章节14:实现字符串变量

50 阅读4分钟

只是打印数字还不够直观,这个时候我们需要字符串变量来展示,如 打印a的值为:1,所以需要添加字符串变量,同时字符串可以通过+连接,也可以将数字和字符串连接 , 用单引号‘hello world’来表示字符串。

回顾上一个语法图 , 现在用QUOTE表示单引号’,增加语法节点stringAst : QUOTE variableAst QUOTE 。

programAst : (assignAst | printCallAst)*
assignAst : variableAst ASSIGN arithAst
printCallAst : print LBRACKET arithAst RBRACKET
arithAst : termAst ((PLUS|MINUS) termAst)* 
termAst : factorAst ((MUL|DIV) factorAst)* 
factorAst: INTEGER | LBRACKET arithAst RBRACKET | variableAst

新的语法图

programAst : (assignAst | printCallAst)*
assignAst : variableAst ASSIGN arithAst
printCallAst : print LBRACKET arithAst RBRACKET
arithAst : termAst ((PLUS|MINUS) termAst)* 
termAst : factorAst ((MUL|DIV) factorAst)* 
factorAst: INTEGER | LBRACKET arithAst RBRACKET | variableAst | stringAst

修改词法解析器

// 词性枚举
public enum TokenType {
    INTEGER // 数字
    , PLUS // 加法运算符
    , EOF // 程序结束
    , MINUS // 减法运算符
    , MUL // 乘法运算符
    , DIV // 除法运算符
    , LBRACKET // 左括号
    , RBRACKET // 右括号
    , ID // 变量
    , ASSIGN // 赋值符号=
    , PRINT // 内置打印函数
    , QUOTE // 单引号 '
    , STRING // 字符串
}

// 词法解析器
public class Lexer {
    private String text; // 输入的程序
    private Integer position; // 记录扫描的位置
    private Character currentChar; // 记录当前扫描的字符
    private Map<String , Token> keyWordMap = new HashMap<>(); // 关键字map

    public Token getNextToken(){  // 获取词法单元
        while(this.currentChar != null){
            if(Character.isDigit(this.currentChar)){
                return this.integer();
            }else if(Character.isWhitespace(currentChar)){
                this.skipWhiteSpace();
            }else if(this.currentChar == '+'){
                Token token = new Token(TokenType.PLUS , "+");
                this.advance();
                return token; 
            }else if(this.currentChar == '-'){
                Token token = new Token(TokenType.MINUS , "-");
                this.advance();
                return token; 
            }else if(this.currentChar == '*'){
                Token token = new Token(TokenType.MUL , "*");
                this.advance();
                return token; 
            }else if(this.currentChar == '/'){
                Token token = new Token(TokenType.DIV , "/");
                this.advance();
                return token; 
            }else if(this.currentChar == '/'){
                Token token = new Token(TokenType.DIV , "/");
                this.advance();
                return token; 
            }else if(this.currentChar == '('){
                Token token = new Token(TokenType.LBRACKET , "(");
                this.advance();
                return token; 
            }else if(this.currentChar == ')'){
                Token token = new Token(TokenType.RBRACKET , ")");
                this.advance();
                return token; 
            }else if(this.currentChar == '='){
                Token token = new Token(TokenType.ASSIGN , "=");
                this.advance();
                return token; 
            }else if(this.currentChar == '\''){
                return this.string();
            }else if(Character.isAlphabetic(currentChar)){
                return variable();
            }else {
                this.error("未知的词法");
            }
        }
        return new Token(TokenType.EOF);
    }
    private Token string() {
        String value = "";
        this.advance();
        while(currentChar != null && currentChar != '\''){
            value += currentChar;
            this.advance();
        }
        this.advance();
        return new Token(TokenType.STRING , value);
    }
    private Token variable(){ // 识别变量
        String value = "";
        while(currentChar != null && Character.isAlphabetic(currentChar)){
            value += currentChar;
            this.advance();
        }
        Token token = keyWordMap.getOrDefault(value, new Token(TokenType.ID, value));
        return token;
    }
    public Token integer(){ // 识别多个数字
        String result = "";
        while(this.currentChar != null && Character.isDigit(this.currentChar)){
            result += this.currentChar;
            this.advance();
        }
        return new Token(TokenType.INTEGER ,Integer.valueOf(result));
    }

    private void skipWhiteSpace(){ // 空格跳过
        while(currentChar != null && Character.isWhitespace(currentChar)){
            this.advance();
        }
    }

    public void advance(){ // 往后走一步
        this.position += 1;
        if(this.position <= this.text.length() - 1){ // 扫描的位置有效
            this.currentChar = text.charAt(this.position);
        }else{ // 扫描完了
            this.currentChar = null;
        }
        
