lexicalanalyzer.cpp

一个用C++写的编译器

/*
   ADT for lexicalAnalyzer,has a symbolTable in it.
*/
#include "stdafx.h"


using namespace std;

LexicalAnalyzer::LexicalAnalyzer()
{}

void LexicalAnalyzer::setFilename(string a)
{
	filename=a;
	input.open(filename.c_str());
	cTable.initiate(filename);//initiate the symbolTable
	line=0;//first line is 0
	cLength=0;
	noT=new symbolElement;
}
symbolElement* LexicalAnalyzer::yylex()
{
	
	cLength=0;//the contents in lexemeArr become meaningless.
	int state=-1;//state -1 means hasn't found the entry for current lexeme.
	char tempC=input.get();//read a character.
	inBuffer(buffer,tempC);//inseert a character into buffer.
	state=findEntry(tempC);//find entry for current lexeme according to the first character.
	while(!DFA(state,tempC))//If can be sure of the token for current lexeme go on looping.
	{
		if(!input.eof()&&cLength<64)
		{
			tempC=input.get();
			inBuffer(buffer,tempC);
		}
		else if((input.eof()||cLength>=64)&&(state!=4&&state!=6&&state!=7&&state!=8&&state!=11&&state!=12
			&&state!=15&&state!=17&&state!=19&&state!=20&&state!=21&&state!=22&&state!=22
			&&state!=23&&state!=25&&state!=26&&state!=27&&state!=29&&state!=30&&state!=34
			&&state!=35&&state!=41&&state!=43))
	    //if meets the end of the source file or the maximum length of an id was reached,
		//but the current state for the analyzing lexeme is not an acceptable one.
		{
			cLength=0;//the string in the buffer is not available.
			skip();//skip the rest of the current string that is read.
			return 0;//Lexical error occurs.
		}
		else
		{
			break;
		}
	}
	if(state==-2)
	//If other error occurs such as undifined symbols was met.
	{
		cLength=0;//the string in the buffer is not available
		return 0;//Lexical error occurs.
	}
	if(cLength>0&&state!=4&&state!=5&&state!=6&&state!=7&&state!=8)
	//If the state for current lexeme is not whitespace or comments
	{
		string a="";
		for(int l=0;l<cLength;l++)
		{
			a+=buffer[l];
		}
		//read the lexeme from buffer put it into a.
		string tok;
		switch(state)
		{
			//these states are acceptable ones.
			case 11:
				tok="div";break;
			case 12:
				if(a=="+")tok="plus";
				if(a=="-")tok="minus";
				if(a=="*")tok="mul";break;
			case 14:
				if(a=="==")tok="eq";
				if(a=="<=")tok="le";
				if(a==">=")tok="ge";break;
			case 15:
				if(a=="=")tok="as";
				if(a=="<")tok="lt";
				if(a==">")tok="gt";break;
			case 17:
				tok="ne";break;
			case 19:
				tok="rp";break;
			case 20:
				tok="lp";break;
			case 21:
				tok="lb";break;
			case 22:
				tok="smc";break;
			case 23:
				tok="rb";break;
			case 25:case 26:case 27:
				tok="id";break;
			case 34:case 35:case 30:case 37:case 38:case 41:case 43:case 29:
				tok="num";break;
			default:tok="undefined";
		}
		//assign the token value for the current lexeme according to its current state.
		curT=cTable.insert(a,tok);//insert the current token into symbolTable.
		return curT;//return the pointer points to the newly inserted token in symbolTable.
	}
	else
	//if the current token is comments or whitespace.
	{
		cLength=0;//the contents in lexemeArr become meaningless.
		return noT;//noT is the special node for this.
	}
}

void LexicalAnalyzer::skip()
/*
  skip the current string until meets 
  a ' 'or '\n'.
  increase the line by 1 if meets '\n'.
*/
{
	char fs;
	input.get(fs);
	while(!input.eof()&&fs!=' '&&fs!='\n')
	{
		input.get(fs);
	}
	if(fs=='\n')
	line++;
}

void LexicalAnalyzer::inBuffer(char buffer[],char a)
//Insert a into buffer.Increase cLength by 1.
{
	buffer[cLength++]=a;
}


int LexicalAnalyzer::findEntry(char a)
//if can't find entry for a, set state -2,means undefined symbol was met.
{
	switch (a)
	{
		case '\n':
		case ' ':
		case '\t':return 5;
		case '/':case '+':case '-':case '*':case '!':case '<':case '>':case '=':
			return 9;
		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':
		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':
			return 24;
		case '0':case '1':case '2':case '3':case '4':case '5':case '6':case '7':case '8':
		case '9':
			return 28;
		case '(':case ')':case '{':case '}':case ';':
			return 18;
		default:
			return -2;

	}
}

bool LexicalAnalyzer::DFA(int &state, char tempC)
/*
  according to the current state and tempC,find the next state.
*/
{
	switch (state)
	{
		case -2:
			goto error;
		case 2:
			if(tempC!='\n')
			{
				state=3;
				return false;
			}
			else
			{
				state=4;
				line++;
				return true;
			}
		case 3:
			if(tempC!='\n')
			{
				state=3;
				return false;
			}
			else
			{
				state=4;
				line++;
				return true;
			}
		case 4:return true;
			
