Compare commits

..
Author SHA1 Message Date
Niels Lohmann ea04c24fdc Fall back to plain-object encoding when _ArrayType_ is not a string
write_bjdata_ndarray() looked up _ArrayType_ by calling get<string_t>()
directly, which throws type_error.302 when the annotation is not a
string (e.g. a number, null, boolean, array, or object). Per the
documented BJData ndarray contract, an object only qualifies for the
compact ndarray encoding if _ArrayType_ names a known type; anything
else must fall back to plain-object encoding, the same way an unknown
type-name string already does.

Add an is_string() check before the get<string_t>() call so a
non-string _ArrayType_ takes the existing "unrecognized type name"
fallback path instead of throwing.

Fixes #5398.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
2026-09-05 21:54:45 +02:00
Niels Lohmann 50e392ab1a Fix to_bjdata() emitting the Draft-3-only 'B' marker in default Draft-2 mode
_ArrayType_ = "byte" mapped unconditionally to the BJData type marker
'B', regardless of the requested bjdata_version. 'B' is defined only by
BJData Draft 3; with the default version (draft2), this produced a
stream that is invalid for Draft 2 and, unlike every other
_ArrayType_, round-tripped back as a binary value instead of the
original annotated object.

Only accept "byte" / emit 'B' when bjdata_version selects Draft 3.
Under Draft 2, fall back to the same plain-object encoding used
elsewhere in this function for other invalid-annotation cases, so the
value round-trips correctly.

Fixes #5404.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
2026-09-05 20:52:54 +02:00
Niels Lohmann d2c2db92a9 Fix to_bjdata() silently truncating out-of-range _ArrayData_ elements
write_bjdata_ndarray() validated that each _ArrayData_ element matched
the number kind (integer vs. float) named by _ArrayType_, but not its
range. An element that did not fit the target C++ type (e.g. 256 for
"uint8") was silently wrapped by the static_cast used to write it, or,
for "single", silently overflowed to infinity.

Range-check each element against the type named by _ArrayType_ before
writing it, reusing the existing fallback path that already encodes
the annotated object as a plain object for other invalid-annotation
cases in this function.

