This is a Haxe version of part 1 of my "small guide on writing interpreters".

The structure is very much alike, with minor changes:

• I make reasonable use of Haxe's ADT enums and abstracts for readability.
• Since Haxe has exception support, this uses throw for throwing errors.