124 lines
3.8 KiB
C++
124 lines
3.8 KiB
C++
#include"Tokenizer.h"
|
|
#include<cctype>
|
|
#include<stdexcept>
|
|
|
|
using namespace std;
|
|
|
|
vector<Token> Tokenizer::run(const string& expr) {
|
|
|
|
expression = expr;
|
|
pos = 0;
|
|
vector<Token> tokens;
|
|
|
|
// 主循环：逐字符读取表达式
|
|
while (pos < expression.length()) {
|
|
|
|
skipWhitespace();// 跳过空白字符
|
|
if (pos >= expression.length()) break;
|
|
|
|
char current = expression[pos];
|
|
|
|
// 如果是数字或一元负号开头，读取完整数字
|
|
if (is_dig(current) || (current == '-' && is_UMC(tokens)))
|
|
tokens.push_back(readNumber());
|
|
|
|
// 如果是字母，读取函数或变量名
|
|
else if (is_let(current))
|
|
tokens.push_back(readIdentifier());
|
|
|
|
// 否则读取操作符或括号
|
|
else
|
|
tokens.push_back(readOperatorOrParen());
|
|
}
|
|
|
|
return tokens;
|
|
}
|
|
|
|
void Tokenizer::skipWhitespace() {
|
|
// 跳过所有空格、制表符等空白字符
|
|
while (pos < expression.length() && isspace(expression[pos]))
|
|
++pos;
|
|
}
|
|
|
|
Token Tokenizer::readNumber() {
|
|
size_t start = pos;
|
|
bool hasDecimal = false;// 标记小数点是否出现过
|
|
|
|
if (expression[pos] == '-')
|
|
++pos; // 若是一元负号，先跳过
|
|
|
|
while (pos < expression.length() && (is_dig(expression[pos]) || expression[pos] == '.')) {
|
|
if (expression[pos] == '.') {
|
|
if (hasDecimal)
|
|
throw runtime_error("数字格式无效：多个小数点");// 小数点重复错误
|
|
hasDecimal = true;
|
|
}
|
|
++pos;
|
|
}
|
|
|
|
string numberStr = expression.substr(start, pos - start);
|
|
return Token(TokenType::Number, numberStr);// 构造数字 Token
|
|
}
|
|
|
|
Token Tokenizer::readIdentifier() {
|
|
size_t start = pos;
|
|
|
|
// 读取连续的字母字符（函数名或变量名）
|
|
while (pos < expression.length() && is_let(expression[pos]))
|
|
++pos;
|
|
|
|
string name = expression.substr(start, pos - start);
|
|
|
|
//识别函数与变量（若后面紧跟‘（’则为函数，否则为变量）
|
|
skipWhitespace();
|
|
if (pos < expression.length() && expression[pos] == '(')
|
|
return Token(TokenType::Function, name);
|
|
else
|
|
return Token(TokenType::Variable, name);
|
|
}
|
|
|
|
Token Tokenizer::readOperatorOrParen() {
|
|
if (expression[pos] == '/' && pos + 1 < expression.length() && expression[pos + 1] == '/') {
|
|
pos += 2; // 跳过两个斜杠
|
|
return Token(TokenType::Operator, "//", 2, Associativity::Left);
|
|
}
|
|
|
|
char c = expression[pos++]; // 读取当前字符并自增指针
|
|
|
|
// 判断字符类型并创建相应 Token
|
|
switch (c) {
|
|
case '+': return Token(TokenType::Operator, "+", 1, Associativity::Left);
|
|
case '-': return Token(TokenType::Operator, "-", 1, Associativity::Left);
|
|
case '*': return Token(TokenType::Operator, "*", 2, Associativity::Left);
|
|
case '/': return Token(TokenType::Operator, "/", 2, Associativity::Left);
|
|
case '%': return Token(TokenType::Operator, "%", 2, Associativity::Left);
|
|
case '^': return Token(TokenType::Operator, "^", 3, Associativity::Right);
|
|
case '!': return Token(TokenType::Operator, "!", 4, Associativity::Right);
|
|
case '(': return Token(TokenType::LeftParen, "(");
|
|
case ')': return Token(TokenType::RightParen, ")");
|
|
default:
|
|
throw runtime_error(string("未知符号 ") + c); // 未知字符错误
|
|
}
|
|
}
|
|
|
|
bool Tokenizer::is_UMC(const vector<Token>& tokens) {
|
|
// 如果是表达式开头，或前一个是运算符或左括号，则当前负号为一元
|
|
if (tokens.empty()) return true;
|
|
Token last = tokens.back();
|
|
return last.is_ope() || last.is_LP();
|
|
}
|
|
|
|
bool Tokenizer::is_dig(char c) const {
|
|
// 判断是否为数字字符
|
|
return std::isdigit(static_cast<unsigned char>(c));
|
|
}
|
|
|
|
bool Tokenizer::is_let(char c) const {
|
|
// 判断是否为英文字母
|
|
return std::isalpha(static_cast<unsigned char>(c));
|
|
}
|
|
|
|
bool Tokenizer::is_OpeCh(char c) const {
|
|
// 判断是否为合法的运算符字符
|
|
return c == '+' || c == '-' || c == '*' || c == '/' || c == '^' || c == '%';
|
|
} |