Regex in Python

Here is a sample file names.txt that we’ll be using regex on.

Love, Kenneth   (555) 555-5555  Teacher, Treehouse  @kennethlove
McFarland, Dave  (555) 555-5554  Teacher, Treehouse
Arthur, King       King, Camelot
Österberg, Sven-Erik       Governor, Norrbotten    @sverik
, Tim        Enchanter, Killer Rabbit Cave
Carson, Ryan  (555) 555-5543  CEO, Treehouse  @ryancarson
Doctor, The       Time Lord, Gallifrey
Exampleson, Example  555-555-5552    Example, Example Co.    @example
Obama, Barack 555 555-5551    President, United States of America @potus44
Chalkley, Andrew    (555) 555-5553  Teacher, Treehouse  @chalkers
Vader, Darth  (555) 555-4444  Sith Lord, Galactic Empire  @darthvader
Fernández de la Vega Sanz, María Teresa     First Deputy Prime Minister, Spanish Govt.

Let’s create our python script file called First things first, you need to import the regex expressions module.

import re

Then we’ll open the names.txt file, read it and close it.

names_file = open("names.txt")
data =

Now we have all the content of the names.txt file in a variable called data which we can now use to run regex on.

Let’s try searching for a name

print(re.match(r'Love', data))
print('Kenneth', data))

r'string' - A raw string that makes writing regular expressions easier.

re.match(pattern, text, flags) - Tries to match a pattern against the beginning of the text., text, flags) - Tries to match a pattern anywhere in the text. Returns the first match.

.match searches for a pattern in the beginning of $data and .search searches for a pattern anywhere in $data.

The result we would get would be something like this:

<_sre.SRE_Match object; span=(0, 4), match='Love'>
<_sre.SRE_Match object; span=(6, 13), match='Kenneth'> 

You can make the script easier to read by defining your search patterns as variables.

first_name = r'Love'
last_name = r'Kenneth'
print(re.match(first_name, data))
print(, data)) 

Escape Characters

Character Meaning
\w any Unicode word character
\W anything that isn’t a Unicode character
\s matches whitespace (spaces, tabs, newlines etc.)
\S anything that isn’t whitespace
\d matches any number 0-9
\D matches anything that isn’t a number
\b matches word boundries (edges of the word)
\B matches anything that isn’t the edge of a word

\w means any Unicode character including numbers, letters both uppercase and lowercase, special characters etc.

Let’s try searching for a phone number now, say (555) 555-5555

number_pattern = r'\(\d\d\d\) \d\d\d-\d\d\d\d'
print(, data))

notice that we have escaped the parantheses ( ). By default, parantheses mean a regex group. To use parantheses in a match pattern, you need to escape them. The output would be:

<_sre.SRE_Match object; span=(40, 54), match='(555) 555-5555'>    


Instead of doing \d\d\d\d\d for 5 numbers, we can specify counts like {3} for somethnig that occurs 3 times and {,3} for something that occurs 0 to 3 times.

Counts Meaning
{3} something that occurs exactly 3 times
{,3} something that occurs 0 to 3 times
{3,} something that occurs 3 or more times
{3,5} something that occurs 3, 4 or 5 times
? something that occurs 0 or 1 times
* something that occurs at least 0 times (there is no upper bound)
+ something that occurs at least once (there is no upper bound)