civicaml/lexer.mll

{
open Lexing
open Parser

exception SyntaxError of string

let next_line lexbuf =
    let pos = lexbuf.lex_curr_p in
    lexbuf.lex_curr_p <- {
        pos with pos_bol = lexbuf.lex_curr_pos;
                 pos_lnum = pos.pos_lnum + 1
    }
}

rule token = parse
    | '#'' '+['0'-'9']+' '+'"'[^'"']+'"'(' '+['1'-'4'])*'\n' as marker {
        (*
         * The C preprocessor inserts so called ``line markers'' into the output.
         * These markers have the following form:
         *
         *  # <linenum> <filename> <flags>
         *
         * This marker indicates that all lines after this marker up to the next
         * marker come from the file <filename> starting at line <linenum>. After
         * the file name can be zero or more flags, these flags are 1, 2, 3, 4.
         * These flags can be ignored.
         *)
        let scan lnum fn =
            let filename = String.sub fn 0 (String.length fn - 1) in
            let pos = lexbuf.lex_curr_p in
            lexbuf.lex_curr_p <- { pos with
                                   pos_fname = filename;
                                   pos_lnum = lnum;
                                   pos_bol = pos.pos_cnum }
        in
        Scanf.sscanf marker "# %d \"%s" scan;
        token lexbuf
    }

    (*| ['('')''['']''{''}'';'','] as literal { literal }*)

    | '('       { LPAREN }
    | ')'       { RPAREN }
    | '['       { LBRACK }
    | ']'       { RBRACK }
    | '{'       { LBRACE }
    | '}'       { RBRACE }
    | ';'       { SEMICOL }
    | ','       { COMMA }

    | '='       { ASSIGN }
    | '!'       { NOT }
    | '+'       { ADD }
    | '-'       { SUB }
    | '*'       { MUL }
    | '/'       { DIV }
    | '%'       { MOD }

    | "<="      { LE }
    | "<"       { LT }
    | ">="      { GE }
    | ">"       { GT }
    | "=="      { EQ }
    | "!="      { NE }
    | "&&"      { AND }
    | "||"      { OR }

    | "if"      { IF }
    | "else"    { ELSE }
    | "do"      { DO }
    | "while"   { WHILE }
    | "for"     { FOR }
    | "return"  { RETURN }
    | "extern"  { EXTERN }
    | "export"  { EXPORT }
    | "int"     { INT }
    | "bool"    { BOOL }
    | "float"   { FLOAT }
    | "void"    { VOID }

    | "true"    { BOOL_CONST true }
    | "false"   { BOOL_CONST false }
    | ['0'-'9']+ as i              { INT_CONST (int_of_string i) }
    | ['0'-'9']+'.'['0'-'9']+ as f { FLOAT_CONST (float_of_string f) }
    | ['A'-'Z''a'-'z']['A'-'Z''a'-'z''0'-'9''_']* as id { ID id }

    | '\r' | '\n' | "\r\n"  { next_line lexbuf; token lexbuf }
    | [' ''\t']+            { token lexbuf }
    | "//"[^'\n']*          { token lexbuf }
    | "/*"                  { comment lexbuf }
    | eof | '\000'          { EOF }

    | _ as chr { raise (SyntaxError ("unexpected char: " ^ Char.escaped chr)) }

(* Multi-line comments *)
and comment = parse
    | '\r' | '\n' | "\r\n"  { next_line lexbuf; comment lexbuf }
    | _                     { comment lexbuf }
    | "*/"                  { token lexbuf }