Data Extractor
Extract emails, URLs, domains, IPs, MAC addresses, dates, phone numbers, hashtags, mentions, file paths or hashes from any text, deduplicated.
No matches found
What the data extractor does
Paste a wall of text — an email thread, a chat export, a CSV dump, a web page, a server log — and this tool pulls out just the pieces you care about. The dropdown offers eleven kinds of thing to look for: Email addresses, URLs, Domains, IP addresses, MAC addresses, Phone numbers, Dates, Hashtags, @ mentions, File paths and Hashes (MD5/SHA), plus a Custom pattern (regular expression) of your own. Pick one and it instantly lists every match, or pick All types and the whole lot comes back at once, grouped under a heading each with its own count. Matches are shown in the order they appear unless you Sort them A → Z or Z → A, and duplicates are removed automatically, so you get a clean, copy-ready list instead of scrolling through the noise by hand. “Separate with” puts one match per line, or joins them with a comma, a semicolon, a tab or a separator you type; Filter keeps only, or drops, the matches containing the words or domains you list. Tick “Lower-case, and merge items that differ only in case” and Ada@Example.com and ada@example.com become one — a URL has only its scheme and host lower-cased, since its path can be case-sensitive, and a file path is left as written. With Email addresses, URLs or Domains picked, a folded Domains table under the result lists each domain with how many of the matches came from it, most first. Under the result, “Send the result to” opens the URL & IP Defanger and Refanger in this tab with every match in it, one per line and without the group headings, ready to defang for a ticket or a report; while it extracts email addresses, Email Normalizer is offered beside it.
How the matching works
Each type uses its own pattern. Emails match a local part, an @, and a dotted domain. URLs cover http, https and www forms, with trailing sentence punctuation trimmed off. Domains find bare host names on their own, using look-around so the domain inside an email address or a URL is not counted twice. Phone numbers accept spaces, dashes, dots and parentheses and are kept only when they carry a realistic 7 to 15 digits and are not already a date or an IP address. Dates cover ISO 8601 with an optional time as well as the common slash forms. MAC addresses want six hex pairs with one consistent separator. File paths cover Windows drive and UNC paths and Unix absolute, home and relative ones. Hashes are hex digests of exactly 32, 40 or 64 characters — MD5, SHA-1 and SHA-256. Hashtags and mentions are anchored so a C# in prose or the @ inside an email address is not mistaken for one. IP addresses are validated octet by octet so 999.1.1.1 is skipped. Everything runs live as you type or edit.
Frequently asked questions
Does it remove duplicate matches?
Yes. Every result list is deduplicated while keeping the original order (until you choose a Sort), so if the same email or URL appears five times you see it once, in the position it first showed up.
Which types of data can it find?
Eleven: Email addresses, URLs, Domains, IP addresses, MAC addresses, Phone numbers, Dates, Hashtags, @ mentions, File paths and Hashes (MD5/SHA). Pick one from the dropdown and the list updates instantly for the text in the box, or pick All types to run every extractor at once and get the results grouped by kind. For anything else, pick Custom pattern (regular expression) and type your own.
Can I extract with my own regular expression?
Yes. Pick Custom pattern (regular expression) and type it in the Pattern box — INV-\d+ pulls every invoice number. It runs through the same engine as the regex tester, so a pattern that could backtrack catastrophically is refused before it runs rather than freezing the tab, and on a very long text the page says how much of it was searched. Duplicates are removed as for every other type, and Sort, Separate with and Filter work on its results too.
Can I download the matches as a CSV file?
Yes. Download CSV, beside the text download, writes the list as it stands after Sort and Filter, each item on a row of its own under a header row in the page’s language. Email addresses, URLs and Domains get a domain column beside each value, and in the All types view a type column says which group each row came from. A value holding a comma is quoted, and a mention or phone number that opens with @ or + cannot run as a spreadsheet formula.
Can it show which domains the addresses came from, or merge addresses that differ only in case?
Yes. With Email addresses, URLs or Domains picked, a folded Domains table under the result lists each domain with how many of the matches came from it, most first. Tick “Lower-case, and merge items that differ only in case” and Ada@Example.com and ada@example.com become one entry; a URL has only its scheme and host lower-cased, since a server may read its path by case, and a file path is left exactly as written.