Object* Character::alphabetical_p(STATE) {
bool found;
int c = codepoint(state, &found);
OnigEncodingType* enc = encoding()->get_encoding();
return RBOOL(found && ONIGENC_IS_CODE_ALPHA(enc, c));
}
Object* Character::lower_p(STATE) {
bool found;
int c = codepoint(state, &found);
OnigEncodingType* enc = encoding()->get_encoding();
return RBOOL(found && ONIGENC_IS_CODE_LOWER(enc, c));
}
Object* Character::punctuation_p(STATE) {
bool found;
int c = codepoint(state, &found);
OnigEncodingType* enc = encoding()->get_encoding();
return RBOOL(found && ONIGENC_IS_CODE_PUNCT(enc, c));
}
inline codepoint encode_raw_byte(char byte)
{
codepoint c = codepoint(byte) & 0xFF;
if (c == 0)
{
throw encoding_error("null or zero byte cannot be raw-data encoded");
}
return c | 0xF800;
}
typename boost::enable_if_c<bit_size<typename std::iterator_traits<T>::value_type>::value >= 21, codepoint>::type next(T& it, T const& end)
{
codepoint c = codepoint(*it);
if (!is_valid(c))
{
throw encoding_error("invalid codepoint");
}
++it;
return c;
}
Object* Character::ascii_p(STATE) {
bool found;
int c = codepoint(state, &found);
return RBOOL(found && ONIGENC_IS_CODE_ASCII(c));
}
std::string Scanner::escape(const char *str, int len, char quote_type) const {
std::string output;
output.reserve(len);
if (quote_type == '\'') {
for (int i = 0; i < len; i++) {
unsigned char ch = str[i];
if (ch == '\\') {
if (++i < len) {
switch (str[i]) {
case '\\': output += "\\"; break;
case '\'': output += '\''; break;
default: {
output += ch;
output += str[i];
break;
}
}
} else {
assert(false);
output += ch;
}
} else {
output += ch;
}
}
} else {
for (int i = 0; i < len; i++) {
unsigned char ch = str[i];
if (ch == '\\') {
if (++i < len) {
switch (str[i]) {
case 'n': output += '\n'; break;
case 't': output += '\t'; break;
case 'r': output += '\r'; break;
case 'v': output += '\v'; break;
case 'f': output += '\f'; break;
case 'e': output += '\033'; break;
case '\\': output += '\\'; break;
case '$': output += '$'; break;
case '"':
case '`':
if (str[i] != quote_type) {
output += '\\';
}
output += str[i];
break;
case 'x':
case 'X': {
if (isxdigit(str[i+1])) {
std::string shex;
shex += str[++i]; // 0th hex digit
if (isxdigit(str[i+1])) {
shex += str[++i]; // 1st hex digit
}
output += strtol(shex.c_str(), nullptr, 16);
} else {
output += ch;
output += str[i];
}
break;
}
case 'u': {
// Unicode escape sequence
// "\u{123456}"
if (str[i+1] != '{') {
// BC for "\u1234" passthrough
output += ch;
output += str[i];
break;
}
bool valid = true;
auto start = str + i + 2;
auto closebrace = strchr(start, '}');
if (closebrace > start) {
for (auto p = start; p < closebrace; ++p) {
if (!isxdigit(*p)) {
valid = false;
break;
}
}
} else {
valid = false;
}
auto fatal = [this](const char *msg) {
auto loc = getLocation();
return ParseTimeFatalException(
loc->file,
loc->r.line0,
"%s", msg);
};
if (!valid) {
throw fatal("Invalid UTF-8 codepoint escape sequence");
}
std::string codepoint(start, closebrace - start);
char *end = nullptr;
int32_t uchar = strtol(codepoint.c_str(), &end, 16);
if ((end && *end) || (uchar > 0x10FFFF)) {
throw fatal(
"Invalid UTF-8 codepoint escape sequence: "
"Codepoint too large");
}
if (uchar <= 0x0007F) {
output += (char)uchar;
} else if (uchar <= 0x007FF) {
output += (char)(0xC0 | ( uchar >> 6 ));
output += (char)(0x80 | ( uchar & 0x3F));
} else if (uchar <= 0x00FFFF) {
output += (char)(0xE0 | ( uchar >> 12 ));
output += (char)(0x80 | ((uchar >> 6) & 0x3F));
output += (char)(0x80 | ( uchar & 0x3F));
} else if (uchar <= 0x10FFFF) {
output += (char)(0xF0 | ( uchar >> 18 ));
output += (char)(0x80 | ((uchar >> 12) & 0x3F));
output += (char)(0x80 | ((uchar >> 6) & 0x3F));
output += (char)(0x80 | ( uchar & 0x3F));
} else {
not_reached();
assert(false);
}
i += codepoint.size() + 2 /* strlen("{}") */;
break;
}
