This parser was livecoded in class between Tuesday and Wednesday (July 12-13 2016). The parser handles the following grammar:
digit ::= '0' | '1' | '2' | '3' | '4' | '5' | '6' | '8' | '9'
int ::= digit digit*
op ::= '+' | '-'
term ::= int | int op term
...with term being the toplevel production.
Blends tokenization with parsing; its input is simply a String.
Assumes that the input string contains only tokens from the above.
This is more correctly a recognizer as opposed to a parser.
That is, it returns whether or not an input can be parsed (as a boolean), as opposed to producing an AST.
While relatively simple, this demonstrates all the core concepts in parsing; more complex grammars merely means the code gets longer, without adding in any fundamentally new ideas.