Does "\d" in regex mean a digit?

python

regex-engineering

promises

callbacks

byAnton Shumikhin·Jan 22, 2025

Yeah, it does. \d in regex stands for any digit (0-9). In Python:

import re

# Let's go hunting for digits!
re.findall(r'\d', 'Example123')  # ['1', '2', '3'] just got caught!

But don't underestimate \d; this guy matches more than just your vanilla Latin digits. It matches all Unicode digit characters, even non-Latin ones!

When \d goes beyond 0-9

In Python 3, \d doesn't stop at matching 0-9. Its talent extends to catching Unicode digit characters too, even those non-Latin charlatans (no offense!) Here's \d in action with Eastern Arabic numerals (٠‎ ١‎ ٢‎ ٣‎ ٤‎ ٥‎ ٦‎ ٧‎ ٨‎ ٩):

# \d is now multilingual!
re.findall(r'\d', '١٢٣')  # Returns ['١', '٢', '٣']

Limiting the Scope

But what if you're old school, and all you want is the good old ASCII digits (0-9)? Simply put them in a character class and voila:

# Back to basics.
re.findall(r'[0-9]', 'Example١٢٣123')  # Returns ['1', '2', '3']...Simple as 123!

The Art of Counting Digits

Wanna catch a bunch of digits, say three, in a row? Pair up \d with quantifiers to go on a digit-spree:

# Gotta catch 'em all...well, just the digits.
re.findall(r'\d{3}', 'Example123456')  # Returns ['123', '456']

Clarifying the \d Confusion

Misconceptions like "\d matches only odd-indexed digits" are like wild rumors—not entirely reliable. Let's clear them once and for all:

Odd \d Behavior Unmasked

Unusual behaviors of \d are usually due to text editor issues or regex plugin quirks, not Python or regex. Think of it as the difference between homemade and store-bought cookies.
Terminal highlight settings can play tricks, making the regex look like it's just walked out of a Halloween party. Don't let these phantoms confuse you!
When seeking help, always share the full story (i.e., the full regex). You wouldn't diagnose a car problem over the phone, would you?

The Dynamic Duo: \d and Metacharacters

\d paired with other regex metacharacters leads to exceptional detective work:

A digit followed by a word character:

re.findall(r'\d\w', '12abc 34def')  # ["These aren't droids. They are" '2a', '4d']

A digit fresh at the start:

re.search(r'^\d', 'abc 123') # Extraterrestrial Life? 'None'
re.search(r'^\d', '123 abc') # Extraterrestrial Life! <re.Match object; span=(0, 1), match='1'>