lex.c

unix环境下实现的cmm语言编译器

#include "cmm.h"

int t;			/*当前的标记*/
char *token;	/*指向标示符*/
Coordinate src; /*标记的位置 */
Symbol tsym;	/*指向临时符号存储的指针*/
static struct symbol tval; /*用于存储临时的符号*/	

enum {BLANK=01,NEWLINE=02,LETTER=04,
	  DIGIT=010,OTHER=020};
static unsigned char map[256] = {
	/* 000 null	*/ 	0,
	/* 001 soh	*/ 	0,
	/* 002 stx	*/ 	0,
	/* 003 etx	*/ 	0,
	/* 004 eot	*/ 	0,
	/* 005 enq	*/ 	0,
	/* 006 ack	*/ 	0,
	/* 007 bel	*/ 	0,
	/* 010 bs	*/ 	0,
	/* 011 ht	*/ 	BLANK,
	/* 012 lf	*/ 	NEWLINE,
	/* 013 vt	*/ 	BLANK,
	/* 014 ff	*/ 	BLANK,
	/* 015 cr	*/ 	0,
	/* 016 so	*/ 	0,
 	/* 017 si	*/ 	0,
	/* 020 dle	*/ 	0,
	/* 021 dc1	*/ 	0,
	/* 022 dc2	*/ 	0,
	/* 023 dc3	*/ 	0,
	/* 024 dc4	*/ 	0,
	/* 025 nak	*/ 	0,
	/* 026 syn	*/ 	0,
	/* 027 etb	*/ 	0,
	/* 030 can	*/ 	0,
	/* 031 em 	*/ 	0,
	/* 032 sub	*/ 	0,
	/* 033 esc	*/ 	0,
	/* 034 fs 	*/ 	0,
	/* 035 gs 	*/ 	0,
	/* 036 rs	*/ 	0,
	/* 037 us	*/ 	0,
	/* 040 sp	*/ 	BLANK,
	/* 041 !	*/ 	OTHER,
	/* 042 "	*/ 	OTHER,
	/* 043 #	*/ 	0,
	/* 044 $	*/ 	0,
	/* 045 %	*/ 	OTHER,
	/* 046 &	*/ 	OTHER,
	/* 047 '	*/ 	OTHER,
	/* 050 (	*/ 	OTHER,
	/* 051 )	*/ 	OTHER,
	/* 052 *	*/ 	OTHER,
	/* 053 +	*/ 	OTHER,
	/* 054 ,	*/ 	OTHER,
	/* 055 -	*/ 	OTHER,
	/* 056 .	*/ 	0,
	/* 057 /	*/ 	OTHER,
	/* 060 0	*/ 	DIGIT,
	/* 061 1	*/ 	DIGIT,
	/* 062 2	*/ 	DIGIT,
	/* 063 3	*/ 	DIGIT,
	/* 064 4	*/ 	DIGIT,
	/* 065 5	*/ 	DIGIT,
	/* 066 6	*/ 	DIGIT,
	/* 067 7	*/ 	DIGIT,
	/* 070 8	*/ 	DIGIT,
	/* 071 9	*/ 	DIGIT,
	/* 072 :	*/ 	0,
	/* 073 ;	*/	OTHER,
	/* 074 <   */	OTHER,
	/* 075 =   */	OTHER,
	/* 076 >   */	OTHER,
	/* 077 ?   */	0,
	/* 100 @   */	0,
	/* 101 A   */	LETTER,
	/* 102 B   */	LETTER,
	/* 103 C   */	LETTER,
	/* 104 D   */	LETTER,
	/* 105 E   */	LETTER,
	/* 106 F   */	LETTER,
	/* 107 G   */	LETTER,
	/* 110 H   */	LETTER,
	/* 111 I   */	LETTER,
	/* 112 J   */	LETTER,
	/* 113 K   */	LETTER,
	/* 114 L   */	LETTER,
	/* 115 M   */	LETTER,
	/* 116 N   */	LETTER,
	/* 117 O   */	LETTER,
	/* 120 P   */	LETTER,
	/* 121 Q   */	LETTER,
	/* 122 R   */	LETTER,
	/* 123 S   */	LETTER,
	/* 124 T   */	LETTER,
	/* 125 U   */	LETTER,
	/* 126 V   */	LETTER,
	/* 127 W   */	LETTER,
	/* 130 X   */	LETTER,
	/* 131 Y   */	LETTER,
	/* 132 Z   */	LETTER,
	/* 133 [   */	OTHER,
	/* 134 \   */	0,
	/* 135 ]   */	OTHER,
	/* 136 ^   */	0,
	/* 137 _   */	LETTER,
	/* 140 `   */	0,
	/* 141 a   */	LETTER,
	/* 142 b   */	LETTER,
	/* 143 c   */	LETTER,
	/* 144 d   */	LETTER,
	/* 145 e   */	LETTER,
	/* 146 f   */	LETTER,
	/* 147 g   */	LETTER,
	/* 150 h   */	LETTER,
	/* 151 i   */	LETTER,
	/* 152 j   */	LETTER,
	/* 153 k   */	LETTER,
	/* 154 l   */	LETTER,
	/* 155 m   */	LETTER,
	/* 156 n   */	LETTER,
	/* 157 o   */	LETTER,
	/* 160 p   */	LETTER,
	/* 161 q   */	LETTER,
	/* 162 r   */	LETTER,
	/* 163 s   */	LETTER,
	/* 164 t   */	LETTER,
	/* 165 u   */	LETTER,
