People who are new to the Python programming language can get a bit confused about the difference between “==” (equality) and Python’s keyword “is” (identity). I have even seen experienced programmers who will find the difference subtle enough that they will introduce logic errors in their code do to a misunderstanding between the two. In this article, we will look at this interesting topic.
Equality in Python
Many programming languages have the concept of equality and several use the double equals sign (“==”) to designate this concept. Let;s take a look at equality in action:
>>> num = 1 >>> num_two = num >>> num == num_two True
Here we create a variable that we call num and assign it to the integer 1. Next we create a second variable called num_two and assign it to the value of num. Finally we ask Python if num and num_two are equal. In this case, Python tells us that this expression is True.
Another way to think of equality is that we are asking Python if two variables contain the same thing. In the example above, they both contain the integer 1. Let’s see what happens when we create two lists with the same values:
>>> list_one = [1, 2, 3] >>> list_two = [1, 2, 3] >>> list_one == list_two True
This worked out the way we expected.
Now let’s see what happens if we ask Python about their identity:
>>> num is num_two True >>> list_one is list_two False
What happened here? The first example returned True, but the second returned False! We will look into that in the next section.
Identity in Python
When you ask Python about whether one object is the same as another object, you are asking if they have the same identity. Are they actually the same object? In the case of num and num_two, the answer is yes. Python provides an easy way to prove it via it’s built-in id() function:
>>> id(num) 10914368 >>> id(num_two) 10914368
The reason that these two variables share the same identity is because we told Python they should back when we assigned num to num_two (i.e. num_two = num). If you come from C or C++, you can think of the identity as a pointer where num and num_two are both pointing to the same place in memory. If you use Python’s id() function on the two list objects, you will quickly see that they have different identities:
>>> id(list_one) 140401050827592 >>> id(list_two) 140401050827976
Thus when you ask Python the question “list_one is list_two”, you receive False. Note that you can also ask Python if one object is not another object:
>>> list_one = [1, 2, 3] >>> list_two = [1, 2, 3] >>> list_one is not list_two True
Let’s take a moment to find out what happens when you mix equality and identity up.
Mixing it up
I know when I was starting out as a Python programmer, these sorts of things would lead to silly mistakes. The reason is that I would see recommended statements like this one:
if obj is None: # do something call_function()
So I would assume naively that you could do something like this:
>>> def func(): return [1, 2, 3] >>> list_one = [1, 2, 3] >>> list_two = func() >>> list_one is list_two False
Of course, that doesn’t work as I now have two different objects with different identities. What I wanted to do here was this:
>>> list_one == list_two True
Another issue that is tangential to this one is when you create two variables that point to the same object, but you think you can work on them independently of each other:
>>> list_one = list_two = [1, 2, 3] >>> list_one == list_two True >>> list_one is list_two True >>> list_two.append(5) >>> list_one [1, 2, 3, 5]
In this example, I created two variables that both pointed at one object. Then I tried adding an element to just list_two. What a lot of beginners don’t realize is that they just added that element to list_one as well. The reason for this is that both list_one and list_two are pointing at the exact same object. This is proven when we asked Python is list_one is list_two and it returned True.
Hopefully by now you understand the differences between equality (==) and identity (is) in Python. Equality is basically just asking if the contents of the two object are the same and in the case of lists, it needs to be in the same order as well. Identity in Python refers to the object you are referring to. In Python, the identity of an object is a unique, constant integer (or long integer) that exists for the length of the object’s life.
- Python documentation on the id() function
- Python documentation of “is”
- StackOverflow: Understanding Python’s â€œisâ€ operator
6 thoughts on “Python 101: Equality vs Identity”
I think you should mention __hash__ and __eq__. Also what is value objects.
Thanks for sharing. Does list_one is list_two equivalent to id(list_one) == id(list_two)?
Yeah, they should be the same in that case. The first is checking the identity while the second is verifying the id number is the same
This is probably a good idea. I am researching it to find a good way to add it without adding a lot of scope to the article.
Great post Mike. This stuff always trips me up, but the way you explained it using the id() function made it make more sense to me.
I now understand that objects are the same if they have the same id.
One way to clear up confusion is to refer to object names as “identifiers”, as opposed to the term “variable” when tutoring/teaching someone python. Calling an object’s name an identifier can make it clear to a developer that an object can be identified in more than one way because the identifier is simply a name used to identify the function, variable or object. ~VN
Comments are closed.