[legacy/frontend] Add lexer, parser, and AST

1 files changed, 95 insertions, 0 deletions

diff --git a/rcc-legacy/src/frontend/lexer.l b/rcc-legacy/src/frontend/lexer.l
new file mode 100644
index 0000000..33359ef
--- /dev/null
+++ b/rcc-legacy/src/frontend/lexer.l
@@ -0,0 +1,95 @@
+/* Source code for RCC-Legacy.
+   Copyright (C) 2026 Richard Qin
+
+This file is part of RCC.
+
+RCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+RCC 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.  See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with RCC; see the file LICENSE.  If not see
+<http://www.gnu.org/licenses/>.  */
+
+%{
+/* lexer.l - SysY2022 Lexer */
+#include "rcc/ast.h"
+#include "parser.tab.h"
+
+#include <cstdlib>
+#include <cstring>
+
+// Line number tracking
+int yylineno_val = 1;
+#define YY_USER_ACTION  yylloc.first_line = yylineno_val;
+
+using namespace rcc;
+%}
+
+%option noyywrap nounput noinput
+
+/* Exclusive start condition: multi-line comments */
+%x COMMENT
+
+%%
+
+ /* ─── Whitespace and Newlines ─── */
+[ \t\r]+        { /* skip whitespace */ }
+\n              { yylineno_val++; }
+
+ /* ─── Comments ─── */
+"//"[^\n]*      { /* single-line comment */ }
+"/*"            { BEGIN(COMMENT); }
+<COMMENT>"*/"   { BEGIN(INITIAL); }
+<COMMENT>\n     { yylineno_val++; }
+<COMMENT>.      { /* eat anything inside comment */ }
+
+ /* ─── Keywords ─── */
+"int"           { return T_INT; }
+"float"         { return T_FLOAT; }
+"void"          { return T_VOID; }
+"const"         { return T_CONST; }
+"if"            { return T_IF; }
+"else"          { return T_ELSE; }
+"while"         { return T_WHILE; }
+"break"         { return T_BREAK; }
+"continue"      { return T_CONTINUE; }
+"return"        { return T_RETURN; }
+
+ /* ─── Integer Constants (Hex > Octal > Decimal, order is important) ─── */
+0[xX][0-9a-fA-F]+       { yylval.int_val = static_cast<int32_t>(strtoul(yytext, nullptr, 16)); return INT_CONST; }
+0[0-7]+                  { yylval.int_val = static_cast<int32_t>(strtoul(yytext, nullptr, 8));  return INT_CONST; }
+[0-9]+                   { yylval.int_val = static_cast<int32_t>(strtoul(yytext, nullptr, 10)); return INT_CONST; }
+
+ /* ─── Floating Point Constants (Hex float + Decimal float) ─── */
+0[xX][0-9a-fA-F]*\.[0-9a-fA-F]+([pP][+-]?[0-9]+)?  { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+0[xX][0-9a-fA-F]+\.[0-9a-fA-F]*([pP][+-]?[0-9]+)?  { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+0[xX][0-9a-fA-F]+[pP][+-]?[0-9]+                     { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+[0-9]*\.[0-9]+([eE][+-]?[0-9]+)?    { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+[0-9]+\.[0-9]*([eE][+-]?[0-9]+)?    { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+[0-9]+[eE][+-]?[0-9]+               { yylval.float_val = strtof(yytext, nullptr); return FLOAT_CONST; }
+
+ /* ─── Identifiers ─── */
+[a-zA-Z_][a-zA-Z0-9_]*  { yylval.str_val = strdup(yytext); return IDENT; }
+
+ /* ─── Multi-character Operators ─── */
+"&&"    { return T_AND; }
+"||"    { return T_OR; }
+"<="    { return T_LE; }
+">="    { return T_GE; }
+"=="    { return T_EQ; }
+"!="    { return T_NE; }
+
+ /* ─── Format Strings (for putf only) ─── */
+\"[^\"]*\"  { yylval.str_val = strdup(yytext); return STRING_CONST; }
+
+ /* ─── Single-character Operators and Delimiters, return ASCII directly ─── */
+.       { return yytext[0]; }
+
+%%