	/* 166 v   */	LETTER,
	/* 167 w   */	LETTER,
	/* 170 x   */	LETTER,
	/* 171 y   */	LETTER,
	/* 172 z   */	LETTER,
	/* 173 {   */	OTHER,
	/* 174 |   */	0,
	/* 175 }   */	OTHER,
	/* 175 ~   */	0,
};
static  char *escape(char *p ,unsigned char *rcp)
{
	switch (*rcp) {
	case 'n':		/*LF*/
		*p = 10;
		break;
	case 't':		/*HT*/
		*p = 9;
		break;
	case 'b':		/*BS*/
		*p = 8;
		break;
	case 'r':		/*CR*/
		*p = 13;
		break;
	case 'f':		/*FF*/
		*p = 12;
		break;
	case 'v':		/*VT*/
		*p = 11;
		break;
	case '\\':		/*'\'*/
		*p = 92;
		break;
	case '\'':
		*p = 39;
		break;
	case '\"':
		*p = 34;
		break;
	case 'a':		/*BEL*/
		*p = 7;
		break;
	case '?':
		*p = 63;
		break;
	default:
		*p = (char)*rcp;
		warning("unknown escape character: '\\%c'\n", *rcp);
		break;
	}
	return p + 1;
}
int gettoken(void)
{
	for(;;) {
		register unsigned char *rcp = cp;
		while(map[*rcp]&BLANK)
			rcp++;
		src.file = file;
		src.x = (char *)rcp - line + 1;
		src.y = lineno + 1;
		cp = rcp + 1;
		switch(*rcp++){
		case '/': if (*rcp == '*') {
				int c = 0;
			  	for (rcp++; *rcp != '/' || c != '*'; )
			  		if (map[*rcp]&NEWLINE) {
			  			if (rcp < limit)
			  				c = *rcp;
			  			cp = rcp + 1;
			  			nexline();
			  			rcp = cp;
			  			if (rcp == limit)
			  				break;
			  		} else
			  			c = *rcp++;
				if (rcp < limit)
					rcp++;
				else
					error("unclosed comment\n");
				cp = rcp;
				continue;
			}
			return '/';
		case '\n': case '\v':case '\r':case '\f':
			cp = rcp ;
			nexline();
			rcp = cp;
			if(rcp >= limit) {	
				tsym = NULL;
				return EOI;
			}
			continue;
		case '+':	case '-':	case '*':	case ';':
		case '%':	case ',':	case '(':	case ')':	
		case '{':	case '}':	case '[':	case ']':
		case '&':
			return rcp[-1];
		case '!':	return *rcp == '=' ? cp++,NEQ: '!';
		case '=':	return *rcp == '=' ? cp++,EQL: '=';
		case '<':	return *rcp == '=' ? cp++,LEQ: '<';
		case '>':	return *rcp == '=' ? cp++,GEQ: '>';
		case 'b': if(rcp[0] == 'r'
			&& rcp[1] == 'e'
			&& rcp[2] == 'a'
			&& rcp[3] == 'k'
			&& !(map[rcp[4]]&(DIGIT|LETTER))) {
				cp = rcp + 4;
				return  BREAK;
			}
			goto id;
		case 'c': if(rcp[0] == 'h'
			&& rcp[1] == 'a'
			&& rcp[2] == 'r'
			&& !(map[rcp[3]] &(DIGIT|LETTER))) {
				cp = rcp + 3;
				return CHAR;
			}
			if(rcp[0] == 'o'
			&& rcp[1] == 'n'
			&& rcp[2] == 't'
			&& rcp[3] == 'i'
			&& rcp[4] == 'n'
			&& rcp[5] == 'u'
			&& rcp[6] == 'e'
			&& !(map[rcp[7]] &(DIGIT|LETTER))) {
				cp = rcp + 7;
				return CONTINUE;
			}
			goto id;
		case 'e': if(rcp[0] == 'l'
			&& rcp[1] == 's'
			&& rcp[2] == 'e'
			&& !(map[rcp[3]] &(DIGIT|LETTER))) {
				cp = rcp + 3;
				return ELSE;
			}
			goto id;
		case 'i': if(rcp[0] == 'f'
			&& !(map[rcp[1]] &(DIGIT|LETTER))) {
				cp = rcp + 1;
				return IF;
			}
