lexer.mll 7.15 KB
Newer Older
1 2
(**************************************************************************)
(*                                                                        *)
Jean-Christophe Filliâtre's avatar
headers  
Jean-Christophe Filliâtre committed
3 4 5 6 7
(*  Copyright (C) 2010-                                                   *)
(*    Francois Bobot                                                      *)
(*    Jean-Christophe Filliatre                                           *)
(*    Johannes Kanig                                                      *)
(*    Andrei Paskevich                                                    *)
8 9 10 11 12 13 14 15 16 17 18
(*                                                                        *)
(*  This software is free software; you can redistribute it and/or        *)
(*  modify it under the terms of the GNU Library General Public           *)
(*  License version 2.1, with the special exception on linking            *)
(*  described in file LICENSE.                                            *)
(*                                                                        *)
(*  This software is distributed in the hope that it will be useful,      *)
(*  but WITHOUT ANY WARRANTY; without even the implied warranty of        *)
(*  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.                  *)
(*                                                                        *)
(**************************************************************************)
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
19 20

{
21
  open Format
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
22
  open Lexing
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
23
  open Term
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
24 25 26
  open Ptree
  open Parser

27 28 29 30 31 32 33
  (* lexical errors *)

  type error = 
    | IllegalCharacter of char
    | UnterminatedComment
    | UnterminatedString

34
  exception Error of error
35 36 37 38 39 40

  let report fmt = function
    | IllegalCharacter c -> fprintf fmt "illegal character %c" c
    | UnterminatedComment -> fprintf fmt "unterminated comment"
    | UnterminatedString -> fprintf fmt "unterminated string"

Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
41 42 43 44 45 46 47 48 49 50 51
  let keywords = Hashtbl.create 97
  let () = 
    List.iter 
      (fun (x,y) -> Hashtbl.add keywords x y)
      [ "absurd", ABSURD;
	"and", AND;
	"as", AS;
	"assert", ASSERT;
	"axiom", AXIOM;
	"begin", BEGIN;
        "check", CHECK;
52
	"clone", CLONE;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
53 54 55 56 57 58
	"do", DO;
	"done", DONE;
        "else", ELSE;
	"end", END;
	"exception", EXCEPTION; 
	"exists", EXISTS;
59
	"export", EXPORT;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
60 61 62 63 64 65 66 67
	"external", EXTERNAL;
        "false", FALSE;
	"for", FOR;
	"forall", FORALL;
	"fun", FUN;
	"function", FUNCTION;
	"goal", GOAL;
	"if", IF;
68
	"import", IMPORT;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
69 70 71
	"in", IN;
	"inductive", INDUCTIVE;
	"invariant", INVARIANT;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
72
	"lemma", LEMMA;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
73 74 75
	"let", LET;
	"logic", LOGIC;
	"match", MATCH;
76
	"namespace", NAMESPACE;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
77 78 79 80 81 82 83 84 85 86 87 88 89
	"not", NOT;
	"of", OF;
	"or", OR;
	"parameter", PARAMETER;
	"predicate", PREDICATE;
	"prop", PROP;
	"raise", RAISE;
	"raises", RAISES;
	"reads", READS;
	"rec", REC;
	"ref", REF;
	"returns", RETURNS;
	"then", THEN;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
90
	"theory", THEORY;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
91 92 93 94
	"true", TRUE;
	"try", TRY;
	"type", TYPE;
	"unit", UNIT;
95
	"use", USE;
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133
	"variant", VARIANT;
	"void", VOID;
	"while", WHILE;
	"with", WITH;
        "writes", WRITES ]
	       
  let newline lexbuf =
    let pos = lexbuf.lex_curr_p in
    lexbuf.lex_curr_p <- 
      { pos with pos_lnum = pos.pos_lnum + 1; pos_bol = pos.pos_cnum }

  let string_buf = Buffer.create 1024

  let char_for_backslash = function
    | 'n' -> '\n'
    | 't' -> '\t'
    | c -> c

  let update_loc lexbuf file line chars =
    let pos = lexbuf.lex_curr_p in
    let new_file = match file with None -> pos.pos_fname | Some s -> s in
    lexbuf.lex_curr_p <- 
      { pos with
	  pos_fname = new_file;
	  pos_lnum = int_of_string line;
	  pos_bol = pos.pos_cnum - int_of_string chars;
      }

  let remove_leading_plus s =
    let n = String.length s in 
    if n > 0 && s.[0] = '+' then String.sub s 1 (n-1) else s

  let option_app f = function None -> None | Some x -> Some (f x)

