int gettok(char **pbuf, int *psz) /* get next input token */
{
int c;
char *buf = *pbuf;
int sz = *psz;
char *bp = buf;
c = input();
if (c == 0)
return 0;
buf[0] = c;
buf[1] = 0;
if (!isalnum(c) && c != '.' && c != '_')
return c;
*bp++ = c;
if (isalpha(c) || c == '_') { /* it's a varname */
for ( ; (c = input()) != 0; ) {
if (bp-buf >= sz)
if (!adjbuf(&buf, &sz, bp-buf+2, 100, &bp, 0))
FATAL( "out of space for name %.10s...", buf );
if (isalnum(c) || c == '_')
*bp++ = c;
else {
*bp = 0;
unput(c);
break;
}
}
*bp = 0;
} else { /* it's a number */
char *rem;
/* read input until can't be a number */
for ( ; (c = input()) != 0; ) {
if (bp-buf >= sz)
if (!adjbuf(&buf, &sz, bp-buf+2, 100, &bp, 0))
FATAL( "out of space for number %.10s...", buf );
if (isdigit(c) || c == 'e' || c == 'E'
|| c == '.' || c == '+' || c == '-')
*bp++ = c;
else {
unput(c);
break;
}
}
*bp = 0;
strtod(buf, &rem); /* parse the number */
unputstr(rem); /* put rest back for later */
rem[0] = 0;
}
*pbuf = buf;
*psz = sz;
return buf[0];
}
int yylex(void)
{
int c;
static char *buf = 0;
static int bufsize = 500;
if (buf == 0 && (buf = (char *) malloc(bufsize)) == NULL)
FATAL( "out of space in yylex" );
if (sc) {
sc = 0;
RET('}');
}
if (reg) {
reg = 0;
return regexpr();
}
/* printf("top\n"); */
for (;;) {
c = gettok(&buf, &bufsize);
/* printf("gettok [%s]\n", buf); */
if (c == 0)
return 0;
if (isalpha(c) || c == '_')
return word(buf);
if (isdigit(c)) {
yylval.cp = setsymtab(buf, tostring(buf), atof(buf), CON|NUM, symtab);
/* should this also have STR set? */
RET(NUMBER);
}
yylval.i = c;
switch (c) {
case '\n': /* {EOL} */
RET(NL);
case '\r': /* assume \n is coming */
case ' ': /* {WS}+ */
case '\t':
break;
case '#': /* #.* strip comments */
while ((c = input()) != '\n' && c != 0)
;
unput(c);
break;
case ';':
RET(';');
case '\\':
if (peek() == '\n') {
input();
} else if (peek() == '\r') {
input(); input(); /* \n */
lineno++;
} else {
RET(c);
}
break;
case '&':
if (peek() == '&') {
input(); RET(AND);
} else
RET('&');
case '|':
if (peek() == '|') {
input(); RET(BOR);
} else
RET('|');
case '!':
if (peek() == '=') {
input(); yylval.i = NE; RET(NE);
} else if (peek() == '~') {
input(); yylval.i = NOTMATCH; RET(MATCHOP);
} else
RET(NOT);
case '~':
yylval.i = MATCH;
RET(MATCHOP);
case '<':
if (peek() == '=') {
input(); yylval.i = LE; RET(LE);
} else {
yylval.i = LT; RET(LT);
}
case '=':
if (peek() == '=') {
input(); yylval.i = EQ; RET(EQ);
} else {
yylval.i = ASSIGN; RET(ASGNOP);
}
case '>':
if (peek() == '=') {
input(); yylval.i = GE; RET(GE);
} else if (peek() == '>') {
input(); yylval.i = APPEND; RET(APPEND);
} else {
yylval.i = GT; RET(GT);
}
case '+':
if (peek() == '+') {
input(); yylval.i = INCR; RET(INCR);
} else if (peek() == '=') {
input(); yylval.i = ADDEQ; RET(ASGNOP);
} else
RET('+');
case '-':
