Test, validate and match regular expressions
Matches URLs with an optional http/https scheme, multi-segment domain names (allowing chars like %, @, ~, + in segments), a 1-20 character TLD/port part, optional path segments, and optional query strings or fragments. It also matches the literal placeholder 'website_not_available'. With global and multiline flags, it is well-suited for scanning logs, scraped data, or documents line-by-line to extract or validate website links, including cases where missing sites are marked with a placeholder.