Merge branch 'develop' into cbor-integer-writer

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

docs/mkdocs/docs/api/basic_json/to_bjdata.md

@@ -123,6 +123,4 @@ Linear in the size of the JSON value `j`.
   that is not valid UTF-8 unchanged, as before; `strict` (the default if
   [`JSON_STRICT_BINARY_UTF8`](../macros/json_strict_binary_utf8.md) is enabled) throws `type_error.316`.
 - Throws `type_error.321` for a discarded value since version 3.13.0; previously, a discarded value nested in an
-  array or object was silently skipped, producing invalid BJData.
-- Writes unsigned integers wider than 64 bits as high-precision numbers since version 3.13.0; previously, they were
-  silently truncated to 64 bits.
+  array or object was silently skipped, producing invalid BJData.

docs/mkdocs/docs/api/basic_json/to_bon8.md

@@ -37,9 +37,8 @@ With (2), the bytes written before the exception remain in the output adapter.
 
 ## Exceptions
 
-- Throws [out_of_range.407](../../home/exceptions.md#jsonexceptionout_of_range407) if `j` contains an integer outside
-  the range of int64 (an unsigned integer above 9223372036854775807, or, with a number type wider than 64 bits, any
-  integer beyond int64), which BON8 cannot represent
+- Throws [out_of_range.407](../../home/exceptions.md#jsonexceptionout_of_range407) if `j` contains an unsigned integer
+  above 9223372036854775807, which BON8 cannot represent
 - Throws [type_error.316](../../home/exceptions.md#jsonexceptiontype_error316) if `j` contains a string that is not
   valid UTF-8
 - Throws [type_error.321](../../home/exceptions.md#jsonexceptiontype_error321) if `j` or a value nested in it is

docs/mkdocs/docs/api/basic_json/to_bson.md

@@ -47,10 +47,6 @@ Strong guarantee: if an exception is thrown, there are no changes in the JSON va
 
 - Throws [`type_error.317`](../../home/exceptions.md#jsonexceptiontype_error317) if the top-level type of the JSON value
   is not an object; example: `"to serialize to BSON, top-level type must be object, but is string"`
-- Throws [`out_of_range.407`](../../home/exceptions.md#jsonexceptionout_of_range407) if `j` contains a signed integer
-  outside the range of int64 or an unsigned integer outside the range of uint64, which is only possible with a number
-  type wider than 64 bits; example:
-  `"integer number 9223372036854775808 cannot be represented by BSON as it does not fit int64"`
 - Throws [`out_of_range.409`](../../home/exceptions.md#jsonexceptionout_of_range409) if a key in the JSON object contains
   a null byte (code point U+0000); example: `"BSON key cannot contain code point U+0000 (at byte 2)"`
 - Throws [`out_of_range.412`](../../home/exceptions.md#jsonexceptionout_of_range412) if the length of a document, array,
@@ -123,5 +119,3 @@ pass before anything is written.
   that is not valid UTF-8 unchanged, as before; `strict` (the default if
   [`JSON_STRICT_BINARY_UTF8`](../macros/json_strict_binary_utf8.md) is enabled) throws `type_error.316` before anything
   is written.
-- Throws `out_of_range.407` for integers that do not fit 64 bits since version 3.13.0; previously, integers of a
-  number type wider than 64 bits were silently truncated.

docs/mkdocs/docs/api/basic_json/to_cbor.md

@@ -46,9 +46,6 @@ Strong guarantee: if an exception is thrown, there are no changes in the JSON va
 
 ## Exceptions
 
-- Throws [`out_of_range.407`](../../home/exceptions.md#jsonexceptionout_of_range407) if `j` contains an integer
-  outside [-2^64, 2^64-1], which is only possible with a number type wider than 64 bits; example:
-  `"integer number 18446744073709551616 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]"`
 - Throws [type_error.316](../../home/exceptions.md#jsonexceptiontype_error316) if a string or object key in `j` is
   not valid UTF-8 and `error_handler` is `strict` (the default only if
   [`JSON_STRICT_BINARY_UTF8`](../macros/json_strict_binary_utf8.md) is enabled)
@@ -93,5 +90,3 @@ Linear in the size of the JSON value `j`.
   [`JSON_STRICT_BINARY_UTF8`](../macros/json_strict_binary_utf8.md) is enabled) throws `type_error.316`.
 - Throws `type_error.321` for a discarded value since version 3.13.0; previously, a discarded value nested in an
   array or object was silently skipped, producing invalid CBOR.
-- Throws `out_of_range.407` for integers that do not fit 64 bits since version 3.13.0; previously, integers of a
-  number type wider than 64 bits were silently truncated.

docs/mkdocs/docs/api/basic_json/to_msgpack.md

@@ -46,9 +46,6 @@ Strong guarantee: if an exception is thrown, there are no changes in the JSON va
 
 ## Exceptions
 
-- Throws [`out_of_range.407`](../../home/exceptions.md#jsonexceptionout_of_range407) if `j` contains an integer
-  outside [-2^63, 2^64-1], which is only possible with a number type wider than 64 bits; example:
-  `"integer number 18446744073709551616 cannot be represented by MessagePack as it does not fit [-2^63, 2^64-1]"`
 - Throws [`out_of_range.412`](../../home/exceptions.md#jsonexceptionout_of_range412) if the length of a string, binary
   value, array, or object exceeds 4294967295, the maximum MessagePack can store; example:
   `"MessagePack length 4294967296 exceeds maximum of 4294967295"`
@@ -115,5 +112,3 @@ Linear in the size of the JSON value `j`.
   `number_unsigned_t`.
 - Throws `type_error.321` for a discarded value since version 3.13.0; previously, a discarded value nested in an
   array or object was silently skipped, producing invalid MessagePack.
-- Throws `out_of_range.407` for integers that do not fit 64 bits since version 3.13.0; previously, integers of a
-  number type wider than 64 bits were silently truncated.

docs/mkdocs/docs/home/exceptions.md

@@ -916,19 +916,9 @@ double-precision number when `number_float_t` is `#!cpp float`.
 
 ### json.exception.out_of_range.407
 
-An integer number cannot be represented by the binary format it is serialized to:
-
-- [BON8](../features/binary_formats/bon8.md) only stores integers that fit into int64, so
-  [`to_bon8`](../api/basic_json/to_bon8.md) throws for unsigned integers greater than 9223372036854775807.
-- [CBOR](../features/binary_formats/cbor.md), [MessagePack](../features/binary_formats/messagepack.md), and
-  [BSON](../features/binary_formats/bson.md) store integers in at most 64 bits. With the default number types, every
-  integer fits, but a [`number_integer_t`](../api/basic_json/number_integer_t.md) or
-  [`number_unsigned_t`](../api/basic_json/number_unsigned_t.md) wider than 64 bits (e.g., `__int128`) can hold values
-  outside the range of the format: [-2^64, 2^64-1] for CBOR, [-2^63, 2^64-1] for MessagePack, and the range of int64
-  (signed integers) or uint64 (unsigned integers) for BSON.
-
-[UBJSON](../features/binary_formats/ubjson.md) and [BJData](../features/binary_formats/bjdata.md) never throw this
-exception, because they serialize integers beyond 64 bits as high-precision numbers.
+An unsigned integer number greater than 9223372036854775807 cannot be represented in the binary format. This exception
+is only thrown by [`to_bon8`](../api/basic_json/to_bon8.md), as [BON8](../features/binary_formats/bon8.md) has no
+unsigned 64-bit integer type.
 
 !!! note "Former occurrences"
 
@@ -937,17 +927,11 @@ exception, because they serialize integers beyond 64 bits as high-precision numb
     - Since version 3.9.0, integer numbers beyond int64 are serialized as high-precision UBJSON numbers.
     - Since version 3.12.0, integer numbers beyond int64 are serialized as uint64 BSON numbers.
 
-    Before version 3.13.0, CBOR, MessagePack, and BSON silently truncated integers wider than 64 bits, and BJData
-    truncated unsigned integers wider than 64 bits.
-
-!!! failure "Example messages"
+!!! failure "Example message"
 
     ```
     integer number 18446744073709551615 cannot be represented by BON8 as it does not fit int64
     ```
-    ```
-    integer number 1267650600228229401496703205376 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]
-    ```
 
 ### json.exception.out_of_range.408
 

include/nlohmann/detail/output/binary_writer.hpp

@@ -23,6 +23,7 @@
     #include <cstdlib> // _byteswap_ushort, _byteswap_ulong, _byteswap_uint64
 #endif
 
+#include <nlohmann/detail/bit_ops.hpp>
 #include <nlohmann/detail/input/binary_reader.hpp>
 #include <nlohmann/detail/input/string_scan.hpp>
 #include <nlohmann/detail/macro_scope.hpp>
@@ -204,36 +205,12 @@ class binary_writer
             }
 
             case value_t::number_integer:
-            {
-                if (j.m_data.m_value.number_integer >= 0)
-                {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "CBOR", "[-2^64, 2^64-1]");
-                    }
-                    // CBOR does not differentiate between positive signed
-                    // integers and unsigned integers
-                    write_cbor_head(0x00, static_cast<std::uint64_t>(j.m_data.m_value.number_integer));
-                }
-                else
-                {
-                    // a negative integer n is encoded as -1 - n
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(-1 - j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "CBOR", "[-2^64, 2^64-1]");
-                    }
-                    write_cbor_head(0x20, static_cast<std::uint64_t>(-1 - j.m_data.m_value.number_integer));
-                }
-                break;
-            }
-
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "CBOR", "uint64");
-                }
-                write_cbor_head(0x00, conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
+                // one case for both, so that an array mixing negative
+                // (number_integer) and non-negative (number_unsigned) values
+                // does not mispredict the switch
+                write_cbor_integer(j);
                 break;
             }
 
@@ -268,13 +245,7 @@ class binary_writer
 
             case value_t::array:
             {
-                // step 1: write control byte and the array size
-                write_cbor_head(0x80, j.m_data.m_value.array->size());
-
-                for (const auto& el : *j.m_data.m_value.array)
-                {
-                    write_cbor(el, depth + 1);
-                }
+                write_cbor_array(j, depth);
                 break;
             }
 
@@ -432,66 +403,12 @@ class binary_writer
             }
 
             case value_t::number_integer:
-            {
-                if (j.m_data.m_value.number_integer >= 0)
-                {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "MessagePack", "[-2^63, 2^64-1]");
-                    }
-                    // MessagePack does not differentiate between positive
-                    // signed integers and unsigned integers.
-                    write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_integer));
-                }
-                else
-                {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_min<std::int64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "MessagePack", "[-2^63, 2^64-1]");
-                    }
-                    if (j.m_data.m_value.number_integer >= -32)
-                    {
-                        // negative fixnum
-                        write_number(static_cast<std::int8_t>(j.m_data.m_value.number_integer));
-                    }
-                    else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int8_t>::min)() &&
-                             j.m_data.m_value.number_integer <= (std::numeric_limits<std::int8_t>::max)())
-                    {
-                        // int 8
-                        oa.write_character(to_char_type(0xD0));
-                        write_number(static_cast<std::int8_t>(j.m_data.m_value.number_integer));
-                    }
-                    else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int16_t>::min)() &&
-                             j.m_data.m_value.number_integer <= (std::numeric_limits<std::int16_t>::max)())
-                    {
-                        // int 16
-                        oa.write_character(to_char_type(0xD1));
-                        write_number(static_cast<std::int16_t>(j.m_data.m_value.number_integer));
-                    }
-                    else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int32_t>::min)() &&
-                             j.m_data.m_value.number_integer <= (std::numeric_limits<std::int32_t>::max)())
-                    {
-                        // int 32
-                        oa.write_character(to_char_type(0xD2));
-                        write_number(static_cast<std::int32_t>(j.m_data.m_value.number_integer));
-                    }
-                    else
-                    {
-                        // int 64
-                        oa.write_character(to_char_type(0xD3));
-                        write_number(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
-                    }
-                }
-                break;
-            }
-
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "MessagePack", "uint64");
-                }
-                write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
+                // one case for both, so that an array mixing negative
+                // (number_integer) and non-negative (number_unsigned) values
+                // does not mispredict the switch
+                write_msgpack_integer(j, std::integral_constant<bool, msgpack_integers_fit_64_bits()>());
                 break;
             }
 
@@ -509,13 +426,7 @@ class binary_writer
 
             case value_t::array:
             {
-                // step 1: write control byte and the array size
-                write_msgpack_array_prefix(j.m_data.m_value.array->size(), j);
-
-                for (const auto& el : *j.m_data.m_value.array)
-                {
-                    write_msgpack(el, depth + 1);
-                }
+                write_msgpack_array(j, depth);
                 break;
             }
 
@@ -848,133 +759,6 @@ class binary_writer
         JSON_THROW(type_error::create(exception_id::discarded_value_used, concat("cannot serialize discarded value to ", format_name), &j));
     }
 
