Remove all special characters, punctuation and spaces from string

python

string-manipulation

data-cleaning

performance-optimization

byAlex Kataev·Mar 4, 2025

Easily remove non-alphanumeric characters from a string in Python using regex:

import re
cleaned = re.sub(r'\W+', '', 'String!@ with# %special* chars&')
print(cleaned)  # 'Stringwithspecialchars'

In the above case, re.sub(r'\W+', '', string) is hunting down non-word characters (\W+), ruthlessly wiping out punctuation and spaces, leaving behind only a crystal-clear alphanumeric sequence.

Detailed approaches and their execution times

Different scenarios might require different strategies for removing special characters from strings. Here are some efficient solutions for disparate situations.

Regex-free strategy: `str.isalnum()`

If you view regex as overkill or prioritize readability, opt for the built-in str.isalnum() method:

# Like a bouncer at a club, only alphanumeric characters will pass!
cleaned = ''.join(e for e in string if e.isalnum())
print(cleaned)  # 'Stringwithspecialchars'

This approach retains the original character order while being remarkably easy to read and understand.

Strategies for bulky strings

For voluminous strings or performance-critical applications, employ [*filter(str.isalnum, string)] in Python 3.5 and beyond:

# It's not just cleaning, it's performing a magic trick!
cleaned = ''.join(filter(str.isalnum, string))
print(cleaned)  # 'Stringwithspecialchars'

Just like a magic show, this unpacking and filter technique can wow you with amazing performance jumps.

List comprehension for the win!

According to initial benchmarks, list comprehension may race ahead of generator expressions:

# It's like jogging through the string and picking up litter
cleaned = ''.join([c for c in string if c.isalnum()])
print(cleaned)  # 'Stringwithspecialchars'

This boost in speed might be distinguishable with larger datasets.

Keeping the spaces intact

Sometimes, you would want to preserve all spaces, and remove only the special characters. Here's a handy modification:

# Like a good gardener, only gets rid of the weeds.
cleaned = re.sub(r'[^\w\s]', '', 'String!@ with# %special* chars&')
print(cleaned)  # 'String with special chars'

Timing various methods with `timeit`

To make a calculated choice, you might want to compare performance using the timeit module:

import timeit

# It's like watching two athletes race!
time_regex = timeit.timeit("re.sub(r'\W+','',string)", setup='import re; string="String with special chars!"')
time_isalnum = timeit.timeit("''.join(e for e in string if e.isalnum())", setup='string="String with special chars!"')

print(f"Regex: {time_regex}s vs str.isalnum(): {time_isalnum}s")

The execution times will guide you towards the optimum choice.

Important considerations and advanced methods

Edge Case Alert!

In the realm of string cleaning, being prepared for edge cases ensures robustness:

Accented characters: Default behavior removes these. But not if you incorporate Unicode properties in Regex!
Underscore handling: Even though \W picks out "_" as a non-word character, you might prefer to retain it.

Special considerations for Data Cleaning

Not only about scrubbing off characters, data cleaning is also about understanding the context and purpose of data use:

Data Integrity: An absolute must. The right data retention avoids corrupting datasets.
Word Boundaries: For further computerized linguistic analysis, spaces are important.

Looking beyond: `str.translate()`

For complex character mapping, don't miss out on str.translate() and maketrans(). They allow character substitutions:

# Like a language interpreter
trans = str.maketrans('', '', string.punctuation)
cleaned = string.translate(trans)
print(cleaned)  # 'String with special chars'