1) Like the original article, this leaves unary operators as an exercise for the reader (though it does note so explicitly and provides a small hint)

2) This looks more like Pratt parsing vs. the approach described in the article

3) I was asking the original author