Fixes #5403.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
2026-09-05 20:43:55 +02:00
5 changed files with 307 additions and 137 deletions
@@ -1647,6 +1647,20 @@ class binary_writer
return 'D'; // float 64
}
/*!
@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
*/
template<typename TargetType>
static bool bjdata_ndarray_value_in_range(const BasicJsonType& el)
{
return el.is_number_unsigned()
? value_in_range_of<TargetType>(el.template get<std::uint64_t>())
: value_in_range_of<TargetType>(el.template get<std::int64_t>());
}
/*!
@return false if the object is successfully converted to a bjdata ndarray, true if the type or size is invalid
*/
@@ -1658,6 +1672,16 @@ class binary_writer
};
string_t key = "_ArrayType_";
// the type name is looked up as a string below; a non-string
// annotation (e.g. a number, null, or an array) cannot name a known
// dtype, so it is treated the same as an unrecognized type name and
// falls back to a plain object encoding instead of throwing
// type_error.302 out of get<string_t>()
if (!value.at(key).is_string())
{
return true;
}
// use get<string_t>() instead of static_cast<string_t> to avoid an
// ambiguous conversion under explicit instantiation on C++17 (see #4825)
auto it = bjdtype.find(value.at(key).template get<string_t>());
@@ -1667,6 +1691,16 @@ class binary_writer
}
CharType dtype = it->second;
// the 'B' (byte) marker is only defined by BJData Draft 3; emitting it
// under the default Draft 2 mode would produce a stream that Draft 2
// readers reject, so such an object falls back to a plain object
// encoding instead (see the "Binary values" section of the BJData
// documentation)
if (dtype == 'B' && bjdata_version != bjdata_version_t::draft3)
{
return true;
}
key = "_ArraySize_";
// the dimensions are written verbatim as the header length below, so a
// value that is not an array cannot produce a valid one: null emits 'Z'
@@ -1731,6 +1765,60 @@ class binary_writer
}
}
// every element is cast to the (possibly narrower) C++ type matching
// dtype below; a value that does not fit that type would silently
// wrap (integers) or overflow to infinity (the "single" precision
// float) instead of being reported, so such an object falls back to
// a plain object encoding as well
for (const auto& el : value.at(key))
{
bool in_range = true;
switch (dtype)
{
case 'U':
case 'C':
case 'B':
in_range = bjdata_ndarray_value_in_range<std::uint8_t>(el);
break;
case 'i':
in_range = bjdata_ndarray_value_in_range<std::int8_t>(el);
break;
case 'u':
in_range = bjdata_ndarray_value_in_range<std::uint16_t>(el);
break;
case 'I':
in_range = bjdata_ndarray_value_in_range<std::int16_t>(el);
break;
case 'm':
in_range = bjdata_ndarray_value_in_range<std::uint32_t>(el);
break;
case 'l':
in_range = bjdata_ndarray_value_in_range<std::int32_t>(el);
break;
case 'M':
in_range = bjdata_ndarray_value_in_range<std::uint64_t>(el);
break;
case 'L':
in_range = bjdata_ndarray_value_in_range<std::int64_t>(el);
break;
case 'd':
{
const auto dval = el.template get<double>();
in_range = !std::isfinite(dval) ||
(dval >= static_cast<double>(std::numeric_limits<float>::lowest()) &&
dval <= static_cast<double>((std::numeric_limits<float>::max)()));
break;
}
default:
// 'D' (double) already spans the full range of number_float_t
break;
}
if (!in_range)
{
return true;
}
}
oa->write_character('[');
oa->write_character('$');
oa->write_character(dtype);
+2 -14
View File
@@ -71,7 +71,7 @@ class serializer
, thousands_sep(loc->thousands_sep == nullptr ? '\0' : std::char_traits<char>::to_char_type(* (loc->thousands_sep)))
, decimal_point(loc->decimal_point == nullptr ? '\0' : std::char_traits<char>::to_char_type(* (loc->decimal_point)))
, indent_char(ichar)
, indent_string()
, indent_string(512, indent_char)
, error_handler(error_handler_)
{}
@@ -126,10 +126,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -203,10 +199,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -268,10 +260,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -1022,7 +1010,7 @@ class serializer
/// the indentation character
const char indent_char;
/// the indentation string (lazily allocated on first use by a pretty-print branch)
/// the indentation string
string_t indent_string;
/// error_handler how to react on decoding errors
+90 -14
View File
@@ -18655,6 +18655,20 @@ class binary_writer
return 'D'; // float 64
}
/*!
@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
*/
template<typename TargetType>
static bool bjdata_ndarray_value_in_range(const BasicJsonType& el)
{
return el.is_number_unsigned()
? value_in_range_of<TargetType>(el.template get<std::uint64_t>())
: value_in_range_of<TargetType>(el.template get<std::int64_t>());
}
/*!
@return false if the object is successfully converted to a bjdata ndarray, true if the type or size is invalid
*/
@@ -18666,6 +18680,16 @@ class binary_writer
};
string_t key = "_ArrayType_";
// the type name is looked up as a string below; a non-string
