/* __ _____ _____ _____ __| | __| | | | JSON for Modern C++ | | |__ | | | | | | version 3.2.0 |_____|_____|_____|_|___| https://github.com/nlohmann/json Licensed under the MIT License . SPDX-License-Identifier: MIT Copyright (c) 2013-2018 Niels Lohmann . Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ #ifndef NLOHMANN_JSON_HPP #define NLOHMANN_JSON_HPP #define NLOHMANN_JSON_VERSION_MAJOR 3 #define NLOHMANN_JSON_VERSION_MINOR 2 #define NLOHMANN_JSON_VERSION_PATCH 0 #include // all_of, find, for_each #include // assert #include // and, not, or #include // nullptr_t, ptrdiff_t, size_t #include // hash, less #include // initializer_list #include // istream, ostream #include // iterator_traits, random_access_iterator_tag #include // accumulate #include // string, stoi, to_string #include // declval, forward, move, pair, swap // #include #ifndef NLOHMANN_JSON_FWD_HPP #define NLOHMANN_JSON_FWD_HPP #include // int64_t, uint64_t #include // map #include // allocator #include // string #include // vector /*! @brief namespace for Niels Lohmann @see https://github.com/nlohmann @since version 1.0.0 */ namespace nlohmann { /*! @brief default JSONSerializer template argument This serializer ignores the template arguments and uses ADL ([argument-dependent lookup](https://en.cppreference.com/w/cpp/language/adl)) for serialization. */ template struct adl_serializer; template class ObjectType = std::map, template class ArrayType = std::vector, class StringType = std::string, class BooleanType = bool, class NumberIntegerType = std::int64_t, class NumberUnsignedType = std::uint64_t, class NumberFloatType = double, template class AllocatorType = std::allocator, template class JSONSerializer = adl_serializer> class basic_json; /*! @brief JSON Pointer A JSON pointer defines a string syntax for identifying a specific value within a JSON document. It can be used with functions `at` and `operator[]`. Furthermore, JSON pointers are the base for JSON patches. @sa [RFC 6901](https://tools.ietf.org/html/rfc6901) @since version 2.0.0 */ template class json_pointer; /*! @brief default JSON class This type is the default specialization of the @ref basic_json class which uses the standard template types. @since version 1.0.0 */ using json = basic_json<>; } #endif // #include // 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 #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 class ObjectType, \ template class ArrayType, \ class StringType, class BooleanType, class NumberIntegerType, \ class NumberUnsignedType, class NumberFloatType, \ template class AllocatorType, \ template class JSONSerializer> #define NLOHMANN_BASIC_JSON_TPL \ basic_json /*! @brief Helper to determine whether there's a key_type for T. This helper is used to tell associative containers apart from other containers such as sequence containers. For instance, `std::map` passes the test as it contains a `mapped_type`, whereas `std::vector` fails the test. @sa http://stackoverflow.com/a/7728728/266378 @since version 1.0.0, overworked in version 2.0.6 */ #define NLOHMANN_JSON_HAS_HELPER(type) \ template struct has_##type { \ private: \ template \ static int detect(U &&); \ static void detect(...); \ public: \ static constexpr bool value = \ std::is_integral()))>::value; \ } // #include #include // not #include // size_t #include // 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 using enable_if_t = typename std::enable_if::type; template using uncvref_t = typename std::remove_cv::type>::type; // implementation of C++14 index_sequence and affiliates // source: https://stackoverflow.com/a/32223343 template struct index_sequence { using type = index_sequence; using value_type = std::size_t; static constexpr std::size_t size() noexcept { return sizeof...(Ints); } }; template struct merge_and_renumber; template struct merge_and_renumber, index_sequence> : index_sequence < I1..., (sizeof...(I1) + I2)... > {}; template 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 using index_sequence_for = make_index_sequence; /* Implementation of two C++17 constructs: conjunction, negation. This is needed to avoid evaluating all the traits in a condition For example: not std::is_same::value and has_value_type::value will not compile when T = void (on MSVC at least). Whereas conjunction>, has_value_type>::value will stop evaluating if negation<...