The Tao of Python

Contents of this talk

  • Object-oriented relationships
  • Relationship rules
  • What is a Python object?
  • Classes as objects
  • Metaclasses
  • What is type?
  • What is object?
  • How are type and object related?
  • The Python objects map

Most of this talk is based on this article by Shalabh Chaturvedi.

Why is this talk useful?

  • Actually it's not terribly useful
  • Deep understanding of the Python object model
  • Clarification of the role and behavior of classes, metaclasses and instances
  • Appreciation of the language on different level
  • Zen-like satisfying moment of understanding
  • Bragging rights :D

Object-oriented relationships

While we introduce many different objects, we only use two kinds of relationships:

  1. is a kind of (solid line): Also known as specialization or inheritance, this relationship exists between two objects when one (the subclass) is a specialized version of the other (the superclass). A snake is a kind of reptile. It has all the traits of a reptile and some specific traits which identify a snake. Terms used: subclass of, superclass of, superclass-subclass or simply is a.
  2. is an instance of (dashed line): Also known as instantiation, this relationship exists between two objects when one (the instance) is a concrete example of what the other specifies (the type). I have a pet snake named Squasher. Squasher is an instance of a snake. Terms used: instance of, type of.

Relationship rules

  1. If A is a subclass of B, and B is a subclass of C, then A is a subclass of C.
  2. If X is an instance of A, and A is a subclass of B, then X is an instance of B.
  3. If B is an instance of M, and A is a subclass of B, then A is an instance of M.

What is a Python object?

An object is an entity with the following characteristic properties:

  1. Identity (i.e. given two names we can say for sure if they refer to one and the same object, or not).
  2. A value - which may include a bunch of attributes (i.e. we can reach other objects through objectname.attributename).
  3. A type - every object has exactly one type. For instance, the object 2 has a type int and the object "joe" has a type str.
  4. One or more bases. A base is similar to a super-class or base-class in object-oriented lingo.
In [1]:
two = 2
<class 'int'>
In [2]:
<class 'type'>
In [3]:
(<class 'object'>,)
In [4]:
['__abs__', '__add__', '__and__', '__bool__', '__ceil__', '__class__', '__delattr__', '__dir__', '__divmod__', '__doc__', '__eq__', '__float__', '__floor__', '__floordiv__', '__format__', '__ge__', '__getattribute__', '__getnewargs__', '__gt__', '__hash__', '__index__', '__init__', '__init_subclass__', '__int__', '__invert__', '__le__', '__lshift__', '__lt__', '__mod__', '__mul__', '__ne__', '__neg__', '__new__', '__or__', '__pos__', '__pow__', '__radd__', '__rand__', '__rdivmod__', '__reduce__', '__reduce_ex__', '__repr__', '__rfloordiv__', '__rlshift__', '__rmod__', '__rmul__', '__ror__', '__round__', '__rpow__', '__rrshift__', '__rshift__', '__rsub__', '__rtruediv__', '__rxor__', '__setattr__', '__sizeof__', '__str__', '__sub__', '__subclasshook__', '__truediv__', '__trunc__', '__xor__', 'bit_length', 'conjugate', 'denominator', 'from_bytes', 'imag', 'numerator', 'real', 'to_bytes']

Rule 1: Everything is an object

And I mean everything. Even things that are "primitive types" in other languages.

  • You can store them in variables
  • You can pass them as parameters to functions
  • You can return them as the result of functions
  • You can construct them at runtime

And more importantly: You can treat every programming construct in a uniform and consistent way

Functions as objects

  • When you use the keyword def, Python creates a function object.
  • Functions can be passed around as arguments to other functions.
  • These functions that take other functions as arguments are called higher order functions.
  • e.g. the map function takes a function and an iterable and applies the function to each item in the iterable.

Classes as objects

  1. When you use the keyword class, Python executes it and creates an object.
  2. This object (the class) is itself capable of creating objects (the instances), and this is why it's a class.
  3. Since classes are objects, they must be generated by something, this is metaclasses.
  4. Since metaclasses are also objects, they must also be generated by something, this is again metaclasses.