-    /*!
-    @brief whether integer @a n does not exceed the maximum of @a TargetType
-
-    The binary formats store integers in at most 64 bits, but number_integer_t
-    and number_unsigned_t may be wider (e.g., __int128). The range of the
-    integer is taken from std::numeric_limits, so a number type whose range
-    fits into @a TargetType is never checked: the function is then constant
-    true, and the default int64_t/uint64_t types pay nothing.
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType n) noexcept
-    {
-        return integer_fits_max<TargetType>(n, std::integral_constant < bool,
-                                            (std::numeric_limits<NumberType>::digits > std::numeric_limits<TargetType>::digits) > ());
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType /*unused*/, std::false_type /*may_exceed*/) noexcept
-    {
-        return true;
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType n, std::true_type /*may_exceed*/) noexcept
-    {
-        // NumberType has more digits than TargetType, so it can hold its maximum
-        return n <= static_cast<NumberType>((std::numeric_limits<TargetType>::max)());
-    }
-
-    /*!
-    @brief whether integer @a n is not below the minimum of @a TargetType;
-           constant true if NumberType cannot hold such a value
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType n) noexcept
-    {
-        return integer_fits_min<TargetType>(n, std::integral_constant < bool, std::numeric_limits<NumberType>::is_signed &&
-                                            (!std::numeric_limits<TargetType>::is_signed || std::numeric_limits<NumberType>::digits > std::numeric_limits<TargetType>::digits) > ());
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType /*unused*/, std::false_type /*may_fall_below*/) noexcept
-    {
-        return true;
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType n, std::true_type /*may_fall_below*/) noexcept
-    {
-        // NumberType is signed and either TargetType is unsigned (minimum 0)
-        // or NumberType has more digits, so it can hold the minimum
-        return n >= static_cast<NumberType>((std::numeric_limits<TargetType>::min)());
-    }
-
-    /*!
-    @brief whether integer @a n lies in the range of @a TargetType
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits(const NumberType n) noexcept
-    {
-        return integer_fits_min<TargetType>(n) && integer_fits_max<TargetType>(n);
-    }
-
-    /*!
-    @brief the decimal digits of the non-negative integer @a n
-    */
-    template<typename NumberType>
-    static std::string integer_to_string(NumberType n, std::false_type /*is_signed*/)
-    {
-        std::string result;
-        do
-        {
-            result.push_back(static_cast<char>('0' + (n % 10)));
-            n = static_cast<NumberType>(n / 10);
-        }
-        while (n != 0);
-        std::reverse(result.begin(), result.end());
-        return result;
-    }
-
-    /*!
-    @brief the decimal digits of the integer @a n, with a leading '-' if it
-           is negative
-    */
-    template<typename NumberType>
-    static std::string integer_to_string(NumberType n, std::true_type /*is_signed*/)
-    {
-        // the remainder of a negative n is not positive; each digit is
-        // negated on its own, because negating the minimum of NumberType
-        // as a whole would overflow
-        const bool negative = n < 0;
-        std::string result;
-        do
-        {
-            const auto remainder = n % 10;
-            result.push_back(static_cast<char>('0' + (remainder < 0 ? -remainder : remainder)));
-            n = static_cast<NumberType>(n / 10);
-        }
-        while (n != 0);
-        if (negative)
-        {
-            result.push_back('-');
-        }
-        std::reverse(result.begin(), result.end());
-        return result;
-    }
-
-    /*!
-    @brief throws because the integer @a j is too large for @a format_name
-    @throw out_of_range.407 always
-    */
-    JSON_HEDLEY_NO_RETURN static void throw_integer_out_of_range(const BasicJsonType& j, const char* format_name, const char* range)
-    {
-        // neither std::to_string(), which has no overload for integer types
-        // wider than 64 bits, nor dump(), which would also instantiate the
-        // serializer for objects and fail for key types it cannot write
-        using integer_limits = std::numeric_limits<typename BasicJsonType::number_integer_t>;
-        using unsigned_limits = std::numeric_limits<typename BasicJsonType::number_unsigned_t>;
-        const std::string number = j.is_number_unsigned()
-                                   ? integer_to_string(j.m_data.m_value.number_unsigned, std::integral_constant<bool, unsigned_limits::is_signed>())
-                                   : integer_to_string(j.m_data.m_value.number_integer, std::integral_constant<bool, integer_limits::is_signed>());
-        static_cast<void>(number); // unused when JSON_NOEXCEPTION is defined
-        static_cast<void>(format_name);
-        static_cast<void>(range);
-        JSON_THROW(out_of_range::create(exception_id::integer_too_large, concat("integer number ", number, " cannot be represented by ", format_name, " as it does not fit ", range), &j));
-    }
-
     void write_msgpack_array_prefix(const std::size_t N, const BasicJsonType& j)
     {
         const auto n = to_msgpack_length(N, j);
@@ -1731,8 +1515,6 @@ class binary_writer
             is neither an object nor an array
     @throw out_of_range.415 if @a j is binary with a subtype that does not fit
            into a byte, before anything is written
-    @throw out_of_range.407 if @a j is an integer that does not fit the 64 bits
-           of a BSON integer, before anything is written
     @throw type_error.316 if @a j is a string that is not valid UTF-8, before
            anything is written
     @throw type_error.321 if @a j is discarded
@@ -1751,17 +1533,9 @@ class binary_writer
                 return 8ul;
 
             case value_t::number_integer:
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits<std::int64_t>(j.m_data.m_value.number_integer)))
-                {
-                    throw_integer_out_of_range(j, "BSON", "int64");
-                }
-                return calc_bson_integer_size(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
+                return calc_bson_integer_size(j.m_data.m_value.number_integer);
 
             case value_t::number_unsigned:
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "BSON", "uint64");
-                }
                 return calc_bson_unsigned_size(conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
 
             case value_t::string:
@@ -1800,9 +1574,8 @@ class binary_writer
             case value_t::number_float:
                 return write_bson_double(name, j.m_data.m_value.number_float);
 
-            // calc_bson_value_size() checked that integers fit 64 bits
             case value_t::number_integer:
-                return write_bson_integer(name, static_cast<std::int64_t>(j.m_data.m_value.number_integer));
+                return write_bson_integer(name, j.m_data.m_value.number_integer);
 
             case value_t::number_unsigned:
                 return write_bson_unsigned(name, conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
@@ -2113,6 +1886,316 @@ class binary_writer
         }
     }
 
+    /*!
+    @brief size class of an unsigned integer from its bit length
+
+    1, 2, 3, or 4 if @a x fits 1, 2, 4, or 8 bytes. Looked up from the number
+    of leading zero bits instead of found by branches: the sizes of a stream
+    of numbers are hard to predict, and a mispredicted branch costs more than
+    writing the number. (A table, as comparisons are vectorized by clang into
+    slower code.)
+    */
+    static unsigned integer_size_class(const std::uint64_t x) noexcept
+    {
+        static constexpr std::array<std::uint8_t, 64> size_class_by_leading_zeros =
+        {
+            {
+                4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
+                3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+                2, 2, 2, 2, 2, 2, 2, 2,
+                1, 1, 1, 1, 1, 1, 1, 1
+            }
+        };
+        // x | 1 keeps count_leading_zeros() away from 0
+        return size_class_by_leading_zeros[static_cast<std::size_t>(count_leading_zeros(x | 1u))];
+    }
+
+    /*!
+    @brief encode @a first_byte and the low-order bytes of @a value that
+           @a size_class calls for (none, or 1, 2, 4, or 8) in network order
+           to @a out
+    @return the number of characters of the encoding; up to
+            @ref integer_encoding_space characters at @a out are overwritten
+
+    The bytes are written as one byte and one 8-byte word whose bytes after
+    the value are scratch that the caller drops, so that writing them does not
+    depend on the size class. For class 0 the word is all scratch; it is
+    shifted by 60 bits rather than 64, which would be undefined.
+    */
+    std::size_t encode_first_byte_and_tail(CharType* out, const std::uint8_t first_byte, const unsigned size_class, const std::uint64_t value) const noexcept
+    {
+        out[0] = to_char_type(first_byte);
+        const unsigned unused_bits = 64u - ((8u << size_class) >> 1u);
+        const std::uint64_t tail = is_little_endian ? byte_swap(value) >> unused_bits : value << unused_bits;
+        std::memcpy(out + 1, &tail, sizeof(tail));
+        return ((std::size_t{1} << size_class) >> 1u) + 1;
+    }
+
+    /// characters encode_first_byte_and_tail() may overwrite: the first byte
+    /// and one 8-byte word
+    static constexpr std::size_t integer_encoding_space = 9;
+
+    /*!
+    @brief the 64 bits of a number_integer or number_unsigned value, and a mask
+           that is all ones if it is negative, else zero
+
+    Selected through masks rather than branches, as the type and the sign of
+    the values in an array of mixed signs alternate unpredictably (non-negative
+    integers are number_unsigned after parsing). Up to 64 bits, the sign is bit
+    63 of the result (a narrower integer is sign-extended); computing it with
+    masks keeps GCC from branching on the type. A wider integer has to be
+    compared.
+    */
+    static std::uint64_t integer_bits(const BasicJsonType& j, std::uint64_t& sign_mask) noexcept
+    {
+        const bool is_signed = j.m_data.m_type == value_t::number_integer;
+        const std::uint64_t raw = is_signed
+                                  ? static_cast<std::uint64_t>(j.m_data.m_value.number_integer)
+                                  : static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned);
+        sign_mask = sizeof(typename BasicJsonType::number_integer_t) <= sizeof(std::uint64_t)
+                    ? (std::uint64_t{0} - static_cast<std::uint64_t>(is_signed)) & (std::uint64_t{0} - (raw >> 63u))
+                    : std::uint64_t{0} - static_cast<std::uint64_t>(is_signed && j.m_data.m_value.number_integer < 0);
+        return raw;
+    }
+
+    /*!
+    @brief encode the head of a CBOR integer to @a out without branching on its
+           size
+    @return the number of characters of the head
+
+    Same output as @ref write_cbor_head. The lengths of strings and containers
+    are mostly predictable, so their heads keep the branches; the sizes of a
+    run of integers are not, so an integer's head is assembled without them.
+    */
+    std::size_t encode_cbor_integer_head(CharType* out, const std::uint8_t major_type, const std::uint64_t argument) const noexcept
+    {
+        // an argument of at most 23 is in size class 1 (one byte) and is moved
+        // to class 0, the first byte, by subtracting 1
+        const auto is_immediate = static_cast<unsigned>(argument <= 0x17u);
+        const unsigned size_class = integer_size_class(argument) - is_immediate;
+
+        // the first byte: the argument itself if it is immediate, else
+        // 0x17 + size_class, selected through a mask
+        const std::uint64_t immediate_mask = std::uint64_t{0} - is_immediate;
+        const auto first_byte = static_cast<std::uint8_t>(major_type + ((argument & immediate_mask) | ((0x17u + size_class) & ~immediate_mask)));
+        return encode_first_byte_and_tail(out, first_byte, size_class, argument);
+    }
+
+    /*!
+    @brief write an array value (head and elements) as CBOR
+
+    Integer elements, the most common scalar, are written without a call to
+    @ref write_cbor. The loop is kept out of line: inlined into write_cbor, its
+    registers would be saved and restored on every call of write_cbor, which
+    is one call per value nested in an object.
+    */
+    JSON_HEDLEY_NEVER_INLINE
+    void write_cbor_array(const BasicJsonType& j, const std::size_t depth)
+    {
+        write_cbor_head(0x80, j.m_data.m_value.array->size());
+        write_array_elements(j, [this](CharType * out, const BasicJsonType & el)
+        {
+            return encode_cbor_integer(out, el);
+        }, [this, depth](const BasicJsonType & el)
+        {
+            write_cbor(el, depth + 1);
+        });
+    }
+
+    /*!
+    @brief write the elements of the array @a j, runs of integers with
+           @a encode_integer and all other values with @a write_value
+
+    Runs of integers are encoded straight into space reserved from the sink
+    for a whole run: the write pointer stays a local, where it would otherwise
+    be reloaded from the sink for every value, as a store of a character may
+    alias it.
+    */
+    template<typename EncodeInteger, typename WriteValue>
+    void write_array_elements(const BasicJsonType& j, EncodeInteger encode_integer, WriteValue write_value)
+    {
+        constexpr std::size_t max_run = max_reserve_size / integer_encoding_space;
+        auto it = j.m_data.m_value.array->begin();
+        const auto end = j.m_data.m_value.array->end();
+        while (it != end)
+        {
+            if (it->is_number_integer())
+            {
+                CharType* const start = oa.reserve(max_run * integer_encoding_space);
+                CharType* out = start;
+                for (std::size_t n = 0; n < max_run && it != end && it->is_number_integer(); ++n, ++it)
+                {
+                    out += encode_integer(out, *it);
+                }
+                oa.commit(static_cast<std::size_t>(out - start));
+            }
+            else
+            {
+                write_value(*it);
+                ++it;
+            }
+        }
+    }
+
+    /*!
+    @brief write a number_integer or number_unsigned value as a CBOR integer
+
+    A negative integer n has major type 1 and the argument -1 - n, which is ~n.
+    Kept out of line, like the other per-value paths, so that write_cbor stays
+    small (runs of integers in an array are encoded inline instead).
+    */
+    JSON_HEDLEY_NEVER_INLINE
+    void write_cbor_integer(const BasicJsonType& j)
+    {
+        oa.commit(encode_cbor_integer(oa.reserve(integer_encoding_space), j));
+    }
+
+    /// encode a number_integer or number_unsigned value to @a out (see
+    /// @ref write_cbor_integer) and return the number of characters
+    std::size_t encode_cbor_integer(CharType* out, const BasicJsonType& j) const noexcept
+    {
+        std::uint64_t sign_mask = 0;
+        const std::uint64_t raw = integer_bits(j, sign_mask);
+        // ~raw is -1 - n in the 64 bits that are written
+        return encode_cbor_integer_head(out, static_cast<std::uint8_t>(sign_mask & 0x20u), raw ^ sign_mask);
+    }
+
+    /*!
+    @brief whether the MessagePack integers of BasicJsonType are written by
+           @ref encode_msgpack_integer
+
+    It picks the format from the low 64 bits; a wider number type keeps the
+    range checks of @ref write_msgpack on the whole value.
+    */
+    static constexpr bool msgpack_integers_fit_64_bits() noexcept
+    {
+        return sizeof(typename BasicJsonType::number_integer_t) <= sizeof(std::uint64_t)
+               && sizeof(typename BasicJsonType::number_unsigned_t) <= sizeof(std::uint64_t);
+    }
+
+    /*!
+    @brief write a number_integer or number_unsigned value as a MessagePack
+           integer without branching on its sign or size
+
+    Same output as @ref write_msgpack: a non-negative integer as positive fixint
+    (< 128) or uint 8/16/32/64, a negative one as negative fixint (>= -32) or
+    int 8/16/32/64. Kept out of line (see @ref write_cbor_integer).
+    */
+    JSON_HEDLEY_NEVER_INLINE
+    void write_msgpack_integer(const BasicJsonType& j, std::true_type /*fits_64_bits*/)
+    {
+        oa.commit(encode_msgpack_integer(oa.reserve(integer_encoding_space), j));
+    }
+
+    /// encode a number_integer or number_unsigned value to @a out (see
+    /// @ref write_msgpack_integer) and return the number of characters
+    std::size_t encode_msgpack_integer(CharType* out, const BasicJsonType& j) const noexcept
+    {
+        std::uint64_t sign_mask = 0;
+        const std::uint64_t raw = integer_bits(j, sign_mask);
+
+        // A negative n fits int 8/16/32 exactly if 2 * (-1 - n) + 1 fits
+        // uint 8/16/32, so x gives the size class of either sign.
+        const std::uint64_t is_negative = sign_mask & 1u;
+        const std::uint64_t x = ((raw ^ sign_mask) << is_negative) | is_negative;
+
+        // A fixint is at most 127, or at least -32 (x at most 63). It is in
+        // size class 1 and is moved to class 0, the first byte, by subtracting 1.
+        const auto is_fixint = static_cast<unsigned>(x <= (127u ^ (sign_mask & 64u)));
+        const unsigned size_class = integer_size_class(x) - is_fixint;
+
+        // the first byte: the low byte of the integer for a fixint, else the
+        // marker 0xCC..0xCF (uint) or 0xD0..0xD3 (int), selected through a mask
+        const std::uint64_t fixint_mask = std::uint64_t{0} - is_fixint;
+        const std::uint64_t marker = 0xCBu + size_class + (sign_mask & 4u);
+        const auto first_byte = static_cast<std::uint8_t>((raw & fixint_mask) | (marker & ~fixint_mask));
+        return encode_first_byte_and_tail(out, first_byte, size_class, raw);
+    }
+
+    /// write a number_integer or number_unsigned value of a number type wider
+    /// than 64 bits as a MessagePack integer, with range checks on the whole value
+    void write_msgpack_integer(const BasicJsonType& j, std::false_type /*fits_64_bits*/)
+    {
+        if (j.m_data.m_type == value_t::number_unsigned)
+        {
+            write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
+            return;
+        }
+
+        if (j.m_data.m_value.number_integer >= 0)
+        {
+            // MessagePack does not differentiate between positive
+            // signed integers and unsigned integers.
+            write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_integer));
+        }
+        else
+        {
+            if (j.m_data.m_value.number_integer >= -32)
+            {
+                // negative fixnum
+                write_number(static_cast<std::int8_t>(j.m_data.m_value.number_integer));
+            }
+            else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int8_t>::min)() &&
+                     j.m_data.m_value.number_integer <= (std::numeric_limits<std::int8_t>::max)())
+            {
+                // int 8
+                oa.write_character(to_char_type(0xD0));
+                write_number(static_cast<std::int8_t>(j.m_data.m_value.number_integer));
+            }
+            else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int16_t>::min)() &&
+                     j.m_data.m_value.number_integer <= (std::numeric_limits<std::int16_t>::max)())
+            {
+                // int 16
+                oa.write_character(to_char_type(0xD1));
+                write_number(static_cast<std::int16_t>(j.m_data.m_value.number_integer));
+            }
+            else if (j.m_data.m_value.number_integer >= (std::numeric_limits<std::int32_t>::min)() &&
+                     j.m_data.m_value.number_integer <= (std::numeric_limits<std::int32_t>::max)())
+            {
+                // int 32
+                oa.write_character(to_char_type(0xD2));
+                write_number(static_cast<std::int32_t>(j.m_data.m_value.number_integer));
+            }
+            else
+            {
+                // int 64
+                oa.write_character(to_char_type(0xD3));
+                write_number(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
+            }
+        }
+    }
+
+    /*!
+    @brief write an array value (head and elements) as MessagePack
+
+    See @ref write_cbor_array.
+    */
+    JSON_HEDLEY_NEVER_INLINE
+    void write_msgpack_array(const BasicJsonType& j, const std::size_t depth)
+    {
+        write_msgpack_array_prefix(j.m_data.m_value.array->size(), j);
+        write_msgpack_array_elements(j, depth, std::integral_constant<bool, msgpack_integers_fit_64_bits()>());
+    }
+
+    void write_msgpack_array_elements(const BasicJsonType& j, const std::size_t depth, std::true_type /*fits_64_bits*/)
+    {
+        write_array_elements(j, [this](CharType * out, const BasicJsonType & el)
+        {
+            return encode_msgpack_integer(out, el);
+        }, [this, depth](const BasicJsonType & el)
+        {
+            write_msgpack(el, depth + 1);
+        });
+    }
+
+    void write_msgpack_array_elements(const BasicJsonType& j, const std::size_t depth, std::false_type /*fits_64_bits*/)
+    {
+        for (const auto& el : *j.m_data.m_value.array)
+        {
+            write_msgpack(el, depth + 1);
+        }
+    }
+
     /*!
     @brief write an object key as a CBOR text string
 