>::value == false Please note that those constructs must be used with caution, since symbols can become very long quickly (which can slow down compilation and cause MSVC internal compiler errors). Only use it when you have to (see example ahead). */ template struct conjunction : std::true_type {}; template struct conjunction : B1 {}; template struct conjunction : std::conditional, B1>::type {}; template struct negation : std::integral_constant {}; // dispatch utility (taken from ranges-v3) template struct priority_tag : priority_tag < N - 1 > {}; template<> struct priority_tag<0> {}; // taken from ranges-v3 template struct static_const { static constexpr T value{}; }; template constexpr T static_const::value; } } // #include #include // not #include // numeric_limits #include // false_type, is_constructible, is_integral, is_same, true_type #include // declval // #include // #include // #include 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 struct is_basic_json : std::false_type {}; NLOHMANN_BASIC_JSON_TPL_DECLARATION struct is_basic_json : std::true_type {}; //////////////////////// // has_/is_ functions // //////////////////////// // source: https://stackoverflow.com/a/37193089/4116453 template struct is_complete_type : std::false_type {}; template struct is_complete_type : std::true_type {}; NLOHMANN_JSON_HAS_HELPER(mapped_type); NLOHMANN_JSON_HAS_HELPER(key_type); NLOHMANN_JSON_HAS_HELPER(value_type); NLOHMANN_JSON_HAS_HELPER(iterator); template struct is_compatible_object_type_impl : std::false_type {}; template struct is_compatible_object_type_impl { static constexpr auto value = std::is_constructible::value and std::is_constructible::value; }; template struct is_compatible_string_type_impl : std::false_type {}; template struct is_compatible_string_type_impl { static constexpr auto value = std::is_same::value and std::is_constructible::value; }; template struct is_compatible_object_type { static auto constexpr value = is_compatible_object_type_impl < conjunction>, has_mapped_type, has_key_type>::value, typename BasicJsonType::object_t, CompatibleObjectType >::value; }; template struct is_compatible_string_type { static auto constexpr value = is_compatible_string_type_impl < conjunction>, has_value_type>::value, typename BasicJsonType::string_t, CompatibleStringType >::value; }; template struct is_basic_json_nested_type { static auto constexpr value = std::is_same::value or std::is_same::value or std::is_same::value or std::is_same::value; }; template struct is_compatible_array_type { static auto constexpr value = conjunction>, negation>, negation>, negation>, has_value_type, has_iterator>::value; }; template struct is_compatible_integer_type_impl : std::false_type {}; template struct is_compatible_integer_type_impl { // is there an assert somewhere on overflows? using RealLimits = std::numeric_limits; using CompatibleLimits = std::numeric_limits; static constexpr auto value = std::is_constructible::value and CompatibleLimits::is_integer and RealLimits::is_signed == CompatibleLimits::is_signed; }; template struct is_compatible_integer_type { static constexpr auto value = is_compatible_integer_type_impl < std::is_integral::value and not std::is_same::value, RealIntegerType, CompatibleNumberIntegerType > ::value; }; // trait checking if JSONSerializer::from_json(json const&, udt&) exists template struct has_from_json { private: // also check the return type of from_json template::from_json( std::declval(), std::declval()))>::value>> static int detect(U&&); static void detect(...); public: static constexpr bool value = std::is_integral>()))>::value; }; // This trait checks if JSONSerializer::from_json(json const&) exists // this overload is used for non-default-constructible user-defined-types template struct has_non_default_from_json { private: template < typename U, typename = enable_if_t::from_json(std::declval()))>::value >> static int detect(U&&); static void detect(...); public: static constexpr bool value = std::is_integral>()))>::value; }; // This trait checks if BasicJsonType::json_serializer::to_json exists template struct has_to_json { private: template::to_json( std::declval(), std::declval()))> static int detect(U&&); static void detect(...); public: static constexpr bool value = std::is_integral>()))>::value; }; template struct is_compatible_complete_type { static constexpr bool value = not std::is_base_of::value and not is_basic_json::value and not is_basic_json_nested_type::value and has_to_json::value; }; template struct is_compatible_type : conjunction, is_compatible_complete_type> { }; } } // #include #include // exception #include // runtime_error #include // 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) {} }; } } // #include #include // array #include // and #include // size_t #include // 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 order = {{ 0 /* null */, 3 /* object */, 4 /* array */, 5 /* string */, 1 /* boolean */, 2 /* integer */, 2 /* unsigned */, 2 /* float */ } }; const auto l_index = static_cast(lhs); const auto r_index = static_cast(rhs); return l_index < order.size() and r_index < order.size() and order[l_index] < order[r_index]; } } } // #include #include // transform #include // array #include // and, not #include // forward_list #include // inserter, front_inserter, end #include // map #include // string #include // tuple, make_tuple #include // is_arithmetic, is_same, is_enum, underlying_type, is_convertible #include // unordered_map #include // pair, declval #include // valarray // #include // #include // #include // #include // #include namespace nlohmann { namespace detail { template 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::value and not std::is_same::value, int> = 0> void get_arithmetic_value(const BasicJsonType& j, ArithmeticType& val) { switch (static_cast(j)) { case value_t::number_unsigned: { val = static_cast(*j.template get_ptr()); break; } case value_t::number_integer: { val = static_cast(*j.template get_ptr()); break; } case value_t::number_float: { val = static_cast(*j.template get_ptr()); break; } default: JSON_THROW(type_error::create(302, "type must be number, but is " + std::string(j.type_name()))); } } template 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(); } template 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(); } template < typename BasicJsonType, typename CompatibleStringType, enable_if_t < is_compatible_string_type::value and not std::is_same::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(); } template void from_json(const BasicJsonType& j, typename BasicJsonType::number_float_t& val) { get_arithmetic_value(j, val); } template void from_json(const BasicJsonType& j, typename BasicJsonType::number_unsigned_t& val) { get_arithmetic_value(j, val); } template void from_json(const BasicJsonType& j, typename BasicJsonType::number_integer_t& val) { get_arithmetic_value(j, val); } template::value, int> = 0> void from_json(const BasicJsonType& j, EnumType& e) { typename std::underlying_type::type val; get_arithmetic_value(j, val); e = static_cast(val); } template void from_json(const BasicJsonType& j, typename BasicJsonType::array_t& arr) { if (JSON_UNLIKELY(not j.is_array())) { JSON_THROW(type_error::create(302, "type must be array, but is " + std::string(j.type_name()))); } arr = *j.template get_ptr(); } // forward_list doesn't have an insert method template::value, int> = 0> void from_json(const BasicJsonType& j, std::forward_list& 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(); }); } // valarray doesn't have an insert method template::value, int> = 0> void from_json(const BasicJsonType& j, std::valarray& 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 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() returns *this, this won't call a from_json // method when value_type is BasicJsonType return i.template get(); }); } template auto from_json_array_impl(const BasicJsonType& j, CompatibleArrayType& arr, priority_tag<1> /*unused*/) -> decltype( arr.reserve(std::declval()), void()) { using std::end; arr.reserve(j.size()); std::transform(j.begin(), j.end(), std::inserter(arr, end(arr)), [](const BasicJsonType & i) { // get() returns *this, this won't call a from_json // method when value_type is BasicJsonType return i.template get(); }); } template void from_json_array_impl(const BasicJsonType& j, std::array& arr, priority_tag<2> /*unused*/) { for (std::size_t i = 0; i < N; ++i) { arr[i] = j.at(i).template get(); } } template < typename BasicJsonType, typename CompatibleArrayType, enable_if_t < is_compatible_array_type::value and not std::is_same::value and std::is_constructible < BasicJsonType, typename CompatibleArrayType::value_type >::value, int > = 0 > void from_json(const BasicJsonType& j, CompatibleArrayType& arr) { 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<2> {}); } template::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(); 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()); }); } // 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::value and not std::is_same::value and not std::is_same::value and not std::is_same::value and not std::is_same::value, int> = 0> void from_json(const BasicJsonType& j, ArithmeticType& val) { switch (static_cast(j)) { case value_t::number_unsigned: { val = static_cast(*j.template get_ptr()); break; } case