static int charliteral(int c)
{
if(!bCanInsertSymbol)
{
//maks: solve the error when colorize "oi/oi"
WASLITERAL = 0;
return c;
}
if (c == '\\') {
switch ((c = EiC_nextchar())) {
case 'n': c = '\n'; break; /* newline */
case 't': c = '\t'; break; /* tabspace */
case 'v': c = '\v'; break; /* vertical tab */
case 'b': c = '\b'; break; /* backspace */
case 'r': c = '\r'; break; /* carriage return */
case 'f': c = '\f'; break; /* formfeed */
case 'a': c = '\a'; break; /* bell */
case '\\': c = '\\'; break; /* backslash */
case '\'': c = '\''; break; /* single quote */
case '"': c = '\"'; break; /* double quote */
case '?': c = '\?'; break; /* question mark */
case 'x': /* string of hex characters */
case 'X':{
int i, val = 0;
while ((i = gethex((c = EiC_nextchar()))) > -1) {
val = val * 16 + i;
}
retract(c);
if (val > 255)
EiC_error("Illegal character hex value");
c = val;
}
break;
default:
if (getoct(c) > -1) { /* octal characters */
int i, val = 0;
while ((i = getoct(c)) > -1) {
val = val * 8 + i;
c = EiC_nextchar();
}
retract(c);
if (val > 255)
EiC_error("Illegal character octal value");
c = val;
} else
EiC_error("Illegal character escape sequence `\\%c'", c);
break;
}
WASLITERAL = 1;
} else
WASLITERAL = 0;
return ((signed char )c);
}
static char *getNextInput()
{
char c, *p;
static char buf[256];
while(isspace(c = EiC_nextchar()));
p = buf;
while( c && c != DONE && !isspace(c)) {
*p++ = c;
c = EiC_nextchar();
}
*p = '\0';
return buf;
}
static void EiC_stringliteral(void)
{
unsigned size, lastsize = 0, c;
char *p=NULL;
lex_str_start_pos = lex_lastpos = lex_curpos; //maks
do {
for (size = 0; ((c = charliteral(EiC_nextchar())) != '\0' || WASLITERAL) &&
!(c == '"' && !WASLITERAL) && size < BSIZE; size++)
EiC_LEXEM[size] = c;
if (lastsize)
p = (char *) xrealloc(p, lastsize + size + 1);
else
p = (char *) xcalloc(size + 1, sizeof(char));
memcpy(&p[lastsize], EiC_LEXEM, size);
lastsize += size;
if(((c != '"') || (c == '"' && WASLITERAL)) && size == BSIZE) {
p[lastsize++] = c;
continue;
}
if (c != '"')
EiC_error("String literal error");
do {
c = EiC_nextchar();
if (c == '\n')
lex_lastpos++, lex_lineno++;
} while (WHITE(c) || c == '\n');
lex_lastpos = lex_curpos;
if (!c)
do
c = EiC_nextchar();
while (WHITE(c));
} while (c == '"' || size == BSIZE);
retract(c);
p[lastsize] = '\0';
token->Val.p.sp = token->Val.p.p = p;
token->Val.p.ep = p + lastsize + 1;
}
static void checkExt(int c)
{ /* check for unsigned and long suffix */
Lseen = Useen = Fseen = 0;
if (c == 'f' || c == 'F')
Fseen = 1;
else if (c == 'u' || c == 'U') {
Useen = 1;
if ((c = EiC_nextchar()) == 'l' || c == 'L')
Lseen = 1;
else
retract(c);
} else if (c == 'l' || c == 'L')
Lseen = 1;
else
retract(c);
}
extern int EiC_lexan(void)
{
int t=0, loop; char c=0, EiC_nextchar();
#ifdef ILOOKAHEAD
token = &EiC_TokenArray[EiC_TokenP];
if(EiC_TokenR > 0) {
EiC_TokenR--;
EiC_TokenI++;
EiC_TokenP=(EiC_TokenP+1)%MAX_TOKENS;
return token->Tok;
}
#else
if (STOKEN != NOTOKEN) {
STOKEN = NOTOKEN;
return token->Tok;
}
#endif
loop = 1;
state = 0;
while (loop) {
switch (state) {
case 0: lex_lastpos = lex_curpos; c = EiC_nextchar();
state = (WHITE(c) ? 0 :
(c == '\n' ? lex_lineno++, 0 :
(c == '<' ? t = LT, 1 :
(c == '>' ? t = GT, 2 :
(c == '+' ? t = '+', 3 :
(c == '-' ? t = '-', 4 :
(c == '|' ? t = BOR, 5 :
(c == '&' ? t = AND, 6 :
(c == '\''? 7 :
(c == '"' ? 8 :
(c == '.' ? 9 :
(c == '/' ? t = '/', c = EiC_nextchar(), 50 :
(c == '%' ? t = '%', c = EiC_nextchar(), 50 :
(c == '*' ? t = '*', c = EiC_nextchar(), 50 :
(c == '=' ? t = ASS, c = EiC_nextchar(), 50 :
(c == '!' ? t = NOT, c = EiC_nextchar(), 50 :
(c == '^' ? t = XOR, c = EiC_nextchar(), 50 :
//(c == '~' ? t = NOTB, c = EiC_nextchar(), 50 : //maks: ~
fail(RELOP, c))))))))))))))))));
