diff options
| author | Daniel Friesel <derf@finalrewind.org> | 2018-09-17 11:10:12 +0200 |
|---|---|---|
| committer | Daniel Friesel <derf@finalrewind.org> | 2018-09-17 11:10:12 +0200 |
| commit | 276b9a93c2e3b9ac58a9ddb5fff4c8299c459222 (patch) | |
| tree | 8b39c09a58a8c3225b234b2b69af08a76d31f9a8 /include/lib/modernjson/detail | |
| parent | 99cea89d798b61398adf33cbc57c11d985ddab77 (diff) | |
add nlohmann modenjson
Diffstat (limited to 'include/lib/modernjson/detail')
27 files changed, 10904 insertions, 0 deletions
diff --git a/include/lib/modernjson/detail/conversions/from_json.hpp b/include/lib/modernjson/detail/conversions/from_json.hpp new file mode 100644 index 0000000..8281dda --- /dev/null +++ b/include/lib/modernjson/detail/conversions/from_json.hpp @@ -0,0 +1,370 @@ +#pragma once + +#include <algorithm> // transform +#include <array> // array +#include <ciso646> // and, not +#include <forward_list> // forward_list +#include <iterator> // inserter, front_inserter, end +#include <map> // map +#include <string> // string +#include <tuple> // tuple, make_tuple +#include <type_traits> // is_arithmetic, is_same, is_enum, underlying_type, is_convertible +#include <unordered_map> // unordered_map +#include <utility> // pair, declval +#include <valarray> // valarray + +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/meta/cpp_future.hpp> +#include <lib/modernjson/detail/meta/type_traits.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename std::nullptr_t& n) +{ + if (JSON_UNLIKELY(not j.is_null())) + { + JSON_THROW(type_error::create(302, "type must be null, but is " + std::string(j.type_name()))); + } + n = nullptr; +} + +// overloads for basic_json template parameters +template<typename BasicJsonType, typename ArithmeticType, + enable_if_t<std::is_arithmetic<ArithmeticType>::value and + not std::is_same<ArithmeticType, typename BasicJsonType::boolean_t>::value, + int> = 0> +void get_arithmetic_value(const BasicJsonType& j, ArithmeticType& val) +{ + switch (static_cast<value_t>(j)) + { + case value_t::number_unsigned: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_unsigned_t*>()); + break; + } + case value_t::number_integer: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_integer_t*>()); + break; + } + case value_t::number_float: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_float_t*>()); + break; + } + + default: + JSON_THROW(type_error::create(302, "type must be number, but is " + std::string(j.type_name()))); + } +} + +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename BasicJsonType::boolean_t& b) +{ + if (JSON_UNLIKELY(not j.is_boolean())) + { + JSON_THROW(type_error::create(302, "type must be boolean, but is " + std::string(j.type_name()))); + } + b = *j.template get_ptr<const typename BasicJsonType::boolean_t*>(); +} + +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename BasicJsonType::string_t& s) +{ + if (JSON_UNLIKELY(not j.is_string())) + { + JSON_THROW(type_error::create(302, "type must be string, but is " + std::string(j.type_name()))); + } + s = *j.template get_ptr<const typename BasicJsonType::string_t*>(); +} + +template < + typename BasicJsonType, typename CompatibleStringType, + enable_if_t < + is_compatible_string_type<BasicJsonType, CompatibleStringType>::value and + not std::is_same<typename BasicJsonType::string_t, + CompatibleStringType>::value, + int > = 0 > +void from_json(const BasicJsonType& j, CompatibleStringType& s) +{ + if (JSON_UNLIKELY(not j.is_string())) + { + JSON_THROW(type_error::create(302, "type must be string, but is " + std::string(j.type_name()))); + } + + s = *j.template get_ptr<const typename BasicJsonType::string_t*>(); +} + +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename BasicJsonType::number_float_t& val) +{ + get_arithmetic_value(j, val); +} + +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename BasicJsonType::number_unsigned_t& val) +{ + get_arithmetic_value(j, val); +} + +template<typename BasicJsonType> +void from_json(const BasicJsonType& j, typename BasicJsonType::number_integer_t& val) +{ + get_arithmetic_value(j, val); +} + +template<typename BasicJsonType, typename EnumType, + enable_if_t<std::is_enum<EnumType>::value, int> = 0> +void from_json(const BasicJsonType& j, EnumType& e) +{ + typename std::underlying_type<EnumType>::type val; + get_arithmetic_value(j, val); + e = static_cast<EnumType>(val); +} + +// forward_list doesn't have an insert method +template<typename BasicJsonType, typename T, typename Allocator, + enable_if_t<std::is_convertible<BasicJsonType, T>::value, int> = 0> +void from_json(const BasicJsonType& j, std::forward_list<T, Allocator>& l) +{ + if (JSON_UNLIKELY(not j.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(j.type_name()))); + } + std::transform(j.rbegin(), j.rend(), + std::front_inserter(l), [](const BasicJsonType & i) + { + return i.template get<T>(); + }); +} + +// valarray doesn't have an insert method +template<typename BasicJsonType, typename T, + enable_if_t<std::is_convertible<BasicJsonType, T>::value, int> = 0> +void from_json(const BasicJsonType& j, std::valarray<T>& l) +{ + if (JSON_UNLIKELY(not j.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(j.type_name()))); + } + l.resize(j.size()); + std::copy(j.m_value.array->begin(), j.m_value.array->end(), std::begin(l)); +} + +template<typename BasicJsonType> +void from_json_array_impl(const BasicJsonType& j, typename BasicJsonType::array_t& arr, priority_tag<3> /*unused*/) +{ + arr = *j.template get_ptr<const typename BasicJsonType::array_t*>(); +} + +template <typename BasicJsonType, typename T, std::size_t N> +auto from_json_array_impl(const BasicJsonType& j, std::array<T, N>& arr, + priority_tag<2> /*unused*/) +-> decltype(j.template get<T>(), void()) +{ + for (std::size_t i = 0; i < N; ++i) + { + arr[i] = j.at(i).template get<T>(); + } +} + +template<typename BasicJsonType, typename CompatibleArrayType> +auto from_json_array_impl(const BasicJsonType& j, CompatibleArrayType& arr, priority_tag<1> /*unused*/) +-> decltype( + arr.reserve(std::declval<typename CompatibleArrayType::size_type>()), + j.template get<typename CompatibleArrayType::value_type>(), + void()) +{ + using std::end; + + arr.reserve(j.size()); + std::transform(j.begin(), j.end(), + std::inserter(arr, end(arr)), [](const BasicJsonType & i) + { + // get<BasicJsonType>() returns *this, this won't call a from_json + // method when value_type is BasicJsonType + return i.template get<typename CompatibleArrayType::value_type>(); + }); +} + +template <typename BasicJsonType, typename CompatibleArrayType> +void from_json_array_impl(const BasicJsonType& j, CompatibleArrayType& arr, + priority_tag<0> /*unused*/) +{ + using std::end; + + std::transform( + j.begin(), j.end(), std::inserter(arr, end(arr)), + [](const BasicJsonType & i) + { + // get<BasicJsonType>() returns *this, this won't call a from_json + // method when value_type is BasicJsonType + return i.template get<typename CompatibleArrayType::value_type>(); + }); +} + +template <typename BasicJsonType, typename CompatibleArrayType, + enable_if_t < + is_compatible_array_type<BasicJsonType, CompatibleArrayType>::value and + not is_compatible_object_type<BasicJsonType, CompatibleArrayType>::value and + not is_compatible_string_type<BasicJsonType, CompatibleArrayType>::value and + not is_basic_json<CompatibleArrayType>::value, + int > = 0 > + +auto from_json(const BasicJsonType& j, CompatibleArrayType& arr) +-> decltype(from_json_array_impl(j, arr, priority_tag<3> {}), +j.template get<typename CompatibleArrayType::value_type>(), +void()) +{ + if (JSON_UNLIKELY(not j.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + + std::string(j.type_name()))); + } + + from_json_array_impl(j, arr, priority_tag<3> {}); +} + +template<typename BasicJsonType, typename CompatibleObjectType, + enable_if_t<is_compatible_object_type<BasicJsonType, CompatibleObjectType>::value, int> = 0> +void from_json(const BasicJsonType& j, CompatibleObjectType& obj) +{ + if (JSON_UNLIKELY(not j.is_object())) + { + JSON_THROW(type_error::create(302, "type must be object, but is " + std::string(j.type_name()))); + } + + auto inner_object = j.template get_ptr<const typename BasicJsonType::object_t*>(); + using value_type = typename CompatibleObjectType::value_type; + std::transform( + inner_object->begin(), inner_object->end(), + std::inserter(obj, obj.begin()), + [](typename BasicJsonType::object_t::value_type const & p) + { + return value_type(p.first, p.second.template get<typename CompatibleObjectType::mapped_type>()); + }); +} + +// overload for arithmetic types, not chosen for basic_json template arguments +// (BooleanType, etc..); note: Is it really necessary to provide explicit +// overloads for boolean_t etc. in case of a custom BooleanType which is not +// an arithmetic type? +template<typename BasicJsonType, typename ArithmeticType, + enable_if_t < + std::is_arithmetic<ArithmeticType>::value and + not std::is_same<ArithmeticType, typename BasicJsonType::number_unsigned_t>::value and + not std::is_same<ArithmeticType, typename BasicJsonType::number_integer_t>::value and + not std::is_same<ArithmeticType, typename BasicJsonType::number_float_t>::value and + not std::is_same<ArithmeticType, typename BasicJsonType::boolean_t>::value, + int> = 0> +void from_json(const BasicJsonType& j, ArithmeticType& val) +{ + switch (static_cast<value_t>(j)) + { + case value_t::number_unsigned: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_unsigned_t*>()); + break; + } + case value_t::number_integer: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_integer_t*>()); + break; + } + case value_t::number_float: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::number_float_t*>()); + break; + } + case value_t::boolean: + { + val = static_cast<ArithmeticType>(*j.template get_ptr<const typename BasicJsonType::boolean_t*>()); + break; + } + + default: + JSON_THROW(type_error::create(302, "type must be number, but is " + std::string(j.type_name()))); + } +} + +template<typename BasicJsonType, typename A1, typename A2> +void from_json(const BasicJsonType& j, std::pair<A1, A2>& p) +{ + p = {j.at(0).template get<A1>(), j.at(1).template get<A2>()}; +} + +template<typename BasicJsonType, typename Tuple, std::size_t... Idx> +void from_json_tuple_impl(const BasicJsonType& j, Tuple& t, index_sequence<Idx...>) +{ + t = std::make_tuple(j.at(Idx).template get<typename std::tuple_element<Idx, Tuple>::type>()...); +} + +template<typename BasicJsonType, typename... Args> +void from_json(const BasicJsonType& j, std::tuple<Args...>& t) +{ + from_json_tuple_impl(j, t, index_sequence_for<Args...> {}); +} + +template <typename BasicJsonType, typename Key, typename Value, typename Compare, typename Allocator, + typename = enable_if_t<not std::is_constructible< + typename BasicJsonType::string_t, Key>::value>> +void from_json(const BasicJsonType& j, std::map<Key, Value, Compare, Allocator>& m) +{ + if (JSON_UNLIKELY(not j.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(j.type_name()))); + } + for (const auto& p : j) + { + if (JSON_UNLIKELY(not p.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(p.type_name()))); + } + m.emplace(p.at(0).template get<Key>(), p.at(1).template get<Value>()); + } +} + +template <typename BasicJsonType, typename Key, typename Value, typename Hash, typename KeyEqual, typename Allocator, + typename = enable_if_t<not std::is_constructible< + typename BasicJsonType::string_t, Key>::value>> +void from_json(const BasicJsonType& j, std::unordered_map<Key, Value, Hash, KeyEqual, Allocator>& m) +{ + if (JSON_UNLIKELY(not j.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(j.type_name()))); + } + for (const auto& p : j) + { + if (JSON_UNLIKELY(not p.is_array())) + { + JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(p.type_name()))); + } + m.emplace(p.at(0).template get<Key>(), p.at(1).template get<Value>()); + } +} + +struct from_json_fn +{ + template<typename BasicJsonType, typename T> + auto operator()(const BasicJsonType& j, T& val) const + noexcept(noexcept(from_json(j, val))) + -> decltype(from_json(j, val), void()) + { + return from_json(j, val); + } +}; +} + +/// namespace to hold default `from_json` function +/// to see why this is required: +/// http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/n4381.html +namespace +{ +constexpr const auto& from_json = detail::static_const<detail::from_json_fn>::value; +} +} diff --git a/include/lib/modernjson/detail/conversions/to_chars.hpp b/include/lib/modernjson/detail/conversions/to_chars.hpp new file mode 100644 index 0000000..a13d258 --- /dev/null +++ b/include/lib/modernjson/detail/conversions/to_chars.hpp @@ -0,0 +1,1095 @@ +#pragma once + +#include <cassert> // assert +#include <ciso646> // or, and, not +#include <cmath> // signbit, isfinite +#include <cstdint> // intN_t, uintN_t +#include <cstring> // memcpy, memmove + +namespace nlohmann +{ +namespace detail +{ + +/*! +@brief implements the Grisu2 algorithm for binary to decimal floating-point +conversion. + +This implementation is a slightly modified version of the reference +implementation which may be obtained from +http://florian.loitsch.com/publications (bench.tar.gz). + +The code is distributed under the MIT license, Copyright (c) 2009 Florian Loitsch. + +For a detailed description of the algorithm see: + +[1] Loitsch, "Printing Floating-Point Numbers Quickly and Accurately with + Integers", Proceedings of the ACM SIGPLAN 2010 Conference on Programming + Language Design and Implementation, PLDI 2010 +[2] Burger, Dybvig, "Printing Floating-Point Numbers Quickly and Accurately", + Proceedings of the ACM SIGPLAN 1996 Conference on Programming Language + Design and Implementation, PLDI 1996 +*/ +namespace dtoa_impl +{ + +template <typename Target, typename Source> +Target reinterpret_bits(const Source source) +{ + static_assert(sizeof(Target) == sizeof(Source), "size mismatch"); + + Target target; + std::memcpy(&target, &source, sizeof(Source)); + return target; +} + +struct diyfp // f * 2^e +{ + static constexpr int kPrecision = 64; // = q + + uint64_t f; + int e; + + constexpr diyfp() noexcept : f(0), e(0) {} + constexpr diyfp(uint64_t f_, int e_) noexcept : f(f_), e(e_) {} + + /*! + @brief returns x - y + @pre x.e == y.e and x.f >= y.f + */ + static diyfp sub(const diyfp& x, const diyfp& y) noexcept + { + assert(x.e == y.e); + assert(x.f >= y.f); + + return diyfp(x.f - y.f, x.e); + } + + /*! + @brief returns x * y + @note The result is rounded. (Only the upper q bits are returned.) + */ + static diyfp mul(const diyfp& x, const diyfp& y) noexcept + { + static_assert(kPrecision == 64, "internal error"); + + // Computes: + // f = round((x.f * y.f) / 2^q) + // e = x.e + y.e + q + + // Emulate the 64-bit * 64-bit multiplication: + // + // p = u * v + // = (u_lo + 2^32 u_hi) (v_lo + 2^32 v_hi) + // = (u_lo v_lo ) + 2^32 ((u_lo v_hi ) + (u_hi v_lo )) + 2^64 (u_hi v_hi ) + // = (p0 ) + 2^32 ((p1 ) + (p2 )) + 2^64 (p3 ) + // = (p0_lo + 2^32 p0_hi) + 2^32 ((p1_lo + 2^32 p1_hi) + (p2_lo + 2^32 p2_hi)) + 2^64 (p3 ) + // = (p0_lo ) + 2^32 (p0_hi + p1_lo + p2_lo ) + 2^64 (p1_hi + p2_hi + p3) + // = (p0_lo ) + 2^32 (Q ) + 2^64 (H ) + // = (p0_lo ) + 2^32 (Q_lo + 2^32 Q_hi ) + 2^64 (H ) + // + // (Since Q might be larger than 2^32 - 1) + // + // = (p0_lo + 2^32 Q_lo) + 2^64 (Q_hi + H) + // + // (Q_hi + H does not overflow a 64-bit int) + // + // = p_lo + 2^64 p_hi + + const uint64_t u_lo = x.f & 0xFFFFFFFF; + const uint64_t u_hi = x.f >> 32; + const uint64_t v_lo = y.f & 0xFFFFFFFF; + const uint64_t v_hi = y.f >> 32; + + const uint64_t p0 = u_lo * v_lo; + const uint64_t p1 = u_lo * v_hi; + const uint64_t p2 = u_hi * v_lo; + const uint64_t p3 = u_hi * v_hi; + + const uint64_t p0_hi = p0 >> 32; + const uint64_t p1_lo = p1 & 0xFFFFFFFF; + const uint64_t p1_hi = p1 >> 32; + const uint64_t p2_lo = p2 & 0xFFFFFFFF; + const uint64_t p2_hi = p2 >> 32; + + uint64_t Q = p0_hi + p1_lo + p2_lo; + + // The full product might now be computed as + // + // p_hi = p3 + p2_hi + p1_hi + (Q >> 32) + // p_lo = p0_lo + (Q << 32) + // + // But in this particular case here, the full p_lo is not required. + // Effectively we only need to add the highest bit in p_lo to p_hi (and + // Q_hi + 1 does not overflow). + + Q += uint64_t{1} << (64 - 32 - 1); // round, ties up + + const uint64_t h = p3 + p2_hi + p1_hi + (Q >> 32); + + return diyfp(h, x.e + y.e + 64); + } + + /*! + @brief normalize x such that the significand is >= 2^(q-1) + @pre x.f != 0 + */ + static diyfp normalize(diyfp x) noexcept + { + assert(x.f != 0); + + while ((x.f >> 63) == 0) + { + x.f <<= 1; + x.e--; + } + + return x; + } + + /*! + @brief normalize x such that the result has the exponent E + @pre e >= x.e and the upper e - x.e bits of x.f must be zero. + */ + static diyfp normalize_to(const diyfp& x, const int target_exponent) noexcept + { + const int delta = x.e - target_exponent; + + assert(delta >= 0); + assert(((x.f << delta) >> delta) == x.f); + + return diyfp(x.f << delta, target_exponent); + } +}; + +struct boundaries +{ + diyfp w; + diyfp minus; + diyfp plus; +}; + +/*! +Compute the (normalized) diyfp representing the input number 'value' and its +boundaries. + +@pre value must be finite and positive +*/ +template <typename FloatType> +boundaries compute_boundaries(FloatType value) +{ + assert(std::isfinite(value)); + assert(value > 0); + + // Convert the IEEE representation into a diyfp. + // + // If v is denormal: + // value = 0.F * 2^(1 - bias) = ( F) * 2^(1 - bias - (p-1)) + // If v is normalized: + // value = 1.F * 2^(E - bias) = (2^(p-1) + F) * 2^(E - bias - (p-1)) + + static_assert(std::numeric_limits<FloatType>::is_iec559, + "internal error: dtoa_short requires an IEEE-754 floating-point implementation"); + + constexpr int kPrecision = std::numeric_limits<FloatType>::digits; // = p (includes the hidden bit) + constexpr int kBias = std::numeric_limits<FloatType>::max_exponent - 1 + (kPrecision - 1); + constexpr int kMinExp = 1 - kBias; + constexpr uint64_t kHiddenBit = uint64_t{1} << (kPrecision - 1); // = 2^(p-1) + + using bits_type = typename std::conditional< kPrecision == 24, uint32_t, uint64_t >::type; + + const uint64_t bits = reinterpret_bits<bits_type>(value); + const uint64_t E = bits >> (kPrecision - 1); + const uint64_t F = bits & (kHiddenBit - 1); + + const bool is_denormal = (E == 0); + const diyfp v = is_denormal + ? diyfp(F, kMinExp) + : diyfp(F + kHiddenBit, static_cast<int>(E) - kBias); + + // Compute the boundaries m- and m+ of the floating-point value + // v = f * 2^e. + // + // Determine v- and v+, the floating-point predecessor and successor if v, + // respectively. + // + // v- = v - 2^e if f != 2^(p-1) or e == e_min (A) + // = v - 2^(e-1) if f == 2^(p-1) and e > e_min (B) + // + // v+ = v + 2^e + // + // Let m- = (v- + v) / 2 and m+ = (v + v+) / 2. All real numbers _strictly_ + // between m- and m+ round to v, regardless of how the input rounding + // algorithm breaks ties. + // + // ---+-------------+-------------+-------------+-------------+--- (A) + // v- m- v m+ v+ + // + // -----------------+------+------+-------------+-------------+--- (B) + // v- m- v m+ v+ + + const bool lower_boundary_is_closer = (F == 0 and E > 1); + const diyfp m_plus = diyfp(2 * v.f + 1, v.e - 1); + const diyfp m_minus = lower_boundary_is_closer + ? diyfp(4 * v.f - 1, v.e - 2) // (B) + : diyfp(2 * v.f - 1, v.e - 1); // (A) + + // Determine the normalized w+ = m+. + const diyfp w_plus = diyfp::normalize(m_plus); + + // Determine w- = m- such that e_(w-) = e_(w+). + const diyfp w_minus = diyfp::normalize_to(m_minus, w_plus.e); + + return {diyfp::normalize(v), w_minus, w_plus}; +} + +// Given normalized diyfp w, Grisu needs to find a (normalized) cached +// power-of-ten c, such that the exponent of the product c * w = f * 2^e lies +// within a certain range [alpha, gamma] (Definition 3.2 from [1]) +// +// alpha <= e = e_c + e_w + q <= gamma +// +// or +// +// f_c * f_w * 2^alpha <= f_c 2^(e_c) * f_w 2^(e_w) * 2^q +// <= f_c * f_w * 2^gamma +// +// Since c and w are normalized, i.e. 2^(q-1) <= f < 2^q, this implies +// +// 2^(q-1) * 2^(q-1) * 2^alpha <= c * w * 2^q < 2^q * 2^q * 2^gamma +// +// or +// +// 2^(q - 2 + alpha) <= c * w < 2^(q + gamma) +// +// The choice of (alpha,gamma) determines the size of the table and the form of +// the digit generation procedure. Using (alpha,gamma)=(-60,-32) works out well +// in practice: +// +// The idea is to cut the number c * w = f * 2^e into two parts, which can be +// processed independently: An integral part p1, and a fractional part p2: +// +// f * 2^e = ( (f div 2^-e) * 2^-e + (f mod 2^-e) ) * 2^e +// = (f div 2^-e) + (f mod 2^-e) * 2^e +// = p1 + p2 * 2^e +// +// The conversion of p1 into decimal form requires a series of divisions and +// modulos by (a power of) 10. These operations are faster for 32-bit than for +// 64-bit integers, so p1 should ideally fit into a 32-bit integer. This can be +// achieved by choosing +// +// -e >= 32 or e <= -32 := gamma +// +// In order to convert the fractional part +// +// p2 * 2^e = p2 / 2^-e = d[-1] / 10^1 + d[-2] / 10^2 + ... +// +// into decimal form, the fraction is repeatedly multiplied by 10 and the digits +// d[-i] are extracted in order: +// +// (10 * p2) div 2^-e = d[-1] +// (10 * p2) mod 2^-e = d[-2] / 10^1 + ... +// +// The multiplication by 10 must not overflow. It is sufficient to choose +// +// 10 * p2 < 16 * p2 = 2^4 * p2 <= 2^64. +// +// Since p2 = f mod 2^-e < 2^-e, +// +// -e <= 60 or e >= -60 := alpha + +constexpr int kAlpha = -60; +constexpr int kGamma = -32; + +struct cached_power // c = f * 2^e ~= 10^k +{ + uint64_t f; + int e; + int k; +}; + +/*! +For a normalized diyfp w = f * 2^e, this function returns a (normalized) cached +power-of-ten c = f_c * 2^e_c, such that the exponent of the product w * c +satisfies (Definition 3.2 from [1]) + + alpha <= e_c + e + q <= gamma. +*/ +inline cached_power get_cached_power_for_binary_exponent(int e) +{ + // Now + // + // alpha <= e_c + e + q <= gamma (1) + // ==> f_c * 2^alpha <= c * 2^e * 2^q + // + // and since the c's are normalized, 2^(q-1) <= f_c, + // + // ==> 2^(q - 1 + alpha) <= c * 2^(e + q) + // ==> 2^(alpha - e - 1) <= c + // + // If c were an exakt power of ten, i.e. c = 10^k, one may determine k as + // + // k = ceil( log_10( 2^(alpha - e - 1) ) ) + // = ceil( (alpha - e - 1) * log_10(2) ) + // + // From the paper: + // "In theory the result of the procedure could be wrong since c is rounded, + // and the computation itself is approximated [...]. In practice, however, + // this simple function is sufficient." + // + // For IEEE double precision floating-point numbers converted into + // normalized diyfp's w = f * 2^e, with q = 64, + // + // e >= -1022 (min IEEE exponent) + // -52 (p - 1) + // -52 (p - 1, possibly normalize denormal IEEE numbers) + // -11 (normalize the diyfp) + // = -1137 + // + // and + // + // e <= +1023 (max IEEE exponent) + // -52 (p - 1) + // -11 (normalize the diyfp) + // = 960 + // + // This binary exponent range [-1137,960] results in a decimal exponent + // range [-307,324]. One does not need to store a cached power for each + // k in this range. For each such k it suffices to find a cached power + // such that the exponent of the product lies in [alpha,gamma]. + // This implies that the difference of the decimal exponents of adjacent + // table entries must be less than or equal to + // + // floor( (gamma - alpha) * log_10(2) ) = 8. + // + // (A smaller distance gamma-alpha would require a larger table.) + + // NB: + // Actually this function returns c, such that -60 <= e_c + e + 64 <= -34. + + constexpr int kCachedPowersSize = 79; + constexpr int kCachedPowersMinDecExp = -300; + constexpr int kCachedPowersDecStep = 8; + + static constexpr cached_power kCachedPowers[] = + { + { 0xAB70FE17C79AC6CA, -1060, -300 }, + { 0xFF77B1FCBEBCDC4F, -1034, -292 }, + { 0xBE5691EF416BD60C, -1007, -284 }, + { 0x8DD01FAD907FFC3C, -980, -276 }, + { 0xD3515C2831559A83, -954, -268 }, + { 0x9D71AC8FADA6C9B5, -927, -260 }, + { 0xEA9C227723EE8BCB, -901, -252 }, + { 0xAECC49914078536D, -874, -244 }, + { 0x823C12795DB6CE57, -847, -236 }, + { 0xC21094364DFB5637, -821, -228 }, + { 0x9096EA6F3848984F, -794, -220 }, + { 0xD77485CB25823AC7, -768, -212 }, + { 0xA086CFCD97BF97F4, -741, -204 }, + { 0xEF340A98172AACE5, -715, -196 }, + { 0xB23867FB2A35B28E, -688, -188 }, + { 0x84C8D4DFD2C63F3B, -661, -180 }, + { 0xC5DD44271AD3CDBA, -635, -172 }, + { 0x936B9FCEBB25C996, -608, -164 }, + { 0xDBAC6C247D62A584, -582, -156 }, + { 0xA3AB66580D5FDAF6, -555, -148 }, + { 0xF3E2F893DEC3F126, -529, -140 }, + { 0xB5B5ADA8AAFF80B8, -502, -132 }, + { 0x87625F056C7C4A8B, -475, -124 }, + { 0xC9BCFF6034C13053, -449, -116 }, + { 0x964E858C91BA2655, -422, -108 }, + { 0xDFF9772470297EBD, -396, -100 }, + { 0xA6DFBD9FB8E5B88F, -369, -92 }, + { 0xF8A95FCF88747D94, -343, -84 }, + { 0xB94470938FA89BCF, -316, -76 }, + { 0x8A08F0F8BF0F156B, -289, -68 }, + { 0xCDB02555653131B6, -263, -60 }, + { 0x993FE2C6D07B7FAC, -236, -52 }, + { 0xE45C10C42A2B3B06, -210, -44 }, + { 0xAA242499697392D3, -183, -36 }, + { 0xFD87B5F28300CA0E, -157, -28 }, + { 0xBCE5086492111AEB, -130, -20 }, + { 0x8CBCCC096F5088CC, -103, -12 }, + { 0xD1B71758E219652C, -77, -4 }, + { 0x9C40000000000000, -50, 4 }, + { 0xE8D4A51000000000, -24, 12 }, + { 0xAD78EBC5AC620000, 3, 20 }, + { 0x813F3978F8940984, 30, 28 }, + { 0xC097CE7BC90715B3, 56, 36 }, + { 0x8F7E32CE7BEA5C70, 83, 44 }, + { 0xD5D238A4ABE98068, 109, 52 }, + { 0x9F4F2726179A2245, 136, 60 }, + { 0xED63A231D4C4FB27, 162, 68 }, + { 0xB0DE65388CC8ADA8, 189, 76 }, + { 0x83C7088E1AAB65DB, 216, 84 }, + { 0xC45D1DF942711D9A, 242, 92 }, + { 0x924D692CA61BE758, 269, 100 }, + { 0xDA01EE641A708DEA, 295, 108 }, + { 0xA26DA3999AEF774A, 322, 116 }, + { 0xF209787BB47D6B85, 348, 124 }, + { 0xB454E4A179DD1877, 375, 132 }, + { 0x865B86925B9BC5C2, 402, 140 }, + { 0xC83553C5C8965D3D, 428, 148 }, + { 0x952AB45CFA97A0B3, 455, 156 }, + { 0xDE469FBD99A05FE3, 481, 164 }, + { 0xA59BC234DB398C25, 508, 172 }, + { 0xF6C69A72A3989F5C, 534, 180 }, + { 0xB7DCBF5354E9BECE, 561, 188 }, + { 0x88FCF317F22241E2, 588, 196 }, + { 0xCC20CE9BD35C78A5, 614, 204 }, + { 0x98165AF37B2153DF, 641, 212 }, + { 0xE2A0B5DC971F303A, 667, 220 }, + { 0xA8D9D1535CE3B396, 694, 228 }, + { 0xFB9B7CD9A4A7443C, 720, 236 }, + { 0xBB764C4CA7A44410, 747, 244 }, + { 0x8BAB8EEFB6409C1A, 774, 252 }, + { 0xD01FEF10A657842C, 800, 260 }, + { 0x9B10A4E5E9913129, 827, 268 }, + { 0xE7109BFBA19C0C9D, 853, 276 }, + { 0xAC2820D9623BF429, 880, 284 }, + { 0x80444B5E7AA7CF85, 907, 292 }, + { 0xBF21E44003ACDD2D, 933, 300 }, + { 0x8E679C2F5E44FF8F, 960, 308 }, + { 0xD433179D9C8CB841, 986, 316 }, + { 0x9E19DB92B4E31BA9, 1013, 324 }, + }; + + // This computation gives exactly the same results for k as + // k = ceil((kAlpha - e - 1) * 0.30102999566398114) + // for |e| <= 1500, but doesn't require floating-point operations. + // NB: log_10(2) ~= 78913 / 2^18 + assert(e >= -1500); + assert(e <= 1500); + const int f = kAlpha - e - 1; + const int k = (f * 78913) / (1 << 18) + (f > 0); + + const int index = (-kCachedPowersMinDecExp + k + (kCachedPowersDecStep - 1)) / kCachedPowersDecStep; + assert(index >= 0); + assert(index < kCachedPowersSize); + static_cast<void>(kCachedPowersSize); // Fix warning. + + const cached_power cached = kCachedPowers[index]; + assert(kAlpha <= cached.e + e + 64); + assert(kGamma >= cached.e + e + 64); + + return cached; +} + +/*! +For n != 0, returns k, such that pow10 := 10^(k-1) <= n < 10^k. +For n == 0, returns 1 and sets pow10 := 1. +*/ +inline int find_largest_pow10(const uint32_t n, uint32_t& pow10) +{ + // LCOV_EXCL_START + if (n >= 1000000000) + { + pow10 = 1000000000; + return 10; + } + // LCOV_EXCL_STOP + else if (n >= 100000000) + { + pow10 = 100000000; + return 9; + } + else if (n >= 10000000) + { + pow10 = 10000000; + return 8; + } + else if (n >= 1000000) + { + pow10 = 1000000; + return 7; + } + else if (n >= 100000) + { + pow10 = 100000; + return 6; + } + else if (n >= 10000) + { + pow10 = 10000; + return 5; + } + else if (n >= 1000) + { + pow10 = 1000; + return 4; + } + else if (n >= 100) + { + pow10 = 100; + return 3; + } + else if (n >= 10) + { + pow10 = 10; + return 2; + } + else + { + pow10 = 1; + return 1; + } +} + +inline void grisu2_round(char* buf, int len, uint64_t dist, uint64_t delta, + uint64_t rest, uint64_t ten_k) +{ + assert(len >= 1); + assert(dist <= delta); + assert(rest <= delta); + assert(ten_k > 0); + + // <--------------------------- delta ----> + // <---- dist ---------> + // --------------[------------------+-------------------]-------------- + // M- w M+ + // + // ten_k + // <------> + // <---- rest ----> + // --------------[------------------+----+--------------]-------------- + // w V + // = buf * 10^k + // + // ten_k represents a unit-in-the-last-place in the decimal representation + // stored in buf. + // Decrement buf by ten_k while this takes buf closer to w. + + // The tests are written in this order to avoid overflow in unsigned + // integer arithmetic. + + while (rest < dist + and delta - rest >= ten_k + and (rest + ten_k < dist or dist - rest > rest + ten_k - dist)) + { + assert(buf[len - 1] != '0'); + buf[len - 1]--; + rest += ten_k; + } +} + +/*! +Generates V = buffer * 10^decimal_exponent, such that M- <= V <= M+. +M- and M+ must be normalized and share the same exponent -60 <= e <= -32. +*/ +inline void grisu2_digit_gen(char* buffer, int& length, int& decimal_exponent, + diyfp M_minus, diyfp w, diyfp M_plus) +{ + static_assert(kAlpha >= -60, "internal error"); + static_assert(kGamma <= -32, "internal error"); + + // Generates the digits (and the exponent) of a decimal floating-point + // number V = buffer * 10^decimal_exponent in the range [M-, M+]. The diyfp's + // w, M- and M+ share the same exponent e, which satisfies alpha <= e <= gamma. + // + // <--------------------------- delta ----> + // <---- dist ---------> + // --------------[------------------+-------------------]-------------- + // M- w M+ + // + // Grisu2 generates the digits of M+ from left to right and stops as soon as + // V is in [M-,M+]. + + assert(M_plus.e >= kAlpha); + assert(M_plus.e <= kGamma); + + uint64_t delta = diyfp::sub(M_plus, M_minus).f; // (significand of (M+ - M-), implicit exponent is e) + uint64_t dist = diyfp::sub(M_plus, w ).f; // (significand of (M+ - w ), implicit exponent is e) + + // Split M+ = f * 2^e into two parts p1 and p2 (note: e < 0): + // + // M+ = f * 2^e + // = ((f div 2^-e) * 2^-e + (f mod 2^-e)) * 2^e + // = ((p1 ) * 2^-e + (p2 )) * 2^e + // = p1 + p2 * 2^e + + const diyfp one(uint64_t{1} << -M_plus.e, M_plus.e); + + uint32_t p1 = static_cast<uint32_t>(M_plus.f >> -one.e); // p1 = f div 2^-e (Since -e >= 32, p1 fits into a 32-bit int.) + uint64_t p2 = M_plus.f & (one.f - 1); // p2 = f mod 2^-e + + // 1) + // + // Generate the digits of the integral part p1 = d[n-1]...d[1]d[0] + + assert(p1 > 0); + + uint32_t pow10; + const int k = find_largest_pow10(p1, pow10); + + // 10^(k-1) <= p1 < 10^k, pow10 = 10^(k-1) + // + // p1 = (p1 div 10^(k-1)) * 10^(k-1) + (p1 mod 10^(k-1)) + // = (d[k-1] ) * 10^(k-1) + (p1 mod 10^(k-1)) + // + // M+ = p1 + p2 * 2^e + // = d[k-1] * 10^(k-1) + (p1 mod 10^(k-1)) + p2 * 2^e + // = d[k-1] * 10^(k-1) + ((p1 mod 10^(k-1)) * 2^-e + p2) * 2^e + // = d[k-1] * 10^(k-1) + ( rest) * 2^e + // + // Now generate the digits d[n] of p1 from left to right (n = k-1,...,0) + // + // p1 = d[k-1]...d[n] * 10^n + d[n-1]...d[0] + // + // but stop as soon as + // + // rest * 2^e = (d[n-1]...d[0] * 2^-e + p2) * 2^e <= delta * 2^e + + int n = k; + while (n > 0) + { + // Invariants: + // M+ = buffer * 10^n + (p1 + p2 * 2^e) (buffer = 0 for n = k) + // pow10 = 10^(n-1) <= p1 < 10^n + // + const uint32_t d = p1 / pow10; // d = p1 div 10^(n-1) + const uint32_t r = p1 % pow10; // r = p1 mod 10^(n-1) + // + // M+ = buffer * 10^n + (d * 10^(n-1) + r) + p2 * 2^e + // = (buffer * 10 + d) * 10^(n-1) + (r + p2 * 2^e) + // + assert(d <= 9); + buffer[length++] = static_cast<char>('0' + d); // buffer := buffer * 10 + d + // + // M+ = buffer * 10^(n-1) + (r + p2 * 2^e) + // + p1 = r; + n--; + // + // M+ = buffer * 10^n + (p1 + p2 * 2^e) + // pow10 = 10^n + // + + // Now check if enough digits have been generated. + // Compute + // + // p1 + p2 * 2^e = (p1 * 2^-e + p2) * 2^e = rest * 2^e + // + // Note: + // Since rest and delta share the same exponent e, it suffices to + // compare the significands. + const uint64_t rest = (uint64_t{p1} << -one.e) + p2; + if (rest <= delta) + { + // V = buffer * 10^n, with M- <= V <= M+. + + decimal_exponent += n; + + // We may now just stop. But instead look if the buffer could be + // decremented to bring V closer to w. + // + // pow10 = 10^n is now 1 ulp in the decimal representation V. + // The rounding procedure works with diyfp's with an implicit + // exponent of e. + // + // 10^n = (10^n * 2^-e) * 2^e = ulp * 2^e + // + const uint64_t ten_n = uint64_t{pow10} << -one.e; + grisu2_round(buffer, length, dist, delta, rest, ten_n); + + return; + } + + pow10 /= 10; + // + // pow10 = 10^(n-1) <= p1 < 10^n + // Invariants restored. + } + + // 2) + // + // The digits of the integral part have been generated: + // + // M+ = d[k-1]...d[1]d[0] + p2 * 2^e + // = buffer + p2 * 2^e + // + // Now generate the digits of the fractional part p2 * 2^e. + // + // Note: + // No decimal point is generated: the exponent is adjusted instead. + // + // p2 actually represents the fraction + // + // p2 * 2^e + // = p2 / 2^-e + // = d[-1] / 10^1 + d[-2] / 10^2 + ... + // + // Now generate the digits d[-m] of p1 from left to right (m = 1,2,...) + // + // p2 * 2^e = d[-1]d[-2]...d[-m] * 10^-m + // + 10^-m * (d[-m-1] / 10^1 + d[-m-2] / 10^2 + ...) + // + // using + // + // 10^m * p2 = ((10^m * p2) div 2^-e) * 2^-e + ((10^m * p2) mod 2^-e) + // = ( d) * 2^-e + ( r) + // + // or + // 10^m * p2 * 2^e = d + r * 2^e + // + // i.e. + // + // M+ = buffer + p2 * 2^e + // = buffer + 10^-m * (d + r * 2^e) + // = (buffer * 10^m + d) * 10^-m + 10^-m * r * 2^e + // + // and stop as soon as 10^-m * r * 2^e <= delta * 2^e + + assert(p2 > delta); + + int m = 0; + for (;;) + { + // Invariant: + // M+ = buffer * 10^-m + 10^-m * (d[-m-1] / 10 + d[-m-2] / 10^2 + ...) * 2^e + // = buffer * 10^-m + 10^-m * (p2 ) * 2^e + // = buffer * 10^-m + 10^-m * (1/10 * (10 * p2) ) * 2^e + // = buffer * 10^-m + 10^-m * (1/10 * ((10*p2 div 2^-e) * 2^-e + (10*p2 mod 2^-e)) * 2^e + // + assert(p2 <= UINT64_MAX / 10); + p2 *= 10; + const uint64_t d = p2 >> -one.e; // d = (10 * p2) div 2^-e + const uint64_t r = p2 & (one.f - 1); // r = (10 * p2) mod 2^-e + // + // M+ = buffer * 10^-m + 10^-m * (1/10 * (d * 2^-e + r) * 2^e + // = buffer * 10^-m + 10^-m * (1/10 * (d + r * 2^e)) + // = (buffer * 10 + d) * 10^(-m-1) + 10^(-m-1) * r * 2^e + // + assert(d <= 9); + buffer[length++] = static_cast<char>('0' + d); // buffer := buffer * 10 + d + // + // M+ = buffer * 10^(-m-1) + 10^(-m-1) * r * 2^e + // + p2 = r; + m++; + // + // M+ = buffer * 10^-m + 10^-m * p2 * 2^e + // Invariant restored. + + // Check if enough digits have been generated. + // + // 10^-m * p2 * 2^e <= delta * 2^e + // p2 * 2^e <= 10^m * delta * 2^e + // p2 <= 10^m * delta + delta *= 10; + dist *= 10; + if (p2 <= delta) + { + break; + } + } + + // V = buffer * 10^-m, with M- <= V <= M+. + + decimal_exponent -= m; + + // 1 ulp in the decimal representation is now 10^-m. + // Since delta and dist are now scaled by 10^m, we need to do the + // same with ulp in order to keep the units in sync. + // + // 10^m * 10^-m = 1 = 2^-e * 2^e = ten_m * 2^e + // + const uint64_t ten_m = one.f; + grisu2_round(buffer, length, dist, delta, p2, ten_m); + + // By construction this algorithm generates the shortest