Files
carbon-lang/executable_semantics/syntax/lexer.lpp
T

156 lines
5.8 KiB
Plaintext

/*
Part of the Carbon Language project, under the Apache License v2.0 with LLVM
Exceptions. See /LICENSE for license information.
SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*/
%{
#include <cstdlib>
#include <iostream>
#include "executable_semantics/syntax/parse_and_lex_context.h"
%}
/* Turn off legacy bits we don't need */
%option noyywrap nounput nodefault noinput
/* maintains the number of the current line read from input in the
global variable yylineno.
*/
%option yylineno
AND "and"
ARROW "->"
AUTO "auto"
BOOL "Bool"
BREAK "break"
CASE "case"
CHOICE "choice"
ONE_LINE_COMMENT \/\/[^\n]*\n
CONTINUE "continue"
DBLARROW "=>"
DEFAULT "default"
ELSE "else"
EQUAL_EQUAL "=="
FALSE "false"
FN "fn"
FNTY "fnty"
IF "if"
INT "Int"
MATCH "match"
NOT "not"
OR "or"
RETURN "return"
STRUCT "struct"
TRUE "true"
TYPE "Type"
VAR "var"
WHILE "while"
identifier [A-Za-z_][A-Za-z0-9_]*
integer_literal [0-9]+
horizontal_whitespace [ \t\r]
%{
// This macro is expanded to run each time a token is recognized.
//
// Advances the current token position by yyleng columns without changing
// the line number.
# define YY_USER_ACTION context.current_token_position.columns(yyleng);
%}
%%
%{
// Code run each time yylex is called.
// Begin with an empty token span starting where its previous end was.
context.current_token_position.step();
%}
{AND} { return yy::parser::make_AND(context.current_token_position); }
{ARROW} { return yy::parser::make_ARROW(context.current_token_position); }
{AUTO} { return yy::parser::make_AUTO(context.current_token_position); }
{BOOL} { return yy::parser::make_BOOL(context.current_token_position); }
{BREAK} { return yy::parser::make_BREAK(context.current_token_position); }
{CASE} { return yy::parser::make_CASE(context.current_token_position); }
{CHOICE} { return yy::parser::make_CHOICE(context.current_token_position); }
{CONTINUE} { return yy::parser::make_CONTINUE(context.current_token_position); }
{DBLARROW} { return yy::parser::make_DBLARROW(context.current_token_position); }
{DEFAULT} { return yy::parser::make_DEFAULT(context.current_token_position); }
{ELSE} { return yy::parser::make_ELSE(context.current_token_position); }
"==" { return yy::parser::make_EQUAL_EQUAL(context.current_token_position); }
{FALSE} { return yy::parser::make_FALSE(context.current_token_position); }
{FN} { return yy::parser::make_FN(context.current_token_position); }
{FNTY} { return yy::parser::make_FNTY(context.current_token_position); }
{IF} { return yy::parser::make_IF(context.current_token_position); }
{INT} { return yy::parser::make_INT(context.current_token_position); }
{MATCH} { return yy::parser::make_MATCH(context.current_token_position); }
{NOT} { return yy::parser::make_NOT(context.current_token_position); }
{OR} { return yy::parser::make_OR(context.current_token_position); }
{RETURN} { return yy::parser::make_RETURN(context.current_token_position); }
{STRUCT} { return yy::parser::make_STRUCT(context.current_token_position); }
{TRUE} { return yy::parser::make_TRUE(context.current_token_position); }
{TYPE} { return yy::parser::make_TYPE(context.current_token_position); }
{VAR} { return yy::parser::make_VAR(context.current_token_position); }
{WHILE} { return yy::parser::make_WHILE(context.current_token_position); }
"=" return yy::parser::make_EQUAL(context.current_token_position);
"-" return yy::parser::make_MINUS(context.current_token_position);
"+" return yy::parser::make_PLUS(context.current_token_position);
"*" return yy::parser::make_STAR(context.current_token_position);
"/" return yy::parser::make_SLASH(context.current_token_position);
"(" return yy::parser::make_LEFT_PARENTHESIS(context.current_token_position);
")" return yy::parser::make_RIGHT_PARENTHESIS(context.current_token_position);
"{" return yy::parser::make_LEFT_CURLY_BRACE(context.current_token_position);
"}" return yy::parser::make_RIGHT_CURLY_BRACE(context.current_token_position);
"[" return yy::parser::make_LEFT_SQUARE_BRACKET(context.current_token_position);
"]" return yy::parser::make_RIGHT_SQUARE_BRACKET(context.current_token_position);
"." return yy::parser::make_PERIOD(context.current_token_position);
"," return yy::parser::make_COMMA(context.current_token_position);
";" return yy::parser::make_SEMICOLON(context.current_token_position);
":" return yy::parser::make_COLON(context.current_token_position);
{identifier} {
int n = strlen(yytext);
auto r = reinterpret_cast<char*>(malloc((n + 1) * sizeof(char)));
strncpy(r, yytext, n + 1);
return yy::parser::make_identifier(r, context.current_token_position);
}
{integer_literal} {
auto r = atof(yytext);
return yy::parser::make_integer_literal(r, context.current_token_position);
}
{ONE_LINE_COMMENT} {
// Advance end by 1 line, resetting the column to zero.
context.current_token_position.lines(1);
// Make the span empty by setting start to end.
context.current_token_position.step();
}
{horizontal_whitespace}+ {
// Make the span empty by setting start to end.
context.current_token_position.step();
}
\n+ {
// Advance end by yyleng lines, resetting the column to zero.
context.current_token_position.lines(yyleng);
// Make the span empty by setting start to end.
context.current_token_position.step();
}
. {
std::cerr << context.current_token_position << ": invalid character '"
<< yytext[0] << "' in source file." << std::endl;
std::exit(1);
}
<<EOF>> {
// A more modern Bison would give us make_EOF.
return yy::parser::make_END_OF_FILE(context.current_token_position);
}
%%