0920编译原理词法分析

#include<stdio.h>
#include<stdlib.h>
#include<string.h>
#define _KEY_WOED_END "waiting for your expanding"     //关键字结束标志

typedef struct
{
	int typenum;
	char * word;
}WORD;

char input[255];   //输入换缓冲区 

char token[255]="";   //单词缓冲区

int p_input;        //输入换缓冲区指针

int p_token;        //单词缓冲区指针

char ch;           //当前所读的字符

char *rwtab[]={"begin","if","then","while","do","end",_KEY_WOED_END};        //C语言关键字

WORD * scaner();    //词法扫描函数，获得关键字






main()
{
	int over=1;
	WORD *oneword;
	oneword=(WORD *)malloc(sizeof(WORD));
	printf("请输入您的字符串（以#作为结束标志）:");
	scanf("%[^#]s",input);                        //读入源程序字符串到缓冲区，以#结束，允许多行输入
	p_input=0;
	printf("您输入的字符串是:%s

",input);
	while(over<1000&&over!=-1)
	{
		oneword=scaner();
		printf("(%d,%s)
",oneword->typenum,oneword->word);
		over=oneword->typenum;
	}
	printf("

备注:数字10代表所输入的不是关键词，只是普通词
");
	printf("     数字20代表您输入的是数字
");
    printf("     数字1000是结束标志

");

}


//需要用到的自编函数参考实现
//从输入缓冲区读取一个字符到ch中

char m_getch(){
	ch=input[p_input];
	p_input=p_input+1;
	return (ch);
}



//去掉空白字符
void getbc()
{
	while(ch==' '||ch==10){
	ch=input[p_input];
	p_input=p_input+1;
	}
}



//拼接单词
void concat()
{
	token[p_token]=ch;
	p_token=p_token+1;
	token[p_token]='';
}



//判断是否字母
int letter()
{
	if(ch>='a'&&ch<='z'||ch>='A'&&ch<='Z')return 1;
	else return 0;
}


//判断是否数字
int digit()
{
	if(ch>='0'&&ch<='9')
		return 1;
	else
		return 0;
}


//检索关键字表格
int reserve()
{
	int i=0;
	for(i=0;i<7;i++)
	{
		if(!strcmp(rwtab[i],token))
		{
			return i+1;
		}
		i=i+1;
	}
	return 10;
}




//回退一个字符
void retract()
{
	p_input=p_input-1;
}




WORD *scaner()
{
	WORD *myword;
	myword=(WORD *)malloc(sizeof(WORD));
	myword->typenum=10;
	myword->word="";
	p_token=0;
	m_getch();
	getbc();
	if(letter())
	{
		while(letter()||digit())
		{
		   concat();
		   m_getch();
		
		}
		retract();
 	    myword->typenum=reserve();
	    myword->word=token;
	    return(myword);
	}
	else if(digit())
	{
		while(digit())
		{
			concat();
			m_getch();
		}
		retract();
		myword->typenum=20;
	    myword->word=token;
		return(myword);
	}
	else
	{
		switch(ch)
		{
		case '=':m_getch();
			if(ch=='=')
			{
				myword->typenum=39;
	            myword->word="==";
		        return(myword);
			}
			retract();
			myword->typenum=21;
	        myword->word="=";
		    return(myword);
			break;
		case '+':
			myword->typenum=22;
	            myword->word="+";
		        return(myword);
				break;
		case '-':
			myword->typenum=23;
	            myword->word="-";
		        return(myword);
				break;
		case '*':
			myword->typenum=24;
	        myword->word="*";
		    return(myword);
		    break;
		case '/':
			myword->typenum=25;
	        myword->word="/";
		    return(myword);
		    break;
		case '(':
			myword->typenum=26;
	        myword->word="(";
		    return(myword);
		    break;
		case ')':
			myword->typenum=27;
	        myword->word=")";
		    return(myword);
		    break;
		case '[':
			myword->typenum=28;
	        myword->word="[";
		    return(myword);
		    break;
		case ']':
			myword->typenum=29;
	        myword->word="]";
		    return(myword);
		    break;
		case '{':
			myword->typenum=30;
	        myword->word="{";
		    return(myword);
		    break;
		case '}':
			myword->typenum=31;
	        myword->word="}";
		    return(myword);
		    break;
		case ',':
			myword->typenum=32;
	        myword->word=",";
		    return(myword);
		    break;
		case ':':
			myword->typenum=33;
	        myword->word=":";
		    return(myword);
		    break;
		case ';':
			myword->typenum=34;
	        myword->word=";";
		    return(myword);
		    break;
		case '>':
			myword->typenum=35;
	        myword->word=">";
		    return(myword);
		    break;
		case '<':
			myword->typenum=36;
	        myword->word="<";
		    return(myword);
		    break;
		case '!':
			m_getch();
			if(ch=='=')
			{
				myword->typenum=40;
				myword->word="!=";
				return(myword);
			}
			retract();
			myword->typenum=-1;
	        myword->word="ERROR";
		    return(myword);
		    break;
		case '':
			myword->typenum=1000;
	        myword->word="OVER";
		    return(myword);
		    break;
		default:
			myword->typenum=-1;
	        myword->word="ERROR";
		    return(myword);
			}
		}
	}