void Scanner::skip_non_token() {
while (1) {
if ((is_space(cur_char()))) {
go_ahead();
} else if (is_comment_start()) {
skip_comment();
} else {
return ;
}
}
}
void Scanner::skip_comment() {
if (!is_comment_start()) {
return ;
}
char c = next_char();
if (c == '/') {
while (cur_char() != '\n' && cur_char() != 0) {
go_ahead();
}
go_ahead();
} else {
go_ahead();
go_ahead();
while (cur_char() != '*' || next_char() != '/') {
go_ahead();
}
go_ahead();
go_ahead();
}
}
void fill_buffer(void)
{
char * p = NULL;
BOOLEAN finished_a_line = false;
/* indent() may be saving the text between "if (...)" and the following
* statement. To do so, it uses another buffer (`save_com'). Switch
* back to the previous buffer here. */
if (bp_save != 0)
{
buf_ptr = bp_save;
buf_end = be_save;
bp_save = be_save = 0;
/* only return if there is really something in this buffer */
if (buf_ptr < buf_end)
{
return;
}
}
if (*in_prog_pos == EOS)
{
cur_line = buf_ptr = in_prog_pos;
had_eof = true;
}
else
{
/* Here if we know there are chars to read. The file is
* NULL-terminated, so we can always look one character ahead
* safely. */
p = cur_line = in_prog_pos;
finished_a_line = false;
do
{
p = skip_horiz_space(p);
/* If we are looking at the beginning of a comment, see
* if it turns off formatting with off-on directives. */
if (is_comment_start(p))
{
p += 2;
p = skip_horiz_space(p);
/* Skip all lines between the indent off and on directives. */
if (strncmp (p, "*INDENT-OFF*", 12) == 0)
{
inhibit_indenting(true);
}
}
while ((*p != EOS) && *p != EOL)
{
p++;
}
/* Here for newline -- finish up unless formatting is off */
if (*p == EOL)
{
finished_a_line = true;
in_prog_pos = p + 1;
}
/* Here for embedded NULLs */
else if ((unsigned int) (p - current_input->data) < current_input->size)
{
WARNING (_("Warning: File %s contains NULL-characters\n"), current_input->name, 0);
p++;
}
/* Here for EOF with no terminating newline char. */
else
{
in_prog_pos = p;
finished_a_line = true;
}
} while (!finished_a_line);
buf_ptr = cur_line;
buf_end = in_prog_pos;
}
}
char *get_token(char *lexeme , int mode){
char *token=(char*)calloc(strlen(lexeme)+50,sizeof(char));
//printf("Getting token\n");
if(is_long(lexeme)){
sprintf(token,"%d",LONG);
}
else if(is_static(lexeme)){
sprintf(token,"%d",STATIC);
}
else if(is_union(lexeme)){
sprintf(token,"%d",UNION);
}
else if(is_default(lexeme)){
sprintf(token,"%d",DEFAULT);
}
else if(is_break(lexeme)){
sprintf(token,"%d",BREAK);
}
else if(is_case(lexeme)){
sprintf(token,"%d",CASE);
}
else if(is_continue(lexeme)){
sprintf(token,"%d",CONTINUE);
}
else if(is_goto(lexeme)){
sprintf(token,"%d",GOTO);
}
else if(is_struct(lexeme)){
sprintf(token,"%d",STRUCT);
}
else if(is_const(lexeme)){
sprintf(token,"%d",CONST);
}
else if(is_void(lexeme)){
sprintf(token,"%d",VOID);
}
else if(is_switch(lexeme)){
sprintf(token,"%d",SWITCH);
}
else if(is_for(lexeme)){
sprintf(token,"%d",FOR);
}
else if(is_while(lexeme)){
sprintf(token,"%d",WHILE);
}
else if(is_do(lexeme)){
sprintf(token,"%d",DO);
}
else if(is_return(lexeme)){
sprintf(token,"%d",RETURN);
}
else if(is_bool(lexeme)){
sprintf(token,"%d",BOOL);
}
else if(is_char(lexeme)){
sprintf(token,"%d",CHAR);
}
else if(is_signed(lexeme)){
sprintf(token,"%d",SIGNED);
}
else if(is_unsigned(lexeme)){
sprintf(token,"%d",UNSIGNED);
}
else if(is_short(lexeme)){
sprintf(token,"%d",SHORT);
}
else if(is_int(lexeme)){
sprintf(token,"%d",INT);
}
else if(is_float(lexeme)){
sprintf(token,"%d",FLOAT);
}
else if(is_double(lexeme)){
sprintf(token,"%d",DOUBLE);
}
else if(is_l_square(lexeme)){
sprintf(token,"%d",L_SQUARE);
}
else if(is_r_square(lexeme)){
sprintf(token,"%d",R_SQUARE);
}
else if(is_l_paraen(lexeme)){
sprintf(token,"%d",L_PARAEN);
}
else if(is_r_paraen(lexeme)){
sprintf(token,"%d",R_PARAEN);
}
else if(is_l_cbrace(lexeme)){
sprintf(token,"%d",L_CBRACE);
}
else if(is_r_cbrace(lexeme)){
sprintf(token,"%d",R_CBRACE);
}
else if(is_comma(lexeme)){
sprintf(token,"%d",COMMA);
}
else if(is_semicol(lexeme)){
sprintf(token,"%d",SEMICOL);
}
else if(is_eq_eq(lexeme)){
sprintf(token,"%d",EQ_EQ);
}
else if(is_lesser(lexeme)){
sprintf(token,"%d",LESSER);
}
else if(is_less_eq(lexeme)){
sprintf(token,"%d",LESS_EQ);
}
else if(is_div(lexeme)){
sprintf(token,"%d",DIV);
}
else if(is_greater(lexeme)){
sprintf(token,"%d",GREATER);
}
else if(is_great_eq(lexeme)){
sprintf(token,"%d",GREAT_EQ);
}
else if(is_plus_eq(lexeme)){
sprintf(token,"%d",PLUS_EQ);
}
else if(is_minus_eq(lexeme)){
sprintf(token,"%d",MINUS_EQ);
}
else if(is_div_eq(lexeme)){
sprintf(token,"%d",DIV_EQ);
}
else if(is_mult_eq(lexeme)){
sprintf(token,"%d",MULT_EQ);
}
else if(is_minus_minus(lexeme)){
sprintf(token,"%d",MINUS_MINUS);
}
else if(is_plus_plus(lexeme)){
sprintf(token,"%d",PLUS_PLUS);
}
else if(is_percent(lexeme)){
sprintf(token,"%d",PERCENT);
}
else if(is_div(lexeme)){
sprintf(token,"%d",DIV);
}
else if(is_mult(lexeme)){
sprintf(token,"%d",MULT);
}
else if(is_minus(lexeme)){
sprintf(token,"%d",MINUS);
}
else if(is_plus(lexeme)){
sprintf(token,"%d",PLUS);
}
else if(is_int_const(lexeme)){
printf("int");
sprintf(token,"%d\t%s",INT_CONST,lexeme);
}
else if(is_flo_const(lexeme)){
printf("float");
sprintf(token,"%d\t%s",FLO_CONST,lexeme);
}
else if(is_comment_start(lexeme)){
sprintf(token,"$start");
}
else if(is_comment_end(lexeme)){
sprintf(token,"$end");
}
else if(is_identifier(lexeme)){
printf("Identifier");
if(mode==1) ht_set( symboltable, lexeme, "1");
sprintf(token,"%d\t%s",IDNTIFIER,lexeme);
}
else sprintf(token,"%d",NOTOK);
return token;
}
