diff --git a/include/nlohmann/detail/input/binary_reader.hpp b/include/nlohmann/detail/input/binary_reader.hpp index 4041cedba..1245ac1d2 100644 --- a/include/nlohmann/detail/input/binary_reader.hpp +++ b/include/nlohmann/detail/input/binary_reader.hpp @@ -95,12 +95,13 @@ class binary_reader @return whether parsing was successful */ JSON_HEDLEY_NON_NULL(3) - bool sax_parse(const input_format_t format, + bool sax_parse(const input_format_t format_, json_sax_t* sax_, const bool strict = true, const cbor_tag_handler_t tag_handler = cbor_tag_handler_t::error) { sax = sax_; + format = format_; bool result = false; switch (format) @@ -118,6 +119,7 @@ class binary_reader break; case input_format_t::ubjson: + case input_format_t::bjdata: result = parse_ubjson_internal(); break; @@ -129,7 +131,7 @@ class binary_reader // strict mode: next byte must be EOF if (result && strict) { - if (format == input_format_t::ubjson) + if (format == input_format_t::ubjson || format == input_format_t::bjdata) { get_ignore_noop(); } @@ -1844,7 +1846,7 @@ class binary_reader get(); // TODO(niels): may we ignore N here? } - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "value"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "value"))) { return false; } @@ -1854,52 +1856,141 @@ class binary_reader case 'U': { std::uint8_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'i': { std::int8_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'I': { std::int16_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'l': { std::int32_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'L': { std::int64_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } default: + if (format == input_format_t::bjdata) + { + switch (current) + { + case 'u': + { + uint16_t len; + return get_number(format, len) and get_string(format, len, result); + } + case 'm': + { + uint32_t len; + return get_number(format, len) and get_string(format, len, result); + } + case 'M': + { + uint64_t len; + return get_number(format, len) and get_string(format, len, result); + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("expected length type specification (U, i, I, l, L); last byte: 0x", last_token), "string"), nullptr)); + exception_message(format, concat("expected length type specification (U, i, I, l, L); last byte: 0x", last_token), "string"), nullptr)); } } + /*! + @param[out] dim an integer vector storing the ND array dimensions + @return whether array creation completed + */ + bool get_ubjson_ndarray_size(std::vector& dim) + { + std::pair size_and_type; + size_t dimlen; + + bool is_optimized = get_ubjson_size_type(size_and_type); + + if (size_and_type.first != string_t::npos) + { + if (size_and_type.second != 0) + { + if (size_and_type.second != 'N') + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen, size_and_type.second))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + } + } + } + else + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + } + } + } + else + { + while (current != ']') + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + get_ignore_noop(); + } + } + return true; + } + /*! @param[out] result determined size @return whether size determination completed */ - bool get_ubjson_size_value(std::size_t& result) + bool get_ubjson_size_value(std::size_t& result, int prefix = 0) { - switch (get_ignore_noop()) + if (prefix == 0) + { + prefix = get_ignore_noop(); + } + + switch (prefix) { case 'U': { std::uint8_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -1910,7 +2001,7 @@ class binary_reader case 'i': { std::int8_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -1921,7 +2012,7 @@ class binary_reader case 'I': { std::int16_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -1932,7 +2023,7 @@ class binary_reader case 'l': { std::int32_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -1943,7 +2034,7 @@ class binary_reader case 'L': { std::int64_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -1953,9 +2044,62 @@ class binary_reader default: { + if (format == input_format_t::bjdata) + { + switch (prefix) + { + case 'u': + { + uint16_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case 'm': + { + uint32_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case 'M': + { + uint64_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case '[': + { + std::vector dim; + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_ndarray_size(dim))) + { + return false; + } + else + { + result = 1; + for (std::size_t i = 0; i < dim.size(); ++i) + { + result *= dim.at(i); + } + } + return true; + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("expected length type specification (U, i, I, l, L) after '#'; last byte: 0x", last_token), "size"), nullptr)); + exception_message(format, concat("expected length type specification (U, i, I, l, L) after '#'; last byte: 0x", last_token), "size"), nullptr)); } } } @@ -1980,7 +2124,7 @@ class binary_reader if (current == '$') { result.second = get(); // must not ignore 'N', because 'N' maybe the type - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "type"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "type"))) { return false; } @@ -1988,13 +2132,13 @@ class binary_reader get_ignore_noop(); if (JSON_HEDLEY_UNLIKELY(current != '#')) { - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "value"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "value"))) { return false; } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, - exception_message(input_format_t::ubjson, concat("expected '#' after type information; last byte: 0x", last_token), "size"), nullptr)); + exception_message(format, concat("expected '#' after type information; last byte: 0x", last_token), "size"), nullptr)); } return get_ubjson_size_value(result.first); @@ -2017,7 +2161,7 @@ class binary_reader switch (prefix) { case std::char_traits::eof(): // EOF - return unexpect_eof(input_format_t::ubjson, "value"); + return unexpect_eof(format, "value"); case 'T': // true return sax->boolean(true); @@ -2030,43 +2174,43 @@ class binary_reader case 'U': { std::uint8_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_unsigned(number); + return get_number(format, number) && sax->number_unsigned(number); } case 'i': { std::int8_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'I': { std::int16_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'l': { std::int32_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'L': { std::int64_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'd': { float number{}; - return get_number(input_format_t::ubjson, number) && sax->number_float(static_cast(number), ""); + return get_number(format, number) && sax->number_float(static_cast(number), ""); } case 'D': { double number{}; - return get_number(input_format_t::ubjson, number) && sax->number_float(static_cast(number), ""); + return get_number(format, number) && sax->number_float(static_cast(number), ""); } case 'H': @@ -2077,7 +2221,7 @@ class binary_reader case 'C': // char { get(); - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "char"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "char"))) { return false; } @@ -2085,7 +2229,7 @@ class binary_reader { auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("byte after 'C' must be in range 0x00..0x7F; last byte: 0x", last_token), "char"), nullptr)); + exception_message(format, concat("byte after 'C' must be in range 0x00..0x7F; last byte: 0x", last_token), "char"), nullptr)); } string_t s(1, static_cast(current)); return sax->string(s); @@ -2105,9 +2249,74 @@ class binary_reader default: // anything else { + if (format == input_format_t::bjdata) + { + switch (prefix) + { + case 'u': + { + uint16_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'm': + { + uint32_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'M': + { + uint64_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'h': + { + const int byte2 = get(); + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "half"))) + { + return false; + } + const int byte1 = get(); + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "half"))) + { + return false; + } + + // code from RFC 7049, Appendix D, Figure 3: + // As half-precision floating-point numbers were only added + // to IEEE 754 in 2008, today's programming platforms often + // still only have limited support for them. It is very + // easy to include at least decoding support for them even + // without such support. An example of a small decoder for + // half-precision floating-point numbers in the C language + // is shown in Fig. 3. + const int half = (byte1 << 8) + byte2; + const double val = [&half] + { + const int exp = (half >> 10) & 0x1F; + const int mant = half & 0x3FF; + JSON_ASSERT(0 <= exp and exp <= 32); + JSON_ASSERT(0 <= mant and mant <= 1024); + switch (exp) + { + case 0: + return std::ldexp(mant, -24); + case 31: + return (mant == 0) + ? std::numeric_limits::infinity() + : std::numeric_limits::quiet_NaN(); + default: + return std::ldexp(mant + 1024, exp - 25); + } + }(); + return sax->number_float((half & 0x8000) != 0 + ? static_cast(-val) + : static_cast(val), ""); + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, - exception_message(input_format_t::ubjson, concat("invalid byte: 0x", last_token), "value"), nullptr)); + exception_message(format, concat("invalid byte: 0x", last_token), "value"), nullptr)); } } } @@ -2267,7 +2476,7 @@ class binary_reader for (std::size_t i = 0; i < size; ++i) { get(); - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "number"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "number"))) { return false; } @@ -2286,7 +2495,7 @@ class binary_reader if (JSON_HEDLEY_UNLIKELY(result_remainder != token_type::end_of_input)) { return sax->parse_error(chars_read, number_string, parse_error::create(115, chars_read, - exception_message(input_format_t::ubjson, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); + exception_message(format, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); } switch (result_number) @@ -2313,7 +2522,7 @@ class binary_reader case token_type::literal_or_value: default: return sax->parse_error(chars_read, number_string, parse_error::create(115, chars_read, - exception_message(input_format_t::ubjson, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); + exception_message(format, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); } } @@ -2377,7 +2586,8 @@ class binary_reader } // reverse byte order prior to conversion if necessary - if (is_little_endian != InputIsLittleEndian) + if ((is_little_endian != InputIsLittleEndian and format != input_format_t::bjdata) or + (is_little_endian == InputIsLittleEndian and format == input_format_t::bjdata)) { vec[sizeof(NumberType) - i - 1] = static_cast(current); } @@ -2514,6 +2724,10 @@ class binary_reader error_msg += "BSON"; break; + case input_format_t::bjdata: + error_msg += "BJData"; + break; + case input_format_t::json: // LCOV_EXCL_LINE default: // LCOV_EXCL_LINE JSON_ASSERT(false); // NOLINT(cert-dcl03-c,hicpp-static-assert,misc-static-assert) LCOV_EXCL_LINE @@ -2535,6 +2749,9 @@ class binary_reader /// whether we can assume little endianness const bool is_little_endian = little_endianness(); + /// sax parser format + input_format_t format; + /// the SAX parser json_sax_t* sax = nullptr; }; diff --git a/include/nlohmann/detail/input/input_adapters.hpp b/include/nlohmann/detail/input/input_adapters.hpp index d196aec54..4c1ad3cef 100644 --- a/include/nlohmann/detail/input/input_adapters.hpp +++ b/include/nlohmann/detail/input/input_adapters.hpp @@ -23,7 +23,7 @@ namespace nlohmann namespace detail { /// the supported input formats -enum class input_format_t { json, cbor, msgpack, ubjson, bson }; +enum class input_format_t { json, cbor, msgpack, ubjson, bson, bjdata }; //////////////////// // input adapters // diff --git a/include/nlohmann/detail/output/binary_writer.hpp b/include/nlohmann/detail/output/binary_writer.hpp index c018b8ccd..c51e2b9ae 100644 --- a/include/nlohmann/detail/output/binary_writer.hpp +++ b/include/nlohmann/detail/output/binary_writer.hpp @@ -38,9 +38,10 @@ class binary_writer @param[in] adapter output adapter to write to */ - explicit binary_writer(output_adapter_t adapter) : oa(std::move(adapter)) + explicit binary_writer(output_adapter_t adapter, const bool is_bjdata_ = false) : oa(std::move(adapter)) { JSON_ASSERT(oa); + is_bjdata = is_bjdata_; } /*! @@ -1334,6 +1335,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('u')); // uint16 - bjdata only + } + write_number(static_cast(n)); + } else if (n <= static_cast((std::numeric_limits::max)())) { if (add_prefix) @@ -1342,6 +1351,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('m')); // uint32 - bjdata only + } + write_number(static_cast(n)); + } else if (n <= static_cast((std::numeric_limits::max)())) { if (add_prefix) @@ -1350,6 +1367,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('M')); // uint64 - bjdata only + } + write_number(static_cast(n)); + } else { if (add_prefix) @@ -1397,6 +1422,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('u')); // uint16 - bjdata only + } + write_number(static_cast(n)); + } else if ((std::numeric_limits::min)() <= n && n <= (std::numeric_limits::max)()) { if (add_prefix) @@ -1405,6 +1438,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('m')); // uint32 - bjdata only + } + write_number(static_cast(n)); + } else if ((std::numeric_limits::min)() <= n && n <= (std::numeric_limits::max)()) { if (add_prefix) @@ -1413,6 +1454,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('M')); // uint64 - bjdata only + } + write_number(static_cast(n)); + } // LCOV_EXCL_START else { @@ -1458,14 +1507,26 @@ class binary_writer { return 'I'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'u'; + } if ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)()) { return 'l'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'm'; + } if ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)()) { return 'L'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'M'; + } // anything else is treated as high-precision number return 'H'; // LCOV_EXCL_LINE } @@ -1484,14 +1545,26 @@ class binary_writer { return 'I'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'u'; + } if (j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) { return 'l'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'm'; + } if (j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) { return 'L'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'M'; + } // anything else is treated as high-precision number return 'H'; // LCOV_EXCL_LINE } @@ -1539,6 +1612,7 @@ class binary_writer @note This function needs to respect the system's endianness, because bytes in CBOR, MessagePack, and UBJSON are stored in network order (big endian) and therefore need reordering on little endian systems. + BJData is similar to UBJSON but uses little endian by default. */ template void write_number(const NumberType n) @@ -1548,7 +1622,7 @@ class binary_writer std::memcpy(vec.data(), &n, sizeof(NumberType)); // step 2: write array to output (with possible reordering) - if (is_little_endian != OutputIsLittleEndian) + if ((!is_bjdata and (is_little_endian != OutputIsLittleEndian)) or (is_bjdata and !is_little_endian)) { // reverse byte order prior to conversion if necessary std::reverse(vec.begin(), vec.end()); @@ -1629,6 +1703,9 @@ class binary_writer /// whether we can assume little endianness const bool is_little_endian = little_endianness(); + /// whether to write in bjdata format + bool is_bjdata = false; + /// the output output_adapter_t oa = nullptr; }; diff --git a/include/nlohmann/json.hpp b/include/nlohmann/json.hpp index 1a2da8d45..d232b2f7a 100644 --- a/include/nlohmann/json.hpp +++ b/include/nlohmann/json.hpp @@ -3965,6 +3965,27 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec binary_writer(o).write_ubjson(j, use_size, use_type); } + static std::vector to_bjdata(const basic_json& j, + const bool use_size = false, + const bool use_type = false) + { + std::vector result; + to_bjdata(j, result, use_size, use_type); + return result; + } + + static void to_bjdata(const basic_json& j, detail::output_adapter o, + const bool use_size = false, const bool use_type = false) + { + binary_writer(o, true).write_ubjson(j, use_size, use_type); + } + + static void to_bjdata(const basic_json& j, detail::output_adapter o, + const bool use_size = false, const bool use_type = false) + { + binary_writer(o, true).write_ubjson(j, use_size, use_type); + } + /// @brief create a BSON serialization of a given JSON value /// @sa https://json.nlohmann.me/api/basic_json/to_bson/ static std::vector to_bson(const basic_json& j) @@ -4155,6 +4176,62 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec return res ? result : basic_json(value_t::discarded); } + + /// @brief create a JSON value from an input in UBJSON format + /// @sa https://json.nlohmann.me/api/basic_json/from_ubjson/ + template + JSON_HEDLEY_WARN_UNUSED_RESULT + static basic_json from_bjdata(InputType&& i, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = detail::input_adapter(std::forward(i)); + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + /// @brief create a JSON value from an input in BJData format + /// @sa https://github.com/NeuroJSON/bjdata/blob/master/Binary_JData_Specification.md + template + JSON_HEDLEY_WARN_UNUSED_RESULT + static basic_json from_bjdata(IteratorType first, IteratorType last, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = detail::input_adapter(std::move(first), std::move(last)); + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + template + JSON_HEDLEY_WARN_UNUSED_RESULT + JSON_HEDLEY_DEPRECATED_FOR(3.8.0, from_bjdata(ptr, ptr + len)) + static basic_json from_bjdata(const T* ptr, std::size_t len, + const bool strict = true, + const bool allow_exceptions = true) + { + return from_bjdata(ptr, ptr + len, strict, allow_exceptions); + } + + JSON_HEDLEY_WARN_UNUSED_RESULT + JSON_HEDLEY_DEPRECATED_FOR(3.8.0, from_bjdata(ptr, ptr + len)) + static basic_json from_bjdata(detail::span_input_adapter&& i, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = i.get(); + // NOLINTNEXTLINE(hicpp-move-const-arg,performance-move-const-arg) + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + /// @brief create a JSON value from an input in BSON format /// @sa https://json.nlohmann.me/api/basic_json/from_bson/ template diff --git a/single_include/nlohmann/json.hpp b/single_include/nlohmann/json.hpp index a29c52997..13c501f70 100644 --- a/single_include/nlohmann/json.hpp +++ b/single_include/nlohmann/json.hpp @@ -5433,7 +5433,7 @@ namespace nlohmann namespace detail { /// the supported input formats -enum class input_format_t { json, cbor, msgpack, ubjson, bson }; +enum class input_format_t { json, cbor, msgpack, ubjson, bson, bjdata }; //////////////////// // input adapters // @@ -8471,12 +8471,13 @@ class binary_reader @return whether parsing was successful */ JSON_HEDLEY_NON_NULL(3) - bool sax_parse(const input_format_t format, + bool sax_parse(const input_format_t format_, json_sax_t* sax_, const bool strict = true, const cbor_tag_handler_t tag_handler = cbor_tag_handler_t::error) { sax = sax_; + format = format_; bool result = false; switch (format) @@ -8494,6 +8495,7 @@ class binary_reader break; case input_format_t::ubjson: + case input_format_t::bjdata: result = parse_ubjson_internal(); break; @@ -8505,7 +8507,7 @@ class binary_reader // strict mode: next byte must be EOF if (result && strict) { - if (format == input_format_t::ubjson) + if (format == input_format_t::ubjson || format == input_format_t::bjdata) { get_ignore_noop(); } @@ -10220,7 +10222,7 @@ class binary_reader get(); // TODO(niels): may we ignore N here? } - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "value"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "value"))) { return false; } @@ -10230,52 +10232,141 @@ class binary_reader case 'U': { std::uint8_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'i': { std::int8_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'I': { std::int16_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'l': { std::int32_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } case 'L': { std::int64_t len{}; - return get_number(input_format_t::ubjson, len) && get_string(input_format_t::ubjson, len, result); + return get_number(format, len) && get_string(format, len, result); } default: + if (format == input_format_t::bjdata) + { + switch (current) + { + case 'u': + { + uint16_t len; + return get_number(format, len) and get_string(format, len, result); + } + case 'm': + { + uint32_t len; + return get_number(format, len) and get_string(format, len, result); + } + case 'M': + { + uint64_t len; + return get_number(format, len) and get_string(format, len, result); + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("expected length type specification (U, i, I, l, L); last byte: 0x", last_token), "string"), nullptr)); + exception_message(format, concat("expected length type specification (U, i, I, l, L); last byte: 0x", last_token), "string"), nullptr)); } } + /*! + @param[out] dim an integer vector storing the ND array dimensions + @return whether array creation completed + */ + bool get_ubjson_ndarray_size(std::vector& dim) + { + std::pair size_and_type; + size_t dimlen; + + bool is_optimized = get_ubjson_size_type(size_and_type); + + if (size_and_type.first != string_t::npos) + { + if (size_and_type.second != 0) + { + if (size_and_type.second != 'N') + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen, size_and_type.second))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + } + } + } + else + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + } + } + } + else + { + while (current != ']') + { + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_size_value(dimlen))) + { + return false; + } + else + { + dim.push_back(dimlen); + } + get_ignore_noop(); + } + } + return true; + } + /*! @param[out] result determined size @return whether size determination completed */ - bool get_ubjson_size_value(std::size_t& result) + bool get_ubjson_size_value(std::size_t& result, int prefix = 0) { - switch (get_ignore_noop()) + if (prefix == 0) + { + prefix = get_ignore_noop(); + } + + switch (prefix) { case 'U': { std::uint8_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -10286,7 +10377,7 @@ class binary_reader case 'i': { std::int8_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -10297,7 +10388,7 @@ class binary_reader case 'I': { std::int16_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -10308,7 +10399,7 @@ class binary_reader case 'l': { std::int32_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -10319,7 +10410,7 @@ class binary_reader case 'L': { std::int64_t number{}; - if (JSON_HEDLEY_UNLIKELY(!get_number(input_format_t::ubjson, number))) + if (JSON_HEDLEY_UNLIKELY(!get_number(format, number))) { return false; } @@ -10329,9 +10420,62 @@ class binary_reader default: { + if (format == input_format_t::bjdata) + { + switch (prefix) + { + case 'u': + { + uint16_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case 'm': + { + uint32_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case 'M': + { + uint64_t number; + if (JSON_HEDLEY_UNLIKELY(not get_number(format, number))) + { + return false; + } + result = static_cast(number); + return true; + } + case '[': + { + std::vector dim; + if (JSON_HEDLEY_UNLIKELY(not get_ubjson_ndarray_size(dim))) + { + return false; + } + else + { + result = 1; + for (std::size_t i = 0; i < dim.size(); ++i) + { + result *= dim.at(i); + } + } + return true; + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("expected length type specification (U, i, I, l, L) after '#'; last byte: 0x", last_token), "size"), nullptr)); + exception_message(format, concat("expected length type specification (U, i, I, l, L) after '#'; last byte: 0x", last_token), "size"), nullptr)); } } } @@ -10356,7 +10500,7 @@ class binary_reader if (current == '$') { result.second = get(); // must not ignore 'N', because 'N' maybe the type - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "type"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "type"))) { return false; } @@ -10364,13 +10508,13 @@ class binary_reader get_ignore_noop(); if (JSON_HEDLEY_UNLIKELY(current != '#')) { - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "value"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "value"))) { return false; } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, - exception_message(input_format_t::ubjson, concat("expected '#' after type information; last byte: 0x", last_token), "size"), nullptr)); + exception_message(format, concat("expected '#' after type information; last byte: 0x", last_token), "size"), nullptr)); } return get_ubjson_size_value(result.first); @@ -10393,7 +10537,7 @@ class binary_reader switch (prefix) { case std::char_traits::eof(): // EOF - return unexpect_eof(input_format_t::ubjson, "value"); + return unexpect_eof(format, "value"); case 'T': // true return sax->boolean(true); @@ -10406,43 +10550,43 @@ class binary_reader case 'U': { std::uint8_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_unsigned(number); + return get_number(format, number) && sax->number_unsigned(number); } case 'i': { std::int8_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'I': { std::int16_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'l': { std::int32_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'L': { std::int64_t number{}; - return get_number(input_format_t::ubjson, number) && sax->number_integer(number); + return get_number(format, number) && sax->number_integer(number); } case 'd': { float number{}; - return get_number(input_format_t::ubjson, number) && sax->number_float(static_cast(number), ""); + return get_number(format, number) && sax->number_float(static_cast(number), ""); } case 'D': { double number{}; - return get_number(input_format_t::ubjson, number) && sax->number_float(static_cast(number), ""); + return get_number(format, number) && sax->number_float(static_cast(number), ""); } case 'H': @@ -10453,7 +10597,7 @@ class binary_reader case 'C': // char { get(); - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "char"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "char"))) { return false; } @@ -10461,7 +10605,7 @@ class binary_reader { auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, - exception_message(input_format_t::ubjson, concat("byte after 'C' must be in range 0x00..0x7F; last byte: 0x", last_token), "char"), nullptr)); + exception_message(format, concat("byte after 'C' must be in range 0x00..0x7F; last byte: 0x", last_token), "char"), nullptr)); } string_t s(1, static_cast(current)); return sax->string(s); @@ -10481,9 +10625,74 @@ class binary_reader default: // anything else { + if (format == input_format_t::bjdata) + { + switch (prefix) + { + case 'u': + { + uint16_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'm': + { + uint32_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'M': + { + uint64_t number; + return get_number(format, number) and sax->number_integer(number); + } + case 'h': + { + const int byte2 = get(); + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "half"))) + { + return false; + } + const int byte1 = get(); + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "half"))) + { + return false; + } + + // code from RFC 7049, Appendix D, Figure 3: + // As half-precision floating-point numbers were only added + // to IEEE 754 in 2008, today's programming platforms often + // still only have limited support for them. It is very + // easy to include at least decoding support for them even + // without such support. An example of a small decoder for + // half-precision floating-point numbers in the C language + // is shown in Fig. 3. + const int half = (byte1 << 8) + byte2; + const double val = [&half] + { + const int exp = (half >> 10) & 0x1F; + const int mant = half & 0x3FF; + JSON_ASSERT(0 <= exp and exp <= 32); + JSON_ASSERT(0 <= mant and mant <= 1024); + switch (exp) + { + case 0: + return std::ldexp(mant, -24); + case 31: + return (mant == 0) + ? std::numeric_limits::infinity() + : std::numeric_limits::quiet_NaN(); + default: + return std::ldexp(mant + 1024, exp - 25); + } + }(); + return sax->number_float((half & 0x8000) != 0 + ? static_cast(-val) + : static_cast(val), ""); + } + } + } auto last_token = get_token_string(); return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, - exception_message(input_format_t::ubjson, concat("invalid byte: 0x", last_token), "value"), nullptr)); + exception_message(format, concat("invalid byte: 0x", last_token), "value"), nullptr)); } } } @@ -10643,7 +10852,7 @@ class binary_reader for (std::size_t i = 0; i < size; ++i) { get(); - if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(input_format_t::ubjson, "number"))) + if (JSON_HEDLEY_UNLIKELY(!unexpect_eof(format, "number"))) { return false; } @@ -10662,7 +10871,7 @@ class binary_reader if (JSON_HEDLEY_UNLIKELY(result_remainder != token_type::end_of_input)) { return sax->parse_error(chars_read, number_string, parse_error::create(115, chars_read, - exception_message(input_format_t::ubjson, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); + exception_message(format, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); } switch (result_number) @@ -10689,7 +10898,7 @@ class binary_reader case token_type::literal_or_value: default: return sax->parse_error(chars_read, number_string, parse_error::create(115, chars_read, - exception_message(input_format_t::ubjson, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); + exception_message(format, concat("invalid number text: ", number_lexer.get_token_string()), "high-precision number"), nullptr)); } } @@ -10753,7 +10962,8 @@ class binary_reader } // reverse byte order prior to conversion if necessary - if (is_little_endian != InputIsLittleEndian) + if ((is_little_endian != InputIsLittleEndian and format != input_format_t::bjdata) or + (is_little_endian == InputIsLittleEndian and format == input_format_t::bjdata)) { vec[sizeof(NumberType) - i - 1] = static_cast(current); } @@ -10890,6 +11100,10 @@ class binary_reader error_msg += "BSON"; break; + case input_format_t::bjdata: + error_msg += "BJData"; + break; + case input_format_t::json: // LCOV_EXCL_LINE default: // LCOV_EXCL_LINE JSON_ASSERT(false); // NOLINT(cert-dcl03-c,hicpp-static-assert,misc-static-assert) LCOV_EXCL_LINE @@ -10911,6 +11125,9 @@ class binary_reader /// whether we can assume little endianness const bool is_little_endian = little_endianness(); + /// sax parser format + input_format_t format; + /// the SAX parser json_sax_t* sax = nullptr; }; @@ -13606,9 +13823,10 @@ class binary_writer @param[in] adapter output adapter to write to */ - explicit binary_writer(output_adapter_t adapter) : oa(std::move(adapter)) + explicit binary_writer(output_adapter_t adapter, const bool is_bjdata_ = false) : oa(std::move(adapter)) { JSON_ASSERT(oa); + is_bjdata = is_bjdata_; } /*! @@ -14902,6 +15120,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('u')); // uint16 - bjdata only + } + write_number(static_cast(n)); + } else if (n <= static_cast((std::numeric_limits::max)())) { if (add_prefix) @@ -14910,6 +15136,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('m')); // uint32 - bjdata only + } + write_number(static_cast(n)); + } else if (n <= static_cast((std::numeric_limits::max)())) { if (add_prefix) @@ -14918,6 +15152,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and n <= static_cast((std::numeric_limits::max)())) + { + if (add_prefix) + { + oa->write_character(to_char_type('M')); // uint64 - bjdata only + } + write_number(static_cast(n)); + } else { if (add_prefix) @@ -14965,6 +15207,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('u')); // uint16 - bjdata only + } + write_number(static_cast(n)); + } else if ((std::numeric_limits::min)() <= n && n <= (std::numeric_limits::max)()) { if (add_prefix) @@ -14973,6 +15223,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('m')); // uint32 - bjdata only + } + write_number(static_cast(n)); + } else if ((std::numeric_limits::min)() <= n && n <= (std::numeric_limits::max)()) { if (add_prefix) @@ -14981,6 +15239,14 @@ class binary_writer } write_number(static_cast(n)); } + else if (is_bjdata and (static_cast((std::numeric_limits::min)()) <= n && n <= static_cast((std::numeric_limits::max)()))) + { + if (add_prefix) + { + oa->write_character(to_char_type('M')); // uint64 - bjdata only + } + write_number(static_cast(n)); + } // LCOV_EXCL_START else { @@ -15026,14 +15292,26 @@ class binary_writer { return 'I'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'u'; + } if ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)()) { return 'l'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'm'; + } if ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)()) { return 'L'; } + if (is_bjdata and ((std::numeric_limits::min)() <= j.m_value.number_integer && j.m_value.number_integer <= (std::numeric_limits::max)())) + { + return 'M'; + } // anything else is treated as high-precision number return 'H'; // LCOV_EXCL_LINE } @@ -15052,14 +15330,26 @@ class binary_writer { return 'I'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'u'; + } if (j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) { return 'l'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'm'; + } if (j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) { return 'L'; } + if (is_bjdata and j.m_value.number_unsigned <= static_cast((std::numeric_limits::max)())) + { + return 'M'; + } // anything else is treated as high-precision number return 'H'; // LCOV_EXCL_LINE } @@ -15107,6 +15397,7 @@ class binary_writer @note This function needs to respect the system's endianness, because bytes in CBOR, MessagePack, and UBJSON are stored in network order (big endian) and therefore need reordering on little endian systems. + BJData is similar to UBJSON but uses little endian by default. */ template void write_number(const NumberType n) @@ -15116,7 +15407,7 @@ class binary_writer std::memcpy(vec.data(), &n, sizeof(NumberType)); // step 2: write array to output (with possible reordering) - if (is_little_endian != OutputIsLittleEndian) + if ((!is_bjdata and (is_little_endian != OutputIsLittleEndian)) or (is_bjdata and !is_little_endian)) { // reverse byte order prior to conversion if necessary std::reverse(vec.begin(), vec.end()); @@ -15197,6 +15488,9 @@ class binary_writer /// whether we can assume little endianness const bool is_little_endian = little_endianness(); + /// whether to write in bjdata format + bool is_bjdata = false; + /// the output output_adapter_t oa = nullptr; }; @@ -21413,6 +21707,27 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec binary_writer(o).write_ubjson(j, use_size, use_type); } + static std::vector to_bjdata(const basic_json& j, + const bool use_size = false, + const bool use_type = false) + { + std::vector result; + to_bjdata(j, result, use_size, use_type); + return result; + } + + static void to_bjdata(const basic_json& j, detail::output_adapter o, + const bool use_size = false, const bool use_type = false) + { + binary_writer(o, true).write_ubjson(j, use_size, use_type); + } + + static void to_bjdata(const basic_json& j, detail::output_adapter o, + const bool use_size = false, const bool use_type = false) + { + binary_writer(o, true).write_ubjson(j, use_size, use_type); + } + /// @brief create a BSON serialization of a given JSON value /// @sa https://json.nlohmann.me/api/basic_json/to_bson/ static std::vector to_bson(const basic_json& j) @@ -21603,6 +21918,62 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec return res ? result : basic_json(value_t::discarded); } + + /// @brief create a JSON value from an input in UBJSON format + /// @sa https://json.nlohmann.me/api/basic_json/from_ubjson/ + template + JSON_HEDLEY_WARN_UNUSED_RESULT + static basic_json from_bjdata(InputType&& i, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = detail::input_adapter(std::forward(i)); + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + /// @brief create a JSON value from an input in BJData format + /// @sa https://github.com/NeuroJSON/bjdata/blob/master/Binary_JData_Specification.md + template + JSON_HEDLEY_WARN_UNUSED_RESULT + static basic_json from_bjdata(IteratorType first, IteratorType last, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = detail::input_adapter(std::move(first), std::move(last)); + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + template + JSON_HEDLEY_WARN_UNUSED_RESULT + JSON_HEDLEY_DEPRECATED_FOR(3.8.0, from_bjdata(ptr, ptr + len)) + static basic_json from_bjdata(const T* ptr, std::size_t len, + const bool strict = true, + const bool allow_exceptions = true) + { + return from_bjdata(ptr, ptr + len, strict, allow_exceptions); + } + + JSON_HEDLEY_WARN_UNUSED_RESULT + JSON_HEDLEY_DEPRECATED_FOR(3.8.0, from_bjdata(ptr, ptr + len)) + static basic_json from_bjdata(detail::span_input_adapter&& i, + const bool strict = true, + const bool allow_exceptions = true) + { + basic_json result; + detail::json_sax_dom_parser sdp(result, allow_exceptions); + auto ia = i.get(); + // NOLINTNEXTLINE(hicpp-move-const-arg,performance-move-const-arg) + const bool res = binary_reader(std::move(ia)).sax_parse(input_format_t::bjdata, &sdp, strict); + return res ? result : basic_json(value_t::discarded); + } + + /// @brief create a JSON value from an input in BSON format /// @sa https://json.nlohmann.me/api/basic_json/from_bson/ template diff --git a/test/src/fuzzer-parse_bjdata.cpp b/test/src/fuzzer-parse_bjdata.cpp new file mode 100644 index 000000000..99e799bd3 --- /dev/null +++ b/test/src/fuzzer-parse_bjdata.cpp @@ -0,0 +1,84 @@ +/* + __ _____ _____ _____ + __| | __| | | | JSON for Modern C++ (fuzz test support) +| | |__ | | | | | | version 3.10.5 +|_____|_____|_____|_|___| https://github.com/nlohmann/json + +This file implements a parser test suitable for fuzz testing. Given a byte +array data, it performs the following steps: + +- j1 = from_bjdata(data) +- vec = to_bjdata(j1) +- j2 = from_bjdata(vec) +- assert(j1 == j2) +- vec2 = to_bjdata(j1, use_size = true, use_type = false) +- j3 = from_bjdata(vec2) +- assert(j1 == j3) +- vec3 = to_bjdata(j1, use_size = true, use_type = true) +- j4 = from_bjdata(vec3) +- assert(j1 == j4) + +The provided function `LLVMFuzzerTestOneInput` can be used in different fuzzer +drivers. + +Licensed under the MIT License . +*/ + +#include +#include +#include + +using json = nlohmann::json; + +// see http://llvm.org/docs/LibFuzzer.html +extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) +{ + try + { + // step 1: parse input + std::vector vec1(data, data + size); + json j1 = json::from_bjdata(vec1); + + try + { + // step 2.1: round trip without adding size annotations to container types + std::vector vec2 = json::to_bjdata(j1, false, false); + + // step 2.2: round trip with adding size annotations but without adding type annonations to container types + std::vector vec3 = json::to_bjdata(j1, true, false); + + // step 2.3: round trip with adding size as well as type annotations to container types + std::vector vec4 = json::to_bjdata(j1, true, true); + + // parse serialization + json j2 = json::from_bjdata(vec2); + json j3 = json::from_bjdata(vec3); + json j4 = json::from_bjdata(vec4); + + // serializations must match + assert(json::to_bjdata(j2, false, false) == vec2); + assert(json::to_bjdata(j3, true, false) == vec3); + assert(json::to_bjdata(j4, true, true) == vec4); + } + catch (const json::parse_error&) + { + // parsing a BJData serialization must not fail + assert(false); + } + } + catch (const json::parse_error&) + { + // parse errors are ok, because input may be random bytes + } + catch (const json::type_error&) + { + // type errors can occur during parsing, too + } + catch (const json::out_of_range&) + { + // out of range errors may happen if provided sizes are excessive + } + + // return 0 - non-zero return values are reserved for future use + return 0; +} diff --git a/test/src/unit-bjdata.cpp b/test/src/unit-bjdata.cpp new file mode 100644 index 000000000..da6983318 --- /dev/null +++ b/test/src/unit-bjdata.cpp @@ -0,0 +1,2981 @@ +/* + __ _____ _____ _____ + __| | __| | | | JSON for Modern C++ (test suite) +| | |__ | | | | | | version 3.10.5 +|_____|_____|_____|_|___| https://github.com/nlohmann/json + +Licensed under the MIT License . +SPDX-License-Identifier: MIT +Copyright (c) 2013-2022 Niels Lohmann . + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +*/ + +#include "doctest_compatibility.h" + +#include +using nlohmann::json; + +#include +#include +#include +#include +#include "test_utils.hpp" + +namespace +{ +class SaxCountdown +{ + public: + explicit SaxCountdown(const int count) : events_left(count) + {} + + bool null() + { + return events_left-- > 0; + } + + bool boolean(bool /*unused*/) + { + return events_left-- > 0; + } + + bool number_integer(json::number_integer_t /*unused*/) + { + return events_left-- > 0; + } + + bool number_unsigned(json::number_unsigned_t /*unused*/) + { + return events_left-- > 0; + } + + bool number_float(json::number_float_t /*unused*/, const std::string& /*unused*/) + { + return events_left-- > 0; + } + + bool string(std::string& /*unused*/) + { + return events_left-- > 0; + } + + bool binary(std::vector& /*unused*/) + { + return events_left-- > 0; + } + + bool start_object(std::size_t /*unused*/) + { + return events_left-- > 0; + } + + bool key(std::string& /*unused*/) + { + return events_left-- > 0; + } + + bool end_object() + { + return events_left-- > 0; + } + + bool start_array(std::size_t /*unused*/) + { + return events_left-- > 0; + } + + bool end_array() + { + return events_left-- > 0; + } + + bool parse_error(std::size_t /*unused*/, const std::string& /*unused*/, const json::exception& /*unused*/) // NOLINT(readability-convert-member-functions-to-static) + { + return false; + } + + private: + int events_left = 0; +}; +} // namespace + +TEST_CASE("BJData") +{ + SECTION("individual values") + { + SECTION("discarded") + { + // discarded values are not serialized + json j = json::value_t::discarded; + const auto result = json::to_bjdata(j); + CHECK(result.empty()); + } + + SECTION("null") + { + json j = nullptr; + std::vector expected = {'Z'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("boolean") + { + SECTION("true") + { + json j = true; + std::vector expected = {'T'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("false") + { + json j = false; + std::vector expected = {'F'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("number") + { + SECTION("signed") + { + SECTION("-9223372036854775808..