About the Extract URLs From Text
Export a chat channel or a fortnight of meeting notes and the links you wanted are scattered through hundreds of lines. This tool pulls out every http and https address, drops the repeats, and stacks the rest one per line, which turns a noisy transcript into a reading list in a single paste.
Matching begins at the scheme and runs until whitespace or one of the characters that normally closes a link: an angle bracket, a quotation mark, an apostrophe, a closing round bracket or a closing square bracket. That is what makes Markdown behave, since a link written as bracketed text followed by a parenthesised address yields the address without the bracket, and it is why links sitting inside HTML attributes come out clean.
Sentence punctuation is the catch. A full stop or a comma directly after a link is a legal URL character, so a link at the end of a sentence is captured with the trailing dot attached and needs trimming afterwards. Bare domains with no scheme are ignored on purpose, since matching those would turn every abbreviation containing a dot into a false positive.
For links inside real markup, with their anchor text, the Link Extractor parses the HTML instead of scanning it. To break one address into scheme, host, path and query, use the URL Parser, and URL Validator to check the ones you keep.
How to use
- Paste the text.
- Copy the list of URLs.
- Parse or validate individual links with the related tools.
Common questions
- Does it find links without http?
- No, only http and https URLs are matched, to avoid false positives on ordinary text with dots.
- Does it extract links from HTML tags?
- It finds the URLs in the text including inside tags, but for anchor text and href attributes specifically, use the Link Extractor.
- Are duplicates removed?
- Yes, each unique URL is listed once.