@@ -2309,7 +2392,7 @@ class binary_writer
         {
             return 'L';
         }
-        if (use_bjdata && std::is_unsigned<NumberType>::value && integer_fits_max<std::uint64_t>(n))
+        if (use_bjdata && std::is_unsigned<NumberType>::value)
         {
             return 'M';
         }
@@ -2434,16 +2517,14 @@ class binary_writer
     @brief checks whether a JSON number fits into @a TargetType
     @param[in] el a JSON number of either the signed or unsigned integer kind
     @return whether @a el's value can be represented by @a TargetType without
-            wrapping, regardless of which of the two kinds it is stored as;
-            false for a value that does not even fit the 64-bit type it is
-            read as (possible for number types wider than 64 bits)
+            wrapping, regardless of which of the two kinds it is stored as
     */
     template<typename TargetType>
     static bool bjdata_ndarray_value_in_range(const BasicJsonType& el)
     {
         return el.is_number_unsigned()
-               ? integer_fits_max<std::uint64_t>(el.m_data.m_value.number_unsigned) && value_in_range_of<TargetType>(el.template get<std::uint64_t>())
-               : integer_fits<std::int64_t>(el.m_data.m_value.number_integer) && value_in_range_of<TargetType>(el.template get<std::int64_t>());
+               ? value_in_range_of<TargetType>(el.template get<std::uint64_t>())
+               : value_in_range_of<TargetType>(el.template get<std::int64_t>());
     }
 
     /*!
@@ -2675,11 +2756,10 @@ class binary_writer
         for (const auto& el : dims)
         {
             // a dimension is read as an unsigned value below, so anything that
-            // is not a non-negative integer in the range of std::uint64_t is
-            // rejected: a non-integer entry would pun unrelated bytes as the
-            // dimension, and a negative or wider one would wrap into a
-            // nonsensical length
-            if (!el.is_number_integer() || !bjdata_ndarray_value_in_range<std::uint64_t>(el))
+            // is not a non-negative integer is rejected: a non-integer entry
+            // would pun unrelated bytes as the dimension, and a negative one
+            // would wrap into a nonsensical length
+            if (!el.is_number_integer() || (!el.is_number_unsigned() && el.template get<std::int64_t>() < 0))
             {
                 return true;
             }
@@ -2811,9 +2891,9 @@ class binary_writer
 
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::int64_t>(j.m_data.m_value.number_unsigned)))
+                if (j.m_data.m_value.number_unsigned > static_cast<typename BasicJsonType::number_unsigned_t>((std::numeric_limits<std::int64_t>::max)()))
                 {
-                    throw_integer_out_of_range(j, "BON8", "int64");
+                    JSON_THROW(out_of_range::create(exception_id::integer_too_large, concat("integer number ", std::to_string(j.m_data.m_value.number_unsigned), " cannot be represented by BON8 as it does not fit int64"), &j));
                 }
                 write_bon8_integer(static_cast<std::int64_t>(j.m_data.m_value.number_unsigned));
                 string_open = false;
@@ -2822,10 +2902,6 @@ class binary_writer
 
             case value_t::number_integer:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits<std::int64_t>(j.m_data.m_value.number_integer)))
-                {
-                    throw_integer_out_of_range(j, "BON8", "int64");
-                }
                 write_bon8_integer(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
                 string_open = false;
                 break;

include/nlohmann/detail/output/output_adapters.hpp

@@ -8,7 +8,10 @@
 
 #pragma once
 
+#include <algorithm> // max
+#include <array> // array
 #include <cstddef> // size_t
+#include <cstring> // memcpy
 #include <memory> // shared_ptr, make_shared
 #include <string> // basic_string
 #include <type_traits> // conditional, integral_constant, is_same
@@ -48,32 +51,170 @@ template<typename CharType> struct output_adapter_protocol
 template<typename CharType>
 using output_adapter_t = std::shared_ptr<output_adapter_protocol<CharType>>;
 
+/// the largest number of characters a writer may reserve() from a sink at once;
+/// output_vector_sink also appends writes of at least this size with insert()
+constexpr std::size_t max_reserve_size = 256;
+
 /// @brief non-virtual output sink writing into a std::vector
 ///
 /// This sink is not part of the virtual output_adapter_protocol hierarchy: it is
 /// passed to binary_writer by value as a template parameter, so
 /// write_character()/write_characters() are ordinary (inlinable) calls with no
 /// vtable lookup and no shared_ptr. It is used for the common
-/// `to_cbor`/`to_msgpack`/... into a std::vector. output_vector_adapter below
-/// wraps this same sink to provide the virtual interface.
+/// `to_cbor`/`to_msgpack`/... into a std::vector.
+///
+/// Most writes are a few bytes - a type marker, or a number of 1, 2, 4, or 8
+/// bytes - and appending those with push_back() or insert() cost more than
+/// encoding them: insert() is an out-of-line call into a routine that can insert
+/// anywhere, and resize() is one as well with libstdc++. The sink therefore keeps
+/// its own write position: it grows the vector geometrically ahead of the
+/// position and copies each write to it, so a write is a capacity check and a
+/// short copy. The destructor shrinks the vector to the bytes actually written,
+/// also when a write throws; shrinking a vector never allocates. The sink owns
+/// that trim, so it can be moved but not copied, and the vector must not be used
+/// while the sink is alive.
 template<typename CharType, typename AllocatorType = std::allocator<CharType>>
 class output_vector_sink
 {
   public:
     explicit output_vector_sink(std::vector<CharType, AllocatorType>& vec) noexcept
+        : v(&vec), position(vec.size())
+    {}
+
+    output_vector_sink(const output_vector_sink&) = delete;
+    output_vector_sink& operator=(const output_vector_sink&) = delete;
+    output_vector_sink& operator=(output_vector_sink&&) = delete;
+
+    output_vector_sink(output_vector_sink&& other) noexcept
+        : v(other.v), position(other.position)
+    {
+        other.v = nullptr;
+    }
+
+    ~output_vector_sink()
+    {
+        if (v != nullptr)
+        {
+            v->resize(position);
+        }
+    }
+
+    // always inlined: GCC otherwise keeps these out of line in a large
+    // translation unit, and a call per byte or number costs more than the write
+    JSON_HEDLEY_ALWAYS_INLINE
+    void write_character(CharType c)
+    {
+        if (JSON_HEDLEY_UNLIKELY(position == v->size()))
+        {
+            grow(1);
+        }
+        (*v)[position++] = c;
+    }
+
+    // no JSON_HEDLEY_NON_NULL here: binary_writer legitimately passes a null
+    // pointer with length 0 for empty strings/binary values; memcpy must not
+    // see it.
+    JSON_HEDLEY_ALWAYS_INLINE
+    void write_characters(const CharType* s, std::size_t length)
+    {
+        if (JSON_HEDLEY_UNLIKELY(v->size() - position < length))
+        {
+            if (length >= max_reserve_size)
+            {
+                append(s, length);
+                return;
+            }
+            grow(length);
+        }
+        if (length != 0)
+        {
+            std::memcpy(v->data() + position, s, length * sizeof(CharType));
+            position += length;
+        }
+    }
+
+    /// @brief space for writing up to @a length characters directly
+    ///
+    /// Returns a pointer to at least @a length (at most @ref max_reserve_size)
+    /// writable characters at the write position; @ref commit then keeps the
+    /// first ones of them. A writer can store whole words there and keep only
+    /// the bytes it needs, and keep its write pointer in a register across many
+    /// values instead of updating the sink after each.
+    JSON_HEDLEY_ALWAYS_INLINE
+    CharType* reserve(const std::size_t length)
+    {
+        JSON_ASSERT(length <= max_reserve_size);
+        if (JSON_HEDLEY_UNLIKELY(v->size() - position < length))
+        {
+            grow(length);
+        }
+        return v->data() + position;
+    }
+
+    /// keep the first @a length characters written to the space from reserve()
+    JSON_HEDLEY_ALWAYS_INLINE
+    void commit(const std::size_t length) noexcept
+    {
+        position += length;
+    }
+
+  private:
+    /// make room for at least @a length more characters after the position;
+    /// kept out of line so that the write functions stay small enough for
+    /// GCC to inline them into the writer
+    JSON_HEDLEY_NEVER_INLINE
+    void grow(const std::size_t length)
+    {
+        // use the capacity the vector already has, else at least double it;
+        // never less than 64 characters, so that a small value does not
+        // reallocate for every few bytes
+        const std::size_t needed = position + length;
+        v->resize((std::max)(std::size_t{64}, needed <= v->capacity()
+                             ? v->capacity()
+                             : (std::max)(needed, v->capacity() * 2)));
+    }
+
+    /// append @a length (at least max_reserve_size) characters that do not fit
+    /// the space ahead of the position: insert() copies them once into the
+    /// grown vector, where grow() would first zero the new space and then copy
+    /// over it - noticeable for a large string or binary value
+    JSON_HEDLEY_NEVER_INLINE
+    void append(const CharType* s, const std::size_t length)
+    {
+        v->resize(position);
+        if (v->capacity() - position < length)
+        {
+            // grow geometrically, with room for the writes that follow, so
+            // that the next grow() need not reallocate right away
+            v->reserve((std::max)(v->capacity() * 2, position + length + max_reserve_size));
+        }
+        v->insert(v->end(), s, s + length);
+        position = v->size();
+    }
+
+    std::vector<CharType, AllocatorType>* v;
+    /// number of characters of *v written so far
+    std::size_t position;
+};
+
+/// output adapter for byte vectors
+template<typename CharType, typename AllocatorType = std::allocator<CharType>>
+class output_vector_adapter : public output_adapter_protocol<CharType>
+{
+  public:
+    explicit output_vector_adapter(std::vector<CharType, AllocatorType>& vec) noexcept
         : v(vec)
     {}
 
