scanner.cpp

编译原理：词法分析器(C++实现) 将PASCAL的源程序

#include<iostream.h>
#include<fstream.h>
#include<string.h>

int m = 0;
int n = 0;
int x[50], y[50];

char * reservechar[] = {"PROGRAM", "CONST", "VAR", "INTEGER", "LONG", "PROCEDURE",
						 "IF", "THEN", "WHILE", "DO", "READ", "WRITE", "BEGIN", "END", "ODD"};
//关键字表

char id[50][10];	//符号表

char cst[50][10];	//常数表

class Scanner
{
private:
	int i, j;	//指针
	int length;
	char ch, buffer[100], strtoken[20];
public:
	Scanner(char str[], int n);
	int isdigit();
	int isletter();
	int reserve();
	int insertid();
	int insertconst();
	void getchar();
	void getbc();
	void concat();
	void retract();
	void scan();
};

Scanner::Scanner(char str[], int n)	//构造函数
{
	strcpy(buffer, str);
	length = n;
	i = j = 0;
}

int Scanner::isdigit()
{
	if(ch>='0' && ch<='9')
		return 1;
	else
		return 0;
}

int Scanner::isletter()
{
	if((ch>='a' && ch<='z') || (ch>='A' && ch<='Z'))
		return 1;
	else
		return 0;
}

void Scanner::getchar()	//将下一输入字符读到ch中，指针后移一字符位置
{
	ch = buffer[i];
	i++;
}

void Scanner::getbc()	//保证ch是一个非空白字符
{
	while(ch == ' ')
		getchar();
}

void Scanner::concat()	//将ch连接到字符串stroken的末尾
{
	strtoken[j] = ch;
	j++;
}

void Scanner::retract()	//置ch为空白字符，指针前移一字符位置
{
	ch = ' ';
	i--;
}

int Scanner::reserve()	//对stroken进行关键字表的查找，返回其编码值
{
	int i, flag = 0;
	for(i=0; i<15; i++)
	{
		if( strncmp(reservechar[i],strupr(strtoken), j) == 0 )
		{
			flag = 1;
			break;
		}
	}
	if(flag == 1) return i+1;
	else return 0;
}

int Scanner::insertid()	//将stroken中的标识符插入符号表，返回在符号表中的位置
{
	for(int a = 0; a < m; a++)
		for(int b = 0; b < x[a]; b++)
			if( strncmp(&id[a][0], strtoken, j) == 0 )
			{
				return m+1;
				break;
			}
	for(a = 0; a < j; a++)
		id[m][a] = strtoken[a];

	x[m] = j;
	m++;
	return m;
}

int Scanner::insertconst()	//将stroken中的常数插入常数表，返回在常数表中的位置
{
	for(int i = 0; i < j; i++)
		cst[n][i] = strtoken[i];
	y[n] = j;
	n++;
	return n;
}

void Scanner::scan()
{
	while(i < length)
	{
		j = 0;
		int code, value;
		strcpy(strtoken, " ");	//置strtoken为空串
		getchar();
		getbc();
		if(isletter())	//如果打头的是字母
		{
			while(isletter() || isdigit())
			{
				concat();
				getchar();
			}
			retract();
			code = reserve();
			if(code == 0)	//如果扫描到的是标识符
			{
				value = insertid();
				cout<<"<34,"<<value<<">"<<'\n';
			}
			else cout<<"<"<<code<<",->"<<'\n';	//如果扫描到的是关键字
		}
		else if(isdigit())	//如果打头的是数字
		{
			while(isdigit())
			{
				concat();
				getchar();
			}
			retract();
			value = insertconst();
			cout<<"<33,"<<value<<">"<<'\n';
		}
		else if(ch == '+')
			cout<<"<16,->"<<'\n'; 
		else if(ch == '-')
			cout<<"<17,->"<<'\n';
		else if(ch == '*')
			cout<<"<18,->"<<'\n';
		else if(ch == '/')
			cout<<"<19,->"<<'\n';
		else if(ch == '=')
			cout<<"<20,->"<<'\n';
		else if(ch == '<')
		{
			getchar();
			if(ch == '>')
				cout<<"<21,->"<<'\n';
			else if(ch == '=')
				cout<<"<23,->"<<'\n';
			else
			{
				retract();
				cout<<"<22,->"<<'\n';
			}
		}
		else if(ch == '>')
		{
			getchar();
			if(ch == '=')
				cout<<"<25,->"<<'\n';
			else
			{
				retract();
				cout<<"<24,->"<<'\n';
			}
		}
		else if(ch == '.')
			cout<<"<26,->"<<'\n';
		else if(ch == ',')
			cout<<"<27,->"<<'\n';
		else if(ch == ';')
			cout<<"<28,->"<<'\n';
		else if(ch == ':')
		{
			getchar();
			if(ch == '=')
				cout<<"<30,->"<<'\n';
			else
			{
				retract();
				cout<<"<29,->"<<'\n';
			}
		}
		else if(ch == '(')
			cout<<"<31,->"<<'\n';
		else if(ch == ')')
			cout<<"<32,->"<<'\n';
		else if(ch == '{')
		{
			while(ch != '}')
				getchar();
		}
 		else cout<<"出错！"<<'\n';
	}
}

void main(void)
{	
	fstream file;
	file.open("G:\My Work\编译原理\词法分析器\source_code.txt", ios::in||ios::nocreate);	//以只读方式打开
	file.unsetf(ios::skipws);	//不跳过文本中的空格
	char buffer[100];
	while(file.getline(buffer, 100))
	{
		Scanner SS(buffer, strlen(buffer));
		SS.scan();
	}

	cout<<"标识符表如下：\n"<<"编号\t"<<"值\n";
	for(int i=0; i<m; i++)
	{
		cout<<i+1<<'\t';
		for(int j=0; j<x[i]; j++)
			cout<<id[i][j];
		cout<<'\n';
	}

	cout<<"常数表如下：\n"<<"编号\t"<<"值\n";
	for(i=0; i<n; i++)
	{
		cout<<i+1<<'\t';
		for(int j=0; j<y[i]; j++)
			cout<<cst[i][j];
		cout<<'\n';
	}
}