possible decimal + // number (Loitsch, Theorem 6.2) which rounds back to w. + // For an input number of precision p, at least + // + // N = 1 + ceil(p * log_10(2)) + // + // decimal digits are sufficient to identify all binary floating-point + // numbers (Matula, "In-and-Out conversions"). + // This implies that the algorithm does not produce more than N decimal + // digits. + // + // N = 17 for p = 53 (IEEE double precision) + // N = 9 for p = 24 (IEEE single precision) +} + +/*! +v = buf * 10^decimal_exponent +len is the length of the buffer (number of decimal digits) +The buffer must be large enough, i.e. >= max_digits10. +*/ +inline void grisu2(char* buf, int& len, int& decimal_exponent, + diyfp m_minus, diyfp v, diyfp m_plus) +{ + assert(m_plus.e == m_minus.e); + assert(m_plus.e == v.e); + + // --------(-----------------------+-----------------------)-------- (A) + // m- v m+ + // + // --------------------(-----------+-----------------------)-------- (B) + // m- v m+ + // + // First scale v (and m- and m+) such that the exponent is in the range + // [alpha, gamma]. + + const cached_power cached = get_cached_power_for_binary_exponent(m_plus.e); + + const diyfp c_minus_k(cached.f, cached.e); // = c ~= 10^-k + + // The exponent of the products is = v.e + c_minus_k.e + q and is in the range [alpha,gamma] + const diyfp w = diyfp::mul(v, c_minus_k); + const diyfp w_minus = diyfp::mul(m_minus, c_minus_k); + const diyfp w_plus = diyfp::mul(m_plus, c_minus_k); + + // ----(---+---)---------------(---+---)---------------(---+---)---- + // w- w w+ + // = c*m- = c*v = c*m+ + // + // diyfp::mul rounds its result and c_minus_k is approximated too. w, w- and + // w+ are now off by a small amount. + // In fact: + // + // w - v * 10^k < 1 ulp + // + // To account for this inaccuracy, add resp. subtract 1 ulp. + // + // --------+---[---------------(---+---)---------------]---+-------- + // w- M- w M+ w+ + // + // Now any number in [M-, M+] (bounds included) will round to w when input, + // regardless of how the input rounding algorithm breaks ties. + // + // And digit_gen generates the shortest possible such number in [M-, M+]. + // Note that this does not mean that Grisu2 always generates the shortest + // possible number in the interval (m-, m+). + const diyfp M_minus(w_minus.f + 1, w_minus.e); + const diyfp M_plus (w_plus.f - 1, w_plus.e ); + + decimal_exponent = -cached.k; // = -(-k) = k + + grisu2_digit_gen(buf, len, decimal_exponent, M_minus, w, M_plus); +} + +/*! +v = buf * 10^decimal_exponent +len is the length of the buffer (number of decimal digits) +The buffer must be large enough, i.e. >= max_digits10. +*/ +template <typename FloatType> +void grisu2(char* buf, int& len, int& decimal_exponent, FloatType value) +{ + static_assert(diyfp::kPrecision >= std::numeric_limits<FloatType>::digits + 3, + "internal error: not enough precision"); + + assert(std::isfinite(value)); + assert(value > 0); + + // If the neighbors (and boundaries) of 'value' are always computed for double-precision + // numbers, all float's can be recovered using strtod (and strtof). However, the resulting + // decimal representations are not exactly "short". + // + // The documentation for 'std::to_chars' (https://en.cppreference.com/w/cpp/utility/to_chars) + // says "value is converted to a string as if by std::sprintf in the default ("C") locale" + // and since sprintf promotes float's to double's, I think this is exactly what 'std::to_chars' + // does. + // On the other hand, the documentation for 'std::to_chars' requires that "parsing the + // representation using the corresponding std::from_chars function recovers value exactly". That + // indicates that single precision floating-point numbers should be recovered using + // 'std::strtof'. + // + // NB: If the neighbors are computed for single-precision numbers, there is a single float + // (7.0385307e-26f) which can't be recovered using strtod. The resulting double precision + // value is off by 1 ulp. +#if 0 + const boundaries w = compute_boundaries(static_cast<double>(value)); +#else + const boundaries w = compute_boundaries(value); +#endif + + grisu2(buf, len, decimal_exponent, w.minus, w.w, w.plus); +} + +/*! +@brief appends a decimal representation of e to buf +@return a pointer to the element following the exponent. +@pre -1000 < e < 1000 +*/ +inline char* append_exponent(char* buf, int e) +{ + assert(e > -1000); + assert(e < 1000); + + if (e < 0) + { + e = -e; + *buf++ = '-'; + } + else + { + *buf++ = '+'; + } + + uint32_t k = static_cast<uint32_t>(e); + if (k < 10) + { + // Always print at least two digits in the exponent. + // This is for compatibility with printf("%g"). + *buf++ = '0'; + *buf++ = static_cast<char>('0' + k); + } + else if (k < 100) + { + *buf++ = static_cast<char>('0' + k / 10); + k %= 10; + *buf++ = static_cast<char>('0' + k); + } + else + { + *buf++ = static_cast<char>('0' + k / 100); + k %= 100; + *buf++ = static_cast<char>('0' + k / 10); + k %= 10; + *buf++ = static_cast<char>('0' + k); + } + + return buf; +} + +/*! +@brief prettify v = buf * 10^decimal_exponent + +If v is in the range [10^min_exp, 10^max_exp) it will be printed in fixed-point +notation. Otherwise it will be printed in exponential notation. + +@pre min_exp < 0 +@pre max_exp > 0 +*/ +inline char* format_buffer(char* buf, int len, int decimal_exponent, + int min_exp, int max_exp) +{ + assert(min_exp < 0); + assert(max_exp > 0); + + const int k = len; + const int n = len + decimal_exponent; + + // v = buf * 10^(n-k) + // k is the length of the buffer (number of decimal digits) + // n is the position of the decimal point relative to the start of the buffer. + + if (k <= n and n <= max_exp) + { + // digits[000] + // len <= max_exp + 2 + + std::memset(buf + k, '0', static_cast<size_t>(n - k)); + // Make it look like a floating-point number (#362, #378) + buf[n + 0] = '.'; + buf[n + 1] = '0'; + return buf + (n + 2); + } + + if (0 < n and n <= max_exp) + { + // dig.its + // len <= max_digits10 + 1 + + assert(k > n); + + std::memmove(buf + (n + 1), buf + n, static_cast<size_t>(k - n)); + buf[n] = '.'; + return buf + (k + 1); + } + + if (min_exp < n and n <= 0) + { + // 0.[000]digits + // len <= 2 + (-min_exp - 1) + max_digits10 + + std::memmove(buf + (2 + -n), buf, static_cast<size_t>(k)); + buf[0] = '0'; + buf[1] = '.'; + std::memset(buf + 2, '0', static_cast<size_t>(-n)); + return buf + (2 + (-n) + k); + } + + if (k == 1) + { + // dE+123 + // len <= 1 + 5 + + buf += 1; + } + else + { + // d.igitsE+123 + // len <= max_digits10 + 1 + 5 + + std::memmove(buf + 2, buf + 1, static_cast<size_t>(k - 1)); + buf[1] = '.'; + buf += 1 + k; + } + + *buf++ = 'e'; + return append_exponent(buf, n - 1); +} + +} // namespace dtoa_impl + +/*! +@brief generates a decimal representation of the floating-point number value in [first, last). + +The format of the resulting decimal representation is similar to printf's %g +format. Returns an iterator pointing past-the-end of the decimal representation. + +@note The input number must be finite, i.e. NaN's and Inf's are not supported. +@note The buffer must be large enough. +@note The result is NOT null-terminated. +*/ +template <typename FloatType> +char* to_chars(char* first, char* last, FloatType value) +{ + static_cast<void>(last); // maybe unused - fix warning + assert(std::isfinite(value)); + + // Use signbit(value) instead of (value < 0) since signbit works for -0. + if (std::signbit(value)) + { + value = -value; + *first++ = '-'; + } + + if (value == 0) // +-0 + { + *first++ = '0'; + // Make it look like a floating-point number (#362, #378) + *first++ = '.'; + *first++ = '0'; + return first; + } + + assert(last - first >= std::numeric_limits<FloatType>::max_digits10); + + // Compute v = buffer * 10^decimal_exponent. + // The decimal digits are stored in the buffer, which needs to be interpreted + // as an unsigned decimal integer. + // len is the length of the buffer, i.e. the number of decimal digits. + int len = 0; + int decimal_exponent = 0; + dtoa_impl::grisu2(first, len, decimal_exponent, value); + + assert(len <= std::numeric_limits<FloatType>::max_digits10); + + // Format the buffer like printf("%.*g", prec, value) + constexpr int kMinExp = -4; + // Use digits10 here to increase compatibility with version 2. + constexpr int kMaxExp = std::numeric_limits<FloatType>::digits10; + + assert(last - first >= kMaxExp + 2); + assert(last - first >= 2 + (-kMinExp - 1) + std::numeric_limits<FloatType>::max_digits10); + assert(last - first >= std::numeric_limits<FloatType>::max_digits10 + 6); + + return dtoa_impl::format_buffer(first, len, decimal_exponent, kMinExp, kMaxExp); +} + +} // namespace detail +} // namespace nlohmann diff --git a/include/lib/modernjson/detail/conversions/to_json.hpp b/include/lib/modernjson/detail/conversions/to_json.hpp new file mode 100644 index 0000000..37210ad --- /dev/null +++ b/include/lib/modernjson/detail/conversions/to_json.hpp @@ -0,0 +1,342 @@ +#pragma once + +#include <ciso646> // or, and, not +#include <iterator> // begin, end +#include <tuple> // tuple, get +#include <type_traits> // is_same, is_constructible, is_floating_point, is_enum, underlying_type +#include <utility> // move, forward, declval, pair +#include <valarray> // valarray +#include <vector> // vector + +#include <lib/modernjson/detail/meta/cpp_future.hpp> +#include <lib/modernjson/detail/meta/type_traits.hpp> +#include <lib/modernjson/detail/value_t.hpp> +#include <lib/modernjson/detail/iterators/iteration_proxy.hpp> + +namespace nlohmann +{ +namespace detail +{ +////////////////// +// constructors // +////////////////// + +template<value_t> struct external_constructor; + +template<> +struct external_constructor<value_t::boolean> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::boolean_t b) noexcept + { + j.m_type = value_t::boolean; + j.m_value = b; + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::string> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, const typename BasicJsonType::string_t& s) + { + j.m_type = value_t::string; + j.m_value = s; + j.assert_invariant(); + } + + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::string_t&& s) + { + j.m_type = value_t::string; + j.m_value = std::move(s); + j.assert_invariant(); + } + + template<typename BasicJsonType, typename CompatibleStringType, + enable_if_t<not std::is_same<CompatibleStringType, typename BasicJsonType::string_t>::value, + int> = 0> + static void construct(BasicJsonType& j, const CompatibleStringType& str) + { + j.m_type = value_t::string; + j.m_value.string = j.template create<typename BasicJsonType::string_t>(str); + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::number_float> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::number_float_t val) noexcept + { + j.m_type = value_t::number_float; + j.m_value = val; + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::number_unsigned> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::number_unsigned_t val) noexcept + { + j.m_type = value_t::number_unsigned; + j.m_value = val; + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::number_integer> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::number_integer_t val) noexcept + { + j.m_type = value_t::number_integer; + j.m_value = val; + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::array> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, const typename BasicJsonType::array_t& arr) + { + j.m_type = value_t::array; + j.m_value = arr; + j.assert_invariant(); + } + + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::array_t&& arr) + { + j.m_type = value_t::array; + j.m_value = std::move(arr); + j.assert_invariant(); + } + + template<typename BasicJsonType, typename CompatibleArrayType, + enable_if_t<not std::is_same<CompatibleArrayType, typename BasicJsonType::array_t>::value, + int> = 0> + static void construct(BasicJsonType& j, const CompatibleArrayType& arr) + { + using std::begin; + using std::end; + j.m_type = value_t::array; + j.m_value.array = j.template create<typename BasicJsonType::array_t>(begin(arr), end(arr)); + j.assert_invariant(); + } + + template<typename BasicJsonType> + static void construct(BasicJsonType& j, const std::vector<bool>& arr) + { + j.m_type = value_t::array; + j.m_value = value_t::array; + j.m_value.array->reserve(arr.size()); + for (const bool x : arr) + { + j.m_value.array->push_back(x); + } + j.assert_invariant(); + } + + template<typename BasicJsonType, typename T, + enable_if_t<std::is_convertible<T, BasicJsonType>::value, int> = 0> + static void construct(BasicJsonType& j, const std::valarray<T>& arr) + { + j.m_type = value_t::array; + j.m_value = value_t::array; + j.m_value.array->resize(arr.size()); + std::copy(std::begin(arr), std::end(arr), j.m_value.array->begin()); + j.assert_invariant(); + } +}; + +template<> +struct external_constructor<value_t::object> +{ + template<typename BasicJsonType> + static void construct(BasicJsonType& j, const typename BasicJsonType::object_t& obj) + { + j.m_type = value_t::object; + j.m_value = obj; + j.assert_invariant(); + } + + template<typename BasicJsonType> + static void construct(BasicJsonType& j, typename BasicJsonType::object_t&& obj) + { + j.m_type = value_t::object; + j.m_value = std::move(obj); + j.assert_invariant(); + } + + template<typename BasicJsonType, typename CompatibleObjectType, + enable_if_t<not std::is_same<CompatibleObjectType, typename BasicJsonType::object_t>::value, int> = 0> + static void construct(BasicJsonType& j, const CompatibleObjectType& obj) + { + using std::begin; + using std::end; + + j.m_type = value_t::object; + j.m_value.object = j.template create<typename BasicJsonType::object_t>(begin(obj), end(obj)); + j.assert_invariant(); + } +}; + +///////////// +// to_json // +///////////// + +template<typename BasicJsonType, typename T, + enable_if_t<std::is_same<T, typename BasicJsonType::boolean_t>::value, int> = 0> +void to_json(BasicJsonType& j, T b) noexcept +{ + external_constructor<value_t::boolean>::construct(j, b); +} + +template<typename BasicJsonType, typename CompatibleString, + enable_if_t<std::is_constructible<typename BasicJsonType::string_t, CompatibleString>::value, int> = 0> +void to_json(BasicJsonType& j, const CompatibleString& s) +{ + external_constructor<value_t::string>::construct(j, s); +} + +template<typename BasicJsonType> +void to_json(BasicJsonType& j, typename BasicJsonType::string_t&& s) +{ + external_constructor<value_t::string>::construct(j, std::move(s)); +} + +template<typename BasicJsonType, typename FloatType, + enable_if_t<std::is_floating_point<FloatType>::value, int> = 0> +void to_json(BasicJsonType& j, FloatType val) noexcept +{ + external_constructor<value_t::number_float>::construct(j, static_cast<typename BasicJsonType::number_float_t>(val)); +} + +template<typename BasicJsonType, typename CompatibleNumberUnsignedType, + enable_if_t<is_compatible_integer_type<typename BasicJsonType::number_unsigned_t, CompatibleNumberUnsignedType>::value, int> = 0> +void to_json(BasicJsonType& j, CompatibleNumberUnsignedType val) noexcept +{ + external_constructor<value_t::number_unsigned>::construct(j, static_cast<typename BasicJsonType::number_unsigned_t>(val)); +} + +template<typename BasicJsonType, typename CompatibleNumberIntegerType, + enable_if_t<is_compatible_integer_type<typename BasicJsonType::number_integer_t, CompatibleNumberIntegerType>::value, int> = 0> +void to_json(BasicJsonType& j, CompatibleNumberIntegerType val) noexcept +{ + external_constructor<value_t::number_integer>::construct(j, static_cast<typename BasicJsonType::number_integer_t>(val)); +} + +template<typename BasicJsonType, typename EnumType, + enable_if_t<std::is_enum<EnumType>::value, int> = 0> +void to_json(BasicJsonType& j, EnumType e) noexcept +{ + using underlying_type = typename std::underlying_type<EnumType>::type; + external_constructor<value_t::number_integer>::construct(j, static_cast<underlying_type>(e)); +} + +template<typename BasicJsonType> +void to_json(BasicJsonType& j, const std::vector<bool>& e) +{ + external_constructor<value_t::array>::construct(j, e); +} + +template <typename BasicJsonType, typename CompatibleArrayType, + enable_if_t<is_compatible_array_type<BasicJsonType, + CompatibleArrayType>::value and + not is_compatible_object_type< + BasicJsonType, CompatibleArrayType>::value and + not is_compatible_string_type<BasicJsonType, CompatibleArrayType>::value and + not is_basic_json<CompatibleArrayType>::value, + int> = 0> +void to_json(BasicJsonType& j, const CompatibleArrayType& arr) +{ + external_constructor<value_t::array>::construct(j, arr); +} + +template<typename BasicJsonType, typename T, + enable_if_t<std::is_convertible<T, BasicJsonType>::value, int> = 0> +void to_json(BasicJsonType& j, const std::valarray<T>& arr) +{ + external_constructor<value_t::array>::construct(j, std::move(arr)); +} + +template<typename BasicJsonType> +void to_json(BasicJsonType& j, typename BasicJsonType::array_t&& arr) +{ + external_constructor<value_t::array>::construct(j, std::move(arr)); +} + +template<typename BasicJsonType, typename CompatibleObjectType, + enable_if_t<is_compatible_object_type<BasicJsonType, CompatibleObjectType>::value and not is_basic_json<CompatibleObjectType>::value, int> = 0> +void to_json(BasicJsonType& j, const CompatibleObjectType& obj) +{ + external_constructor<value_t::object>::construct(j, obj); +} + +template<typename BasicJsonType> +void to_json(BasicJsonType& j, typename BasicJsonType::object_t&& obj) +{ + external_constructor<value_t::object>::construct(j, std::move(obj)); +} + +template < + typename BasicJsonType, typename T, std::size_t N, + enable_if_t<not std::is_constructible<typename BasicJsonType::string_t, + const T (&)[N]>::value, + int> = 0 > +void to_json(BasicJsonType& j, const T (&arr)[N]) +{ + external_constructor<value_t::array>::construct(j, arr); +} + +template<typename BasicJsonType, typename... Args> +void to_json(BasicJsonType& j, const std::pair<Args...>& p) +{ + j = {p.first, p.second}; +} + +// for https://github.com/nlohmann/json/pull/1134 +template<typename BasicJsonType, typename T, + enable_if_t<std::is_same<T, typename iteration_proxy<typename BasicJsonType::iterator>::iteration_proxy_internal>::value, int> = 0> +void to_json(BasicJsonType& j, T b) noexcept +{ + j = {{b.key(), b.value()}}; +} + +template<typename BasicJsonType, typename Tuple, std::size_t... Idx> +void to_json_tuple_impl(BasicJsonType& j, const Tuple& t, index_sequence<Idx...>) +{ + j = {std::get<Idx>(t)...}; +} + +template<typename BasicJsonType, typename... Args> +void to_json(BasicJsonType& j, const std::tuple<Args...>& t) +{ + to_json_tuple_impl(j, t, index_sequence_for<Args...> {}); +} + +struct to_json_fn +{ + template<typename BasicJsonType, typename T> + auto operator()(BasicJsonType& j, T&& val) const noexcept(noexcept(to_json(j, std::forward<T>(val)))) + -> decltype(to_json(j, std::forward<T>(val)), void()) + { + return to_json(j, std::forward<T>(val)); + } +}; +} + +/// namespace to hold default `to_json` function +namespace +{ +constexpr const auto& to_json = detail::static_const<detail::to_json_fn>::value; +} +} diff --git a/include/lib/modernjson/detail/exceptions.hpp b/include/lib/modernjson/detail/exceptions.hpp new file mode 100644 index 0000000..b73d7b1 --- /dev/null +++ b/include/lib/modernjson/detail/exceptions.hpp @@ -0,0 +1,330 @@ +#pragma once + +#include <exception> // exception +#include <stdexcept> // runtime_error +#include <string> // to_string + +namespace nlohmann +{ +namespace detail +{ +//////////////// +// exceptions // +//////////////// + +/*! +@brief general exception of the @ref basic_json class + +This class is an extension of `std::exception` objects with a member @a id for +exception ids. It is used as the base class for all exceptions thrown by the +@ref basic_json class. This class can hence be used as "wildcard" to catch +exceptions. + +Subclasses: +- @ref parse_error for exceptions indicating a parse error +- @ref invalid_iterator for exceptions indicating errors with iterators +- @ref type_error for exceptions indicating executing a member function with + a wrong type +- @ref out_of_range for exceptions indicating access out of the defined range +- @ref other_error for exceptions indicating other library errors + +@internal +@note To have nothrow-copy-constructible exceptions, we internally use + `std::runtime_error` which can cope with arbitrary-length error messages. + Intermediate strings are built with static functions and then passed to + the actual constructor. +@endinternal + +@liveexample{The following code shows how arbitrary library exceptions can be +caught.,exception} + +@since version 3.0.0 +*/ +class exception : public std::exception +{ + public: + /// returns the explanatory string + const char* what() const noexcept override + { + return m.what(); + } + + /// the id of the exception + const int id; + + protected: + exception(int id_, const char* what_arg) : id(id_), m(what_arg) {} + + static std::string name(const std::string& ename, int id_) + { + return "[json.exception." + ename + "." + std::to_string(id_) + "] "; + } + + private: + /// an exception object as storage for error messages + std::runtime_error m; +}; + +/*! +@brief exception indicating a parse error + +This exception is thrown by the library when a parse error occurs. Parse errors +can occur during the deserialization of JSON text, CBOR, MessagePack, as well +as when using JSON Patch. + +Member @a byte holds the byte index of the last read character in the input +file. + +Exceptions have ids 1xx. + +name / id | example message | description +------------------------------ | --------------- | ------------------------- +json.exception.parse_error.101 | parse error at 2: unexpected end of input; expected string literal | This error indicates a syntax error while deserializing a JSON text. The error message describes that an unexpected token (character) was encountered, and the member @a byte indicates the error position. +json.exception.parse_error.102 | parse error at 14: missing or wrong low surrogate | JSON uses the `\uxxxx` format to describe Unicode characters. Code points above above 0xFFFF are split into two `\uxxxx` entries ("surrogate pairs"). This error indicates that the surrogate pair is incomplete or contains an invalid code point. +json.exception.parse_error.103 | parse error: code points above 0x10FFFF are invalid | Unicode supports code points up to 0x10FFFF. Code points above 0x10FFFF are invalid. +json.exception.parse_error.104 | parse error: JSON patch must be an array of objects | [RFC 6902](https://tools.ietf.org/html/rfc6902) requires a JSON Patch document to be a JSON document that represents an array of objects. +json.exception.parse_error.105 | parse error: operation must have string member 'op' | An operation of a JSON Patch document must contain exactly one "op" member, whose value indicates the operation to perform. Its value must be one of "add", "remove", "replace", "move", "copy", or "test"; other values are errors. +json.exception.parse_error.106 | parse error: array index '01' must not begin with '0' | An array index in a JSON Pointer ([RFC 6901](https://tools.ietf.org/html/rfc6901)) may be `0` or any number without a leading `0`. +json.exception.parse_error.107 | parse error: JSON pointer must be empty or begin with '/' - was: 'foo' | A JSON Pointer must be a Unicode string containing a sequence of zero or more reference tokens, each prefixed by a `/` character. +json.exception.parse_error.108 | parse error: escape character '~' must be followed with '0' or '1' | In a JSON Pointer, only `~0` and `~1` are valid escape sequences. +json.exception.parse_error.109 | parse error: array index 'one' is not a number | A JSON Pointer array index must be a number. +json.exception.parse_error.110 | parse error at 1: cannot read 2 bytes from vector | When parsing CBOR or MessagePack, the byte vector ends before the complete value has been read. +json.exception.parse_error.112 | parse error at 1: error reading CBOR; last byte: 0xF8 | Not all types of CBOR or MessagePack are supported. This exception occurs if an unsupported byte was read. +json.exception.parse_error.113 | parse error at 2: expected a CBOR string; last byte: 0x98 | While parsing a map key, a value that is not a string has been read. + +@note For an input with n bytes, 1 is the index of the first character and n+1 + is the index of the terminating null byte or the end of file. This also + holds true when reading a byte vector (CBOR or MessagePack). + +@liveexample{The following code shows how a `parse_error` exception can be +caught.,parse_error} + +@sa @ref exception for the base class of the library exceptions +@sa @ref invalid_iterator for exceptions indicating errors with iterators +@sa @ref type_error for exceptions indicating executing a member function with + a wrong type +@sa @ref out_of_range for exceptions indicating access out of the defined range +@sa @ref other_error for exceptions indicating other library errors + +@since version 3.0.0 +*/ +class parse_error : public exception +{ + public: + /*! + @brief create a parse error exception + @param[in] id_ the id of the exception + @param[in] byte_ the byte index where the error occurred (or 0 if the + position cannot be determined) + @param[in] what_arg the explanatory string + @return parse_error object + */ + static parse_error create(int id_, std::size_t byte_, const std::string& what_arg) + { + std::string w = exception::name("parse_error", id_) + "parse error" + + (byte_ != 0 ? (" at " + std::to_string(byte_)) : "") + + ": " + what_arg; + return parse_error(id_, byte_, w.c_str()); + } + + /*! + @brief byte index of the parse error + + The byte index of the last read character in the input file. + + @note For an input with n bytes, 1 is the index of the first character and + n+1 is the index of the terminating null byte or the end of file. + This also holds true when reading a byte vector (CBOR or MessagePack). + */ + const std::size_t byte; + + private: + parse_error(int id_, std::size_t byte_, const char* what_arg) + : exception(id_, what_arg), byte(byte_) {} +}; + +/*! +@brief exception indicating errors with iterators + +This exception is thrown if iterators passed to a library function do not match +the expected semantics. + +Exceptions have ids 2xx. + +name / id | example message | description +----------------------------------- | --------------- | ------------------------- +json.exception.invalid_iterator.201 | iterators are not compatible | The iterators passed to constructor @ref basic_json(InputIT first, InputIT last) are not compatible, meaning they do not belong to the same container. Therefore, the range (@a first, @a last) is invalid. +json.exception.invalid_iterator.202 | iterator does not fit current value | In an erase or insert function, the passed iterator @a pos does not belong to the JSON value for which the function was called. It hence does not define a valid position for the deletion/insertion. +json.exception.invalid_iterator.203 | iterators do not fit current value | Either iterator passed to function @ref erase(IteratorType first, IteratorType last) does not belong to the JSON value from which values shall be erased. It hence does not define a valid range to delete values from. +json.exception.invalid_iterator.204 | iterators out of range | When an iterator range for a primitive type (number, boolean, or string) is passed to a constructor or an erase function, this range has to be exactly (@ref begin(), @ref end()), because this is the only way the single stored value is expressed. All other ranges are invalid. +json.exception.invalid_iterator.205 | iterator out of range | When an iterator for a primitive type (number, boolean, or string) is passed to an erase function, the iterator has to be the @ref begin() iterator, because it is the only way to address the stored value. All other iterators are invalid. +json.exception.invalid_iterator.206 | cannot construct with iterators from null | The iterators passed to constructor @ref basic_json(InputIT first, InputIT last) belong to a JSON null value and hence to not define a valid range. +json.exception.invalid_iterator.207 | cannot use key() for non-object iterators | The key() member function can only be used on iterators belonging to a JSON object, because other types do not have a concept of a key. +json.exception.invalid_iterator.208 | cannot use operator[] for object iterators | The operator[] to specify a concrete offset cannot be used on iterators belonging to a JSON object, because JSON objects are unordered. +json.exception.invalid_iterator.209 | cannot use offsets with object iterators | The offset operators (+, -, +=, -=) cannot be used on iterators belonging to a JSON object, because JSON objects are unordered. +json.exception.invalid_iterator.210 | iterators do not fit | The iterator range passed to the insert function are not compatible, meaning they do not belong to the same container. Therefore, the range (@a first, @a last) is invalid. +json.exception.invalid_iterator.211 | passed iterators may not belong to container | The iterator range passed to the insert function must not be a subrange of the container to insert to. +json.exception.invalid_iterator.212 | cannot compare iterators of different containers | When two iterators are compared, they must belong to the same container. +json.exception.invalid_iterator.213 | cannot compare order of object iterators | The order of object iterators cannot be compared, because JSON objects are unordered. +json.exception.invalid_iterator.214 | cannot get value | Cannot get value for iterator: Either the iterator belongs to a null value or it is an iterator to a primitive type (number, boolean, or string), but the iterator is different to @ref begin(). + +@liveexample{The following code shows how an `invalid_iterator` exception can be +caught.,invalid_iterator} + +@sa @ref exception for the base class of the library exceptions +@sa @ref parse_error for exceptions indicating a parse error +@sa @ref type_error for exceptions indicating executing a member function with + a wrong type +@sa @ref out_of_range for exceptions indicating access out of the defined range +@sa @ref other_error for exceptions indicating other library errors + +@since version 3.0.0 +*/ +class invalid_iterator : public exception +{ + public: + static invalid_iterator create(int id_, const std::string& what_arg) + { + std::string w = exception::name("invalid_iterator", id_) + what_arg; + return invalid_iterator(id_, w.c_str()); + } + + private: + invalid_iterator(int id_, const char* what_arg) + : exception(id_, what_arg) {} +}; + +/*! +@brief exception indicating executing a member function with a wrong type + +This exception is thrown in case of a type error; that is, a library function is +executed on a JSON value whose type does not match the expected semantics. + +Exceptions have ids 3xx. + +name / id | example message | description +----------------------------- | --------------- | ------------------------- +json.exception.type_error.301 | cannot create object from initializer list | To create an object from an initializer list, the initializer list must consist only of a list of pairs whose first element is a string. When this constraint is violated, an array is created instead. +json.exception.type_error.302 | type must be object, but is array | During implicit or explicit value conversion, the JSON type must be compatible to the target type. For instance, a JSON string can only be converted into string types, but not into numbers or boolean types. +json.exception.type_error.303 | incompatible ReferenceType for get_ref, actual type is object | To retrieve a reference to a value stored in a @ref basic_json object with @ref get_ref, the type of the reference must match the value type. For instance, for a JSON array, the @a ReferenceType must be @ref array_t&. +json.exception.type_error.304 | cannot use at() with string | The @ref at() member functions can only be executed for certain JSON types. +json.exception.type_error.305 | cannot use operator[] with string | The @ref operator[] member functions can only be executed for certain JSON types. +json.exception.type_error.306 | cannot use value() with string | The @ref value() member functions can only be executed for certain JSON types. +json.exception.type_error.307 | cannot use erase() with string | The @ref erase() member functions can only be executed for certain JSON types. +json.exception.type_error.308 | cannot use push_back() with string | The @ref push_back() and @ref operator+= member functions can only be executed for certain JSON types. +json.exception.type_error.309 | cannot use insert() with | The @ref insert() member functions can only be executed for certain JSON types. +json.exception.type_error.310 | cannot use swap() with number | The @ref swap() member functions can only be executed for certain JSON types. +json.exception.type_error.311 | cannot use emplace_back() with string | The @ref emplace_back() member function can only be executed for certain JSON types. +json.exception.type_error.312 | cannot use update() with string | The @ref update() member functions can only be executed for certain JSON types. +json.exception.type_error.313 | invalid value to unflatten | The @ref unflatten function converts an object whose keys are JSON Pointers back into an arbitrary nested JSON value. The JSON Pointers must not overlap, because then the resulting value would not be well defined. +json.exception.type_error.314 | only objects can be unflattened | The @ref unflatten function only works for an object whose keys are JSON Pointers. +json.exception.type_error.315 | values in object must be primitive | The @ref unflatten function only works for an object whose keys are JSON Pointers and whose values are primitive. +json.exception.type_error.316 | invalid UTF-8 byte at index 10: 0x7E | The @ref dump function only works with UTF-8 encoded strings; that is, if you assign a `std::string` to a JSON value, make sure it is UTF-8 encoded. | + +@liveexample{The following code shows how a `type_error` exception can be +caught.,type_error} + +@sa @ref exception for the base class of the library exceptions +@sa @ref parse_error for exceptions indicating a parse error +@sa @ref invalid_iterator for exceptions indicating errors with iterators +@sa @ref out_of_range for exceptions indicating access out of the defined range +@sa @ref other_error for exceptions indicating other library errors + +@since version 3.0.0 +*/ +class type_error : public exception +{ + public: + static type_error create(int id_, const std::string& what_arg) + { + std::string w = exception::name("type_error", id_) + what_arg; + return type_error(id_, w.c_str()); + } + + private: + type_error(int id_, const char* what_arg) : exception(id_, what_arg) {} +}; + +/*! +@brief exception indicating access out of the defined range + +This exception is thrown in case a library function is called on an input +parameter that exceeds the expected range, for instance in case of array +indices or nonexisting object keys. + +Exceptions have ids 4xx. + +name / id | example message | description +------------------------------- | --------------- | ------------------------- +json.exception.out_of_range.401 | array index 3 is out of range | The provided array index @a i is larger than @a size-1. +json.exception.out_of_range.402 | array index '-' (3) is out of range | The special array index `-` in a JSON Pointer never describes a valid element of the array, but the index past the end. That is, it can only be used to add elements at this position, but not to read it. +json.exception.out_of_range.403 | key 'foo' not found | The provided key was not found in the JSON object. +json.exception.out_of_range.404 | unresolved reference token 'foo' | A reference token in a JSON Pointer could not be resolved. +json.exception.out_of_range.405 | JSON pointer has no parent | The JSON Patch operations 'remove' and 'add' can not be applied to the root element of the JSON value. +json.exception.out_of_range.406 | number overflow parsing '10E1000' | A parsed number could not be stored as without changing it to NaN or INF. +json.exception.out_of_range.407 | number overflow serializing '9223372036854775808' | UBJSON only supports integers numbers up to 9223372036854775807. | +json.exception.out_of_range.408 | excessive array size: 8658170730974374167 | The size (following `#`) of an UBJSON array or object exceeds the maximal capacity. | + +@liveexample{The following code shows how an `out_of_range` exception can be +caught.,out_of_range} + +@sa @ref exception for the base class of the library exceptions +@sa @ref parse_error for exceptions indicating a parse error +@sa @ref invalid_iterator for exceptions indicating errors with iterators +@sa @ref type_error for exceptions indicating executing a member function with + a wrong type +@sa @ref other_error for exceptions indicating other library errors + +@since version 3.0.0 +*/ +class out_of_range : public exception +{ + public: + static out_of_range create(int id_, const std::string& what_arg) + { + std::string w = exception::name("out_of_range", id_) + what_arg; + return out_of_range(id_, w.c_str()); + } + + private: + out_of_range(int id_, const char* what_arg) : exception(id_, what_arg) {} +}; + +/*! +@brief exception indicating other library errors + +This exception is thrown in case of errors that cannot be classified with the +other exception types. + +Exceptions have ids 5xx. + +name / id | example message | description +------------------------------ | --------------- | ------------------------- +json.exception.other_error.501 | unsuccessful: {"op":"test","path":"/baz", "value":"bar"} | A JSON Patch operation 'test' failed. The unsuccessful operation is also printed. + +@sa @ref exception for the base class of the library exceptions +@sa @ref parse_error for exceptions indicating a parse error +@sa @ref invalid_iterator for exceptions indicating errors with iterators +@sa @ref type_error for exceptions indicating executing a member function with + a wrong type +@sa @ref out_of_range for exceptions indicating access out of the defined range + +@liveexample{The following code shows how an `other_error` exception can be +caught.,other_error} + +@since version 3.0.0 +*/ +class other_error : public exception +{ + public: + static other_error create(int id_, const std::string& what_arg) + { + std::string w = exception::name("other_error", id_) + what_arg; + return other_error(id_, w.c_str()); + } + + private: + other_error(int id_, const char* what_arg) : exception(id_, what_arg) {} +}; +} +} diff --git a/include/lib/modernjson/detail/input/binary_reader.hpp b/include/lib/modernjson/detail/input/binary_reader.hpp new file mode 100644 index 0000000..103cbf7 --- /dev/null +++ b/include/lib/modernjson/detail/input/binary_reader.hpp @@ -0,0 +1,1697 @@ +#pragma once + +#include <algorithm> // generate_n +#include <array> // array +#include <cassert> // assert +#include <cmath> // ldexp +#include <cstddef> // size_t +#include <cstdint> // uint8_t, uint16_t, uint32_t, uint64_t +#include <cstdio> // snprintf +#include <cstring> // memcpy +#include <iterator> // back_inserter +#include <limits> // numeric_limits +#include <string> // char_traits, string +#include <utility> // make_pair, move + +#include <lib/modernjson/detail/input/input_adapters.hpp> +#include <lib/modernjson/detail/input/json_sax.hpp> +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/meta/is_sax.