URL and Email Extractor: Extract Links and E-mail Addresses from Text, HTML or a File

URL and Email Extractor: Get All the Links and E-mail Addresses Out of a Text, a Web Page or a File

You paste a text, or open a file, and the tool gives you a clean list of every web address that is in it, one per line, or, if you choose so, a clean list of every e-mail address. It finds the links for you, so you do not have to read through the text and copy them one by one. Everything happens in your browser: the text and the files you use are never uploaded.

When do you need it

The tool is for the moment when you have a lot of text with links hidden in it and you need only the links:

• A long email, a chat or a document. You want the list of all the links that were sent to you, to open them, check them, save them or send them on.
• The code of a web page. Open a page, press Ctrl+U to see its source code, copy it here, and you get every address that is in the page: links to other pages, images, scripts and style sheets.
• A text file, a log or an export. A notes file, a Markdown article, a CSV table, a server log or a saved page may have hundreds of links among other words. Open the file and get the list.
• A check of the links of a site (SEO). Which sites does a page link to? Write your own domain in the field, and the links to your own site are left out, so you see only the links that lead away from it. Choose "Only the domains" to get a list of the sites without the long addresses.
• Cleaning a list. You have a long list of links with repeats. Keep the option "Show every link once" on, and the repeats are gone.

After that you can check what the links return with the HTTP status code checker, or read the %XX codes in them with the URL decoder and encoder.

Extract e-mail addresses

Switch the field "Find" to "E-mail addresses", and the same box gives you a list of the e-mail addresses of the text. The tool finds an address anywhere in a text, in a letter, a list of contacts, a CSV table, the source of a page or a log, and not only in the links that start with mailto:, because most addresses on the web are written as plain text.

• Without repeats. "Show every address once" removes the repeats, and [email protected] and [email protected] are one address.
• Sorted. "Sort the list from A to Z" puts the addresses in the order of the alphabet.
• Only the domains. Choose "Only the domains" to get the part after the @, for example to see which mail services or companies the addresses belong to.
• Without your own addresses. Write your domain in the field "Your own domain", and the addresses at it and at its subdomains are left out.

An address is a name, the sign @ and a domain with a zone. The name can have letters of any alphabet, digits and the signs . _ % + -, and the zone can be long (.photography) or international (пример.рф). A comma, a bracket, a quote or a full stop after the address is not a part of it. The tool leaves out what only looks like an address: the names of pictures like [email protected], and the user name in a web address like https://user:[email protected].

It does not find an address that is hidden on purpose: name [at] example [dot] com, an address drawn as a picture or built by a script on the page. Think before you collect the addresses of other people: sending them mass mail without their permission is against the law in many countries (GDPR in Europe, CAN-SPAM in the USA and others).

How to use it

1. Choose what to find: links or e-mail addresses.
2. Paste the text into the box, drop a file on it, or press "Open a file". Text, HTML, Markdown, CSV, JSON, XML and log files work, up to 10 MB.
3. The list appears at once under the box, with the number of links (or addresses) found, the number without repeats and the number of different domains.
4. Choose how to show it: every link once or all of them, sorted or in the order of the text, the full links or only the domains.
5. Press Copy, or download the list as a .txt file.

What counts as a link

• Addresses that start with http://, https:// or ftp://. You can also tick the option for addresses that start with www. and have no http://.
• Links in HTML (href="...", src="...") and in Markdown ([text](address)) are found; the quotes and brackets around them are not part of the result.
• The punctuation of a sentence is not part of a link: in "visit http://example.com/page." the full stop is left out. A bracket that belongs to the address, as in https://en.wikipedia.org/wiki/Rust_(programming_language), stays.
• Addresses with non-English letters, a port, a user name, an IP address or localhost are found.

Limitations

• Only full addresses are found. A link like /about or ../page.html in HTML has no site name in it, so it is not listed. An address with no http:// and no www., such as example.com/page, is not found either, because it cannot be told from an ordinary word with a dot.
• In the mode "Links" an e-mail address is not a link, so mailto:[email protected] gives nothing. Switch to "E-mail addresses" to get the addresses.
• In the mode "E-mail addresses" a name like user@localhost, with no dot in the domain, is not an address, and the tool does not check that the mailbox exists. A text that was cut in the middle of an address gives a cut address.
• A link that is broken over two lines is read as two pieces. Addresses where the slashes are written as \/ (as in some JSON) are not found.
• The tool does not check whether a link works, only that it is written like an address.