static int atoh(unsigned int *result,
const char * const s, const size_t length)
{
size_t i = 0;
unsigned int val = 0;
while ((i < length) && csxdigit(s[i])) {
if (csdigit(s[i]))
val = val*0x10 + (s[i]-'0');
else if (csupper(s[i]))
val = val*0x10 + (s[i]-'A') + 10;
else
val = val*0x10 + (s[i]-'a') + 10;
i++;
}
if (i != length)
return 0;
*result = val;
return 1;
}
const char *check_unicode_name(const char *u)
{
if (*u != 'u')
return 0;
const char *p = ++u;
for (;;) {
int val = 0;
const char *start = p;
for (;;) {
// only uppercase hex digits allowed
if (!csxdigit(*p))
return 0;
if (csdigit(*p))
val = val*0x10 + (*p-'0');
else if (csupper(*p))
val = val*0x10 + (*p-'A'+10);
else
return 0;
// biggest Unicode value is U+10FFFF
if (val > 0x10FFFF)
return 0;
p++;
if (*p == '\0' || *p == '_')
break;
}
// surrogates not allowed
if ((val >= 0xD800 && val <= 0xDBFF) || (val >= 0xDC00 && val <= 0xDFFF))
return 0;
if (val > 0xFFFF) {
if (*start == '0') // no leading zeros allowed if > 0xFFFF
return 0;
}
else if (p - start != 4) // otherwise, check for exactly 4 hex digits
return 0;
if (*p == '\0')
break;
p++;
}
return u;
}
int get_token(int lookup_flag)
{
context_buffer.clear();
for (;;) {
int n = 0;
int bol = input_stack::bol();
int c = input_stack::get_char();
if (bol && c == command_char) {
token_buffer.clear();
token_buffer += c;
// the newline is not part of the token
for (;;) {
c = input_stack::peek_char();
if (c == EOF || c == '\n')
break;
input_stack::get_char();
token_buffer += char(c);
}
context_buffer = token_buffer;
return COMMAND_LINE;
}
switch (c) {
case EOF:
return EOF;
case ' ':
case '\t':
break;
case '\\':
{
int d = input_stack::peek_char();
if (d != '\n') {
context_buffer = '\\';
return '\\';
}
input_stack::get_char();
break;
}
case '#':
do {
c = input_stack::get_char();
} while (c != '\n' && c != EOF);
if (c == '\n')
context_buffer = '\n';
return c;
case '"':
context_buffer = '"';
token_buffer.clear();
for (;;) {
c = input_stack::get_char();
if (c == '\\') {
context_buffer += '\\';
c = input_stack::peek_char();
if (c == '"') {
input_stack::get_char();
token_buffer += '"';
context_buffer += '"';
}
else
token_buffer += '\\';
}
else if (c == '\n') {
error("newline in string");
break;
}
else if (c == EOF) {
error("missing `\"'");
break;
}
else if (c == '"') {
context_buffer += '"';
break;
}
else {
context_buffer += char(c);
token_buffer += char(c);
}
}
return TEXT;
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
{
int overflow = 0;
n = 0;
for (;;) {
if (n > (INT_MAX - 9)/10) {
overflow = 1;
break;
}
n *= 10;
n += c - '0';
context_buffer += char(c);
c = input_stack::peek_char();
if (c == EOF || !csdigit(c))
break;
c = input_stack::get_char();
}
token_double = n;
if (overflow) {
for (;;) {
token_double *= 10.0;
token_double += c - '0';
context_buffer += char(c);
c = input_stack::peek_char();
if (c == EOF || !csdigit(c))
break;
c = input_stack::get_char();
}
// if somebody asks for 1000000000000th, we will silently
// give them INT_MAXth
double temp = token_double; // work around gas 1.34/sparc bug
if (token_double > INT_MAX)
n = INT_MAX;
else
n = int(temp);
}
}
switch (c) {
case 'i':
case 'I':
context_buffer += char(c);
input_stack::get_char();
return NUMBER;
case '.':
{
context_buffer += '.';
input_stack::get_char();
got_dot:
double factor = 1.0;
for (;;) {
c = input_stack::peek_char();
if (c == EOF || !csdigit(c))
break;
input_stack::get_char();
context_buffer += char(c);
factor /= 10.0;
if (c != '0')
token_double += factor*(c - '0');
}
if (c != 'e' && c != 'E') {
if (c == 'i' || c == 'I') {
context_buffer += char(c);
input_stack::get_char();
}
return NUMBER;
}
}
// fall through
case 'e':
case 'E':
{
int echar = c;
input_stack::get_char();
c = input_stack::peek_char();
int sign = '+';
if (c == '+' || c == '-') {
sign = c;
input_stack::get_char();
c = input_stack::peek_char();
if (c == EOF || !csdigit(c)) {
input_stack::push_back(sign);
input_stack::push_back(echar);
return NUMBER;
}
context_buffer += char(echar);
