Test, validate and match regular expressions
Matches a three-token sequence separated by single whitespace characters: a word of exactly 3 characters, a word of exactly 2 characters, and a word of one or more characters, each captured in its own group. With global and multiline flags, it finds all such sequences on every line. Useful for parsing structured codes like 'LAX CA Airport' or extracting abbreviated identifiers from formatted text.