Domain Extractor
Domain Extractor — runs in your browser. Inputs are not stored.
Extract domains
Enter values, then press Calculate.
Input stays in the browser and is not sent to the server.
How to use
- Paste logs, notes, or one URL per line into text / URL list.
- Press Extract. Host patterns are found with or without http/https.
- Results are lowercased, a leading
www.is stripped, and the same domain appears once. - Extracted and unique domains show above. In this version they are the same number.
- Reset clears input and the list. No network lookup is performed.
- You can further sort the list alphabetically and dedupe. For Hangul domains, convert to Punycode first, then extract.
Key concepts
Domain extraction picks ASCII hostnames out of body text. http:// or https:// and a leading www. are only hints; the host string is kept in lowercase. Subdomains such as blog.example.com are kept; only www is removed. Paths, queries, ports, and IPv4 are not added to the domain list. Extracted count and unique domains are the same in this version.
The regex is mostly English and digit labels, so Hangul IDN may be missed. Convert to Punycode (xn--) and extract again. It does not harvest the part after @ in emails—use the email extractor for that. Input stays in the browser; there is no DNS lookup. Competitor URL lists or access logs become a unique host inventory.
Example
https://www.Example.com/a, blog.example.com, and www.example.com become two lines: example.com and blog.example.com. Capitals and www are normalized; the same host appears once. 192.168.0.1 does not match the host pattern and is correctly omitted. Paste the result into list sorter for an alphabetical report list. Addresses with and without www merge into one host line.
Related: Email extractor · List sorter · Find and replace
Frequently asked questions
Can I extract only the domain from a URL?
Yes. Scheme and www are ignored; hosts become a unique lowercase list. Path and query are discarded.
Are subdomains kept?
Yes. Only <code>www</code> is removed; other subdomains such as <code>api.example.com</code> stay as their own line.
Are Hangul domains extracted?
This pattern is ASCII-label oriented. For Hangul IDN, convert to Punycode <code>xn--</code> first, then extract.
Are email domains captured too?
If they look like a host in the text they might match, but this is not an address harvester. Use the email extractor for mail.
Are IP addresses treated as domains?
IPv4 does not match this host pattern and is correctly left out. Filter numeric addresses separately.
Is the domain list sent to a server?
No. A regex runs only in the browser with no DNS lookup, and the host list is not transmitted. It is a draft for inventory cleanup.
Last reviewed: 2026-09-04