Regular Expressions (Regex) - Quick Reference

A practical guide to regex syntax for BRE and ERE, covering anchors, quantifiers, character sets, and word boundaries for use in shell tools and scripts.

Features

• Covers both Basic (BRE) and Extended (ERE) regex variants
• Ready-to-use patterns for grep, egrep, and shell scripting
• Side-by-side syntax comparison to avoid common pitfalls

Prerequisites

1. A Unix-like system (Linux, macOS, WSL)
2. grep available in $PATH (pre-installed on most systems)
3. Basic familiarity with the terminal

Overview

Regex has two main variants in Unix tools. Knowing which one your tool uses avoids silent errors.

Variant	Command	Notes
BRE	grep	Default; requires \ to activate {, (, |
ERE	grep -E or egrep	Cleaner syntax; +, ?, |, () work natively

Always use double quotes when writing regex patterns on the command line to prevent shell expansion.

Usage

BRE - Basic Regular Expressions

Literal Characters

grep "a" file.txt                # matches the letter "a"

Anchors

grep "^start" file.txt           # line starts with "start"
grep "end$" file.txt             # line ends with "end"
grep "^$" file.txt               # empty lines only

Wildcards

grep "a.b" file.txt              # any single character between a and b
grep "a.*b" file.txt             # any sequence of characters between a and b (including empty)

Quantifiers

* is the only quantifier native to BRE. ? and + require ERE.

grep "ab*c" file.txt             # zero or more "b" between a and c

Character Sets

grep "[abc]" file.txt            # matches a, b, or c
grep "[a-z]" file.txt            # any lowercase letter
grep "[A-Z]" file.txt            # any uppercase letter
grep "[0-9]" file.txt            # any digit
grep "[^abc]" file.txt           # any character except a, b, or c

Grouping

BRE requires backslashes to activate grouping.

grep "\(foo\|bar\)" file.txt     # matches "foo" or "bar" (BRE OR syntax)
grep "\(re\)*start" file.txt     # zero or more repetitions of "re" before "start"

Practical BRE Examples

grep "^$" file.txt               # find all empty lines
grep "^[A-Z]" file.txt           # lines starting with an uppercase letter
grep "[0-9][0-9][0-9]" file.txt  # lines containing at least 3 consecutive digits
grep "\.[a-z]*$" file.txt        # lines ending with a file extension

---

ERE - Extended Regular Expressions

ERE includes all BRE features with a simpler syntax and additional operators. Use grep -E or egrep.

Quantifiers

grep -E "colou?r" file.txt          # zero or one "u" - matches "color" and "colour"
grep -E "ab+" file.txt              # one or more "b"
grep -E "ab*" file.txt              # zero or more "b"

OR Operator

grep -E "cat|dog" file.txt          # matches "cat" or "dog"
grep -E "error|warn|fail" file.txt  # matches any of the three words

Grouping

No backslashes needed for grouping in ERE.

grep -E "(foo|bar)baz" file.txt     # matches "foobaz" or "barbaz"
grep -E "(ha)+" file.txt            # one or more repetitions of "ha"

Repeat Counts

grep -E "a{3}" file.txt             # exactly 3 "a"
grep -E "a{2,4}" file.txt           # between 2 and 4 "a"
grep -E "a{2,}" file.txt            # at least 2 "a"
grep -E "[0-9]{4}" file.txt         # exactly 4 consecutive digits (e.g., year)
grep -E "[0-9]{2,3}" file.txt       # 2 or 3 consecutive digits

Word Boundaries and Character Classes

grep -E "\bword\b" file.txt         # exact word match - avoids partial matches
grep -E "\b[pP]\w+\b" file.txt      # all words starting with "p" or "P"
grep -E "\w+" file.txt              # one or more alphanumeric characters or underscore

\w is equivalent to [A-Za-z0-9_].
\b matches the boundary between a word character and a non-word character.

Practical ERE Examples

# Match an IP address pattern
grep -E "([0-9]{1,3}\.){3}[0-9]{1,3}" file.txt

# Match an email address pattern
grep -E "\w+@\w+\.[a-z]{2,}" file.txt

# Match lines with at least one word starting with a capital letter
grep -E "\b[A-Z]\w+" file.txt

# Match date format YYYY-MM-DD
grep -E "[0-9]{4}-[0-9]{2}-[0-9]{2}" file.txt

# Case-insensitive match with -i flag
grep -Ei "error" file.txt

---

BRE vs ERE - Syntax Comparison

Feature	BRE	ERE
Grouping	\(...\)	(...)
OR operator	|	|
Zero or one	not supported natively	?
One or more	not supported natively	+
Repeat count	\{n,m\}	{n,m}
Character class	[aeiou]	[aeiou]
Alternation in group	\(a|e\)	(a|e)

Key rule: (a|e|i|o|u) requires ERE. Use [aeiou] for BRE compatibility - same result, simpler syntax.

---

Notes

• grep -E and egrep are equivalent - prefer grep -E for portability
• grep -i makes the match case-insensitive (works with both BRE and ERE)
• grep -v inverts the match - prints lines that do NOT match
• grep -o prints only the matched portion, not the full line
• grep -n prefixes output with the line number
• grep -c counts the number of matching lines

# Combine flags
grep -En "\b[A-Z]\w{4,}\b" file.txt    # ERE, words starting uppercase, at least 5 chars
grep -iv "debug" app.log                # case-insensitive, exclude lines with "debug"
grep -on "[0-9]+" file.txt             # print only matched numbers, with line numbers

---

Links

• GNU grep manual
• regex101 - online regex tester
• POSIX BRE/ERE specification

---

Created by Muller Matos