How links are pulled out of text
Use this when links are scattered through an email, a set of meeting notes or a server log and you need just the addresses. The results can be joined with line breaks or commas so they paste straight into a spreadsheet, and a table groups them by domain so you can see at a glance which sites dominate.
Addresses inside prose usually carry punctuation at the end. Periods, commas and quotation marks are trimmed from the tail, while a closing bracket is only trimmed when the address contains no matching opening bracket. That way a link written inside parentheses does not swallow the closing one, and a link whose own path contains brackets survives intact. Addresses beginning with www are matched too, and you can optionally prefix them with https so every result is clickable.
One limitation is worth knowing. The tool matches patterns with a regular expression, so internationalised domains written in non-Latin scripts, bare addresses with neither a scheme nor a www prefix, and links broken across two lines may be missed or captured only in part. Always skim the result list before relying on it. Removing duplicates merges addresses that differ only in letter case and reports how many were combined. Everything runs in your browser.
Frequently asked questions
Periods, commas and quotation marks are trimmed off. A closing bracket is trimmed only when the address has no matching opening bracket, so links with brackets in the path stay whole.
Not reliably. Regular expressions cannot cover every non-Latin domain form, so addresses in other scripts may be missed or partly captured. Check the list yourself when they appear.