-2147483649 (int64)") + { + std::vector numbers; + numbers.push_back((std::numeric_limits::min)()); + numbers.push_back(-1000000000000000000LL); + numbers.push_back(-100000000000000000LL); + numbers.push_back(-10000000000000000LL); + numbers.push_back(-1000000000000000LL); + numbers.push_back(-100000000000000LL); + numbers.push_back(-10000000000000LL); + numbers.push_back(-1000000000000LL); + numbers.push_back(-100000000000LL); + numbers.push_back(-10000000000LL); + numbers.push_back(-2147483649LL); + for (auto i : numbers) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('L')); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + expected.push_back(static_cast((i >> 32) & 0xff)); + expected.push_back(static_cast((i >> 40) & 0xff)); + expected.push_back(static_cast((i >> 48) & 0xff)); + expected.push_back(static_cast((i >> 56) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 9); + + // check individual bytes + CHECK(result[0] == 'L'); + int64_t restored = (static_cast(result[8]) << 070) + + (static_cast(result[7]) << 060) + + (static_cast(result[6]) << 050) + + (static_cast(result[5]) << 040) + + (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("-2147483648..-32769 (int32)") + { + std::vector numbers; + numbers.push_back(-32769); + numbers.push_back(-100000); + numbers.push_back(-1000000); + numbers.push_back(-10000000); + numbers.push_back(-100000000); + numbers.push_back(-1000000000); + numbers.push_back(-2147483647 - 1); // https://stackoverflow.com/a/29356002/266378 + for (auto i : numbers) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('l')); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 5); + + // check individual bytes + CHECK(result[0] == 'l'); + int32_t restored = (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("-32768..-129 (int16)") + { + for (int32_t i = -32768; i <= -129; ++i) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('I')); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'I'); + auto restored = static_cast(((result[2] << 8) + result[1])); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("-9263 (int16)") + { + json j = -9263; + std::vector expected = {'I', 0xd1, 0xdb}; + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'I'); + auto restored = static_cast(((result[2] << 8) + result[1])); + CHECK(restored == -9263); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("-128..-1 (int8)") + { + for (auto i = -128; i <= -1; ++i) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back('i'); + expected.push_back(static_cast(i)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 2); + + // check individual bytes + CHECK(result[0] == 'i'); + CHECK(static_cast(result[1]) == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("0..127 (int8)") + { + for (size_t i = 0; i <= 127; ++i) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('i')); + expected.push_back(static_cast(i)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 2); + + // check individual bytes + CHECK(result[0] == 'i'); + CHECK(result[1] == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("128..255 (uint8)") + { + for (size_t i = 128; i <= 255; ++i) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('U')); + expected.push_back(static_cast(i)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 2); + + // check individual bytes + CHECK(result[0] == 'U'); + CHECK(result[1] == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("256..32767 (int16)") + { + for (size_t i = 256; i <= 32767; ++i) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('I')); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'I'); + auto restored = static_cast(static_cast(result[2]) * 256 + static_cast(result[1])); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("32768..65535 (uint16)") + { + for (uint32_t i : + { + 32768u, 55555u, 65535u + }) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('u')); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'u'); + auto restored = static_cast(static_cast(result[2]) * 256 + static_cast(result[1])); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("65536..2147483647 (int32)") + { + for (uint32_t i : + { + 65536u, 77777u, 2147483647u + }) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back('l'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 5); + + // check individual bytes + CHECK(result[0] == 'l'); + uint32_t restored = (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("2147483648..4294967295 (uint32)") + { + for (uint32_t i : + { + 2147483648u, 3333333333u, 4294967295u + }) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back('m'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 5); + + // check individual bytes + CHECK(result[0] == 'm'); + uint32_t restored = (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("4294967296..9223372036854775807 (int64)") + { + std::vector v = {4294967296LU, 9223372036854775807LU}; + for (uint64_t i : v) + { + CAPTURE(i) + + // create JSON value with integer number + json j = -1; + j.get_ref() = static_cast(i); + + // check type + CHECK(j.is_number_integer()); + + // create expected byte vector + std::vector expected; + expected.push_back('L'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 010) & 0xff)); + expected.push_back(static_cast((i >> 020) & 0xff)); + expected.push_back(static_cast((i >> 030) & 0xff)); + expected.push_back(static_cast((i >> 040) & 0xff)); + expected.push_back(static_cast((i >> 050) & 0xff)); + expected.push_back(static_cast((i >> 060) & 0xff)); + expected.push_back(static_cast((i >> 070) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 9); + + // check individual bytes + CHECK(result[0] == 'L'); + uint64_t restored = (static_cast(result[8]) << 070) + + (static_cast(result[7]) << 060) + + (static_cast(result[6]) << 050) + + (static_cast(result[5]) << 040) + + (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + } + + SECTION("unsigned") + { + SECTION("0..127 (int8)") + { + for (size_t i = 0; i <= 127; ++i) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('i'); + expected.push_back(static_cast(i)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 2); + + // check individual bytes + CHECK(result[0] == 'i'); + auto restored = static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("128..255 (uint8)") + { + for (size_t i = 128; i <= 255; ++i) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('U'); + expected.push_back(static_cast(i)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 2); + + // check individual bytes + CHECK(result[0] == 'U'); + auto restored = static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("256..32767 (int16)") + { + for (size_t i = 256; i <= 32767; ++i) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('I'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'I'); + auto restored = static_cast(static_cast(result[2]) * 256 + static_cast(result[1])); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("32768..65535 (uint16)") + { + for (uint32_t i : + { + 32768u, 55555u, 65535u + }) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('u'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 3); + + // check individual bytes + CHECK(result[0] == 'u'); + auto restored = static_cast(static_cast(result[2]) * 256 + static_cast(result[1])); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + SECTION("65536..2147483647 (int32)") + { + for (uint32_t i : + { + 65536u, 77777u, 2147483647u + }) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('l'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 5); + + // check individual bytes + CHECK(result[0] == 'l'); + uint32_t restored = (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("2147483648..4294967295 (uint32)") + { + for (uint32_t i : + { + 2147483648u, 3333333333u, 4294967295u + }) + { + CAPTURE(i) + + // create JSON value with unsigned integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('m'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 8) & 0xff)); + expected.push_back(static_cast((i >> 16) & 0xff)); + expected.push_back(static_cast((i >> 24) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 5); + + // check individual bytes + CHECK(result[0] == 'm'); + uint32_t restored = (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("4294967296..9223372036854775807 (int64)") + { + std::vector v = {4294967296ul, 9223372036854775807ul}; + for (uint64_t i : v) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('L'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 010) & 0xff)); + expected.push_back(static_cast((i >> 020) & 0xff)); + expected.push_back(static_cast((i >> 030) & 0xff)); + expected.push_back(static_cast((i >> 040) & 0xff)); + expected.push_back(static_cast((i >> 050) & 0xff)); + expected.push_back(static_cast((i >> 060) & 0xff)); + expected.push_back(static_cast((i >> 070) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 9); + + // check individual bytes + CHECK(result[0] == 'L'); + uint64_t restored = (static_cast(result[8]) << 070) + + (static_cast(result[7]) << 060) + + (static_cast(result[6]) << 050) + + (static_cast(result[5]) << 040) + + (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("9223372036854775808..18446744073709551615 (uint64)") + { + std::vector v = {9223372036854775808ull, 18446744073709551615ull}; + for (uint64_t i : v) + { + CAPTURE(i) + + // create JSON value with integer number + json j = i; + + // check type + CHECK(j.is_number_unsigned()); + + // create expected byte vector + std::vector expected; + expected.push_back('M'); + expected.push_back(static_cast(i & 0xff)); + expected.push_back(static_cast((i >> 010) & 0xff)); + expected.push_back(static_cast((i >> 020) & 0xff)); + expected.push_back(static_cast((i >> 030) & 0xff)); + expected.push_back(static_cast((i >> 040) & 0xff)); + expected.push_back(static_cast((i >> 050) & 0xff)); + expected.push_back(static_cast((i >> 060) & 0xff)); + expected.push_back(static_cast((i >> 070) & 0xff)); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == 9); + + // check individual bytes + CHECK(result[0] == 'M'); + uint64_t restored = (static_cast(result[8]) << 070) + + (static_cast(result[7]) << 060) + + (static_cast(result[6]) << 050) + + (static_cast(result[5]) << 040) + + (static_cast(result[4]) << 030) + + (static_cast(result[3]) << 020) + + (static_cast(result[2]) << 010) + + static_cast(result[1]); + CHECK(restored == i); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + } + SECTION("float64") + { + SECTION("3.1415925") + { + double v = 3.1415925; + json j = v; + std::vector expected = + { + 'D', 0xfc, 0xde, 0xa6, 0x3f, 0xfb, 0x21, 0x09, 0x40 + }; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result) == v); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("half-precision float") + { + SECTION("simple half floats") + { + CHECK(json::parse("0.0") == json::from_bjdata(std::vector({'h', 0x00, 0x00}))); + CHECK(json::parse("-0.0") == json::from_bjdata(std::vector({'h', 0x00, 0x80}))); + CHECK(json::parse("1.0") == json::from_bjdata(std::vector({'h', 0x00, 0x3c}))); + CHECK(json::parse("1.5") == json::from_bjdata(std::vector({'h', 0x00, 0x3e}))); + CHECK(json::parse("65504.0") == json::from_bjdata(std::vector({'h', 0xff, 0x7b}))); + } + } + + SECTION("half-precision float (edge cases)") + { + SECTION("exp = 0b00000") + { + SECTION("0 (0 00000 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x00})); + json::number_float_t d{j}; + CHECK(d == 0.0); + } + + SECTION("-0 (1 00000 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x80})); + json::number_float_t d{j}; + CHECK(d == -0.0); + } + + SECTION("2**-24 (0 00000 0000000001)") + { + json j = json::from_bjdata(std::vector({'h', 0x01, 0x00})); + json::number_float_t d{j}; + CHECK(d == std::pow(2.0, -24.0)); + } + } + + SECTION("exp = 0b11111") + { + SECTION("infinity (0 11111 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x7c})); + json::number_float_t d{j}; + CHECK(d == std::numeric_limits::infinity()); + CHECK(j.dump() == "null"); + } + + SECTION("-infinity (1 11111 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0xfc})); + json::number_float_t d{j}; + CHECK(d == -std::numeric_limits::infinity()); + CHECK(j.dump() == "null"); + } + } + + SECTION("other values from https://en.wikipedia.org/wiki/Half-precision_floating-point_format") + { + SECTION("1 (0 01111 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x3c})); + json::number_float_t d{j}; + CHECK(d == 1); + } + + SECTION("-2 (1 10000 0000000000)") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0xc0})); + json::number_float_t d{j}; + CHECK(d == -2); + } + + SECTION("65504 (0 11110 1111111111)") + { + json j = json::from_bjdata(std::vector({'h', 0xff, 0x7b})); + json::number_float_t d{j}; + CHECK(d == 65504); + } + } + + SECTION("infinity") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x7c})); + json::number_float_t d{j}; + CHECK(!std::isfinite(d)); + CHECK(j.dump() == "null"); + } + + SECTION("NaN") + { + json j = json::from_bjdata(std::vector({'h', 0x00, 0x7e })); + json::number_float_t d{j}; + CHECK(std::isnan(d)); + CHECK(j.dump() == "null"); + } + } + + SECTION("high-precision number") + { + SECTION("unsigned integer number") + { + std::vector vec = {'H', 'i', 0x14, '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0'}; + const auto j = json::from_bjdata(vec); + CHECK(j.is_number_unsigned()); + CHECK(j.dump() == "12345678901234567890"); + } + + SECTION("signed integer number") + { + std::vector vec = {'H', 'i', 0x13, '-', '1', '2', '3', '4', '5', '6', '7', '8', '9', '0', '1', '2', '3', '4', '5', '6', '7', '8'}; + const auto j = json::from_bjdata(vec); + CHECK(j.is_number_integer()); + CHECK(j.dump() == "-123456789012345678"); + } + + SECTION("floating-point number") + { + std::vector vec = {'H', 'i', 0x16, '3', '.', '1', '4', '1', '5', '9', '2', '6', '5', '3', '5', '8', '9', '7', '9', '3', '2', '3', '8', '4', '6'}; + const auto j = json::from_bjdata(vec); + CHECK(j.is_number_float()); + CHECK(j.dump() == "3.141592653589793"); + } + + SECTION("errors") + { + // error while parsing length + std::vector vec0 = {'H', 'i'}; + CHECK(json::from_bjdata(vec0, true, false).is_discarded()); + // error while parsing string + std::vector vec1 = {'H', 'i', '1'}; + CHECK(json::from_bjdata(vec1, true, false).is_discarded()); + + json _; + std::vector vec2 = {'H', 'i', 2, '1', 'A', '3'}; + CHECK_THROWS_WITH_AS(_ = json::from_bjdata(vec2), "[json.exception.parse_error.115] parse error at byte 5: syntax error while parsing BJData high-precision number: invalid number text: 1A", json::parse_error); + std::vector vec3 = {'H', 'i', 2, '1', '.'}; + CHECK_THROWS_WITH_AS(_ = json::from_bjdata(vec3), "[json.exception.parse_error.115] parse error at byte 5: syntax error while parsing BJData high-precision number: invalid number text: 1.", json::parse_error); + std::vector vec4 = {'H', 2, '1', '0'}; + CHECK_THROWS_WITH_AS(_ = json::from_bjdata(vec4), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing BJData size: expected length type specification (U, i, I, l, L) after '#'; last byte: 0x02", json::parse_error); + } + } + } + + SECTION("string") + { + SECTION("N = 0..127") + { + for (size_t N = 0; N <= 127; ++N) + { + CAPTURE(N) + + // create JSON value with string containing of N * 'x' + const auto s = std::string(N, 'x'); + json j = s; + + // create expected byte vector + std::vector expected; + expected.push_back('S'); + expected.push_back('i'); + expected.push_back(static_cast(N)); + for (size_t i = 0; i < N; ++i) + { + expected.push_back('x'); + } + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == N + 3); + // check that no null byte is appended + if (N > 0) + { + CHECK(result.back() != '\x00'); + } + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("N = 128..255") + { + for (size_t N = 128; N <= 255; ++N) + { + CAPTURE(N) + + // create JSON value with string containing of N * 'x' + const auto s = std::string(N, 'x'); + json j = s; + + // create expected byte vector + std::vector expected; + expected.push_back('S'); + expected.push_back('U'); + expected.push_back(static_cast(N)); + for (size_t i = 0; i < N; ++i) + { + expected.push_back('x'); + } + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == N + 3); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("N = 256..32767") + { + for (size_t N : + { + 256u, 999u, 1025u, 3333u, 2048u, 32767u + }) + { + CAPTURE(N) + + // create JSON value with string containing of N * 'x' + const auto s = std::string(N, 'x'); + json j = s; + + // create expected byte vector (hack: create string first) + std::vector expected(N, 'x'); + // reverse order of commands, because we insert at begin() + expected.insert(expected.begin(), static_cast((N >> 8) & 0xff)); + expected.insert(expected.begin(), static_cast(N & 0xff)); + expected.insert(expected.begin(), 'I'); + expected.insert(expected.begin(), 'S'); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == N + 4); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("N = 32768..65535") + { + for (size_t N : + { + 32768u, 55555u, 65535u + }) + { + CAPTURE(N) + + // create JSON value with string containing of N * 'x' + const auto s = std::string(N, 'x'); + json j = s; + + // create expected byte vector (hack: create string first) + std::vector expected(N, 'x'); + // reverse order of commands, because we insert at begin() + expected.insert(expected.begin(), static_cast((N >> 8) & 0xff)); + expected.insert(expected.begin(), static_cast(N & 0xff)); + expected.insert(expected.begin(), 'u'); + expected.insert(expected.begin(), 'S'); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == N + 4); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("N = 65536..2147483647") + { + for (size_t N : + { + 65536u, 77777u, 1048576u + }) + { + CAPTURE(N) + + // create JSON value with string containing of N * 'x' + const auto s = std::string(N, 'x'); + json j = s; + + // create expected byte vector (hack: create string first) + std::vector expected(N, 'x'); + // reverse order of commands, because we insert at begin() + expected.insert(expected.begin(), static_cast((N >> 24) & 0xff)); + expected.insert(expected.begin(), static_cast((N >> 16) & 0xff)); + expected.insert(expected.begin(), static_cast((N >> 8) & 0xff)); + expected.insert(expected.begin(), static_cast(N & 0xff)); + expected.insert(expected.begin(), 'l'); + expected.insert(expected.begin(), 'S'); + + // compare result + size + const auto result = json::to_bjdata(j); + CHECK(result == expected); + CHECK(result.size() == N + 6); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + } + + + SECTION("binary") + { + SECTION("N = 0..127") + { + for (std::size_t N = 0; N <= 127; ++N) + { + CAPTURE(N) + + // create JSON value with byte array containing of N * 'x' + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('[')); + if (N != 0) + { + expected.push_back(static_cast('$')); + expected.push_back(static_cast('U')); + } + expected.push_back(static_cast('#')); + expected.push_back(static_cast('i')); + expected.push_back(static_cast(N)); + for (size_t i = 0; i < N; ++i) + { + expected.push_back(0x78); + } + + // compare result + size + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + if (N == 0) + { + CHECK(result.size() == N + 4); + } + else + { + CHECK(result.size() == N + 6); + } + + // check that no null byte is appended + if (N > 0) + { + CHECK(result.back() != '\x00'); + } + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + + SECTION("N = 128..255") + { + for (std::size_t N = 128; N <= 255; ++N) + { + CAPTURE(N) + + // create JSON value with byte array containing of N * 'x' + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + // create expected byte vector + std::vector expected; + expected.push_back(static_cast('[')); + expected.push_back(static_cast('$')); + expected.push_back(static_cast('U')); + expected.push_back(static_cast('#')); + expected.push_back(static_cast('U')); + expected.push_back(static_cast(N)); + for (size_t i = 0; i < N; ++i) + { + expected.push_back(0x78); + } + + // compare result + size + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + CHECK(result.size() == N + 6); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + + SECTION("N = 256..32767") + { + for (std::size_t N : + { + 256u, 999u, 1025u, 3333u, 2048u, 32767u + }) + { + CAPTURE(N) + + // create JSON value with byte array containing of N * 'x' + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + // create expected byte vector + std::vector expected(N + 7, 'x'); + expected[0] = '['; + expected[1] = '$'; + expected[2] = 'U'; + expected[3] = '#'; + expected[4] = 'I'; + expected[5] = static_cast(N & 0xFF); + expected[6] = static_cast((N >> 8) & 0xFF); + + // compare result + size + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + CHECK(result.size() == N + 7); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + + SECTION("N = 32768..65535") + { + for (std::size_t N : + { + 32768u, 55555u, 65535u + }) + { + CAPTURE(N) + + // create JSON value with byte array containing of N * 'x' + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + // create expected byte vector + std::vector expected(N + 7, 'x'); + expected[0] = '['; + expected[1] = '$'; + expected[2] = 'U'; + expected[3] = '#'; + expected[4] = 'u'; + expected[5] = static_cast(N & 0xFF); + expected[6] = static_cast((N >> 8) & 0xFF); + + // compare result + size + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + CHECK(result.size() == N + 7); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + + SECTION("N = 65536..2147483647") + { + for (std::size_t N : + { + 65536u, 77777u, 1048576u + }) + { + CAPTURE(N) + + // create JSON value with byte array containing of N * 'x' + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + // create expected byte vector + std::vector expected(N + 9, 'x'); + expected[0] = '['; + expected[1] = '$'; + expected[2] = 'U'; + expected[3] = '#'; + expected[4] = 'l'; + expected[5] = static_cast(N & 0xFF); + expected[6] = static_cast((N >> 8) & 0xFF); + expected[7] = static_cast((N >> 16) & 0xFF); + expected[8] = static_cast((N >> 24) & 0xFF); + + // compare result + size + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + CHECK(result.size() == N + 9); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + + SECTION("Other Serializations") + { + const std::size_t N = 10; + const auto s = std::vector(N, 'x'); + json j = json::binary(s); + + SECTION("No Count No Type") + { + std::vector expected; + expected.push_back(static_cast('[')); + for (std::size_t i = 0; i < N; ++i) + { + expected.push_back(static_cast('U')); + expected.push_back(static_cast(0x78)); + } + expected.push_back(static_cast(']')); + + // compare result + size + const auto result = json::to_bjdata(j, false, false); + CHECK(result == expected); + CHECK(result.size() == N + 12); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + + SECTION("Yes Count No Type") + { + std::vector expected; + expected.push_back(static_cast('[')); + expected.push_back(static_cast('#')); + expected.push_back(static_cast('i')); + expected.push_back(static_cast(N)); + + for (size_t i = 0; i < N; ++i) + { + expected.push_back(static_cast('U')); + expected.push_back(static_cast(0x78)); + } + + // compare result + size + const auto result = json::to_bjdata(j, true, false); + CHECK(result == expected); + CHECK(result.size() == N + 14); + // check that no null byte is appended + CHECK(result.back() != '\x00'); + + // roundtrip only works to an array of numbers + json j_out = s; + CHECK(json::from_bjdata(result) == j_out); + CHECK(json::from_bjdata(result, true, false) == j_out); + } + } + } + SECTION("array") + { + SECTION("empty") + { + SECTION("size=false type=false") + { + json j = json::array(); + std::vector expected = {'[', ']'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = json::array(); + std::vector expected = {'[', '#', 'i', 0}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = json::array(); + std::vector expected = {'[', '#', 'i', 0}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("[null]") + { + SECTION("size=false type=false") + { + json j = {nullptr}; + std::vector expected = {'[', 'Z', ']'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = {nullptr}; + std::vector expected = {'[', '#', 'i', 1, 'Z'}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = {nullptr}; + std::vector expected = {'[', '$', 'Z', '#', 'i', 1}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("[1,2,3,4,5]") + { + SECTION("size=false type=false") + { + json j = json::parse("[1,2,3,4,5]"); + std::vector expected = {'[', 'i', 1, 'i', 2, 'i', 3, 'i', 4, 'i', 5, ']'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = json::parse("[1,2,3,4,5]"); + std::vector expected = {'[', '#', 'i', 5, 'i', 1, 'i', 2, 'i', 3, 'i', 4, 'i', 5}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = json::parse("[1,2,3,4,5]"); + std::vector expected = {'[', '$', 'i', '#', 'i', 5, 1, 2, 3, 4, 5}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("[[[[]]]]") + { + SECTION("size=false type=false") + { + json j = json::parse("[[[[]]]]"); + std::vector expected = {'[', '[', '[', '[', ']', ']', ']', ']'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = json::parse("[[[[]]]]"); + std::vector expected = {'[', '#', 'i', 1, '[', '#', 'i', 1, '[', '#', 'i', 1, '[', '#', 'i', 0}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = json::parse("[[[[]]]]"); + std::vector expected = {'[', '$', '[', '#', 'i', 1, '$', '[', '#', 'i', 1, '$', '[', '#', 'i', 1, '#', 'i', 0}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("array with uint16_t elements") + { + SECTION("size=false type=false") + { + json j(257, nullptr); + std::vector expected(j.size() + 2, 'Z'); // all null + expected[0] = '['; // opening array + expected[258] = ']'; // closing array + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j(257, nullptr); + std::vector expected(j.size() + 5, 'Z'); // all null + expected[0] = '['; // opening array + expected[1] = '#'; // array size + expected[2] = 'I'; // int16 + expected[3] = 0x01; // 0x0101, first byte + expected[4] = 0x01; // 0x0101, second byte + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j(257, nullptr); + std::vector expected = {'[', '$', 'Z', '#', 'I', 0x01, 0x01}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("array with uint32_t elements") + { + SECTION("size=false type=false") + { + json j(65793, nullptr); + std::vector expected(j.size() + 2, 'Z'); // all null + expected[0] = '['; // opening array + expected[65794] = ']'; // closing array + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j(65793, nullptr); + std::vector expected(j.size() + 7, 'Z'); // all null + expected[0] = '['; // opening array + expected[1] = '#'; // array size + expected[2] = 'l'; // int32 + expected[3] = 0x01; // 0x00010101, fourth byte + expected[4] = 0x01; // 0x00010101, third byte + expected[5] = 0x01; // 0x00010101, second byte + expected[6] = 0x00; // 0x00010101, first byte + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j(65793, nullptr); + std::vector expected = {'[', '$', 'Z', '#', 'l', 0x01, 0x01, 0x01, 0x00}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + } + + SECTION("object") + { + SECTION("empty") + { + SECTION("size=false type=false") + { + json j = json::object(); + std::vector expected = {'{', '}'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = json::object(); + std::vector expected = {'{', '#', 'i', 0}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = json::object(); + std::vector expected = {'{', '#', 'i', 0}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("{\"\":null}") + { + SECTION("size=false type=false") + { + json j = {{"", nullptr}}; + std::vector expected = {'{', 'i', 0, 'Z', '}'}; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = {{"", nullptr}}; + std::vector expected = {'{', '#', 'i', 1, 'i', 0, 'Z'}; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = {{"", nullptr}}; + std::vector expected = {'{', '$', 'Z', '#', 'i', 1, 'i', 0}; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + + SECTION("{\"a\": {\"b\": {\"c\": {}}}}") + { + SECTION("size=false type=false") + { + json j = json::parse(R"({"a": {"b": {"c": {}}}})"); + std::vector expected = + { + '{', 'i', 1, 'a', '{', 'i', 1, 'b', '{', 'i', 1, 'c', '{', '}', '}', '}', '}' + }; + const auto result = json::to_bjdata(j); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=false") + { + json j = json::parse(R"({"a": {"b": {"c": {}}}})"); + std::vector expected = + { + '{', '#', 'i', 1, 'i', 1, 'a', '{', '#', 'i', 1, 'i', 1, 'b', '{', '#', 'i', 1, 'i', 1, 'c', '{', '#', 'i', 0 + }; + const auto result = json::to_bjdata(j, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + + SECTION("size=true type=true") + { + json j = json::parse(R"({"a": {"b": {"c": {}}}})"); + std::vector expected = + { + '{', '$', '{', '#', 'i', 1, 'i', 1, 'a', '$', '{', '#', 'i', 1, 'i', 1, 'b', '$', '{', '#', 'i', 1, 'i', 1, 'c', '#', 'i', 0 + }; + const auto result = json::to_bjdata(j, true, true); + CHECK(result == expected); + + // roundtrip + CHECK(json::from_bjdata(result) == j); + CHECK(json::from_bjdata(result, true, false) == j); + } + } + } + } + + SECTION("errors") + { + SECTION("strict mode") + { + std::vector vec = {'Z', 'Z'}; + SECTION("non-strict mode") + { + const auto result = json::from_bjdata(vec, false); + CHECK(result == json()); + } + + SECTION("strict mode") + { + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(vec), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vec), + "[json.exception.parse_error.110] parse error at byte 2: syntax error while parsing BJData value: expected end of input; last byte: 0x5A"); + } + } + + SECTION("excessive size") + { + SECTION("array") + { + std::vector v_bjdata = {'[', '$', 'Z', '#', 'L', 0x78, 0x28, 0x00, 0x68, 0x28, 0x69, 0x69, 0x17}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v_bjdata), json::out_of_range&); + + json j; + nlohmann::detail::json_sax_dom_callback_parser scp(j, [](int /*unused*/, json::parse_event_t /*unused*/, const json& /*unused*/) + { + return true; + }); + CHECK_THROWS_AS(_ = json::sax_parse(v_bjdata, &scp, json::input_format_t::bjdata), json::out_of_range&); + } + + SECTION("object") + { + std::vector v_bjdata = {'{', '$', 'Z', '#', 'L', 0x78, 0x28, 0x00, 0x68, 0x28, 0x69, 0x69, 0x17}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v_bjdata), json::out_of_range&); + + json j; + nlohmann::detail::json_sax_dom_callback_parser scp(j, [](int /*unused*/, json::parse_event_t /*unused*/, const json& /*unused*/) + { + return true; + }); + CHECK_THROWS_AS(_ = json::sax_parse(v_bjdata, &scp, json::input_format_t::bjdata), json::out_of_range&); + } + } + } + + SECTION("SAX aborts") + { + SECTION("start_array()") + { + std::vector v = {'[', 'T', 'F', ']'}; + SaxCountdown scp(0); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + + SECTION("start_object()") + { + std::vector v = {'{', 'i', 3, 'f', 'o', 'o', 'F', '}'}; + SaxCountdown scp(0); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + + SECTION("key() in object") + { + std::vector v = {'{', 'i', 3, 'f', 'o', 'o', 'F', '}'}; + SaxCountdown scp(1); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + + SECTION("start_array(len)") + { + std::vector v = {'[', '#', 'i', '2', 'T', 'F'}; + SaxCountdown scp(0); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + + SECTION("start_object(len)") + { + std::vector v = {'{', '#', 'i', '1', 3, 'f', 'o', 'o', 'F'}; + SaxCountdown scp(0); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + + SECTION("key() in object with length") + { + std::vector v = {'{', 'i', 3, 'f', 'o', 'o', 'F', '}'}; + SaxCountdown scp(1); + CHECK(!json::sax_parse(v, &scp, json::input_format_t::bjdata)); + } + } + + SECTION("parsing values") + { + SECTION("strings") + { + // create a single-character string for all number types + std::vector s_i = {'S', 'i', 1, 'a'}; + std::vector s_U = {'S', 'U', 1, 'a'}; + std::vector s_I = {'S', 'I', 1, 0, 'a'}; + std::vector s_l = {'S', 'l', 1, 0, 0, 0, 'a'}; + std::vector s_L = {'S', 'L', 1, 0, 0, 0, 0, 0, 0, 0, 'a'}; + + // check if string is parsed correctly to "a" + CHECK(json::from_bjdata(s_i) == "a"); + CHECK(json::from_bjdata(s_U) == "a"); + CHECK(json::from_bjdata(s_I) == "a"); + CHECK(json::from_bjdata(s_l) == "a"); + CHECK(json::from_bjdata(s_L) == "a"); + + // roundtrip: output should be optimized + CHECK(json::to_bjdata(json::from_bjdata(s_i)) == s_i); + CHECK(json::to_bjdata(json::from_bjdata(s_U)) == s_i); + CHECK(json::to_bjdata(json::from_bjdata(s_I)) == s_i); + CHECK(json::to_bjdata(json::from_bjdata(s_l)) == s_i); + CHECK(json::to_bjdata(json::from_bjdata(s_L)) == s_i); + } + + SECTION("number") + { + SECTION("float") + { + // float32 + std::vector v_d = {'d', 0xd0, 0x0f, 0x49, 0x40}; + CHECK(json::from_bjdata(v_d) == 3.14159f); + + // float64 + std::vector v_D = {'D', 0x6e, 0x86, 0x1b, 0xf0, 0xf9, 0x21, 0x09, 0x40}; + CHECK(json::from_bjdata(v_D) == 3.14159); + + // float32 is serialized as float64 as the library does not support float32 + CHECK(json::to_bjdata(json::from_bjdata(v_d)) == json::to_bjdata(3.14159f)); + } + } + + SECTION("array") + { + SECTION("optimized version (length only)") + { + // create vector with two elements of the same type + std::vector v_TU = {'[', '#', 'U', 2, 'T', 'T'}; + std::vector v_T = {'[', '#', 'i', 2, 'T', 'T'}; + std::vector v_F = {'[', '#', 'i', 2, 'F', 'F'}; + std::vector v_Z = {'[', '#', 'i', 2, 'Z', 'Z'}; + std::vector v_i = {'[', '#', 'i', 2, 'i', 0x7F, 'i', 0x7F}; + std::vector v_U = {'[', '#', 'i', 2, 'U', 0xFF, 'U', 0xFF}; + std::vector v_I = {'[', '#', 'i', 2, 'I', 0xFF, 0x7F, 'I', 0xFF, 0x7F}; + std::vector v_l = {'[', '#', 'i', 2, 'l', 0xFF, 0xFF, 0xFF, 0x7F, 'l', 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_L = {'[', '#', 'i', 2, 'L', 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F, 'L', 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_D = {'[', '#', 'i', 2, 'D', 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40, 'D', 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40}; + std::vector v_S = {'[', '#', 'i', 2, 'S', 'i', 1, 'a', 'S', 'i', 1, 'a'}; + std::vector v_C = {'[', '#', 'i', 2, 'C', 'a', 'C', 'a'}; + + // check if vector is parsed correctly + CHECK(json::from_bjdata(v_TU) == json({true, true})); + CHECK(json::from_bjdata(v_T) == json({true, true})); + CHECK(json::from_bjdata(v_F) == json({false, false})); + CHECK(json::from_bjdata(v_Z) == json({nullptr, nullptr})); + CHECK(json::from_bjdata(v_i) == json({127, 127})); + CHECK(json::from_bjdata(v_U) == json({255, 255})); + CHECK(json::from_bjdata(v_I) == json({32767, 32767})); + CHECK(json::from_bjdata(v_l) == json({2147483647, 2147483647})); + CHECK(json::from_bjdata(v_L) == json({9223372036854775807, 9223372036854775807})); + CHECK(json::from_bjdata(v_D) == json({3.1415926, 3.1415926})); + CHECK(json::from_bjdata(v_S) == json({"a", "a"})); + CHECK(json::from_bjdata(v_C) == json({"a", "a"})); + + // roundtrip: output should be optimized + CHECK(json::to_bjdata(json::from_bjdata(v_T), true) == v_T); + CHECK(json::to_bjdata(json::from_bjdata(v_F), true) == v_F); + CHECK(json::to_bjdata(json::from_bjdata(v_Z), true) == v_Z); + CHECK(json::to_bjdata(json::from_bjdata(v_i), true) == v_i); + CHECK(json::to_bjdata(json::from_bjdata(v_U), true) == v_U); + CHECK(json::to_bjdata(json::from_bjdata(v_I), true) == v_I); + CHECK(json::to_bjdata(json::from_bjdata(v_l), true) == v_l); + CHECK(json::to_bjdata(json::from_bjdata(v_L), true) == v_L); + CHECK(json::to_bjdata(json::from_bjdata(v_D), true) == v_D); + CHECK(json::to_bjdata(json::from_bjdata(v_S), true) == v_S); + CHECK(json::to_bjdata(json::from_bjdata(v_C), true) == v_S); // char is serialized to string + } + + SECTION("optimized version (type and length)") + { + // create vector with two elements of the same type + std::vector v_N = {'[', '$', 'N', '#', 'i', 2}; + std::vector v_T = {'[', '$', 'T', '#', 'i', 2}; + std::vector v_F = {'[', '$', 'F', '#', 'i', 2}; + std::vector v_Z = {'[', '$', 'Z', '#', 'i', 2}; + std::vector v_i = {'[', '$', 'i', '#', 'i', 2, 0x7F, 0x7F}; + std::vector v_U = {'[', '$', 'U', '#', 'i', 2, 0xFF, 0xFF}; + std::vector v_I = {'[', '$', 'I', '#', 'i', 2, 0xFF, 0x7F, 0xFF, 0x7F}; + std::vector v_l = {'[', '$', 'l', '#', 'i', 2, 0xFF, 0xFF, 0xFF, 0x7F, 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_L = {'[', '$', 'L', '#', 'i', 2, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_D = {'[', '$', 'D', '#', 'i', 2, 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40, 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40}; + std::vector v_S = {'[', '$', 'S', '#', 'i', 2, 'i', 1, 'a', 'i', 1, 'a'}; + std::vector v_C = {'[', '$', 'C', '#', 'i', 2, 'a', 'a'}; + + // check if vector is parsed correctly + CHECK(json::from_bjdata(v_N) == json::array()); + CHECK(json::from_bjdata(v_T) == json({true, true})); + CHECK(json::from_bjdata(v_F) == json({false, false})); + CHECK(json::from_bjdata(v_Z) == json({nullptr, nullptr})); + CHECK(json::from_bjdata(v_i) == json({127, 127})); + CHECK(json::from_bjdata(v_U) == json({255, 255})); + CHECK(json::from_bjdata(v_I) == json({32767, 32767})); + CHECK(json::from_bjdata(v_l) == json({2147483647, 2147483647})); + CHECK(json::from_bjdata(v_L) == json({9223372036854775807, 9223372036854775807})); + CHECK(json::from_bjdata(v_D) == json({3.1415926, 3.1415926})); + CHECK(json::from_bjdata(v_S) == json({"a", "a"})); + CHECK(json::from_bjdata(v_C) == json({"a", "a"})); + + // roundtrip: output should be optimized + std::vector v_empty = {'[', '#', 'i', 0}; + CHECK(json::to_bjdata(json::from_bjdata(v_N), true, true) == v_empty); + CHECK(json::to_bjdata(json::from_bjdata(v_T), true, true) == v_T); + CHECK(json::to_bjdata(json::from_bjdata(v_F), true, true) == v_F); + CHECK(json::to_bjdata(json::from_bjdata(v_Z), true, true) == v_Z); + CHECK(json::to_bjdata(json::from_bjdata(v_i), true, true) == v_i); + CHECK(json::to_bjdata(json::from_bjdata(v_U), true, true) == v_U); + CHECK(json::to_bjdata(json::from_bjdata(v_I), true, true) == v_I); + CHECK(json::to_bjdata(json::from_bjdata(v_l), true, true) == v_l); + CHECK(json::to_bjdata(json::from_bjdata(v_L), true, true) == v_L); + CHECK(json::to_bjdata(json::from_bjdata(v_D), true, true) == v_D); + CHECK(json::to_bjdata(json::from_bjdata(v_S), true, true) == v_S); + CHECK(json::to_bjdata(json::from_bjdata(v_C), true, true) == v_S); // char is serialized to string + } + + SECTION("optimized ndarray (type and vector-size)") + { + // create vector with two elements of the same type + std::vector v_N = {'[', '$', 'N', '#', '[', '$', 'i', '#', 'i', 2, 1, 2}; + std::vector v_T = {'[', '$', 'T', '#', '[', '$', 'i', '#', 'i', 2, 1, 2}; + std::vector v_F = {'[', '$', 'F', '#', '[', '$', 'i', '#', 'i', 2, 1, 2}; + std::vector v_Z = {'[', '$', 'Z', '#', '[', '$', 'i', '#', 'i', 2, 1, 2}; + std::vector v_i = {'[', '$', 'i', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0x7F, 0x7F}; + std::vector v_U = {'[', '$', 'U', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0xFF, 0xFF}; + std::vector v_I = {'[', '$', 'I', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0xFF, 0x7F, 0xFF, 0x7F}; + std::vector v_l = {'[', '$', 'l', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0xFF, 0xFF, 0xFF, 0x7F, 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_L = {'[', '$', 'L', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F}; + std::vector v_D = {'[', '$', 'D', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40, 0x4a, 0xd8, 0x12, 0x4d, 0xfb, 0x21, 0x09, 0x40}; + std::vector v_S = {'[', '$', 'S', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 'i', 1, 'a', 'i', 1, 'a'}; + std::vector v_C = {'[', '$', 'C', '#', '[', '$', 'i', '#', 'i', 2, 1, 2, 'a', 'a'}; + + // check if vector is parsed correctly + CHECK(json::from_bjdata(v_N) == json::array()); + CHECK(json::from_bjdata(v_T) == json({true, true})); + CHECK(json::from_bjdata(v_F) == json({false, false})); + CHECK(json::from_bjdata(v_Z) == json({nullptr, nullptr})); + CHECK(json::from_bjdata(v_i) == json({127, 127})); + CHECK(json::from_bjdata(v_U) == json({255, 255})); + CHECK(json::from_bjdata(v_I) == json({32767, 32767})); + CHECK(json::from_bjdata(v_l) == json({2147483647, 2147483647})); + CHECK(json::from_bjdata(v_L) == json({9223372036854775807, 9223372036854775807})); + CHECK(json::from_bjdata(v_D) == json({3.1415926, 3.1415926})); + CHECK(json::from_bjdata(v_S) == json({"a", "a"})); + CHECK(json::from_bjdata(v_C) == json({"a", "a"})); + +#ifdef BJDATA_TEST_ROUNDTRIP // round-trip to vectorized size (ndarray) is not yet supported + // roundtrip: output should be optimized + std::vector v_empty = {'[', '#', 'i', 0}; + CHECK(json::to_bjdata(json::from_bjdata(v_N), true, true) == v_empty); + CHECK(json::to_bjdata(json::from_bjdata(v_T), true, true) == v_T); + CHECK(json::to_bjdata(json::from_bjdata(v_F), true, true) == v_F); + CHECK(json::to_bjdata(json::from_bjdata(v_Z), true, true) == v_Z); + CHECK(json::to_bjdata(json::from_bjdata(v_i), true, true) == v_i); + CHECK(json::to_bjdata(json::from_bjdata(v_U), true, true) == v_U); + CHECK(json::to_bjdata(json::from_bjdata(v_I), true, true) == v_I); + CHECK(json::to_bjdata(json::from_bjdata(v_l), true, true) == v_l); + CHECK(json::to_bjdata(json::from_bjdata(v_L), true, true) == v_L); + CHECK(json::to_bjdata(json::from_bjdata(v_D), true, true) == v_D); + CHECK(json::to_bjdata(json::from_bjdata(v_S), true, true) == v_S); + CHECK(json::to_bjdata(json::from_bjdata(v_C), true, true) == v_S); // char is serialized to string +#endif + + } + } + } + + SECTION("parse errors") + { + SECTION("empty byte vector") + { + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(std::vector()), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(std::vector()), + "[json.exception.parse_error.110] parse error at byte 1: syntax error while parsing BJData value: unexpected end of input"); + } + + SECTION("char") + { + SECTION("eof after C byte") + { + std::vector v = {'C'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.110] parse error at byte 2: syntax error while parsing BJData char: unexpected end of input"); + } + + SECTION("byte out of range") + { + std::vector v = {'C', 130}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing BJData char: byte after 'C' must be in range 0x00..0x7F; last byte: 0x82"); + } + } + + SECTION("strings") + { + SECTION("eof after S byte") + { + std::vector v = {'S'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.110] parse error at byte 2: syntax error while parsing BJData value: unexpected end of input"); + } + + SECTION("invalid byte") + { + std::vector v = {'S', '1', 'a'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing BJData string: expected length type specification (U, i, I, l, L); last byte: 0x31"); + } + } + + SECTION("array") + { + SECTION("optimized array: no size following type") + { + std::vector v = {'[', '$', 'i', 2}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.112] parse error at byte 4: syntax error while parsing BJData size: expected '#' after type information; last byte: 0x02"); + } + } + + SECTION("strings") + { + std::vector vS = {'S'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(vS), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vS), "[json.exception.parse_error.110] parse error at byte 2: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vS, true, false).is_discarded()); + + std::vector v = {'S', 'i', '2', 'a'}; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.110] parse error at byte 5: syntax error while parsing BJData string: unexpected end of input"); + CHECK(json::from_bjdata(v, true, false).is_discarded()); + + std::vector vC = {'C'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vC), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vC), "[json.exception.parse_error.110] parse error at byte 2: syntax error while parsing BJData char: unexpected end of input"); + CHECK(json::from_bjdata(vC, true, false).is_discarded()); + } + + SECTION("sizes") + { + std::vector vU = {'[', '#', 'U'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(vU), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vU), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vU, true, false).is_discarded()); + + std::vector vi = {'[', '#', 'i'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vi), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vi), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vi, true, false).is_discarded()); + + std::vector vI = {'[', '#', 'I'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vI), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vI), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vI, true, false).is_discarded()); + + std::vector vl = {'[', '#', 'l'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vl), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vl), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vl, true, false).is_discarded()); + + std::vector vL = {'[', '#', 'L'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vL), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vL), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vL, true, false).is_discarded()); + + std::vector v0 = {'[', '#', 'T', ']'}; + CHECK_THROWS_AS(_ = json::from_bjdata(v0), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v0), "[json.exception.parse_error.113] parse error at byte 3: syntax error while parsing BJData size: expected length type specification (U, i, I, l, L) after '#'; last byte: 0x54"); + CHECK(json::from_bjdata(v0, true, false).is_discarded()); + } + + SECTION("types") + { + std::vector v0 = {'[', '$'}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(v0), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v0), "[json.exception.parse_error.110] parse error at byte 3: syntax error while parsing BJData type: unexpected end of input"); + CHECK(json::from_bjdata(v0, true, false).is_discarded()); + + std::vector vi = {'[', '$', '#'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vi), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vi), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vi, true, false).is_discarded()); + + std::vector vT = {'[', '$', 'T'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vT), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vT), "[json.exception.parse_error.110] parse error at byte 4: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vT, true, false).is_discarded()); + } + + SECTION("arrays") + { + std::vector vST = {'[', '$', 'i', '#', 'i', 2, 1}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(vST), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vST), "[json.exception.parse_error.110] parse error at byte 8: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vST, true, false).is_discarded()); + + std::vector vS = {'[', '#', 'i', 2, 'i', 1}; + CHECK_THROWS_AS(_ = json::from_bjdata(vS), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vS), "[json.exception.parse_error.110] parse error at byte 7: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vS, true, false).is_discarded()); + + std::vector v = {'[', 'i', 2, 'i', 1}; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.110] parse error at byte 6: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(v, true, false).is_discarded()); + } + + SECTION("objects") + { + std::vector vST = {'{', '$', 'i', '#', 'i', 2, 'i', 1, 'a', 1}; + json _; + CHECK_THROWS_AS(_ = json::from_bjdata(vST), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vST), "[json.exception.parse_error.110] parse error at byte 11: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vST, true, false).is_discarded()); + + std::vector vT = {'{', '$', 'i', 'i', 1, 'a', 1}; + CHECK_THROWS_AS(_ = json::from_bjdata(vT), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vT), "[json.exception.parse_error.112] parse error at byte 4: syntax error while parsing BJData size: expected '#' after type information; last byte: 0x69"); + CHECK(json::from_bjdata(vT, true, false).is_discarded()); + + std::vector vS = {'{', '#', 'i', 2, 'i', 1, 'a', 'i', 1}; + CHECK_THROWS_AS(_ = json::from_bjdata(vS), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vS), "[json.exception.parse_error.110] parse error at byte 10: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vS, true, false).is_discarded()); + + std::vector v = {'{', 'i', 1, 'a', 'i', 1}; + CHECK_THROWS_AS(_ = json::from_bjdata(v), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v), "[json.exception.parse_error.110] parse error at byte 7: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(v, true, false).is_discarded()); + + std::vector v2 = {'{', 'i', 1, 'a', 'i', 1, 'i'}; + CHECK_THROWS_AS(_ = json::from_bjdata(v2), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v2), "[json.exception.parse_error.110] parse error at byte 8: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(v2, true, false).is_discarded()); + + std::vector v3 = {'{', 'i', 1, 'a'}; + CHECK_THROWS_AS(_ = json::from_bjdata(v3), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(v3), "[json.exception.parse_error.110] parse error at byte 5: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(v3, true, false).is_discarded()); + + std::vector vST1 = {'{', '$', 'd', '#', 'i', 2, 'i', 1, 'a'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vST1), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vST1), "[json.exception.parse_error.110] parse error at byte 10: syntax error while parsing BJData number: unexpected end of input"); + CHECK(json::from_bjdata(vST1, true, false).is_discarded()); + + std::vector vST2 = {'{', '#', 'i', 2, 'i', 1, 'a'}; + CHECK_THROWS_AS(_ = json::from_bjdata(vST2), json::parse_error&); + CHECK_THROWS_WITH(_ = json::from_bjdata(vST2), "[json.exception.parse_error.110] parse error at byte 8: syntax error while parsing BJData value: unexpected end of input"); + CHECK(json::from_bjdata(vST2, true, false).is_discarded()); + } + } + + SECTION("writing optimized values") + { + SECTION("integer") + { + SECTION("array of i") + { + json j = {1, -1}; + std::vector expected = {'[', '$', 'i', '#', 'i', 2, 1, 0xff}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + + SECTION("array of U") + { + json j = {200, 201}; + std::vector expected = {'[', '$', 'U', '#', 'i', 2, 0xC8, 0xC9}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + + SECTION("array of I") + { + json j = {30000, -30000}; + std::vector expected = {'[', '$', 'I', '#', 'i', 2, 0x30, 0x75, 0xd0, 0x8a}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + + SECTION("array of l") + { + json j = {70000, -70000}; + std::vector expected = {'[', '$', 'l', '#', 'i', 2, 0x70, 0x11, 0x01, 0x00, 0x90, 0xEE, 0xFE, 0xFF}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + + SECTION("array of L") + { + json j = {5000000000, -5000000000}; + std::vector expected = {'[', '$', 'L', '#', 'i', 2, 0x00, 0xF2, 0x05, 0x2A, 0x01, 0x00, 0x00, 0x00, 0x00, 0x0E, 0xFA, 0xD5, 0xFE, 0xFF, 0xFF, 0xFF}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + } + + SECTION("unsigned integer") + { + SECTION("array of i") + { + json j = {1u, 2u}; + std::vector expected = {'[', '$', 'i', '#', 'i', 2, 1, 2}; + std::vector expected_size = {'[', '#', 'i', 2, 'i', 1, 'i', 2}; + CHECK(json::to_bjdata(j, true, true) == expected); + CHECK(json::to_bjdata(j, true) == expected_size); + } + + SECTION("array of U") + { + json j = {200u, 201u}; + std::vector expected = {'[', '$', 'U', '#', 'i', 2, 0xC8, 0xC9}; + std::vector expected_size = {'[', '#', 'i', 2, 'U', 0xC8, 'U', 0xC9}; + CHECK(json::to_bjdata(j, true, true) == expected); + CHECK(json::to_bjdata(j, true) == expected_size); + } + + SECTION("array of I") + { + json j = {30000u, 30001u}; + std::vector expected = {'[', '$', 'I', '#', 'i', 2, 0x30, 0x75, 0x31, 0x75}; + std::vector expected_size = {'[', '#', 'i', 2, 'I', 0x30, 0x75, 'I', 0x31, 0x75}; + CHECK(json::to_bjdata(j, true, true) == expected); + CHECK(json::to_bjdata(j, true) == expected_size); + } + + SECTION("array of l") + { + json j = {70000u, 70001u}; + std::vector expected = {'[', '$', 'l', '#', 'i', 2, 0x70, 0x11, 0x01, 0x00, 0x71, 0x11, 0x01, 0x00}; + std::vector expected_size = {'[', '#', 'i', 2, 'l', 0x70, 0x11, 0x01, 0x00, 'l', 0x71, 0x11, 0x01, 0x00}; + CHECK(json::to_bjdata(j, true, true) == expected); + CHECK(json::to_bjdata(j, true) == expected_size); + } + + SECTION("array of L") + { + json j = {5000000000u, 5000000001u}; + std::vector expected = {'[', '$', 'L', '#', 'i', 2, 0x00, 0xF2, 0x05, 0x2A, 0x01, 0x00, 0x00, 0x00, 0x01, 0xF2, 0x05, 0x2A, 0x01, 0x00, 0x00, 0x00}; + std::vector expected_size = {'[', '#', 'i', 2, 'L', 0x00, 0xF2, 0x05, 0x2A, 0x01, 0x00, 0x00, 0x00, 'L', 0x01, 0xF2, 0x05, 0x2A, 0x01, 0x00, 0x00, 0x00}; + CHECK(json::to_bjdata(j, true, true) == expected); + CHECK(json::to_bjdata(j, true) == expected_size); + } + } + + SECTION("discarded") + { + json j = {json::value_t::discarded, json::value_t::discarded}; + std::vector expected = {'[', '$', 'N', '#', 'i', 2}; + CHECK(json::to_bjdata(j, true, true) == expected); + } + } +} + +TEST_CASE("Universal Binary JSON Specification Examples 1") +{ + SECTION("Null Value") + { + json j = {{"passcode", nullptr}}; + std::vector v = {'{', 'i', 8, 'p', 'a', 's', 's', 'c', 'o', 'd', 'e', 'Z', '}'}; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("No-Op Value") + { + json j = {"foo", "bar", "baz"}; + std::vector v = {'[', 'S', 'i', 3, 'f', 'o', 'o', + 'S', 'i', 3, 'b', 'a', 'r', + 'S', 'i', 3, 'b', 'a', 'z', ']' + }; + std::vector v2 = {'[', 'S', 'i', 3, 'f', 'o', 'o', 'N', + 