编译原理：C++词法分析器

编译原理之C++词法分析器
1.明确实验目的：
对于每一个以文件形式给出的C或C++项目，对其进行词法分析
如图所示
2.分析实验需求：
第一：需要对文件进行读取，则需要用到文件读取流
第二：既然是词法分析，那就涉及到需要对每一个字符进行分析和归类
第三：既然是对每个字符进行分析，那代码的架构就是基于每一个字符，所以函数的编写是基于读取到的每一个字符。
3.进行实验：
第一：首先我需要规出词法的5个大类

1.关键字
const char*KeyWord[] = { "void","char","int","float","bool","w_char","include","enum",
	                        "iostream","scanf","main","printf","struct","union",
	                        "class","typedef","std","long","short","signed","unsigned",
	                         "const","volatile", "do","for","while", 
	                         "break","continue","return","goto",  
	                         "auto","register","static","extern","inline",
							"if","else","switch","case","default",
							"new","delete","sizeof",
							"private","protected","public",
							"this","friend","true","false","template","typename",
							"using","namespace","throw","try","catch","operator"};

2.标识符
3.常数
4.运算符
char Operator[8] = { '+','-','*','/','>','<','=','!' };

5.分隔符
char Separater[8] = { ';',',','{','}','[',']','(',')' };

第二：我需要写出实验中所要用到的函数名和它的功能

bool isKey()                   //判断是否为关键字的嵌套函数
char KeyOrNotKeyJudgeProcess() //判断是否为关键字的主函数
char DigitJudgeProcess()	   //判断是否为常数
char OtherJudgeProcess()       //其他判断：是终结符？注释？头文件？等其他

为什么没有判断为标识符的函数？
其实我偷了个懒，对于任何一个完整的单词来说，它只可能为关键字或者标识符，所以我把标识符的判断放在了KeyOrNotKeyJudgeProcess()函数中。详情见函数详细代码。

第三：根据代码思路进行流程式分析：
1.main函数

int main() {
	if ((fp = fopen("C:\\Users\\XXX\\Desktop\\data.txt", "r")) == NULL)//文件流函数读取文件
	{
		cout << "打开文件失败！！" << endl;
		return 0;
	}
	buffer = fgetc(fp);
	while (buffer != EOF) {
		if (buffer == ' ' || buffer == '\n' || buffer == '\t')//直接跳过对空格、换行符、结束符的读取
		{
			buffer = fgetc(fp);
		}
		else if (isalpha(buffer))//库函数：如果读取到的字符是字母，则进入函数
		{
			buffer = KeyOrNotKeyJudgeProcess(buffer);
		}
		else if (isdigit(buffer))
		{
			buffer = DigitJudgeProcess(buffer);
		}
		else
		{
			buffer = OtherJudgeProcess(buffer);
		}
	}

	return 0;
}

需要注意的是，我对空格、换行符、结束符进行了冷处理，如果读到这三类，则不进行任何判断处理，而是直接跳过它，接着读它下一个字符。
2.isKey()和KeyOrNotKeyJudgeProcess()

bool isKey(char*key) //判断是否为关键字
{
	for (int i = 0; i < sizeof(KeyWord) / sizeof(KeyWord[0]); i++) {
		if (strcmp(key, KeyWord[i]) == 0) {
			return true;
		}
	}
	return false;
}


char KeyOrNotKeyJudgeProcess(char buffer)
{
	int i = -1;
	char  alphatp[20];
	while ((isalpha(buffer)) || (isdigit(buffer)) || buffer == '_') {
		//当读到的第一个字符是字母或数字或下划线时，继续读取一个完整的单词，对它调用函数进行判断
		alphatp[++i] = buffer;
		buffer = fgetc(fp);
	}
	alphatp[i + 1] = '\0';
	if (isKey(alphatp))
	{
		cout << alphatp << "    ------>1.关键字" << endl;
		id = 1;
	}
	else
	{
		cout << alphatp << "    ------>2.标识符" << endl;
		id = 2;
	}
	return (buffer);
}

需要注意的是KeyOrNotKeyJudgeProcess，其中对读取一个完整单词的操作。

3.DigitJudgeProcess函数（此函数并无特别要注意的地方）

char DigitJudgeProcess(char buffer)
{
	int i = -1;
	char digittp[20];
	while ((isdigit(buffer)) || buffer == '.' || buffer == 'e')//1.小数点  2.科学计数法
	{
		digittp[++i] = buffer;
		buffer = fgetc(fp);
	}
	digittp[i + 1] = '\0';
	cout << digittp << "    ------>3.常数" << endl;
	id = 3;
	return (buffer);
}

