Useful data is often buried in unstructured text — email addresses inside a pasted thread, URLs scattered through a document, order numbers in a support log. Extraction pulls out just the pattern you want, deduplicated and ready to use.
How to use the Text Extractor
- Paste the text containing the data you want
- Choose what to extract — emails, URLs, numbers or hashtags
- Copy the deduplicated list of matches
What can be extracted
- Email addresses — matched on the standard local@domain.tld shape.
- URLs — http and https links, including query strings.
- Numbers — integers and decimals, useful for pulling figures out of reports.
- Hashtags — # followed by word characters, for social content analysis.
Results are deduplicated, so an address appearing forty times in a quoted email chain comes out once.
The limits of pattern matching
Extraction is regex-based, which means it matches shapes rather than validating meaning. A few consequences worth knowing:
An extracted email address is well-formed, not real — it may bounce. Fully validating email syntax by regex is famously impossible; the practical patterns everyone uses accept some invalid addresses and reject some exotic valid ones.
URL extraction stops at whitespace, so a link broken across two lines by word wrapping extracts as two fragments. Numbers do not know about units or thousands separators — '1,500' may extract as 1 and 500 depending on formatting.
Always eyeball the output before feeding it into anything that sends mail.
Frequently asked questions
Are extracted email addresses guaranteed valid?
No. They match the standard pattern, which means they are well-formed — not that they exist or accept mail. Verify before sending anything to them.
Why did a URL come out truncated?
Matching stops at whitespace. If the source text wrapped a long URL across two lines, each fragment matches separately. Join the lines before extracting.
Is my text sent to a server?
No. Extraction runs entirely in your browser, which matters here because the input is often a pasted email thread or support log containing personal data.
Can I extract a custom pattern?
Not with this tool — it handles the four common cases. For an arbitrary pattern, build it in the Regex Tester and use Find & Replace to isolate the matches.