'S', 'i', 3, 'b', 'a', 'r', 'N', 'N', 'N', + 'S', 'i', 3, 'b', 'a', 'z', 'N', 'N', ']' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + CHECK(json::from_bjdata(v2) == j); + } + + SECTION("Boolean Types") + { + json j = {{"authorized", true}, {"verified", false}}; + std::vector v = {'{', 'i', 10, 'a', 'u', 't', 'h', 'o', 'r', 'i', 'z', 'e', 'd', 'T', + 'i', 8, 'v', 'e', 'r', 'i', 'f', 'i', 'e', 'd', 'F', '}' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Numeric Types") + { + json j = + { + {"int8", 16}, + {"uint8", 255}, + {"int16", 32767}, + {"int32", 2147483647}, + {"int64", 9223372036854775807}, + {"float64", 113243.7863123} + }; + std::vector v = {'{', + 'i', 7, 'f', 'l', 'o', 'a', 't', '6', '4', 'D', 0xcf, 0x34, 0xbc, 0x94, 0xbc, 0xa5, 0xfb, 0x40, + 'i', 5, 'i', 'n', 't', '1', '6', 'I', 0xff, 0x7f, + 'i', 5, 'i', 'n', 't', '3', '2', 'l', 0xff, 0xff, 0xff, 0x7f, + 'i', 5, 'i', 'n', 't', '6', '4', 'L', 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f, + 'i', 4, 'i', 'n', 't', '8', 'i', 16, + 'i', 5, 'u', 'i', 'n', 't', '8', 'U', 0xff, + '}' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Char Type") + { + json j = {{"rolecode", "a"}, {"delim", ";"}}; + std::vector v = {'{', 'i', 5, 'd', 'e', 'l', 'i', 'm', 'C', ';', 'i', 8, 'r', 'o', 'l', 'e', 'c', 'o', 'd', 'e', 'C', 'a', '}'}; + //CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("String Type") + { + SECTION("English") + { + json j = "hello"; + std::vector v = {'S', 'i', 5, 'h', 'e', 'l', 'l', 'o'}; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Russian") + { + json j = "привет"; + std::vector v = {'S', 'i', 12, 0xD0, 0xBF, 0xD1, 0x80, 0xD0, 0xB8, 0xD0, 0xB2, 0xD0, 0xB5, 0xD1, 0x82}; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Russian") + { + json j = "مرحبا"; + std::vector v = {'S', 'i', 10, 0xD9, 0x85, 0xD8, 0xB1, 0xD8, 0xAD, 0xD8, 0xA8, 0xD8, 0xA7}; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + } + + SECTION("Array Type") + { + SECTION("size=false type=false") + { + // note the float has been replaced by a double + json j = {nullptr, true, false, 4782345193, 153.132, "ham"}; + std::vector v = {'[', 'Z', 'T', 'F', 'L', 0xE9, 0xCB, 0x0C, 0x1D, 0x01, 0x00, 0x00, 0x00, 'D', 0x4e, 0x62, 0x10, 0x58, 0x39, 0x24, 0x63, 0x40, 'S', 'i', 3, 'h', 'a', 'm', ']'}; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("size=true type=false") + { + // note the float has been replaced by a double + json j = {nullptr, true, false, 4782345193, 153.132, "ham"}; + std::vector v = {'[', '#', 'i', 6, 'Z', 'T', 'F', 'L', 0xE9, 0xCB, 0x0C, 0x1D, 0x01, 0x00, 0x00, 0x00, 'D', 0x4e, 0x62, 0x10, 0x58, 0x39, 0x24, 0x63, 0x40, 'S', 'i', 3, 'h', 'a', 'm'}; + CHECK(json::to_bjdata(j, true) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("size=true type=true") + { + // note the float has been replaced by a double + json j = {nullptr, true, false, 4782345193, 153.132, "ham"}; + std::vector v = {'[', '#', 'i', 6, 'Z', 'T', 'F', 'L', 0xE9, 0xCB, 0x0C, 0x1D, 0x01, 0x00, 0x00, 0x00, 'D', 0x4e, 0x62, 0x10, 0x58, 0x39, 0x24, 0x63, 0x40, 'S', 'i', 3, 'h', 'a', 'm'}; + CHECK(json::to_bjdata(j, true, true) == v); + CHECK(json::from_bjdata(v) == j); + } + } + + SECTION("Object Type") + { + SECTION("size=false type=false") + { + json j = + { + { + "post", { + {"id", 1137}, + {"author", "rkalla"}, + {"timestamp", 1364482090592}, + {"body", "I totally agree!"} + } + } + }; + std::vector v = {'{', 'i', 4, 'p', 'o', 's', 't', '{', + 'i', 6, 'a', 'u', 't', 'h', 'o', 'r', 'S', 'i', 6, 'r', 'k', 'a', 'l', 'l', 'a', + 'i', 4, 'b', 'o', 'd', 'y', 'S', 'i', 16, 'I', ' ', 't', 'o', 't', 'a', 'l', 'l', 'y', ' ', 'a', 'g', 'r', 'e', 'e', '!', + 'i', 2, 'i', 'd', 'I', 0x71, 0x04, + 'i', 9, 't', 'i', 'm', 'e', 's', 't', 'a', 'm', 'p', 'L', 0x60, 0x66, 0x78, 0xB1, 0x3D, 0x01, 0x00, 0x00, + '}', '}' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("size=true type=false") + { + json j = + { + { + "post", { + {"id", 1137}, + {"author", "rkalla"}, + {"timestamp", 1364482090592}, + {"body", "I totally agree!"} + } + } + }; + std::vector v = {'{', '#', 'i', 1, 'i', 4, 'p', 'o', 's', 't', '{', '#', 'i', 4, + 'i', 6, 'a', 'u', 't', 'h', 'o', 'r', 'S', 'i', 6, 'r', 'k', 'a', 'l', 'l', 'a', + 'i', 4, 'b', 'o', 'd', 'y', 'S', 'i', 16, 'I', ' ', 't', 'o', 't', 'a', 'l', 'l', 'y', ' ', 'a', 'g', 'r', 'e', 'e', '!', + 'i', 2, 'i', 'd', 'I', 0x71, 0x04, + 'i', 9, 't', 'i', 'm', 'e', 's', 't', 'a', 'm', 'p', 'L', 0x60, 0x66, 0x78, 0xB1, 0x3D, 0x01, 0x00, 0x00, + }; + CHECK(json::to_bjdata(j, true) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("size=true type=true") + { + json j = + { + { + "post", { + {"id", 1137}, + {"author", "rkalla"}, + {"timestamp", 1364482090592}, + {"body", "I totally agree!"} + } + } + }; + std::vector v = {'{', '$', '{', '#', 'i', 1, 'i', 4, 'p', 'o', 's', 't', '#', 'i', 4, + 'i', 6, 'a', 'u', 't', 'h', 'o', 'r', 'S', 'i', 6, 'r', 'k', 'a', 'l', 'l', 'a', + 'i', 4, 'b', 'o', 'd', 'y', 'S', 'i', 16, 'I', ' ', 't', 'o', 't', 'a', 'l', 'l', 'y', ' ', 'a', 'g', 'r', 'e', 'e', '!', + 'i', 2, 'i', 'd', 'I', 0x71, 0x04, + 'i', 9, 't', 'i', 'm', 'e', 's', 't', 'a', 'm', 'p', 'L', 0x60, 0x66, 0x78, 0xB1, 0x3D, 0x01, 0x00, 0x00, + }; + CHECK(json::to_bjdata(j, true, true) == v); + CHECK(json::from_bjdata(v) == j); + } + } + + SECTION("Optimized Format") + { + SECTION("Array Example") + { + SECTION("No Optimization") + { + // note the floats have been replaced by doubles + json j = {29.97, 31.13, 67.0, 2.113, 23.888}; + std::vector v = {'[', + 'D', 0xb8, 0x1e, 0x85, 0xeb, 0x51, 0xf8, 0x3d, 0x40, + 'D', 0xe1, 0x7a, 0x14, 0xae, 0x47, 0x21, 0x3f, 0x40, + 'D', 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 'D', 0x81, 0x95, 0x43, 0x8b, 0x6c, 0xe7, 0x00, 0x40, + 'D', 0x17, 0xd9, 0xce, 0xf7, 0x53, 0xe3, 0x37, 0x40, + ']' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Optimized with count") + { + // note the floats have been replaced by doubles + json j = {29.97, 31.13, 67.0, 2.113, 23.888}; + std::vector v = {'[', '#', 'i', 5, + 'D', 0xb8, 0x1e, 0x85, 0xeb, 0x51, 0xf8, 0x3d, 0x40, + 'D', 0xe1, 0x7a, 0x14, 0xae, 0x47, 0x21, 0x3f, 0x40, + 'D', 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 'D', 0x81, 0x95, 0x43, 0x8b, 0x6c, 0xe7, 0x00, 0x40, + 'D', 0x17, 0xd9, 0xce, 0xf7, 0x53, 0xe3, 0x37, 0x40, + }; + CHECK(json::to_bjdata(j, true) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Optimized with type & count") + { + // note the floats have been replaced by doubles + json j = {29.97, 31.13, 67.0, 2.113, 23.888}; + std::vector v = {'[', '$', 'D', '#', 'i', 5, + 0xb8, 0x1e, 0x85, 0xeb, 0x51, 0xf8, 0x3d, 0x40, + 0xe1, 0x7a, 0x14, 0xae, 0x47, 0x21, 0x3f, 0x40, + 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 0x81, 0x95, 0x43, 0x8b, 0x6c, 0xe7, 0x00, 0x40, + 0x17, 0xd9, 0xce, 0xf7, 0x53, 0xe3, 0x37, 0x40, + }; + CHECK(json::to_bjdata(j, true, true) == v); + CHECK(json::from_bjdata(v) == j); + } + } + + SECTION("Object Example") + { + SECTION("No Optimization") + { + // note the floats have been replaced by doubles + json j = { {"lat", 29.976}, {"long", 31.131}, {"alt", 67.0} }; + std::vector v = {'{', + 'i', 3, 'a', 'l', 't', 'D', 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 'i', 3, 'l', 'a', 't', 'D', 0x60, 0xe5, 0xd0, 0x22, 0xdb, 0xf9, 0x3d, 0x40, + 'i', 4, 'l', 'o', 'n', 'g', 'D', 0xa8, 0xc6, 0x4b, 0x37, 0x89, 0x21, 0x3f, 0x40, + '}' + }; + CHECK(json::to_bjdata(j) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Optimized with count") + { + // note the floats have been replaced by doubles + json j = { {"lat", 29.976}, {"long", 31.131}, {"alt", 67.0} }; + std::vector v = {'{', '#', 'i', 3, + 'i', 3, 'a', 'l', 't', 'D', 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 'i', 3, 'l', 'a', 't', 'D', 0x60, 0xe5, 0xd0, 0x22, 0xdb, 0xf9, 0x3d, 0x40, + 'i', 4, 'l', 'o', 'n', 'g', 'D', 0xa8, 0xc6, 0x4b, 0x37, 0x89, 0x21, 0x3f, 0x40, + }; + CHECK(json::to_bjdata(j, true) == v); + CHECK(json::from_bjdata(v) == j); + } + + SECTION("Optimized with type & count") + { + // note the floats have been replaced by doubles + json j = { {"lat", 29.976}, {"long", 31.131}, {"alt", 67.0} }; + std::vector v = {'{', '$', 'D', '#', 'i', 3, + 'i', 3, 'a', 'l', 't', 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x50, 0x40, + 'i', 3, 'l', 'a', 't', 0x60, 0xe5, 0xd0, 0x22, 0xdb, 0xf9, 0x3d, 0x40, + 'i', 4, 'l', 'o', 'n', 'g', 0xa8, 0xc6, 0x4b, 0x37, 0x89, 0x21, 0x3f, 0x40, + }; + CHECK(json::to_bjdata(j, true, true) == v); + CHECK(json::from_bjdata(v) == j); + } + } + + SECTION("Special Cases (Null, No-Op and Boolean)") + { + SECTION("Array") + { + std::vector v = {'[', '$', 'N', '#', 'I', 0x00, 0x02}; + CHECK(json::from_bjdata(v) == json::array()); + } + + SECTION("Object") + { + std::vector v = {'{', '$', 'Z', '#', 'i', 3, 'i', 4, 'n', 'a', 'm', 'e', 'i', 8, 'p', 'a', 's', 's', 'w', 'o', 'r', 'd', 'i', 5, 'e', 'm', 'a', 'i', 'l'}; + CHECK(json::from_bjdata(v) == json({ {"name", nullptr}, {"password", nullptr}, {"email", nullptr} })); + } + } + } +} + +#if !defined(JSON_NOEXCEPTION) +TEST_CASE("all BJData first bytes") +{ + // these bytes will fail immediately with exception parse_error.112 + std::set supported = + { + 'T', 'F', 'Z', 'U', 'i', 'I', 'l', 'L', 'd', 'D', 'C', 'S', '[', '{', 'N', 'H', 'u', 'm', 'M', 'h' + }; + + for (auto i = 0; i < 256; ++i) + { + const auto byte = static_cast(i); + CAPTURE(byte) + + try + { + auto res = json::from_bjdata(std::vector(1, byte)); + } + catch (const json::parse_error& e) + { + // check that parse_error.112 is only thrown if the + // first byte is not in the supported set + INFO_WITH_TEMP(e.what()); + if (supported.find(byte) == supported.end()) + { + CHECK(e.id == 112); + } + else + { + CHECK(e.id != 112); + } + } + } +} +#endif + +#if defined(BJDATA_TEST_ROUNDTRIP) +TEST_CASE("BJData roundtrips" * doctest::skip()) +{ + SECTION("input from self-generated BJData files") + { + for (std::string filename : + { + TEST_DATA_DIRECTORY "/json_nlohmann_tests/all_unicode.json", + TEST_DATA_DIRECTORY "/json.org/1.json", + TEST_DATA_DIRECTORY "/json.org/2.json", + TEST_DATA_DIRECTORY "/json.org/3.json", + TEST_DATA_DIRECTORY "/json.org/4.json", + TEST_DATA_DIRECTORY "/json.org/5.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip01.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip02.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip03.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip04.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip05.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip06.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip07.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip08.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip09.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip10.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip11.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip12.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip13.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip14.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip15.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip16.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip17.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip18.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip19.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip20.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip21.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip22.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip23.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip24.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip25.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip26.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip27.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip28.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip29.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip30.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip31.json", + TEST_DATA_DIRECTORY "/json_roundtrip/roundtrip32.json", + TEST_DATA_DIRECTORY "/json_testsuite/sample.json", + TEST_DATA_DIRECTORY "/json_tests/pass1.json", + TEST_DATA_DIRECTORY "/json_tests/pass2.json", + TEST_DATA_DIRECTORY "/json_tests/pass3.json" + }) + { + CAPTURE(filename) + + { + INFO_WITH_TEMP(filename + ": std::vector"); + // parse JSON file + std::ifstream f_json(filename); + json j1 = json::parse(f_json); + + // parse BJData file + auto packed = utils::read_binary_file(filename + ".bjdata"); + json j2; + CHECK_NOTHROW(j2 = json::from_bjdata(packed)); + + // compare parsed JSON values + CHECK(j1 == j2); + } + + { + INFO_WITH_TEMP(filename + ": std::ifstream"); + // parse JSON file + std::ifstream f_json(filename); + json j1 = json::parse(f_json); + + // parse BJData file + std::ifstream f_bjdata(filename + ".bjdata", std::ios::binary); + json j2; + CHECK_NOTHROW(j2 = json::from_bjdata(f_bjdata)); + + // compare parsed JSON values + CHECK(j1 == j2); + } + + { + INFO_WITH_TEMP(filename + ": uint8_t* and size"); + // parse JSON file + std::ifstream f_json(filename); + json j1 = json::parse(f_json); + + // parse BJData file + auto packed = utils::read_binary_file(filename + ".bjdata"); + json j2; + CHECK_NOTHROW(j2 = json::from_bjdata({packed.data(), packed.size()})); + + // compare parsed JSON values + CHECK(j1 == j2); + } + + { + INFO_WITH_TEMP(filename + ": output to output adapters"); + // parse JSON file + std::ifstream f_json(filename); + json j1 = json::parse(f_json); + + // parse BJData file + auto packed = utils::read_binary_file(filename + ".bjdata"); + + { + INFO_WITH_TEMP(filename + ": output adapters: std::vector"); + std::vector vec; + json::to_bjdata(j1, vec); + CHECK(vec == packed); + } + } + } + } +} +#endif \ No newline at end of file