| 1 | (*****************************************************************************) |
| 2 | (* src: Jon Harrop. |
| 3 | * |
| 4 | * "Certain applications are extremely well suited to functional |
| 5 | * programming and parsing is one of them. Specifically, the ability to |
| 6 | * write functional combinators that allow parsers for everything from |
| 7 | * integers up to symbolic expressions to be composed is more general |
| 8 | * and provides more opportunity for code reuse than the use of |
| 9 | * conventional parser generators such as ocamllex and ocamlyacc. This |
| 10 | * article explains how parser combinators may be designed and |
| 11 | * implemented in OCaml, using the standard example of a calculator." |
| 12 | * |
| 13 | * pad: a few bugfixes. I also put more restrictive and descriptive types. |
| 14 | * |
| 15 | *) |
| 16 | |
| 17 | (*****************************************************************************) |
| 18 | |
| 19 | (* A generic parser takes a list of stuff (either char for lexical |
| 20 | * parser or tokens for grammar parser) and return something and the |
| 21 | * remaing list of stuff. *) |
| 22 | type ('a, 'b) genp = 'a list -> 'b * 'a list |
| 23 | val val_of_parser : 'b * 'a list -> 'b |
| 24 | |
| 25 | (* lexer = parser of char list *) |
| 26 | (* type 'a lexer = (char, 'a) genp *) |
| 27 | |
| 28 | (* grammer = parser ot tokens *) |
| 29 | (* type 'a pparser = (token, 'a) genp *) |
| 30 | |
| 31 | |
| 32 | val ( ||| ) : ('a, 'b) genp -> ('a, 'b) genp -> ('a, 'b) genp |
| 33 | (* ('a -> 'b) -> ('a -> 'b) -> 'a -> 'b *) |
| 34 | val ( +++ ) : ('a, 'b) genp -> ('a, 'c) genp -> ('a, 'b * 'c) genp |
| 35 | (* ('a -> 'b * 'c) -> ('c -> 'd * 'e) -> 'a -> ('b * 'd) * 'e *) |
| 36 | |
| 37 | val many : ('a, 'b) genp -> ('a, 'b list) genp |
| 38 | (* ('a -> 'b * 'a) -> 'a -> 'b list * 'a *) |
| 39 | |
| 40 | val ( >| ) : ('a, 'b) genp -> ('b -> 'c) -> ('a, 'c) genp |
| 41 | (* ('a -> 'b * 'c) -> ('b -> 'd) -> 'a -> 'd * 'c *) |
| 42 | |
| 43 | (* was called 'some', but confusing *) |
| 44 | val pred : ('a -> bool) -> ('a, 'a) genp |
| 45 | (* ('a -> bool) -> 'a list -> 'a * 'a list *) |
| 46 | |
| 47 | val a : 'a -> ('a, 'a) genp |
| 48 | (* 'a -> 'a list -> 'a * 'a list *) |
| 49 | |
| 50 | val several : ('a -> bool) -> ('a, 'a list) genp |
| 51 | (* ('a -> bool) -> 'a list -> 'a list * 'a list *) |
| 52 | |
| 53 | |
| 54 | module Abstr : sig |
| 55 | type t |
| 56 | val x : t |
| 57 | end |
| 58 | |
| 59 | val fin : ('a, Abstr.t) genp |
| 60 | (* 'a list -> Abstr.t * 'b list *) |
| 61 | |
| 62 | |
| 63 | val digit : char -> bool |
| 64 | val alpha : char -> bool |
| 65 | val symbol : char -> bool |
| 66 | val alphanum : char -> bool |
| 67 | val space : char -> bool |
| 68 | |
| 69 | val alphanum_underscore : char -> bool |
| 70 | val alphanum_minus : char -> bool |
| 71 | val alphanum_under_minus : char -> bool |
| 72 | |
| 73 | val collect : char * char list -> string |
| 74 | val list_of_string : string -> char list |
| 75 | |
| 76 | |
| 77 | (*****************************************************************************) |
| 78 | type token = |
| 79 | | IDENT of string |
| 80 | | KWD of string |
| 81 | | INT of string |
| 82 | | SYM of string |
| 83 | | STR of string |
| 84 | |
| 85 | val string_of_token : token -> string |
| 86 | |
| 87 | type lexer = (char, token) genp |
| 88 | |
| 89 | val rawident : lexer |
| 90 | (* char list -> token * char list *) |
| 91 | val rawnumber : lexer |
| 92 | (* char list -> token * char list *) |
| 93 | |
| 94 | val rawsymbol : lexer |
| 95 | |
| 96 | (* not space, not digit *) |
| 97 | val rawkeyword : lexer |
| 98 | (* char list -> token * char list *) |
| 99 | |
| 100 | val rawstring : lexer |
| 101 | |
| 102 | val lex_gen : lexer -> string -> token list |
| 103 | |
| 104 | (*****************************************************************************) |
| 105 | val token : lexer |
| 106 | (* char list -> token * char list *) |
| 107 | val tokens : (char, token list) genp |
| 108 | (* char list -> token list * char list *) |
| 109 | |
| 110 | val alltokens : (char, token list) genp |
| 111 | (* char list -> token list * 'a list *) |
| 112 | |
| 113 | val lex : string -> token list |
| 114 | |
| 115 | |
| 116 | (*****************************************************************************) |
| 117 | (* cant use parser as it's a reseverd word *) |
| 118 | type 'a pparser = (token, 'a) genp |
| 119 | |
| 120 | val ident : string pparser |
| 121 | (* token list -> string * token list *) |
| 122 | val int : string pparser |
| 123 | (* token list -> string * token list *) |
| 124 | val string : string pparser |
| 125 | |
| 126 | type expr = |
| 127 | | Int of int |
| 128 | | Var of string |
| 129 | | Add of expr * expr |
| 130 | | Mul of expr * expr |
| 131 | |
| 132 | val atom : expr pparser |
| 133 | (* token list -> expr * token list *) |
| 134 | val factor : expr pparser |
| 135 | (* token list -> expr * token list *) |
| 136 | val term : expr pparser |
| 137 | (* token list -> expr * token list *) |
| 138 | val expr : expr pparser |
| 139 | (* token list -> expr * 'a list *) |
| 140 | |
| 141 | val parse : 'a pparser -> string -> 'a |
| 142 | (* (token list -> 'a * 'b) -> string -> 'a *) |
| 143 | |
| 144 | |
| 145 | (*****************************************************************************) |
| 146 | |
| 147 | module Infix : sig |
| 148 | val ( ||| ) : ('a, 'b) genp -> ('a, 'b) genp -> ('a, 'b) genp |
| 149 | (* ('a -> 'b) -> ('a -> 'b) -> 'a -> 'b *) |
| 150 | val ( +++ ) : ('a, 'b) genp -> ('a, 'c) genp -> ('a, 'b * 'c) genp |
| 151 | (* ('a -> 'b * 'c) -> ('c -> 'd * 'e) -> 'a -> ('b * 'd) * 'e *) |
| 152 | val ( >| ) : ('a, 'b) genp -> ('b -> 'c) -> ('a, 'c) genp |
| 153 | (* ('a -> 'b * 'c) -> ('b -> 'd) -> 'a -> 'd * 'c *) |
| 154 | end |