lexical_analyzer.h

这个是编译原理的词法分析的实验报告！很不错

#ifndef LEXICAL_H
#define LEXICAL_H

#include "global.h"
#include "error.h"
#include "symbol.h"


long tokenval=NONE;				
int lineno=1;								//line num
int position=1;								//position




char lexbuf[BSIZE];							//buf ,


to_parser &lexan(FILE *file);
char * get_reserved(int p);					//get reserved_word string



to_parser &lexan(FILE *file)
{
	int t;		//used to receive char



struct to_parser q;


	while(1)
	{
		t=getc(file);						//读一个字符，文件读
			//t=getchar();std	
		if(t==' ' || t=='\t')				//white space or tab
			;
		else if(t=='\n')					//return 
		{lineno++;position=1;}
	
////////////////////////////////////////////////////////////////////
		else if(isdigit(t))					//digit
		{
			ungetc(t,file);					//倒车
				//ungetc(t,stdin);	std	

			fscanf(file, "%ld",&tokenval);	//读文件
																		//for debug
			q.token=NUM;
			q.line_no=lineno;
			q.position=position;
			q.attr.long_attr=tokenval;
			position++;
		
			return q;
		}
//////////////////////////////////////////////////////////////////////
		else if(isalpha(t))					//id or keywords
		{
			int p;							//used as return value
			int b=0;						//the position of buf
			while(isalnum(t))
			{
				lexbuf[b]=t;
				t=getc(file);
				b++;
				if(b>=BSIZE)
					error("compiler error");
			}
			lexbuf[b]=EOS;
			if(t!=EOF)
				ungetc(t,file);
			p=lookup(lexbuf);
			if(p==0)
			{
				p=insert(lexbuf,ID);
				q.token=ID;

			}
			else if(p<7)
			{
				q.token=p;

			}
			else
			{
				q.token=ID;

			}			
			tokenval=p;
		
			q.line_no=lineno;
			q.position=position;
			q.attr.str_attr=lexbuf;
			position++;

			return q;
		}
////////////////////////////////////////////////////////////////////////////
		else if(t==EOF)			//end of file
		{
			q.token=DONE;
			return q;
		}

////////////////////////////////////////////////////////////////////////////
		else if(t=='+' || t=='-' || t=='*'  )	// operator
		{
		
			q.token=OPERATOR;
			q.line_no=lineno;
			q.position=position;
			sprintf(lexbuf,"%c",t);
			q.attr.str_attr=lexbuf;
		
			position++;
		
			return q;

		}
		else if(t=='/')
		{
			char next_t;
			next_t=getc(file);
			if(next_t=='/')			//不考虑 注释内容，
			{	fgets(lexbuf,1000,file); position=1; lineno++;}
			else 
			{
				ungetc(next_t,file);
				
				q.token=OPERATOR;
				q.line_no=lineno;
				q.position=position;
				sprintf(lexbuf,"%c",t);
				q.attr.str_attr=lexbuf;
		
				position++;
			}
		}

////////////////////////////////////////////////////////////////////////////
		else if(t=='=' ||t=='>' || t=='<' ||t=='!')		//operator
		{
			char next_t;
			next_t=getc(file);
			if(next_t!='=')
			{
				ungetc(next_t,file);
				sprintf(lexbuf,"%c",t);
				q.attr.str_attr=lexbuf;
			}
			else 
			{
				sprintf(lexbuf,"%c%c",t,next_t);
				q.attr.str_attr=lexbuf;
			}

			q.token=OPERATOR;
			q.line_no=lineno;
			q.position=position;

			position++;
				
			return q;

		}
//////////////////////////////////////////////////////////////////////////////////
		else if(t=='(' || t==')' || t==';' ||t=='{' || t=='}')
		{
			
			q.token=DELIMITER;
			q.line_no=lineno;
			q.position=position;
			q.attr.char_attr=t;

				position++;
			
			return q;
		
		}
//////////////////////////////////////////////////////////////////////////////
		else		//other invalid character
		{
			tokenval=NONE;
			char next_t;
			next_t=getc(file);
			if(next_t==' ' ||next_t=='/t')
				position++;
			else
				ungetc(next_t,file);


		}


		
			
	
	}

}

char * get_reserved(int p)
{

	switch(p)
	{
	case 1: return "IF";
	case 2: return "THEN";
	case 3: return "WHILE";
	case 4: return "REAL";
	case 5: return "INT";
	case 6: return "ELSE";

	}
}



#endif