Method validateXMLString

src/Base/Persistence.cpp:157–197

! * In XML not all valid Unicode characters are allowed. Replace all * disallowed characters with '_' */

 * disallowed characters with '_'
 */
std::string Persistence::validateXMLString(const std::string& str)
{
    // Decode UTF-8 into code points and filter for XML 1.0 validity, replacing invalid or
    // discouraged code points with '_'.
    std::string out;
    out.reserve(str.size());

    const auto* data = reinterpret_cast<const uint8_t*>(str.data());  // NOLINT
    const int32_t len = static_cast<int32_t>(str.size());

    for (int32_t i = 0; i < len;) {
        UChar32 cp = 0;
        U8_NEXT(data, i, len, cp);
        if (cp < 0) {
            out.push_back('_');
            continue;
        }

        const char32_t c32 = static_cast<char32_t>(cp);
        const bool ok = std::ranges::any_of(validRanges, [c32](const auto& r) {
            return c32 >= r.first && c32 <= r.second;
        });
        const bool discouraged = std::ranges::any_of(discouragedRanges, [c32](const auto& r) {
            return c32 >= r.first && c32 <= r.second;
        });

        const char32_t emit = (ok && !discouraged) ? c32 : U'_';
        uint8_t buf[8] {};
        int32_t outLen = 0;
        UBool isError = false;
        U8_APPEND(buf, outLen, static_cast<int32_t>(sizeof(buf)), static_cast<UChar32>(emit), isError);
        if (isError) {
            out.push_back('_');
        }
        else {
            out.append(reinterpret_cast<const char*>(buf), static_cast<std::size_t>(outLen));  // NOLINT
        }
    }

    return out;
}

void Persistence::dumpToStream(std::ostream& stream, int compression)
{