scan.c

tiny编译器++

/****************************************************/
/* File: scan.c                                     */
/* The scanner implementation for the TINY compiler */
/* Compiler Construction: Principles and Practice   */
/* Kenneth C. Louden                                */
/****************************************************/

#include "globals.h"
#include "util.h"
#include "scan.h"

/* states in scanner DFA */
typedef enum
{
	START, INASSIGN, SQM,/* new add state */INCOMMENT, INNUM, INID, DONE
} StateType;

/* lexeme of identifier or reserved word */
char tokenString[MAXTOKENLEN + 1];

/* BUFLEN = length of the input buffer for
 source code lines */
#define BUFLEN 256

static char lineBuf[BUFLEN]; /* holds the current line */
static int linepos = 0; /* current position in LineBuf */
static int bufsize = 0; /* current size of buffer string */
static int EOF_flag = FALSE; /* corrects ungetNextChar behavior on EOF */

/* getNextChar fetches the next non-blank character
 from lineBuf, reading in a new line if lineBuf is
 exhausted */
static int getNextChar(void)
{
	if (!(linepos < bufsize))
	{
		lineno++;
		if (fgets(lineBuf, BUFLEN - 1, source))
		{
			bufsize = strlen(lineBuf);
			if (EchoSource)
			{
				/* new add code */
				if (10 == lineBuf[bufsize - 1])
				{
					fprintf(listing, "%4d: %s", lineno, lineBuf);
				}
				else
				{
					fprintf(listing, "%4d: %s\n", lineno, lineBuf);
				}
				/* end add */
			}
			linepos = 0;
			return lineBuf[linepos++];
		}
		else
		{
			EOF_flag = TRUE;
			return EOF;
		}
	}
	else
		return lineBuf[linepos++];
}

/* ungetNextChar backtracks one character
 in lineBuf */
static void ungetNextChar(void)
{
	if (!EOF_flag)
		linepos--;
}

/* ungetNextToken backtracks one token
 in lineBuf */
void ungetNextToken(void)
{
	if (!EOF_flag)
	{
		linepos -= strlen(tokenString);
	}
}

/* lookup table of reserved words */
static struct
{
	char* str;
	TokenType tok;
} reservedWords[MAXRESERVED] =
{
{ "if", IF },
{ "then", THEN },
{ "else", ELSE },
{ "end", END },
{ "repeat", REPEAT },
{ "until", UNTIL },
{ "read", READ },
{ "write", WRITE },
{ "true", TINYTRUE },
{ "false", TINYFALSE },
{ "or", OR },
{ "and", AND },
{ "not", NOT },
{ "int", INT },
{ "bool", BOOL },
{ "string", STRING },
{ "while", WHILE },
{ "do", DO } };

/* lookup an identifier to see if it is a reserved word */
/* uses linear search */
static TokenType reservedLookup(char * s)
{
	int i;
	for (i = 0; i < MAXRESERVED; i++)
		if (!strcmp(s, reservedWords[i].str))
			return reservedWords[i].tok;
	return ID;
}

/****************************************/
/* the primary function of the scanner  */
/****************************************/
/* function getToken returns the
 * next token in source file
 */
TokenType getToken(void)
{ /* index for storing into tokenString */
	int tokenStringIndex = 0;
	/* holds current token to be returned */
	TokenType currentToken;
	/* current state - always begins at START */
	StateType state = START;
	/* flag to indicate save to tokenString */
	int save;
	while (state != DONE)
	{
		int c = getNextChar();
		save = TRUE;
		switch (state)
		{
		case START:
			if (isdigit(c))
				state = INNUM;
			else if (isalpha(c))
				state = INID;
			else if (c == ':')
				state = INASSIGN;
			else if ((c == ' ') || (c == '\t') || (c == '\n'))
				save = FALSE;
			/* new add code */
			else if (c == '\'')
			{
				save = FALSE;
				state = SQM;
			}
			/* end add */
			else if (c == '{')
			{
				save = FALSE;
				state = INCOMMENT;
			}
			else
			{
				state = DONE;
				switch (c)
				{
				case EOF:
					save = FALSE;
					currentToken = ENDFILE;
					break;
				case '=':
					currentToken = EQ;
					break;
				case '<':
					currentToken = LT;
					break;
				case '+':
					currentToken = PLUS;
					break;
				case '-':
					currentToken = MINUS;
					break;
				case '*':
					currentToken = TIMES;
					break;
				case '/':
					currentToken = OVER;
					break;
				case '(':
					currentToken = LPAREN;
					break;
				case ')':
					currentToken = RPAREN;
					break;
					/* new add code */
				case ',':
					currentToken = COMMA;
					break;
					/* end add */
				case ';':
					currentToken = SEMI;
					break;
				default:
					currentToken = ERROR;
					break;
				}
			}
			break;
			/* new add code*/
		case SQM:
			save = TRUE;
			if (c == '\n' || c == EOF)
			{
				state = DONE;
				currentToken = ERROR;
			}
			else if (c == '\'')
			{
				save = FALSE;
				state = DONE;
				currentToken = STR;
			}
			break;
			/* end add */
		case INCOMMENT:
			save = FALSE;
			if (c == EOF)
			{
				state = DONE;
				/* modify */
				currentToken = ERROR;
			}
			else if (c == '}')
				state = START;
			break;
		case INASSIGN:
			state = DONE;
			if (c == '=')
				currentToken = ASSIGN;
			else
			{ /* backup in the input */
				ungetNextChar();
				save = FALSE;
				currentToken = ERROR;
			}
			break;
		case INNUM:
			if (!isdigit(c))
			{ /* backup in the input */
				ungetNextChar();
				save = FALSE;
				state = DONE;
				currentToken = NUM;
			}
			break;
		case INID:
			if (!isalpha(c))
			{ /* backup in the input */
				ungetNextChar();
				save = FALSE;
				state = DONE;
				currentToken = ID;
			}
			break;
		case DONE:
		default: /* should never happen */
			fprintf(listing, "Scanner Bug: state= %d\n", state);
			state = DONE;
			currentToken = ERROR;
			break;
		}
		if ((save) && (tokenStringIndex <= MAXTOKENLEN))
			tokenString[tokenStringIndex++] = (char) c;
		if (state == DONE)
		{
			tokenString[tokenStringIndex] = '\0';
			if (currentToken == ID)
				currentToken = reservedLookup(tokenString);
		}
	}
	if (TraceScan)
	{
		fprintf(listing, "\t%d: ", lineno);
		printToken(currentToken, tokenString);
	}
	return currentToken;
} /* end getToken */