Extract emails, URLs, phone numbers and more from text
Pull email addresses, links, numbers or matching references out of a pasted document. Review the matches, remove repeats and copy a list or download it as TXT, CSV or JSON.
Runs in your browser — your input is not uploaded to ToolsNow.
Optional. Added to whatever is selected above.
The text you enter is processed locally in your browser and is not uploaded to ToolsNow.
Site files and analytics use separate network connections. Ads are off. Privacy & Cookies
Check the matches. An email-shaped string may not be a working address. A number may be an order reference or date rather than a phone number. These are possible matches, not verified contacts.
Extracted URLs are shown as text, not clickable links. Large inputs or complex custom patterns may reach a processing limit. Read any warning before treating the results as a complete list.
How to use
- Paste Text to search. Use the text from an email thread, a log or a column of notes. This tool searches pasted text rather than fetching a web page.
- Choose What to extract. Select one or more types. For references such as INV-0042, try INV-\d{4} in Custom pattern (regular expression), without surrounding slashes.
- Review the matches. Remove duplicates is on by default. Check Ignore case when comparing, especially for case-sensitive URL paths or identifiers. Show surrounding text adds context to built-in matches.
- Copy or download. Copy results and .txt give one value per line. The .csv and .json downloads include types and counts as well. Wait for custom matching to finish before exporting.
Questions
Are the extracted contacts verified?
No. The patterns recognise likely formats without checking deliverability or ownership. [email protected] has the shape of an email address, for example. Phone-like numbers are filtered to 7–15 digits, but can still be order references or other numbers; review the context before using them.
Why was a link, hashtag or IP address missed?
The built-in patterns cover limited formats. Links need http://, https://, ftp:// or www.; a bare example.com is not enough. Hashtags use an ASCII-letter pattern, and the IPv6 pattern does not cover every compressed form and can return partial matches. Check those results against the original text.
How are duplicates compared?
Matches with the same type and value are merged, keeping the first spelling and a count. Ignore case when comparing also merges values such as [email protected] and [email protected]. Turn it off if case matters: URL paths such as /Report and /report may lead to different resources.
What can a custom pattern return?
It returns the full text of each non-empty JavaScript regex match, not just a capture group. INV-\d{4} returns INV-0042. Custom matches are appended after the built-in results; the Order setting and surrounding-text option do not apply to those appended rows, and the Total matches summary counts only built-in hits.
What happens if a custom pattern is slow?
Custom matching runs in a Web Worker with a two-second timeout and a 20,000-match cap. A timeout or syntax problem is reported; if workers are unavailable, the pattern is not run. Built-in extraction has a separate 100,000-match guard per type and still runs on the page thread, so large inputs can slow the page.
Does exporting include more than the visible 500 rows?
Yes. Copies and downloads include all collected results, including rows below the display limit. A match-limit warning means collection was cut short, so an export is not necessarily every match in the original document. Links are exported as text, without being opened.
Is the searched text uploaded?
The input is searched in your browser without uploading it. That refers to the document contents, not page analytics; those are described separately in the site’s privacy policy.
Published by ToolsNow. How tools and sources are checked · Report a problem