Test, validate and match regular expressions
Matches URLs beginning with http, https, or ftp protocols. It captures three groups: the protocol scheme, the host/domain portion (any characters up to a slash or line break), and optionally the path including its leading slash. With global and multiline flags enabled, it can extract multiple URLs from multi-line text, making it useful for link detection, URL extraction from documents, or basic URL parsing tasks.