scaner.h

学编译原理时的课程作业,里面包含抽象语法树和程序源代码.

#ifndef SCANER_H
#define SCANER_H
#include "macro.h"
#include "Pair.h"
using namespace std;

deque<Pair *> pairs;

class KeyWord{
public:
	string word;
	int id;
	KeyWord(string word = "", int id = 0){
		this->word = word;
		this->id = id;
	}
};


class Scaner{
private:
	KeyWord keyword[6];
	string token;
	ifstream &in;
	int lineNo ;
public:
	Scaner(ifstream &in): in(in){
		keyword[0] = KeyWord("else", 0);
		keyword[1] = KeyWord("if", 1);
		keyword[2] = KeyWord("int", 2);
		keyword[3] = KeyWord("return", 3);
		keyword[4] = KeyWord("void", 4);
		keyword[5] = KeyWord("while", 5);
		
	
		token = "";
		lineNo = 1;
	}

	int lookup(string word)	{

		for(int i=0;i<6;i++) {

			if(word == keyword[i].word)
			return i;
		}
		return -1;
	}

	void outPut(int id ,string word) {
	 
		cout << "(" << ((id<10)?" ":"") << id << "," << word << ")" << endl;
		
		if( B11 == id || B12 == id) {// when it is a quote

			pairs.push_back(new Pair(id));
		}
		else if (id == INT){
			
			pairs.push_back(new Pair(id, atoi(word.c_str())));//value type
		}
		else if (id == ADD || id == MINUS || MULT == id) {
			
			pairs.push_back(new Pair(id));
		}
		else {
			
			cout << "Illegal char!" << endl;
		}
	}

	
	void display_char(char ERR_CH) {
	
		cout<< "undeclared identifler "<< ERR_CH << " at " << lineNo << " line!" << endl;
	}

	bool isalpha(char c) {

		if( (c>='a'&&c<='z') || (c>='A'&&c<='Z') )
			return true;
		else
			return false;
	}

	bool isdigit(char c) {

		if(c>='0'&&c<='9')
			return true;
		else
			return false;
	}
	
	void scan() {
		char ch;
		int c;
 
		while(!in.eof()) {
			
			in.get(ch);
			if(in.eof()) 
				break;

			if(isalpha(ch))/*当以字母打头时*/ {
				token = "";
				token += ch;
				
				in.get(ch);				
				
				while(isalpha(ch)) {
					
					token += ch;
					in.get(ch);
					if (in.eof()) {

						break;
					}
				}

				in.putback(ch);
				c=lookup(token);
			
				if(c==-1) //当不是关键字
					outPut(ID,token);
				else//是关键字
					outPut(c,keyword[c].word);
			}
			else if(isdigit(ch))/*当以数字打头时*/{
				
				token = "";
				token += ch;
				in.get(ch);
				
				while(isdigit(ch))/*取得其后所有数字*/{
				
					token += ch;
					in.get(ch);
					if (in.eof()) {
				
						break;
					}
				}
			
				in.putback(ch);
				outPut(INT,token);
		}
		else//第一个字不是数字，不是ID
			switch(ch)
			{
				case '<': 
					in.get(ch);
					if(ch=='=') 
						outPut(LE,"<=");
					else {
						
						in.putback(ch);
						outPut(L,"<");	
					}
					break;

				case '=': 
					in.get(ch);
					if(ch=='=')
						outPut(EQ, "==");
					else {
						in.putback(ch);
						outPut(ASS, "=");
					}
					break;	

				case '>':
					in.get(ch);
					if(ch=='=')
						outPut(ME,">=");
					else {
						in.putback(ch);
						outPut(M,">");
					}
					break;

				case '!': 
					
					in.get(ch);
					if(ch=='=')              
						outPut(NE,"!=");
					else {
						in.putback(ch);
						display_char('!');
					}
					break;

				case '+':
					outPut(ADD, "+");
					break;
	
				case '-':
					outPut(MINUS, "-");
					break;

				case '*':
					outPut(MULT, "*");
					break;

				case ';':
					outPut(SECO, ";");
					break;

				case ',':
					outPut(COMA, ",");
					break;
	
				case '/':
					in.get(ch);   //删除程序中的注释
					if(ch=='*') {//注释开始
						A: do
							{
								in.get(ch);
								if (in.eof()) {
									break;
								}
								if('\n' == ch)
									lineNo++;
							}while(ch !='*');
							//遇到*

							in.get(ch);
							while ('*' == ch) {
								in.get(ch);
								if (in.eof()) {
									break;
								}
							}

							if('/' != ch)//取完所有的*后如果紧接着的不是/
								goto A;
							//如果是/，则注释结束
					}
					else {//仅是除号
						in.putback(ch);
						outPut(DIV,"/");
					}
					break;
						 
				case '(':
					outPut(B11,"(");
					break;

				case ')':
					outPut(B12,")");
					break;

				case '[':
					outPut(B21,"[");
					break;
		
				case ']':
					outPut(B22,"]");
					break;

				case '{':
					outPut(B31,"{");
					break;
			
				case '}':
					outPut(B32,"}");
					break;

				case ' ' : 
					break;
					//删除程序中的空格
			
				case '\n':
					lineNo++;
					break;   //删除程序中的回车,并记录程序编译到第几行  
			
				case '	':
					break;      //删除程序中的横向制表符

				case -1 : 
					break;      //删除文件尾符号

				default : display_char(ch);
					break;
			}

		}

		return;
	}

	~Scaner(){

		in.close();
	}
};

#endif