Build, test, and debug regular expressions
Intended to parse YouTube embed shortcodes in WordPress plugin readme.txt files, matching patterns like [youtube http://www.youtube.com/watch?v=VIDEOID]. It captures the 'youtube' keyword, optional whitespace, and the URL, using a lookahead to require a v= video parameter. Note: the URL portion is wrapped in square brackets, creating a character class rather than a literal pattern, so it actually matches any run of characters from that set (letters, slashes, dots, ?, =, etc.) between '[youtube ' and ']'.