C++实现sysy语言的词法分析

 编译原理的小组作业捏，随手分享一下，防止以后自己再看代码和地铁老人一样(doge)

tips：空格是不影响词法分析判断捏，部分通过运算符分隔词

判断常量的DFA图贴在这里了：（判断标识符的DFA就不贴了（比较简单捏））

 结果图：

 

 

 

 

 

#include<iostream>
#include<fstream>
using namespace std;

#define MOD 9999971

//储存关键词的哈希值
long long keyword[] = { 17235,349057, 4980080,4621252,138277,183,1110314,4183007,417166,105582,3836897,335129,6707534,137956,15579,3820577,8388370,4471584,4597517,7809209,439,2873918,5928846,6174355,7517928,440078};

int last_state;//用来记录上次分析的词后面可不可以跟一个有符号数（0表示可以跟，1表示不可以跟）//懒得改了(doge)

//用于识别标识符，关键词已在里面通过哈希判断
bool ckIdentifier(string s) {
	int state = 1; int index = 0; int len = s.length(); long long tmp = 0; long long xs = 1;
	while (index < len && (state == 1 || state == 2 || state == 3)) {
		if (state == 1) {
			if (isalpha(s[index]) || s[index] == '_') {
				tmp += (s[index] - 'a' + 1) * xs; xs *= 29; tmp %= MOD;
				state = 2;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 2) {
			if (isalpha(s[index]) || isdigit(s[index]) || s[index] == '_') {
				tmp += (s[index] - 'a' + 1) * xs; xs *= 29; tmp %= MOD;
				index++;
			}
			else {
				return false;
			}
		}
	}
	for (int i = 0; i < 25; i++) {
		if (tmp == keyword[i]) {
			cout << s << "   " << "keyword" << endl;
			last_state = 1;
			return true;
		}
	}
	if (state == 2) {
		cout << s << "   " << "identifier" << endl;
		last_state = 1;
		return true;
	}
}

//用于识别运算符，直接判就好啦(doge)
bool ckoperator(string s) {
	if (s == "+" || s == "-" || s == "*" || s == "/" || s == "=" || s == "%" || s == "==" || s == "!=" || s == "<" || s == ">" || s == "<=" || s == ">=" || s == "!" || s == "&&" || s == "||") {
		cout << s << "   " << "operator" << endl;
		return true;
	}
	return false;
}

//用于识别边界符，直接判
bool ckborder(string s) {
	if (s == "(" || s == ")" || s == "{" || s == "}" || s == ";" || s == ",") {
		cout << s << "   " << "border" << endl;
		last_state = 1;
		return true;
	}
	return false;
}

//用于识别常量值，（包含了有符号整数，无符号整数，有符号浮点数，无符号浮点数，无符号八进制数，无符号十六进制数）
bool ckconst(string s) {
	int state = 0; int index = 0; int len = s.length();
	while (index < len && (state == 0 || state == 1 || state == 2 || state == 3 || state == 4 || state == 5 || state == 6 || state == 7 || state == 8 || state == 9 || state == 10 || state == 11)) {
		if (state == 0) {
			if (s[index] == '+' || s[index] == '-') {
				state = 1;
				index++;
			}
			else if (s[index] != '0' && isdigit(s[index])) {
				state = 3;
				index++;
			}
			else if (s[index] == '0') {
				state = 8;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 1) {
			if (isdigit(s[index])) {
				state = 2;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 2) {
			if (isdigit(s[index])) {
				index++;
			}
			else if (s[index] == '.') {
				state = 4;
				index++;
			}
		}
		else if (state == 3) {
			if (isdigit(s[index])) {
				index++;
			}
			else if (s[index] == '.') {
				state = 5;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 4) {
			if (isdigit(s[index])) {
				state = 7;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 5) {
			if (isdigit(s[index])) {
				state = 6;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 6) {
			if (isdigit(s[index])) {
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 7) {
			if (isdigit(s[index])) {
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 8) {
			if (s[index] == '.') {
				state = 5;
				index++;
			}
			else if (isdigit(s[index])) {
				state = 9;
				index++;
			}
			else if (s[index] == 'x' || s[index] == 'X') {
				state = 10;
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 9) {
			if (isdigit(s[index])) {
				index++;
			}
			else {
				return false;
			}
		}
		else if (state == 10) {
		    if (isdigit(s[index]) || s[index] >= 'a' && s[index] <= 'f' || s[index] >= 'A' && s[index] <= 'F') {
				state = 11;
				index++;
			}
			else {
				return false;
			}
        }
		else if (state == 11) {
			if (isdigit(s[index]) || s[index] >= 'a' && s[index] <= 'f' || s[index] >= 'A' && s[index] <= 'F') {
				index++;
			}
			else {
				return false;
			}
        }
	}
	if (state == 2) {
		last_state = 1;
		cout << s << "   signedinteger"  << endl;
		return true;
	}
	else if (state == 3 || state == 8) {
		last_state = 1;
		cout << s << "   unsignedinteger"  << endl;
		return true;
	}
	else if (state == 7) {
		last_state = 1;
		cout << s << "   signedfloat"  << endl;
		return true;
	}
	else if (state == 6) {
		last_state = 1;
		cout << s << "   unsignedfloat"  << endl;
		return true;
	}
	else if (state == 9) {
		last_state = 1;
		cout << s << "   octalinteger" << endl;
	}
	else if (state == 11) {
		last_state = 1;
		cout << s << "   hexinteger" << endl;
	}
	else {
		return false;
	}
}