		case 5:
			if(tempC=='\n')
			{
				state=7;
				line++;
				return false;
			}
			else if(tempC==' ')
			{
				state=6;
				return false;
			}
			else if(tempC=='\t')
			{
				state=8;
				return false;
			}
			else
			{
				goto error;
			}
		case 6:
			if(tempC==' ')
			{
				state=6;
				return false;
			}
			else if(tempC=='\t')
			{
				state=8;
				return false;
			}
			else if(tempC=='\n')
			{
				state=7;
				line++;
				return false;
			}
			else
			{
				cLength--;
				input.unget();
				return true;
			}
		case 7:
			if(tempC==' ')
			{
				state=6;
				return false;
			}
			else if(tempC=='\n')
			{
				state=7;
				line++;
				return false;
			}
			else if(tempC=='\t')
			{
				state=8;
				return false;
			}
			else
			{
				cLength--;
				input.unget();
				return true;
			}
		case 8:
			if(tempC==' ')
			{
				state=6;
				return false;
			}
			else if(tempC=='\n')
			{
				state=7;
				line++;
				return false;
			}
			else if(tempC=='/t')
			{
				state=8;
				return false;
			}
			else
			{
				cLength--;
				input.unget();
				return true;
			}
		case 9:
			if(tempC=='/')
			{
				state=10;
				return false;
			}
			else if(tempC=='+'||tempC=='-'||tempC=='*')
			{
				state=12;
				return true;
			}
			else if(tempC=='='||tempC=='<'||tempC=='>')
			{
				state=13;
				return false;
			}
			else if(tempC=='!')
			{
				state=16;
				return false;
			}
			else
			{
				goto error;
			}
		case 10:
			{
				if(tempC!='/')
				{
					state=11;
					cLength--;
					input.unget();
					return true;
				}
				else
				{
					state=2;
					return false;
				}
			}
		case 13:
			if(tempC=='=')
			{
				state=14;
				return true;
			}
			else
			{
					state=15;
					cLength--;
					input.unget();
					return true;
			}
		case 16:
			if(tempC=='=')
			{
				state=17;
				return true;
			}
			else
			{
				goto error;
			}
		case 18:
			if(tempC=='(')
			{
				state=20;
				return true;
			}
			else if(tempC==')')
			{
				state=19;
				return true;
			}
			else if(tempC=='{')
			{
				state=21;
				return true;
			}
			else if(tempC=='}')
			{
				state=23;
				return true;
			}
			else if(tempC==';')
			{
				state=22;
				return true;
			}
			else 
			{
				goto error;
			}
		case 24:
			if(isalpha(tempC))
			{
				state=25;
				return false;
			}
			else
			{
				goto error;
			}
		case 25:
			if(isalpha(tempC))
			{
				state=26;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=27;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 26:
			if(isalpha(tempC))
			{
				state=26;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=27;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 27:
			if(isalpha(tempC))
			{
				state=26;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=27;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 28:
			if(isdigit(tempC))
			{
				state=29;
				return false;
			}
			else
			{
				goto error;
			}
		case 29:
			if(tempC=='E'||tempC=='e')
			{
				state=31;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=30;
				return false;
			}
			else if(tempC=='.')
			{
				state=36;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 30:
			if(isdigit(tempC))
			{
				state=30;
				return false;
			}
			else if(tempC=='E'||tempC=='e')
			{
				state=31;
				return false;
			}
			else if(tempC=='.')
			{
				state=36;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 31:
			if(tempC=='+')
			{
				state=32;
				return false;
			}
			else if(tempC=='-')
			{
				state=33;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=34;
				return false;
			}
			else
			{
				goto error;
			}
		case 32:
			if(isdigit(tempC))
			{
				state=34;
				return false;
			}
			else
			{
				goto error;
			}
		case 33:
			if(isdigit(tempC))
			{
				state=34;
				return false;
			}
			else
			{
				goto error;
			}
		case 34:
			if(isdigit(tempC))
			{
				state=35;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 35:
			if(isdigit(tempC))
			{
				state=35;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 36:
			if(isdigit(tempC))
			{
				state=37;
				return false;
			}
			else
			{
				goto error;
			}
		case 37:
			if(isdigit(tempC))
			{
				state=38;
				return false;
			}
			else if(tempC=='E'||tempC=='e')
			{
				state=39;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 38:
			if(isdigit(tempC))
			{
				state=38;
				return false;
			}
			else if(tempC=='E'||tempC=='e')
			{
				state=39;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 39:
			if(tempC=='+')
			{
				state=40;
				return false;
			}
			else if(isdigit(tempC))
			{
				state=41;
				return false;
			}
			else if(tempC=='-')
			{
				state=42;
				return false;
			}
			else
			{
				goto error;
			}
		case 40:
			if(isdigit(tempC))
			{
				state=41;
				return false;
			}
			else
			{
				goto error;
			}
		case 41:
			if(isdigit(tempC))
			{
				state=43;
				return false;
			}
			else
			{
					cLength--;
					input.unget();
					return true;
			}
		case 42:
			if(isdigit(tempC))
			{
				state=41;
				return false;
			}
			else
			{
				goto error;
			}
		case 43:
			if(isdigit(tempC))
			{
				state=43;
				return false;
			}
			else 
			{
					cLength--;
					input.unget();
					return true;
			}
	}




	error:
		{
			for(;cLength>0;cLength--)
			{
				input.unget();
			}
			//if error occurs,push back the character in buffer.
			skip();
			//skip the string.
			state=-2;
			//set state -2.
			return true;
			//be sure of an occuring error.
		}
}
bool LexicalAnalyzer::isEnd()
{
	input.get();
	if(input.eof())
	{
		input.unget();
		return true;
	}
	else
	{
		input.unget();
		return false;
	}
}

void LexicalAnalyzer::ungetT()
{
	for(;cLength>0;cLength--)
	{
		input.unget();
	}
}

LexicalAnalyzer::~LexicalAnalyzer()
{
	delete noT;
}