lexical1.cpp

编译器很好用

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

	char *reserve[21]={"and","begin","const","div","do","else","end","function","if","integer",		
		"not","or","procedure","program","read","real","then","type","var","while","write"};	//保留字表
	
	char* symbol[20];		//简易符号表
	int num[20];			//常量表
	char line[100];				//放入文件的一行
	static int ii=0,jj=0;//当前文件读头位置，行未标志，行号

int scanner(ifstream&);
void getnbc(ifstream&, char&);
void getchar(ifstream&, char&);
int search_reserve(char*);
int search_symbol(char*);
int search_num(char*);
void retract(char&);

	
void main( )
{
	int end;
	ifstream sf("c:\\file.txt");
	if(sf.fail())
	{
		cerr<<"error opening file file.txt\n";
		return;
	}
	end=scanner(sf);
	cout<<end<<endl;
	while(end!=26)
	{	
	end=scanner(sf);
	cout<<end<<endl;
	}	
	sf.close();
}

int scanner(ifstream& sf)
{
	char token[11];
	int token_num=0;
	int c;
	char current;
	getnbc(sf,current);

	switch(current)
	{
	case'a':
	case'b':
	case'c':
	case'd':
	case'e':
	case'f':
	case'g':
	case'h':
	case'i':
	case'j':
	case'k':
	case'l':
	case'm':
	case'n':
	case'o':
	case'p':
	case'q':
	case'r':
	case's':
	case't':
	case'u':
	case'v':
	case'w':
	case'x':
	case'y':
	case'z':	token[token_num]=current;	
				token_num++;
				getchar(sf,current);
				while((current<='z'&&current>='a')||(current<='9'&&current>='0'))
				{
					if(token_num<10)
					{
						token[token_num]=current;
						token_num++;
					}
					getchar(sf,current);
				}
				token[token_num]='\0';
				retract(current);
				c=search_reserve(token);
				if(c==-1)
				{
					cout<<token<<"  |";
					return 21;
				}
				else
				{
					cout<<token<<"  |";
					return c;
				}
				break;

	case'0':
	case'1':
	case'2':
	case'3':
	case'4':
	case'5':
	case'6':
	case'7':
	case'8':
	case'9':	token[token_num]=current;	
				token_num++;
				getchar(sf,current);
				while(current<='9'&&current>='0')
				{
					if(token_num<10)
					{
						token[token_num]=current;
						token_num++;
					}
					getchar(sf,current);
				}
				token[token_num]='\0';
				retract(current);
				cout<<token<<"  |";
				return 22;
				break;
	case',':	cout<<','<<"  |";
				return 23;
	case';':	cout<<';'<<"  |";
				return 24;
	case':':	getchar(sf,current);
				if(current=='=')
				{
					cout<<':'<<'='<<"  |";
					return 44;
				}else
				{
					retract(current);
					cout<<':'<<"  |";
					return 25;
				}
	case'.':	getchar(sf,current);
				if(current=='.')
				{
					cout<<'.'<<'.'<<"  |";
					return 31;
				}else
				{
					retract(current);
					cout<<'.'<<"  |";
					return 26;
				}
	case'(':	cout<<'('<<"  |";
				return 27;
	case')':	cout<<')'<<"  |";
				return 28;
	case'[':	cout<<'['<<"  |";
				return 29;
	case']':	cout<<']'<<"  |";
				return 30;
	//缺单目加
	//缺单目减
	case'+':	cout<<'+'<<"  |";
				return 34;
	case'-':	cout<<'-'<<"  |";
				return 35;	
	case'*':	cout<<'*'<<"  |";
				return 36;
	case'/':	cout<<'/'<<"  |";
				return 37;	
	case'=':	cout<<'='<<"  |";
				return 38;
	case'<':	getchar(sf,current);
				if(current=='>')
				{
					cout<<'<'<<'>'<<"  |";
					return 41;
				}
				if(current=='=')
				{
					cout<<'<'<<'='<<"  |";
					return 42;
				}
				else
				{
					retract(current);
					cout<<'<'<<"  |";
					return 39;
				}
	case'>':	getchar(sf,current);
				if(current=='=')
				{
					cout<<'>'<<'='<<"  |";
					return 43;
				}else
				{
					retract(current);
					cout<<'>'<<"  |";
					return 40;
				}
	case'{':	cout<<'{'<<"  |";
				return 45;
	case'}':	cout<<'}'<<"  |";
				return 46;
	case'#':	cout<<'#'<<"  |";
				return 47;
	default:	cout<<"error"<<"  |";
				return -1;
	}
}

void getnbc(ifstream& sf, char& current)
{
GET:	getchar(sf,current);
	
	if(current==' '||current=='\0'||current=='\t')		//滤掉空格，回车，tab键
	{
		do{
			getchar(sf,current);
		}while(current==' '||current=='\0'||current=='\t');
	}
	
	if(current=='{')
	{
		do{
			getchar(sf,current);
		}while(current!='}');
		goto GET;
	}

	
}

void getchar(ifstream& sf, char& current)
{
	static int lineno=1;
	if(ii==jj)
	{
		ii=jj=0;
		sf.getline(line,100);
		cout<<endl<<"line"<<lineno<<":   "<<line<<endl;
		
		for( ;line[jj]!='\0';jj++);

		lineno=lineno+1;
		current=line[ii];
		return;
	}
	
	current=line[++ii];
	if(current>='A'&&current<='Z')
		current+=32;
}

int search_reserve(char* token)
{
	for(int i=0;i<=20;i++)
	{
		if(strcmp(token,reserve[i])==0)
			return i;
	}
	
	return search_symbol(token);
}
int search_symbol(char* token)
{
	for(int j=0;j<=19;j++)
	{
		if(symbol[j]==NULL)
		{
			symbol[j]=new char[11];
			strcpy(symbol[j],token);
				return -1;
		}
		if(strcmp(symbol[j],token)==0)
			return -1;
	}

	cout<<"symbol table is full!"<<endl;
		return -1;
}

void retract(char& current)
{
	ii--;
	current=line[ii];
}