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +/////////////////// +// binary reader // +/////////////////// + +/*! +@brief deserialization of CBOR, MessagePack, and UBJSON values +*/ +template<typename BasicJsonType, typename SAX = json_sax_dom_parser<BasicJsonType>> +class binary_reader +{ + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + using json_sax_t = SAX; + + public: + /*! + @brief create a binary reader + + @param[in] adapter input adapter to read from + */ + explicit binary_reader(input_adapter_t adapter) : ia(std::move(adapter)) + { + (void)detail::is_sax_static_asserts<SAX, BasicJsonType> {}; + assert(ia); + } + + /*! + @param[in] format the binary format to parse + @param[in] sax_ a SAX event processor + @param[in] strict whether to expect the input to be consumed completed + + @return + */ + bool sax_parse(const input_format_t format, + json_sax_t* sax_, + const bool strict = true) + { + sax = sax_; + bool result = false; + + switch (format) + { + case input_format_t::cbor: + result = parse_cbor_internal(); + break; + + case input_format_t::msgpack: + result = parse_msgpack_internal(); + break; + + case input_format_t::ubjson: + result = parse_ubjson_internal(); + break; + + // LCOV_EXCL_START + default: + assert(false); + // LCOV_EXCL_STOP + } + + // strict mode: next byte must be EOF + if (result and strict) + { + if (format == input_format_t::ubjson) + { + get_ignore_noop(); + } + else + { + get(); + } + + if (JSON_UNLIKELY(current != std::char_traits<char>::eof())) + { + return sax->parse_error(chars_read, get_token_string(), parse_error::create(110, chars_read, "expected end of input")); + } + } + + return result; + } + + /*! + @brief determine system byte order + + @return true if and only if system's byte order is little endian + + @note from http://stackoverflow.com/a/1001328/266378 + */ + static constexpr bool little_endianess(int num = 1) noexcept + { + return (*reinterpret_cast<char*>(&num) == 1); + } + + private: + /*! + @param[in] get_char whether a new character should be retrieved from the + input (true, default) or whether the last read + character should be considered instead + + @return whether a valid CBOR value was passed to the SAX parser + */ + bool parse_cbor_internal(const bool get_char = true) + { + switch (get_char ? get() : current) + { + // EOF + case std::char_traits<char>::eof(): + return unexpect_eof(); + + // Integer 0x00..0x17 (0..23) + case 0x00: + case 0x01: + case 0x02: + case 0x03: + case 0x04: + case 0x05: + case 0x06: + case 0x07: + case 0x08: + case 0x09: + case 0x0A: + case 0x0B: + case 0x0C: + case 0x0D: + case 0x0E: + case 0x0F: + case 0x10: + case 0x11: + case 0x12: + case 0x13: + case 0x14: + case 0x15: + case 0x16: + case 0x17: + return sax->number_unsigned(static_cast<number_unsigned_t>(current)); + + case 0x18: // Unsigned integer (one-byte uint8_t follows) + { + uint8_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0x19: // Unsigned integer (two-byte uint16_t follows) + { + uint16_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0x1A: // Unsigned integer (four-byte uint32_t follows) + { + uint32_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0x1B: // Unsigned integer (eight-byte uint64_t follows) + { + uint64_t number; + return get_number(number) and sax->number_unsigned(number); + } + + // Negative integer -1-0x00..-1-0x17 (-1..-24) + case 0x20: + case 0x21: + case 0x22: + case 0x23: + case 0x24: + case 0x25: + case 0x26: + case 0x27: + case 0x28: + case 0x29: + case 0x2A: + case 0x2B: + case 0x2C: + case 0x2D: + case 0x2E: + case 0x2F: + case 0x30: + case 0x31: + case 0x32: + case 0x33: + case 0x34: + case 0x35: + case 0x36: + case 0x37: + return sax->number_integer(static_cast<int8_t>(0x20 - 1 - current)); + + case 0x38: // Negative integer (one-byte uint8_t follows) + { + uint8_t number; + return get_number(number) and sax->number_integer(static_cast<number_integer_t>(-1) - number); + } + + case 0x39: // Negative integer -1-n (two-byte uint16_t follows) + { + uint16_t number; + return get_number(number) and sax->number_integer(static_cast<number_integer_t>(-1) - number); + } + + case 0x3A: // Negative integer -1-n (four-byte uint32_t follows) + { + uint32_t number; + return get_number(number) and sax->number_integer(static_cast<number_integer_t>(-1) - number); + } + + case 0x3B: // Negative integer -1-n (eight-byte uint64_t follows) + { + uint64_t number; + return get_number(number) and sax->number_integer(static_cast<number_integer_t>(-1) + - static_cast<number_integer_t>(number)); + } + + // UTF-8 string (0x00..0x17 bytes follow) + case 0x60: + case 0x61: + case 0x62: + case 0x63: + case 0x64: + case 0x65: + case 0x66: + case 0x67: + case 0x68: + case 0x69: + case 0x6A: + case 0x6B: + case 0x6C: + case 0x6D: + case 0x6E: + case 0x6F: + case 0x70: + case 0x71: + case 0x72: + case 0x73: + case 0x74: + case 0x75: + case 0x76: + case 0x77: + case 0x78: // UTF-8 string (one-byte uint8_t for n follows) + case 0x79: // UTF-8 string (two-byte uint16_t for n follow) + case 0x7A: // UTF-8 string (four-byte uint32_t for n follow) + case 0x7B: // UTF-8 string (eight-byte uint64_t for n follow) + case 0x7F: // UTF-8 string (indefinite length) + { + string_t s; + return get_cbor_string(s) and sax->string(s); + } + + // array (0x00..0x17 data items follow) + case 0x80: + case 0x81: + case 0x82: + case 0x83: + case 0x84: + case 0x85: + case 0x86: + case 0x87: + case 0x88: + case 0x89: + case 0x8A: + case 0x8B: + case 0x8C: + case 0x8D: + case 0x8E: + case 0x8F: + case 0x90: + case 0x91: + case 0x92: + case 0x93: + case 0x94: + case 0x95: + case 0x96: + case 0x97: + return get_cbor_array(static_cast<std::size_t>(current & 0x1F)); + + case 0x98: // array (one-byte uint8_t for n follows) + { + uint8_t len; + return get_number(len) and get_cbor_array(static_cast<std::size_t>(len)); + } + + case 0x99: // array (two-byte uint16_t for n follow) + { + uint16_t len; + return get_number(len) and get_cbor_array(static_cast<std::size_t>(len)); + } + + case 0x9A: // array (four-byte uint32_t for n follow) + { + uint32_t len; + return get_number(len) and get_cbor_array(static_cast<std::size_t>(len)); + } + + case 0x9B: // array (eight-byte uint64_t for n follow) + { + uint64_t len; + return get_number(len) and get_cbor_array(static_cast<std::size_t>(len)); + } + + case 0x9F: // array (indefinite length) + return get_cbor_array(std::size_t(-1)); + + // map (0x00..0x17 pairs of data items follow) + case 0xA0: + case 0xA1: + case 0xA2: + case 0xA3: + case 0xA4: + case 0xA5: + case 0xA6: + case 0xA7: + case 0xA8: + case 0xA9: + case 0xAA: + case 0xAB: + case 0xAC: + case 0xAD: + case 0xAE: + case 0xAF: + case 0xB0: + case 0xB1: + case 0xB2: + case 0xB3: + case 0xB4: + case 0xB5: + case 0xB6: + case 0xB7: + return get_cbor_object(static_cast<std::size_t>(current & 0x1F)); + + case 0xB8: // map (one-byte uint8_t for n follows) + { + uint8_t len; + return get_number(len) and get_cbor_object(static_cast<std::size_t>(len)); + } + + case 0xB9: // map (two-byte uint16_t for n follow) + { + uint16_t len; + return get_number(len) and get_cbor_object(static_cast<std::size_t>(len)); + } + + case 0xBA: // map (four-byte uint32_t for n follow) + { + uint32_t len; + return get_number(len) and get_cbor_object(static_cast<std::size_t>(len)); + } + + case 0xBB: // map (eight-byte uint64_t for n follow) + { + uint64_t len; + return get_number(len) and get_cbor_object(static_cast<std::size_t>(len)); + } + + case 0xBF: // map (indefinite length) + return get_cbor_object(std::size_t(-1)); + + case 0xF4: // false + return sax->boolean(false); + + case 0xF5: // true + return sax->boolean(true); + + case 0xF6: // null + return sax->null(); + + case 0xF9: // Half-Precision Float (two-byte IEEE 754) + { + const int byte1 = get(); + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + const int byte2 = get(); + if (JSON_UNLIKELY(not unexpect_eof())) + { + 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; + assert(0 <= exp and exp <= 32); + assert(0 <= mant and mant <= 1024); + switch (exp) + { + case 0: + return std::ldexp(mant, -24); + case 31: + return (mant == 0) + ? std::numeric_limits<double>::infinity() + : std::numeric_limits<double>::quiet_NaN(); + default: + return std::ldexp(mant + 1024, exp - 25); + } + }(); + return sax->number_float((half & 0x8000) != 0 + ? static_cast<number_float_t>(-val) + : static_cast<number_float_t>(val), ""); + } + + case 0xFA: // Single-Precision Float (four-byte IEEE 754) + { + float number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + case 0xFB: // Double-Precision Float (eight-byte IEEE 754) + { + double number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + default: // anything else (0xFF is handled inside the other types) + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, "error reading CBOR; last byte: 0x" + last_token)); + } + } + } + + /*! + @return whether a valid MessagePack value was passed to the SAX parser + */ + bool parse_msgpack_internal() + { + switch (get()) + { + // EOF + case std::char_traits<char>::eof(): + return unexpect_eof(); + + // positive fixint + case 0x00: + case 0x01: + case 0x02: + case 0x03: + case 0x04: + case 0x05: + case 0x06: + case 0x07: + case 0x08: + case 0x09: + case 0x0A: + case 0x0B: + case 0x0C: + case 0x0D: + case 0x0E: + case 0x0F: + case 0x10: + case 0x11: + case 0x12: + case 0x13: + case 0x14: + case 0x15: + case 0x16: + case 0x17: + case 0x18: + case 0x19: + case 0x1A: + case 0x1B: + case 0x1C: + case 0x1D: + case 0x1E: + case 0x1F: + case 0x20: + case 0x21: + case 0x22: + case 0x23: + case 0x24: + case 0x25: + case 0x26: + case 0x27: + case 0x28: + case 0x29: + case 0x2A: + case 0x2B: + case 0x2C: + case 0x2D: + case 0x2E: + case 0x2F: + case 0x30: + case 0x31: + case 0x32: + case 0x33: + case 0x34: + case 0x35: + case 0x36: + case 0x37: + case 0x38: + case 0x39: + case 0x3A: + case 0x3B: + case 0x3C: + case 0x3D: + case 0x3E: + case 0x3F: + case 0x40: + case 0x41: + case 0x42: + case 0x43: + case 0x44: + case 0x45: + case 0x46: + case 0x47: + case 0x48: + case 0x49: + case 0x4A: + case 0x4B: + case 0x4C: + case 0x4D: + case 0x4E: + case 0x4F: + case 0x50: + case 0x51: + case 0x52: + case 0x53: + case 0x54: + case 0x55: + case 0x56: + case 0x57: + case 0x58: + case 0x59: + case 0x5A: + case 0x5B: + case 0x5C: + case 0x5D: + case 0x5E: + case 0x5F: + case 0x60: + case 0x61: + case 0x62: + case 0x63: + case 0x64: + case 0x65: + case 0x66: + case 0x67: + case 0x68: + case 0x69: + case 0x6A: + case 0x6B: + case 0x6C: + case 0x6D: + case 0x6E: + case 0x6F: + case 0x70: + case 0x71: + case 0x72: + case 0x73: + case 0x74: + case 0x75: + case 0x76: + case 0x77: + case 0x78: + case 0x79: + case 0x7A: + case 0x7B: + case 0x7C: + case 0x7D: + case 0x7E: + case 0x7F: + return sax->number_unsigned(static_cast<number_unsigned_t>(current)); + + // fixmap + case 0x80: + case 0x81: + case 0x82: + case 0x83: + case 0x84: + case 0x85: + case 0x86: + case 0x87: + case 0x88: + case 0x89: + case 0x8A: + case 0x8B: + case 0x8C: + case 0x8D: + case 0x8E: + case 0x8F: + return get_msgpack_object(static_cast<std::size_t>(current & 0x0F)); + + // fixarray + case 0x90: + case 0x91: + case 0x92: + case 0x93: + case 0x94: + case 0x95: + case 0x96: + case 0x97: + case 0x98: + case 0x99: + case 0x9A: + case 0x9B: + case 0x9C: + case 0x9D: + case 0x9E: + case 0x9F: + return get_msgpack_array(static_cast<std::size_t>(current & 0x0F)); + + // fixstr + case 0xA0: + case 0xA1: + case 0xA2: + case 0xA3: + case 0xA4: + case 0xA5: + case 0xA6: + case 0xA7: + case 0xA8: + case 0xA9: + case 0xAA: + case 0xAB: + case 0xAC: + case 0xAD: + case 0xAE: + case 0xAF: + case 0xB0: + case 0xB1: + case 0xB2: + case 0xB3: + case 0xB4: + case 0xB5: + case 0xB6: + case 0xB7: + case 0xB8: + case 0xB9: + case 0xBA: + case 0xBB: + case 0xBC: + case 0xBD: + case 0xBE: + case 0xBF: + { + string_t s; + return get_msgpack_string(s) and sax->string(s); + } + + case 0xC0: // nil + return sax->null(); + + case 0xC2: // false + return sax->boolean(false); + + case 0xC3: // true + return sax->boolean(true); + + case 0xCA: // float 32 + { + float number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + case 0xCB: // float 64 + { + double number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + case 0xCC: // uint 8 + { + uint8_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0xCD: // uint 16 + { + uint16_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0xCE: // uint 32 + { + uint32_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0xCF: // uint 64 + { + uint64_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 0xD0: // int 8 + { + int8_t number; + return get_number(number) and sax->number_integer(number); + } + + case 0xD1: // int 16 + { + int16_t number; + return get_number(number) and sax->number_integer(number); + } + + case 0xD2: // int 32 + { + int32_t number; + return get_number(number) and sax->number_integer(number); + } + + case 0xD3: // int 64 + { + int64_t number; + return get_number(number) and sax->number_integer(number); + } + + case 0xD9: // str 8 + case 0xDA: // str 16 + case 0xDB: // str 32 + { + string_t s; + return get_msgpack_string(s) and sax->string(s); + } + + case 0xDC: // array 16 + { + uint16_t len; + return get_number(len) and get_msgpack_array(static_cast<std::size_t>(len)); + } + + case 0xDD: // array 32 + { + uint32_t len; + return get_number(len) and get_msgpack_array(static_cast<std::size_t>(len)); + } + + case 0xDE: // map 16 + { + uint16_t len; + return get_number(len) and get_msgpack_object(static_cast<std::size_t>(len)); + } + + case 0xDF: // map 32 + { + uint32_t len; + return get_number(len) and get_msgpack_object(static_cast<std::size_t>(len)); + } + + // negative fixint + case 0xE0: + case 0xE1: + case 0xE2: + case 0xE3: + case 0xE4: + case 0xE5: + case 0xE6: + case 0xE7: + case 0xE8: + case 0xE9: + case 0xEA: + case 0xEB: + case 0xEC: + case 0xED: + case 0xEE: + case 0xEF: + case 0xF0: + case 0xF1: + case 0xF2: + case 0xF3: + case 0xF4: + case 0xF5: + case 0xF6: + case 0xF7: + case 0xF8: + case 0xF9: + case 0xFA: + case 0xFB: + case 0xFC: + case 0xFD: + case 0xFE: + case 0xFF: + return sax->number_integer(static_cast<int8_t>(current)); + + default: // anything else + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, "error reading MessagePack; last byte: 0x" + last_token)); + } + } + } + + /*! + @param[in] get_char whether a new character should be retrieved from the + input (true, default) or whether the last read + character should be considered instead + + @return whether a valid UBJSON value was passed to the SAX parser + */ + bool parse_ubjson_internal(const bool get_char = true) + { + return get_ubjson_value(get_char ? get_ignore_noop() : current); + } + + /*! + @brief get next character from the input + + This function provides the interface to the used input adapter. It does + not throw in case the input reached EOF, but returns a -'ve valued + `std::char_traits<char>::eof()` in that case. + + @return character read from the input + */ + int get() + { + ++chars_read; + return (current = ia->get_character()); + } + + /*! + @return character read from the input after ignoring all 'N' entries + */ + int get_ignore_noop() + { + do + { + get(); + } + while (current == 'N'); + + return current; + } + + /* + @brief read a number from the input + + @tparam NumberType the type of the number + @param[out] result number of type @a NumberType + + @return whether conversion completed + + @note This function needs to respect the system's endianess, because + bytes in CBOR, MessagePack, and UBJSON are stored in network order + (big endian) and therefore need reordering on little endian systems. + */ + template<typename NumberType> + bool get_number(NumberType& result) + { + // step 1: read input into array with system's byte order + std::array<uint8_t, sizeof(NumberType)> vec; + for (std::size_t i = 0; i < sizeof(NumberType); ++i) + { + get(); + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + + // reverse byte order prior to conversion if necessary + if (is_little_endian) + { + vec[sizeof(NumberType) - i - 1] = static_cast<uint8_t>(current); + } + else + { + vec[i] = static_cast<uint8_t>(current); // LCOV_EXCL_LINE + } + } + + // step 2: convert array into number of type T and return + std::memcpy(&result, vec.data(), sizeof(NumberType)); + return true; + } + + /*! + @brief create a string by reading characters from the input + + @tparam NumberType the type of the number + @param[in] len number of characters to read + @param[out] string created by reading @a len bytes + + @return whether string creation completed + + @note We can not reserve @a len bytes for the result, because @a len + may be too large. Usually, @ref unexpect_eof() detects the end of + the input before we run out of string memory. + */ + template<typename NumberType> + bool get_string(const NumberType len, string_t& result) + { + bool success = true; + std::generate_n(std::back_inserter(result), len, [this, &success]() + { + get(); + if (JSON_UNLIKELY(not unexpect_eof())) + { + success = false; + } + return static_cast<char>(current); + }); + return success; + } + + /*! + @brief reads a CBOR string + + This function first reads starting bytes to determine the expected + string length and then copies this number of bytes into a string. + Additionally, CBOR's strings with indefinite lengths are supported. + + @param[out] result created string + + @return whether string creation completed + */ + bool get_cbor_string(string_t& result) + { + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + + switch (current) + { + // UTF-8 string (0x00..0x17 bytes follow) + case 0x60: + case 0x61: + case 0x62: + case 0x63: + case 0x64: + case 0x65: + case 0x66: + case 0x67: + case 0x68: + case 0x69: + case 0x6A: + case 0x6B: + case 0x6C: + case 0x6D: + case 0x6E: + case 0x6F: + case 0x70: + case 0x71: + case 0x72: + case 0x73: + case 0x74: + case 0x75: + case 0x76: + case 0x77: + { + return get_string(current & 0x1F, result); + } + + case 0x78: // UTF-8 string (one-byte uint8_t for n follows) + { + uint8_t len; + return get_number(len) and get_string(len, result); + } + + case 0x79: // UTF-8 string (two-byte uint16_t for n follow) + { + uint16_t len; + return get_number(len) and get_string(len, result); + } + + case 0x7A: // UTF-8 string (four-byte uint32_t for n follow) + { + uint32_t len; + return get_number(len) and get_string(len, result); + } + + case 0x7B: // UTF-8 string (eight-byte uint64_t for n follow) + { + uint64_t len; + return get_number(len) and get_string(len, result); + } + + case 0x7F: // UTF-8 string (indefinite length) + { + while (get() != 0xFF) + { + string_t chunk; + if (not get_cbor_string(chunk)) + { + return false; + } + result.append(chunk); + } + return true; + } + + default: + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, "expected a CBOR string; last byte: 0x" + last_token)); + } + } + } + + /*! + @param[in] len the length of the array or std::size_t(-1) for an + array of indefinite size + @return whether array creation completed + */ + bool get_cbor_array(const std::size_t len) + { + if (JSON_UNLIKELY(not sax->start_array(len))) + { + return false; + } + + if (len != std::size_t(-1)) + for (std::size_t i = 0; i < len; ++i) + { + if (JSON_UNLIKELY(not parse_cbor_internal())) + { + return false; + } + } + else + { + while (get() != 0xFF) + { + if (JSON_UNLIKELY(not parse_cbor_internal(false))) + { + return false; + } + } + } + + return sax->end_array(); + } + + /*! + @param[in] len the length of the object or std::size_t(-1) for an + object of indefinite size + @return whether object creation completed + */ + bool get_cbor_object(const std::size_t len) + { + if (not JSON_UNLIKELY(sax->start_object(len))) + { + return false; + } + + string_t key; + if (len != std::size_t(-1)) + { + for (std::size_t i = 0; i < len; ++i) + { + get(); + if (JSON_UNLIKELY(not get_cbor_string(key) or not sax->key(key))) + { + return false; + } + + if (JSON_UNLIKELY(not parse_cbor_internal())) + { + return false; + } + key.clear(); + } + } + else + { + while (get() != 0xFF) + { + if (JSON_UNLIKELY(not get_cbor_string(key) or not sax->key(key))) + { + return false; + } + + if (JSON_UNLIKELY(not parse_cbor_internal())) + { + return false; + } + key.clear(); + } + } + + return sax->end_object(); + } + + /*! + @brief reads a MessagePack string + + This function first reads starting bytes to determine the expected + string length and then copies this number of bytes into a string. + + @param[out] result created string + + @return whether string creation completed + */ + bool get_msgpack_string(string_t& result) + { + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + + switch (current) + { + // fixstr + case 0xA0: + case 0xA1: + case 0xA2: + case 0xA3: + case 0xA4: + case 0xA5: + case 0xA6: + case 0xA7: + case 0xA8: + case 0xA9: + case 0xAA: + case 0xAB: + case 0xAC: + case 0xAD: + case 0xAE: + case 0xAF: + case 0xB0: + case 0xB1: + case 0xB2: + case 0xB3: + case 0xB4: + case 0xB5: + case 0xB6: + case 0xB7: + case 0xB8: + case 0xB9: + case 0xBA: + case 0xBB: + case 0xBC: + case 0xBD: + case 0xBE: + case 0xBF: + { + return get_string(current & 0x1F, result); + } + + case 0xD9: // str 8 + { + uint8_t len; + return get_number(len) and get_string(len, result); + } + + case 0xDA: // str 16 + { + uint16_t len; + return get_number(len) and get_string(len, result); + } + + case 0xDB: // str 32 + { + uint32_t len; + return get_number(len) and get_string(len, result); + } + + default: + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, "expected a MessagePack string; last byte: 0x" + last_token)); + } + } + } + + /*! + @param[in] len the length of the array + @return whether array creation completed + */ + bool get_msgpack_array(const std::size_t len) + { + if (JSON_UNLIKELY(not sax->start_array(len))) + { + return false; + } + + for (std::size_t i = 0; i < len; ++i) + { + if (JSON_UNLIKELY(not parse_msgpack_internal())) + { + return false; + } + } + + return sax->end_array(); + } + + /*! + @param[in] len the length of the object + @return whether object creation completed + */ + bool get_msgpack_object(const std::size_t len) + { + if (JSON_UNLIKELY(not sax->start_object(len))) + { + return false; + } + + string_t key; + for (std::size_t i = 0; i < len; ++i) + { + get(); + if (JSON_UNLIKELY(not get_msgpack_string(key) or not sax->key(key))) + { + return false; + } + + if (JSON_UNLIKELY(not parse_msgpack_internal())) + { + return false; + } + key.clear(); + } + + return sax->end_object(); + } + + /*! + @brief reads a UBJSON string + + This function is either called after reading the 'S' byte explicitly + indicating a string, or in case of an object key where the 'S' byte can be + left out. + + @param[out] result created string + @param[in] get_char whether a new character should be retrieved from the + input (true, default) or whether the last read + character should be considered instead + + @return whether string creation completed + */ + bool get_ubjson_string(string_t& result, const bool get_char = true) + { + if (get_char) + { + get(); // TODO: may we ignore N here? + } + + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + + switch (current) + { + case 'U': + { + uint8_t len; + return get_number(len) and get_string(len, result); + } + + case 'i': + { + int8_t len; + return get_number(len) and get_string(len, result); + } + + case 'I': + { + int16_t len; + return get_number(len) and get_string(len, result); + } + + case 'l': + { + int32_t len; + return get_number(len) and get_string(len, result); + } + + case 'L': + { + int64_t len; + return get_number(len) and get_string(len, result); + } + + default: + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, "expected a UBJSON string; last byte: 0x" + last_token)); + } + } + + /*! + @param[out] result determined size + @return whether size determination completed + */ + bool get_ubjson_size_value(std::size_t& result) + { + switch (get_ignore_noop()) + { + case 'U': + { + uint8_t number; + if (JSON_UNLIKELY(not get_number(number))) + { + return false; + } + result = static_cast<std::size_t>(number); + return true; + } + + case 'i': + { + int8_t number; + if (JSON_UNLIKELY(not get_number(number))) + { + return false; + } + result = static_cast<std::size_t>(number); + return true; + } + + case 'I': + { + int16_t number; + if (JSON_UNLIKELY(not get_number(number))) + { + return false; + } + result = static_cast<std::size_t>(number); + return true; + } + + case 'l': + { + int32_t number; + if (JSON_UNLIKELY(not get_number(number))) + { + return false; + } + result = static_cast<std::size_t>(number); + return true; + } + + case 'L': + { + int64_t number; + if (JSON_UNLIKELY(not get_number(number))) + { + return false; + } + result = static_cast<std::size_t>(number); + return true; + } + + default: + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, "byte after '#' must denote a number type; last byte: 0x" + last_token)); + } + } + } + + /*! + @brief determine the type and size for a container + + In the optimized UBJSON format, a type and a size can be provided to allow + for a more compact representation. + + @param[out] result pair of the size and the type + + @return whether pair creation completed + */ + bool get_ubjson_size_type(std::pair<std::size_t, int>& result) + { + result.first = string_t::npos; // size + result.second = 0; // type + + get_ignore_noop(); + + if (current == '$') + { + result.second = get(); // must not ignore 'N', because 'N' maybe the type + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + + get_ignore_noop(); + if (JSON_UNLIKELY(current != '#')) + { + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, "expected '#' after UBJSON type information; last byte: 0x" + last_token)); + } + + return get_ubjson_size_value(result.first); + } + else if (current == '#') + { + return get_ubjson_size_value(result.first); + } + return true; + } + + /*! + @param prefix the previously read or set type prefix + @return whether value creation completed + */ + bool get_ubjson_value(const int prefix) + { + switch (prefix) + { + case std::char_traits<char>::eof(): // EOF + return unexpect_eof(); + + case 'T': // true + return sax->boolean(true); + case 'F': // false + return sax->boolean(false); + + case 'Z': // null + return sax->null(); + + case 'U': + { + uint8_t number; + return get_number(number) and sax->number_unsigned(number); + } + + case 'i': + { + int8_t number; + return get_number(number) and sax->number_integer(number); + } + + case 'I': + { + int16_t number; + return get_number(number) and sax->number_integer(number); + } + + case 'l': + { + int32_t number; + return get_number(number) and sax->number_integer(number); + } + + case 'L': + { + int64_t number; + return get_number(number) and sax->number_integer(number); + } + + case 'd': + { + float number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + case 'D': + { + double number; + return get_number(number) and sax->number_float(static_cast<number_float_t>(number), ""); + } + + case 'C': // char + { + get(); + if (JSON_UNLIKELY(not unexpect_eof())) + { + return false; + } + if (JSON_UNLIKELY(current > 127)) + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(113, chars_read, "byte after 'C' must be in range 0x00..0x7F; last byte: 0x" + last_token)); + } + string_t s(1, static_cast<char>(current)); + return sax->string(s); + } + + case 'S': // string + { + string_t s; + return get_ubjson_string(s) and sax->string(s); + } + + case '[': // array + return get_ubjson_array(); + + case '{': // object + return get_ubjson_object(); + + default: // anything else + { + auto last_token = get_token_string(); + return sax->parse_error(chars_read, last_token, parse_error::create(112, chars_read, "error reading UBJSON; last byte: 0x" + last_token)); + } + } + } + + /*! + @return whether array creation completed + */ + bool get_ubjson_array() + { + std::pair<std::size_t, int> size_and_type; + if (JSON_UNLIKELY(not get_ubjson_size_type(size_and_type))) + { + return false; + } + + if (size_and_type.first != string_t::npos) + { + if (JSON_UNLIKELY(not sax->start_array(size_and_type.first))) + { + return false; + } + + 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_UNLIKELY(not get_ubjson_value(size_and_type.second))) + { + return false; + } + } + } + } + else + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_UNLIKELY(not parse_ubjson_internal())) + { + return false; + } + } + } + } + else + { + if (JSON_UNLIKELY(not sax->start_array(std::size_t(-1)))) + { + return false; + } + + while (current != ']') + { + if (JSON_UNLIKELY(not parse_ubjson_internal(false))) + { + return false; + } + get_ignore_noop(); + } + } + + return sax->end_array(); + } + + /*! + @return whether object creation completed + */ + bool get_ubjson_object() + { + std::pair<std::size_t, int> size_and_type; + if (JSON_UNLIKELY(not get_ubjson_size_type(size_and_type))) + { + return false; + } + + string_t key; + if (size_and_type.first != string_t::npos) + { + if (JSON_UNLIKELY(not sax->start_object(size_and_type.first))) + { + return false; + } + + if (size_and_type.second != 0) + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_UNLIKELY(not get_ubjson_string(key) or not sax->key(key))) + { + return false; + } + if (JSON_UNLIKELY(not get_ubjson_value(size_and_type.second))) + { + return false; + } + key.clear(); + } + } + else + { + for (std::size_t i = 0; i < size_and_type.first; ++i) + { + if (JSON_UNLIKELY(not get_ubjson_string(key) or not sax->key(key))) + { + return false; + } + if (JSON_UNLIKELY(not parse_ubjson_internal())) + { + return false; + } + key.clear(); + } + } + } + else + { + if (JSON_UNLIKELY(not sax->start_object(std::size_t(-1)))) + { + return false; + } + + while (current != '}') + { + if (JSON_UNLIKELY(not get_ubjson_string(key, false) or not sax->key(key))) + { + return false; + } + if (JSON_UNLIKELY(not parse_ubjson_internal())) + { + return false; + } + get_ignore_noop(); + key.clear(); + } + } + + return sax->end_object(); + } + + /*! + @return whether the last read character is not EOF + */ + bool unexpect_eof() const + { + if (JSON_UNLIKELY(current == std::char_traits<char>::eof())) + { + return sax->parse_error(chars_read, "<end of file>", parse_error::create(110, chars_read, "unexpected end of input")); + } + return true; + } + + /*! + @return a string representation of the last read byte + */ + std::string get_token_string() const + { + char cr[3]; + snprintf(cr, 3, "%.2hhX", static_cast<unsigned char>(current)); + return std::string{cr}; + } + + private: + /// input adapter + input_adapter_t ia = nullptr; + + /// the current character + int current = std::char_traits<char>::eof(); + + /// the number of characters read + std::size_t chars_read = 0; + + /// whether we can assume little endianess + const bool is_little_endian = little_endianess(); + + /// the SAX parser + json_sax_t* sax = nullptr; +}; +} +} diff --git a/include/lib/modernjson/detail/input/input_adapters.hpp b/include/lib/modernjson/detail/input/input_adapters.hpp new file mode 100644 index 0000000..5abaee3 --- /dev/null +++ b/include/lib/modernjson/detail/input/input_adapters.hpp @@ -0,0 +1,377 @@ +#pragma once + +#include <cassert> // assert +#include <cstddef> // size_t +#include <cstring> // strlen +#include <istream> // istream +#include <iterator> // begin, end, iterator_traits, random_access_iterator_tag, distance, next +#include <memory> // shared_ptr, make_shared, addressof +#include <numeric> // accumulate +#include <string> // string, char_traits +#include <type_traits> // enable_if, is_base_of, is_pointer, is_integral, remove_pointer +#include <utility> // pair, declval + +#include <lib/modernjson/detail/macro_scope.hpp> + +namespace nlohmann +{ +namespace detail +{ +/// the supported input formats +enum class input_format_t { json, cbor, msgpack, ubjson }; + +//////////////////// +// input adapters // +//////////////////// + +/*! +@brief abstract input adapter interface + +Produces a stream of std::char_traits<char>::int_type characters from a +std::istream, a buffer, or some other input type. Accepts the return of +exactly one non-EOF character for future input. The int_type characters +returned consist of all valid char values as positive values (typically +unsigned char), plus an EOF value outside that range, specified by the value +of the function std::char_traits<char>::eof(). This value is typically -1, but +could be any arbitrary value which is not a valid char value. +*/ +struct input_adapter_protocol +{ + /// get a character [0,255] or std::char_traits<char>::eof(). + virtual std::char_traits<char>::int_type get_character() = 0; + virtual ~input_adapter_protocol() = default; +}; + +/// a type to simplify interfaces +using input_adapter_t = std::shared_ptr<input_adapter_protocol>; + +/*! +Input adapter for a (caching) istream. Ignores a UFT Byte Order Mark at +beginning of input. Does not support changing the underlying std::streambuf +in mid-input. Maintains underlying std::istream and std::streambuf to support +subsequent use of standard std::istream operations to process any input +characters following those used in parsing the JSON input. Clears the +std::istream flags; any input errors (e.g., EOF) will be detected by the first +subsequent call for input from the std::istream. +*/ +class input_stream_adapter : public input_adapter_protocol +{ + public: + ~input_stream_adapter() override + { + // clear stream flags; we use underlying streambuf I/O, do not + // maintain ifstream flags + is.clear(); + } + + explicit input_stream_adapter(std::istream& i) + : is(i), sb(*i.rdbuf()) + {} + + // delete because of pointer members + input_stream_adapter(const input_stream_adapter&) = delete; + input_stream_adapter& operator=(input_stream_adapter&) = delete; + + // std::istream/std::streambuf use std::char_traits<char>::to_int_type, to + // ensure that std::char_traits<char>::eof() and the character 0xFF do not + // end up as the same value, eg. 0xFFFFFFFF. + std::char_traits<char>::int_type get_character() override + { + return sb.sbumpc(); + } + + private: + /// the associated input stream + std::istream& is; + std::streambuf& sb; +}; + +/// input adapter for buffer input +class input_buffer_adapter : public input_adapter_protocol +{ + public: + input_buffer_adapter(const char* b, const std::size_t l) + : cursor(b), limit(b + l) + {} + + // delete because of pointer members + input_buffer_adapter(const input_buffer_adapter&) = delete; + input_buffer_adapter& operator=(input_buffer_adapter&) = delete; + + std::char_traits<char>::int_type get_character() noexcept override + { + if (JSON_LIKELY(cursor < limit)) + { + return std::char_traits<char>::to_int_type(*(cursor++)); + } + + return std::char_traits<char>::eof(); + } + + private: + /// pointer to the current character + const char* cursor; + /// pointer past the last character + const char* const limit; +}; + +template<typename WideStringType> +class wide_string_input_adapter : public input_adapter_protocol +{ + public: + explicit wide_string_input_adapter(const WideStringType& w) : str(w) {} + + std::char_traits<char>::int_type get_character() noexcept override + { + // check if buffer needs to be filled + if (utf8_bytes_index == utf8_bytes_filled) + { + if (sizeof(typename WideStringType::value_type) == 2) + { + fill_buffer_utf16(); + } + else + { + fill_buffer_utf32(); + } + + assert(utf8_bytes_filled > 0); + assert(utf8_bytes_index == 0); + } + + // use buffer + assert(utf8_bytes_filled > 0); + assert(utf8_bytes_index < utf8_bytes_filled); + return utf8_bytes[utf8_bytes_index++]; + } + + private: + void fill_buffer_utf16() + { + utf8_bytes_index = 0; + + if (current_wchar == str.size()) + { + utf8_bytes[0] = std::char_traits<char>::eof(); + utf8_bytes_filled = 1; + } + else + { + // get the current character + const int wc = static_cast<int>(str[current_wchar++]); + + // UTF-16 to UTF-8 encoding + if (wc < 0x80) + { + utf8_bytes[0] = wc; + utf8_bytes_filled = 1; + } + else if (wc <= 0x7FF) + { + utf8_bytes[0] = 0xC0 | ((wc >> 6)); + utf8_bytes[1] = 0x80 | (wc & 0x3F); + utf8_bytes_filled = 2; + } + else if (0xD800 > wc or wc >= 0xE000) + { + utf8_bytes[0] = 0xE0 | ((wc >> 12)); + utf8_bytes[1] = 0x80 | ((wc >> 6) & 0x3F); + utf8_bytes[2] = 0x80 | (wc & 0x3F); + utf8_bytes_filled = 3; + } + else + { + if (current_wchar < str.size()) + { + const int wc2 = static_cast<int>(str[current_wchar++]); + const int charcode = 0x10000 + (((wc & 0x3FF) << 10) | (wc2 & 0x3FF)); + utf8_bytes[0] = 0xf0 | (charcode >> 18); + utf8_bytes[1] = 0x80 | ((charcode >> 12) & 0x3F); + utf8_bytes[2] = 0x80 | ((charcode >> 6) & 0x3F); + utf8_bytes[3] = 0x80 | (charcode & 0x3F); + utf8_bytes_filled = 4; + } + else + { + // unknown character + ++current_wchar; + utf8_bytes[0] = wc; + utf8_bytes_filled = 1; + } + } + } + } + + void fill_buffer_utf32() + { + utf8_bytes_index = 0; + + if (current_wchar == str.size()) + { + utf8_bytes[0] = std::char_traits<char>::eof(); + utf8_bytes_filled = 1; + } + else + { + // get the current character + const int wc = static_cast<int>(str[current_wchar++]); + + // UTF-32 to UTF-8 encoding + if (wc < 0x80) + { + utf8_bytes[0] = wc; + utf8_bytes_filled = 1; + } + else if (wc <= 0x7FF) + { + utf8_bytes[0] = 0xC0 | ((wc >> 6) & 0x1F); + utf8_bytes[1] = 0x80 | (wc & 0x3F); + utf8_bytes_filled = 2; + } + else if (wc <= 0xFFFF) + { + utf8_bytes[0] = 0xE0 | ((wc >> 12) & 0x0F); + utf8_bytes[1] = 0x80 | ((wc >> 6) & 0x3F); + utf8_bytes[2] = 0x80 | (wc & 0x3F); + utf8_bytes_filled = 3; + } + else if (wc <= 0x10FFFF) + { + utf8_bytes[0] = 0xF0 | ((wc >> 18 ) & 0x07); + utf8_bytes[1] = 0x80 | ((wc >> 12) & 0x3F); + utf8_bytes[2] = 0x80 | ((wc >> 6) & 0x3F); + utf8_bytes[3] = 0x80 | (wc & 0x3F); + utf8_bytes_filled = 4; + } + else + { + // unknown character + utf8_bytes[0] = wc; + utf8_bytes_filled = 1; + } + } + } + + private: + /// the wstring to process + const WideStringType& str; + + /// index of the current wchar in str + std::size_t current_wchar = 0; + + /// a buffer for UTF-8 bytes + std::array<std::char_traits<char>::int_type, 4> utf8_bytes = {{0, 0, 0, 0}}; + + /// index to the utf8_codes array for the next valid byte + std::size_t utf8_bytes_index = 0; + /// number of valid bytes in the utf8_codes array + std::size_t utf8_bytes_filled = 0; +}; + +class input_adapter +{ + public: + // native support + + /// input adapter for input stream + input_adapter(std::istream& i) + : ia(std::make_shared<input_stream_adapter>(i)) {} + + /// input adapter for input stream + input_adapter(std::istream&& i) + : ia(std::make_shared<input_stream_adapter>(i)) {} + + input_adapter(const std::wstring& ws) + : ia(std::make_shared<wide_string_input_adapter<std::wstring>>(ws)) {} + + input_adapter(const std::u16string& ws) + : ia(std::make_shared<wide_string_input_adapter<std::u16string>>(ws)) {} + + input_adapter(const std::u32string& ws) + : ia(std::make_shared<wide_string_input_adapter<std::u32string>>(ws)) {} + + /// input adapter for buffer + template<typename CharT, + typename std::enable_if< + std::is_pointer<CharT>::value and + std::is_integral<typename std::remove_pointer<CharT>::type>::value and + sizeof(typename std::remove_pointer<CharT>::type) == 1, + int>::type = 0> + input_adapter(CharT b, std::size_t l) + : ia(std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(b), l)) {} + + // derived support + + /// input adapter for string literal + template<typename CharT, + typename std::enable_if< + std::is_pointer<CharT>::value and + std::is_integral<typename std::remove_pointer<CharT>::type>::value and + sizeof(typename std::remove_pointer<CharT>::type) == 1, + int>::type = 0> + input_adapter(CharT b) + : input_adapter(reinterpret_cast<const char*>(b), + std::strlen(reinterpret_cast<const char*>(b))) {} + + /// input adapter for iterator range with contiguous storage + template<class IteratorType, + typename std::enable_if< + std::is_same<typename std::iterator_traits<IteratorType>::iterator_category, std::random_access_iterator_tag>::value, + int>::type = 0> + input_adapter(IteratorType first, IteratorType last) + { +#ifndef NDEBUG + // assertion to check that the iterator range is indeed contiguous, + // see http://stackoverflow.com/a/35008842/266378 for more discussion + const auto is_contiguous = std::accumulate( + first, last, std::pair<bool, int>(true, 0), + [&first](std::pair<bool, int> res, decltype(*first) val) + { + res.first &= (val == *(std::next(std::addressof(*first), res.second++))); + return res; + }).first; + assert(is_contiguous); +#endif + + // assertion to check that each element is 1 byte long + static_assert( + sizeof(typename std::iterator_traits<IteratorType>::value_type) == 1, + "each element in the iterator range must have the size of 1 byte"); + + const auto len = static_cast<size_t>(std::distance(first, last)); + if (JSON_LIKELY(len > 0)) + { + // there is at least one element: use the address of first + ia = std::make_shared<input_buffer_adapter>(reinterpret_cast<const char*>(&(*first)), len); + } + else + { + // the address of first cannot be used: use nullptr + ia = std::make_shared<input_buffer_adapter>(nullptr, len); + } + } + + /// input adapter for array + template<class T, std::size_t N> + input_adapter(T (&array)[N]) + : input_adapter(std::begin(array), std::end(array)) {} + + /// input adapter for contiguous container + template<class ContiguousContainer, typename + std::enable_if<not std::is_pointer<ContiguousContainer>::value and + std::is_base_of<std::random_access_iterator_tag, typename std::iterator_traits<decltype(std::begin(std::declval<ContiguousContainer const>()))>::iterator_category>::value, + int>::type = 0> + input_adapter(const ContiguousContainer& c) + : input_adapter(std::begin(c), std::end(c)) {} + + operator input_adapter_t() + { + return ia; + } + + private: + /// the actual adapter + input_adapter_t ia = nullptr; +}; +} +} diff --git a/include/lib/modernjson/detail/input/json_sax.hpp b/include/lib/modernjson/detail/input/json_sax.hpp new file mode 100644 index 0000000..968fcd0 --- /dev/null +++ b/include/lib/modernjson/detail/input/json_sax.hpp @@ -0,0 +1,702 @@ +#pragma once + +#include <cstddef> +#include <string> +#include <vector> + +#include <lib/modernjson/detail/input/parser.hpp> +#include <lib/modernjson/detail/exceptions.hpp> + +namespace nlohmann +{ + +/*! +@brief SAX interface + +This class describes the SAX interface used by @ref nlohmann::json::sax_parse. +Each function is called in different situations while the input is parsed. The +boolean return value informs the parser whether to continue processing the +input. +*/ +template<typename BasicJsonType> +struct json_sax +{ + /// type for (signed) integers + using number_integer_t = typename BasicJsonType::number_integer_t; + /// type for unsigned integers + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + /// type for floating-point numbers + using number_float_t = typename BasicJsonType::number_float_t; + /// type for strings + using string_t = typename BasicJsonType::string_t; + + /*! + @brief a null value was read + @return whether parsing should proceed + */ + virtual bool null() = 0; + + /*! + @brief a boolean value was read + @param[in] val boolean value + @return whether parsing should proceed + */ + virtual bool boolean(bool val) = 0; + + /*! + @brief an integer number was read + @param[in] val integer value + @return whether parsing should proceed + */ + virtual bool number_integer(number_integer_t val) = 0; + + /*! + @brief an unsigned integer number was read + @param[in] val unsigned integer value + @return whether parsing should proceed + */ + virtual bool number_unsigned(number_unsigned_t val) = 0; + + /*! + @brief an floating-point number was read + @param[in] val floating-point value + @param[in] s raw token value + @return whether parsing should proceed + */ + virtual bool number_float(number_float_t val, const string_t& s) = 0; + + /*! + @brief a string was read + @param[in] val string value + @return whether parsing should proceed + @note It is safe to move the passed string. + */ + virtual bool string(string_t& val) = 0; + + /*! + @brief the beginning of an object was read + @param[in] elements number of object elements or -1 if unknown + @return whether parsing should proceed + @note binary formats may report the number of elements + */ + virtual bool start_object(std::size_t elements) = 0; + + /*! + @brief an object key was read + @param[in] val object key + @return whether parsing should proceed + @note It is safe to move the passed string. + */ + virtual bool key(string_t& val) = 0; + + /*! + @brief the end of an object was read + @return whether parsing should proceed + */ + virtual bool end_object() = 0; + + /*! + @brief the beginning of an array was read + @param[in] elements number of array elements or -1 if unknown + @return whether parsing should proceed + @note binary formats may report the number of elements + */ + virtual bool start_array(std::size_t elements) = 0; + + /*! + @brief the end of an array was read + @return whether parsing should proceed + */ + virtual bool end_array() = 0; + + /*! + @brief a parse error occurred + @param[in] position the position in the input where the error occurs + @param[in] last_token the last read token + @param[in] error_msg a detailed error message + @return whether parsing should proceed (must return false) + */ + virtual bool parse_error(std::size_t position, + const std::string& last_token, + const detail::exception& ex) = 0; + + virtual ~json_sax() = default; +}; + + +namespace detail +{ +/*! +@brief SAX implementation to create a JSON value from SAX events + +This class implements the @ref json_sax interface and processes the SAX events +to create a JSON value which makes it basically a DOM parser. The structure or +hierarchy of the JSON value is managed by the stack `ref_stack` which contains +a pointer to the respective array or object for each recursion depth. + +After successful parsing, the value that is passed by reference to the +constructor contains the parsed value. + +@tparam BasicJsonType the JSON type +*/ +template<typename BasicJsonType> +class json_sax_dom_parser +{ + public: + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + + /*! + @param[in, out] r reference to a JSON value that is manipulated while + parsing + @param[in] allow_exceptions_ whether parse errors yield exceptions + */ + explicit json_sax_dom_parser(BasicJsonType& r, const bool allow_exceptions_ = true) + : root(r), allow_exceptions(allow_exceptions_) + {} + + bool null() + { + handle_value(nullptr); + return true; + } + + bool boolean(bool val) + { + handle_value(val); + return true; + } + + bool number_integer(number_integer_t val) + { + handle_value(val); + return true; + } + + bool number_unsigned(number_unsigned_t val) + { + handle_value(val); + return true; + } + + bool number_float(number_float_t val, const string_t&) + { + handle_value(val); + return true; + } + + bool string(string_t& val) + { + handle_value(val); + return true; + } + + bool start_object(std::size_t len) + { + ref_stack.push_back(handle_value(BasicJsonType::value_t::object)); + + if (JSON_UNLIKELY(len != std::size_t(-1) and len > ref_stack.back()->max_size())) + { + JSON_THROW(out_of_range::create(408, + "excessive object size: " + std::to_string(len))); + } + + return true; + } + + bool key(string_t& val) + { + // add null at given key and store the reference for later + object_element = &(ref_stack.back()->m_value.object->operator[](val)); + return true; + } + + bool end_object() + { + ref_stack.pop_back(); + return true; + } + + bool start_array(std::size_t len) + { + ref_stack.push_back(handle_value(BasicJsonType::value_t::array)); + + if (JSON_UNLIKELY(len != std::size_t(-1) and len > ref_stack.back()->max_size())) + { + JSON_THROW(out_of_range::create(408, + "excessive array size: " + std::to_string(len))); + } + + return true; + } + + bool end_array() + { + ref_stack.pop_back(); + return true; + } + + bool parse_error(std::size_t, const std::string&, + const detail::exception& ex) + { + errored = true; + if (allow_exceptions) + { + // determine the proper exception type from the id + switch ((ex.id / 100) % 100) + { + case 1: + JSON_THROW(*reinterpret_cast<const detail::parse_error*>(&ex)); + case 4: + JSON_THROW(*reinterpret_cast<const detail::out_of_range*>(&ex)); + // LCOV_EXCL_START + case 2: + JSON_THROW(*reinterpret_cast<const detail::invalid_iterator*>(&ex)); + case 3: + JSON_THROW(*reinterpret_cast<const detail::type_error*>(&ex)); + case 5: + JSON_THROW(*reinterpret_cast<const detail::other_error*>(&ex)); + default: + assert(false); + // LCOV_EXCL_STOP + } + } + return false; + } + + constexpr bool is_errored() const + { + return errored; + } + + private: + /*! + @invariant If the ref stack is empty, then the passed value will be the new + root. + @invariant If the ref stack contains a value, then it is an array or an + object to which we can add elements + */ + template<typename Value> + BasicJsonType* handle_value(Value&& v) + { + if (ref_stack.empty()) + { + root = BasicJsonType(std::forward<Value>(v)); + return &root; + } + else + { + assert(ref_stack.back()->is_array() or ref_stack.back()->is_object()); + if (ref_stack.back()->is_array()) + { + ref_stack.back()->m_value.array->emplace_back(std::forward<Value>(v)); + return &(ref_stack.back()->m_value.array->back()); + } + else + { + assert(object_element); + *object_element = BasicJsonType(std::forward<Value>(v)); + return object_element; + } + } + } + + /// the parsed JSON value + BasicJsonType& root; + /// stack to model hierarchy of values + std::vector<BasicJsonType*> ref_stack; + /// helper to hold the reference for the next object element + BasicJsonType* object_element = nullptr; + /// whether a syntax error occurred + bool errored = false; + /// whether to throw exceptions in case of errors + const bool allow_exceptions = true; +}; + +template<typename BasicJsonType> +class json_sax_dom_callback_parser +{ + public: + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + using parser_callback_t = typename BasicJsonType::parser_callback_t; + using parse_event_t = typename BasicJsonType::parse_event_t; + + json_sax_dom_callback_parser(BasicJsonType& r, + const parser_callback_t cb, + const bool allow_exceptions_ = true) + : root(r), callback(cb), allow_exceptions(allow_exceptions_) + { + keep_stack.push_back(true); + } + + bool null() + { + handle_value(nullptr); + return true; + } + + bool boolean(bool val) + { + handle_value(val); + return true; + } + + bool number_integer(number_integer_t val) + { + handle_value(val); + return true; + } + + bool number_unsigned(number_unsigned_t val) + { + handle_value(val); + return true; + } + + bool number_float(number_float_t val, const string_t&) + { + handle_value(val); + return true; + } + + bool string(string_t& val) + { + handle_value(val); + return true; + } + + bool start_object(std::size_t len) + { + // check callback for object start + const bool keep = callback(static_cast<int>(ref_stack.size()), parse_event_t::object_start, discarded); + keep_stack.push_back(keep); + + auto val = handle_value(BasicJsonType::value_t::object, true); + ref_stack.push_back(val.second); + + // check object limit + if (ref_stack.back()) + { + if (JSON_UNLIKELY(len != std::size_t(-1) and len > ref_stack.back()->max_size())) + { + JSON_THROW(out_of_range::create(408, + "excessive object size: " + std::to_string(len))); + } + } + + return true; + } + + bool key(string_t& val) + { + BasicJsonType k = BasicJsonType(val); + + // check callback for key + const bool keep = callback(static_cast<int>(ref_stack.size()), parse_event_t::key, k); + key_keep_stack.push_back(keep); + + // add discarded value at given key and store the reference for later + if (keep and ref_stack.back()) + { + object_element = &(ref_stack.back()->m_value.object->operator[](val) = discarded); + } + + return true; + } + + bool end_object() + { + if (ref_stack.back()) + { + if (not callback(static_cast<int>(ref_stack.size()) - 1, parse_event_t::object_end, *ref_stack.back())) + { + // discard object + *ref_stack.back() = discarded; + } + } + + assert(not ref_stack.empty()); + assert(not keep_stack.empty()); + ref_stack.pop_back(); + keep_stack.pop_back(); + + if (not ref_stack.empty() and ref_stack.back()) + { + // remove discarded value + if (ref_stack.back()->is_object()) + { + for (auto it = ref_stack.back()->begin(); it != ref_stack.back()->end(); ++it) + { + if (it->is_discarded()) + { + ref_stack.back()->erase(it); + break; + } + } + } + } + + return true; + } + + bool start_array(std::size_t len) + { + const bool keep = callback(static_cast<int>(ref_stack.size()), parse_event_t::array_start, discarded); + keep_stack.push_back(keep); + + auto val = handle_value(BasicJsonType::value_t::array, true); + ref_stack.push_back(val.second); + + // check array limit + if (ref_stack.back()) + { + if (JSON_UNLIKELY(len != std::size_t(-1) and len > ref_stack.back()->max_size())) + { + JSON_THROW(out_of_range::create(408, + "excessive array size: " + std::to_string(len))); + } + } + + return true; + } + + bool end_array() + { + bool keep = true; + + if (ref_stack.back()) + { + keep = callback(static_cast<int>(ref_stack.size()) - 1, parse_event_t::array_end, *ref_stack.back()); + if (not keep) + { + // discard array + *ref_stack.back() = discarded; + } + } + + assert(not ref_stack.empty()); + assert(not keep_stack.empty()); + ref_stack.pop_back(); + keep_stack.pop_back(); + + // remove discarded value + if (not keep and not ref_stack.empty()) + { + if (ref_stack.back()->is_array()) + { + ref_stack.back()->m_value.array->pop_back(); + } + } + + return true; + } + + bool parse_error(std::size_t, const std::string&, + const detail::exception& ex) + { + errored = true; + if (allow_exceptions) + { + // determine the proper exception type from the id + switch ((ex.id / 100) % 100) + { + case 1: + JSON_THROW(*reinterpret_cast<const detail::parse_error*>(&ex)); + case 4: + JSON_THROW(*reinterpret_cast<const detail::out_of_range*>(&ex)); + // LCOV_EXCL_START + case 2: + JSON_THROW(*reinterpret_cast<const detail::invalid_iterator*>(&ex)); + case 3: + JSON_THROW(*reinterpret_cast<const detail::type_error*>(&ex)); + case 5: + JSON_THROW(*reinterpret_cast<const detail::other_error*>(&ex)); + default: + assert(false); + // LCOV_EXCL_STOP + } + } + return false; + } + + constexpr bool is_errored() const + { + return errored; + } + + private: + /*! + @param[in] v value to add to the JSON value we build during parsing + @param[in] skip_callback whether we should skip calling the callback + function; this is required after start_array() and + start_object() SAX events, because otherwise we would call the + callback function with an empty array or object, respectively. + + @invariant If the ref stack is empty, then the passed value will be the new + root. + @invariant If the ref stack contains a value, then it is an array or an + object to which we can add elements + + @return pair of boolean (whether value should be kept) and pointer (to the + passed value in the ref_stack hierarchy; nullptr if not kept) + */ + template<typename Value> + std::pair<bool, BasicJsonType*> handle_value(Value&& v, const bool skip_callback = false) + { + assert(not keep_stack.empty()); + + // do not handle this value if we know it would be added to a discarded + // container + if (not keep_stack.back()) + { + return {false, nullptr}; + } + + // create value + auto value = BasicJsonType(std::forward<Value>(v)); + + // check callback + const bool keep = skip_callback or callback(static_cast<int>(ref_stack.size()), parse_event_t::value, value); + + // do not handle this value if we just learnt it shall be discarded + if (not keep) + { + return {false, nullptr}; + } + + if (ref_stack.empty()) + { + root = std::move(value); + return {true, &root}; + } + else + { + // skip this value if we already decided to skip the parent + // (https://github.com/nlohmann/json/issues/971#issuecomment-413678360) + if (not ref_stack.back()) + { + return {false, nullptr}; + } + + assert(ref_stack.back()->is_array() or ref_stack.back()->is_object()); + if (ref_stack.back()->is_array()) + { + ref_stack.back()->m_value.array->push_back(std::move(value)); + return {true, &(ref_stack.back()->m_value.array->back())}; + } + else + { + // check if we should store an element for the current key + assert(not key_keep_stack.empty()); + const bool store_element = key_keep_stack.back(); + key_keep_stack.pop_back(); + + if (not store_element) + { + return {false, nullptr}; + } + + assert(object_element); + *object_element = std::move(value); + return {true, object_element}; + } + } + } + + /// the parsed JSON value + BasicJsonType& root; + /// stack to model hierarchy of values + std::vector<BasicJsonType*> ref_stack; + /// stack to manage which values to keep + std::vector<bool> keep_stack; + /// stack to manage which object keys to keep + std::vector<bool> key_keep_stack; + /// helper to hold the reference for the next object element + BasicJsonType* object_element = nullptr; + /// whether a syntax error occurred + bool errored = false; + /// callback function + const parser_callback_t callback = nullptr; + /// whether to throw exceptions in case of errors + const bool allow_exceptions = true; + /// a discarded value for the callback + BasicJsonType discarded = BasicJsonType::value_t::discarded; +}; + +template<typename BasicJsonType> +class json_sax_acceptor +{ + public: + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + + bool null() + { + return true; + } + + bool boolean(bool) + { + return true; + } + + bool number_integer(number_integer_t) + { + return true; + } + + bool number_unsigned(number_unsigned_t) + { + return true; + } + + bool number_float(number_float_t, const string_t&) + { + return true; + } + + bool string(string_t&) + { + return true; + } + + bool start_object(std::size_t = std::size_t(-1)) + { + return true; + } + + bool key(string_t&) + { + return true; + } + + bool end_object() + { + return true; + } + + bool start_array(std::size_t = std::size_t(-1)) + { + return true; + } + + bool end_array() + { + return true; + } + + bool parse_error(std::size_t, const std::string&, const detail::exception&) + { + return false; + } +}; +} + +} diff --git a/include/lib/modernjson/detail/input/lexer.hpp b/include/lib/modernjson/detail/input/lexer.hpp new file mode 100644 index 0000000..9606211 --- /dev/null +++ b/include/lib/modernjson/detail/input/lexer.hpp @@ -0,0 +1,1333 @@ +#pragma once + +#include <clocale> // localeconv +#include <cstddef> // size_t +#include <cstdlib> // strtof, strtod, strtold, strtoll, strtoull +#include <cstdio> // snprintf +#include <initializer_list> // initializer_list +#include <string> // char_traits, string +#include <vector> // vector + +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/input/input_adapters.hpp> + +namespace nlohmann +{ +namespace detail +{ +/////////// +// lexer // +/////////// + +/*! +@brief lexical analysis + +This class organizes the lexical analysis during JSON deserialization. +*/ +template<typename BasicJsonType> +class lexer +{ + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + + public: + /// token types for the parser + enum class token_type + { + uninitialized, ///< indicating the scanner is uninitialized + literal_true, ///< the `true` literal + literal_false, ///< the `false` literal + literal_null, ///< the `null` literal + value_string, ///< a string -- use get_string() for actual value + value_unsigned, ///< an unsigned integer -- use get_number_unsigned() for actual value + value_integer, ///< a signed integer -- use get_number_integer() for actual value + value_float, ///< an floating point number -- use get_number_float() for actual value + begin_array, ///< the character for array begin `[` + begin_object, ///< the character for object begin `{` + end_array, ///< the character for array end `]` + end_object, ///< the character for object end `}` + name_separator, ///< the name separator `:` + value_separator, ///< the value separator `,` + parse_error, ///< indicating a parse error + end_of_input, ///< indicating the end of the input buffer + literal_or_value ///< a literal or the begin of a value (only for diagnostics) + }; + + /// return name of values of type token_type (only used for errors) + static const char* token_type_name(const token_type t) noexcept + { + switch (t) + { + case token_type::uninitialized: + return "<uninitialized>"; + case token_type::literal_true: + return "true literal"; + case token_type::literal_false: + return "false literal"; + case token_type::literal_null: + return "null literal"; + case token_type::value_string: + return "string literal"; + case lexer::token_type::value_unsigned: + case lexer::token_type::value_integer: + case lexer::token_type::value_float: + return "number literal"; + case token_type::begin_array: + return "'['"; + case token_type::begin_object: + return "'{'"; + case token_type::end_array: + return "']'"; + case token_type::end_object: + return "'}'"; + case token_type::name_separator: + return "':'"; + case token_type::value_separator: + return "','"; + case token_type::parse_error: + return "<parse error>"; + case token_type::end_of_input: + return "end of input"; + case token_type::literal_or_value: + return "'[', '{', or a literal"; + // LCOV_EXCL_START + default: // catch non-enum values + return "unknown token"; + // LCOV_EXCL_STOP + } + } + + explicit lexer(detail::input_adapter_t&& adapter) + : ia(std::move(adapter)), decimal_point_char(get_decimal_point()) {} + + // delete because of pointer members + lexer(const lexer&) = delete; + lexer& operator=(lexer&) = delete; + + private: + ///////////////////// + // locales + ///////////////////// + + /// return the locale-dependent decimal point + static char get_decimal_point() noexcept + { + const auto loc = localeconv(); + assert(loc != nullptr); + return (loc->decimal_point == nullptr) ? '.' : *(loc->decimal_point); + } + + ///////////////////// + // scan functions + ///////////////////// + + /*! + @brief get codepoint from 4 hex characters following `\u` + + For input "\u c1 c2 c3 c4" the codepoint is: + (c1 * 0x1000) + (c2 * 0x0100) + (c3 * 0x0010) + c4 + = (c1 << 12) + (c2 << 8) + (c3 << 4) + (c4 << 0) + + Furthermore, the possible characters '0'..'9', 'A'..'F', and 'a'..'f' + must be converted to the integers 0x0..0x9, 0xA..0xF, 0xA..0xF, resp. The + conversion is done by subtracting the offset (0x30, 0x37, and 0x57) + between the ASCII value of the character and the desired integer value. + + @return codepoint (0x0000..0xFFFF) or -1 in case of an error (e.g. EOF or + non-hex character) + */ + int get_codepoint() + { + // this function only makes sense after reading `\u` + assert(current == 'u'); + int codepoint = 0; + + const auto factors = { 12, 8, 4, 0 }; + for (const auto factor : factors) + { + get(); + + if (current >= '0' and current <= '9') + { + codepoint += ((current - 0x30) << factor); + } + else if (current >= 'A' and current <= 'F') + { + codepoint += ((current - 0x37) << factor); + } + else if (current >= 'a' and current <= 'f') + { + codepoint += ((current - 0x57) << factor); + } + else + { + return -1; + } + } + + assert(0x0000 <= codepoint and codepoint <= 0xFFFF); + return codepoint; + } + + /*! + @brief check if the next byte(s) are inside a given range + + Adds the current byte and, for each passed range, reads a new byte and + checks if it is inside the range. If a violation was detected, set up an + error message and return false. Otherwise, return true. + + @param[in] ranges list of integers; interpreted as list of pairs of + inclusive lower and upper bound, respectively + + @pre The passed list @a ranges must have 2, 4, or 6 elements; that is, + 1, 2, or 3 pairs. This precondition is enforced by an assertion. + + @return true if and only if no range violation was detected + */ + bool next_byte_in_range(std::initializer_list<int> ranges) + { + assert(ranges.size() == 2 or ranges.size() == 4 or ranges.size() == 6); + add(current); + + for (auto range = ranges.begin(); range != ranges.end(); ++range) + { + get(); + if (JSON_LIKELY(*range <= current and current <= *(++range))) + { + add(current); + } + else + { + error_message = "invalid string: ill-formed UTF-8 byte"; + return false; + } + } + + return true; + } + + /*! + @brief scan a string literal + + This function scans a string according to Sect. 7 of RFC 7159. While + scanning, bytes are escaped and copied into buffer token_buffer. Then the + function returns successfully, token_buffer is *not* null-terminated (as it + may contain \0 bytes), and token_buffer.size() is the number of bytes in the + string. + + @return token_type::value_string if string could be successfully scanned, + token_type::parse_error otherwise + + @note In case of errors, variable error_message contains a textual + description. + */ + token_type scan_string() + { + // reset token_buffer (ignore opening quote) + reset(); + + // we entered the function by reading an open quote + assert(current == '\"'); + + while (true) + { + // get next character + switch (get()) + { + // end of file while parsing string + case std::char_traits<char>::eof(): + { + error_message = "invalid string: missing closing quote"; + return token_type::parse_error; + } + + // closing quote + case '\"': + { + return token_type::value_string; + } + + // escapes + case '\\': + { + switch (get()) + { + // quotation mark + case '\"': + add('\"'); + break; + // reverse solidus + case '\\': + add('\\'); + break; + // solidus + case '/': + add('/'); + break; + // backspace + case 'b': + add('\b'); + break; + // form feed + case 'f': + add('\f'); + break; + // line feed + case 'n': + add('\n'); + break; + // carriage return + case 'r': + add('\r'); + break; + // tab + case 't': + add('\t'); + break; + + // unicode escapes + case 'u': + { + const int codepoint1 = get_codepoint(); + int codepoint = codepoint1; // start with codepoint1 + + if (JSON_UNLIKELY(codepoint1 == -1)) + { + error_message = "invalid string: '\\u' must be followed by 4 hex digits"; + return token_type::parse_error; + } + + // check if code point is a high surrogate + if (0xD800 <= codepoint1 and codepoint1 <= 0xDBFF) + { + // expect next \uxxxx entry + if (JSON_LIKELY(get() == '\\' and get() == 'u')) + { + const int codepoint2 = get_codepoint(); + + if (JSON_UNLIKELY(codepoint2 == -1)) + { + error_message = "invalid string: '\\u' must be followed by 4 hex digits"; + return token_type::parse_error; + } + + // check if codepoint2 is a low surrogate + if (JSON_LIKELY(0xDC00 <= codepoint2 and codepoint2 <= 0xDFFF)) + { + // overwrite codepoint + codepoint = + // high surrogate occupies the most significant 22 bits + (codepoint1 << 10) + // low surrogate occupies the least significant 15 bits + + codepoint2 + // there is still the 0xD800, 0xDC00 and 0x10000 noise + // in the result so we have to subtract with: + // (0xD800 << 10) + DC00 - 0x10000 = 0x35FDC00 + - 0x35FDC00; + } + else + { + error_message = "invalid string: surrogate U+DC00..U+DFFF must be followed by U+DC00..U+DFFF"; + return token_type::parse_error; + } + } + else + { + error_message = "invalid string: surrogate U+DC00..U+DFFF must be followed by U+DC00..U+DFFF"; + return token_type::parse_error; + } + } + else + { + if (JSON_UNLIKELY(0xDC00 <= codepoint1 and codepoint1 <= 0xDFFF)) + { + error_message = "invalid string: surrogate U+DC00..U+DFFF must follow U+D800..U+DBFF"; + return token_type::parse_error; + } + } + + // result of the above calculation yields a proper codepoint + assert(0x00 <= codepoint and codepoint <= 0x10FFFF); + + // translate codepoint into bytes + if (codepoint < 0x80) + { + // 1-byte characters: 0xxxxxxx (ASCII) + add(codepoint); + } + else if (codepoint <= 0x7FF) + { + // 2-byte characters: 110xxxxx 10xxxxxx + add(0xC0 | (codepoint >> 6)); + add(0x80 | (codepoint & 0x3F)); + } + else if (codepoint <= 0xFFFF) + { + // 3-byte characters: 1110xxxx 10xxxxxx 10xxxxxx + add(0xE0 | (codepoint >> 12)); + add(0x80 | ((codepoint >> 6) & 0x3F)); + add(0x80 | (codepoint & 0x3F)); + } + else + { + // 4-byte characters: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx + add(0xF0 | (codepoint >> 18)); + add(0x80 | ((codepoint >> 12) & 0x3F)); + add(0x80 | ((codepoint >> 6) & 0x3F)); + add(0x80 | (codepoint & 0x3F)); + } + + break; + } + + // other characters after escape + default: + error_message = "invalid string: forbidden character after backslash"; + return token_type::parse_error; + } + + break; + } + + // invalid control characters + case 0x00: + case 0x01: + case 0x02: + case 0x03: + case 0x04: + case 0x05: + case 0x06: + case 0x07: + case 0x08: + case 0x09: + case 0x0A: + case 0x0B: + case 0x0C: + case 0x0D: + case 0x0E: + case 0x0F: + case 0x10: + case 0x11: + case 0x12: + case 0x13: + case 0x14: + case 0x15: + case 0x16: + case 0x17: + case 0x18: + case 0x19: + case 0x1A: + case 0x1B: + case 0x1C: + case 0x1D: + case 0x1E: + case 0x1F: + { + error_message = "invalid string: control character must be escaped"; + return token_type::parse_error; + } + + // U+0020..U+007F (except U+0022 (quote) and U+005C (backspace)) + case 0x20: + case 0x21: + case 0x23: + case 0x24: + case 0x25: + case 0x26: + case 0x27: + case 0x28: + case 0x29: + case 0x2A: + case 0x2B: + case 0x2C: + case 0x2D: + case 0x2E: + case 0x2F: + case 0x30: + case 0x31: + case 0x32: + case 0x33: + case 0x34: + case 0x35: + case 0x36: + case 0x37: + case 0x38: + case 0x39: + case 0x3A: + case 0x3B: + case 0x3C: + case 0x3D: + case 0x3E: + case 0x3F: + case 0x40: + case 0x41: + case 0x42: + case 0x43: + case 0x44: + case 0x45: + case 0x46: + case 0x47: + case 0x48: + case 0x49: + case 0x4A: + case 0x4B: + case 0x4C: + case 0x4D: + case 0x4E: + case 0x4F: + case 0x50: + case 0x51: + case 0x52: + case 0x53: + case 0x54: + case 0x55: + case 0x56: + case 0x57: + case 0x58: + case 0x59: + case 0x5A: + case 0x5B: + case 0x5D: + case 0x5E: + case 0x5F: + case 0x60: + case 0x61: + case 0x62: + case 0x63: + case 0x64: + case 0x65: + case 0x66: + case 0x67: + case 0x68: + case 0x69: + case 0x6A: + case 0x6B: + case 0x6C: + case 0x6D: + case 0x6E: + case 0x6F: + case 0x70: + case 0x71: + case 0x72: + case 0x73: + case 0x74: + case 0x75: + case 0x76: + case 0x77: + case 0x78: + case 0x79: + case 0x7A: + case 0x7B: + case 0x7C: + case 0x7D: + case 0x7E: + case 0x7F: + { + add(current); + break; + } + + // U+0080..U+07FF: bytes C2..DF 80..BF + case 0xC2: + case 0xC3: + case 0xC4: + case 0xC5: + case 0xC6: + case 0xC7: + case 0xC8: + case 0xC9: + case 0xCA: + case 0xCB: + case 0xCC: + case 0xCD: + case 0xCE: + case 0xCF: + case 0xD0: + case 0xD1: + case 0xD2: + case 0xD3: + case 0xD4: + case 0xD5: + case 0xD6: + case 0xD7: + case 0xD8: + case 0xD9: + case 0xDA: + case 0xDB: + case 0xDC: + case 0xDD: + case 0xDE: + case 0xDF: + { + if (JSON_UNLIKELY(not next_byte_in_range({0x80, 0xBF}))) + { + return token_type::parse_error; + } + break; + } + + // U+0800..U+0FFF: bytes E0 A0..BF 80..BF + case 0xE0: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0xA0, 0xBF, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // U+1000..U+CFFF: bytes E1..EC 80..BF 80..BF + // U+E000..U+FFFF: bytes EE..EF 80..BF 80..BF + case 0xE1: + case 0xE2: + case 0xE3: + case 0xE4: + case 0xE5: + case 0xE6: + case 0xE7: + case 0xE8: + case 0xE9: + case 0xEA: + case 0xEB: + case 0xEC: + case 0xEE: + case 0xEF: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0x80, 0xBF, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // U+D000..U+D7FF: bytes ED 80..9F 80..BF + case 0xED: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0x80, 0x9F, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // U+10000..U+3FFFF F0 90..BF 80..BF 80..BF + case 0xF0: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0x90, 0xBF, 0x80, 0xBF, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF + case 0xF1: + case 0xF2: + case 0xF3: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0x80, 0xBF, 0x80, 0xBF, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // U+100000..U+10FFFF F4 80..8F 80..BF 80..BF + case 0xF4: + { + if (JSON_UNLIKELY(not (next_byte_in_range({0x80, 0x8F, 0x80, 0xBF, 0x80, 0xBF})))) + { + return token_type::parse_error; + } + break; + } + + // remaining bytes (80..C1 and F5..FF) are ill-formed + default: + { + error_message = "invalid string: ill-formed UTF-8 byte"; + return token_type::parse_error; + } + } + } + } + + static void strtof(float& f, const char* str, char** endptr) noexcept + { + f = std::strtof(str, endptr); + } + + static void strtof(double& f, const char* str, char** endptr) noexcept + { + f = std::strtod(str, endptr); + } + + static void strtof(long double& f, const char* str, char** endptr) noexcept + { + f = std::strtold(str, endptr); + } + + /*! + @brief scan a number literal + + This function scans a string according to Sect. 6 of RFC 7159. + + The function is realized with a deterministic finite state machine derived + from the grammar described in RFC 7159. Starting in state "init", the + input is read and used to determined the next state. Only state "done" + accepts the number. State "error" is a trap state to model errors. In the + table below, "anything" means any character but the ones listed before. + + state | 0 | 1-9 | e E | + | - | . | anything + ---------|----------|----------|----------|---------|---------|----------|----------- + init | zero | any1 | [error] | [error] | minus | [error] | [error] + minus | zero | any1 | [error] | [error] | [error] | [error] | [error] + zero | done | done | exponent | done | done | decimal1 | done + any1 | any1 | any1 | exponent | done | done | decimal1 | done + decimal1 | decimal2 | [error] | [error] | [error] | [error] | [error] | [error] + decimal2 | decimal2 | decimal2 | exponent | done | done | done | done + exponent | any2 | any2 | [error] | sign | sign | [error] | [error] + sign | any2 | any2 | [error] | [error] | [error] | [error] | [error] + any2 | any2 | any2 | done | done | done | done | done + + The state machine is realized with one label per state (prefixed with + "scan_number_") and `goto` statements between them. The state machine + contains cycles, but any cycle can be left when EOF is read. Therefore, + the function is guaranteed to terminate. + + During scanning, the read bytes are stored in token_buffer. This string is + then converted to a signed integer, an unsigned integer, or a + floating-point number. + + @return token_type::value_unsigned, token_type::value_integer, or + token_type::value_float if number could be successfully scanned, + token_type::parse_error otherwise + + @note The scanner is independent of the current locale. Internally, the + locale's decimal point is used instead of `.` to work with the + locale-dependent converters. + */ + token_type scan_number() + { + // reset token_buffer to store the number's bytes + reset(); + + // the type of the parsed number; initially set to unsigned; will be + // changed if minus sign, decimal point or exponent is read + token_type number_type = token_type::value_unsigned; + + // state (init): we just found out we need to scan a number + switch (current) + { + case '-': + { + add(current); + goto scan_number_minus; + } + + case '0': + { + add(current); + goto scan_number_zero; + } + + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any1; + } + + // LCOV_EXCL_START + default: + { + // all other characters are rejected outside scan_number() + assert(false); + } + // LCOV_EXCL_STOP + } + +scan_number_minus: + // state: we just parsed a leading minus sign + number_type = token_type::value_integer; + switch (get()) + { + case '0': + { + add(current); + goto scan_number_zero; + } + + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any1; + } + + default: + { + error_message = "invalid number; expected digit after '-'"; + return token_type::parse_error; + } + } + +scan_number_zero: + // state: we just parse a zero (maybe with a leading minus sign) + switch (get()) + { + case '.': + { + add(decimal_point_char); + goto scan_number_decimal1; + } + + case 'e': + case 'E': + { + add(current); + goto scan_number_exponent; + } + + default: + goto scan_number_done; + } + +scan_number_any1: + // state: we just parsed a number 0-9 (maybe with a leading minus sign) + switch (get()) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any1; + } + + case '.': + { + add(decimal_point_char); + goto scan_number_decimal1; + } + + case 'e': + case 'E': + { + add(current); + goto scan_number_exponent; + } + + default: + goto scan_number_done; + } + +scan_number_decimal1: + // state: we just parsed a decimal point + number_type = token_type::value_float; + switch (get()) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_decimal2; + } + + default: + { + error_message = "invalid number; expected digit after '.'"; + return token_type::parse_error; + } + } + +scan_number_decimal2: + // we just parsed at least one number after a decimal point + switch (get()) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_decimal2; + } + + case 'e': + case 'E': + { + add(current); + goto scan_number_exponent; + } + + default: + goto scan_number_done; + } + +scan_number_exponent: + // we just parsed an exponent + number_type = token_type::value_float; + switch (get()) + { + case '+': + case '-': + { + add(current); + goto scan_number_sign; + } + + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any2; + } + + default: + { + error_message = + "invalid number; expected '+', '-', or digit after exponent"; + return token_type::parse_error; + } + } + +scan_number_sign: + // we just parsed an exponent sign + switch (get()) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any2; + } + + default: + { + error_message = "invalid number; expected digit after exponent sign"; + return token_type::parse_error; + } + } + +scan_number_any2: + // we just parsed a number after the exponent or exponent sign + switch (get()) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + { + add(current); + goto scan_number_any2; + } + + default: + goto scan_number_done; + } + +scan_number_done: + // unget the character after the number (we only read it to know that + // we are done scanning a number) + unget(); + + char* endptr = nullptr; + errno = 0; + + // try to parse integers first and fall back to floats + if (number_type == token_type::value_unsigned) + { + const auto x = std::strtoull(token_buffer.data(), &endptr, 10); + + // we checked the number format before + assert(endptr == token_buffer.data() + token_buffer.size()); + + if (errno == 0) + { + value_unsigned = static_cast<number_unsigned_t>(x); + if (value_unsigned == x) + { + return token_type::value_unsigned; + } + } + } + else if (number_type == token_type::value_integer) + { + const auto x = std::strtoll(token_buffer.data(), &endptr, 10); + + // we checked the number format before + assert(endptr == token_buffer.data() + token_buffer.size()); + + if (errno == 0) + { + value_integer = static_cast<number_integer_t>(x); + if (value_integer == x) + { + return token_type::value_integer; + } + } + } + + // this code is reached if we parse a floating-point number or if an + // integer conversion above failed + strtof(value_float, token_buffer.data(), &endptr); + + // we checked the number format before + assert(endptr == token_buffer.data() + token_buffer.size()); + + return token_type::value_float; + } + + /*! + @param[in] literal_text the literal text to expect + @param[in] length the length of the passed literal text + @param[in] return_type the token type to return on success + */ + token_type scan_literal(const char* literal_text, const std::size_t length, + token_type return_type) + { + assert(current == literal_text[0]); + for (std::size_t i = 1; i < length; ++i) + { + if (JSON_UNLIKELY(get() != literal_text[i])) + { + error_message = "invalid literal"; + return token_type::parse_error; + } + } + return return_type; + } + + ///////////////////// + // input management + ///////////////////// + + /// reset token_buffer; current character is beginning of token + void reset() noexcept + { + token_buffer.clear(); + token_string.clear(); + token_string.push_back(std::char_traits<char>::to_char_type(current)); + } + + /* + @brief get next character from the input + + This function provides the interface to the used input adapter. It does + not throw in case the input reached EOF, but returns a + `std::char_traits<char>::eof()` in that case. Stores the scanned characters + for use in error messages. + + @return character read from the input + */ + std::char_traits<char>::int_type get() + { + ++chars_read; + if (next_unget) + { + // just reset the next_unget variable and work with current + next_unget = false; + } + else + { + current = ia->get_character(); + } + + if (JSON_LIKELY(current != std::char_traits<char>::eof())) + { + token_string.push_back(std::char_traits<char>::to_char_type(current)); + } + return current; + } + + /*! + @brief unget current character (read it again on next get) + + We implement unget by setting variable next_unget to true. The input is not + changed - we just simulate ungetting by modifying chars_read and + token_string. The next call to get() will behave as if the unget character + is read again. + */ + void unget() + { + next_unget = true; + --chars_read; + if (JSON_LIKELY(current != std::char_traits<char>::eof())) + { + assert(token_string.size() != 0); + token_string.pop_back(); + } + } + + /// add a character to token_buffer + void add(int c) + { + token_buffer.push_back(std::char_traits<char>::to_char_type(c)); + } + + public: + ///////////////////// + // value getters + ///////////////////// + + /// return integer value + constexpr number_integer_t get_number_integer() const noexcept + { + return value_integer; + } + + /// return unsigned integer value + constexpr number_unsigned_t get_number_unsigned() const noexcept + { + return value_unsigned; + } + + /// return floating-point value + constexpr number_float_t get_number_float() const noexcept + { + return value_float; + } + + /// return current string value (implicitly resets the token; useful only once) + string_t& get_string() + { + return token_buffer; + } + + ///////////////////// + // diagnostics + ///////////////////// + + /// return position of last read token + constexpr std::size_t get_position() const noexcept + { + return chars_read; + } + + /// return the last read token (for errors only). Will never contain EOF + /// (an arbitrary value that is not a valid char value, often -1), because + /// 255 may legitimately occur. May contain NUL, which should be escaped. + std::string get_token_string() const + { + // escape control characters + std::string result; + for (const auto c : token_string) + { + if ('\x00' <= c and c <= '\x1F') + { + // escape control characters + char cs[9]; + snprintf(cs, 9, "<U+%.4X>", static_cast<unsigned char>(c)); + result += cs; + } + else + { + // add character as is + result.push_back(c); + } + } + + return result; + } + + /// return syntax error message + constexpr const char* get_error_message() const noexcept + { + return error_message; + } + + ///////////////////// + // actual scanner + ///////////////////// + + /*! + @brief skip the UTF-8 byte order mark + @return true iff there is no BOM or the correct BOM has been skipped + */ + bool skip_bom() + { + if (get() == 0xEF) + { + if (get() == 0xBB and get() == 0xBF) + { + // we completely parsed the BOM + return true; + } + else + { + // after reading 0xEF, an unexpected character followed + return false; + } + } + else + { + // the first character is not the beginning of the BOM; unget it to + // process is later + unget(); + return true; + } + } + + token_type scan() + { + // initially, skip the BOM + if (chars_read == 0 and not skip_bom()) + { + error_message = "invalid BOM; must be 0xEF 0xBB 0xBF if given"; + return token_type::parse_error; + } + + // read next character and ignore whitespace + do + { + get(); + } + while (current == ' ' or current == '\t' or current == '\n' or current == '\r'); + + switch (current) + { + // structural characters + case '[': + return token_type::begin_array; + case ']': + return token_type::end_array; + case '{': + return token_type::begin_object; + case '}': + return token_type::end_object; + case ':': + return token_type::name_separator; + case ',': + return token_type::value_separator; + + // literals + case 't': + return scan_literal("true", 4, token_type::literal_true); + case 'f': + return scan_literal("false", 5, token_type::literal_false); + case 'n': + return scan_literal("null", 4, token_type::literal_null); + + // string + case '\"': + return scan_string(); + + // number + case '-': + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + return scan_number(); + + // end of input (the null byte is needed when parsing from + // string literals) + case '\0': + case std::char_traits<char>::eof(): + return token_type::end_of_input; + + // error + default: + error_message = "invalid literal"; + return token_type::parse_error; + } + } + + private: + /// input adapter + detail::input_adapter_t ia = nullptr; + + /// the current character + std::char_traits<char>::int_type current = std::char_traits<char>::eof(); + + /// whether the next get() call should just return current + bool next_unget = false; + + /// the number of characters read + std::size_t chars_read = 0; + + /// raw input token string (for error messages) + std::vector<char> token_string {}; + + /// buffer for variable-length tokens (numbers, strings) + string_t token_buffer {}; + + /// a description of occurred lexer errors + const char* error_message = ""; + + // number values + number_integer_t value_integer = 0; + number_unsigned_t value_unsigned = 0; + number_float_t value_float = 0; + + /// the decimal point + const char decimal_point_char = '.'; +}; +} +} diff --git a/include/lib/modernjson/detail/input/parser.hpp b/include/lib/modernjson/detail/input/parser.hpp new file mode 100644 index 0000000..a622077 --- /dev/null +++ b/include/lib/modernjson/detail/input/parser.hpp @@ -0,0 +1,488 @@ +#pragma once + +#include <cassert> // assert +#include <cmath> // isfinite +#include <cstdint> // uint8_t +#include <functional> // function +#include <string> // string +#include <utility> // move + +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/meta/is_sax.hpp> +#include <lib/modernjson/detail/input/input_adapters.hpp> +#include <lib/modernjson/detail/input/json_sax.hpp> +#include <lib/modernjson/detail/input/lexer.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +//////////// +// parser // +//////////// + +/*! +@brief syntax analysis + +This class implements a recursive decent parser. +*/ +template<typename BasicJsonType> +class parser +{ + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + using lexer_t = lexer<BasicJsonType>; + using token_type = typename lexer_t::token_type; + + public: + enum class parse_event_t : uint8_t + { + /// the parser read `{` and started to process a JSON object + object_start, + /// the parser read `}` and finished processing a JSON object + object_end, + /// the parser read `[` and started to process a JSON array + array_start, + /// the parser read `]` and finished processing a JSON array + array_end, + /// the parser read a key of a value in an object + key, + /// the parser finished reading a JSON value + value + }; + + using parser_callback_t = + std::function<bool(int depth, parse_event_t event, BasicJsonType& parsed)>; + + /// a parser reading from an input adapter + explicit parser(detail::input_adapter_t&& adapter, + const parser_callback_t cb = nullptr, + const bool allow_exceptions_ = true) + : callback(cb), m_lexer(std::move(adapter)), allow_exceptions(allow_exceptions_) + { + // read first token + get_token(); + } + + /*! + @brief public parser interface + + @param[in] strict whether to expect the last token to be EOF + @param[in,out] result parsed JSON value + + @throw parse_error.101 in case of an unexpected token + @throw parse_error.102 if to_unicode fails or surrogate error + @throw parse_error.103 if to_unicode fails + */ + void parse(const bool strict, BasicJsonType& result) + { + if (callback) + { + json_sax_dom_callback_parser<BasicJsonType> sdp(result, callback, allow_exceptions); + sax_parse_internal(&sdp); + result.assert_invariant(); + + // in strict mode, input must be completely read + if (strict and (get_token() != token_type::end_of_input)) + { + sdp.parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::end_of_input))); + } + + // in case of an error, return discarded value + if (sdp.is_errored()) + { + result = value_t::discarded; + return; + } + + // set top-level value to null if it was discarded by the callback + // function + if (result.is_discarded()) + { + result = nullptr; + } + } + else + { + json_sax_dom_parser<BasicJsonType> sdp(result, allow_exceptions); + sax_parse_internal(&sdp); + result.assert_invariant(); + + // in strict mode, input must be completely read + if (strict and (get_token() != token_type::end_of_input)) + { + sdp.parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::end_of_input))); + } + + // in case of an error, return discarded value + if (sdp.is_errored()) + { + result = value_t::discarded; + return; + } + } + } + + /*! + @brief public accept interface + + @param[in] strict whether to expect the last token to be EOF + @return whether the input is a proper JSON text + */ + bool accept(const bool strict = true) + { + json_sax_acceptor<BasicJsonType> sax_acceptor; + return sax_parse(&sax_acceptor, strict); + } + + template <typename SAX> + bool sax_parse(SAX* sax, const bool strict = true) + { + (void)detail::is_sax_static_asserts<SAX, BasicJsonType> {}; + const bool result = sax_parse_internal(sax); + + // strict mode: next byte must be EOF + if (result and strict and (get_token() != token_type::end_of_input)) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::end_of_input))); + } + + return result; + } + + private: + template <typename SAX> + bool sax_parse_internal(SAX* sax) + { + // stack to remember the hieararchy of structured values we are parsing + // true = array; false = object + std::vector<bool> states; + // value to avoid a goto (see comment where set to true) + bool skip_to_state_evaluation = false; + + while (true) + { + if (not skip_to_state_evaluation) + { + // invariant: get_token() was called before each iteration + switch (last_token) + { + case token_type::begin_object: + { + if (JSON_UNLIKELY(not sax->start_object(std::size_t(-1)))) + { + return false; + } + + // closing } -> we are done + if (get_token() == token_type::end_object) + { + if (JSON_UNLIKELY(not sax->end_object())) + { + return false; + } + break; + } + + // parse key + if (JSON_UNLIKELY(last_token != token_type::value_string)) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::value_string))); + } + else + { + if (JSON_UNLIKELY(not sax->key(m_lexer.get_string()))) + { + return false; + } + } + + // parse separator (:) + if (JSON_UNLIKELY(get_token() != token_type::name_separator)) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::name_separator))); + } + + // remember we are now inside an object + states.push_back(false); + + // parse values + get_token(); + continue; + } + + case token_type::begin_array: + { + if (JSON_UNLIKELY(not sax->start_array(std::size_t(-1)))) + { + return false; + } + + // closing ] -> we are done + if (get_token() == token_type::end_array) + { + if (JSON_UNLIKELY(not sax->end_array())) + { + return false; + } + break; + } + + // remember we are now inside an array + states.push_back(true); + + // parse values (no need to call get_token) + continue; + } + + case token_type::value_float: + { + const auto res = m_lexer.get_number_float(); + + if (JSON_UNLIKELY(not std::isfinite(res))) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + out_of_range::create(406, "number overflow parsing '" + m_lexer.get_token_string() + "'")); + } + else + { + if (JSON_UNLIKELY(not sax->number_float(res, m_lexer.get_string()))) + { + return false; + } + break; + } + } + + case token_type::literal_false: + { + if (JSON_UNLIKELY(not sax->boolean(false))) + { + return false; + } + break; + } + + case token_type::literal_null: + { + if (JSON_UNLIKELY(not sax->null())) + { + return false; + } + break; + } + + case token_type::literal_true: + { + if (JSON_UNLIKELY(not sax->boolean(true))) + { + return false; + } + break; + } + + case token_type::value_integer: + { + if (JSON_UNLIKELY(not sax->number_integer(m_lexer.get_number_integer()))) + { + return false; + } + break; + } + + case token_type::value_string: + { + if (JSON_UNLIKELY(not sax->string(m_lexer.get_string()))) + { + return false; + } + break; + } + + case token_type::value_unsigned: + { + if (JSON_UNLIKELY(not sax->number_unsigned(m_lexer.get_number_unsigned()))) + { + return false; + } + break; + } + + case token_type::parse_error: + { + // using "uninitialized" to avoid "expected" message + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::uninitialized))); + } + + default: // the last token was unexpected + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::literal_or_value))); + } + } + } + else + { + skip_to_state_evaluation = false; + } + + // we reached this line after we successfully parsed a value + if (states.empty()) + { + // empty stack: we reached the end of the hieararchy: done + return true; + } + else + { + if (states.back()) // array + { + // comma -> next value + if (get_token() == token_type::value_separator) + { + // parse a new value + get_token(); + continue; + } + + // closing ] + if (JSON_LIKELY(last_token == token_type::end_array)) + { + if (JSON_UNLIKELY(not sax->end_array())) + { + return false; + } + + // We are done with this array. Before we can parse a + // new value, we need to evaluate the new state first. + // By setting skip_to_state_evaluation to false, we + // are effectively jumping to the beginning of this if. + assert(not states.empty()); + states.pop_back(); + skip_to_state_evaluation = true; + continue; + } + else + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::end_array))); + } + } + else // object + { + // comma -> next value + if (get_token() == token_type::value_separator) + { + // parse key + if (JSON_UNLIKELY(get_token() != token_type::value_string)) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::value_string))); + } + else + { + if (JSON_UNLIKELY(not sax->key(m_lexer.get_string()))) + { + return false; + } + } + + // parse separator (:) + if (JSON_UNLIKELY(get_token() != token_type::name_separator)) + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::name_separator))); + } + + // parse values + get_token(); + continue; + } + + // closing } + if (JSON_LIKELY(last_token == token_type::end_object)) + { + if (JSON_UNLIKELY(not sax->end_object())) + { + return false; + } + + // We are done with this object. Before we can parse a + // new value, we need to evaluate the new state first. + // By setting skip_to_state_evaluation to false, we + // are effectively jumping to the beginning of this if. + assert(not states.empty()); + states.pop_back(); + skip_to_state_evaluation = true; + continue; + } + else + { + return sax->parse_error(m_lexer.get_position(), + m_lexer.get_token_string(), + parse_error::create(101, m_lexer.get_position(), exception_message(token_type::end_object))); + } + } + } + } + } + + /// get next token from lexer + token_type get_token() + { + return (last_token = m_lexer.scan()); + } + + std::string exception_message(const token_type expected) + { + std::string error_msg = "syntax error - "; + if (last_token == token_type::parse_error) + { + error_msg += std::string(m_lexer.get_error_message()) + "; last read: '" + + m_lexer.get_token_string() + "'"; + } + else + { + error_msg += "unexpected " + std::string(lexer_t::token_type_name(last_token)); + } + + if (expected != token_type::uninitialized) + { + error_msg += "; expected " + std::string(lexer_t::token_type_name(expected)); + } + + return error_msg; + } + + private: + /// callback function + const parser_callback_t callback = nullptr; + /// the type of the last read token + token_type last_token = token_type::uninitialized; + /// the lexer + lexer_t m_lexer; + /// whether to throw exceptions in case of errors + const bool allow_exceptions = true; +}; +} +} diff --git a/include/lib/modernjson/detail/iterators/internal_iterator.hpp b/include/lib/modernjson/detail/iterators/internal_iterator.hpp new file mode 100644 index 0000000..95f056d --- /dev/null +++ b/include/lib/modernjson/detail/iterators/internal_iterator.hpp @@ -0,0 +1,25 @@ +#pragma once + +#include <lib/modernjson/detail/iterators/primitive_iterator.hpp> + +namespace nlohmann +{ +namespace detail +{ +/*! +@brief an iterator value + +@note This structure could easily be a union, but MSVC currently does not allow +unions members with complex constructors, see https://github.com/nlohmann/json/pull/105. +*/ +template<typename BasicJsonType> struct internal_iterator +{ + /// iterator for JSON objects + typename BasicJsonType::object_t::iterator object_iterator {}; + /// iterator for JSON arrays + typename BasicJsonType::array_t::iterator array_iterator {}; + /// generic iterator for all other types + primitive_iterator_t primitive_iterator {}; +}; +} +} diff --git a/include/lib/modernjson/detail/iterators/iter_impl.hpp b/include/lib/modernjson/detail/iterators/iter_impl.hpp new file mode 100644 index 0000000..9e4795a --- /dev/null +++ b/include/lib/modernjson/detail/iterators/iter_impl.hpp @@ -0,0 +1,614 @@ +#pragma once + +#include <ciso646> // not +#include <iterator> // iterator, random_access_iterator_tag, bidirectional_iterator_tag, advance, next +#include <type_traits> // conditional, is_const, remove_const + +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/iterators/internal_iterator.hpp> +#include <lib/modernjson/detail/iterators/primitive_iterator.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/meta/cpp_future.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +// forward declare, to be able to friend it later on +template<typename IteratorType> class iteration_proxy; + +/*! +@brief a template for a bidirectional iterator for the @ref basic_json class + +This class implements a both iterators (iterator and const_iterator) for the +@ref basic_json class. + +@note An iterator is called *initialized* when a pointer to a JSON value has + been set (e.g., by a constructor or a copy assignment). If the iterator is + default-constructed, it is *uninitialized* and most methods are undefined. + **The library uses assertions to detect calls on uninitialized iterators.** + +@requirement The class satisfies the following concept requirements: +- +[BidirectionalIterator](https://en.cppreference.com/w/cpp/named_req/BidirectionalIterator): + The iterator that can be moved can be moved in both directions (i.e. + incremented and decremented). + +@since version 1.0.0, simplified in version 2.0.9, change to bidirectional + iterators in version 3.0.0 (see https://github.com/nlohmann/json/issues/593) +*/ +template<typename BasicJsonType> +class iter_impl +{ + /// allow basic_json to access private members + friend iter_impl<typename std::conditional<std::is_const<BasicJsonType>::value, typename std::remove_const<BasicJsonType>::type, const BasicJsonType>::type>; + friend BasicJsonType; + friend iteration_proxy<iter_impl>; + + using object_t = typename BasicJsonType::object_t; + using array_t = typename BasicJsonType::array_t; + // make sure BasicJsonType is basic_json or const basic_json + static_assert(is_basic_json<typename std::remove_const<BasicJsonType>::type>::value, + "iter_impl only accepts (const) basic_json"); + + public: + + /// The std::iterator class template (used as a base class to provide typedefs) is deprecated in C++17. + /// The C++ Standard has never required user-defined iterators to derive from std::iterator. + /// A user-defined iterator should provide publicly accessible typedefs named + /// iterator_category, value_type, difference_type, pointer, and reference. + /// Note that value_type is required to be non-const, even for constant iterators. + using iterator_category = std::bidirectional_iterator_tag; + + /// the type of the values when the iterator is dereferenced + using value_type = typename BasicJsonType::value_type; + /// a type to represent differences between iterators + using difference_type = typename BasicJsonType::difference_type; + /// defines a pointer to the type iterated over (value_type) + using pointer = typename std::conditional<std::is_const<BasicJsonType>::value, + typename BasicJsonType::const_pointer, + typename BasicJsonType::pointer>::type; + /// defines a reference to the type iterated over (value_type) + using reference = + typename std::conditional<std::is_const<BasicJsonType>::value, + typename BasicJsonType::const_reference, + typename BasicJsonType::reference>::type; + + /// default constructor + iter_impl() = default; + + /*! + @brief constructor for a given JSON instance + @param[in] object pointer to a JSON object for this iterator + @pre object != nullptr + @post The iterator is initialized; i.e. `m_object != nullptr`. + */ + explicit iter_impl(pointer object) noexcept : m_object(object) + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + m_it.object_iterator = typename object_t::iterator(); + break; + } + + case value_t::array: + { + m_it.array_iterator = typename array_t::iterator(); + break; + } + + default: + { + m_it.primitive_iterator = primitive_iterator_t(); + break; + } + } + } + + /*! + @note The conventional copy constructor and copy assignment are implicitly + defined. Combined with the following converting constructor and + assignment, they support: (1) copy from iterator to iterator, (2) + copy from const iterator to const iterator, and (3) conversion from + iterator to const iterator. However conversion from const iterator + to iterator is not defined. + */ + + /*! + @brief converting constructor + @param[in] other non-const iterator to copy from + @note It is not checked whether @a other is initialized. + */ + iter_impl(const iter_impl<typename std::remove_const<BasicJsonType>::type>& other) noexcept + : m_object(other.m_object), m_it(other.m_it) {} + + /*! + @brief converting assignment + @param[in,out] other non-const iterator to copy from + @return const/non-const iterator + @note It is not checked whether @a other is initialized. + */ + iter_impl& operator=(const iter_impl<typename std::remove_const<BasicJsonType>::type>& other) noexcept + { + m_object = other.m_object; + m_it = other.m_it; + return *this; + } + + private: + /*! + @brief set the iterator to the first value + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + void set_begin() noexcept + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + m_it.object_iterator = m_object->m_value.object->begin(); + break; + } + + case value_t::array: + { + m_it.array_iterator = m_object->m_value.array->begin(); + break; + } + + case value_t::null: + { + // set to end so begin()==end() is true: null is empty + m_it.primitive_iterator.set_end(); + break; + } + + default: + { + m_it.primitive_iterator.set_begin(); + break; + } + } + } + + /*! + @brief set the iterator past the last value + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + void set_end() noexcept + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + m_it.object_iterator = m_object->m_value.object->end(); + break; + } + + case value_t::array: + { + m_it.array_iterator = m_object->m_value.array->end(); + break; + } + + default: + { + m_it.primitive_iterator.set_end(); + break; + } + } + } + + public: + /*! + @brief return a reference to the value pointed to by the iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + reference operator*() const + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + assert(m_it.object_iterator != m_object->m_value.object->end()); + return m_it.object_iterator->second; + } + + case value_t::array: + { + assert(m_it.array_iterator != m_object->m_value.array->end()); + return *m_it.array_iterator; + } + + case value_t::null: + JSON_THROW(invalid_iterator::create(214, "cannot get value")); + + default: + { + if (JSON_LIKELY(m_it.primitive_iterator.is_begin())) + { + return *m_object; + } + + JSON_THROW(invalid_iterator::create(214, "cannot get value")); + } + } + } + + /*! + @brief dereference the iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + pointer operator->() const + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + assert(m_it.object_iterator != m_object->m_value.object->end()); + return &(m_it.object_iterator->second); + } + + case value_t::array: + { + assert(m_it.array_iterator != m_object->m_value.array->end()); + return &*m_it.array_iterator; + } + + default: + { + if (JSON_LIKELY(m_it.primitive_iterator.is_begin())) + { + return m_object; + } + + JSON_THROW(invalid_iterator::create(214, "cannot get value")); + } + } + } + + /*! + @brief post-increment (it++) + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl const operator++(int) + { + auto result = *this; + ++(*this); + return result; + } + + /*! + @brief pre-increment (++it) + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl& operator++() + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + std::advance(m_it.object_iterator, 1); + break; + } + + case value_t::array: + { + std::advance(m_it.array_iterator, 1); + break; + } + + default: + { + ++m_it.primitive_iterator; + break; + } + } + + return *this; + } + + /*! + @brief post-decrement (it--) + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl const operator--(int) + { + auto result = *this; + --(*this); + return result; + } + + /*! + @brief pre-decrement (--it) + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl& operator--() + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + { + std::advance(m_it.object_iterator, -1); + break; + } + + case value_t::array: + { + std::advance(m_it.array_iterator, -1); + break; + } + + default: + { + --m_it.primitive_iterator; + break; + } + } + + return *this; + } + + /*! + @brief comparison: equal + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator==(const iter_impl& other) const + { + // if objects are not the same, the comparison is undefined + if (JSON_UNLIKELY(m_object != other.m_object)) + { + JSON_THROW(invalid_iterator::create(212, "cannot compare iterators of different containers")); + } + + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + return (m_it.object_iterator == other.m_it.object_iterator); + + case value_t::array: + return (m_it.array_iterator == other.m_it.array_iterator); + + default: + return (m_it.primitive_iterator == other.m_it.primitive_iterator); + } + } + + /*! + @brief comparison: not equal + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator!=(const iter_impl& other) const + { + return not operator==(other); + } + + /*! + @brief comparison: smaller + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator<(const iter_impl& other) const + { + // if objects are not the same, the comparison is undefined + if (JSON_UNLIKELY(m_object != other.m_object)) + { + JSON_THROW(invalid_iterator::create(212, "cannot compare iterators of different containers")); + } + + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + JSON_THROW(invalid_iterator::create(213, "cannot compare order of object iterators")); + + case value_t::array: + return (m_it.array_iterator < other.m_it.array_iterator); + + default: + return (m_it.primitive_iterator < other.m_it.primitive_iterator); + } + } + + /*! + @brief comparison: less than or equal + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator<=(const iter_impl& other) const + { + return not other.operator < (*this); + } + + /*! + @brief comparison: greater than + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator>(const iter_impl& other) const + { + return not operator<=(other); + } + + /*! + @brief comparison: greater than or equal + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + bool operator>=(const iter_impl& other) const + { + return not operator<(other); + } + + /*! + @brief add to iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl& operator+=(difference_type i) + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + JSON_THROW(invalid_iterator::create(209, "cannot use offsets with object iterators")); + + case value_t::array: + { + std::advance(m_it.array_iterator, i); + break; + } + + default: + { + m_it.primitive_iterator += i; + break; + } + } + + return *this; + } + + /*! + @brief subtract from iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl& operator-=(difference_type i) + { + return operator+=(-i); + } + + /*! + @brief add to iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl operator+(difference_type i) const + { + auto result = *this; + result += i; + return result; + } + + /*! + @brief addition of distance and iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + friend iter_impl operator+(difference_type i, const iter_impl& it) + { + auto result = it; + result += i; + return result; + } + + /*! + @brief subtract from iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + iter_impl operator-(difference_type i) const + { + auto result = *this; + result -= i; + return result; + } + + /*! + @brief return difference + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + difference_type operator-(const iter_impl& other) const + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + JSON_THROW(invalid_iterator::create(209, "cannot use offsets with object iterators")); + + case value_t::array: + return m_it.array_iterator - other.m_it.array_iterator; + + default: + return m_it.primitive_iterator - other.m_it.primitive_iterator; + } + } + + /*! + @brief access to successor + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + reference operator[](difference_type n) const + { + assert(m_object != nullptr); + + switch (m_object->m_type) + { + case value_t::object: + JSON_THROW(invalid_iterator::create(208, "cannot use operator[] for object iterators")); + + case value_t::array: + return *std::next(m_it.array_iterator, n); + + case value_t::null: + JSON_THROW(invalid_iterator::create(214, "cannot get value")); + + default: + { + if (JSON_LIKELY(m_it.primitive_iterator.get_value() == -n)) + { + return *m_object; + } + + JSON_THROW(invalid_iterator::create(214, "cannot get value")); + } + } + } + + /*! + @brief return the key of an object iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + const typename object_t::key_type& key() const + { + assert(m_object != nullptr); + + if (JSON_LIKELY(m_object->is_object())) + { + return m_it.object_iterator->first; + } + + JSON_THROW(invalid_iterator::create(207, "cannot use key() for non-object iterators")); + } + + /*! + @brief return the value of an iterator + @pre The iterator is initialized; i.e. `m_object != nullptr`. + */ + reference value() const + { + return operator*(); + } + + private: + /// associated JSON instance + pointer m_object = nullptr; + /// the actual iterator of the associated instance + internal_iterator<typename std::remove_const<BasicJsonType>::type> m_it; +}; +} +} diff --git a/include/lib/modernjson/detail/iterators/iteration_proxy.hpp b/include/lib/modernjson/detail/iterators/iteration_proxy.hpp new file mode 100644 index 0000000..2da7154 --- /dev/null +++ b/include/lib/modernjson/detail/iterators/iteration_proxy.hpp @@ -0,0 +1,128 @@ +#pragma once + +#include <cstddef> // size_t +#include <string> // string, to_string +#include <iterator> // input_iterator_tag + +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +/// proxy class for the items() function +template<typename IteratorType> class iteration_proxy +{ + private: + /// helper class for iteration + class iteration_proxy_internal + { + public: + using difference_type = std::ptrdiff_t; + using value_type = iteration_proxy_internal; + using pointer = iteration_proxy_internal*; + using reference = iteration_proxy_internal&; + using iterator_category = std::input_iterator_tag; + + private: + /// the iterator + IteratorType anchor; + /// an index for arrays (used to create key names) + std::size_t array_index = 0; + /// last stringified array index + mutable std::size_t array_index_last = 0; + /// a string representation of the array index + mutable std::string array_index_str = "0"; + /// an empty string (to return a reference for primitive values) + const std::string empty_str = ""; + + public: + explicit iteration_proxy_internal(IteratorType it) noexcept : anchor(it) {} + + iteration_proxy_internal(const iteration_proxy_internal&) = default; + iteration_proxy_internal& operator=(const iteration_proxy_internal&) = default; + + /// dereference operator (needed for range-based for) + iteration_proxy_internal& operator*() + { + return *this; + } + + /// increment operator (needed for range-based for) + iteration_proxy_internal& operator++() + { + ++anchor; + ++array_index; + + return *this; + } + + /// equality operator (needed for InputIterator) + bool operator==(const iteration_proxy_internal& o) const noexcept + { + return anchor == o.anchor; + } + + /// inequality operator (needed for range-based for) + bool operator!=(const iteration_proxy_internal& o) const noexcept + { + return anchor != o.anchor; + } + + /// return key of the iterator + const std::string& key() const + { + assert(anchor.m_object != nullptr); + + switch (anchor.m_object->type()) + { + // use integer array index as key + case value_t::array: + { + if (array_index != array_index_last) + { + array_index_str = std::to_string(array_index); + array_index_last = array_index; + } + return array_index_str; + } + + // use key from the object + case value_t::object: + return anchor.key(); + + // use an empty key for all primitive types + default: + return empty_str; + } + } + + /// return value of the iterator + typename IteratorType::reference value() const + { + return anchor.value(); + } + }; + + /// the container to iterate + typename IteratorType::reference container; + + public: + /// construct iteration proxy from a container + explicit iteration_proxy(typename IteratorType::reference cont) noexcept + : container(cont) {} + + /// return iterator begin (needed for range-based for) + iteration_proxy_internal begin() noexcept + { + return iteration_proxy_internal(container.begin()); + } + + /// return iterator end (needed for range-based for) + iteration_proxy_internal end() noexcept + { + return iteration_proxy_internal(container.end()); + } +}; +} +} diff --git a/include/lib/modernjson/detail/iterators/json_reverse_iterator.hpp b/include/lib/modernjson/detail/iterators/json_reverse_iterator.hpp new file mode 100644 index 0000000..2750de4 --- /dev/null +++ b/include/lib/modernjson/detail/iterators/json_reverse_iterator.hpp @@ -0,0 +1,119 @@ +#pragma once + +#include <cstddef> // ptrdiff_t +#include <iterator> // reverse_iterator +#include <utility> // declval + +namespace nlohmann +{ +namespace detail +{ +////////////////////// +// reverse_iterator // +////////////////////// + +/*! +@brief a template for a reverse iterator class + +@tparam Base the base iterator type to reverse. Valid types are @ref +iterator (to create @ref reverse_iterator) and @ref const_iterator (to +create @ref const_reverse_iterator). + +@requirement The class satisfies the following concept requirements: +- +[BidirectionalIterator](https://en.cppreference.com/w/cpp/named_req/BidirectionalIterator): + The iterator that can be moved can be moved in both directions (i.e. + incremented and decremented). +- [OutputIterator](https://en.cppreference.com/w/cpp/named_req/OutputIterator): + It is possible to write to the pointed-to element (only if @a Base is + @ref iterator). + +@since version 1.0.0 +*/ +template<typename Base> +class json_reverse_iterator : public std::reverse_iterator<Base> +{ + public: + using difference_type = std::ptrdiff_t; + /// shortcut to the reverse iterator adapter + using base_iterator = std::reverse_iterator<Base>; + /// the reference type for the pointed-to element + using reference = typename Base::reference; + + /// create reverse iterator from iterator + explicit json_reverse_iterator(const typename base_iterator::iterator_type& it) noexcept + : base_iterator(it) {} + + /// create reverse iterator from base class + explicit json_reverse_iterator(const base_iterator& it) noexcept : base_iterator(it) {} + + /// post-increment (it++) + json_reverse_iterator const operator++(int) + { + return static_cast<json_reverse_iterator>(base_iterator::operator++(1)); + } + + /// pre-increment (++it) + json_reverse_iterator& operator++() + { + return static_cast<json_reverse_iterator&>(base_iterator::operator++()); + } + + /// post-decrement (it--) + json_reverse_iterator const operator--(int) + { + return static_cast<json_reverse_iterator>(base_iterator::operator--(1)); + } + + /// pre-decrement (--it) + json_reverse_iterator& operator--() + { + return static_cast<json_reverse_iterator&>(base_iterator::operator--()); + } + + /// add to iterator + json_reverse_iterator& operator+=(difference_type i) + { + return static_cast<json_reverse_iterator&>(base_iterator::operator+=(i)); + } + + /// add to iterator + json_reverse_iterator operator+(difference_type i) const + { + return static_cast<json_reverse_iterator>(base_iterator::operator+(i)); + } + + /// subtract from iterator + json_reverse_iterator operator-(difference_type i) const + { + return static_cast<json_reverse_iterator>(base_iterator::operator-(i)); + } + + /// return difference + difference_type operator-(const json_reverse_iterator& other) const + { + return base_iterator(*this) - base_iterator(other); + } + + /// access to successor + reference operator[](difference_type n) const + { + return *(this->operator+(n)); + } + + /// return the key of an object iterator + auto key() const -> decltype(std::declval<Base>().key()) + { + auto it = --this->base(); + return it.key(); + } + + /// return the value of an iterator + reference value() const + { + auto it = --this->base(); + return it.operator * (); + } +}; +} +} diff --git a/include/lib/modernjson/detail/iterators/primitive_iterator.hpp b/include/lib/modernjson/detail/iterators/primitive_iterator.hpp new file mode 100644 index 0000000..db3f897 --- /dev/null +++ b/include/lib/modernjson/detail/iterators/primitive_iterator.hpp @@ -0,0 +1,120 @@ +#pragma once + +#include <cstddef> // ptrdiff_t +#include <limits> // numeric_limits + +namespace nlohmann +{ +namespace detail +{ +/* +@brief an iterator for primitive JSON types + +This class models an iterator for primitive JSON types (boolean, number, +string). It's only purpose is to allow the iterator/const_iterator classes +to "iterate" over primitive values. Internally, the iterator is modeled by +a `difference_type` variable. Value begin_value (`0`) models the begin, +end_value (`1`) models past the end. +*/ +class primitive_iterator_t +{ + private: + using difference_type = std::ptrdiff_t; + static constexpr difference_type begin_value = 0; + static constexpr difference_type end_value = begin_value + 1; + + /// iterator as signed integer type + difference_type m_it = (std::numeric_limits<std::ptrdiff_t>::min)(); + + public: + constexpr difference_type get_value() const noexcept + { + return m_it; + } + + /// set iterator to a defined beginning + void set_begin() noexcept + { + m_it = begin_value; + } + + /// set iterator to a defined past the end + void set_end() noexcept + { + m_it = end_value; + } + + /// return whether the iterator can be dereferenced + constexpr bool is_begin() const noexcept + { + return m_it == begin_value; + } + + /// return whether the iterator is at end + constexpr bool is_end() const noexcept + { + return m_it == end_value; + } + + friend constexpr bool operator==(primitive_iterator_t lhs, primitive_iterator_t rhs) noexcept + { + return lhs.m_it == rhs.m_it; + } + + friend constexpr bool operator<(primitive_iterator_t lhs, primitive_iterator_t rhs) noexcept + { + return lhs.m_it < rhs.m_it; + } + + primitive_iterator_t operator+(difference_type n) noexcept + { + auto result = *this; + result += n; + return result; + } + + friend constexpr difference_type operator-(primitive_iterator_t lhs, primitive_iterator_t rhs) noexcept + { + return lhs.m_it - rhs.m_it; + } + + primitive_iterator_t& operator++() noexcept + { + ++m_it; + return *this; + } + + primitive_iterator_t const operator++(int) noexcept + { + auto result = *this; + ++m_it; + return result; + } + + primitive_iterator_t& operator--() noexcept + { + --m_it; + return *this; + } + + primitive_iterator_t const operator--(int) noexcept + { + auto result = *this; + --m_it; + return result; + } + + primitive_iterator_t& operator+=(difference_type n) noexcept + { + m_it += n; + return *this; + } + + primitive_iterator_t& operator-=(difference_type n) noexcept + { + m_it -= n; + return *this; + } +}; +} +} diff --git a/include/lib/modernjson/detail/json_pointer.hpp b/include/lib/modernjson/detail/json_pointer.hpp new file mode 100644 index 0000000..5689384 --- /dev/null +++ b/include/lib/modernjson/detail/json_pointer.hpp @@ -0,0 +1,696 @@ +#pragma once + +#include <cassert> // assert +#include <numeric> // accumulate +#include <string> // string +#include <vector> // vector + +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +template<typename BasicJsonType> +class json_pointer +{ + // allow basic_json to access private members + NLOHMANN_BASIC_JSON_TPL_DECLARATION + friend class basic_json; + + public: + /*! + @brief create JSON pointer + + Create a JSON pointer according to the syntax described in + [Section 3 of RFC6901](https://tools.ietf.org/html/rfc6901#section-3). + + @param[in] s string representing the JSON pointer; if omitted, the empty + string is assumed which references the whole JSON value + + @throw parse_error.107 if the given JSON pointer @a s is nonempty and does + not begin with a slash (`/`); see example below + + @throw parse_error.108 if a tilde (`~`) in the given JSON pointer @a s is + not followed by `0` (representing `~`) or `1` (representing `/`); see + example below + + @liveexample{The example shows the construction several valid JSON pointers + as well as the exceptional behavior.,json_pointer} + + @since version 2.0.0 + */ + explicit json_pointer(const std::string& s = "") + : reference_tokens(split(s)) + {} + + /*! + @brief return a string representation of the JSON pointer + + @invariant For each JSON pointer `ptr`, it holds: + @code {.cpp} + ptr == json_pointer(ptr.to_string()); + @endcode + + @return a string representation of the JSON pointer + + @liveexample{The example shows the result of `to_string`., + json_pointer__to_string} + + @since version 2.0.0 + */ + std::string to_string() const noexcept + { + return std::accumulate(reference_tokens.begin(), reference_tokens.end(), + std::string{}, + [](const std::string & a, const std::string & b) + { + return a + "/" + escape(b); + }); + } + + /// @copydoc to_string() + operator std::string() const + { + return to_string(); + } + + /*! + @param[in] s reference token to be converted into an array index + + @return integer representation of @a s + + @throw out_of_range.404 if string @a s could not be converted to an integer + */ + static int array_index(const std::string& s) + { + std::size_t processed_chars = 0; + const int res = std::stoi(s, &processed_chars); + + // check if the string was completely read + if (JSON_UNLIKELY(processed_chars != s.size())) + { + JSON_THROW(detail::out_of_range::create(404, "unresolved reference token '" + s + "'")); + } + + return res; + } + + private: + /*! + @brief remove and return last reference pointer + @throw out_of_range.405 if JSON pointer has no parent + */ + std::string pop_back() + { + if (JSON_UNLIKELY(is_root())) + { + JSON_THROW(detail::out_of_range::create(405, "JSON pointer has no parent")); + } + + auto last = reference_tokens.back(); + reference_tokens.pop_back(); + return last; + } + + /// return whether pointer points to the root document + bool is_root() const + { + return reference_tokens.empty(); + } + + json_pointer top() const + { + if (JSON_UNLIKELY(is_root())) + { + JSON_THROW(detail::out_of_range::create(405, "JSON pointer has no parent")); + } + + json_pointer result = *this; + result.reference_tokens = {reference_tokens[0]}; + return result; + } + + /*! + @brief create and return a reference to the pointed to value + + @complexity Linear in the number of reference tokens. + + @throw parse_error.109 if array index is not a number + @throw type_error.313 if value cannot be unflattened + */ + BasicJsonType& get_and_create(BasicJsonType& j) const + { + using size_type = typename BasicJsonType::size_type; + auto result = &j; + + // in case no reference tokens exist, return a reference to the JSON value + // j which will be overwritten by a primitive value + for (const auto& reference_token : reference_tokens) + { + switch (result->m_type) + { + case detail::value_t::null: + { + if (reference_token == "0") + { + // start a new array if reference token is 0 + result = &result->operator[](0); + } + else + { + // start a new object otherwise + result = &result->operator[](reference_token); + } + break; + } + + case detail::value_t::object: + { + // create an entry in the object + result = &result->operator[](reference_token); + break; + } + + case detail::value_t::array: + { + // create an entry in the array + JSON_TRY + { + result = &result->operator[](static_cast<size_type>(array_index(reference_token))); + } + JSON_CATCH(std::invalid_argument&) + { + JSON_THROW(detail::parse_error::create(109, 0, "array index '" + reference_token + "' is not a number")); + } + break; + } + + /* + The following code is only reached if there exists a reference + token _and_ the current value is primitive. In this case, we have + an error situation, because primitive values may only occur as + single value; that is, with an empty list of reference tokens. + */ + default: + JSON_THROW(detail::type_error::create(313, "invalid value to unflatten")); + } + } + + return *result; + } + + /*! + @brief return a reference to the pointed to value + + @note This version does not throw if a value is not present, but tries to + create nested values instead. For instance, calling this function + with pointer `"/this/that"` on a null value is equivalent to calling + `operator[]("this").operator[]("that")` on that value, effectively + changing the null value to an object. + + @param[in] ptr a JSON value + + @return reference to the JSON value pointed to by the JSON pointer + + @complexity Linear in the length of the JSON pointer. + + @throw parse_error.106 if an array index begins with '0' + @throw parse_error.109 if an array index was not a number + @throw out_of_range.404 if the JSON pointer can not be resolved + */ + BasicJsonType& get_unchecked(BasicJsonType* ptr) const + { + using size_type = typename BasicJsonType::size_type; + for (const auto& reference_token : reference_tokens) + { + // convert null values to arrays or objects before continuing + if (ptr->m_type == detail::value_t::null) + { + // check if reference token is a number + const bool nums = + std::all_of(reference_token.begin(), reference_token.end(), + [](const char x) + { + return (x >= '0' and x <= '9'); + }); + + // change value to array for numbers or "-" or to object otherwise + *ptr = (nums or reference_token == "-") + ? detail::value_t::array + : detail::value_t::object; + } + + switch (ptr->m_type) + { + case detail::value_t::object: + { + // use unchecked object access + ptr = &ptr->operator[](reference_token); + break; + } + + case detail::value_t::array: + { + // error condition (cf. RFC 6901, Sect. 4) + if (JSON_UNLIKELY(reference_token.size() > 1 and reference_token[0] == '0')) + { + JSON_THROW(detail::parse_error::create(106, 0, + "array index '" + reference_token + + "' must not begin with '0'")); + } + + if (reference_token == "-") + { + // explicitly treat "-" as index beyond the end + ptr = &ptr->operator[](ptr->m_value.array->size()); + } + else + { + // convert array index to number; unchecked access + JSON_TRY + { + ptr = &ptr->operator[]( + static_cast<size_type>(array_index(reference_token))); + } + JSON_CATCH(std::invalid_argument&) + { + JSON_THROW(detail::parse_error::create(109, 0, "array index '" + reference_token + "' is not a number")); + } + } + break; + } + + default: + JSON_THROW(detail::out_of_range::create(404, "unresolved reference token '" + reference_token + "'")); + } + } + + return *ptr; + } + + /*! + @throw parse_error.106 if an array index begins with '0' + @throw parse_error.109 if an array index was not a number + @throw out_of_range.402 if the array index '-' is used + @throw out_of_range.404 if the JSON pointer can not be resolved + */ + BasicJsonType& get_checked(BasicJsonType* ptr) const + { + using size_type = typename BasicJsonType::size_type; + for (const auto& reference_token : reference_tokens) + { + switch (ptr->m_type) + { + case detail::value_t::object: + { + // note: at performs range check + ptr = &ptr->at(reference_token); + break; + } + + case detail::value_t::array: + { + if (JSON_UNLIKELY(reference_token == "-")) + { + // "-" always fails the range check + JSON_THROW(detail::out_of_range::create(402, + "array index '-' (" + std::to_string(ptr->m_value.array->size()) + + ") is out of range")); + } + + // error condition (cf. RFC 6901, Sect. 4) + if (JSON_UNLIKELY(reference_token.size() > 1 and reference_token[0] == '0')) + { + JSON_THROW(detail::parse_error::create(106, 0, + "array index '" + reference_token + + "' must not begin with '0'")); + } + + // note: at performs range check + JSON_TRY + { + ptr = &ptr->at(static_cast<size_type>(array_index(reference_token))); + } + JSON_CATCH(std::invalid_argument&) + { + JSON_THROW(detail::parse_error::create(109, 0, "array index '" + reference_token + "' is not a number")); + } + break; + } + + default: + JSON_THROW(detail::out_of_range::create(404, "unresolved reference token '" + reference_token + "'")); + } + } + + return *ptr; + } + + /*! + @brief return a const reference to the pointed to value + + @param[in] ptr a JSON value + + @return const reference to the JSON value pointed to by the JSON + pointer + + @throw parse_error.106 if an array index begins with '0' + @throw parse_error.109 if an array index was not a number + @throw out_of_range.402 if the array index '-' is used + @throw out_of_range.404 if the JSON pointer can not be resolved + */ + const BasicJsonType& get_unchecked(const BasicJsonType* ptr) const + { + using size_type = typename BasicJsonType::size_type; + for (const auto& reference_token : reference_tokens) + { + switch (ptr->m_type) + { + case detail::value_t::object: + { + // use unchecked object access + ptr = &ptr->operator[](reference_token); + break; + } + + case detail::value_t::array: + { + if (JSON_UNLIKELY(reference_token == "-")) + { + // "-" cannot be used for const access + JSON_THROW(detail::out_of_range::create(402, + "array index '-' (" + std::to_string(ptr->m_value.array->size()) + + ") is out of range")); + } + + // error condition (cf. RFC 6901, Sect. 4) + if (JSON_UNLIKELY(reference_token.size() > 1 and reference_token[0] == '0')) + { + JSON_THROW(detail::parse_error::create(106, 0, + "array index '" + reference_token + + "' must not begin with '0'")); + } + + // use unchecked array access + JSON_TRY + { + ptr = &ptr->operator[]( + static_cast<size_type>(array_index(reference_token))); + } + JSON_CATCH(std::invalid_argument&) + { + JSON_THROW(detail::parse_error::create(109, 0, "array index '" + reference_token + "' is not a number")); + } + break; + } + + default: + JSON_THROW(detail::out_of_range::create(404, "unresolved reference token '" + reference_token + "'")); + } + } + + return *ptr; + } + + /*! + @throw parse_error.106 if an array index begins with '0' + @throw parse_error.109 if an array index was not a number + @throw out_of_range.402 if the array index '-' is used + @throw out_of_range.404 if the JSON pointer can not be resolved + */ + const BasicJsonType& get_checked(const BasicJsonType* ptr) const + { + using size_type = typename BasicJsonType::size_type; + for (const auto& reference_token : reference_tokens) + { + switch (ptr->m_type) + { + case detail::value_t::object: + { + // note: at performs range check + ptr = &ptr->at(reference_token); + break; + } + + case detail::value_t::array: + { + if (JSON_UNLIKELY(reference_token == "-")) + { + // "-" always fails the range check + JSON_THROW(detail::out_of_range::create(402, + "array index '-' (" + std::to_string(ptr->m_value.array->size()) + + ") is out of range")); + } + + // error condition (cf. RFC 6901, Sect. 4) + if (JSON_UNLIKELY(reference_token.size() > 1 and reference_token[0] == '0')) + { + JSON_THROW(detail::parse_error::create(106, 0, + "array index '" + reference_token + + "' must not begin with '0'")); + } + + // note: at performs range check + JSON_TRY + { + ptr = &ptr->at(static_cast<size_type>(array_index(reference_token))); + } + JSON_CATCH(std::invalid_argument&) + { + JSON_THROW(detail::parse_error::create(109, 0, "array index '" + reference_token + "' is not a number")); + } + break; + } + + default: + JSON_THROW(detail::out_of_range::create(404, "unresolved reference token '" + reference_token + "'")); + } + } + + return *ptr; + } + + /*! + @brief split the string input to reference tokens + + @note This function is only called by the json_pointer constructor. + All exceptions below are documented there. + + @throw parse_error.107 if the pointer is not empty or begins with '/' + @throw parse_error.108 if character '~' is not followed by '0' or '1' + */ + static std::vector<std::string> split(const std::string& reference_string) + { + std::vector<std::string> result; + + // special case: empty reference string -> no reference tokens + if (reference_string.empty()) + { + return result; + } + + // check if nonempty reference string begins with slash + if (JSON_UNLIKELY(reference_string[0] != '/')) + { + JSON_THROW(detail::parse_error::create(107, 1, + "JSON pointer must be empty or begin with '/' - was: '" + + reference_string + "'")); + } + + // extract the reference tokens: + // - slash: position of the last read slash (or end of string) + // - start: position after the previous slash + for ( + // search for the first slash after the first character + std::size_t slash = reference_string.find_first_of('/', 1), + // set the beginning of the first reference token + start = 1; + // we can stop if start == string::npos+1 = 0 + start != 0; + // set the beginning of the next reference token + // (will eventually be 0 if slash == std::string::npos) + start = slash + 1, + // find next slash + slash = reference_string.find_first_of('/', start)) + { + // use the text between the beginning of the reference token + // (start) and the last slash (slash). + auto reference_token = reference_string.substr(start, slash - start); + + // check reference tokens are properly escaped + for (std::size_t pos = reference_token.find_first_of('~'); + pos != std::string::npos; + pos = reference_token.find_first_of('~', pos + 1)) + { + assert(reference_token[pos] == '~'); + + // ~ must be followed by 0 or 1 + if (JSON_UNLIKELY(pos == reference_token.size() - 1 or + (reference_token[pos + 1] != '0' and + reference_token[pos + 1] != '1'))) + { + JSON_THROW(detail::parse_error::create(108, 0, "escape character '~' must be followed with '0' or '1'")); + } + } + + // finally, store the reference token + unescape(reference_token); + result.push_back(reference_token); + } + + return result; + } + + /*! + @brief replace all occurrences of a substring by another string + + @param[in,out] s the string to manipulate; changed so that all + occurrences of @a f are replaced with @a t + @param[in] f the substring to replace with @a t + @param[in] t the string to replace @a f + + @pre The search string @a f must not be empty. **This precondition is + enforced with an assertion.** + + @since version 2.0.0 + */ + static void replace_substring(std::string& s, const std::string& f, + const std::string& t) + { + assert(not f.empty()); + for (auto pos = s.find(f); // find first occurrence of f + pos != std::string::npos; // make sure f was found + s.replace(pos, f.size(), t), // replace with t, and + pos = s.find(f, pos + t.size())) // find next occurrence of f + {} + } + + /// escape "~"" to "~0" and "/" to "~1" + static std::string escape(std::string s) + { + replace_substring(s, "~", "~0"); + replace_substring(s, "/", "~1"); + return s; + } + + /// unescape "~1" to tilde and "~0" to slash (order is important!) + static void unescape(std::string& s) + { + replace_substring(s, "~1", "/"); + replace_substring(s, "~0", "~"); + } + + /*! + @param[in] reference_string the reference string to the current value + @param[in] value the value to consider + @param[in,out] result the result object to insert values to + + @note Empty objects or arrays are flattened to `null`. + */ + static void flatten(const std::string& reference_string, + const BasicJsonType& value, + BasicJsonType& result) + { + switch (value.m_type) + { + case detail::value_t::array: + { + if (value.m_value.array->empty()) + { + // flatten empty array as null + result[reference_string] = nullptr; + } + else + { + // iterate array and use index as reference string + for (std::size_t i = 0; i < value.m_value.array->size(); ++i) + { + flatten(reference_string + "/" + std::to_string(i), + value.m_value.array->operator[](i), result); + } + } + break; + } + + case detail::value_t::object: + { + if (value.m_value.object->empty()) + { + // flatten empty object as null + result[reference_string] = nullptr; + } + else + { + // iterate object and use keys as reference string + for (const auto& element : *value.m_value.object) + { + flatten(reference_string + "/" + escape(element.first), element.second, result); + } + } + break; + } + + default: + { + // add primitive value with its reference string + result[reference_string] = value; + break; + } + } + } + + /*! + @param[in] value flattened JSON + + @return unflattened JSON + + @throw parse_error.109 if array index is not a number + @throw type_error.314 if value is not an object + @throw type_error.315 if object values are not primitive + @throw type_error.313 if value cannot be unflattened + */ + static BasicJsonType + unflatten(const BasicJsonType& value) + { + if (JSON_UNLIKELY(not value.is_object())) + { + JSON_THROW(detail::type_error::create(314, "only objects can be unflattened")); + } + + BasicJsonType result; + + // iterate the JSON object values + for (const auto& element : *value.m_value.object) + { + if (JSON_UNLIKELY(not element.second.is_primitive())) + { + JSON_THROW(detail::type_error::create(315, "values in object must be primitive")); + } + + // assign value to reference pointed to by JSON pointer; Note that if + // the JSON pointer is "" (i.e., points to the whole value), function + // get_and_create returns a reference to result itself. An assignment + // will then create a primitive value. + json_pointer(element.first).get_and_create(result) = element.second; + } + + return result; + } + + friend bool operator==(json_pointer const& lhs, + json_pointer const& rhs) noexcept + { + return (lhs.reference_tokens == rhs.reference_tokens); + } + + friend bool operator!=(json_pointer const& lhs, + json_pointer const& rhs) noexcept + { + return not (lhs == rhs); + } + + /// the reference tokens + std::vector<std::string> reference_tokens; +}; +} diff --git a/include/lib/modernjson/detail/json_ref.hpp b/include/lib/modernjson/detail/json_ref.hpp new file mode 100644 index 0000000..8285be2 --- /dev/null +++ b/include/lib/modernjson/detail/json_ref.hpp @@ -0,0 +1,63 @@ +#pragma once + +#include <initializer_list> +#include <utility> + +namespace nlohmann +{ +namespace detail +{ +template<typename BasicJsonType> +class json_ref +{ + public: + using value_type = BasicJsonType; + + json_ref(value_type&& value) + : owned_value(std::move(value)), value_ref(&owned_value), is_rvalue(true) + {} + + json_ref(const value_type& value) + : value_ref(const_cast<value_type*>(&value)), is_rvalue(false) + {} + + json_ref(std::initializer_list<json_ref> init) + : owned_value(init), value_ref(&owned_value), is_rvalue(true) + {} + + template<class... Args> + json_ref(Args&& ... args) + : owned_value(std::forward<Args>(args)...), value_ref(&owned_value), is_rvalue(true) + {} + + // class should be movable only + json_ref(json_ref&&) = default; + json_ref(const json_ref&) = delete; + json_ref& operator=(const json_ref&) = delete; + + value_type moved_or_copied() const + { + if (is_rvalue) + { + return std::move(*value_ref); + } + return *value_ref; + } + + value_type const& operator*() const + { + return *static_cast<value_type const*>(value_ref); + } + + value_type const* operator->() const + { + return static_cast<value_type const*>(value_ref); + } + + private: + mutable value_type owned_value = nullptr; + value_type* value_ref = nullptr; + const bool is_rvalue; +}; +} +} diff --git a/include/lib/modernjson/detail/macro_scope.hpp b/include/lib/modernjson/detail/macro_scope.hpp new file mode 100644 index 0000000..80fbece --- /dev/null +++ b/include/lib/modernjson/detail/macro_scope.hpp @@ -0,0 +1,104 @@ +#pragma once + +// This file contains all internal macro definitions +// You MUST include macro_unscope.hpp at the end of json.hpp to undef all of them + +// exclude unsupported compilers +#if !defined(JSON_SKIP_UNSUPPORTED_COMPILER_CHECK) + #if defined(__clang__) + #if (__clang_major__ * 10000 + __clang_minor__ * 100 + __clang_patchlevel__) < 30400 + #error "unsupported Clang version - see https://github.com/nlohmann/json#supported-compilers" + #endif + #elif defined(__GNUC__) && !(defined(__ICC) || defined(__INTEL_COMPILER)) + #if (__GNUC__ * 10000 + __GNUC_MINOR__ * 100 + __GNUC_PATCHLEVEL__) < 40900 + #error "unsupported GCC version - see https://github.com/nlohmann/json#supported-compilers" + #endif + #endif +#endif + +// disable float-equal warnings on GCC/clang +#if defined(__clang__) || defined(__GNUC__) || defined(__GNUG__) + #pragma GCC diagnostic push + #pragma GCC diagnostic ignored "-Wfloat-equal" +#endif + +// disable documentation warnings on clang +#if defined(__clang__) + #pragma GCC diagnostic push + #pragma GCC diagnostic ignored "-Wdocumentation" +#endif + +// allow for portable deprecation warnings +#if defined(__clang__) || defined(__GNUC__) || defined(__GNUG__) + #define JSON_DEPRECATED __attribute__((deprecated)) +#elif defined(_MSC_VER) + #define JSON_DEPRECATED __declspec(deprecated) +#else + #define JSON_DEPRECATED +#endif + +// allow to disable exceptions +#if (defined(__cpp_exceptions) || defined(__EXCEPTIONS) || defined(_CPPUNWIND)) && !defined(JSON_NOEXCEPTION) + #define JSON_THROW(exception) throw exception + #define JSON_TRY try + #define JSON_CATCH(exception) catch(exception) + #define JSON_INTERNAL_CATCH(exception) catch(exception) +#else + #define JSON_THROW(exception) std::abort() + #define JSON_TRY if(true) + #define JSON_CATCH(exception) if(false) + #define JSON_INTERNAL_CATCH(exception) if(false) +#endif + +// override exception macros +#if defined(JSON_THROW_USER) + #undef JSON_THROW + #define JSON_THROW JSON_THROW_USER +#endif +#if defined(JSON_TRY_USER) + #undef JSON_TRY + #define JSON_TRY JSON_TRY_USER +#endif +#if defined(JSON_CATCH_USER) + #undef JSON_CATCH + #define JSON_CATCH JSON_CATCH_USER + #undef JSON_INTERNAL_CATCH + #define JSON_INTERNAL_CATCH JSON_CATCH_USER +#endif +#if defined(JSON_INTERNAL_CATCH_USER) + #undef JSON_INTERNAL_CATCH + #define JSON_INTERNAL_CATCH JSON_INTERNAL_CATCH_USER +#endif + +// manual branch prediction +#if defined(__clang__) || defined(__GNUC__) || defined(__GNUG__) + #define JSON_LIKELY(x) __builtin_expect(!!(x), 1) + #define JSON_UNLIKELY(x) __builtin_expect(!!(x), 0) +#else + #define JSON_LIKELY(x) x + #define JSON_UNLIKELY(x) x +#endif + +// C++ language standard detection +#if (defined(__cplusplus) && __cplusplus >= 201703L) || (defined(_HAS_CXX17) && _HAS_CXX17 == 1) // fix for issue #464 + #define JSON_HAS_CPP_17 + #define JSON_HAS_CPP_14 +#elif (defined(__cplusplus) && __cplusplus >= 201402L) || (defined(_HAS_CXX14) && _HAS_CXX14 == 1) + #define JSON_HAS_CPP_14 +#endif + +// Ugly macros to avoid uglier copy-paste when specializing basic_json. They +// may be removed in the future once the class is split. + +#define NLOHMANN_BASIC_JSON_TPL_DECLARATION \ + template<template<typename, typename, typename...> class ObjectType, \ + template<typename, typename...> class ArrayType, \ + class StringType, class BooleanType, class NumberIntegerType, \ + class NumberUnsignedType, class NumberFloatType, \ + template<typename> class AllocatorType, \ + template<typename, typename = void> class JSONSerializer> + +#define NLOHMANN_BASIC_JSON_TPL \ + basic_json<ObjectType, ArrayType, StringType, BooleanType, \ + NumberIntegerType, NumberUnsignedType, NumberFloatType, \ + AllocatorType, JSONSerializer> diff --git a/include/lib/modernjson/detail/macro_unscope.hpp b/include/lib/modernjson/detail/macro_unscope.hpp new file mode 100644 index 0000000..4c5aa91 --- /dev/null +++ b/include/lib/modernjson/detail/macro_unscope.hpp @@ -0,0 +1,22 @@ +#pragma once + +// restore GCC/clang diagnostic settings +#if defined(__clang__) || defined(__GNUC__) || defined(__GNUG__) + #pragma GCC diagnostic pop +#endif +#if defined(__clang__) + #pragma GCC diagnostic pop +#endif + +// clean up +#undef JSON_INTERNAL_CATCH +#undef JSON_CATCH +#undef JSON_THROW +#undef JSON_TRY +#undef JSON_LIKELY +#undef JSON_UNLIKELY +#undef JSON_DEPRECATED +#undef JSON_HAS_CPP_14 +#undef JSON_HAS_CPP_17 +#undef NLOHMANN_BASIC_JSON_TPL_DECLARATION +#undef NLOHMANN_BASIC_JSON_TPL diff --git a/include/lib/modernjson/detail/meta/cpp_future.hpp b/include/lib/modernjson/detail/meta/cpp_future.hpp new file mode 100644 index 0000000..fa7478b --- /dev/null +++ b/include/lib/modernjson/detail/meta/cpp_future.hpp @@ -0,0 +1,63 @@ +#pragma once + +#include <ciso646> // not +#include <cstddef> // size_t +#include <type_traits> // conditional, enable_if, false_type, integral_constant, is_constructible, is_integral, is_same, remove_cv, remove_reference, true_type + +namespace nlohmann +{ +namespace detail +{ +// alias templates to reduce boilerplate +template<bool B, typename T = void> +using enable_if_t = typename std::enable_if<B, T>::type; + +template<typename T> +using uncvref_t = typename std::remove_cv<typename std::remove_reference<T>::type>::type; + +// implementation of C++14 index_sequence and affiliates +// source: https://stackoverflow.com/a/32223343 +template<std::size_t... Ints> +struct index_sequence +{ + using type = index_sequence; + using value_type = std::size_t; + static constexpr std::size_t size() noexcept + { + return sizeof...(Ints); + } +}; + +template<class Sequence1, class Sequence2> +struct merge_and_renumber; + +template<std::size_t... I1, std::size_t... I2> +struct merge_and_renumber<index_sequence<I1...>, index_sequence<I2...>> + : index_sequence < I1..., (sizeof...(I1) + I2)... > {}; + +template<std::size_t N> +struct make_index_sequence + : merge_and_renumber < typename make_index_sequence < N / 2 >::type, + typename make_index_sequence < N - N / 2 >::type > {}; + +template<> struct make_index_sequence<0> : index_sequence<> {}; +template<> struct make_index_sequence<1> : index_sequence<0> {}; + +template<typename... Ts> +using index_sequence_for = make_index_sequence<sizeof...(Ts)>; + +// dispatch utility (taken from ranges-v3) +template<unsigned N> struct priority_tag : priority_tag < N - 1 > {}; +template<> struct priority_tag<0> {}; + +// taken from ranges-v3 +template<typename T> +struct static_const +{ + static constexpr T value{}; +}; + +template<typename T> +constexpr T static_const<T>::value; +} +} diff --git a/include/lib/modernjson/detail/meta/detected.hpp b/include/lib/modernjson/detail/meta/detected.hpp new file mode 100644 index 0000000..78b25d5 --- /dev/null +++ b/include/lib/modernjson/detail/meta/detected.hpp @@ -0,0 +1,56 @@ +#pragma once + +#include <type_traits> + +#include <lib/modernjson/detail/meta/void_t.hpp> + +// http://en.cppreference.com/w/cpp/experimental/is_detected +namespace nlohmann +{ +namespace detail +{ +struct nonesuch +{ + nonesuch() = delete; + ~nonesuch() = delete; + nonesuch(nonesuch const&) = delete; + void operator=(nonesuch const&) = delete; +}; + +template <class Default, + class AlwaysVoid, + template <class...> class Op, + class... Args> +struct detector +{ + using value_t = std::false_type; + using type = Default; +}; + +template <class Default, template <class...> class Op, class... Args> +struct detector<Default, void_t<Op<Args...>>, Op, Args...> +{ + using value_t = std::true_type; + using type = Op<Args...>; +}; + +template <template <class...> class Op, class... Args> +using is_detected = typename detector<nonesuch, void, Op, Args...>::value_t; + +template <template <class...> class Op, class... Args> +using detected_t = typename detector<nonesuch, void, Op, Args...>::type; + +template <class Default, template <class...> class Op, class... Args> +using detected_or = detector<Default, void, Op, Args...>; + +template <class Default, template <class...> class Op, class... Args> +using detected_or_t = typename detected_or<Default, Op, Args...>::type; + +template <class Expected, template <class...> class Op, class... Args> +using is_detected_exact = std::is_same<Expected, detected_t<Op, Args...>>; + +template <class To, template <class...> class Op, class... Args> +using is_detected_convertible = + std::is_convertible<detected_t<Op, Args...>, To>; +} +} diff --git a/include/lib/modernjson/detail/meta/is_sax.hpp b/include/lib/modernjson/detail/meta/is_sax.hpp new file mode 100644 index 0000000..63f9793 --- /dev/null +++ b/include/lib/modernjson/detail/meta/is_sax.hpp @@ -0,0 +1,141 @@ +#pragma once + +#include <cstdint> // size_t +#include <utility> // declval + +#include <lib/modernjson/detail/meta/detected.hpp> +#include <lib/modernjson/detail/meta/type_traits.hpp> + +namespace nlohmann +{ +namespace detail +{ +template <typename T> +using null_function_t = decltype(std::declval<T&>().null()); + +template <typename T> +using boolean_function_t = + decltype(std::declval<T&>().boolean(std::declval<bool>())); + +template <typename T, typename Integer> +using number_integer_function_t = + decltype(std::declval<T&>().number_integer(std::declval<Integer>())); + +template <typename T, typename Unsigned> +using number_unsigned_function_t = + decltype(std::declval<T &>().number_unsigned(std::declval<Unsigned>())); + +template <typename T, typename Float, typename String> +using number_float_function_t = decltype(std::declval<T &>().number_float( + std::declval<Float>(), std::declval<const String &>())); + +template <typename T, typename String> +using string_function_t = + decltype(std::declval<T &>().string(std::declval<String &>())); + +template <typename T> +using start_object_function_t = + decltype(std::declval<T &>().start_object(std::declval<std::size_t>())); + +template <typename T, typename String> +using key_function_t = + decltype(std::declval<T &>().key(std::declval<String &>())); + +template <typename T> +using end_object_function_t = decltype(std::declval<T &>().end_object()); + +template <typename T> +using start_array_function_t = + decltype(std::declval<T &>().start_array(std::declval<std::size_t>())); + +template <typename T> +using end_array_function_t = decltype(std::declval<T &>().end_array()); + +template <typename T, typename Exception> +using parse_error_function_t = decltype(std::declval<T &>().parse_error( + std::declval<std::size_t>(), std::declval<const std::string &>(), + std::declval<const Exception &>())); + +template <typename SAX, typename BasicJsonType> +struct is_sax +{ +private: + static_assert(is_basic_json<BasicJsonType>::value, + "BasicJsonType must be of type basic_json<...>"); + + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + using exception_t = typename BasicJsonType::exception; + +public: + static constexpr bool value = + is_detected_exact<bool, null_function_t, SAX>::value && + is_detected_exact<bool, boolean_function_t, SAX>::value && + is_detected_exact<bool, number_integer_function_t, SAX, + number_integer_t>::value && + is_detected_exact<bool, number_unsigned_function_t, SAX, + number_unsigned_t>::value && + is_detected_exact<bool, number_float_function_t, SAX, number_float_t, + string_t>::value && + is_detected_exact<bool, string_function_t, SAX, string_t>::value && + is_detected_exact<bool, start_object_function_t, SAX>::value && + is_detected_exact<bool, key_function_t, SAX, string_t>::value && + is_detected_exact<bool, end_object_function_t, SAX>::value && + is_detected_exact<bool, start_array_function_t, SAX>::value && + is_detected_exact<bool, end_array_function_t, SAX>::value && + is_detected_exact<bool, parse_error_function_t, SAX, exception_t>::value; +}; + +template <typename SAX, typename BasicJsonType> +struct is_sax_static_asserts +{ +private: + static_assert(is_basic_json<BasicJsonType>::value, + "BasicJsonType must be of type basic_json<...>"); + + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + using number_float_t = typename BasicJsonType::number_float_t; + using string_t = typename BasicJsonType::string_t; + using exception_t = typename BasicJsonType::exception; + +public: + static_assert(is_detected_exact<bool, null_function_t, SAX>::value, + "Missing/invalid function: bool null()"); + static_assert(is_detected_exact<bool, boolean_function_t, SAX>::value, + "Missing/invalid function: bool boolean(bool)"); + static_assert(is_detected_exact<bool, boolean_function_t, SAX>::value, + "Missing/invalid function: bool boolean(bool)"); + static_assert( + is_detected_exact<bool, number_integer_function_t, SAX, + number_integer_t>::value, + "Missing/invalid function: bool number_integer(number_integer_t)"); + static_assert( + is_detected_exact<bool, number_unsigned_function_t, SAX, + number_unsigned_t>::value, + "Missing/invalid function: bool number_unsigned(number_unsigned_t)"); + static_assert(is_detected_exact<bool, number_float_function_t, SAX, + number_float_t, string_t>::value, + "Missing/invalid function: bool number_float(number_float_t, const string_t&)"); + static_assert( + is_detected_exact<bool, string_function_t, SAX, string_t>::value, + "Missing/invalid function: bool string(string_t&)"); + static_assert(is_detected_exact<bool, start_object_function_t, SAX>::value, + "Missing/invalid function: bool start_object(std::size_t)"); + static_assert(is_detected_exact<bool, key_function_t, SAX, string_t>::value, + "Missing/invalid function: bool key(string_t&)"); + static_assert(is_detected_exact<bool, end_object_function_t, SAX>::value, + "Missing/invalid function: bool end_object()"); + static_assert(is_detected_exact<bool, start_array_function_t, SAX>::value, + "Missing/invalid function: bool start_array(std::size_t)"); + static_assert(is_detected_exact<bool, end_array_function_t, SAX>::value, + "Missing/invalid function: bool end_array()"); + static_assert( + is_detected_exact<bool, parse_error_function_t, SAX, exception_t>::value, + "Missing/invalid function: bool parse_error(std::size_t, const " + "std::string&, const exception&)"); +}; +} +} diff --git a/include/lib/modernjson/detail/meta/type_traits.hpp b/include/lib/modernjson/detail/meta/type_traits.hpp new file mode 100644 index 0000000..e5b6986 --- /dev/null +++ b/include/lib/modernjson/detail/meta/type_traits.hpp @@ -0,0 +1,237 @@ +#pragma once + +#include <ciso646> // not +#include <limits> // numeric_limits +#include <type_traits> // false_type, is_constructible, is_integral, is_same, true_type +#include <utility> // declval + +#include <lib/modernjson/json_fwd.h> +#include <lib/modernjson/detail/meta/cpp_future.hpp> +#include <lib/modernjson/detail/meta/detected.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> + +namespace nlohmann +{ +/*! +@brief detail namespace with internal helper functions + +This namespace collects functions that should not be exposed, +implementations of some @ref basic_json methods, and meta-programming helpers. + +@since version 2.1.0 +*/ +namespace detail +{ +///////////// +// helpers // +///////////// + +template<typename> struct is_basic_json : std::false_type {}; + +NLOHMANN_BASIC_JSON_TPL_DECLARATION +struct is_basic_json<NLOHMANN_BASIC_JSON_TPL> : std::true_type {}; + +////////////////////////// +// aliases for detected // +////////////////////////// + +template <typename T> +using mapped_type_t = typename T::mapped_type; + +template <typename T> +using key_type_t = typename T::key_type; + +template <typename T> +using value_type_t = typename T::value_type; + +template <typename T> +using difference_type_t = typename T::difference_type; + +template <typename T> +using pointer_t = typename T::pointer; + +template <typename T> +using reference_t = typename T::reference; + +template <typename T> +using iterator_category_t = typename T::iterator_category; + +template <typename T> +using iterator_t = typename T::iterator; + +template <typename T, typename... Args> +using to_json_function = decltype(T::to_json(std::declval<Args>()...)); + +template <typename T, typename... Args> +using from_json_function = decltype(T::from_json(std::declval<Args>()...)); + +/////////////////// +// is_ functions // +/////////////////// + +template <typename T, typename = void> +struct is_iterator_traits : std::false_type {}; + +template <typename T> +struct is_iterator_traits<std::iterator_traits<T>> +{ + private: + using traits = std::iterator_traits<T>; + + public: + static constexpr auto value = + is_detected<value_type_t, traits>::value && + is_detected<difference_type_t, traits>::value && + is_detected<pointer_t, traits>::value && + is_detected<iterator_category_t, traits>::value && + is_detected<reference_t, traits>::value; +}; + +// source: https://stackoverflow.com/a/37193089/4116453 + +template <typename T, typename = void> +struct is_complete_type : std::false_type {}; + +template <typename T> +struct is_complete_type<T, decltype(void(sizeof(T)))> : std::true_type {}; + +template <typename BasicJsonType, typename CompatibleObjectType, + typename = void> +struct is_compatible_object_type_impl : std::false_type {}; + +template <typename BasicJsonType, typename CompatibleObjectType> +struct is_compatible_object_type_impl < + BasicJsonType, CompatibleObjectType, + enable_if_t<is_detected<mapped_type_t, CompatibleObjectType>::value and + is_detected<key_type_t, CompatibleObjectType>::value >> +{ + + using object_t = typename BasicJsonType::object_t; + + // macOS's is_constructible does not play well with nonesuch... + static constexpr bool value = + std::is_constructible<typename object_t::key_type, + typename CompatibleObjectType::key_type>::value and + std::is_constructible<typename object_t::mapped_type, + typename CompatibleObjectType::mapped_type>::value; +}; + +template <typename BasicJsonType, typename CompatibleObjectType> +struct is_compatible_object_type + : is_compatible_object_type_impl<BasicJsonType, CompatibleObjectType> {}; + +template <typename BasicJsonType, typename CompatibleStringType, + typename = void> +struct is_compatible_string_type_impl : std::false_type {}; + +template <typename BasicJsonType, typename CompatibleStringType> +struct is_compatible_string_type_impl < + BasicJsonType, CompatibleStringType, + enable_if_t<is_detected_exact<typename BasicJsonType::string_t::value_type, + value_type_t, CompatibleStringType>::value >> +{ + static constexpr auto value = + std::is_constructible<typename BasicJsonType::string_t, CompatibleStringType>::value; +}; + +template <typename BasicJsonType, typename CompatibleStringType> +struct is_compatible_string_type + : is_compatible_string_type_impl<BasicJsonType, CompatibleStringType> {}; + +template <typename BasicJsonType, typename CompatibleArrayType, typename = void> +struct is_compatible_array_type_impl : std::false_type {}; + +template <typename BasicJsonType, typename CompatibleArrayType> +struct is_compatible_array_type_impl < + BasicJsonType, CompatibleArrayType, + enable_if_t<is_detected<value_type_t, CompatibleArrayType>::value and + is_detected<iterator_t, CompatibleArrayType>::value >> +{ + // This is needed because json_reverse_iterator has a ::iterator type... + // Therefore it is detected as a CompatibleArrayType. + // The real fix would be to have an Iterable concept. + static constexpr bool value = not is_iterator_traits<std::iterator_traits<CompatibleArrayType>>::value; +}; + +template <typename BasicJsonType, typename CompatibleArrayType> +struct is_compatible_array_type + : is_compatible_array_type_impl<BasicJsonType, CompatibleArrayType> {}; + +template <typename RealIntegerType, typename CompatibleNumberIntegerType, + typename = void> +struct is_compatible_integer_type_impl : std::false_type {}; + +template <typename RealIntegerType, typename CompatibleNumberIntegerType> +struct is_compatible_integer_type_impl < + RealIntegerType, CompatibleNumberIntegerType, + enable_if_t<std::is_integral<RealIntegerType>::value and + std::is_integral<CompatibleNumberIntegerType>::value and + not std::is_same<bool, CompatibleNumberIntegerType>::value >> +{ + // is there an assert somewhere on overflows? + using RealLimits = std::numeric_limits<RealIntegerType>; + using CompatibleLimits = std::numeric_limits<CompatibleNumberIntegerType>; + + static constexpr auto value = + std::is_constructible<RealIntegerType, + CompatibleNumberIntegerType>::value and + CompatibleLimits::is_integer and + RealLimits::is_signed == CompatibleLimits::is_signed; +}; + +template <typename RealIntegerType, typename CompatibleNumberIntegerType> +struct is_compatible_integer_type + : is_compatible_integer_type_impl<RealIntegerType, + CompatibleNumberIntegerType> {}; + +// trait checking if JSONSerializer<T>::from_json(json const&, udt&) exists +template<typename BasicJsonType, typename T> +struct has_from_json +{ + using serializer = typename BasicJsonType::template json_serializer<T, void>; + + static constexpr bool value = + is_detected_exact<void, from_json_function, serializer, + const BasicJsonType&, T&>::value; +}; + +// This trait checks if JSONSerializer<T>::from_json(json const&) exists +// this overload is used for non-default-constructible user-defined-types +template<typename BasicJsonType, typename T> +struct has_non_default_from_json +{ + using serializer = typename BasicJsonType::template json_serializer<T, void>; + + static constexpr bool value = + is_detected_exact<T, from_json_function, serializer, + const BasicJsonType&>::value; +}; + +// This trait checks if BasicJsonType::json_serializer<T>::to_json exists +template<typename BasicJsonType, typename T> +struct has_to_json +{ + using serializer = typename BasicJsonType::template json_serializer<T, void>; + + static constexpr bool value = + is_detected_exact<void, to_json_function, serializer, BasicJsonType&, + T>::value; +}; + +template <typename BasicJsonType, typename CompatibleType, typename = void> +struct is_compatible_type_impl: std::false_type {}; + +template <typename BasicJsonType, typename CompatibleType> +struct is_compatible_type_impl < + BasicJsonType, CompatibleType, + enable_if_t<is_complete_type<CompatibleType>::value >> +{ + static constexpr bool value = + has_to_json<BasicJsonType, CompatibleType>::value; +}; + +template <typename BasicJsonType, typename CompatibleType> +struct is_compatible_type + : is_compatible_type_impl<BasicJsonType, CompatibleType> {}; +} +} diff --git a/include/lib/modernjson/detail/meta/void_t.hpp b/include/lib/modernjson/detail/meta/void_t.hpp new file mode 100644 index 0000000..2528ea8 --- /dev/null +++ b/include/lib/modernjson/detail/meta/void_t.hpp @@ -0,0 +1,13 @@ +#pragma once + +namespace nlohmann +{ +namespace detail +{ +template <typename ...Ts> struct make_void +{ + using type = void; +}; +template <typename ...Ts> using void_t = typename make_void<Ts...>::type; +} +} diff --git a/include/lib/modernjson/detail/output/binary_writer.hpp b/include/lib/modernjson/detail/output/binary_writer.hpp new file mode 100644 index 0000000..1726e0c --- /dev/null +++ b/include/lib/modernjson/detail/output/binary_writer.hpp @@ -0,0 +1,949 @@ +#pragma once + +#include <algorithm> // reverse +#include <array> // array +#include <cstdint> // uint8_t, uint16_t, uint32_t, uint64_t +#include <cstring> // memcpy +#include <limits> // numeric_limits + +#include <lib/modernjson/detail/input/binary_reader.hpp> +#include <lib/modernjson/detail/output/output_adapters.hpp> + +namespace nlohmann +{ +namespace detail +{ +/////////////////// +// binary writer // +/////////////////// + +/*! +@brief serialization to CBOR and MessagePack values +*/ +template<typename BasicJsonType, typename CharType> +class binary_writer +{ + public: + /*! + @brief create a binary writer + + @param[in] adapter output adapter to write to + */ + explicit binary_writer(output_adapter_t<CharType> adapter) : oa(adapter) + { + assert(oa); + } + + /*! + @brief[in] j JSON value to serialize + */ + void write_cbor(const BasicJsonType& j) + { + switch (j.type()) + { + case value_t::null: + { + oa->write_character(static_cast<CharType>(0xF6)); + break; + } + + case value_t::boolean: + { + oa->write_character(j.m_value.boolean + ? static_cast<CharType>(0xF5) + : static_cast<CharType>(0xF4)); + break; + } + + case value_t::number_integer: + { + if (j.m_value.number_integer >= 0) + { + // CBOR does not differentiate between positive signed + // integers and unsigned integers. Therefore, we used the + // code from the value_t::number_unsigned case here. + if (j.m_value.number_integer <= 0x17) + { + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x18)); + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x19)); + write_number(static_cast<uint16_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x1A)); + write_number(static_cast<uint32_t>(j.m_value.number_integer)); + } + else + { + oa->write_character(static_cast<CharType>(0x1B)); + write_number(static_cast<uint64_t>(j.m_value.number_integer)); + } + } + else + { + // The conversions below encode the sign in the first + // byte, and the value is converted to a positive number. + const auto positive_number = -1 - j.m_value.number_integer; + if (j.m_value.number_integer >= -24) + { + write_number(static_cast<uint8_t>(0x20 + positive_number)); + } + else if (positive_number <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x38)); + write_number(static_cast<uint8_t>(positive_number)); + } + else if (positive_number <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x39)); + write_number(static_cast<uint16_t>(positive_number)); + } + else if (positive_number <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x3A)); + write_number(static_cast<uint32_t>(positive_number)); + } + else + { + oa->write_character(static_cast<CharType>(0x3B)); + write_number(static_cast<uint64_t>(positive_number)); + } + } + break; + } + + case value_t::number_unsigned: + { + if (j.m_value.number_unsigned <= 0x17) + { + write_number(static_cast<uint8_t>(j.m_value.number_unsigned)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x18)); + write_number(static_cast<uint8_t>(j.m_value.number_unsigned)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x19)); + write_number(static_cast<uint16_t>(j.m_value.number_unsigned)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x1A)); + write_number(static_cast<uint32_t>(j.m_value.number_unsigned)); + } + else + { + oa->write_character(static_cast<CharType>(0x1B)); + write_number(static_cast<uint64_t>(j.m_value.number_unsigned)); + } + break; + } + + case value_t::number_float: + { + oa->write_character(get_cbor_float_prefix(j.m_value.number_float)); + write_number(j.m_value.number_float); + break; + } + + case value_t::string: + { + // step 1: write control byte and the string length + const auto N = j.m_value.string->size(); + if (N <= 0x17) + { + write_number(static_cast<uint8_t>(0x60 + N)); + } + else if (N <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x78)); + write_number(static_cast<uint8_t>(N)); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x79)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x7A)); + write_number(static_cast<uint32_t>(N)); + } + // LCOV_EXCL_START + else if (N <= (std::numeric_limits<uint64_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x7B)); + write_number(static_cast<uint64_t>(N)); + } + // LCOV_EXCL_STOP + + // step 2: write the string + oa->write_characters( + reinterpret_cast<const CharType*>(j.m_value.string->c_str()), + j.m_value.string->size()); + break; + } + + case value_t::array: + { + // step 1: write control byte and the array size + const auto N = j.m_value.array->size(); + if (N <= 0x17) + { + write_number(static_cast<uint8_t>(0x80 + N)); + } + else if (N <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x98)); + write_number(static_cast<uint8_t>(N)); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x99)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x9A)); + write_number(static_cast<uint32_t>(N)); + } + // LCOV_EXCL_START + else if (N <= (std::numeric_limits<uint64_t>::max)()) + { + oa->write_character(static_cast<CharType>(0x9B)); + write_number(static_cast<uint64_t>(N)); + } + // LCOV_EXCL_STOP + + // step 2: write each element + for (const auto& el : *j.m_value.array) + { + write_cbor(el); + } + break; + } + + case value_t::object: + { + // step 1: write control byte and the object size + const auto N = j.m_value.object->size(); + if (N <= 0x17) + { + write_number(static_cast<uint8_t>(0xA0 + N)); + } + else if (N <= (std::numeric_limits<uint8_t>::max)()) + { + oa->write_character(static_cast<CharType>(0xB8)); + write_number(static_cast<uint8_t>(N)); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + oa->write_character(static_cast<CharType>(0xB9)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + oa->write_character(static_cast<CharType>(0xBA)); + write_number(static_cast<uint32_t>(N)); + } + // LCOV_EXCL_START + else if (N <= (std::numeric_limits<uint64_t>::max)()) + { + oa->write_character(static_cast<CharType>(0xBB)); + write_number(static_cast<uint64_t>(N)); + } + // LCOV_EXCL_STOP + + // step 2: write each element + for (const auto& el : *j.m_value.object) + { + write_cbor(el.first); + write_cbor(el.second); + } + break; + } + + default: + break; + } + } + + /*! + @brief[in] j JSON value to serialize + */ + void write_msgpack(const BasicJsonType& j) + { + switch (j.type()) + { + case value_t::null: // nil + { + oa->write_character(static_cast<CharType>(0xC0)); + break; + } + + case value_t::boolean: // true and false + { + oa->write_character(j.m_value.boolean + ? static_cast<CharType>(0xC3) + : static_cast<CharType>(0xC2)); + break; + } + + case value_t::number_integer: + { + if (j.m_value.number_integer >= 0) + { + // MessagePack does not differentiate between positive + // signed integers and unsigned integers. Therefore, we used + // the code from the value_t::number_unsigned case here. + if (j.m_value.number_unsigned < 128) + { + // positive fixnum + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint8_t>::max)()) + { + // uint 8 + oa->write_character(static_cast<CharType>(0xCC)); + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint16_t>::max)()) + { + // uint 16 + oa->write_character(static_cast<CharType>(0xCD)); + write_number(static_cast<uint16_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint32_t>::max)()) + { + // uint 32 + oa->write_character(static_cast<CharType>(0xCE)); + write_number(static_cast<uint32_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint64_t>::max)()) + { + // uint 64 + oa->write_character(static_cast<CharType>(0xCF)); + write_number(static_cast<uint64_t>(j.m_value.number_integer)); + } + } + else + { + if (j.m_value.number_integer >= -32) + { + // negative fixnum + write_number(static_cast<int8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer >= (std::numeric_limits<int8_t>::min)() and + j.m_value.number_integer <= (std::numeric_limits<int8_t>::max)()) + { + // int 8 + oa->write_character(static_cast<CharType>(0xD0)); + write_number(static_cast<int8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer >= (std::numeric_limits<int16_t>::min)() and + j.m_value.number_integer <= (std::numeric_limits<int16_t>::max)()) + { + // int 16 + oa->write_character(static_cast<CharType>(0xD1)); + write_number(static_cast<int16_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer >= (std::numeric_limits<int32_t>::min)() and + j.m_value.number_integer <= (std::numeric_limits<int32_t>::max)()) + { + // int 32 + oa->write_character(static_cast<CharType>(0xD2)); + write_number(static_cast<int32_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_integer >= (std::numeric_limits<int64_t>::min)() and + j.m_value.number_integer <= (std::numeric_limits<int64_t>::max)()) + { + // int 64 + oa->write_character(static_cast<CharType>(0xD3)); + write_number(static_cast<int64_t>(j.m_value.number_integer)); + } + } + break; + } + + case value_t::number_unsigned: + { + if (j.m_value.number_unsigned < 128) + { + // positive fixnum + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint8_t>::max)()) + { + // uint 8 + oa->write_character(static_cast<CharType>(0xCC)); + write_number(static_cast<uint8_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint16_t>::max)()) + { + // uint 16 + oa->write_character(static_cast<CharType>(0xCD)); + write_number(static_cast<uint16_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint32_t>::max)()) + { + // uint 32 + oa->write_character(static_cast<CharType>(0xCE)); + write_number(static_cast<uint32_t>(j.m_value.number_integer)); + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint64_t>::max)()) + { + // uint 64 + oa->write_character(static_cast<CharType>(0xCF)); + write_number(static_cast<uint64_t>(j.m_value.number_integer)); + } + break; + } + + case value_t::number_float: + { + oa->write_character(get_msgpack_float_prefix(j.m_value.number_float)); + write_number(j.m_value.number_float); + break; + } + + case value_t::string: + { + // step 1: write control byte and the string length + const auto N = j.m_value.string->size(); + if (N <= 31) + { + // fixstr + write_number(static_cast<uint8_t>(0xA0 | N)); + } + else if (N <= (std::numeric_limits<uint8_t>::max)()) + { + // str 8 + oa->write_character(static_cast<CharType>(0xD9)); + write_number(static_cast<uint8_t>(N)); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + // str 16 + oa->write_character(static_cast<CharType>(0xDA)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + // str 32 + oa->write_character(static_cast<CharType>(0xDB)); + write_number(static_cast<uint32_t>(N)); + } + + // step 2: write the string + oa->write_characters( + reinterpret_cast<const CharType*>(j.m_value.string->c_str()), + j.m_value.string->size()); + break; + } + + case value_t::array: + { + // step 1: write control byte and the array size + const auto N = j.m_value.array->size(); + if (N <= 15) + { + // fixarray + write_number(static_cast<uint8_t>(0x90 | N)); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + // array 16 + oa->write_character(static_cast<CharType>(0xDC)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + // array 32 + oa->write_character(static_cast<CharType>(0xDD)); + write_number(static_cast<uint32_t>(N)); + } + + // step 2: write each element + for (const auto& el : *j.m_value.array) + { + write_msgpack(el); + } + break; + } + + case value_t::object: + { + // step 1: write control byte and the object size + const auto N = j.m_value.object->size(); + if (N <= 15) + { + // fixmap + write_number(static_cast<uint8_t>(0x80 | (N & 0xF))); + } + else if (N <= (std::numeric_limits<uint16_t>::max)()) + { + // map 16 + oa->write_character(static_cast<CharType>(0xDE)); + write_number(static_cast<uint16_t>(N)); + } + else if (N <= (std::numeric_limits<uint32_t>::max)()) + { + // map 32 + oa->write_character(static_cast<CharType>(0xDF)); + write_number(static_cast<uint32_t>(N)); + } + + // step 2: write each element + for (const auto& el : *j.m_value.object) + { + write_msgpack(el.first); + write_msgpack(el.second); + } + break; + } + + default: + break; + } + } + + /*! + @param[in] j JSON value to serialize + @param[in] use_count whether to use '#' prefixes (optimized format) + @param[in] use_type whether to use '$' prefixes (optimized format) + @param[in] add_prefix whether prefixes need to be used for this value + */ + void write_ubjson(const BasicJsonType& j, const bool use_count, + const bool use_type, const bool add_prefix = true) + { + switch (j.type()) + { + case value_t::null: + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('Z')); + } + break; + } + + case value_t::boolean: + { + if (add_prefix) + oa->write_character(j.m_value.boolean + ? static_cast<CharType>('T') + : static_cast<CharType>('F')); + break; + } + + case value_t::number_integer: + { + write_number_with_ubjson_prefix(j.m_value.number_integer, add_prefix); + break; + } + + case value_t::number_unsigned: + { + write_number_with_ubjson_prefix(j.m_value.number_unsigned, add_prefix); + break; + } + + case value_t::number_float: + { + write_number_with_ubjson_prefix(j.m_value.number_float, add_prefix); + break; + } + + case value_t::string: + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('S')); + } + write_number_with_ubjson_prefix(j.m_value.string->size(), true); + oa->write_characters( + reinterpret_cast<const CharType*>(j.m_value.string->c_str()), + j.m_value.string->size()); + break; + } + + case value_t::array: + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('[')); + } + + bool prefix_required = true; + if (use_type and not j.m_value.array->empty()) + { + assert(use_count); + const CharType first_prefix = ubjson_prefix(j.front()); + const bool same_prefix = std::all_of(j.begin() + 1, j.end(), + [this, first_prefix](const BasicJsonType & v) + { + return ubjson_prefix(v) == first_prefix; + }); + + if (same_prefix) + { + prefix_required = false; + oa->write_character(static_cast<CharType>('$')); + oa->write_character(first_prefix); + } + } + + if (use_count) + { + oa->write_character(static_cast<CharType>('#')); + write_number_with_ubjson_prefix(j.m_value.array->size(), true); + } + + for (const auto& el : *j.m_value.array) + { + write_ubjson(el, use_count, use_type, prefix_required); + } + + if (not use_count) + { + oa->write_character(static_cast<CharType>(']')); + } + + break; + } + + case value_t::object: + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('{')); + } + + bool prefix_required = true; + if (use_type and not j.m_value.object->empty()) + { + assert(use_count); + const CharType first_prefix = ubjson_prefix(j.front()); + const bool same_prefix = std::all_of(j.begin(), j.end(), + [this, first_prefix](const BasicJsonType & v) + { + return ubjson_prefix(v) == first_prefix; + }); + + if (same_prefix) + { + prefix_required = false; + oa->write_character(static_cast<CharType>('$')); + oa->write_character(first_prefix); + } + } + + if (use_count) + { + oa->write_character(static_cast<CharType>('#')); + write_number_with_ubjson_prefix(j.m_value.object->size(), true); + } + + for (const auto& el : *j.m_value.object) + { + write_number_with_ubjson_prefix(el.first.size(), true); + oa->write_characters( + reinterpret_cast<const CharType*>(el.first.c_str()), + el.first.size()); + write_ubjson(el.second, use_count, use_type, prefix_required); + } + + if (not use_count) + { + oa->write_character(static_cast<CharType>('}')); + } + + break; + } + + default: + break; + } + } + + private: + /* + @brief write a number to output input + + @param[in] n number of type @a NumberType + @tparam NumberType the type of the number + + @note This function needs to respect the system's endianess, because bytes + in CBOR, MessagePack, and UBJSON are stored in network order (big + endian) and therefore need reordering on little endian systems. + */ + template<typename NumberType> + void write_number(const NumberType n) + { + // step 1: write number to array of length NumberType + std::array<CharType, sizeof(NumberType)> vec; + std::memcpy(vec.data(), &n, sizeof(NumberType)); + + // step 2: write array to output (with possible reordering) + if (is_little_endian) + { + // reverse byte order prior to conversion if necessary + std::reverse(vec.begin(), vec.end()); + } + + oa->write_characters(vec.data(), sizeof(NumberType)); + } + + // UBJSON: write number (floating point) + template<typename NumberType, typename std::enable_if< + std::is_floating_point<NumberType>::value, int>::type = 0> + void write_number_with_ubjson_prefix(const NumberType n, + const bool add_prefix) + { + if (add_prefix) + { + oa->write_character(get_ubjson_float_prefix(n)); + } + write_number(n); + } + + // UBJSON: write number (unsigned integer) + template<typename NumberType, typename std::enable_if< + std::is_unsigned<NumberType>::value, int>::type = 0> + void write_number_with_ubjson_prefix(const NumberType n, + const bool add_prefix) + { + if (n <= static_cast<uint64_t>((std::numeric_limits<int8_t>::max)())) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('i')); // int8 + } + write_number(static_cast<uint8_t>(n)); + } + else if (n <= (std::numeric_limits<uint8_t>::max)()) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('U')); // uint8 + } + write_number(static_cast<uint8_t>(n)); + } + else if (n <= static_cast<uint64_t>((std::numeric_limits<int16_t>::max)())) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('I')); // int16 + } + write_number(static_cast<int16_t>(n)); + } + else if (n <= static_cast<uint64_t>((std::numeric_limits<int32_t>::max)())) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('l')); // int32 + } + write_number(static_cast<int32_t>(n)); + } + else if (n <= static_cast<uint64_t>((std::numeric_limits<int64_t>::max)())) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('L')); // int64 + } + write_number(static_cast<int64_t>(n)); + } + else + { + JSON_THROW(out_of_range::create(407, "number overflow serializing " + std::to_string(n))); + } + } + + // UBJSON: write number (signed integer) + template<typename NumberType, typename std::enable_if< + std::is_signed<NumberType>::value and + not std::is_floating_point<NumberType>::value, int>::type = 0> + void write_number_with_ubjson_prefix(const NumberType n, + const bool add_prefix) + { + if ((std::numeric_limits<int8_t>::min)() <= n and n <= (std::numeric_limits<int8_t>::max)()) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('i')); // int8 + } + write_number(static_cast<int8_t>(n)); + } + else if (static_cast<int64_t>((std::numeric_limits<uint8_t>::min)()) <= n and n <= static_cast<int64_t>((std::numeric_limits<uint8_t>::max)())) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('U')); // uint8 + } + write_number(static_cast<uint8_t>(n)); + } + else if ((std::numeric_limits<int16_t>::min)() <= n and n <= (std::numeric_limits<int16_t>::max)()) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('I')); // int16 + } + write_number(static_cast<int16_t>(n)); + } + else if ((std::numeric_limits<int32_t>::min)() <= n and n <= (std::numeric_limits<int32_t>::max)()) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('l')); // int32 + } + write_number(static_cast<int32_t>(n)); + } + else if ((std::numeric_limits<int64_t>::min)() <= n and n <= (std::numeric_limits<int64_t>::max)()) + { + if (add_prefix) + { + oa->write_character(static_cast<CharType>('L')); // int64 + } + write_number(static_cast<int64_t>(n)); + } + // LCOV_EXCL_START + else + { + JSON_THROW(out_of_range::create(407, "number overflow serializing " + std::to_string(n))); + } + // LCOV_EXCL_STOP + } + + /*! + @brief determine the type prefix of container values + + @note This function does not need to be 100% accurate when it comes to + integer limits. In case a number exceeds the limits of int64_t, + this will be detected by a later call to function + write_number_with_ubjson_prefix. Therefore, we return 'L' for any + value that does not fit the previous limits. + */ + CharType ubjson_prefix(const BasicJsonType& j) const noexcept + { + switch (j.type()) + { + case value_t::null: + return 'Z'; + + case value_t::boolean: + return j.m_value.boolean ? 'T' : 'F'; + + case value_t::number_integer: + { + if ((std::numeric_limits<int8_t>::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits<int8_t>::max)()) + { + return 'i'; + } + else if ((std::numeric_limits<uint8_t>::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits<uint8_t>::max)()) + { + return 'U'; + } + else if ((std::numeric_limits<int16_t>::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits<int16_t>::max)()) + { + return 'I'; + } + else if ((std::numeric_limits<int32_t>::min)() <= j.m_value.number_integer and j.m_value.number_integer <= (std::numeric_limits<int32_t>::max)()) + { + return 'l'; + } + else // no check and assume int64_t (see note above) + { + return 'L'; + } + } + + case value_t::number_unsigned: + { + if (j.m_value.number_unsigned <= (std::numeric_limits<int8_t>::max)()) + { + return 'i'; + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<uint8_t>::max)()) + { + return 'U'; + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<int16_t>::max)()) + { + return 'I'; + } + else if (j.m_value.number_unsigned <= (std::numeric_limits<int32_t>::max)()) + { + return 'l'; + } + else // no check and assume int64_t (see note above) + { + return 'L'; + } + } + + case value_t::number_float: + return get_ubjson_float_prefix(j.m_value.number_float); + + case value_t::string: + return 'S'; + + case value_t::array: + return '['; + + case value_t::object: + return '{'; + + default: // discarded values + return 'N'; + } + } + + static constexpr CharType get_cbor_float_prefix(float) + { + return static_cast<CharType>(0xFA); // Single-Precision Float + } + + static constexpr CharType get_cbor_float_prefix(double) + { + return static_cast<CharType>(0xFB); // Double-Precision Float + } + + static constexpr CharType get_msgpack_float_prefix(float) + { + return static_cast<CharType>(0xCA); // float 32 + } + + static constexpr CharType get_msgpack_float_prefix(double) + { + return static_cast<CharType>(0xCB); // float 64 + } + + static constexpr CharType get_ubjson_float_prefix(float) + { + return 'd'; // float 32 + } + + static constexpr CharType get_ubjson_float_prefix(double) + { + return 'D'; // float 64 + } + + private: + /// whether we can assume little endianess + const bool is_little_endian = binary_reader<BasicJsonType>::little_endianess(); + + /// the output + output_adapter_t<CharType> oa = nullptr; +}; +} +} diff --git a/include/lib/modernjson/detail/output/output_adapters.hpp b/include/lib/modernjson/detail/output/output_adapters.hpp new file mode 100644 index 0000000..ff86a6e --- /dev/null +++ b/include/lib/modernjson/detail/output/output_adapters.hpp @@ -0,0 +1,113 @@ +#pragma once + +#include <algorithm> // copy +#include <cstddef> // size_t +#include <ios> // streamsize +#include <iterator> // back_inserter +#include <memory> // shared_ptr, make_shared +#include <ostream> // basic_ostream +#include <string> // basic_string +#include <vector> // vector + +namespace nlohmann +{ +namespace detail +{ +/// abstract output adapter interface +template<typename CharType> struct output_adapter_protocol +{ + virtual void write_character(CharType c) = 0; + virtual void write_characters(const CharType* s, std::size_t length) = 0; + virtual ~output_adapter_protocol() = default; +}; + +/// a type to simplify interfaces +template<typename CharType> +using output_adapter_t = std::shared_ptr<output_adapter_protocol<CharType>>; + +/// output adapter for byte vectors +template<typename CharType> +class output_vector_adapter : public output_adapter_protocol<CharType> +{ + public: + explicit output_vector_adapter(std::vector<CharType>& vec) : v(vec) {} + + void write_character(CharType c) override + { + v.push_back(c); + } + + void write_characters(const CharType* s, std::size_t length) override + { + std::copy(s, s + length, std::back_inserter(v)); + } + + private: + std::vector<CharType>& v; +}; + +/// output adapter for output streams +template<typename CharType> +class output_stream_adapter : public output_adapter_protocol<CharType> +{ + public: + explicit output_stream_adapter(std::basic_ostream<CharType>& s) : stream(s) {} + + void write_character(CharType c) override + { + stream.put(c); + } + + void write_characters(const CharType* s, std::size_t length) override + { + stream.write(s, static_cast<std::streamsize>(length)); + } + + private: + std::basic_ostream<CharType>& stream; +}; + +/// output adapter for basic_string +template<typename CharType, typename StringType = std::basic_string<CharType>> +class output_string_adapter : public output_adapter_protocol<CharType> +{ + public: + explicit output_string_adapter(StringType& s) : str(s) {} + + void write_character(CharType c) override + { + str.push_back(c); + } + + void write_characters(const CharType* s, std::size_t length) override + { + str.append(s, length); + } + + private: + StringType& str; +}; + +template<typename CharType, typename StringType = std::basic_string<CharType>> +class output_adapter +{ + public: + output_adapter(std::vector<CharType>& vec) + : oa(std::make_shared<output_vector_adapter<CharType>>(vec)) {} + + output_adapter(std::basic_ostream<CharType>& s) + : oa(std::make_shared<output_stream_adapter<CharType>>(s)) {} + + output_adapter(StringType& s) + : oa(std::make_shared<output_string_adapter<CharType, StringType>>(s)) {} + + operator output_adapter_t<CharType>() + { + return oa; + } + + private: + output_adapter_t<CharType> oa = nullptr; +}; +} +} diff --git a/include/lib/modernjson/detail/output/serializer.hpp b/include/lib/modernjson/detail/output/serializer.hpp new file mode 100644 index 0000000..7c6abb8 --- /dev/null +++ b/include/lib/modernjson/detail/output/serializer.hpp @@ -0,0 +1,631 @@ +#pragma once + +#include <algorithm> // reverse, remove, fill, find, none_of +#include <array> // array +#include <cassert> // assert +#include <ciso646> // and, or +#include <clocale> // localeconv, lconv +#include <cmath> // labs, isfinite, isnan, signbit +#include <cstddef> // size_t, ptrdiff_t +#include <cstdint> // uint8_t +#include <cstdio> // snprintf +#include <limits> // numeric_limits +#include <string> // string +#include <type_traits> // is_same + +#include <lib/modernjson/detail/exceptions.hpp> +#include <lib/modernjson/detail/conversions/to_chars.hpp> +#include <lib/modernjson/detail/macro_scope.hpp> +#include <lib/modernjson/detail/meta/cpp_future.hpp> +#include <lib/modernjson/detail/output/output_adapters.hpp> +#include <lib/modernjson/detail/value_t.hpp> + +namespace nlohmann +{ +namespace detail +{ +/////////////////// +// serialization // +/////////////////// + +template<typename BasicJsonType> +class serializer +{ + using string_t = typename BasicJsonType::string_t; + using number_float_t = typename BasicJsonType::number_float_t; + using number_integer_t = typename BasicJsonType::number_integer_t; + using number_unsigned_t = typename BasicJsonType::number_unsigned_t; + static constexpr uint8_t UTF8_ACCEPT = 0; + static constexpr uint8_t UTF8_REJECT = 1; + + public: + /*! + @param[in] s output stream to serialize to + @param[in] ichar indentation character to use + */ + serializer(output_adapter_t<char> s, const char ichar) + : o(std::move(s)), loc(std::localeconv()), + thousands_sep(loc->thousands_sep == nullptr ? '\0' : * (loc->thousands_sep)), + decimal_point(loc->decimal_point == nullptr ? '\0' : * (loc->decimal_point)), + indent_char(ichar), indent_string(512, indent_char) + {} + + // delete because of pointer members + serializer(const serializer&) = delete; + serializer& operator=(const serializer&) = delete; + + /*! + @brief internal implementation of the serialization function + + This function is called by the public member function dump and organizes + the serialization internally. The indentation level is propagated as + additional parameter. In case of arrays and objects, the function is + called recursively. + + - strings and object keys are escaped using `escape_string()` + - integer numbers are converted implicitly via `operator<<` + - floating-point numbers are converted to a string using `"%g"` format + + @param[in] val value to serialize + @param[in] pretty_print whether the output shall be pretty-printed + @param[in] indent_step the indent level + @param[in] current_indent the current indent level (only used internally) + */ + void dump(const BasicJsonType& val, const bool pretty_print, + const bool ensure_ascii, + const unsigned int indent_step, + const unsigned int current_indent = 0) + { + switch (val.m_type) + { + case value_t::object: + { + if (val.m_value.object->empty()) + { + o->write_characters("{}", 2); + return; + } + + if (pretty_print) + { + o->write_characters("{\n", 2); + + // variable to hold indentation for recursive calls + const auto new_indent = current_indent + indent_step; + if (JSON_UNLIKELY(indent_string.size() < new_indent)) + { + indent_string.resize(indent_string.size() * 2, ' '); + } + + // first n-1 elements + auto i = val.m_value.object->cbegin(); + for (std::size_t cnt = 0; cnt < val.m_value.object->size() - 1; ++cnt, ++i) + { + o->write_characters(indent_string.c_str(), new_indent); + o->write_character('\"'); + dump_escaped(i->first, ensure_ascii); + o->write_characters("\": ", 3); + dump(i->second, true, ensure_ascii, indent_step, new_indent); + o->write_characters(",\n", 2); + } + + // last element + assert(i != val.m_value.object->cend()); + assert(std::next(i) == val.m_value.object->cend()); + o->write_characters(indent_string.c_str(), new_indent); + o->write_character('\"'); + dump_escaped(i->first, ensure_ascii); + o->write_characters("\": ", 3); + dump(i->second, true, ensure_ascii, indent_step, new_indent); + + o->write_character('\n'); + o->write_characters(indent_string.c_str(), current_indent); + o->write_character('}'); + } + else + { + o->write_character('{'); + + // first n-1 elements + auto i = val.m_value.object->cbegin(); + for (std::size_t cnt = 0; cnt < val.m_value.object->size() - 1; ++cnt, ++i) + { + o->write_character('\"'); + dump_escaped(i->first, ensure_ascii); + o->write_characters("\":", 2); + dump(i->second, false, ensure_ascii, indent_step, current_indent); + o->write_character(','); + } + + // last element + assert(i != val.m_value.object->cend()); + assert(std::next(i) == val.m_value.object->cend()); + o->write_character('\"'); + dump_escaped(i->first, ensure_ascii); + o->write_characters("\":", 2); + dump(i->second, false, ensure_ascii, indent_step, current_indent); + + o->write_character('}'); + } + + return; + } + + case value_t::array: + { + if (val.m_value.array->empty()) + { + o->write_characters("[]", 2); + return; + } + + if (pretty_print) + { + o->write_characters("[\n", 2); + + // variable to hold indentation for recursive calls + const auto new_indent = current_indent + indent_step; + if (JSON_UNLIKELY(indent_string.size() < new_indent)) + { + indent_string.resize(indent_string.size() * 2, ' '); + } + + // first n-1 elements + for (auto i = val.m_value.array->cbegin(); + i != val.m_value.array->cend() - 1; ++i) + { + o->write_characters(indent_string.c_str(), new_indent); + dump(*i, true, ensure_ascii, indent_step, new_indent); + o->write_characters(",\n", 2); + } + + // last element + assert(not val.m_value.array->empty()); + o->write_characters(indent_string.c_str(), new_indent); + dump(val.m_value.array->back(), true, ensure_ascii, indent_step, new_indent); + + o->write_character('\n'); + o->write_characters(indent_string.c_str(), current_indent); + o->write_character(']'); + } + else + { + o->write_character('['); + + // first n-1 elements + for (auto i = val.m_value.array->cbegin(); + i != val.m_value.array->cend() - 1; ++i) + { + dump(*i, false, ensure_ascii, indent_step, current_indent); + o->write_character(','); + } + + // last element + assert(not val.m_value.array->empty()); + dump(val.m_value.array->back(), false, ensure_ascii, indent_step, current_indent); + + o->write_character(']'); + } + + return; + } + + case value_t::string: + { + o->write_character('\"'); + dump_escaped(*val.m_value.string, ensure_ascii); + o->write_character('\"'); + return; + } + + case value_t::boolean: + { + if (val.m_value.boolean) + { + o->write_characters("true", 4); + } + else + { + o->write_characters("false", 5); + } + return; + } + + case value_t::number_integer: + { + dump_integer(val.m_value.number_integer); + return; + } + + case value_t::number_unsigned: + { + dump_integer(val.m_value.number_unsigned); + return; + } + + case value_t::number_float: + { + dump_float(val.m_value.number_float); + return; + } + + case value_t::discarded: + { + o->write_characters("<discarded>", 11); + return; + } + + case value_t::null: + { + o->write_characters("null", 4); + return; + } + } + } + + private: + /*! + @brief dump escaped string + + Escape a string by replacing certain special characters by a sequence of an + escape character (backslash) and another character and other control + characters by a sequence of "\u" followed by a four-digit hex + representation. The escaped string is written to output stream @a o. + + @param[in] s the string to escape + @param[in] ensure_ascii whether to escape non-ASCII characters with + \uXXXX sequences + + @complexity Linear in the length of string @a s. + */ + void dump_escaped(const string_t& s, const bool ensure_ascii) + { + uint32_t codepoint; + uint8_t state = UTF8_ACCEPT; + std::size_t bytes = 0; // number of bytes written to string_buffer + + for (std::size_t i = 0; i < s.size(); ++i) + { + const auto byte = static_cast<uint8_t>(s[i]); + + switch (decode(state, codepoint, byte)) + { + case UTF8_ACCEPT: // decode found a new code point + { + switch (codepoint) + { + case 0x08: // backspace + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = 'b'; + break; + } + + case 0x09: // horizontal tab + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = 't'; + break; + } + + case 0x0A: // newline + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = 'n'; + break; + } + + case 0x0C: // formfeed + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = 'f'; + break; + } + + case 0x0D: // carriage return + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = 'r'; + break; + } + + case 0x22: // quotation mark + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = '\"'; + break; + } + + case 0x5C: // reverse solidus + { + string_buffer[bytes++] = '\\'; + string_buffer[bytes++] = '\\'; + break; + } + + default: + { + // escape control characters (0x00..0x1F) or, if + // ensure_ascii parameter is used, non-ASCII characters + if ((codepoint <= 0x1F) or (ensure_ascii and (codepoint >= 0x7F))) + { + if (codepoint <= 0xFFFF) + { + std::snprintf(string_buffer.data() + bytes, 7, "\\u%04x", + static_cast<uint16_t>(codepoint)); + bytes += 6; + } + else + { + std::snprintf(string_buffer.data() + bytes, 13, "\\u%04x\\u%04x", + static_cast<uint16_t>(0xD7C0 + (codepoint >> 10)), + static_cast<uint16_t>(0xDC00 + (codepoint & 0x3FF))); + bytes += 12; + } + } + else + { + // copy byte to buffer (all previous bytes + // been copied have in default case above) + string_buffer[bytes++] = s[i]; + } + break; + } + } + + // write buffer and reset index; there must be 13 bytes + // left, as this is the maximal number of bytes to be + // written ("\uxxxx\uxxxx\0") for one code point + if (string_buffer.size() - bytes < 13) + { + o->write_characters(string_buffer.data(), bytes); + bytes = 0; + } + break; + } + + case UTF8_REJECT: // decode found invalid UTF-8 byte + { + std::string sn(3, '\0'); + snprintf(&sn[0], sn.size(), "%.2X", byte); + JSON_THROW(type_error::create(316, "invalid UTF-8 byte at index " + std::to_string(i) + ": 0x" + sn)); + } + + default: // decode found yet incomplete multi-byte code point + { + if (not ensure_ascii) + { + // code point will not be escaped - copy byte to buffer + string_buffer[bytes++] = s[i]; + } + break; + } + } + } + + if (JSON_LIKELY(state == UTF8_ACCEPT)) + { + // write buffer + if (bytes > 0) + { + o->write_characters(string_buffer.data(), bytes); + } + } + else + { + // we finish reading, but do not accept: string was incomplete + std::string sn(3, '\0'); + snprintf(&sn[0], sn.size(), "%.2X", static_cast<uint8_t>(s.back())); + JSON_THROW(type_error::create(316, "incomplete UTF-8 string; last byte: 0x" + sn)); + } + } + + /*! + @brief dump an integer + + Dump a given integer to output stream @a o. Works internally with + @a number_buffer. + + @param[in] x integer number (signed or unsigned) to dump + @tparam NumberType either @a number_integer_t or @a number_unsigned_t + */ + template<typename NumberType, detail::enable_if_t< + std::is_same<NumberType, number_unsigned_t>::value or + std::is_same<NumberType, number_integer_t>::value, + int> = 0> + void dump_integer(NumberType x) + { + // special case for "0" + if (x == 0) + { + o->write_character('0'); + return; + } + + const bool is_negative = (x <= 0) and (x != 0); // see issue #755 + std::size_t i = 0; + + while (x != 0) + { + // spare 1 byte for '\0' + assert(i < number_buffer.size() - 1); + + const auto digit = std::labs(static_cast<long>(x % 10)); + number_buffer[i++] = static_cast<char>('0' + digit); + x /= 10; + } + + if (is_negative) + { + // make sure there is capacity for the '-' + assert(i < number_buffer.size() - 2); + number_buffer[i++] = '-'; + } + + std::reverse(number_buffer.begin(), number_buffer.begin() + i); + o->write_characters(number_buffer.data(), i); + } + + /*! + @brief dump a floating-point number + + Dump a given floating-point number to output stream @a o. Works internally + with @a number_buffer. + + @param[in] x floating-point number to dump + */ + void dump_float(number_float_t x) + { + // NaN / inf + if (not std::isfinite(x)) + { + o->write_characters("null", 4); + return; + } + + // If number_float_t is an IEEE-754 single or double precision number, + // use the Grisu2 algorithm to produce short numbers which are + // guaranteed to round-trip, using strtof and strtod, resp. + // + // NB: The test below works if <long double> == <double>. + static constexpr bool is_ieee_single_or_double + = (std::numeric_limits<number_float_t>::is_iec559 and std::numeric_limits<number_float_t>::digits == 24 and std::numeric_limits<number_float_t>::max_exponent == 128) or + (std::numeric_limits<number_float_t>::is_iec559 and std::numeric_limits<number_float_t>::digits == 53 and std::numeric_limits<number_float_t>::max_exponent == 1024); + + dump_float(x, std::integral_constant<bool, is_ieee_single_or_double>()); + } + + void dump_float(number_float_t x, std::true_type /*is_ieee_single_or_double*/) + { + char* begin = number_buffer.data(); + char* end = ::nlohmann::detail::to_chars(begin, begin + number_buffer.size(), x); + + o->write_characters(begin, static_cast<size_t>(end - begin)); + } + + void dump_float(number_float_t x, std::false_type /*is_ieee_single_or_double*/) + { + // get number of digits for a float -> text -> float round-trip + static constexpr auto d = std::numeric_limits<number_float_t>::max_digits10; + + // the actual conversion + std::ptrdiff_t len = snprintf(number_buffer.data(), number_buffer.size(), "%.*g", d, x); + + // negative value indicates an error + assert(len > 0); + // check if buffer was large enough + assert(static_cast<std::size_t>(len) < number_buffer.size()); + + // erase thousands separator + if (thousands_sep != '\0') + { + const auto end = std::remove(number_buffer.begin(), + number_buffer.begin() + len, thousands_sep); + std::fill(end, number_buffer.end(), '\0'); + assert((end - number_buffer.begin()) <= len); + len = (end - number_buffer.begin()); + } + + // convert decimal point to '.' + if (decimal_point != '\0' and decimal_point != '.') + { + const auto dec_pos = std::find(number_buffer.begin(), number_buffer.end(), decimal_point); + if (dec_pos != number_buffer.end()) + { + *dec_pos = '.'; + } + } + + o->write_characters(number_buffer.data(), static_cast<std::size_t>(len)); + + // determine if need to append ".0" + const bool value_is_int_like = + std::none_of(number_buffer.begin(), number_buffer.begin() + len + 1, + [](char c) + { + return (c == '.' or c == 'e'); + }); + + if (value_is_int_like) + { + o->write_characters(".0", 2); + } + } + + /*! + @brief check whether a string is UTF-8 encoded + + The function checks each byte of a string whether it is UTF-8 encoded. The + result of the check is stored in the @a state parameter. The function must + be called initially with state 0 (accept). State 1 means the string must + be rejected, because the current byte is not allowed. If the string is + completely processed, but the state is non-zero, the string ended + prematurely; that is, the last byte indicated more bytes should have + followed. + + @param[in,out] state the state of the decoding + @param[in,out] codep codepoint (valid only if resulting state is UTF8_ACCEPT) + @param[in] byte next byte to decode + @return new state + + @note The function has been edited: a std::array is used. + + @copyright Copyright (c) 2008-2009 Bjoern Hoehrmann <bjoern@hoehrmann.de> + @sa http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ + */ + static uint8_t decode(uint8_t& state, uint32_t& codep, const uint8_t byte) noexcept + { + static const std::array<uint8_t, 400> utf8d = + { + { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 00..1F + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 20..3F + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 40..5F + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 60..7F + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, 9, // 80..9F + 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, // A0..BF + 8, 8, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, // C0..DF + 0xA, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x3, 0x4, 0x3, 0x3, // E0..EF + 0xB, 0x6, 0x6, 0x6, 0x5, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, 0x8, // F0..FF + 0x0, 0x1, 0x2, 0x3, 0x5, 0x8, 0x7, 0x1, 0x1, 0x1, 0x4, 0x6, 0x1, 0x1, 0x1, 0x1, // s0..s0 + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, // s1..s2 + 1, 2, 1, 1, 1, 1, 1, 2, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, // s3..s4 + 1, 2, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 1, 3, 1, 1, 1, 1, 1, 1, // s5..s6 + 1, 3, 1, 1, 1, 1, 1, 3, 1, 3, 1, 1, 1, 1, 1, 1, 1, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 // s7..s8 + } + }; + + const uint8_t type = utf8d[byte]; + + codep = (state != UTF8_ACCEPT) + ? (byte & 0x3fu) | (codep << 6) + : static_cast<uint32_t>(0xff >> type) & (byte); + + state = utf8d[256u + state * 16u + type]; + return state; + } + + private: + /// the output of the serializer + output_adapter_t<char> o = nullptr; + + /// a (hopefully) large enough character buffer + std::array<char, 64> number_buffer{{}}; + + /// the locale + const std::lconv* loc = nullptr; + /// the locale's thousand separator character + const char thousands_sep = '\0'; + /// the locale's decimal point character + const char decimal_point = '\0'; + + /// string buffer + std::array<char, 512> string_buffer{{}}; + + /// the indentation character + const char indent_char; + /// the indentation string + string_t indent_string; +}; +} +} diff --git a/include/lib/modernjson/detail/value_t.hpp b/include/lib/modernjson/detail/value_t.hpp new file mode 100644 index 0000000..326367c --- /dev/null +++ b/include/lib/modernjson/detail/value_t.hpp @@ -0,0 +1,76 @@ +#pragma once + +#include <array> // array +#include <ciso646> // and +#include <cstddef> // size_t +#include <cstdint> // uint8_t + +namespace nlohmann +{ +namespace detail +{ +/////////////////////////// +// JSON type enumeration // +/////////////////////////// + +/*! +@brief the JSON type enumeration + +This enumeration collects the different JSON types. It is internally used to +distinguish the stored values, and the functions @ref basic_json::is_null(), +@ref basic_json::is_object(), @ref basic_json::is_array(), +@ref basic_json::is_string(), @ref basic_json::is_boolean(), +@ref basic_json::is_number() (with @ref basic_json::is_number_integer(), +@ref basic_json::is_number_unsigned(), and @ref basic_json::is_number_float()), +@ref basic_json::is_discarded(), @ref basic_json::is_primitive(), and +@ref basic_json::is_structured() rely on it. + +@note There are three enumeration entries (number_integer, number_unsigned, and +number_float), because the library distinguishes these three types for numbers: +@ref basic_json::number_unsigned_t is used for unsigned integers, +@ref basic_json::number_integer_t is used for signed integers, and +@ref basic_json::number_float_t is used for floating-point numbers or to +approximate integers which do not fit in the limits of their respective type. + +@sa @ref basic_json::basic_json(const value_t value_type) -- create a JSON +value with the default value for a given type + +@since version 1.0.0 +*/ +enum class value_t : std::uint8_t +{ + null, ///< null value + object, ///< object (unordered set of name/value pairs) + array, ///< array (ordered collection of values) + string, ///< string value + boolean, ///< boolean value + number_integer, ///< number value (signed integer) + number_unsigned, ///< number value (unsigned integer) + number_float, ///< number value (floating-point) + discarded ///< discarded by the the parser callback function +}; + +/*! +@brief comparison operator for JSON types + +Returns an ordering that is similar to Python: +- order: null < boolean < number < object < array < string +- furthermore, each type is not smaller than itself +- discarded values are not comparable + +@since version 1.0.0 +*/ +inline bool operator<(const value_t lhs, const value_t rhs) noexcept +{ + static constexpr std::array<std::uint8_t, 8> order = {{ + 0 /* null */, 3 /* object */, 4 /* array */, 5 /* string */, + 1 /* boolean */, 2 /* integer */, 2 /* unsigned */, 2 /* float */ + } + }; + + const auto l_index = static_cast<std::size_t>(lhs); + const auto r_index = static_cast<std::size_t>(rhs); + return l_index < order.size() and r_index < order.size() and order[l_index] < order[r_index]; +} +} +} |