value_t::number_integer: { val = static_cast(*j.template get_ptr()); break; } case value_t::number_float: { val = static_cast(*j.template get_ptr()); break; } case value_t::boolean: { val = static_cast(*j.template get_ptr()); break; } default: JSON_THROW(type_error::create(302, "type must be number, but is " + std::string(j.type_name()))); } } template void from_json(const BasicJsonType& j, std::pair& p) { p = {j.at(0).template get(), j.at(1).template get()}; } template void from_json_tuple_impl(const BasicJsonType& j, Tuple& t, index_sequence) { t = std::make_tuple(j.at(Idx).template get::type>()...); } template void from_json(const BasicJsonType& j, std::tuple& t) { from_json_tuple_impl(j, t, index_sequence_for {}); } template ::value>> void from_json(const BasicJsonType& j, std::map& 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(), p.at(1).template get()); } } template ::value>> void from_json(const BasicJsonType& j, std::unordered_map& 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(), p.at(1).template get()); } } struct from_json_fn { private: template auto call(const BasicJsonType& j, T& val, priority_tag<1> /*unused*/) const noexcept(noexcept(from_json(j, val))) -> decltype(from_json(j, val), void()) { return from_json(j, val); } template void call(const BasicJsonType& /*unused*/, T& /*unused*/, priority_tag<0> /*unused*/) const noexcept { static_assert(sizeof(BasicJsonType) == 0, "could not find from_json() method in T's namespace"); #ifdef _MSC_VER // MSVC does not show a stacktrace for the above assert using decayed = uncvref_t; static_assert(sizeof(typename decayed::force_msvc_stacktrace) == 0, "forcing MSVC stacktrace to show which T we're talking about."); #endif } public: template void operator()(const BasicJsonType& j, T& val) const noexcept(noexcept(std::declval().call(j, val, priority_tag<1> {}))) { return call(j, val, priority_tag<1> {}); } }; } /// 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::value; } } // #include #include // or, and, not #include // begin, end #include // tuple, get #include // is_same, is_constructible, is_floating_point, is_enum, underlying_type #include // move, forward, declval, pair #include // valarray #include // vector // #include // #include // #include // #include #include // size_t #include // string, to_string #include // input_iterator_tag // #include namespace nlohmann { namespace detail { /// proxy class for the items() function template 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()); } }; } } namespace nlohmann { namespace detail { ////////////////// // constructors // ////////////////// template struct external_constructor; template<> struct external_constructor { template 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 { template 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 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::value, int> = 0> static void construct(BasicJsonType& j, const CompatibleStringType& str) { j.m_type = value_t::string; j.m_value.string = j.template create(str); j.assert_invariant(); } }; template<> struct external_constructor { template 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 { template 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 { template 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 { template 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 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::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(begin(arr), end(arr)); j.assert_invariant(); } template static void construct(BasicJsonType& j, const std::vector& 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::value, int> = 0> static void construct(BasicJsonType& j, const std::valarray& 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 { template 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 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::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(begin(obj), end(obj)); j.assert_invariant(); } }; ///////////// // to_json // ///////////// template::value, int> = 0> void to_json(BasicJsonType& j, T b) noexcept { external_constructor::construct(j, b); } template::value, int> = 0> void to_json(BasicJsonType& j, const CompatibleString& s) { external_constructor::construct(j, s); } template void to_json(BasicJsonType& j, typename BasicJsonType::string_t&& s) { external_constructor::construct(j, std::move(s)); } template::value, int> = 0> void to_json(BasicJsonType& j, FloatType val) noexcept { external_constructor::construct(j, static_cast(val)); } template::value, int> = 