# Regex tester for Next.js & Supabase text. In your browser.

Test JavaScript patterns with highlights, named and numbered groups, replacement text and a short cheat sheet. No uploads — free, no sign-up.

Page: https://saascode.ai/tools/regex-tester

## How to use

1. Enter a JavaScript pattern without slash delimiters, flags, a test string and optional replacement.
2. Choose Test regex. An isolated worker finds matches and performs replacement; expensive patterns are stopped after 750 ms.
3. Inspect highlighted matches, capture groups and UTF-16 indices. Copy matches or the replaced string locally.

## Scope

JavaScript engine only. Browser flag support varies; timeout and display caps are explicit. Regex testing does not prove production ReDoS safety.

## References

- [JavaScript RegExp](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp)

## Build your SaaS

[Build from SaaSCode Base](https://saascode.ai/free)

## Related tools

- [Base64 encode / decode](https://saascode.ai/tools/base64)
- [Markdown editor & table generator](https://saascode.ai/tools/markdown-editor)
- [Unix timestamp converter](https://saascode.ai/tools/unix-timestamp-converter)
- [All free tools](https://saascode.ai/tools)

## FAQ

### Is this free, and does anything leave my browser?

Free, no sign-up. Inputs and outputs stay in your browser, with no processing endpoint or upload. Copy and download are local. Usage events contain only a fixed tool name and action, never your data.

### Is this the same syntax as PostgreSQL regex?

No. This runs your browser's JavaScript RegExp engine. PostgreSQL expressions, PCRE and other engines differ. Test a database expression in your own database before relying on it.

### Why do I see only one match?

Without the g flag, JavaScript exec and replace return or replace the first match. Add g for all nonoverlapping matches. Sticky y requires a match at the current cursor rather than searching forward.

### How do replacement groups work?

Use $1 for the first numbered capture and $<name> for a named capture. $& inserts the full match and $$ inserts a dollar sign. These use native JavaScript string replacement rules, not a function callback.

### What prevents a slow pattern from freezing the page?

Matching and replacement run in a worker that is terminated after 750 ms. Patterns are limited to 2 KB, test strings to 100 KB, replacements to 10 KB and outputs to 1 MB. This is a local testing limit, not a proof a pattern is safe in production.

### What about empty matches and Unicode?

Empty matches use a visible vertical marker and advance correctly over Unicode code points with u or v flags. Offsets count UTF-16 units. Capture indices use the d flag automatically. Unsupported or conflicting flags produce an error; the match display caps at 1,000.