}

let newline = '\n'
let space = [' ' '\t' '\r']
134 135 136
let lalpha = ['a'-'z' '_']
let ualpha = ['A'-'Z']
let alpha = lalpha | ualpha
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
137
let digit = ['0'-'9']
138 139
let lident = lalpha (alpha | digit | '\'')*
let uident = ualpha (alpha | digit | '\'')*
140 141 142 143 144 145 146 147 148 149
let decimal_literal =
  ['0'-'9'] ['0'-'9' '_']*
let hex_literal =
  '0' ['x' 'X'] ['0'-'9' 'A'-'F' 'a'-'f']['0'-'9' 'A'-'F' 'a'-'f' '_']*
let oct_literal =
  '0' ['o' 'O'] ['0'-'7'] ['0'-'7' '_']*
let bin_literal =
  '0' ['b' 'B'] ['0'-'1'] ['0'-'1' '_']*
let int_literal =
  decimal_literal | hex_literal | oct_literal | bin_literal
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
150 151 152 153 154 155 156 157 158 159
let hexadigit = ['0'-'9' 'a'-'f' 'A'-'F']

rule token = parse
  | "#" space* ("\"" ([^ '\010' '\013' '"' ]* as file) "\"")?
    space* (digit+ as line) space* (digit+ as char) space* "#"
      { update_loc lexbuf file line char; token lexbuf }
  | newline 
      { newline lexbuf; token lexbuf }
  | space+  
      { token lexbuf }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
160 161
  | '_'
      { UNDERSCORE }
162 163 164 165
  | lident as id  
      { try Hashtbl.find keywords id with Not_found -> LIDENT id }
  | uident as id  
      { UIDENT id }
166
  | int_literal as s
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
167 168 169 170 171 172 173 174 175 176
      { INTEGER s }
  | (digit+ as i) ("" as f) ['e' 'E'] (['-' '+']? digit+ as e)
  | (digit+ as i) '.' (digit* as f) (['e' 'E'] (['-' '+']? digit+ as e))?
  | (digit* as i) '.' (digit+ as f) (['e' 'E'] (['-' '+']? digit+ as e))?
      { FLOAT (RConstDecimal (i, f, option_app remove_leading_plus e)) }
  | '0' ['x' 'X'] ((hexadigit* as i) '.' (hexadigit+ as f) 
                  |(hexadigit+ as i) '.' (hexadigit* as f)
		  |(hexadigit+ as i) ("" as f))
    ['p' 'P'] (['-' '+']? digit+ as e)
      { FLOAT (RConstHexa (i, f, remove_leading_plus e)) }
177 178
  | "(*)"
      { LIDENT "*" }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202
  | "(*"
      { comment lexbuf; token lexbuf }
  | "'"
      { QUOTE }
  | ","
      { COMMA }
  | "("
      { LEFTPAR }
  | ")"
      { RIGHTPAR }
  | "!"
      { BANG }
  | ":"
      { COLON }
  | ";"
      { SEMICOLON }
  | ":="
      { COLONEQUAL }
  | "->"
      { ARROW }
  | "<->"
      { LRARROW }
  | "="
      { EQUAL }
203
  | "<>" | "<" | "<=" | ">" | ">=" as s
204 205 206
      { INFIXOP0 s }
  | "+" 
      { INFIXOP2 "+" }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
207 208
  | "-"
      { MINUS }
209 210
  | "*" | "/" | "%" as c
      { INFIXOP3 (String.make 1 c) }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
  | "@"
      { AT }
  | "."
      { DOT }
  | "["
      { LEFTSQ }
  | "]"
      { RIGHTSQ }
  | "{"
      { LEFTB }
  | "}"
      { RIGHTB }
  | "{{"
      { LEFTBLEFTB }
  | "}}"
      { RIGHTBRIGHTB }
  | "|"
      { BAR }
  | "||"
      { BARBAR }
  | "&&" 
      { AMPAMP }
  | "=>"
      { BIGARROW }
  | "\""
      { Buffer.clear string_buf; string lexbuf }
  | eof 
      { EOF }
  | _ as c
240
      { raise (Error (IllegalCharacter c)) }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
241 242 243 244 245 246 247 248 249

and comment = parse
  | "*)" 
      { () }
  | "(*" 
      { comment lexbuf; comment lexbuf }
  | newline 
      { newline lexbuf; comment lexbuf }
  | eof
250
      { raise (Error UnterminatedComment) }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
251 252 253 254 255 256 257 258 259 260 261
  | _ 
      { comment lexbuf }

and string = parse
  | "\""
      { STRING (Buffer.contents string_buf) }
  | "\\" (_ as c)
      { Buffer.add_char string_buf (char_for_backslash c); string lexbuf }
  | newline 
      { newline lexbuf; Buffer.add_char string_buf '\n'; string lexbuf }
  | eof
262
      { raise (Error UnterminatedString) }
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
  | _ as c
      { Buffer.add_char string_buf c; string lexbuf }



{

  let loc lb = (lexeme_start_p lb, lexeme_end_p lb)

  let with_location f lb =
    try f lb with e -> raise (Loc.Located (loc lb, e))

  let parse_lexpr = with_location (lexpr token)
  let parse_logic_file = with_location (logic_file token)

  let lexpr_of_string s = parse_lexpr (from_string s)
}

(*
Local Variables: 
283
compile-command: "unset LANG; make -C ../.. test"
Jean-Christophe Filliâtre's avatar
Jean-Christophe Filliâtre committed
284 285 286
End: 
*)