cifaanalyse.java

tiger编译器的驱动程序的描述 tiger编译器的驱动程序的描述

package ecust.edu.cn.pl0;

import java.io.BufferedReader;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStreamReader;

public class CiFaAnalyse// 词法分析类
{
	String line; // string变量用于存放读取的一行源代码
	int lineLength;// 存放line中的字符数
	int lineNum = 0;// 存放line的号码
	public char Cha; // 存放读取的字符
	int index;// 存放当前读取的字符位置
	int al = 10; // 定义标识符长度
	int nmax = 4;// 定义整数位数
	public int errorNumber = 0; // 用于保存词法分析中的错误数目
	BufferedReader in;
	static token word[];

	public CiFaAnalyse(String filename) {          //构造函数,根据传过来的文件名.建立一个输入流
		// 定义保留字表word
		word = new token[13];
		word[0] = new token("begin", "beginsym");
		word[1] = new token("call", "callsym");
		word[2] = new token("const", "constsym");
		word[3] = new token("do", "dosym");
		word[4] = new token("end", "endsym");
		word[5] = new token("if", "ifsym");
		word[6] = new token("odd", "oddsym");
		word[7] = new token("procedure", "procsym");
		word[8] = new token("read", "readsym");
		word[9] = new token("then", "thensym");
		word[10] = new token("var", "varsym");
		word[11] = new token("while", "whilesym");
		word[12] = new token("write", "writesym");
		// 根据文件名建立文件输入流
		try {
			FileInputStream fis = new FileInputStream(filename);
			in = new BufferedReader(new InputStreamReader(fis));
			line = in.readLine().trim(); // 读取一行源代码.并去除开头和结尾的空格
			lineLength = line.length();
			lineNum = 1;
			while (lineLength == 0) { // 如果此行源代码为空，则继续读取下一行
				line = in.readLine().trim();
				lineLength = line.length();
				lineNum++;
			}
			index = 0;
			GetCh();
		} catch (FileNotFoundException a) {
			System.out.println("没有找到文件！！");
		} catch (IOException b) {
			System.out.println("文件读取错误");
		}
	}

	public int getErroNumber() {           //返回发生错误的个数 
		return errorNumber;
	}

	public void GetCh()// 利用方法GetCh()在原文件中取得一个符号
	{
		try {
			if (index == lineLength)// 如果到达行尾,则重新读取一行源码
			{ // 其实读取有效行可以写一个方法
				line = in.readLine().trim();
				lineLength = line.length();
				lineNum++;
				while (lineLength == 0) {
					line = in.readLine().trim();
					lineLength = line.length();
					lineNum++;
				}
				index = 0;
				Cha = line.charAt(index);
				index++;
			} else {
				Cha = line.charAt(index);
				index++;
			}
		} catch (NullPointerException e) {
			System.out
					.println("*****************************源程序读取完毕*******************************");
		} catch (IOException c) {
		}
	}

	public token GetToken()// 返回一个token对象 其中名字和类型已经赋值(相当于getsym（）)
	{

		int charNum1;// 记录当前单词的开始索引
		int charNum2;// 记录当前单词的结束索引
		String tokenName;// 当前单词
		int n = 0; // 用于记录保留字数组的下标
		while (Cha == ' ') {
			GetCh();
		}
		if (Cha >= 'a' && Cha <= 'z') {
			charNum1 = index - 1;
			while ((Character.isLetter(Cha) || Character.isDigit(Cha))
					&& index < lineLength) {
				GetCh();
			}
			charNum2 = index;
			if ((charNum2 - charNum1) > al) {
				new erro(lineNum, 19);
				errorNumber++;
				tokenName = line.substring(charNum1, charNum1 + al);
			}// 标识符长度越界
			else {
				if (Character.isLetter(Cha) || Character.isDigit(Cha)) {
					tokenName = line.substring(charNum1);
					GetCh();
				} else {
					tokenName = line.substring(charNum1, charNum2 - 1);
				}
			}
			// 判断保留字过程
			while (n < 13 && (!word[n].getNam().equals(tokenName))) {
				n++;
			}
			if (n == 13) // 如果保留字表中没有找到 则为标识符
			{
				return new token(tokenName, "ident", lineNum);
			} else {
				word[n].setLineNum(lineNum);
				return word[n];
			}
		} // 判断是否是整数
		else if (Character.isDigit(Cha)) {
			charNum1 = index - 1;
			charNum2 = index;
			while (Character.isDigit(Cha) && index < lineLength) {
				charNum2 = index;
				GetCh();
			}
			if (Character.isDigit(Cha))// 判断while为什么停止循环的,如果cha还是数字，则表明是因为此行读完而停止的，如果cha不是数字，表示因为读到非数字而停止
			{
				tokenName = line.substring(charNum1);
				GetCh();
			} else
				tokenName = line.substring(charNum1, charNum2);
			if ((charNum2 - charNum1) > nmax) {
				new erro(lineNum, 8);
				errorNumber++;
				tokenName = tokenName.substring(0, nmax);
			}
			return new token(tokenName, "number", lineNum);
		} else if (Cha == ':') {
			GetCh();
			if (Cha == '=') {
				GetCh();
				return new token(":=", "becomes", lineNum);

			} else
				return new token(":", "null", lineNum);
		} else if (Cha == '<') {
			GetCh();
			if (Cha == '=') {
				GetCh();
				return new token("<=", "leg", lineNum);

			} else
				return new token("<", "lss", lineNum);
		} else if (Cha == '>') {
			GetCh();
			if (Cha == '=') {
				GetCh();
				return new token(">=", "geq", lineNum);
			} else
				return new token(">", "gtr", lineNum);
		} else {
			String sym;
			String sr = Character.toString(Cha);
			switch (Cha) {
			case '+':
				sym = "plus";
				break;
			case '-':
				sym = "minus";
				break;
			case '*':
				sym = "times";
				break;
			case '/':
				sym = "slash";
				break;
			case '(':
				sym = "lparen";
				break;
			case ')':
				sym = "rparen";
				break;
			case '=':
				sym = "eql";
				break;
			case '#':
				sym = "neq";
				break;
			case ',':
				sym = "comma";
				break;
			case '.':
				sym = "period";
				break;
			case ';':
				sym = "semicolon";
				break;
			default:
				sym = "nul";
			}
			GetCh();
			return new token(sr, sym, lineNum);
		}
	}
}