4.OtherJudgeProcess（代码冗余较多）

char OtherJudgeProcess(char buffer)
{
	char ch[20];
	ch[0] = buffer;
	ch[1] = '\0';
	if (ch[0] == ',' || ch[0] == ';' || ch[0] == '{' || ch[0] == '}' || ch[0] == '(' || ch[0] == ')') {
		/*界符*/
		cout << ch << "    ------>5.分隔符" << endl;
		buffer = fgetc(fp);
		id = 5;
		return buffer;
	}
	if (ch[0] == '*' || ch[0] == '/' || ch[0] == '%') {
		// 单⽬运算符, %可能是printf的格式说明符
		if (ch[0] == '%') {
			buffer = fgetc(fp);
			if (buffer == 's' || buffer == 'd' || buffer == 'c' || buffer == 'f') {
				ch[1] = buffer;
				ch[2] = '\0';
				cout << ch << "    ------>格式说明" << endl;
				id = 4;
				return buffer;
			}
		}
		// *的判断，是否是*=
		if (ch[0] == '*') {
			buffer = fgetc(fp);
			if (buffer == '=') {
				ch[1] = buffer;
				ch[2] = '\0';
				cout << ch << "    ------>4.运算符" << endl;
				buffer = fgetc(fp);
				id = 4;
				return buffer;
			}
		}
		// /可能是除号可能是注释--//或者/*
		if (ch[0] == '/') {
			buffer = fgetc(fp);
			if (buffer == '=') {
				ch[1] = buffer;
				ch[2] = '\0';
				cout << ch << "    ------>4.运算符" << endl;
				buffer = fgetc(fp);
				id = 4;
				return buffer;
			}
			if (buffer == '*' || buffer == '/') {
				int i = 1;
				ch[1] = buffer;
				if (ch[1] == '*') {
					while (ch[i] != '/') {
						buffer = fgetc(fp);
						ch[++i] = buffer;
					}
					ch[i + 1] = '\0';
					cout << ch << "    ------>注释内容" << endl;
					buffer = fgetc(fp);
					id = 4;
					return buffer;
				}
				else {
					while (ch[i] != '\n') {
						buffer = fgetc(fp);
						ch[++i] = buffer;
					}
					ch[i + 1] = '\0';
					cout << ch << "    ------>注释内容" << endl;
					buffer = fgetc(fp);
					id = 4;
					return buffer;
				}
			}
		}
		cout << ch << "    ------>4.运算符" << endl;
		buffer = fgetc(fp);
		id = 4;
		return buffer;
	}
	if (ch[0] == '!' || ch[0] == '=' || ch[0] == '>' || ch[0] == '<') {
		buffer = fgetc(fp);
		if (buffer == '=') {
			ch[1] = buffer;
			ch[2] = '\0';
			cout << ch << "    ------>4.运算符" << endl;
			buffer = fgetc(fp);
			id = 4;
			return buffer;
		}
		else {
			cout << ch << "    ------>4.运算符" << endl;
			id = 4;
			return buffer;
		}
	}
	if (ch[0] == '+' || ch[0] == '-') {
		if (id == 4) {
			// 当 +/- 出现在 =或者其他运算符之后就代表的是数值的正负
			buffer = fgetc(fp);
			int i = 1;
			ch[i] = buffer;
			if (isdigit(ch[i])) {
				while (isdigit(ch[i]) || ch[i] == 'e' || ch[i] == '.') {
					buffer = fgetc(fp);
					ch[++i] = buffer;
				}
				ch[i + 1] = '\0';
				cout << ch << "    ------>3.常数" << endl;
				buffer = fgetc(fp);
				id = 3;
				return buffer;
			}
			buffer = fgetc(fp);
			id = 4;
			return buffer;
		}
		if (ch[0] == '+') {
			buffer = fgetc(fp);
			if (buffer == '+' || buffer == '=') {
				ch[1] = buffer;
				ch[2] = '\0';
				cout << ch << "    ------>4.运算符" << endl;
				buffer = fgetc(fp);
				id = 4;
				return buffer;
			}
			ch[1] = '\0';
			cout << ch << "    ------>4.运算符" << endl;
			buffer = fgetc(fp);
			return buffer;
		}
		if (ch[0] == '-') {
			buffer = fgetc(fp);
			if (buffer == '-' || buffer == '=') {
				ch[1] = buffer;
				ch[2] = '\0';
				cout << ch << "    ------>4.运算符" << endl;
				buffer = fgetc(fp);
				id = 4;
				return buffer;
			}
			ch[1] = '\0';
			cout << ch << "    ------>4.运算符" << endl;
			buffer = fgetc(fp);
			return buffer;
		}
	}
	if (ch[0] == '#') {
		int i = 1;
		buffer = fgetc(fp);
		if (buffer == 'i') {
			ch[i] = buffer;
			while (ch[i] != '>') {
				buffer = fgetc(fp);
				ch[++i] = buffer;
			}
			ch[i + 1] = '\0';
			cout << ch << "    ------>头文件" << endl;
			buffer = fgetc(fp);
			return buffer;
		}
		if (buffer == 'd') {
			// #define
			ch[i] = buffer;
			while (isalpha(ch[i])) {
				buffer = fgetc(fp);
				ch[++i] = buffer;
			}
			ch[i + 1] = '\0';
			cout << ch << "    ------>预定义" << endl;
			buffer = fgetc(fp);
			return buffer;
		}
	}
	if (ch[0] == '[' || ch[0] == ']' || ch[0] == '\'' || ch[0] == '\"' || ch[0] == '&') {
		cout << ch << "    ------>4.运算符" << endl;
		buffer = fgetc(fp);
		id = 4;
		return buffer;
	}
	if (ch[0] == '\\') {
		buffer = fgetc(fp);
		if (buffer == 'n' || buffer == 't') {
			ch[1] = buffer;
			ch[2] = '\0';
			cout << ch << "    ------>输出控制符" << endl;
			buffer = fgetc(fp);
			return buffer;
		}
	}
	if (ch[0] == '.') {
		buffer = fgetc(fp);
		cout << ch << "    ------>类运算符" << endl;
		return buffer;
	}
}

整合之后，一个简单的C++词法分析器就实现了。
由于一开始没有考虑到小数点，所以此分析器无法对调用类方法的小数点进行分析归类。