A chunk is either a token tag, a rule tag, or a span of literal text within a tree pattern.

The method ParseTreePatternMatcher#split(String) returns a list of chunks in preparation for creating a token stream by ParseTreePatternMatcher#tokenize(String). From there, we get a parse tree from with ParseTreePatternMatcher#compile(String, int). These chunks are converted to RuleTagToken, TokenTagToken, or the regular tokens of the text surrounding the tags.