Merge pull request '完成g4文件' (#1) from ptabmhn4l/nudt-compiler-cpp:develop into develop

完成了g4，并加入了脚本

scripts/run_test.sh

@@ -0,0 +1,62 @@
+#!/bin/bash
+
+# 项目根目录
+PROJECT_ROOT="/home/wu/nudt-compiler-cpp"
+
+# 输出目录
+TEST_RESULT_DIR="$PROJECT_ROOT/test/test_result"
+
+# 编译g4文件生成C++代码
+echo "=== 编译ANTLR grammar文件 ==="
+java -jar "$PROJECT_ROOT/third_party/antlr-4.13.2-complete.jar" \
+    -Dlanguage=Cpp \
+    -visitor \
+    -no-listener \
+    -Xexact-output-dir \
+    -o "$PROJECT_ROOT/build/generated/antlr4" \
+    "$PROJECT_ROOT/src/antlr4/SysY.g4"
+
+if [ $? -ne 0 ]; then
+    echo "错误：ANTLR编译失败"
+    exit 1
+fi
+
+# 构建项目
+echo "=== 构建项目 ==="
+cmake -S "$PROJECT_ROOT" -B "$PROJECT_ROOT/build" \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DCOMPILER_PARSE_ONLY=ON
+
+if [ $? -ne 0 ]; then
+    echo "错误：CMake配置失败"
+    exit 1
+fi
+
+cmake --build "$PROJECT_ROOT/build" -j "$(nproc)"
+
+if [ $? -ne 0 ]; then
+    echo "错误：项目构建失败"
+    exit 1
+fi
+
+# 遍历测试用例
+echo "=== 运行测试用例 ==="
+find "$PROJECT_ROOT/test/test_case" -name "*.sy" | while read -r test_file; do
+    # 计算相对路径
+    relative_path=$(realpath --relative-to="$PROJECT_ROOT/test/test_case" "$test_file")
+    # 计算输出文件路径
+    output_file="$TEST_RESULT_DIR/$relative_path.tree"
+    # 创建输出目录
+    mkdir -p "$(dirname "$output_file")"
+    
+    # 运行编译器
+    echo "处理：$relative_path"
+    "$PROJECT_ROOT/build/bin/compiler" --emit-parse-tree "$test_file" > "$output_file"
+    
+    if [ $? -ne 0 ]; then
+        echo "警告：处理 $relative_path 时出错"
+    fi
+done
+
+echo "=== 测试完成 ==="
+echo "结果保存在：$TEST_RESULT_DIR"

src/antlr4/SysY.g4

@@ -1,68 +1,65 @@
-// SysY 子集语法：支持形如
-// int main() { int a = 1; int b = 2; return a + b; }
-// 的最小返回表达式编译。
-
-// 后续需要自行添加
-
 grammar SysY;
 
-/*===-------------------------------------------===*/
-/* Lexer rules                                     */
-/*===-------------------------------------------===*/
-
-INT: 'int';
-RETURN: 'return';
-
-ASSIGN: '=';
-ADD: '+';
+compUnit
+    : (funcDef | decl)+ EOF
+    ;
 
-LPAREN: '(';
-RPAREN: ')';
-LBRACE: '{';
-RBRACE: '}';
-SEMICOLON: ';';
+decl
+    : constDecl
+    | varDecl
+    ;
 
-ID: [a-zA-Z_][a-zA-Z_0-9]*;
-ILITERAL: [0-9]+;
+constDecl
+    : Const bType constDef (Comma constDef)* Semi
+    ;
 
-WS: [ \t\r\n] -> skip;
-LINECOMMENT: '//' ~[\r\n]* -> skip;
-BLOCKCOMMENT: '/*' .*? '*/' -> skip;
+bType
+    : Int
+    | Float
+    ;
 
-/*===-------------------------------------------===*/
-/* Syntax rules                                    */
-/*===-------------------------------------------===*/
+constDef
+    : Ident (L_BRAKT constExp R_BRAKT)* Assign constInitVal
+    ;
 
-compUnit
-    : funcDef EOF
+constInitVal
+    : constExp
+    | L_BRACE (constInitVal (Comma constInitVal)*)? R_BRACE
     ;
 
-decl
-    : btype varDef SEMICOLON
+varDecl
+    : bType varDef (Comma varDef)* Semi
     ;
 
-btype
-    : INT
+funcDef
+    : funcType Ident L_PAREN (funcFParams)? R_PAREN block
     ;
 
 varDef
-    : lValue (ASSIGN initValue)?
+    : Ident (L_BRAKT constExp R_BRAKT)* (Assign initVal)?
     ;
 
-initValue
+initVal
     : exp
+    | L_BRACE (initVal (Comma initVal)*)? R_BRACE
     ;
 
-funcDef
-    : funcType ID LPAREN RPAREN blockStmt
+funcType
+    : Void
+    | Int
+    | Float
     ;
 
