Python 3.8 came out two and a half years ago and I have yet to really lean in on the walrus operator. Partly because it always seemed like something kinda silly (my use cases) to require a python version bump for, and partly because I really didn't understand it the best. Primarily I have wanted to use it in comprehensions, but I did not really understand how.

Now that Python 3.6 is end of life, and most folks are using at least 3.8 it seems time to learn and use it.

What's a Walrus

:=

The assignment operator in python is more commonly referred to as the walrus operator due to how := looks like a walrus. It allows you to assign and use a variable in a single expression.

This example from the docs avoids a second call to the len function.

if (n := len(a)) > 10:
    print(f"List is too long ({n} elements, expected <= 10)")

Let's get some data

without a walrus

In this example we are going to do a dict comp to generate a map of content from urls, only if their status code is 200. When doing this in a dictionary comprehension we end up needing to hit the url twice for successful urls. Once for the filter and once for the data going into the dictionary.

{
    url: requests.get(url).content
    for url in ["https://waylonwalker.com/", "https://waylonwalker.com/broken"]
    if requests.get(url).status_code == 200
}

Gimme some walrus

using walrus in a dict comp

Using the walrus operator := list comp allows us to only put things into the dictionary that we want to keep, and not hit the url twice.

{
    url: r.content
    for url in ["https://waylonwalker.com/", "https://waylonwalker.com/broken"]
    if (r := requests.get(url)).status_code == 200
}

FIN

The walrus is a nice to have option to save on extra function/network calls, and micro optimize your code without adding much extra.