Therefore: Objects are instances of classes, classes are instances of metaclasses and metaclasses are instances of themselves.

Metaclasses (these are objects too!)

  1. Metaclasses are the "stuff" that creates classes.
  2. You define classes in order to create objects, right?
  3. We learned that Python classes are objects.
  4. Well, metaclasses are what creates these objects. They are the classes' classes, you can picture them this way:
    • MyClass = MetaClass()
    • MyObject = MyClass()

Metaclasses are deeper magic than 99% of users should ever worry about. If you wonder whether you need them, you don’t (the people who actually need them know with certainty that they need them, and don’t need an explanation about why). Tim Peters

What is type?

  1. Remember the function type? The good old function that lets you know what type an object is.
  2. type can also create classes on the fly. type can take the description of a class as parameters, and return a class as type(name, bases, dct).
    • name is a string giving the name of the class to be constructed.
    • bases is a tuple giving the parent classes of the class to be constructed.
    • dct is a dictionary of the attributes and methods of the class to be constructed.
  3. Why the heck is it written in lowercase, and not Type? Consistency with str, the class that creates strings objects, and int the class that creates integer objects. type is just the class that creates class objects.
In [5]:
class A:
a = A()
<class '__main__.A'>
<class 'type'>
(<class 'object'>,)
In [6]:
A = type('A', (), {})
a = A()
print(isinstance(a, A), isinstance(a, object), issubclass(A, object))
<class '__main__.A'>
<class 'type'>
(<class 'object'>,)
True True True
In [7]:
def f():
    """My name is f."""
<class 'function'>
<class 'type'>
(<class 'object'>,)
My name is f.

The power of type

  1. Everything is an object in Python, and they are all either instances of classes or instances of metaclasses.
  2. type is the metaclass Python uses to create (i.e. instantiate) all classes and metaclasses, including type itself.
  3. type is actually its own metaclass. This is not something you could reproduce in pure Python, and is done by cheating a little bit at the implementation level.

What is object?

  1. object is the class that all classes inherit from.
  2. All classes including object are subclasses of themselves.
  3. All classes including object are subclasses of object. object.__bases__ is an empty tuple.
  4. All classes except object will have object in __bases__ in a class in their inheritance hierarchy.

Kinds of objects

  • There are two kinds of objects in Python:
    1. Type objects - can create instances, can be subclassed. e.g. type, object, int, str, list.
    2. Non-type objects - cannot create instances, cannot be subclassed. e.g. 1, "hello", [1, 2, 3].
  • type and object are two primitive objects of the system.
  • objectname.__class__ exists for every object and points the type of the object.
  • objectname.__bases__ exists for every type object and points the superclasses of the object. It is empty only for object.


  1. All classes and metaclasses including object are subclasses of object.
  2. All classes and metaclasses including type are instances of type.
  3. All objects including object are instances of object.
In [8]:
issubclass(type, object) # Recap rule #1
In [9]:
issubclass(object, object) # Recap rule #1
In [10]:
issubclass(object, type) # Recap rule #1
In [11]:
isinstance(object, type) # Recap rule #2
In [12]:
isinstance(type, type) # Recap rule #2
In [13]:
isinstance(type, object) # Recap rule #3
In [14]:
isinstance(object, object) # Recap rule #3

In [15]:
from IPython.display import Image, display
display(Image(url='figures/mind_blown.gif', width=400))

The Python Objects Map

Rant: JavaScript

Rant: Ruby

Ruby inherited the Perl philosophy of having more than one way to do the same thing. I inherited that philosophy from Larry Wall, who is my hero actually. Y. Matsumoto, creator of Ruby.

Matz chose to sacrifice first-class functions just so he could make parentheses optional.

Methods are a fundamental part of Ruby's syntax, but they are not values that Ruby programs can operate on. That is, Ruby's methods are not objects in the way that strings, numbers, and arrays are. From The Ruby Programming Language Book.

Thank you!