Extract URLs from Text

The tool finds every piece of text that starts with http:// or https:// and takes everything after it up to the next space, line break, quote mark, angle bracket or closing bracket. Each link goes on its own line in the order it first appears, and exact repeats are removed.

Paths, query strings and fragments are kept whole, so https://example.com/page?id=1&x=2 and https://example.com/#frag come out complete. Links wrapped in quotes, angle brackets or round brackets are pulled out cleanly, which makes it work on pasted HTML source, Markdown, emails and chat exports. Non-Latin domains and paths, such as Japanese ones, are kept as typed.

• Links without the scheme. www.example.com and example.net are skipped. If your text has these, use Find and Replace to change www. to https://www. first.
• Capitalised schemes. HTTPS://EXAMPLE.COM is not matched, because the search for http is case-sensitive. Don't lowercase the whole text to fix this, because paths can be case-sensitive. Instead, find https:// and replace it with https:// in Find and Replace with Case sensitive unticked, which rewrites only the scheme.
• Other schemes. ftp://, mailto: and app links are ignored. For email addresses, use Extract Email Addresses.

A full stop or comma straight after a link is treated as part of it. In a real run, this sentence:

gives:

Three things went wrong. The first link carries a trailing comma, which changes the last parameter's value. The second is cut short at the closing bracket, which breaks links like Wikipedia's that contain brackets. And https://example.com. with a full stop is listed separately from the same link without one. Check the end of each line before you use the list.

The duplicate check only removes identical strings. These are all kept as separate entries even though they may lead to the same page:

• https://example.com/page and https://example.com/page/
• https://example.com/page and https://example.com/page?utm_source=x
• http:// and https:// versions of one address

For a link audit, sort the output with Sort Lines Alphabetically so the variants sit together, then decide which ones to merge. Once the trailing punctuation is fixed, Remove Duplicate Lines catches any exact copies that appear.