			if(rcp[0] == 'n'
			&& rcp[1] == 't'
			&& !(map[rcp[2]] &(DIGIT|LETTER))) {
				cp = rcp + 2;
				return INT;
			}
			goto id;
		case 'r': if(rcp[0] == 'e'
			&& rcp[1] == 't'
			&& rcp[2] == 'u'
			&& rcp[3] == 'r'
			&& rcp[4] == 'n'
			&& !(map[rcp[5]] &(DIGIT|LETTER))) {
				cp = rcp + 5;
				return RETURN;
			}
			goto id;
		case 'v': if(rcp[0] == 'o'
			&& rcp[1] == 'i'
			&& rcp[2] == 'd'
			&& !(map[rcp[3]] &(DIGIT|LETTER))) {
				cp = rcp + 3;
				return VOID;
			}
			goto id;
		case 'w': if(rcp[0] == 'h'
			&& rcp[1] == 'i'
			&& rcp[2] == 'l'
			&& rcp[3] == 'e'
			&& !(map[rcp[4]] &(DIGIT|LETTER))) {
				cp = rcp + 4;
				return WHILE;
			}
			goto id;
		case 's': if (rcp[0] == 'i'
			&& rcp[1] == 'z'
			&& rcp[2] == 'e'
			&& rcp[3] == 'o'
			&& rcp[4] == 'f'
			&& !(map[rcp[5]] &(DIGIT|LETTER))) {
				cp = rcp + 5;
				return SIZEOF;
			}
			goto id;
		case 'a':	case 'd':	case 'f':	case 'g':
		case 'h':	case 'j':	case 'k':	case 'l':
		case 'm':	case 'n':	case 'o':	case 'p':
		case 'q':	case 't':	case 'u':
		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':	case '_':
		id: {
			token = (char *)rcp - 1;
			while(map[*rcp]&(DIGIT | LETTER))
				rcp++;
			cp = rcp;
			token = string(token,(char *)rcp - token);
			return ID;
			}
		case '0':	case '1':	case '2':	case '3':
		case '4':	case '5':	case '6':	case '7':
		case '8':	case '9':
		{	int n = 0;
			rcp--;
			while(map[*rcp]&DIGIT){
				n = n*10+(*rcp-'0');
				rcp++;
			}
			cp = rcp;
			tsym = &tval;
			tsym->name = token = stringd(n);
			tsym->type = inttype;
			tsym->u.c.v.i = n;
			return ICON;
		}
		case '\'':{ char temp; /*字符常量*/
			token = &temp;
			if ('\\' == *rcp) 
				escape(token, ++rcp);
			else
				temp = (char)*rcp;
			tsym = &tval;
			tsym->name = token = string(&temp,1);
			tsym->type = chartype;
			tsym->u.c.v.c = temp;
			if(*++rcp != '\'')
				error("missing symbol '\n");
			else
				rcp++;
			cp = rcp;
		 }return CCON;
		case '"': {/*字符串常量*/
			char buffer[512];
			int n;
			token = buffer;
			for (; *rcp != '"' && token < buffer + sizeof buffer; rcp++) {
				if(*rcp == '\n'||*rcp == '\v'||*rcp == '\r'||*rcp == '\f') {
					if(rcp == limit) {
						error("missing symbol '\"'\n");
						break;
					}
					*token++ = *rcp; 
					lineno++;
				}else if ('\\' == *rcp) /*处理转义字符*/
					token = escape(token, ++rcp);
				else 
					*token++ = *rcp; 
			}
			n = token - buffer;
			token = string(buffer,n);
			tsym = &tval;
			tsym->u.c.v.p = token;
			tsym->type = array(chartype, n, 0);
			if(*rcp == '"') rcp++;
			cp = rcp;
		 }return SCON;
		default:
			if(map[rcp[-1]] == 0) {
				if(rcp[-1] < ' ' || rcp[-1] >= 0177)
					error("illegal character '\\0%d'\n",rcp[-1]);
				else
					error("illegal character '%c'\n", rcp[-1]);
			}	
		}
				
	}
}