break;
case 1: /* get <, <= and << */
if ((c = EiC_nextchar()) == '<') t = LSHT;
else state = 50;
break;
case 2: /* get >, >= and >> */
if ((c = EiC_nextchar()) == '>') t = RSHT;
else state = 50;
break;
case 3: c = EiC_nextchar(); /* get +, += or ++ */
if (c == '+') t = INC, state = 60;
else state = 50;
break;
case 4: c = EiC_nextchar(); /* get -, -= -- */
state = 60;
if (c == '-') t = DEC;
else if (c == '>') t = RARROW;
else state = 50;
break;
case 5: c = EiC_nextchar(); /* get |, |= or || */
if (c == '|') t = LOR, state = 60;
else state = 50;
break;
case 6: c = EiC_nextchar(); /* get &, &= or && */
if (c == '&') t = LAND, state = 60;
else state = 50;
break;
case 7:token->Val.ival = charliteral(EiC_nextchar()); /* char_constants */
t = TOKEN_CHAR;
if (EiC_nextchar() != '\'')
EiC_error("Missing single quote '");
state = 60;
break;
case 8: EiC_stringliteral(); /* string literals */
token->Tok = STR;
/*return STR;*/ loop = 0; break;
case 9: c = EiC_nextchar();
t = '.';
if(DIGIT(c))
state = 22;
else
state = 60;
retract(c);
break;
case 10: c = EiC_nextchar(); /* identifiers and keywords */
state = (LETTER(c) ? 11 :
(c == '_' ? 11 : fail(ID, c)));
break;
case 11: c = EiC_nextchar();
state = (LETTER(c) ? 11 :
(DIGIT(c) ? 11 :
(c == '_' ? 11 : 12)));
break;
case 12: retract(c); success(ID); /*return (token->Tok);*/ loop = 0; break;
case 20: c = EiC_nextchar(); /* integers and reals */
state = (c == '0' ? 30 :
(DIGIT(c) ? 21 : fail(TOKEN_INT, c)));
break;
case 21: c = EiC_nextchar();
state = (DIGIT(c) ? 21 :
(c == '.' ? 22 :
(c == 'e' ? 23 :
(c == 'E' ? 23 : 25))));
break;
case 22: c = EiC_nextchar();
state = (DIGIT(c) ? 22 :
(c == 'e' ? 23 :
(c == 'E' ? 23 : 26)));
break;
case 23: c = EiC_nextchar();
state = (c == '+' ? 24 :
(c == '-' ? 24 :
(DIGIT(c) ? 24 : fail(TOKEN_FLOAT, c) /* ??? */ )));
break;
case 24: c = EiC_nextchar();
state = (DIGIT(c) ? 24 : 26);
break;
case 25: checkExt(c); success(TOKEN_INT); /*return (token->Tok);*/ loop = 0; break;
case 26: checkExt(c); success(TOKEN_FLOAT); /*return (token->Tok);*/ loop = 0; break;
case 27: checkExt(c); success(HEX); /*return (token->Tok);*/ loop = 0; break;
case 28: checkExt(c); success(OCTAL); /*return (token->Tok);*/ loop = 0; break;
case 30: /* check for octal and hex numbers */
if ((c = EiC_nextchar()) == 'x' || c == 'X') {
while (gethex((c = EiC_nextchar())) > -1);
state = 27;
break;
}
if (c != '.' && c != 'e' && c != 'E') {
while (getoct(c) > -1)
c = EiC_nextchar();
state = 28;
break;
}
retract(c); state = 21; break;
case 50: /* mix with equal's */
if (c == '=')
switch (t) {
case '+': t = ADDEQ; break; /* += */
case '-': t = SUBEQ; break; /* -= */
case '/': t = DIVEQ; break; /* /= */
case '*': t = MULEQ; break; /* *= */
case '%': t = MODEQ; break; /* %= */
case ASS: t = EQ; break; /* == */
case GT: t = GE; break; /* >= */
case LT: t = LE; break; /* <= */
case NOT: t = NE; break; /* != */
case RSHT:t = RSHTEQ; break; /* >>= */
case LSHT:t = LSHTEQ; break; /* <<= */
case AND: t = ANDEQ; break; /* &= */
case BOR: t = BOREQ; break; /* |= */
case XOR: t = XOREQ; break; /* ^= */
//case NOTB: t = NOTBEQ; break; /* maks ~= */
default: retract(c);
}
else if(c == '/' && t == '/') //maks
{
//C++ comment
//Only for colorize
//Comments are removed by preprocessor before parser
do
{
c = EiC_nextchar();
} while(c && c != '\n');
retract(c);
success(MISC);
token->Tok = TOKEN_COMMENT;
loop = 0;
break;
}
else retract(c);
state = 60;
break;
case 60: success(MISC); token->Tok = t; /*return (token->Tok);*/ loop = 0; break;
case 100: token->Tok = EiC_nextchar(); /*return (token->Tok);*/ loop = 0; break;
}
}
#ifdef ILOOKAHEAD
if(EiC_TokenI<MAX_TOKENS)
EiC_TokenI++;
EiC_TokenP = (EiC_TokenP +1)%MAX_TOKENS;
#endif
return token->Tok;
}
