Regular Expressions

A regular expression (regex) is a pattern for matching text: "a digit", "an email-looking thing", "a word at the start of a line". In Ruby, regexes are written between slashes: /pattern/.

Building blocks

Example

text = "Order #1042 shipped on 2025-01-15 to ada@example.com"

p text.match?(/\d{4}-\d{2}-\d{2}/)
p text[/#(\d+)/, 1]
p text.scan(/\d+/)

if (m = text.match(/(?<year>\d{4})-(?<month>\d{2})-(?<day>\d{2})/))
  puts "Year #{m[:year]}, month #{m[:month]}"
end

p text.gsub(/\d/, "#")
p "a1b22c333".split(/\d+/)

Expected output

true
"1042"
["1042", "2025", "01", "15"]
Year 2025, month 01
"Order ##### shipped on ####-##-## to ada@example.com"
["a", "b", "c"]

• match? returns true/false — the fastest check.
• text[regex, 1] returns the first capture group.
• scan returns every match.
• (?<name>...) is a named capture group.
• =~ and case/when also accept regexes.

Validating input: anchor the pattern

ZIP = /\A\d{5}\z/

p "12345".match?(ZIP)
p "12345x".match?(ZIP)
p "ab\n12345".match?(/^\d{5}$/)

true
false
true

In Ruby, ^ and $ mean start and end of a line, not of the whole string — so the last line matches even though the string isn't a ZIP code. For validation, always use \A and \z.

Try patterns interactively in the Regex Tester before putting them in code.

Common mistakes

• Using ^...$ to validate input — a classic security bug in Ruby apps. Use \A...\z.
• Trying to validate emails completely with a regex. A simple "something@something.something" check plus sending a confirmation email is more reliable.

Practice

Extract every hashtag (words starting with #) from "Loving #ruby and #rails today #dev".

p "Loving #ruby and #rails today #dev".scan(/#\w+/)

["#ruby", "#rails", "#dev"]