    }
    public void error(String msg){ // 报错函数
        throw new RuntimeException(msg);
    }
    public Lexer(String text) {// 构造器
        this.text = text;
        this.position = 0;
        this.currentChar = text.charAt(this.position);
        keyWordMap.put("print", new Token(TokenType.PRINT , "print"));
    }
}

修改语法解析器

@Data
public class StringAst extends Ast{
    private String value;

    public StringAst(Token token) {
        this.value = (String) token.getValue();
    }
    
}

// 语法解析器
public class Parser {
    private Lexer lexer ; // 词法解析器
    private Token currentToken; // 当前的词法单元
    public Parser(Lexer lexer) {
        this.lexer = lexer;
        this.currentToken = this.lexer.getNextToken();
    }
    public Ast programAst(){ // 程序节点
        // programAst : (assignAst | printCallAst)*
        ProgramAst ast = new ProgramAst();
        while (Arrays.asList(TokenType.ID,TokenType.PRINT).contains(this.currentToken.getType()) ) {
            if(this.currentToken.getType() == TokenType.ID){
                ast.add(this.assignAst());
            }else if(this.currentToken.getType() == TokenType.PRINT){
                ast.add(this.printCallAst());
            }
        }
        return ast;
    }
    
    public Ast printCallAst(){
        // printCallAst : print LBRACKET arithAst RBRACKET
        this.eat(TokenType.PRINT);
        this.eat(TokenType.LBRACKET);
        Ast ast = this.arithAst();
        this.eat(TokenType.RBRACKET);
        return new PrintCallAst(ast);
    }

    public Ast assignAst(){
        // assignAst : variableAst ASSIGN arithAst
        Token id = this.currentToken;
        this.eat(TokenType.ID);
        this.eat(TokenType.ASSIGN);
        Ast right = this.arithAst();
        return new AssignAst((String)id.getValue() , right);
    }
    public Ast arithAst(){
        // arithAst : termAst ((PLUS|MINUS) termAst)* 
        Ast node = this.termAst();
        while(Arrays.asList(TokenType.PLUS,TokenType.MINUS).contains(this.currentToken.getType())){
            Token op = this.currentToken;
            if(op.getType() == TokenType.PLUS){
                this.eat(TokenType.PLUS);
            }else if(op.getType() == TokenType.MINUS){
                this.eat(TokenType.MINUS);
            }
            node = new ArithAst(node ,op.getType(),this.termAst());
        }
        return node;
    }
    public Ast termAst(){
        // termAst : factorAst ((MUL|DIV) factorAst)*
        Ast node = this.factorAst();
        while(Arrays.asList(TokenType.MUL,TokenType.DIV).contains(this.currentToken.getType())){
            Token op = this.currentToken;
            if(op.getType() == TokenType.MUL){
                this.eat(TokenType.MUL);
            }else if(op.getType() == TokenType.DIV){
                this.eat(TokenType.DIV);
            }
            node = new TermAst(node ,op.getType(),this.factorAst());
        }
        return node;
    }
    public Ast factorAst(){
         // factorAst: INTEGER | LBRACKET assignAst RBRACKET | variableAst
        Token left = this.currentToken;
        if(left.getType() == TokenType.INTEGER){
            this.eat(TokenType.INTEGER);
            return new FactorAst((Integer)left.getValue());
        }else if(left.getType() == TokenType.LBRACKET){
            this.eat(TokenType.LBRACKET);
            Ast ast = this.arithAst();
            this.eat(TokenType.RBRACKET);
            return ast;
        }else if(left.getType() == TokenType.ID){
            this.eat(TokenType.ID);
            return new VariableAst(left);
        }else if(left.getType() == TokenType.STRING){
            this.eat(TokenType.STRING);
            return new StringAst(left);
        }
        this.error("语法错误");
        return null;
    }
    public void eat(TokenType tokenType){ // 确认当前的词性是否正确
        if(tokenType == this.currentToken.getType()){
            this.currentToken = this.lexer.getNextToken();
        }else{
            this.error("语法错误");
        }
    }
    public void error(String msg){ // 报错函数
        throw new RuntimeException(msg);
    }
    public Ast parse(){ // 获取语法树
        return this.programAst();
    }
}