context_buffer += char(sign);
}
else {
if (c == EOF || !csdigit(c)) {
input_stack::push_back(echar);
return NUMBER;
}
context_buffer += char(echar);
}
input_stack::get_char();
context_buffer += char(c);
n = c - '0';
for (;;) {
c = input_stack::peek_char();
if (c == EOF || !csdigit(c))
break;
input_stack::get_char();
context_buffer += char(c);
n = n*10 + (c - '0');
}
if (sign == '-')
n = -n;
if (c == 'i' || c == 'I') {
context_buffer += char(c);
input_stack::get_char();
}
token_double *= pow(10.0, n);
return NUMBER;
}
case 'n':
input_stack::get_char();
c = input_stack::peek_char();
if (c == 'd') {
input_stack::get_char();
token_int = n;
context_buffer += "nd";
return ORDINAL;
}
input_stack::push_back('n');
return NUMBER;
case 'r':
input_stack::get_char();
c = input_stack::peek_char();
if (c == 'd') {
input_stack::get_char();
token_int = n;
context_buffer += "rd";
return ORDINAL;
}
input_stack::push_back('r');
return NUMBER;
case 't':
input_stack::get_char();
c = input_stack::peek_char();
if (c == 'h') {
input_stack::get_char();
token_int = n;
context_buffer += "th";
return ORDINAL;
}
input_stack::push_back('t');
return NUMBER;
case 's':
input_stack::get_char();
c = input_stack::peek_char();
if (c == 't') {
input_stack::get_char();
token_int = n;
context_buffer += "st";
return ORDINAL;
}
input_stack::push_back('s');
return NUMBER;
default:
return NUMBER;
}
break;
case '\'':
{
c = input_stack::peek_char();
if (c == 't') {
input_stack::get_char();
c = input_stack::peek_char();
if (c == 'h') {
input_stack::get_char();
context_buffer = "'th";
return TH;
}
else
input_stack::push_back('t');
}
context_buffer = "'";
return '\'';
}
case '.':
{
c = input_stack::peek_char();
if (c != EOF && csdigit(c)) {
n = 0;
token_double = 0.0;
context_buffer = '.';
goto got_dot;
}
return get_token_after_dot(c);
}
case '<':
c = input_stack::peek_char();
if (c == '-') {
input_stack::get_char();
c = input_stack::peek_char();
if (c == '>') {
input_stack::get_char();
context_buffer = "<->";
return DOUBLE_ARROW_HEAD;
}
context_buffer = "<-";
return LEFT_ARROW_HEAD;
}
else if (c == '=') {
input_stack::get_char();
context_buffer = "<=";
return LESSEQUAL;
}
context_buffer = "<";
return '<';
case '-':
c = input_stack::peek_char();
if (c == '>') {
input_stack::get_char();
context_buffer = "->";
return RIGHT_ARROW_HEAD;
}
context_buffer = "-";
return '-';
case '!':
c = input_stack::peek_char();
if (c == '=') {
input_stack::get_char();
context_buffer = "!=";
return NOTEQUAL;
}
context_buffer = "!";
return '!';
case '>':
c = input_stack::peek_char();
if (c == '=') {
input_stack::get_char();
context_buffer = ">=";
return GREATEREQUAL;
}
context_buffer = ">";
return '>';
case '=':
c = input_stack::peek_char();
if (c == '=') {
input_stack::get_char();
context_buffer = "==";
return EQUALEQUAL;
}
context_buffer = "=";
return '=';
case '&':
c = input_stack::peek_char();
if (c == '&') {
input_stack::get_char();
context_buffer = "&&";
return ANDAND;
}
context_buffer = "&";
return '&';
case '|':
c = input_stack::peek_char();
if (c == '|') {
input_stack::get_char();
context_buffer = "||";
return OROR;
}
context_buffer = "|";
return '|';
default:
if (c != EOF && csalpha(c)) {
token_buffer.clear();
token_buffer = c;
for (;;) {
c = input_stack::peek_char();
if (c == EOF || (!csalnum(c) && c != '_'))
break;
input_stack::get_char();
token_buffer += char(c);
}
int tok = lookup_keyword(token_buffer.contents(),
token_buffer.length());
if (tok != 0) {
context_buffer = token_buffer;
return tok;
}
char *def = 0;
if (lookup_flag) {
token_buffer += '\0';
def = macro_table.lookup(token_buffer.contents());
token_buffer.set_length(token_buffer.length() - 1);
if (def) {
if (c == '(') {
input_stack::get_char();
interpolate_macro_with_args(def);
}
else
input_stack::push(new macro_input(def));
}
}
if (!def) {
context_buffer = token_buffer;
if (csupper(token_buffer[0]))
return LABEL;
else
return VARIABLE;
}
}
else {
context_buffer = char(c);
return (unsigned char)c;
}
break;
}
}
}
