Regex Tester

How to test a regular expression

1. Type your pattern in the box between the two slashes, for example \d{3}-\d{4}. Enter only the pattern: the slashes are already drawn for you, so typing your own would make them part of the pattern.
2. Pick the flags you need (g is on by default) and paste your test text below. If the pattern is invalid, the engine's own error message appears in red.
3. Matches are highlighted instantly, with a numbered list showing each match, its starting index and the value of every capture group. Copy matches copies the matched text, one match per line.

Understanding the flags

The four flags offered are: g (global) finds every match instead of stopping at the first; i (ignore case) treats A and a the same; m (multiline) makes ^ and $ anchor to the start and end of every line, not just the whole string; and s (dot all) lets . also match newline characters. You can combine them freely, and the active flags are shown after the closing slash. The tool does not offer the u (unicode), y (sticky) or d flags, which matters for the Unicode note below.

Worked examples

• Phone fragments: \d{3}-\d{4} on Call 555-0142 or 555-0199 gives two matches, at index 5 and index 17.
• Capture groups: (\d{4})-(\d{2})-(\d{2}) on 2026-10-07 gives one match with $1=2026, $2=10, $3=07.
• Greedy vs lazy: on <b>bold</b> the pattern <.+> matches the entire string as one match, because .+ grabs as much as it can. Changing it to <.+?> makes the quantifier lazy and finds two matches, <b> and </b>.
• Lines with m: with the m flag on, ^\w+ matches the first word of every line. Without m it matches only the first word of the whole text.

Capture groups

Anything you wrap in parentheses becomes a capture group. For example, (\w+)@(\w+) matched against an email captures the user part as group 1 and the domain part as group 2. The match list shows $1, $2 and so on for each match, and a group that did not take part in the match, such as one inside an unused alternative, shows as undefined. Named groups written as (?<year>\d{4}) work, but are listed by number in this tool.

JavaScript flavor: differences from other languages

• Use ^ and $ for the start and end of the text. \A and \Z from other languages do not mean that in JavaScript.
• Python-style named groups (?P<name>…) are invalid; JavaScript uses (?<name>…).
• Possessive quantifiers and atomic groups, found in PCRE and Java, do not exist in JavaScript.
• Without the u flag, Unicode property escapes such as \p{L} do not work as property classes, and emoji or other characters outside the basic plane are treated as two units. If you need those, test the same pattern in your target environment.

Common mistakes

• Forgetting to escape special characters. A literal dot must be \.; an unescaped . matches any character, so 1.5 also matches 125.
• Forgetting the g flag and wondering why only one match is found.
• Catastrophic backtracking. Nested quantifiers such as (a+)+$ against 30 letters a followed by a single ! can take billions of steps. This tester runs on the page's main thread with no timeout, so such a pattern can freeze the tab until the browser gives up. Test risky patterns on short inputs first, and avoid nested repetition of the same characters in production code.
• Typing escape sequences in the test box. The test string is raw text, so typing \n there inserts a backslash and the letter n, not a line break. Press Enter to create a real newline.

Limits of this tool

Matching stops at 5,000 matches, and only the first 300 appear in the numbered list, although the highlighted text shows everything found. A zero-length match, such as \b or ^, is drawn as a thin bar. There is no replace or substitution preview, so check your replacement string in your own code.

Is it private and safe?

Yes. Toolyard runs the regular expression inside your browser on your own device using the built-in JavaScript engine. Your pattern and test text are never uploaded anywhere, which makes it safe for log files, code snippets and private data.

Related guides

See the step-by-step guide to testing a regular expression, learn how backslashes work in JSON escape characters, or use the URL encoder when your matches need to go into a link.