-funcType
-    : INT
+funcFParams
+    : funcFParam (Comma funcFParam)*
+    ;
+
+funcFParam
+    : bType Ident (L_BRAKT R_BRAKT (L_BRAKT exp R_BRAKT)*)?
     ;
 
-blockStmt
-    : LBRACE blockItem* RBRACE
+block
+    : L_BRACE (blockItem)* R_BRACE
     ;
 
 blockItem
@@ -71,28 +68,182 @@ blockItem
     ;
 
 stmt
-    : returnStmt
+    : lVar Assign exp Semi
+    | (exp)? Semi
+    | block
+    | If L_PAREN cond R_PAREN stmt (Else stmt)?
+    | While L_PAREN cond R_PAREN stmt
+    | Break Semi
+    | Continue Semi
+    | Return (exp)? Semi
     ;
 
-returnStmt
-    : RETURN exp SEMICOLON
+exp
+    : addExp
     ;
 
-exp
-    : LPAREN exp RPAREN          # parenExp
-    | var                        # varExp
-    | number                     # numberExp
-    | exp ADD exp                # additiveExp
+cond
+    : lOrExp
     ;
 
-var
-    : ID
+lVar
+    : Ident (L_BRAKT exp R_BRAKT)*
     ;
 
-lValue
-    : ID
+primaryExp
+    : L_PAREN exp R_PAREN
+    | lVar
+    | number
     ;
 
 number
-    : ILITERAL
+    : IntConst
+    | FloatConst
+    ;
+
+unaryExp
+    : primaryExp
+    | Ident L_PAREN (funcRParams)? R_PAREN
+    | unaryOp unaryExp
+    ;
+
+unaryOp : AddOp | '!';
+
+funcRParams
+    : exp (Comma exp)*
+    ;
+
+mulExp
+    : unaryExp (MulOp unaryExp)*
+    ;
+
+addExp
+    : mulExp (AddOp mulExp)*
+    ;
+
+relExp
+    : addExp (RelOp addExp)*
+    ;
+
+eqExp
+    : relExp (EqOp relExp)*
+    ;
+
+lAndExp
+    : eqExp ('&&' eqExp)*
+    ;
+
+lOrExp
+    : lAndExp ('||' lAndExp)*
+    ;
+
+constExp
+    : addExp
+    ;
+
+Void    : 'void';
+Int     : 'int';
+Float   : 'float';
+Const   : 'const';
+If      : 'if';
+Else    : 'else';
+While   : 'while';
+Break   : 'break';
+Continue: 'continue';
+Return  : 'return';
+
+Comma   : ',';
+Semi    : ';';
+L_PAREN : '(';
+R_PAREN : ')';
+L_BRAKT : '[';
+R_BRAKT : ']';
+L_BRACE : '{';
+R_BRACE : '}';
+Assign  : '=';
+
+MulOp   : '*' | '/' | '%';
+AddOp   : '+' | '-';
+RelOp   : '<' | '>' | '<=' | '>=';
+EqOp    : '==' | '!=';
+
+Ident
+    : [a-zA-Z_][a-zA-Z_0-9]*
+    ;
+
+IntConst
+    : DecInt
+    | OctInt
+    | HexInt
+    ;
+
+fragment DecInt
+    : NonZeroD Digit*
+    ;
+
+fragment OctInt
+    : '0' OctDigit*
+    ;
+
+fragment HexInt
+    : HexPre HexDigit+
+    ;
+
+FloatConst
+    : DecFloat
+	| HexFloat
+    ;
+
+fragment DecFloat
+	: Fraction Exponent?
+	| DiSeq Exponent
+	;
+
+fragment HexFloat
+	: HexPre (HexFraction | HexDiSeq) BiExponent
+	;
+
+fragment Fraction
+	: DiSeq? '.' DiSeq
+	| DiSeq '.'
+	;
+
+fragment Exponent
+	: [eE] ('+' | '-')? DiSeq
+	;
+
+fragment DiSeq
+	: Digit+
+	;
+
+fragment HexFraction
+	: HexDiSeq? '.' HexDiSeq
+	| HexDiSeq '.'
+	;
+
+fragment BiExponent
+	: [pP] ('+' | '-')? DiSeq
+	;
+
+fragment HexDiSeq
+	: HexDigit+
+	;
+
+fragment NonZeroD: [1-9];
+fragment Digit   : [0-9];
+fragment OctDigit: [0-7];
+fragment HexPre  : '0' [xX];
+fragment HexDigit: [0-9A-Fa-f];
+
+
+WS
+    : [ \t\r\n]+ -> skip
+    ;
+
+COMMENT
+    : '//' ~[\r\n]* -> skip
+    ;
+
+BLOCK_COMMENT
+    : '/*' .*? '*/' -> skip
     ;