// integer ::= digit+
inline Token
Lexer::integer()
{
assert(is_decimal_digit(peek()));
digit();
while (is_decimal_digit(peek()))
digit();
return on_integer();
}
/*----------------------------------------------------------------------*/
static void read_number(lex_state_t *l, token_t *token) {
int i;
double d;
int digit;
int after_dot;
int after_dot_divisor;
/* read the left part up to the dot */
i = 0;
while (is_decimal_digit(*l->ptr)) {
digit = (*l->ptr - '0');
i *= 10;
i += digit;
++l->ptr;
}
if (*l->ptr == '.') {
/* read the right part after the dot */
d = i;
++l->ptr;
after_dot = 0;
after_dot_divisor = 1;
while (is_decimal_digit(*l->ptr)) {
digit = (*l->ptr - '0');
after_dot *= 10;
after_dot += digit;
after_dot_divisor *= 10;
++l->ptr;
}
/* read suffix */
if (*l->ptr == 'f') {
++l->ptr;
token->type = TK_FLOAT_LITERAL;
token->u.f = (float)i + (float)after_dot / (float)after_dot_divisor;
} else {
token->type = TK_DOUBLE_LITERAL;
token->u.d = (double)i + (double)after_dot / (double)after_dot_divisor;
}
} else {
token->type = TK_INT_LITERAL;
token->u.i = i;
}
}
// digit ::= [0-9]
inline void
Lexer::digit()
{
assert(is_decimal_digit(peek()));
get();
}
/*----------------------------------------------------------------------*/
static int is_hexadecimal_digit(char c) {
return is_decimal_digit(c) ||
(c >= 'a' && c <= 'f') ||
(c >= 'A' && c <= 'F');
}
/*----------------------------------------------------------------------*/
static int next_token(lex_state_t *l) {
token_t *token;
char c;
char cn;
token = &l->out->tokens[l->out->token_count];
skip_comments_and_whitespace(l);
token->source_pos = l->ptr - l->in->source;
c = *(l->ptr + 0);
cn = *(l->ptr + 1);
if (c == '\0') {
return 0;
} else if (c == '\"') {
read_string(l, token);
} else if (is_decimal_digit(c)) {
read_number(l, token);
} else if (is_identifier_char(c)) {
read_identifier(l, token);
match_keyword(token);
} else if (c == '<' && cn == '=') {
token->type = TK_LESS_EQUAL;
l->ptr += 2;
} else if (c == '=' && cn == '=') {
token->type = TK_EQUALS_EQUALS;
l->ptr += 2;
} else if (c == '>' && cn == '=') {
token->type = TK_GREATER_EQUAL;
l->ptr += 2;
} else if (c == '+') {
token->type = TK_PLUS;
++l->ptr;
} else if (c == '-') {
token->type = TK_MINUS;
++l->ptr;
} else if (c == '*') {
token->type = TK_STAR;
++l->ptr;
} else if (c == '/') {
token->type = TK_SLASH;
++l->ptr;
} else if (c == '=') {
token->type = TK_EQUALS;
++l->ptr;
} else if (c == '<') {
token->type = TK_LESS;
++l->ptr;
} else if (c == '>') {
token->type = TK_GREATER;
++l->ptr;
} else if (c == '(') {
token->type = TK_LBRACKET;
++l->ptr;
} else if (c == ')') {
token->type = TK_RBRACKET;
++l->ptr;
} else if (c == ',') {
token->type = TK_COMMA;
++l->ptr;
} else if (c == ':') {
token->type = TK_COLON;
++l->ptr;
} else if (c == ';') {
token->type = TK_SEMICOLON;
++l->ptr;
} else {
error(l);
}
return 1;
}
// Returns true if c can appear in the rest of an identifier.
inline bool
is_identifier_rest(char c)
{
return is_identifier_start(c) || is_decimal_digit(c);
}