if (peek() == '-') {
input(); yylval.i = DECR; RET(DECR);
} else if (peek() == '=') {
input(); yylval.i = SUBEQ; RET(ASGNOP);
} else
RET('-');
case '*':
if (peek() == '=') { /* *= */
input(); yylval.i = MULTEQ; RET(ASGNOP);
} else if (peek() == '*') { /* ** or **= */
input(); /* eat 2nd * */
if (peek() == '=') {
input(); yylval.i = POWEQ; RET(ASGNOP);
} else {
RET(POWER);
}
} else
RET('*');
case '/':
RET('/');
case '%':
if (peek() == '=') {
input(); yylval.i = MODEQ; RET(ASGNOP);
} else
RET('%');
case '^':
if (peek() == '=') {
input(); yylval.i = POWEQ; RET(ASGNOP);
} else
RET(POWER);
case '$':
/* BUG: awkward, if not wrong */
c = gettok(&buf, &bufsize);
if (isalpha(c)) {
if (strcmp(buf, "NF") == 0) { /* very special */
unputstr("(NF)");
RET(INDIRECT);
}
c = peek();
if (c == '(' || c == '[' || (infunc && isarg(buf) >= 0)) {
unputstr(buf);
RET(INDIRECT);
}
yylval.cp = setsymtab(buf, "", 0.0, STR|NUM, symtab);
RET(IVAR);
} else if (c == 0) { /* */
SYNTAX( "unexpected end of input after $" );
RET(';');
} else {
unputstr(buf);
RET(INDIRECT);
}
case '}':
if (--bracecnt < 0)
SYNTAX( "extra }" );
sc = 1;
RET(';');
case ']':
if (--brackcnt < 0)
SYNTAX( "extra ]" );
RET(']');
case ')':
if (--parencnt < 0)
SYNTAX( "extra )" );
RET(')');
case '{':
bracecnt++;
RET('{');
case '[':
brackcnt++;
RET('[');
case '(':
parencnt++;
RET('(');
case '"':
return string(); /* BUG: should be like tran.c ? */
default:
RET(c);
}
}
}
int gettok(char **pbuf, int *psz) /* get next input token */
{
int c, retc;
char *buf = *pbuf;
int sz = *psz;
char *bp = buf;
c = input();
if (c == 0)
return 0;
buf[0] = c;
buf[1] = 0;
if (!isalnum(c) && c != '.' && c != '_')
return c;
*bp++ = c;
if (isalpha(c) || c == '_') { /* it's a varname */
for ( ; (c = input()) != 0; ) {
if (bp-buf >= sz)
if (!adjbuf(&buf, &sz, bp-buf+2, 100, &bp, 0))
FATAL( "out of space for name %.10s...", buf );
if (isalnum(c) || c == '_')
*bp++ = c;
else {
*bp = 0;
unput(c);
break;
}
}
*bp = 0;
retc = 'a'; /* alphanumeric */
} else { /* maybe it's a number, but could be . */
char *rem;
/* read input until can't be a number */
for ( ; (c = input()) != 0; ) {
if (bp-buf >= sz)
if (!adjbuf(&buf, &sz, bp-buf+2, 100, &bp, 0))
FATAL( "out of space for number %.10s...", buf );
if (isdigit(c) || c == 'e' || c == 'E'
|| c == '.' || c == '+' || c == '-')
*bp++ = c;
else {
unput(c);
break;
}
}
*bp = 0;
strtod(buf, &rem); /* parse the number */
if (rem == buf) { /* it wasn't a valid number at all */
buf[1] = 0; /* return one character as token */
retc = buf[0]; /* character is its own type */
unputstr(rem+1); /* put rest back for later */
} else { /* some prefix was a number */
unputstr(rem); /* put rest back for later */
rem[0] = 0; /* truncate buf after number part */
retc = '0'; /* type is number */
}
}
*pbuf = buf;
*psz = sz;
return retc;
}