0> void to_json(BasicJsonType& j, CompatibleNumberUnsignedType val) noexcept { external_constructor::construct(j, static_cast(val)); } template::value, int> = 0> void to_json(BasicJsonType& j, CompatibleNumberIntegerType val) noexcept { external_constructor::construct(j, static_cast(val)); } template::value, int> = 0> void to_json(BasicJsonType& j, EnumType e) noexcept { using underlying_type = typename std::underlying_type::type; external_constructor::construct(j, static_cast(e)); } template void to_json(BasicJsonType& j, const std::vector& e) { external_constructor::construct(j, e); } template::value or std::is_same::value, int> = 0> void to_json(BasicJsonType& j, const CompatibleArrayType& arr) { external_constructor::construct(j, arr); } template::value, int> = 0> void to_json(BasicJsonType& j, const std::valarray& arr) { external_constructor::construct(j, std::move(arr)); } template void to_json(BasicJsonType& j, typename BasicJsonType::array_t&& arr) { external_constructor::construct(j, std::move(arr)); } template::value, int> = 0> void to_json(BasicJsonType& j, const CompatibleObjectType& obj) { external_constructor::construct(j, obj); } template void to_json(BasicJsonType& j, typename BasicJsonType::object_t&& obj) { external_constructor::construct(j, std::move(obj)); } template::value, int> = 0> void to_json(BasicJsonType& j, T (&arr)[N]) { external_constructor::construct(j, arr); } template void to_json(BasicJsonType& j, const std::pair& p) { j = {p.first, p.second}; } // for https://github.com/nlohmann/json/pull/1134 template::iteration_proxy_internal>::value, int> = 0> void to_json(BasicJsonType& j, T b) noexcept { j = {{b.key(), b.value()}}; } template void to_json_tuple_impl(BasicJsonType& j, const Tuple& t, index_sequence) { j = {std::get(t)...}; } template void to_json(BasicJsonType& j, const std::tuple& t) { to_json_tuple_impl(j, t, index_sequence_for {}); } struct to_json_fn { private: template auto call(BasicJsonType& j, T&& val, priority_tag<1> /*unused*/) const noexcept(noexcept(to_json(j, std::forward(val)))) -> decltype(to_json(j, std::forward(val)), void()) { return to_json(j, std::forward(val)); } template void call(BasicJsonType& /*unused*/, T&& /*unused*/, priority_tag<0> /*unused*/) const noexcept { static_assert(sizeof(BasicJsonType) == 0, "could not find to_json() method in T's namespace"); #ifdef _MSC_VER // MSVC does not show a stacktrace for the above assert using decayed = uncvref_t; static_assert(sizeof(typename decayed::force_msvc_stacktrace) == 0, "forcing MSVC stacktrace to show which T we're talking about."); #endif } public: template void operator()(BasicJsonType& j, T&& val) const noexcept(noexcept(std::declval().call(j, std::forward(val), priority_tag<1> {}))) { return call(j, std::forward(val), priority_tag<1> {}); } }; } /// namespace to hold default `to_json` function namespace { constexpr const auto& to_json = detail::static_const::value; } } // #include #include // assert #include // size_t #include // strlen #include // istream #include // begin, end, iterator_traits, random_access_iterator_tag, distance, next #include // shared_ptr, make_shared, addressof #include // accumulate #include // string, char_traits #include // enable_if, is_base_of, is_pointer, is_integral, remove_pointer #include // pair, declval // #include 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::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::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::eof(). virtual std::char_traits::int_type get_character() = 0; virtual ~input_adapter_protocol() = default; }; /// a type to simplify interfaces using input_adapter_t = std::shared_ptr; /*! 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::to_int_type, to // ensure that std::char_traits::eof() and the character 0xFF do not // end up as the same value, eg. 0xFFFFFFFF. std::char_traits::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::int_type get_character() noexcept override { if (JSON_LIKELY(cursor < limit)) { return std::char_traits::to_int_type(*(cursor++)); } return std::char_traits::eof(); } private: /// pointer to the current character const char* cursor; /// pointer past the last character const char* const limit; }; template class wide_string_input_adapter : public input_adapter_protocol { public: explicit wide_string_input_adapter(const WideStringType& w) : str(w) {} std::char_traits::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::eof(); utf8_bytes_filled = 1; } else { // get the current character const int wc = static_cast(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(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::eof(); utf8_bytes_filled = 1; } else { // get the current character const int wc = static_cast(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::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(i)) {} /// input adapter for input stream input_adapter(std::istream&& i) : ia(std::make_shared(i)) {} input_adapter(const std::wstring& ws) : ia(std::make_shared>(ws)) {} input_adapter(const std::u16string& ws) : ia(std::make_shared>(ws)) {} input_adapter(const std::u32string& ws) : ia(std::make_shared>(ws)) {} /// input adapter for buffer template::value and std::is_integral::type>::value and sizeof(typename std::remove_pointer::type) == 1, int>::type = 0> input_adapter(CharT b, std::size_t l) : ia(std::make_shared(reinterpret_cast(b), l)) {} // derived support /// input adapter for string literal template::value and std::is_integral::type>::value and sizeof(typename std::remove_pointer::type) == 1, int>::type = 0> input_adapter(CharT b) : input_adapter(reinterpret_cast(b), std::strlen(reinterpret_cast(b))) {} /// input adapter for iterator range with contiguous storage template::iterator_category, std::random_access_iterator_tag>::value, int>::type = 0> input_adapter(IteratorType first, IteratorType last) { // assertion to check that the iterator range is indeed contiguous, // see http://stackoverflow.com/a/35008842/266378 for more discussion assert(std::accumulate( first, last, std::pair(true, 0), [&first](std::pair res, decltype(*first) val) { res.first &= (val == *(std::next(std::addressof(*first), res.second++))); return res; }).first); // assertion to check that each element is 1 byte long static_assert( sizeof(typename std::iterator_traits::value_type) == 1, "each element in the iterator range must have the size of 1 byte"); const auto len = static_cast(std::distance(first, last)); if (JSON_LIKELY(len > 0)) { // there is at least one element: use the address of first ia = std::make_shared(reinterpret_cast(&(*first)), len); } else { // the address of first cannot be used: use nullptr ia = std::make_shared(nullptr, len); } } /// input adapter for array template input_adapter(T (&array)[N]) : input_adapter(std::begin(array), std::end(array)) {} /// input adapter for contiguous container template::value and std::is_base_of()))>::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; }; } } // #include #include // localeconv #include // size_t #include // strtof, strtod, strtold, strtoll, strtoull #include // snprintf #include // initializer_list #include // char_traits, string #include // vector // #include // #include namespace nlohmann { namespace detail { /////////// // lexer // /////////// /*! @brief lexical analysis This class organizes the lexical analysis during JSON deserialization. */ template 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 ""; 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 ""; 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 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::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(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(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::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::eof()` in that case. Stores the scanned characters for use in error messages. @return character read from the input */ std::char_traits::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::eof())) { token_string.push_back(std::char_traits::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::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::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, "", static_cast(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::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::int_type current = std::char_traits::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 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 = '.'; }; } } // #include #include // assert #include // isfinite #include // uint8_t #include // function #include // string #include // move // #include // #include // #include #include // size_t #include // declval // #include #include // #include namespace nlohmann { namespace detail { template using void_t = void; } } // 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 Op, class... Args> struct detector { using value_t = std::false_type; using type = Default; }; template class Op, class... Args> struct detector>, Op, Args...> { using value_t = std::true_type; using type = Op; }; template