-    void write_character(CharType c)
+    void write_character(CharType c) override
     {
         v.push_back(c);
     }
 
     // no JSON_HEDLEY_NON_NULL here: binary_writer legitimately passes a null
-    // pointer with length 0 for empty strings/binary values. Appending an empty
-    // range is a no-op; the type-erased path tolerates this via the (unattributed)
-    // virtual base, and the concrete sink must do the same.
-    void write_characters(const CharType* s, std::size_t length)
+    // pointer with length 0 for empty strings/binary values; appending an empty
+    // range is a no-op
+    void write_characters(const CharType* s, std::size_t length) override
     {
         v.insert(v.end(), s, s + length);
     }
@@ -82,33 +223,6 @@ class output_vector_sink
     std::vector<CharType, AllocatorType>& v;
 };
 
-/// output adapter for byte vectors
-///
-/// The appending itself lives in output_vector_sink; this class only adds the
-/// virtual output_adapter_protocol interface on top of it, so both the
-/// type-erased and the templated path share one implementation.
-template<typename CharType, typename AllocatorType = std::allocator<CharType>>
-class output_vector_adapter : public output_adapter_protocol<CharType>
-{
-  public:
-    explicit output_vector_adapter(std::vector<CharType, AllocatorType>& vec) noexcept
-        : sink(vec)
-    {}
-
-    void write_character(CharType c) override
-    {
-        sink.write_character(c);
-    }
-
-    void write_characters(const CharType* s, std::size_t length) override
-    {
-        sink.write_characters(s, length);
-    }
-
-  private:
-    output_vector_sink<CharType, AllocatorType> sink;
-};
-
 #ifndef JSON_NO_IO
 /// output adapter for output streams
 template<typename CharType>
@@ -188,8 +302,24 @@ class output_adapter_sink
         oa->write_characters(s, length);
     }
 
+    /// space for writing up to @a length characters (see
+    /// output_vector_sink::reserve)
+    CharType* reserve(const std::size_t length) noexcept
+    {
+        JSON_ASSERT(length <= max_reserve_size);
+        static_cast<void>(length);
+        return reserved.data();
+    }
+
+    /// write the first @a length characters written to the space from reserve()
+    void commit(const std::size_t length)
+    {
+        oa->write_characters(reserved.data(), length);
+    }
+
   private:
     output_adapter_t<CharType> oa;
+    std::array<CharType, max_reserve_size> reserved{};
 };
 
 /// @brief whether std::basic_string<CharType> has a non-deprecated std::char_traits

include/nlohmann/detail/output/serializer.hpp

@@ -1525,7 +1525,7 @@ class serializer
         // without leading zeros
         if (abs_value < 10000000000000000u)
         {
-            const auto value = static_cast<std::uint64_t>(abs_value); // fits: abs_value < 10^16
+            const std::uint64_t value = abs_value;
             const std::uint64_t upper = value / 100000000u;
             const std::uint64_t first = dtoa_impl::eight_digit_bytes(upper != 0 ? upper : value);
             const auto leading = static_cast<unsigned>(count_leading_zeros(first) / 8); // (first is not 0)

include/nlohmann/detail/value_t.hpp

@@ -14,6 +14,7 @@
 #include <cstdint> // uint8_t
 #include <limits> // numeric_limits
 #include <string> // string
+#include <type_traits> // is_signed
 
 #include <nlohmann/detail/macro_scope.hpp>
 #if JSON_HAS_THREE_WAY_COMPARISON
@@ -174,16 +175,13 @@ FloatType compare_integer_with_float(const IntegerType i, const FloatType f) noe
 
     // values of IntegerType lie in [-bound, bound) when signed and in
     // [0, bound) when unsigned; digits excludes the sign bit, so bound is a
-    // power of two that the float represents exactly; the signedness comes
-    // from numeric_limits as well, because std::is_signed is false for class
-    // types such as 128-bit or multiprecision integers
-    using limits = std::numeric_limits<IntegerType>;
-    const FloatType bound = std::ldexp(static_cast<FloatType>(1), limits::digits);
+    // power of two that the float represents exactly
+    const FloatType bound = std::ldexp(static_cast<FloatType>(1), std::numeric_limits<IntegerType>::digits);
     if (f >= bound)
     {
         return ordered(-1);
     }
-    if (limits::is_signed ? (f < -bound) : (f < static_cast<FloatType>(0)))
+    if (std::is_signed<IntegerType>::value ? (f < -bound) : (f < static_cast<FloatType>(0)))
     {
         return ordered(1);
     }

single_include/nlohmann/json.hpp

@@ -289,6 +289,7 @@
 #include <cstdint> // uint8_t
 #include <limits> // numeric_limits
 #include <string> // string
+#include <type_traits> // is_signed
 
 // #include <nlohmann/detail/macro_scope.hpp>
 //     __ _____ _____ _____
@@ -3395,16 +3396,13 @@ FloatType compare_integer_with_float(const IntegerType i, const FloatType f) noe
 
     // values of IntegerType lie in [-bound, bound) when signed and in
     // [0, bound) when unsigned; digits excludes the sign bit, so bound is a
-    // power of two that the float represents exactly; the signedness comes
-    // from numeric_limits as well, because std::is_signed is false for class
-    // types such as 128-bit or multiprecision integers
-    using limits = std::numeric_limits<IntegerType>;
-    const FloatType bound = std::ldexp(static_cast<FloatType>(1), limits::digits);
+    // power of two that the float represents exactly
+    const FloatType bound = std::ldexp(static_cast<FloatType>(1), std::numeric_limits<IntegerType>::digits);
     if (f >= bound)
     {
         return ordered(-1);
     }
-    if (limits::is_signed ? (f < -bound) : (f < static_cast<FloatType>(0)))
+    if (std::is_signed<IntegerType>::value ? (f < -bound) : (f < static_cast<FloatType>(0)))
     {
         return ordered(1);
     }
@@ -23124,10 +23122,6 @@ class binary_writer
             {
                 if (j.m_data.m_value.number_integer >= 0)
                 {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "CBOR", "[-2^64, 2^64-1]");
-                    }
                     // CBOR does not differentiate between positive signed
                     // integers and unsigned integers
                     write_cbor_head(0x00, static_cast<std::uint64_t>(j.m_data.m_value.number_integer));
@@ -23135,10 +23129,6 @@ class binary_writer
                 else
                 {
                     // a negative integer n is encoded as -1 - n
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(-1 - j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "CBOR", "[-2^64, 2^64-1]");
-                    }
                     write_cbor_head(0x20, static_cast<std::uint64_t>(-1 - j.m_data.m_value.number_integer));
                 }
                 break;
@@ -23146,10 +23136,6 @@ class binary_writer
 
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "CBOR", "uint64");
-                }
                 write_cbor_head(0x00, conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
                 break;
             }
@@ -23352,20 +23338,12 @@ class binary_writer
             {
                 if (j.m_data.m_value.number_integer >= 0)
                 {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "MessagePack", "[-2^63, 2^64-1]");
-                    }
                     // MessagePack does not differentiate between positive
                     // signed integers and unsigned integers.
                     write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_integer));
                 }
                 else
                 {
-                    if (JSON_HEDLEY_UNLIKELY(!integer_fits_min<std::int64_t>(j.m_data.m_value.number_integer)))
-                    {
-                        throw_integer_out_of_range(j, "MessagePack", "[-2^63, 2^64-1]");
-                    }
                     if (j.m_data.m_value.number_integer >= -32)
                     {
                         // negative fixnum
@@ -23404,10 +23382,6 @@ class binary_writer
 
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "MessagePack", "uint64");
-                }
                 write_msgpack_unsigned(static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
                 break;
             }
@@ -23765,133 +23739,6 @@ class binary_writer
         JSON_THROW(type_error::create(exception_id::discarded_value_used, concat("cannot serialize discarded value to ", format_name), &j));
     }
 
-    /*!
-    @brief whether integer @a n does not exceed the maximum of @a TargetType
-
-    The binary formats store integers in at most 64 bits, but number_integer_t
-    and number_unsigned_t may be wider (e.g., __int128). The range of the
-    integer is taken from std::numeric_limits, so a number type whose range
-    fits into @a TargetType is never checked: the function is then constant
-    true, and the default int64_t/uint64_t types pay nothing.
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType n) noexcept
-    {
-        return integer_fits_max<TargetType>(n, std::integral_constant < bool,
-                                            (std::numeric_limits<NumberType>::digits > std::numeric_limits<TargetType>::digits) > ());
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType /*unused*/, std::false_type /*may_exceed*/) noexcept
-    {
-        return true;
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_max(const NumberType n, std::true_type /*may_exceed*/) noexcept
-    {
-        // NumberType has more digits than TargetType, so it can hold its maximum
-        return n <= static_cast<NumberType>((std::numeric_limits<TargetType>::max)());
-    }
-
-    /*!
-    @brief whether integer @a n is not below the minimum of @a TargetType;
-           constant true if NumberType cannot hold such a value
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType n) noexcept
-    {
-        return integer_fits_min<TargetType>(n, std::integral_constant < bool, std::numeric_limits<NumberType>::is_signed &&
-                                            (!std::numeric_limits<TargetType>::is_signed || std::numeric_limits<NumberType>::digits > std::numeric_limits<TargetType>::digits) > ());
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType /*unused*/, std::false_type /*may_fall_below*/) noexcept
-    {
-        return true;
-    }
-
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits_min(const NumberType n, std::true_type /*may_fall_below*/) noexcept
-    {
-        // NumberType is signed and either TargetType is unsigned (minimum 0)
-        // or NumberType has more digits, so it can hold the minimum
-        return n >= static_cast<NumberType>((std::numeric_limits<TargetType>::min)());
-    }
-
-    /*!
-    @brief whether integer @a n lies in the range of @a TargetType
-    */
-    template<typename TargetType, typename NumberType>
-    static constexpr bool integer_fits(const NumberType n) noexcept
-    {
-        return integer_fits_min<TargetType>(n) && integer_fits_max<TargetType>(n);
-    }
-
-    /*!
-    @brief the decimal digits of the non-negative integer @a n
-    */
-    template<typename NumberType>
-    static std::string integer_to_string(NumberType n, std::false_type /*is_signed*/)
-    {
-        std::string result;
-        do
-        {
-            result.push_back(static_cast<char>('0' + (n % 10)));
-            n = static_cast<NumberType>(n / 10);
-        }
-        while (n != 0);
-        std::reverse(result.begin(), result.end());
-        return result;
-    }
-
-    /*!
-    @brief the decimal digits of the integer @a n, with a leading '-' if it
-           is negative
-    */
-    template<typename NumberType>
-    static std::string integer_to_string(NumberType n, std::true_type /*is_signed*/)
-    {
-        // the remainder of a negative n is not positive; each digit is
-        // negated on its own, because negating the minimum of NumberType
-        // as a whole would overflow
-        const bool negative = n < 0;
-        std::string result;
-        do
-        {
-            const auto remainder = n % 10;
-            result.push_back(static_cast<char>('0' + (remainder < 0 ? -remainder : remainder)));
-            n = static_cast<NumberType>(n / 10);
-        }
-        while (n != 0);
-        if (negative)
-        {
-            result.push_back('-');
-        }
-        std::reverse(result.begin(), result.end());
-        return result;
-    }
-
-    /*!
-    @brief throws because the integer @a j is too large for @a format_name
-    @throw out_of_range.407 always
-    */
-    JSON_HEDLEY_NO_RETURN static void throw_integer_out_of_range(const BasicJsonType& j, const char* format_name, const char* range)
-    {
-        // neither std::to_string(), which has no overload for integer types
-        // wider than 64 bits, nor dump(), which would also instantiate the
-        // serializer for objects and fail for key types it cannot write
-        using integer_limits = std::numeric_limits<typename BasicJsonType::number_integer_t>;
-        using unsigned_limits = std::numeric_limits<typename BasicJsonType::number_unsigned_t>;
-        const std::string number = j.is_number_unsigned()
-                                   ? integer_to_string(j.m_data.m_value.number_unsigned, std::integral_constant<bool, unsigned_limits::is_signed>())
-                                   : integer_to_string(j.m_data.m_value.number_integer, std::integral_constant<bool, integer_limits::is_signed>());
-        static_cast<void>(number); // unused when JSON_NOEXCEPTION is defined
-        static_cast<void>(format_name);
-        static_cast<void>(range);
-        JSON_THROW(out_of_range::create(exception_id::integer_too_large, concat("integer number ", number, " cannot be represented by ", format_name, " as it does not fit ", range), &j));
-    }
-
     void write_msgpack_array_prefix(const std::size_t N, const BasicJsonType& j)
     {
         const auto n = to_msgpack_length(N, j);
@@ -24648,8 +24495,6 @@ class binary_writer
             is neither an object nor an array
     @throw out_of_range.415 if @a j is binary with a subtype that does not fit
            into a byte, before anything is written
-    @throw out_of_range.407 if @a j is an integer that does not fit the 64 bits
-           of a BSON integer, before anything is written
     @throw type_error.316 if @a j is a string that is not valid UTF-8, before
            anything is written
     @throw type_error.321 if @a j is discarded
@@ -24668,17 +24513,9 @@ class binary_writer
                 return 8ul;
 
             case value_t::number_integer:
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits<std::int64_t>(j.m_data.m_value.number_integer)))
-                {
-                    throw_integer_out_of_range(j, "BSON", "int64");
-                }
-                return calc_bson_integer_size(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
+                return calc_bson_integer_size(j.m_data.m_value.number_integer);
 
             case value_t::number_unsigned:
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::uint64_t>(j.m_data.m_value.number_unsigned)))
-                {
-                    throw_integer_out_of_range(j, "BSON", "uint64");
-                }
                 return calc_bson_unsigned_size(conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
 
             case value_t::string:
@@ -24717,9 +24554,8 @@ class binary_writer
             case value_t::number_float:
                 return write_bson_double(name, j.m_data.m_value.number_float);
 
-            // calc_bson_value_size() checked that integers fit 64 bits
             case value_t::number_integer:
-                return write_bson_integer(name, static_cast<std::int64_t>(j.m_data.m_value.number_integer));
+                return write_bson_integer(name, j.m_data.m_value.number_integer);
 
             case value_t::number_unsigned:
                 return write_bson_unsigned(name, conditional_static_cast<std::uint64_t>(j.m_data.m_value.number_unsigned));
@@ -25226,7 +25062,7 @@ class binary_writer
         {
             return 'L';
         }
-        if (use_bjdata && std::is_unsigned<NumberType>::value && integer_fits_max<std::uint64_t>(n))
+        if (use_bjdata && std::is_unsigned<NumberType>::value)
         {
             return 'M';
         }
@@ -25351,16 +25187,14 @@ class binary_writer
     @brief checks whether a JSON number fits into @a TargetType
     @param[in] el a JSON number of either the signed or unsigned integer kind
     @return whether @a el's value can be represented by @a TargetType without
-            wrapping, regardless of which of the two kinds it is stored as;
-            false for a value that does not even fit the 64-bit type it is
-            read as (possible for number types wider than 64 bits)
+            wrapping, regardless of which of the two kinds it is stored as
     */
     template<typename TargetType>
     static bool bjdata_ndarray_value_in_range(const BasicJsonType& el)
     {
         return el.is_number_unsigned()
-               ? integer_fits_max<std::uint64_t>(el.m_data.m_value.number_unsigned) && value_in_range_of<TargetType>(el.template get<std::uint64_t>())
-               : integer_fits<std::int64_t>(el.m_data.m_value.number_integer) && value_in_range_of<TargetType>(el.template get<std::int64_t>());
+               ? value_in_range_of<TargetType>(el.template get<std::uint64_t>())
+               : value_in_range_of<TargetType>(el.template get<std::int64_t>());
     }
 
     /*!
@@ -25592,11 +25426,10 @@ class binary_writer
         for (const auto& el : dims)
         {
             // a dimension is read as an unsigned value below, so anything that
-            // is not a non-negative integer in the range of std::uint64_t is
-            // rejected: a non-integer entry would pun unrelated bytes as the
-            // dimension, and a negative or wider one would wrap into a
-            // nonsensical length
-            if (!el.is_number_integer() || !bjdata_ndarray_value_in_range<std::uint64_t>(el))
+            // is not a non-negative integer is rejected: a non-integer entry
+            // would pun unrelated bytes as the dimension, and a negative one
+            // would wrap into a nonsensical length
+            if (!el.is_number_integer() || (!el.is_number_unsigned() && el.template get<std::int64_t>() < 0))
             {
                 return true;
             }
@@ -25728,9 +25561,9 @@ class binary_writer
 
             case value_t::number_unsigned:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits_max<std::int64_t>(j.m_data.m_value.number_unsigned)))
+                if (j.m_data.m_value.number_unsigned > static_cast<typename BasicJsonType::number_unsigned_t>((std::numeric_limits<std::int64_t>::max)()))
                 {
-                    throw_integer_out_of_range(j, "BON8", "int64");
+                    JSON_THROW(out_of_range::create(exception_id::integer_too_large, concat("integer number ", std::to_string(j.m_data.m_value.number_unsigned), " cannot be represented by BON8 as it does not fit int64"), &j));
                 }
                 write_bon8_integer(static_cast<std::int64_t>(j.m_data.m_value.number_unsigned));
                 string_open = false;
@@ -25739,10 +25572,6 @@ class binary_writer
 
             case value_t::number_integer:
             {
-                if (JSON_HEDLEY_UNLIKELY(!integer_fits<std::int64_t>(j.m_data.m_value.number_integer)))
-                {
-                    throw_integer_out_of_range(j, "BON8", "int64");
-                }
                 write_bon8_integer(static_cast<std::int64_t>(j.m_data.m_value.number_integer));
                 string_open = false;
                 break;
@@ -29627,7 +29456,7 @@ class serializer
         // without leading zeros
         if (abs_value < 10000000000000000u)
         {
-            const auto value = static_cast<std::uint64_t>(abs_value); // fits: abs_value < 10^16
+            const std::uint64_t value = abs_value;
             const std::uint64_t upper = value / 100000000u;
             const std::uint64_t first = dtoa_impl::eight_digit_bytes(upper != 0 ? upper : value);
             const auto leading = static_cast<unsigned>(count_leading_zeros(first) / 8); // (first is not 0)

tests/src/unit-binary_writer_sinks.cpp

@@ -11,8 +11,10 @@
 #include <nlohmann/json.hpp>
 using nlohmann::json;
 
+#include <array>
 #include <cstdint>
 #include <limits>
+#include <sstream>
 #include <string>
 #include <vector>
 
@@ -35,6 +37,17 @@ std::vector<json> test_values()
         big_object[std::to_string(i)] = i;
     }
 
+    // large strings and binary values between small ones: a write too large
+    // for the space the vector sink has ahead takes a different path than a
+    // small one, and the small writes after it must still land behind it
+    json large_payloads = json::array();
+    for (int i = 0; i < 50; ++i)
+    {
+        large_payloads.push_back(std::string(300, static_cast<char>('a' + (i % 26))));
+        large_payloads.push_back(i);
+        large_payloads.push_back(json::binary(std::vector<std::uint8_t>(static_cast<std::size_t>(100 * i), static_cast<std::uint8_t>(i))));
+    }
+
     return
     {
         json(nullptr), json(true), json(false),
@@ -46,7 +59,8 @@ std::vector<json> test_values()
         json::array(), json::object(),
         json::array({1, 2, 3}), json({{"a", 1}, {"b", nullptr}}),
         json({{"nested", {{"deep", json::array({1, "two", 3.0, nullptr})}}}}),
-        big_array, big_object
+        big_array, big_object, large_payloads,
+        json::binary(std::vector<std::uint8_t>(5000, 0x42))
     };
 }
 
@@ -170,6 +184,258 @@ TEST_CASE("binary writer output sinks")
     }
 }
 
+namespace
+{
+
+// the CBOR head of an integer, encoded independently of binary_writer
+void reference_cbor_head(std::vector<std::uint8_t>& out, const std::uint8_t major_type, const std::uint64_t argument)
+{
+    if (argument <= 23)
+    {
+        out.push_back(static_cast<std::uint8_t>(major_type + argument));
+        return;
+    }
+
+    int length = 8;
+    std::uint8_t additional = 27;
+    if (argument <= 0xFF)
+    {
+        length = 1;
+        additional = 24;
+    }
+    else if (argument <= 0xFFFF)
+    {
+        length = 2;
+        additional = 25;
+    }
+    else if (argument <= 0xFFFFFFFF)
+    {
+        length = 4;
+        additional = 26;
+    }
+
+    out.push_back(static_cast<std::uint8_t>(major_type + additional));
+    for (int i = length - 1; i >= 0; --i)
+    {
+        out.push_back(static_cast<std::uint8_t>(argument >> (8 * i)));
+    }
+}
+
+// the CBOR encoding of an integer value
+void reference_cbor_integer(std::vector<std::uint8_t>& out, const json& j)
+{
+    if (j.is_number_unsigned())
+    {
+        reference_cbor_head(out, 0x00, j.get<std::uint64_t>());
+        return;
+    }
+
+    const auto n = j.get<std::int64_t>();
+    if (n >= 0)
+    {
+        reference_cbor_head(out, 0x00, static_cast<std::uint64_t>(n));
+    }
+    else
+    {
+        reference_cbor_head(out, 0x20, static_cast<std::uint64_t>(-1 - n));
+    }
+}
+
+// the MessagePack encoding of an integer value, encoded independently of
+// binary_writer
+void reference_msgpack_integer(std::vector<std::uint8_t>& out, const json& j)
+{
+    const auto put = [&out](const std::uint8_t marker, const std::uint64_t bits, const int length)
+    {
+        out.push_back(marker);
+        for (int i = length - 1; i >= 0; --i)
+        {
+            out.push_back(static_cast<std::uint8_t>(bits >> (8 * i)));
+        }
+    };
+
+    if (j.is_number_unsigned() || j.get<std::int64_t>() >= 0)
+    {
+        const auto u = j.get<std::uint64_t>();
+        if (u < 128)
+        {
+            out.push_back(static_cast<std::uint8_t>(u));
+        }
+        else if (u <= 0xFF)
+        {
+            put(0xCC, u, 1);
+        }
+        else if (u <= 0xFFFF)
+        {
+            put(0xCD, u, 2);
+        }
+        else if (u <= 0xFFFFFFFF)
+        {
+            put(0xCE, u, 4);
+        }
+        else
+        {
+            put(0xCF, u, 8);
+        }
+        return;
+    }
+
+    const auto n = j.get<std::int64_t>();
+    const auto bits = static_cast<std::uint64_t>(n);
+    if (n >= -32)
+    {
+        out.push_back(static_cast<std::uint8_t>(bits));
+    }
+    else if (n >= -128)
+    {
+        put(0xD0, bits, 1);
+    }
+    else if (n >= -32768)
+    {
+        put(0xD1, bits, 2);
+    }
+    else if (n >= -2147483648LL)
+    {
+        put(0xD2, bits, 4);
+    }
+    else
+    {
+        put(0xD3, bits, 8);
+    }
+}
+
+// integers at every boundary of the CBOR and MessagePack sizes, signed and
+// unsigned
+std::vector<json> boundary_integers()
+{
+    std::vector<json> result;
+    const std::array<std::uint64_t, 38> boundaries =
+    {
+        {
+            0, 1, 22, 23, 24, 25, 31, 32, 33, 126, 127, 128, 129, 254, 255, 256, 257,
+            32766, 32767, 32768, 32769, 65534, 65535, 65536, 65537,
+            2147483646ULL, 2147483647ULL, 2147483648ULL, 2147483649ULL,
+            4294967294ULL, 4294967295ULL, 4294967296ULL, 4294967297ULL,
+            9223372036854775806ULL, 9223372036854775807ULL, 9223372036854775808ULL,
+            18446744073709551614ULL, 18446744073709551615ULL
+        }
+    };
+    for (const std::uint64_t u : boundaries)
+    {
+        result.emplace_back(u);
+        if (u <= static_cast<std::uint64_t>((std::numeric_limits<std::int64_t>::max)()))
+        {
+            const auto n = static_cast<std::int64_t>(u);
+            result.emplace_back(n);
+            result.emplace_back(-n);
+            result.emplace_back(-n - 1);
+        }
+    }
+    return result;
+}
+
+} // namespace
+
+// Arrays of integers are written in runs straight into space reserved from the
+// sink, with branch-free encodings; check them against independent encoders for
+// runs shorter and longer than one reservation, broken up by other values, and
+// through each sink.
+TEST_CASE("CBOR and MessagePack integer arrays")
+{
+    const std::vector<json> integers = boundary_integers();
+
+    for (const std::size_t length :
+            {
+                std::size_t{1}, std::size_t{27}, std::size_t{28}, std::size_t{29}, std::size_t{57}, std::size_t{1000}
+            })
+    {
+        for (const std::size_t break_every :
+                {
+                    std::size_t{0}, std::size_t{1}, std::size_t{7}, std::size_t{28}
+                })
+        {
+            CAPTURE(length)
+            CAPTURE(break_every)
+
+            json array = json::array();
+            std::vector<std::uint8_t> expected_cbor;
+            std::vector<std::uint8_t> expected_msgpack;
+            for (std::size_t i = 0; i < length; ++i)
+            {
+                if (break_every != 0 && i % (break_every + 1) == break_every)
+                {
+                    array.push_back("x");
+                    expected_cbor.push_back(0x61);
+                    expected_cbor.push_back('x');
+                    expected_msgpack.push_back(0xA1);
+                    expected_msgpack.push_back('x');
+                }
+                else
+                {
+                    const json& value = integers[(i * 7) % integers.size()];
+                    array.push_back(value);
+                    reference_cbor_integer(expected_cbor, value);
+                    reference_msgpack_integer(expected_msgpack, value);
+                }
+            }
+
+            std::vector<std::uint8_t> head;
+            reference_cbor_head(head, 0x80, length);
+            expected_cbor.insert(expected_cbor.begin(), head.begin(), head.end());
+
+            head.clear();
+            if (length <= 15)
+            {
+                head.push_back(static_cast<std::uint8_t>(0x90 + length));
+            }
+            else
+            {
+                head = {0xDC, static_cast<std::uint8_t>(length >> 8), static_cast<std::uint8_t>(length)};
+            }
+            expected_msgpack.insert(expected_msgpack.begin(), head.begin(), head.end());
+
+            CHECK(json::to_cbor(array) == expected_cbor);
+            CHECK(json::to_msgpack(array) == expected_msgpack);
+
+            std::vector<std::uint8_t> through_adapter = {0xAA};
+            json::to_cbor(array, through_adapter);
+            CHECK(std::vector<std::uint8_t>(through_adapter.begin() + 1, through_adapter.end()) == expected_cbor);
+            through_adapter = {0xAA};
+            json::to_msgpack(array, through_adapter);
+            CHECK(std::vector<std::uint8_t>(through_adapter.begin() + 1, through_adapter.end()) == expected_msgpack);
+
+            std::ostringstream stream;
+            json::to_cbor(array, stream);
+            std::string streamed = stream.str();
+            CHECK(std::vector<std::uint8_t>(streamed.begin(), streamed.end()) == expected_cbor);
+            stream.str("");
+            json::to_msgpack(array, stream);
+            streamed = stream.str();
+            CHECK(std::vector<std::uint8_t>(streamed.begin(), streamed.end()) == expected_msgpack);
+
+            CHECK(json::from_cbor(expected_cbor) == array);
+            CHECK(json::from_msgpack(expected_msgpack) == array);
+
+            // the same array as object members
+            const json object = {{"a", array}, {"b", array}};
+            CHECK(json::from_cbor(json::to_cbor(object)) == object);
+            CHECK(json::from_msgpack(json::to_msgpack(object)) == object);
+        }
+    }
+
+    // single values, outside an array
+    for (const auto& value : integers)
+    {
+        CAPTURE(value)
+        std::vector<std::uint8_t> expected;
+        reference_cbor_integer(expected, value);
+        CHECK(json::to_cbor(value) == expected);
+        expected.clear();
+        reference_msgpack_integer(expected, value);
+        CHECK(json::to_msgpack(value) == expected);
+    }
+}
+
 // binary_reserve_hint() is documented as a *lower* bound on the serialized size,
 // so that reserving it up front can never leave the returned vector holding
 // capacity beyond what the value actually needs.

tests/src/unit-custom-array-type.cpp

@@ -9,9 +9,7 @@
 #include "doctest_compatibility.h"
 
 #include <nlohmann/json.hpp>
-using nlohmann::json;
 
-#include <cstdint>
 #include <deque>
 #include <map>
 #include <memory>
@@ -166,47 +164,3 @@ TEST_CASE("array type without at()")
     CHECK(j.at(no_at_json::json_pointer("/1")) == 2);
     CHECK_THROWS_AS(j.at(no_at_json::json_pointer("/3")), no_at_json::out_of_range);
 }
-
-TEST_CASE("regression test #5476 - array type without reserve()")
-{
-    // the capacity reserved for definite-length arrays must not require the
-    // array type to have a reserve() member function
-    SECTION("std::deque")
-    {
-        const auto j = deque_json::parse(R"({"a":[1,[2,3]],"b":[]})");
-        CHECK(j.dump() == R"({"a":[1,[2,3]],"b":[]})");
-
-        // the binary formats pass a definite length to start_array()
-        CHECK(deque_json::from_cbor(deque_json::to_cbor(j)) == j);
-        CHECK(deque_json::from_msgpack(deque_json::to_msgpack(j)) == j);
-        CHECK(deque_json::from_bon8(deque_json::to_bon8(j)) == j);
-
-        // parse() instantiates the callback parser as well, which reserves too
-        const auto with_callback = deque_json::parse(R"([1,2,3])", [](int /*depth*/, deque_json::parse_event_t /*event*/, deque_json& /*parsed*/) noexcept
-        {
-            return true;
-        });
-        CHECK(with_callback == deque_json({1, 2, 3}));
-    }
-
-    SECTION("std::vector still reserves")
-    {
-        json array = json::array();
-        for (int i = 0; i < 100; ++i)
-        {
-            array.push_back(i);
-        }
-
-        const auto j = json::from_cbor(json::to_cbor(array));
-        CHECK(j == array);
-        CHECK(j.get_ref<const json::array_t&>().capacity() >= 100);
-    }
-
-    SECTION("the reservation stays capped")
-    {
-        // CBOR array announcing 2^32-1 elements, but truncated right after the
-        // header: the input must be rejected without reserving that capacity
-        const std::vector<std::uint8_t> truncated = {0x9A, 0xFF, 0xFF, 0xFF, 0xFF};
-        CHECK(json::from_cbor(truncated, true, false).is_discarded());
-    }
-}

tests/src/unit-custom-number-types.cpp

@@ -1,367 +0,0 @@
-//     __ _____ _____ _____
-//  __|  |   __|     |   | |  JSON for Modern C++ (supporting code)
-// |  |  |__   |  |  | | | |  version 3.12.0
-// |_____|_____|_____|_|___|  https://github.com/nlohmann/json
-//
-// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
-// SPDX-License-Identifier: MIT
-
-#include "doctest_compatibility.h"
-
-#include <nlohmann/json.hpp>
-
-#include <cstdint>
-#include <limits>
-#include <map>
-#include <string>
-#include <type_traits>
-#include <vector>
-
-#if JSON_HAS_THREE_WAY_COMPARISON
-    #include <compare>
-#endif
-
-namespace custom_number_types
-{
-
-// a signed integer of class type: std::is_signed is only true for arithmetic
-// types, so the library has to take the signedness from std::numeric_limits,
-// as it does for 128-bit and multiprecision integer classes such as
-// absl::int128 or boost::multiprecision::cpp_int
-class class_int
-{
-  public:
-    // trivial, like the 128-bit integer classes: the value is stored in a union
-    class_int() = default;
-
-    // implicit from built-in integers and explicit from floats, like absl::int128;
-    // non-template overloads (as absl::int128 has) because clang 3.4 and 3.5
-    // reject basic_json's defaulted data() for a number type with template
-    // constructors
-    class_int(int v) : value(v) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    class_int(unsigned v) : value(v) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    class_int(long v) : value(v) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    class_int(unsigned long v) : value(static_cast<std::int64_t>(v)) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    class_int(long long v) : value(v) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    class_int(unsigned long long v) : value(static_cast<std::int64_t>(v)) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
-    explicit class_int(double v) : value(static_cast<std::int64_t>(v)) {}
-
-    template<typename T, typename std::enable_if<std::is_arithmetic<T>::value, int>::type = 0>
-    explicit operator T() const
-    {
-        return static_cast<T>(value);
-    }
-
-    friend bool operator==(class_int lhs, class_int rhs)
-    {
-        return lhs.value == rhs.value;
-    }
-    friend bool operator!=(class_int lhs, class_int rhs)
-    {
-        return lhs.value != rhs.value;
-    }
-    friend bool operator<(class_int lhs, class_int rhs)
-    {
-        return lhs.value < rhs.value;
-    }
-#if JSON_HAS_THREE_WAY_COMPARISON
-    friend std::strong_ordering operator<=>(class_int lhs, class_int rhs) // *NOPAD*
-    {
-        return lhs.value <=> rhs.value; // *NOPAD*
-    }
-#endif
-
-  private:
-    std::int64_t value;
-};
-
-} // namespace custom_number_types
-
-using custom_number_types::class_int;
-
-namespace std
-{
-// only the members the library uses
-template<>
-class numeric_limits<class_int>
-{
-  public:
-    static constexpr bool is_signed = true;
-    static constexpr int digits = std::numeric_limits<std::int64_t>::digits;
-};
-} // namespace std
-
-namespace
-{
-
-using class_int_json = nlohmann::basic_json<std::map, std::vector, std::string, bool, class_int, std::uint64_t, double>;
-
-class_int_json make_class_int(std::int64_t v)
-{
-    class_int_json j(class_int_json::value_t::number_integer);
-    j.get_ref<class_int&>() = class_int(v);
-    return j;
-}
-
-} // namespace
-
-// the serializer cannot print an integer of class type, so doctest must not
-// try when an assertion fails
-namespace doctest
-{
-template<>
-struct StringMaker<class_int_json>
-{
-    static String convert(const class_int_json& j)
-    {
-        return j.type_name();
-    }
-};
-} // namespace doctest
-
-// __int128 as number type needs the std::numeric_limits (for the range checks)
-// and std::is_integral (for the serializer) specializations, which libc++
-// always provides and libstdc++ only outside strict ISO modes; the MSVC
-// standard library has none
-#if defined(__SIZEOF_INT128__) && (defined(_LIBCPP_VERSION) || defined(__GLIBCXX_TYPE_INT_N_0))
-    #define JSON_TEST_INT128_NUMBER_TYPES 1
-#else
-    #define JSON_TEST_INT128_NUMBER_TYPES 0
-#endif
-
-#if JSON_TEST_INT128_NUMBER_TYPES
-namespace
-{
-
-// __extension__ keeps -Wpedantic from flagging the non-standard type
-__extension__ typedef __int128 int128; // NOLINT(modernize-use-using)
-__extension__ typedef unsigned __int128 uint128; // NOLINT(modernize-use-using)
-
-static_assert(std::numeric_limits<int128>::digits == 127 && std::numeric_limits<uint128>::digits == 128 &&
-              std::is_integral<int128>::value && std::is_integral<uint128>::value, // NOLINT(misc-redundant-expression)
-              "__int128 is not fully supported by the standard library");
-
-using wide_json = nlohmann::basic_json<std::map, std::vector, std::string, bool, int128, uint128, double>;
-
-wide_json make_int(int128 v)
-{
-    wide_json j(wide_json::value_t::number_integer);
-    j.get_ref<int128&>() = v;
-    return j;
-}
-
-wide_json make_uint(uint128 v)
-{
-    wide_json j(wide_json::value_t::number_unsigned);
-    j.get_ref<uint128&>() = v;
-    return j;
-}
-
-wide_json wrap(const wide_json& value)
-{
-    wide_json o(wide_json::value_t::object);
-    o["v"] = value;
-    return o;
-}
-
-} // namespace
-
-TEST_CASE("custom number types: integers beyond 64 bits")
-{
-    using out_of_range = wide_json::out_of_range;
-
-    const int128 two_64 = static_cast<int128>(1) << 64;
-    const int128 two_100 = static_cast<int128>(1) << 100;
-    const int128 max_int64 = (std::numeric_limits<std::int64_t>::max)();
-    const int128 min_int64 = (std::numeric_limits<std::int64_t>::min)();
-    const uint128 max_uint64 = (std::numeric_limits<std::uint64_t>::max)();
-
-    // before the range checks, these values were silently truncated, e.g.,
-    // 2^100 was written as 0
-    const wide_json int_big = make_int(two_100);
-    const wide_json int_big_negative = make_int(-two_100);
-    const wide_json uint_big = make_uint(static_cast<uint128>(two_100));
-    std::vector<std::uint8_t> _;
-
-    SECTION("CBOR")
-    {
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(int_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(int_big_negative), "[json.exception.out_of_range.407] integer number -1267650600228229401496703205376 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(uint_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by CBOR as it does not fit uint64", out_of_range&);
-        // the extremes of the 128-bit types, whose digits are produced without
-        // negating the minimum as a whole
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(make_int((std::numeric_limits<int128>::min)())), "[json.exception.out_of_range.407] integer number -170141183460469231731687303715884105728 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(make_int((std::numeric_limits<int128>::max)())), "[json.exception.out_of_range.407] integer number 170141183460469231731687303715884105727 cannot be represented by CBOR as it does not fit [-2^64, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_cbor(make_uint((std::numeric_limits<uint128>::max)())), "[json.exception.out_of_range.407] integer number 340282366920938463463374607431768211455 cannot be represented by CBOR as it does not fit uint64", out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_cbor(make_int(two_64)), out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_cbor(make_int(-two_64 - 1)), out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_cbor(make_uint(static_cast<uint128>(max_uint64) + 1)), out_of_range&);
-
-        // CBOR's integers cover [-2^64, 2^64-1]
-        CHECK(wide_json::to_cbor(make_int(two_64 - 1)) == std::vector<std::uint8_t>({0x1B, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-        CHECK(wide_json::to_cbor(make_int(-two_64)) == std::vector<std::uint8_t>({0x3B, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-        CHECK(wide_json::to_cbor(make_uint(max_uint64)) == std::vector<std::uint8_t>({0x1B, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-    }
-
-    SECTION("MessagePack")
-    {
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_msgpack(int_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by MessagePack as it does not fit [-2^63, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_msgpack(int_big_negative), "[json.exception.out_of_range.407] integer number -1267650600228229401496703205376 cannot be represented by MessagePack as it does not fit [-2^63, 2^64-1]", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_msgpack(uint_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by MessagePack as it does not fit uint64", out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_msgpack(make_int(two_64)), out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_msgpack(make_int(min_int64 - 1)), out_of_range&);
-
-        // MessagePack's integers cover [-2^63, 2^64-1]
-        CHECK(wide_json::to_msgpack(make_int(two_64 - 1)) == std::vector<std::uint8_t>({0xCF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-        CHECK(wide_json::to_msgpack(make_int(min_int64)) == std::vector<std::uint8_t>({0xD3, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}));
-        CHECK(wide_json::to_msgpack(make_uint(max_uint64)) == std::vector<std::uint8_t>({0xCF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-    }
-
-    SECTION("BSON")
-    {
-        // wrap() stores the value at /v, which JSON_DIAGNOSTICS adds to the message
-#if JSON_DIAGNOSTICS
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(int_big)), "[json.exception.out_of_range.407] (/v) integer number 1267650600228229401496703205376 cannot be represented by BSON as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(int_big_negative)), "[json.exception.out_of_range.407] (/v) integer number -1267650600228229401496703205376 cannot be represented by BSON as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(uint_big)), "[json.exception.out_of_range.407] (/v) integer number 1267650600228229401496703205376 cannot be represented by BSON as it does not fit uint64", out_of_range&);
-#else
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(int_big)), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by BSON as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(int_big_negative)), "[json.exception.out_of_range.407] integer number -1267650600228229401496703205376 cannot be represented by BSON as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bson(wrap(uint_big)), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by BSON as it does not fit uint64", out_of_range&);
-#endif
-        CHECK_THROWS_AS(_ = wide_json::to_bson(wrap(make_int(max_int64 + 1))), out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_bson(wrap(make_int(min_int64 - 1))), out_of_range&);
-
-        // nested values are checked as well, before anything is written
-        wide_json nested(wide_json::value_t::object);
-        nested["a"] = wide_json::array({wrap(int_big)});
-        std::vector<std::uint8_t> out;
-        CHECK_THROWS_AS(wide_json::to_bson(nested, out), out_of_range&);
-        CHECK(out.empty());
-
-        CHECK(wide_json::from_bson(wide_json::to_bson(wrap(make_int(max_int64)))) == wrap(make_int(max_int64)));
-        CHECK(wide_json::from_bson(wide_json::to_bson(wrap(make_int(min_int64)))) == wrap(make_int(min_int64)));
-        CHECK_NOTHROW(wide_json::to_bson(wrap(make_uint(max_uint64))));
-    }
-
-    SECTION("BON8")
-    {
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bon8(int_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by BON8 as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bon8(int_big_negative), "[json.exception.out_of_range.407] integer number -1267650600228229401496703205376 cannot be represented by BON8 as it does not fit int64", out_of_range&);
-        CHECK_THROWS_WITH_AS(_ = wide_json::to_bon8(uint_big), "[json.exception.out_of_range.407] integer number 1267650600228229401496703205376 cannot be represented by BON8 as it does not fit int64", out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_bon8(make_int(max_int64 + 1)), out_of_range&);
-        CHECK_THROWS_AS(_ = wide_json::to_bon8(make_int(min_int64 - 1)), out_of_range&);
-
-        CHECK(wide_json::from_bon8(wide_json::to_bon8(make_int(max_int64))) == make_int(max_int64));
-        CHECK(wide_json::from_bon8(wide_json::to_bon8(make_int(min_int64))) == make_int(min_int64));
-    }
-
-    SECTION("UBJSON and BJData")
-    {
-        // integers beyond 64 bits are written exactly as high-precision numbers
-        const std::string digits = "1267650600228229401496703205376";
-        std::vector<std::uint8_t> expected = {'H', 'i', static_cast<std::uint8_t>(digits.size())};
-        expected.insert(expected.end(), digits.begin(), digits.end());
-        CHECK(wide_json::to_ubjson(int_big) == expected);
-        CHECK(wide_json::to_ubjson(uint_big) == expected);
-        CHECK(wide_json::to_bjdata(int_big) == expected);
-        // BJData's uint64 marker 'M' was used for any unsigned value beyond
-        // int64, which truncated the ones beyond 64 bits
-        CHECK(wide_json::to_bjdata(uint_big) == expected);
-
-        // the uint64 marker is still used where the value fits
-        CHECK(wide_json::to_bjdata(make_uint(max_uint64)) == std::vector<std::uint8_t>({'M', 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF}));
-
-        // the elements of an optimized container are written the same way;
-        // BJData does not allow 'H' as the type of an optimized container
-        std::vector<std::uint8_t> expected_ubjson_array = {'[', '$', 'H', '#', 'i', 2};
-        std::vector<std::uint8_t> expected_bjdata_array = {'[', '#', 'i', 2};
-        for (int i = 0; i < 2; ++i)
-        {
-            expected_ubjson_array.insert(expected_ubjson_array.end(), expected.begin() + 1, expected.end());
-            expected_bjdata_array.insert(expected_bjdata_array.end(), expected.begin(), expected.end());
-        }
-        CHECK(wide_json::to_ubjson(wide_json::array({uint_big, uint_big}), true, true) == expected_ubjson_array);
-        CHECK(wide_json::to_bjdata(wide_json::array({uint_big, uint_big}), true, true) == expected_bjdata_array);
-    }
-
-    SECTION("BJData ND-array")
-    {
-        // an ND-array element beyond 64 bits does not fit any dtype, so the
-        // annotated object is written as a plain object instead of truncating
-        // the element into range
-        wide_json element(wide_json::value_t::object);
-        element["_ArrayType_"] = "uint8";
-        element["_ArraySize_"] = wide_json::array({make_uint(2), make_uint(1)});
-        element["_ArrayData_"] = wide_json::array({make_uint(1), make_uint(static_cast<uint128>(two_64) + 1)});
-        const auto element_bytes = wide_json::to_bjdata(element, true, true);
-        REQUIRE(!element_bytes.empty());
-        CHECK(element_bytes[0] == '{');
-
-        wide_json signed_element = element;
-        signed_element["_ArrayType_"] = "int8";
-        signed_element["_ArrayData_"] = wide_json::array({make_int(1), make_int(-two_64 + 1)});
-        const auto signed_element_bytes = wide_json::to_bjdata(signed_element, true, true);
-        REQUIRE(!signed_element_bytes.empty());
-        CHECK(signed_element_bytes[0] == '{');
-
-        // the same for a dimension beyond 64 bits, which wrapped into a
-        // dimension matching the size of _ArrayData_ (here: 1)
-        wide_json dimension = element;
-        dimension["_ArraySize_"] = wide_json::array({make_uint(2), make_uint(static_cast<uint128>(two_64) + 1)});
-        dimension["_ArrayData_"] = wide_json::array({make_uint(1), make_uint(2)});
-        const auto dimension_bytes = wide_json::to_bjdata(dimension, true, true);
-        REQUIRE(!dimension_bytes.empty());
-        CHECK(dimension_bytes[0] == '{');
-
-        // in range, the ND-array is still written
-        wide_json fits = element;
-        fits["_ArrayData_"] = wide_json::array({make_uint(1), make_uint(2)});
-        const auto fits_bytes = wide_json::to_bjdata(fits, true, true);
-        REQUIRE(!fits_bytes.empty());
-        CHECK(fits_bytes[0] == '[');
-    }
-}
-
-#endif
-
-TEST_CASE("custom number types")
-{
-    SECTION("signed integer of class type compared with a float")
-    {
-        // std::is_signed is false for a class type, which made the comparison
-        // treat any float below zero as less than every integer
-        const class_int_json minus_five = make_class_int(-5);
-        const class_int_json minus_two = make_class_int(-2);
-        const class_int_json five = make_class_int(5);
-        const class_int_json minus_two_and_a_half = -2.5;
-        const class_int_json two_and_a_half = 2.5;
-        const class_int_json huge_negative = -1e30;
-        const class_int_json huge_positive = 1e30;
-
-        CHECK(minus_five < minus_two_and_a_half);
-        CHECK_FALSE(minus_two_and_a_half < minus_five);
-        CHECK(minus_two_and_a_half > minus_five);
-        CHECK(minus_five <= minus_two_and_a_half);
-        CHECK(minus_two_and_a_half >= minus_five);
-        CHECK(minus_five != minus_two_and_a_half);
-
-        CHECK(minus_two_and_a_half < minus_two);
-        CHECK_FALSE(minus_two < minus_two_and_a_half);
-
-        CHECK(two_and_a_half < five);
-        CHECK(minus_five < two_and_a_half);
-
-        // floats beyond the integer's range
-        CHECK(huge_negative < minus_five);
-        CHECK_FALSE(minus_five < huge_negative);
-        CHECK(five < huge_positive);
-        CHECK_FALSE(huge_positive < five);
-
-        // equality
-        const class_int_json minus_two_float = -2.0;
-        CHECK(minus_two == minus_two_float);
-        CHECK(minus_two_float == minus_two);
-    }
-
-}

tests/src/unit-msgpack-custom-types.cpp

@@ -1,296 +0,0 @@
-//     __ _____ _____ _____
-//  __|  |   __|     |   | |  JSON for Modern C++ (supporting code)
-// |  |  |__   |  |  | | | |  version 3.12.0
-// |_____|_____|_____|_|___|  https://github.com/nlohmann/json
-//
-// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
-// SPDX-License-Identifier: MIT
-
-// MessagePack tests with basic_json specializations other than nlohmann::json;
-// they are kept apart from unit-msgpack.cpp because every specialization adds
-// many sections to the object file, and MinGW cannot link objects with more
-// than 65535 sections (see the clang job in .github/workflows/windows.yml)
-
-#include "doctest_compatibility.h"
-
-#include <nlohmann/json.hpp>
-using nlohmann::json;
-
-#include <cstdint> // SIZE_MAX, UINT32_MAX
-#include <functional>
-#include <limits>
-#include <map>
-#include <memory>
-#include <string>
-#include <utility>
-#include <vector>
-
-// the fake sizes below do not fit into a 32-bit std::size_t
-// with clang and libstdc++ 10, the std::filesystem::path conversion that
-// C++17 builds consider for every string type is ambiguous for a class
-// derived from std::string, so the string case is not tested there
-#if !(defined(__clang__) && defined(_GLIBCXX_RELEASE) && _GLIBCXX_RELEASE < 11)
-    #define JSON_TEST_BEYOND_UINT32_STRING 1
-#endif
-
-#if SIZE_MAX > UINT32_MAX
-template<typename T, typename A = std::allocator<T>>
-struct huge_array : std::vector<T, A>
-{
-    using base = std::vector<T, A>;
-    using base::base;
-
-    bool fake_size = false;
-
-    std::size_t size() const noexcept
-    {
-        if (fake_size)
-        {
-            return (std::numeric_limits<std::uint32_t>::max)() + 1ULL;
-        }
-
-        return base::size();
-    }
-};
-
-using huge_array_json = nlohmann::json::with_array_t<huge_array>;
-
-TEST_CASE("MessagePack Size above uint32 for array")
-{
-    huge_array_json j = huge_array_json::array();
-
-    j.push_back(1);
-    j.push_back(2);
-    j.push_back(3);
-
-    auto& array = j.get_ref<huge_array_json::array_t&>();
-    array.fake_size = true;
-
-    // write into a caller-owned vector: to_msgpack(j) reserves space based on
-    // the (faked) element count, which fails with bad_alloc on Windows
-    std::vector<std::uint8_t> result;
-    CHECK_THROWS_WITH_AS(
-        huge_array_json::to_msgpack(j, result),
-        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
-        json::out_of_range&);
-
-    array.fake_size = false;
-}
-
-template<typename K, typename V,
-         typename C = std::less<K>,
-         typename A = std::allocator<std::pair<const K, V>>>
-                                     struct huge_map : std::map<K, V, C, A>
-{
-    using base = std::map<K, V, C, A>;
-    using base::base;
-
-    bool fake_size = false;
-
-    std::size_t size() const noexcept
-    {
-        if (fake_size)
-        {
-            return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
-        }
-
-        return base::size();
-    }
-};
-
-using huge_object_json = nlohmann::json::with_object_t<huge_map>;
-
-TEST_CASE("MessagePack Size above uint32 for object")
-{
-
-    huge_object_json j = huge_object_json::object();
-
-    j["one"] = 1;
-    j["two"] = 2;
-
-    auto& object = j.get_ref<huge_object_json::object_t&>();
-    object.fake_size = true;
-
-    // write into a caller-owned vector: to_msgpack(j) reserves space based on
-    // the (faked) element count, which fails with bad_alloc on Windows
-    std::vector<std::uint8_t> result;
-    CHECK_THROWS_WITH_AS(
-        huge_object_json::to_msgpack(j, result),
-        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
-        json::out_of_range&);
-
-    object.fake_size = false;
-}
-
-#ifdef JSON_TEST_BEYOND_UINT32_STRING
-struct huge_string : std::string
-{
-    using std::string::string;
-
-    std::size_t size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
-    {
-        return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
-    }
-};
-
-using huge_string_json = nlohmann::json::with_string_t<huge_string>;
-
-TEST_CASE("MessagePack Size above uint32 for string")
-{
-    const huge_string_json j = "hello";
-
-    CHECK_THROWS_WITH_AS(
-        huge_string_json::to_msgpack(j),
-        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
-        json::out_of_range&);
-}
-#endif
-
-struct huge_binary : std::vector<std::uint8_t>
-{
-    using std::vector<std::uint8_t>::vector;
-
-    std::size_t size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
-    {
-        return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
-    }
-};
-
-using huge_binary_json = nlohmann::json::with_binary_t<huge_binary>;
-
-TEST_CASE("MessagePack Size above uint32 for binary")
-{
-
-    huge_binary_json j = huge_binary_json::binary(huge_binary{});
-
-    j.get_binary().push_back(0x01);
-    j.get_binary().push_back(0x02);
-
-    CHECK_THROWS_WITH_AS(
-        huge_binary_json::to_msgpack(j),
-        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
-        json::out_of_range&);
-}
-#endif
-
-namespace
-{
-// types that report a size beyond UINT32_MAX without allocating that much
-// memory, so the MessagePack length limit can be tested cheaply; see the
-// similar types in unit-bson.cpp
-std::size_t beyond_uint32_size()
-{
-    return static_cast<std::size_t>((std::numeric_limits<std::uint32_t>::max)()) + 1;
-}
-
-class beyond_uint32_binary_t : public std::vector<std::uint8_t>
-{
-  public:
-    using std::vector<std::uint8_t>::vector;
-
-    size_type size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
-    {
-        return beyond_uint32_size();
-    }
-};
-
-#ifdef JSON_TEST_BEYOND_UINT32_STRING
-class beyond_uint32_string_t : public std::string
-{
-  public:
-    using std::string::string;
-
-    size_type size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
-    {
-        return beyond_uint32_size();
-    }
-};
-
-using beyond_uint32_string_json = nlohmann::json::with_string_t<beyond_uint32_string_t>;
-#endif
-
-using beyond_uint32_binary_json = nlohmann::json::with_binary_t<beyond_uint32_binary_t>;
-} // namespace
-
-TEST_CASE("MessagePack lengths beyond UINT32_MAX cannot be serialized")
-{
-    // MessagePack stores the length of a string, binary value, array, or
-    // object in at most 32 bits; a larger one used to be written without any
-    // length at all
-#if SIZE_MAX > UINT32_MAX
-    {
-        const char* const expected = "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295";
-
-        const beyond_uint32_binary_json binary = beyond_uint32_binary_json::binary(beyond_uint32_binary_t{});
-        CHECK_THROWS_WITH_AS(beyond_uint32_binary_json::to_msgpack(binary), expected, beyond_uint32_binary_json::out_of_range&);
-
-        const beyond_uint32_binary_json ext = beyond_uint32_binary_json::binary(beyond_uint32_binary_t{}, 42);
-        CHECK_THROWS_WITH_AS(beyond_uint32_binary_json::to_msgpack(ext), expected, beyond_uint32_binary_json::out_of_range&);
-
-#ifdef JSON_TEST_BEYOND_UINT32_STRING
-        // created from its type rather than from a beyond_uint32_string_t:
-        // that would consider the std::filesystem::path conversion, which
-        // libstdc++ 10 cannot decide for a class derived from std::string
-        const beyond_uint32_string_json string(beyond_uint32_string_json::value_t::string);
-        CHECK_THROWS_WITH_AS(beyond_uint32_string_json::to_msgpack(string), expected, beyond_uint32_string_json::out_of_range&);
-#endif
-    }
-#endif
-}
-
-TEST_CASE("MessagePack numbers use the active union member (see #5644)")
-{
-    // when number_integer_t is narrower than number_unsigned_t, to_msgpack()
-    // used to read the union member that was not the active one, writing
-    // wrong bytes for some values; std::int64_t/std::uint64_t (the default
-    // types, where both members have the same width) were not affected
-    using int32_json = nlohmann::json::with_integers_t<std::int32_t, std::uint64_t>;
-    using int16_json = nlohmann::json::with_integers_t<std::int16_t, std::uint64_t>;
-
-    SECTION("number_integer_t = std::int32_t")
-    {
-        SECTION("6442450944 (uint 64; the low 32 bits used to be sign-extended)")
-        {
-            const int32_json j = 6442450944ULL;
-            CHECK(j.is_number_unsigned());
-
-            std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x80, 0x00, 0x00, 0x00};
-            const auto result = int32_json::to_msgpack(j);
-            CHECK(result == expected);
-            CHECK(int32_json::from_msgpack(result) == j);
-        }
-
-        SECTION("4294967496 (uint 64; the low 32 bits used to be the whole value)")
-        {
-            const int32_json j = 4294967496ULL;
-            CHECK(j.is_number_unsigned());
-
-            std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0xc8};
-            const auto result = int32_json::to_msgpack(j);
-            CHECK(result == expected);
-            CHECK(int32_json::from_msgpack(result) == j);
-        }
-    }
-
-    SECTION("number_integer_t = std::int16_t, 98304 (uint 32)")
-    {
-        const int16_json j = 98304ULL;
-        CHECK(j.is_number_unsigned());
-
-        std::vector<uint8_t> const expected{0xce, 0x00, 0x01, 0x80, 0x00};
-        const auto result = int16_json::to_msgpack(j);
-        CHECK(result == expected);
-        CHECK(int16_json::from_msgpack(result) == j);
-    }
-
-    SECTION("default types (std::int64_t/std::uint64_t) are unaffected")
-    {
-        const json j = 4294967496ULL;
-        CHECK(j.is_number_unsigned());
-
-        std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0xc8};
-        const auto result = json::to_msgpack(j);
-        CHECK(result == expected);
-        CHECK(json::from_msgpack(result) == j);
-    }
-}

tests/src/unit-msgpack.cpp

@@ -2173,6 +2173,276 @@ TEST_CASE("MessagePack roundtrips" * doctest::skip())
     }
 }
 
+// the fake sizes below do not fit into a 32-bit std::size_t
+// with clang and libstdc++ 10, the std::filesystem::path conversion that
+// C++17 builds consider for every string type is ambiguous for a class
+// derived from std::string, so the string case is not tested there
+#if !(defined(__clang__) && defined(_GLIBCXX_RELEASE) && _GLIBCXX_RELEASE < 11)
+    #define JSON_TEST_BEYOND_UINT32_STRING 1
+#endif
+
+#if SIZE_MAX > UINT32_MAX
+template<typename T, typename A = std::allocator<T>>
+struct huge_array : std::vector<T, A>
+{
+    using base = std::vector<T, A>;
+    using base::base;
+
+    bool fake_size = false;
+
+    std::size_t size() const noexcept
+    {
+        if (fake_size)
+        {
+            return (std::numeric_limits<std::uint32_t>::max)() + 1ULL;
+        }
+
+        return base::size();
+    }
+};
+
+using huge_array_json = nlohmann::json::with_array_t<huge_array>;
+
+TEST_CASE("MessagePack Size above uint32 for array")
+{
+    huge_array_json j = huge_array_json::array();
+
+    j.push_back(1);
+    j.push_back(2);
+    j.push_back(3);
+
+    auto& array = j.get_ref<huge_array_json::array_t&>();
+    array.fake_size = true;
+
+    // write into a caller-owned vector: to_msgpack(j) reserves space based on
+    // the (faked) element count, which fails with bad_alloc on Windows
+    std::vector<std::uint8_t> result;
+    CHECK_THROWS_WITH_AS(
+        huge_array_json::to_msgpack(j, result),
+        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
+        json::out_of_range&);
+
+    array.fake_size = false;
+}
+
+template<typename K, typename V,
+         typename C = std::less<K>,
+         typename A = std::allocator<std::pair<const K, V>>>
+                                     struct huge_map : std::map<K, V, C, A>
+{
+    using base = std::map<K, V, C, A>;
+    using base::base;
+
+    bool fake_size = false;
+
+    std::size_t size() const noexcept
+    {
+        if (fake_size)
+        {
+            return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
+        }
+
+        return base::size();
+    }
+};
+
+using huge_object_json = nlohmann::json::with_object_t<huge_map>;
+
+TEST_CASE("MessagePack Size above uint32 for object")
+{
+
+    huge_object_json j = huge_object_json::object();
+
+    j["one"] = 1;
+    j["two"] = 2;
+
+    auto& object = j.get_ref<huge_object_json::object_t&>();
+    object.fake_size = true;
+
+    // write into a caller-owned vector: to_msgpack(j) reserves space based on
+    // the (faked) element count, which fails with bad_alloc on Windows
+    std::vector<std::uint8_t> result;
+    CHECK_THROWS_WITH_AS(
+        huge_object_json::to_msgpack(j, result),
+        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
+        json::out_of_range&);
+
+    object.fake_size = false;
+}
+
+#ifdef JSON_TEST_BEYOND_UINT32_STRING
+struct huge_string : std::string
+{
+    using std::string::string;
+
+    std::size_t size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
+    {
+        return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
+    }
+};
+
+using huge_string_json = nlohmann::json::with_string_t<huge_string>;
+
+TEST_CASE("MessagePack Size above uint32 for string")
+{
+    const huge_string_json j = "hello";
+
+    CHECK_THROWS_WITH_AS(
+        huge_string_json::to_msgpack(j),
+        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
+        json::out_of_range&);
+}
+#endif
+
+struct huge_binary : std::vector<std::uint8_t>
+{
+    using std::vector<std::uint8_t>::vector;
+
+    std::size_t size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
+    {
+        return static_cast<std::size_t>(UINT32_MAX) + 1ULL;
+    }
+};
+
+using huge_binary_json = nlohmann::json::with_binary_t<huge_binary>;
+
+TEST_CASE("MessagePack Size above uint32 for binary")
+{
+
+    huge_binary_json j = huge_binary_json::binary(huge_binary{});
+
+    j.get_binary().push_back(0x01);
+    j.get_binary().push_back(0x02);
+
+    CHECK_THROWS_WITH_AS(
+        huge_binary_json::to_msgpack(j),
+        "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295",
+        json::out_of_range&);
+}
+#endif
+
+namespace
+{
+// types that report a size beyond UINT32_MAX without allocating that much
+// memory, so the MessagePack length limit can be tested cheaply; see the
+// similar types in unit-bson.cpp
+std::size_t beyond_uint32_size()
+{
+    return static_cast<std::size_t>((std::numeric_limits<std::uint32_t>::max)()) + 1;
+}
+
+class beyond_uint32_binary_t : public std::vector<std::uint8_t>
+{
+  public:
+    using std::vector<std::uint8_t>::vector;
+
+    size_type size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
+    {
+        return beyond_uint32_size();
+    }
+};
+
+#ifdef JSON_TEST_BEYOND_UINT32_STRING
+class beyond_uint32_string_t : public std::string
+{
+  public:
+    using std::string::string;
+
+    size_type size() const noexcept // NOLINT(readability-convert-member-functions-to-static)
+    {
+        return beyond_uint32_size();
+    }
+};
+
+using beyond_uint32_string_json = nlohmann::json::with_string_t<beyond_uint32_string_t>;
+#endif
+
+using beyond_uint32_binary_json = nlohmann::json::with_binary_t<beyond_uint32_binary_t>;
+} // namespace
+
+TEST_CASE("MessagePack lengths beyond UINT32_MAX cannot be serialized")
+{
+    // MessagePack stores the length of a string, binary value, array, or
+    // object in at most 32 bits; a larger one used to be written without any
+    // length at all
+#if SIZE_MAX > UINT32_MAX
+    {
+        const char* const expected = "[json.exception.out_of_range.412] MessagePack length 4294967296 exceeds maximum of 4294967295";
+
+        const beyond_uint32_binary_json binary = beyond_uint32_binary_json::binary(beyond_uint32_binary_t{});
+        CHECK_THROWS_WITH_AS(beyond_uint32_binary_json::to_msgpack(binary), expected, beyond_uint32_binary_json::out_of_range&);
+
+        const beyond_uint32_binary_json ext = beyond_uint32_binary_json::binary(beyond_uint32_binary_t{}, 42);
+        CHECK_THROWS_WITH_AS(beyond_uint32_binary_json::to_msgpack(ext), expected, beyond_uint32_binary_json::out_of_range&);
+
+#ifdef JSON_TEST_BEYOND_UINT32_STRING
+        // created from its type rather than from a beyond_uint32_string_t:
+        // that would consider the std::filesystem::path conversion, which
+        // libstdc++ 10 cannot decide for a class derived from std::string
+        const beyond_uint32_string_json string(beyond_uint32_string_json::value_t::string);
+        CHECK_THROWS_WITH_AS(beyond_uint32_string_json::to_msgpack(string), expected, beyond_uint32_string_json::out_of_range&);
+#endif
+    }
+#endif
+}
+
+TEST_CASE("MessagePack numbers use the active union member (see #5644)")
+{
+    // when number_integer_t is narrower than number_unsigned_t, to_msgpack()
+    // used to read the union member that was not the active one, writing
+    // wrong bytes for some values; std::int64_t/std::uint64_t (the default
+    // types, where both members have the same width) were not affected
+    using int32_json = nlohmann::json::with_integers_t<std::int32_t, std::uint64_t>;
+    using int16_json = nlohmann::json::with_integers_t<std::int16_t, std::uint64_t>;
+
+    SECTION("number_integer_t = std::int32_t")
+    {
+        SECTION("6442450944 (uint 64; the low 32 bits used to be sign-extended)")
+        {
+            const int32_json j = 6442450944ULL;
+            CHECK(j.is_number_unsigned());
+
+            std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x80, 0x00, 0x00, 0x00};
+            const auto result = int32_json::to_msgpack(j);
+            CHECK(result == expected);
+            CHECK(int32_json::from_msgpack(result) == j);
+        }
+
+        SECTION("4294967496 (uint 64; the low 32 bits used to be the whole value)")
+        {
+            const int32_json j = 4294967496ULL;
+            CHECK(j.is_number_unsigned());
+
+            std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0xc8};
+            const auto result = int32_json::to_msgpack(j);
+            CHECK(result == expected);
+            CHECK(int32_json::from_msgpack(result) == j);
+        }
+    }
+
+    SECTION("number_integer_t = std::int16_t, 98304 (uint 32)")
+    {
+        const int16_json j = 98304ULL;
+        CHECK(j.is_number_unsigned());
+
+        std::vector<uint8_t> const expected{0xce, 0x00, 0x01, 0x80, 0x00};
+        const auto result = int16_json::to_msgpack(j);
+        CHECK(result == expected);
+        CHECK(int16_json::from_msgpack(result) == j);
+    }
+
+    SECTION("default types (std::int64_t/std::uint64_t) are unaffected")
+    {
+        const json j = 4294967496ULL;
+        CHECK(j.is_number_unsigned());
+
+        std::vector<uint8_t> const expected{0xcf, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0xc8};
+        const auto result = json::to_msgpack(j);
+        CHECK(result == expected);
+        CHECK(json::from_msgpack(result) == j);
+    }
+}
+
 TEST_CASE("MessagePack large strings and binaries (chunked reader)")
 {
     // get_msgpack_string()/get_msgpack_binary() both read through get_binary(),

tests/src/unit-regression3.cpp

@@ -43,6 +43,7 @@ using ordered_json = nlohmann::ordered_json;
 #endif
 
 #include <cstdio>
+#include <deque>
 #include <list>
 #include <type_traits>
 #include <utility>
@@ -866,6 +867,52 @@ TEST_CASE("issue #5402 - update(merge_objects=true) overwrites a primitive with
 }
 
 
+TEST_CASE("regression test #5476 - array type without reserve()")
+{
+    // the capacity reserved for definite-length arrays must not require the
+    // array type to have a reserve() member function
+    using deque_json = nlohmann::json::with_array_t<std::deque>;
+
+    SECTION("std::deque")
+    {
+        const auto j = deque_json::parse(R"({"a":[1,[2,3]],"b":[]})");
+        CHECK(j.dump() == R"({"a":[1,[2,3]],"b":[]})");
+
+        // the binary formats pass a definite length to start_array()
+        CHECK(deque_json::from_cbor(deque_json::to_cbor(j)) == j);
+        CHECK(deque_json::from_msgpack(deque_json::to_msgpack(j)) == j);
+        CHECK(deque_json::from_bon8(deque_json::to_bon8(j)) == j);
+
+        // parse() instantiates the callback parser as well, which reserves too
+        const auto with_callback = deque_json::parse(R"([1,2,3])", [](int /*depth*/, deque_json::parse_event_t /*event*/, deque_json& /*parsed*/) noexcept
+        {
+            return true;
+        });
+        CHECK(with_callback == deque_json({1, 2, 3}));
+    }
+
+    SECTION("std::vector still reserves")
+    {
+        json array = json::array();
+        for (int i = 0; i < 100; ++i)
+        {
+            array.push_back(i);
+        }
+
+        const auto j = json::from_cbor(json::to_cbor(array));
+        CHECK(j == array);
+        CHECK(j.get_ref<const json::array_t&>().capacity() >= 100);
+    }
+
+    SECTION("the reservation stays capped")
+    {
+        // CBOR array announcing 2^32-1 elements, but truncated right after the
+        // header: the input must be rejected without reserving that capacity
+        const std::vector<std::uint8_t> truncated = {0x9A, 0xFF, 0xFF, 0xFF, 0xFF};
+        CHECK(json::from_cbor(truncated, true, false).is_discarded());
+    }
+}
+
 TEST_CASE("issue #5317 - nested indefinite-length CBOR string chunks are rejected")
 {
     json _;