Python endswith() Tutorial – Can We Use Regular Expressions?

by
5/5 - (1 vote)

While refactoring my Python code, I thought of the following question.

Can You Use a Regular Expression with the Python endswith() Method?

The simple answer is no because if you can use a regex, you won’t even need endswith()! Instead, use the re.match(regex, string) function from the re module. For example, re.match("^.*(coffee|cafe)$", tweet) checks whether a single-line string stored in variable tweet ends with either 'coffee' or 'cafe'.

In fact, I realized that using a regex with the endswith() method doesn’t make sense. Why? If you want to use regular expressions, use functions from the re module. That’s what they were created for! Regular expressions are infinitely more powerful than the endswith() method!

(Reading time 6 minutes — or watch the video to learn about the string.endswith() method)

Python endswith() - Super Simple Tutorial with Twitter Example

Do you want to master the regex superpower? Check out my new book The Smartest Way to Learn Regular Expressions in Python with the innovative 3-step approach for active learning: (1) study a book chapter, (2) solve a code puzzle, and (3) watch an educational chapter video.

How Does the Python endswith() Method Work?

Here’s an overview of the string.endswith method:

str.endswith(prefix[, start[, end]])
prefixrequiredString value to be searched at the beginning of string str.
startoptionalIndex of the first position where prefix is to be checked. Default: start=0.
endoptionalIndex of the last position where prefix is to be checked. Default: end=len(str)-1.

Let’s look at some examples using the Python endswith method. In each one, I will modify the code to show different use cases. Let’s start with the most basic scenario. 

Python endswith() Most Basic Example

Suppose you have a list of strings where each string is a tweet.  

tweets = ["to thine own self be true",
          "coffee break python",
          "i like coffee"]

Let’s say you work in the coffee industry and you want to get all tweets that end with the string "coffee". You’ll use the endswith method with a single argument:

>>> for tweet in tweets:
...   if tweet.endswith("coffee"):
...       print(tweet)
i like coffee

The endswith method has two optional arguments: start and end. You can use these two arguments to check whether a substring from the original string ends with your argument. Need an example that explains both arguments?

Python endswith() Optional Arguments

The endswith method has two optional arguments: start and end. You can use these to define a range of indices to check. Per default, endswith checks the entire string. Let’s look at some examples.

The start argument tells endswith() where to begin searching. The default value is 0, i.e., it begins at the start of the string. So, the following code outputs the same result as above:

>>> for tweet in tweets:
...   if tweet.endswith("coffee", 0):
...       print(tweet)
i like coffee

What happens if we set start=8? 

>>> for tweet in tweets:
...   if tweet.endswith("coffee", 8):
...       print(tweet)

Why doesn’t it print anything? By calling the find() method, we see that the substring 'coffee' begins at index 7.

>>> 'i like coffee'.find('coffee')
7

But tweet.endsswith("coffee", 8) starts looking from index 8. So the result is False and nothing is printed.

Let’s add another argument – the end index – to the last snippet:

>>> for tweet in tweets:
...   if tweet.startswith("coffee", 7, 9):
...       print(tweet)
ï»ż

Nothing is printed on the console. This is because we are only searching over two characters – beginning at index 7 (inclusive) and ending at index 9 (exclusive). But we are searching for 'coffee' and it is 6 characters long. As 6 > 2, endswith() doesn’t find any matches and so returns nothing. 

Now that you know everything about Python’s endswith method, let’s go back to our original question:

Can I Use A Regular Expression with the Python endswith() Method?

No. The endswith() method does not allow for a regular expressions. You can only search for a string. 

A regular expression can describe an infinite set of matching strings. For example, '*A' matches all words ending with 'A'. This can be computationally expensive. So, for performance reasons, it makes sense that endswith() doesn’t accept regular expressions. 

Related article: Python Regex Superpower – The Ultimate Guide

But is it also true that endswith only accepts a single string as argument? Not at all. It is possible to do the following:

Python endswith() Tuple – Check For Multiple Strings 

>>> for tweet in tweets:
...   if tweet.endswith(("coffee", "python")):
...       print(tweet)
coffee break python
i like coffee

This snippet prints all strings that end with either "coffee" or "python". It is pretty efficient too. Unfortunately, you can only check a finite set of arguments. If you need to check an infinite set, you cannot use this method.