int main()
{
	string s;
	char c[40];
	ifstream inFile("in.txt", ios::in | ios::binary); //文件输入
	while (inFile.getline(c,40)) {
		s = "";
		for (int i = 0; i < 40; i++) {
			if (c[i] != '\0' && c[i] != '\r') {
				s += c[i];
			}
			else {
				break;
			}
		} //把字符数组转成我熟悉的字符串
		string tmp = ""; int len = s.length();
		for (int i = 0; i < len; i++) {
			if (s[i] == ' ') { //遇到一个空格就要判断一下之前所存的tmp是否符号词法规定捏
				if (ckIdentifier(tmp) || ckoperator(tmp) || ckborder(tmp) || ckconst(tmp)) {

				} 
				else {
					cout << "wrong" << endl;
				}
				tmp = "";
				continue;
			}
			else if (s[i] == '_' || isalpha(s[i]) || isdigit(s[i]) || s[i] == '.') //这种特殊符号就直接加上好了
				tmp += s[i];
			else if (s[i] == ',' || s[i] == ';' || s[i] == '{' || s[i] == '}' || s[i] == '(' || s[i] == ')' || s[i] == '+' || s[i] == '-' || s[i] == '*' || s[i] == '/' || s[i] == '%' || s[i] == '=' || s[i] == '!' || s[i] == '<' || s[i] == '>' || s[i] == '&' || s[i] == '|') { 
				last_state = 0;
				if (ckIdentifier(tmp) || ckoperator(tmp) || ckborder(tmp) || ckconst(tmp)) {

				} //碰到以上符号也要判断一下tmp是否符合词法规定捏
				else {
					cout << "wrong" << "   " << endl;
				}
				tmp = "";
				tmp += s[i];
				if (i < len && ((s[i] == '=' && s[i + 1] == '=') || (s[i] == '!' && s[i + 1] == '=') || (s[i] == '<' && s[i + 1] == '=') || (s[i] == '>' && s[i + 1] == '=') || (s[i] == '&' && s[i + 1] == '&') || (s[i] == '|' && s[i + 1] == '|'))) {
					tmp += s[i + 1]; i++;
				} //出现那种长度为2的运算符
				else if (last_state == 0 && (i == 0 || s[i - 1] != ')') && (i < len && ((s[i] == '+' && isdigit(s[i + 1])) || (s[i] == '-' && isdigit(s[i + 1]))))) {
					tmp += s[i + 1]; i++;
					continue;
				} //这个是用来加上有符号捏，防止出现a+b却把+b识别为正数的情况
				if (ckIdentifier(tmp) || ckoperator(tmp) || ckborder(tmp) || ckconst(tmp)) {

				}
				else {
					cout << "wrong" << "   " << endl;
				}
				tmp = "";
			}
			else {
				tmp += s[i];
			} //出现一些乱七八糟的字母也得加入tmp，eg:@ab需要被识别为错误
		}
		if (ckIdentifier(tmp) || ckoperator(tmp) || ckborder(tmp) || ckconst(tmp)) {

		} //循环结束，判断最后一下tmp
		else {
			cout << "wrong" << "   " << endl;
		}
	}
	return 0;
}