|
This module implements a simple ``standard'' lexical analyzer, presented as a function from character streams to token streams. It implements roughly the lexical conventions of Caml, but is parameterized by the set of keywords of your language.
type token =
Kwd of string
| Ident of string
| Int of int
| Float of float
| String of string
| Char of char
The type of tokens. The lexical classes are:IntandFloatfor integer and floating-point numbers;Stringfor string literals, enclosed in double quotes;Charfor character literals, enclosed in single quotes;Identfor identifiers (either sequences of letters, digits, underscores and quotes, or sequences of ``operator characters'' such as+,*, etc); andKwdfor keywords (either identifiers or single ``special characters'' such as(,}, etc).
val make_lexer: string list -> (char Stream.t -> token Stream.t)
Construct the lexer function. The first argument is the list of keywords. An identifiersis returned asKwd sifsbelongs to this list, and asIdent sotherwise. A special charactersis returned asKwd sifsbelongs to this list, and cause a lexical error (exceptionParse_error) otherwise. Blanks and newlines are skipped. Comments delimited by(*and*)are skipped as well, and can be nested.
Example: a lexer suitable for a desk calculator is obtained bylet lexer = make_lexer ["+";"-";"*";"/";"let";"="; "("; ")"]The associated parser would be a function fromtoken streamto, for instance,int, and would have rules such as:let parse_expr = parser [< 'Int n >] -> n | [< 'Kwd "("; n = parse_expr; 'Kwd ")" >] -> n | [< n1 = parse_expr; n2 = parse_remainder n1 >] -> n2 and parse_remainder n1 = parser [< 'Kwd "+"; n2 = parse_expr >] -> n1+n2 | ...