How to match word boundaries with \b in Perl regex?

Understanding Word Boundaries (\b) in Perl Regex

In Perl regular expressions, the \b metasequence is used to match a word boundary, which is a position between a word character and a non-word character (or the start/end of the string). It does not match an actual character but a position, making it very useful when you want to find whole words or ensure a pattern appears at word edges.

In Perl, "word characters" are defined as alphanumeric characters plus underscore ([A-Za-z0-9_]). A \b matches the transition point between these word characters and anything else (spaces, punctuation, start/end of string).

Key points about \b in Perl Regex:

• \b matches a zero-width position, not a character.
• It matches either the start or end boundary of a "word".
• Useful to match whole words without matching them inside other words.
• Opposite is \B, which matches "non-boundary" positions.
• Works in both scalar and list contexts, but context affects what regex returns.

Example: Matching Whole Words with \b

Suppose you want to find the word cat in a string but avoid matching category or scatter. Using \bcat\b will only match standalone cat.

#!/usr/bin/perl
use strict;
use warnings;

my $text = "The cat sat on the scatter plot near the category.";

# Match whole word "cat" using \b
while ($text =~ /\bcat\b/g) {
    print "Found 'cat' at position ", pos($text) - length("cat"), "\n";
}

# Show failure against substrings
if ($text =~ /\bcat/) {
    print "Matches start of a word 'cat' in '$&'\n";
}

if ($text =~ /cat\b/) {
    print "Matches end of a word 'cat' in '$&'\n";
}

This script prints:

Found 'cat' at position 4
Matches start of a word 'cat'
Matches end of a word 'cat'

Here the first /\bcat\b/ matches only the standalone "cat". The other two examples show you can use \b at start or end to match word starts or ends specifically.

Common Pitfalls and Gotchas

• Unicode and word boundaries: Older Perl versions (before 5.14) may not handle Unicode word boundaries correctly with \b. Perl 5.14+ improved Unicode support.
• Using \b inside character classes: Never put \b inside []. For example, [a-z\b] is different and means backspace ASCII (0x08) character, not word boundary.
• Escaping \b in double-quoted strings: If you use \b in double quotes like "\bcat\b", the \b becomes a backspace character unless escaped as "\\b". Use single quotes or double backslashes.

Summary

The \b anchor is a powerful tool to match word boundaries in Perl regex. It helps you match whole words by verifying the position where a word character transitions to a non-word character. Proper usage avoids partial word matches and makes your regex more precise and readable.