Need every link from a long email thread, a document or a chat export? Copying them one by one is slow and error-prone. Here is how to extract them all at once.
From ordinary text
Paste the text into Extract URLs. It finds every address starting with http://, https:// or www., lists each one on its own line and removes duplicates. Copy or download the list.
The punctuation problem
The hardest part of extracting links from prose is deciding where each URL ends. In the sentence
See https://example.com/docs, or the wiki (https://example.org/Foo_(bar)).
a naive extractor returns https://example.com/docs, with the comma, and https://example.org/Foo_(bar)). with two extra characters. Both links break.
The extractor trims trailing sentence punctuation (full stops, commas, colons, question marks and quotes) and only removes a closing bracket when the URL has no matching opening bracket. That keeps Wikipedia-style addresses intact while dropping the bracket that closes your sentence.
From HTML
In HTML, most links are inside href attributes, and many are relative, like /about. Two approaches:
- If you only need absolute links, paste the HTML source straight into the extractor.
- For every link along with its text, run HTML to Text with Show link URLs on. Each link appears as “text (address)”, readable and ready to review.
Decoding encoded links
Tracking and redirect links often contain the real destination percent-encoded, as in https://example.com/redirect?to=https%3A%2F%2Fexample.org%2Fpage. Paste the link into the URL Decoder to make it readable.
Cleaning the list
- To remove near-duplicates such as the same URL with and without a trailing slash, sort the list with Sort Lines so they sit next to each other, then edit.
- To pull email addresses at the same time, run Extract Emails on the same text.
Limits of pattern matching
Extraction finds text that looks like a link; it can’t check whether the link works. Bare domains without http or www, such as “example.com”, are skipped on purpose, because they are easily confused with file names and abbreviations.