for my anti-python argument by investigating exactly what objects are equal to each other (generate a truth table). Actually, the python truth table is actually, very reasonable.

Also, I wanted to be confident of what values are considered True or False.

The one best practice related to this – that a colleague taught me – is to use x is None instead of use if x == None: ... because None could be equal to any object depending on its implementation of __eq__.

This is a pure and beautiful sight compared to to Javascript or PHP. It is a little surprising to me that [] == [] since I would expect them to be two different list instances, but I would expect equality for (,) = (,) (identical tuples).

There are a few nuances when considering truthiness, however. If you’ve got code that reads if x: ... The following values for x are treated as false:

• None
• False
• zero of any numeric type. Like these: 0, 0L, 0.0, 0j.
• any empty string, tuple, or array.  '', (), [].
• an empty dictionary, for example, {}.
• instances of user-defined classes, if the class defines a __nonzero__() or __len__() method and that method returns the integer 0 or boolean value False

Everything else is truthy (considered equal to True).

Here’s the code I used to generate the tables above.