Orpheus

Extract Emails, URLs and Numbers

Pull structured values out of unstructured text.

Options
Remove duplicates
Sort alphabetically
Separate with
Result

Everything is processed in this tab. Nothing you paste is sent anywhere.

Every tool runs entirely in your browser. Your files are never uploaded to a server.

Extraction scans text for anything matching a pattern and lists the matches. Emails are found by the local part, an @ and a domain with a dot; URLs by the http or https prefix.

How to use Email and URL Extractor

  1. Paste your text. An email thread, a web page, a log — structure does not matter.
  2. Choose what to extract. Emails, URLs, domains, IPs, phone numbers or plain numbers.
  3. Pick a separator. Semicolons paste straight into most mail clients.

About extracting emails and URLs from text

Extracting structured values from unstructured text is pattern matching, and the interesting part is where to set the tolerance. An email address can be validated exactly — RFC 5322 defines the grammar — but the regex that implements it faithfully runs to several kilobytes, matches quoted local parts and comments that no mail provider accepts, and still cannot tell you whether the mailbox exists. For pulling addresses out of prose that precision buys nothing, so a permissive pattern that matches the shape people actually write is the better tool. Phone numbers have the opposite problem: there is no grammar to be exact about. Formatting varies by country and by habit, with brackets, spaces, dots and dashes all in use, and numbers frequently written with no separators at all. Any pattern loose enough to catch real numbers also catches years, invoice totals and identifiers, which is why the results are filtered by digit count rather than by shape. URLs sit in between. The http or https prefix makes them easy to find and the ambiguity is at the other end, because a link at the close of a sentence absorbs the punctuation that follows it. Trimming a trailing period, comma or closing bracket is right nearly always and wrong for the rare address that genuinely ends in one. It is worth being explicit about use. Recovering addresses from your own correspondence is ordinary; harvesting them from pages you did not receive and mailing them is unlawful in most jurisdictions, whatever the tool makes easy.

Frequently asked questions

Will it find every email address?
Almost all real ones. The pattern is deliberately permissive rather than RFC-exact, because a fully compliant address regex runs to several kilobytes and matches forms no mail server accepts. Obfuscated addresses written as "name at example dot com" are not found, by design.
Why do some phone numbers get missed or over-matched?
Because phone formatting has no consistent grammar across countries — brackets, spaces, dots and dashes all appear, and so do numbers with no separators at all. Matches are filtered to those with at least seven digits, which removes years and prices while keeping real numbers.
Can I use this to build a mailing list?
Extracting addresses from a page you did not receive is scraping, and sending unsolicited mail to them is unlawful in most jurisdictions under GDPR, CAN-SPAM and similar rules. The legitimate use is recovering addresses from your own correspondence.
Why is a trailing full stop removed from URLs?
Because a link at the end of a sentence absorbs the punctuation. A trailing period, comma or closing bracket is almost always prose rather than part of the address, so it is trimmed — a rare exception is a URL genuinely ending in a bracket.
Is my text uploaded?
No. Everything runs in your browser. That matters here more than on most tools, because the input is frequently a mail thread or a log file containing information that should not be pasted into an unknown server.

Last updated