Regex Tester

How to use this tool

Enter a pattern without surrounding slashes, choose flags, and supply test text. Select Test to list match positions and captures.

Example

Input

Pattern: \bcat\b | Text: a cat and a cat

Output

2 matches: cat at index 2; cat at index 12

What does this tool do?

Regular expressions describe patterns in text. This tester uses the JavaScript RegExp engine, so its behavior may differ from PCRE, Python or database engines.

Common mistakes and limitations

g finds all matches; i ignores case; m changes line anchors; s lets dot match newlines; u enables Unicode mode. Expensive patterns are stopped after a short timeout. Output is capped at 1,000 matches.

Start with a small matching example

A literal character usually matches itself, a character class such as [0-9] matches one listed character, and a quantifier controls repetition. The + quantifier means one or more, * means zero or more, and {2,4} supplies a range. Parentheses capture a group unless you use a non-capturing form. Begin with text that should match and text that should not; a pattern that matches one happy-path sample is not necessarily a useful validator.

Pattern: ^([A-Z]{2})-([0-9]{3})$
Text: AB-123
Groups: AB and 123

Flags, anchors and escaping

The i flag ignores case, g searches for multiple matches, m changes line-anchor behavior, and s allows a dot to match line terminators. A caret and dollar anchor a pattern to boundaries according to its flags; they do not make every possible Unicode input safe. Enter the pattern itself here without JavaScript slash delimiters. A literal dot needs a backslash. If you later place that same pattern inside a JavaScript string, the string layer may require another backslash.

JavaScript flavor and costly patterns

This tester uses the browser’s JavaScript RegExp implementation. Syntax from PCRE, Python or a database regex engine may behave differently or be rejected. Nested ambiguous repetition can cause excessive backtracking, so testing runs in a disposable worker with a timeout. A timeout is a reason to simplify the pattern, not to raise the limit indefinitely. Matching untrusted large text in production still needs input limits and careful pattern design. Empty matches also need special handling when iterating.

Build a small regression set

After finding a pattern, test empty input, leading and trailing whitespace, repeated delimiters and non-ASCII text. Include near misses rather than only valid examples. If the pattern will run in another environment, repeat those examples in that exact engine. Regex can be useful for a simple shape, but parsing JSON, HTML or a complex protocol with a single pattern often misses nesting and escaping rules.

Frequently asked questions