修改解释器

// 目标执行器
public class Interpreter {
    private Parser parser; // 语法解析器
    private Map<String , Object> symbolMap = new HashMap<>(); // 符号表
    public Interpreter(Parser parser) {
        this.parser = parser;
    }
    public void visitProgramAst(Ast ast){ // 访问programAst节点
        ProgramAst programAst = (ProgramAst) ast; 
        List<Ast> statementList = programAst.getStatementList();
        for(Ast statment : statementList){
            this.visit(statment);
        }
    }
    public void visitPrintCallAst(Ast ast){
        PrintCallAst printCallAst = (PrintCallAst) ast;
        System.out.println(this.visit(printCallAst.getArithAst()));
    }
    public void visitAssignAst(Ast ast){
        AssignAst assignAst = (AssignAst) ast;
        String name = assignAst.getLeftValue();
        Object value =  this.visit(assignAst.getRightValue());
        symbolMap.put(name,value);
    }
    public Object visitArithAst(Ast ast){
        ArithAst arithAst = (ArithAst) ast;
        if(arithAst.getOp() == TokenType.PLUS){
            Object leftObj = this.visit(arithAst.getLeftValue());
            Object rightObj = this.visit(arithAst.getRightValue());
            if(leftObj instanceof String || rightObj instanceof String){
                String left = leftObj instanceof String ? (String) leftObj : String.valueOf((Integer)leftObj);
                String right = rightObj instanceof String ? (String) rightObj : String.valueOf((Integer)rightObj);
                return left + right;
            }else {
                return (Integer) leftObj + (Integer)rightObj;
            }
        }else if(arithAst.getOp() == TokenType.MINUS){
            return (Integer)this.visit(arithAst.getLeftValue()) - (Integer)this.visit(arithAst.getRightValue()); // 减法计算
        }
        return null;
    }   
    public Object visitTermAst(Ast ast){
        TermAst termAst = (TermAst) ast; 
        if(termAst.getOp() == TokenType.MUL){
            return (Integer)this.visit(termAst.getLeftValue()) * (Integer)this.visit(termAst.getRightValue()); // 乘法计算
        }else if(termAst.getOp() == TokenType.DIV){
            return (Integer)this.visit(termAst.getLeftValue()) / (Integer)this.visit(termAst.getRightValue()); // 除法计算
        }
        return null;
    }
    public Object visitStringAst(Ast ast){
        StringAst stringAst = (StringAst) ast;
        return stringAst.getValue();
    }
    public Object visitFactorAst(Ast ast){
        FactorAst factorAst = (FactorAst) ast;
        return factorAst.getValue();
    }
    public Object visitVariableAst(Ast ast){
        VariableAst variableAst = (VariableAst) ast;
        return symbolMap.get(variableAst.getValue()); // 从符号表获取对应的值
    }
    public Object visit(Ast ast){ // 使用反射通过类名调用对应的函数
        String methodName = "visit" + ast.getClass().getSimpleName();
        try {
            Method method = this.getClass().getDeclaredMethod(methodName , Ast.class );
            return method.invoke(this, ast);
        } catch (Exception e) {
            e.printStackTrace();
        } 
        return null;
    }
    public void expr() {
        Ast ast = parser.parse(); // 获取语法树
        this.visit(ast); // 遍历获取结果
        System.out.println(symbolMap);
    }
}

执行测试

LJS.txt
a = 1
b = 1 + 2
c = a + 1
d = a * b + c * 2
f = a * ( b + c ) * 2
e = 'hello world'
print('a的值为:' + a)
print('b的值为:' + b)
print(e + '_' + a)

Path resourcePath = Paths.get("study-python","demo","src", "main", "resources").toAbsolutePath();
Lexer lexer = new Lexer(FileUtil.readUtf8String(resourcePath + "\\LJS.txt"));
Parser parser = new Parser(lexer);
Interpreter interpreter = new Interpreter(parser);
interpreter.expr();

控制台打印

a的值为:1
b的值为:3
hello world_1
{a=1, b=3, c=2, d=7, e=hello world, f=10}