What Happens If I Pass A Regular Expression To endswith()?

Let’s check whether a tweet ends with any version of the "coffee" string. In other words, we want to apply the regex ".+coff*". This greedily matches any character one or more times, then 'coff' plus an arbitrary number of characters. Thus we match strings that end with "coffee", "coffees" and "coffe".

>>> tweets = ["to thine own self be true",
              "coffee break python",
              "i like coffee",
              "i love coffe",
              "what's better than one coffee? two coffees!"]

>>> for tweet in tweets:
        if tweet.endswith(".+coff*"):
          print(tweet)
# No output :(

This doesn’t work. In regular expressions, * is a wildcard and represents any character. But in the endswith() method, it just means the star character *. Since none of the tweets end with the literal string "coff*", Python prints nothing to the screen.

So you might ask:

What Are The Alternatives to Using Regular Expressions in endswith()?

There is one alternative that is simple and clean: use the re module. This is Python’s built-in module built to work with regular expressions.

>>> import re
>>> tweets = ["to thine own self be true",
              "coffee break python",
              "i like coffee",
              "i love coffe",
              "what's better than one coffee? two coffees!"]
# Success!
>>> for tweet in tweets:
        if re.match(".+coff*", tweet):
          print(tweet)
i like coffee
i love coffe
what’s better than one coffee? two coffees!

Success! We’ve now printed all the tweets we expected. That is, all tweets that end with "coff" plus an arbitrary number of characters.

Note that this method is quite slow. Evaluating regular expressions is an expensive operation. But the clarity of the code has improved and we got the result we wanted. Slow and successful is better than fast and unsuccessful.

The function re.match() takes two arguments. First, the regular expression to be matched. Second, the string you want to search. If a matching substring is found, it returns True. If not, it returns False. In this case, it returns False for "to thine own self be true" and "coffee break python". It returns True for the rest. 

So let’s summarize the article.

Can You Use a Regular Expression with the Python endswith() Method?

No, you cannot use a regular expression with the Python endswith function. But you can use the Python regular expression module re instead. It’s as simple as calling the function match(s1, s2). This finds the regular expression s1 in the string s2.

Python endswith() List

Given that we can pass a tuple to endswith(), what happens if we pass a list? 

>>> s = 'cobra'
>>> if s.endswith(['a', 'b', 'c']):
        print('yay!')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: endswith first arg must be str or a tuple of str, not list

Python raises a TypeError. We can only pass a tuple to endswith(). So if we have a list of prefixes we want to check, we can call tuple() before passing it to endswith.

>>> if s.endswith(tuple(['a', 'b', 'c'])):
        print('yay!')
yay!

This works well and is fine performance wise. Yet, one of Python’s key features is its flexibility. So is it possible to get the same outcome without changing our list of letters to a tuple? Of course it is! 

We have two options:

  1. any() + list comprehension
  2. any() + map()

The any() function is a way to combine logical or statements together. It takes one argument – an iterable of conditional statements. So instead of writing

if s.endswith('a') or s.endswith('b') or s.endswith('c'):
    # some code

We write

# any takes 1 argument - an iterable
if any([s.endswith('a'),
        s.endswith('b'),
        s.endswith('c')]):
    # some code

This is much nicer to read and is especially useful if you are using many mathematical statements. We can improve this by first creating a list of conditions and passing this to any(). 

letters = ['a', 'b', 'c']
conditions = [s.endswith(l) for l in letters]

if any(conditions):
    # do something

Alternatively, we can use map instead of a list comprehension

letters = ['a', 'b', 'c']
if any(map(s.endswith, letters)):
    # do something

Both have the same outcome. We personally prefer list comprehensions and think they are more readable. But choose whichever you prefer.  

Regex Humor

Wait, forgot to escape a space. Wheeeeee[taptaptap]eeeeee. (source)

Python Regex Course

Google engineers are regular expression masters. The Google search engine is a massive text-processing engine that extracts value from trillions of webpages.  

Facebook engineers are regular expression masters. Social networks like Facebook, WhatsApp, and Instagram connect humans via text messages. 

Amazon engineers are regular expression masters. Ecommerce giants ship products based on textual product descriptions.  Regular expressions ​rule the game ​when text processing ​meets computer science. 

If you want to become a regular expression master too, check out the most comprehensive Python regex course on the planet: