URL Extractor
Paste an article, an email, a reference list or a page of notes, and the tool lists every web link it finds, once each. Switch on Domains only to get just the host names, then copy the result one per line or comma-separated. Free, no sign-up, and the text never leaves your browser.
Options
How to Use the URL Extractor
- Paste your text. Anything that contains links: a reference list, an email, a document you copied from, a page of notes. The list updates as you type.
- Choose full URLs or domains. Leave Domains only unticked to keep each complete link, or tick it to reduce every link to its host name.
- Pick the output format. One per line suits a spreadsheet column; Comma-separated gives a single line.
- Copy the result. The URLs found (or Domains found) card tells you how many unique entries there are.
What Is Recognised as a Link
The tool picks up anything that starts with http://, https:// or www. and runs on until the next space, angle bracket or quotation mark. Punctuation that clings to the end — a full stop, comma, semicolon, colon or quotation mark, or a closing bracket with no opening partner inside the link — is trimmed off, so a link at the end of a sentence or inside brackets comes out clean.
Repeats are removed, and the comparison ignores case, so https://Example.com/a and https://example.com/a count as one link; the first spelling is kept. Remember that the path of a real URL can be case-sensitive, so two links that differ only in capitals in the path may be different pages, and only the first would be listed.
With Domains only on, each link is reduced to the host name: https://example.com/guide?id=4 becomes example.com, and a port number is dropped. A www. prefix is kept, so www.example.org and example.org are listed as two domains. Domains are de-duplicated too, so a hundred links to one site give a single line.
A Worked Example
Input
See https://example.com/guide?id=4#top, www.example.org/page. Also http://sub.example.co.uk:8080/a/b; and https://EXAMPLE.com/guide?id=4#top. Not links: example.com and ftp://files.example.com/x.zip
Output (full URLs)
https://example.com/guide?id=4#top www.example.org/page http://sub.example.co.uk:8080/a/b
Output (Domains only)
example.com www.example.org sub.example.co.uk
URLs found: 3. Four links matched, but the capitalised second copy of the example.com link is dropped as a repeat. The trailing comma, full stop and semicolon are gone. A bare example.com with no http:// or www. is not recognised, and neither is an ftp:// address.
Limits to Know About
- Closing brackets are kept only when they balance. A Wikipedia address such as Citation_(style) keeps its bracket because the link contains the opening one too, while a bracket that merely closes a parenthesis around the link is dropped. A link that ends in an unmatched bracket is shortened by one character, so check addresses like that by hand.
- Only three prefixes count. Addresses without http://, https:// or www. are not matched, even if they look like domains.
- It does not test the links. A listed address may be dead or redirect elsewhere. The tool finds links; it does not visit them.
Which Tool Do You Need?
| If you want… | Use |
|---|---|
| Links or domains pulled out of a text | This page |
| Email addresses pulled out of a text | Email Extractor |
| A citation built from one web page's address | Website Citation Generator |
| BibTeX built from a URL | URL to BibTeX |
| Only the lines of a list that contain a given site | Line Filter |
Frequently Asked Questions
- How do I extract all URLs from a text?
- Paste the text into the box. Every link that starts with http://, https:// or www. is listed straight away, with repeats removed, and a count under URLs found. Use Copy Result to take the list.
- How do I get a list of domains from text?
- Tick Domains only. Each link is reduced to its host name and repeated domains are collapsed, so you get one line for each site mentioned. Subdomains such as sub.example.co.uk are kept in full.
- Why is a link missing from the results?
- The usual cause is that it has no http://, https:// or www. in front, as in a bare example.com, or that it uses another scheme such as ftp://. Add the prefix to the text and it will be found.
- When is a closing bracket kept at the end of a link?
- A closing bracket stays if the link holds a matching opening bracket, as in Citation_(style). Otherwise it is stripped, like a full stop or comma, so that links inside parentheses come out clean. Only an address that really ends in an unmatched bracket is shortened by one character.
- Are links in different capitals treated as duplicates?
- Yes, case is ignored when repeats are removed, and the first spelling is kept. If a site has case-sensitive paths, two such links would be reduced to one.
- Does it check whether the links work?
- No. It only recognises text that looks like a link. To check a source, open it in a browser, or for citations run it through a citation generator.
- Is my text uploaded?
- No. The scan happens in your browser and the text is not sent or stored.