Method append_utf8

core/string/ustring.cpp:1766–1938

}

Error String::append_utf8(const char *p_utf8, int p_len, bool p_skip_cr) {
	if (!p_utf8) {
		return ERR_INVALID_DATA;
	}

	/* HANDLE BOM (Byte Order Mark) */
	if (p_len < 0 || p_len >= 3) {
		bool has_bom = uint8_t(p_utf8[0]) == 0xef && uint8_t(p_utf8[1]) == 0xbb && uint8_t(p_utf8[2]) == 0xbf;
		if (has_bom) {
			//8-bit encoding, byte order has no meaning in UTF-8, just skip it
			if (p_len >= 0) {
				p_len -= 3;
			}
			p_utf8 += 3;
		}
	}

	if (p_len < 0) {
		p_len = strlen(p_utf8);
	}

	const int prev_length = length();
	// If all utf8 characters maps to ASCII, then the max size will be p_len, and we add +1 for the null termination.
	resize_uninitialized(prev_length + p_len + 1);
	char32_t *dst = ptrw() + prev_length;

	Error result = Error::OK;

	const uint8_t *ptrtmp = (uint8_t *)p_utf8;
	const uint8_t *ptr_limit = (uint8_t *)p_utf8 + p_len;

	while (ptrtmp < ptr_limit && *ptrtmp) {
		uint8_t c = *ptrtmp;

		if (p_skip_cr && c == '\r') {
			++ptrtmp;
			continue;
		}
		uint32_t unicode = _replacement_char;
		uint32_t size = 1;

		if ((c & 0b10000000) == 0) {
			unicode = c;
			if (unicode > 0x7F) {
				unicode = _replacement_char;
				print_unicode_error(vformat("Invalid unicode codepoint (%d)", unicode), true);
				result = Error::ERR_INVALID_DATA;
			}
		} else if ((c & 0b11100000) == 0b11000000) {
			if (ptrtmp + 1 >= ptr_limit) {
				print_unicode_error(vformat("Missing %x UTF-8 continuation byte", c), true);
				result = Error::ERR_INVALID_DATA;
			} else {
				uint8_t c2 = *(ptrtmp + 1);

				if ((c2 & 0b11000000) == 0b10000000) {
					unicode = (uint32_t)((c & 0b00011111) << 6) | (uint32_t)(c2 & 0b00111111);