Using Regular Expressions in Python

If you need to do pattern matching on a string within Python, using a regular expression will be the best way to do it. Here are some basic examples to help you with this.

Import Library

To use any of the regular expressions you’ll need to add this library:

import re

Regex Compiler

Create the regex object using the compile() method.

phone_regex = re.compile(r'\d\d\d-\d\d\d-\d\d\d\d')

This creates or defines the regex object which we can then use against a string.

The r here indicates a raw string.

I recommend building and testing your regex with a tool like http://regexpal.com.

Matching Regex Objects

Once the regex object is defined you can use the search() method like this:

phone_regex = re.compile(r'\d\d\d-\d\d\d-\d\d\d\d')
search_results = phone_regex.search('My number is 310-429-4019.')

The results of this are:

>>> print('Results: ' + search_results.group())
Results: 310-429-4019

The search() method will return the first match it finds in the string.

The results are stored in the group() method of your variable.

Using Parenthesis to Form Groups

If you need to match on a larger string but want to only extract a portion of the string, you can use parenthesis to make more groups.

phone_regex = re.compile(r'(\d\d\d)-(\d\d\d-\d\d\d\d)')
search_results = phone_regex.search('Call me at 310-429-4019.')

This has created 4 groups shown here:

>>> print('Results: ' + search_results.group())
Results: 310-429-4019
>>> print('Results: ' + search_results.group(0))
Results: 310-429-4019
>>> print('Results: ' + search_results.group(1))
Results: 310
>>> print('Results: ' + search_results.group(2))
Results: 429-4019

Without Regex Compile

An alternative way to do this is to skip the regex compile() method and use this syntax instead:

my_string = "Name: Fred, Occupation: Bronto Crane Operator"
parse_data = re.search('Name: (\w*), Occupation: (.*)', my_string)

Results:

>>> print(parse_data.group())
Name: Fred, Occupation: Bronto
>>> print(parse_data.group(1))
Fred
>>> print(parse_data.group(2))
Bronto Crane Operator

Ignore Case

Use the re.IGNORECASE option to ignore capitalization. Example:

fred_regex = re.compile(r'fred', re.IGNORECASE)
search_results = fred_regex.search('His name is Fred.')

This produces the following results:

>>> print(search_results.group())
Fred

Pro Tip: You can use the shortcut re.I instead of re.IGNORECASE.

Using findall()

While the search() method finds the first occurrence, the findall() method returns every occurrence.

phone_regex = re.compile(r'\d\d\d-\d\d\d-\d\d\d\d')
search_results = phone_regex.findall('You can call me at 310-429-4019 or 310-777-4777.')

This turns the search_results variable into a list and has the following results:

>>> print(search_results)
['310-429-4019', '310-777-4777']
>>> print(search_results[0])
310-429-4019
>>> print(search_results[1])
310-777-4777

Substituting Using a Regex

You can use the sub() method to substitute a string matching a regex. Example:

phone_regex = re.compile(r'\d\d\d-\d\d\d-\d\d\d\d')
scrubbed_phone_numbers = phone_regex.sub('XXX-XXX-XXXX', 'My number is 310-429-4019.')
print(scrubbed_phone_numbers)

This prints My number is XXX-XXX-XXXX..