// annotation (e.g. a number, null, or an array) cannot name a known
// dtype, so it is treated the same as an unrecognized type name and
// falls back to a plain object encoding instead of throwing
// type_error.302 out of get<string_t>()
if (!value.at(key).is_string())
{
return true;
}
// use get<string_t>() instead of static_cast<string_t> to avoid an
// ambiguous conversion under explicit instantiation on C++17 (see #4825)
auto it = bjdtype.find(value.at(key).template get<string_t>());
@@ -18675,6 +18699,16 @@ class binary_writer
}
CharType dtype = it->second;
// the 'B' (byte) marker is only defined by BJData Draft 3; emitting it
// under the default Draft 2 mode would produce a stream that Draft 2
// readers reject, so such an object falls back to a plain object
// encoding instead (see the "Binary values" section of the BJData
// documentation)
if (dtype == 'B' && bjdata_version != bjdata_version_t::draft3)
{
return true;
}
key = "_ArraySize_";
// the dimensions are written verbatim as the header length below, so a
// value that is not an array cannot produce a valid one: null emits 'Z'
@@ -18739,6 +18773,60 @@ class binary_writer
}
}
// every element is cast to the (possibly narrower) C++ type matching
// dtype below; a value that does not fit that type would silently
// wrap (integers) or overflow to infinity (the "single" precision
// float) instead of being reported, so such an object falls back to
// a plain object encoding as well
for (const auto& el : value.at(key))
{
bool in_range = true;
switch (dtype)
{
case 'U':
case 'C':
case 'B':
in_range = bjdata_ndarray_value_in_range<std::uint8_t>(el);
break;
case 'i':
in_range = bjdata_ndarray_value_in_range<std::int8_t>(el);
break;
case 'u':
in_range = bjdata_ndarray_value_in_range<std::uint16_t>(el);
break;
case 'I':
in_range = bjdata_ndarray_value_in_range<std::int16_t>(el);
break;
case 'm':
in_range = bjdata_ndarray_value_in_range<std::uint32_t>(el);
break;
case 'l':
in_range = bjdata_ndarray_value_in_range<std::int32_t>(el);
break;
case 'M':
in_range = bjdata_ndarray_value_in_range<std::uint64_t>(el);
break;
case 'L':
in_range = bjdata_ndarray_value_in_range<std::int64_t>(el);
break;
case 'd':
{
const auto dval = el.template get<double>();
in_range = !std::isfinite(dval) ||
(dval >= static_cast<double>(std::numeric_limits<float>::lowest()) &&
dval <= static_cast<double>((std::numeric_limits<float>::max)()));
break;
}
default:
// 'D' (double) already spans the full range of number_float_t
break;
}
if (!in_range)
{
return true;
}
}
oa->write_character('[');
oa->write_character('$');
oa->write_character(dtype);
@@ -20150,7 +20238,7 @@ class serializer
, thousands_sep(loc->thousands_sep == nullptr ? '\0' : std::char_traits<char>::to_char_type(* (loc->thousands_sep)))
, decimal_point(loc->decimal_point == nullptr ? '\0' : std::char_traits<char>::to_char_type(* (loc->decimal_point)))
, indent_char(ichar)
, indent_string()
, indent_string(512, indent_char)
, error_handler(error_handler_)
{}
@@ -20205,10 +20293,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -20282,10 +20366,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -20347,10 +20427,6 @@ class serializer
// variable to hold indentation for recursive calls
const auto new_indent = current_indent + indent_step;
if (JSON_HEDLEY_UNLIKELY(indent_string.empty()))
{
indent_string.resize(512, indent_char);
}
if (JSON_HEDLEY_UNLIKELY(indent_string.size() < new_indent))
{
indent_string.resize(indent_string.size() * 2, ' ');
@@ -21101,7 +21177,7 @@ class serializer
/// the indentation character
const char indent_char;
/// the indentation string (lazily allocated on first use by a pretty-print branch)
/// the indentation string
string_t indent_string;
/// error_handler how to react on decoding errors
+127 -2
View File
@@ -2586,7 +2586,12 @@ TEST_CASE("BJData")
CHECK(json::to_bjdata(json::from_bjdata(v_d), true, true) == v_d);
CHECK(json::to_bjdata(json::from_bjdata(v_D), true, true) == v_D);
CHECK(json::to_bjdata(json::from_bjdata(v_C), true, true) == v_C);
CHECK(json::to_bjdata(json::from_bjdata(v_B), true, true) == v_B);
// v_B uses the Draft-3-only 'B' marker, so it round-trips only when
// Draft 3 is explicitly selected (see GitHub issue #5404); the
// default Draft 2 falls back to a plain object instead, covered by
// the "ndarray with _ArrayType_ "byte" is gated by the BJData draft
// version" section below
CHECK(json::to_bjdata(json::from_bjdata(v_B), true, true, json::bjdata_version_t::draft3) == v_B);
}
SECTION("ndarray with data not matching _ArrayType_ is written as an object")
@@ -2629,8 +2634,10 @@ TEST_CASE("BJData")
// the C++ API stores an int literal as number_integer, so _ArrayType_
// names the wire type rather than the storage. Both storages have to
// produce the same typed array for every type.
// "byte" is checked separately below since it additionally requires
// BJData Draft 3 to be selected explicitly (see GitHub issue #5404).
for (const char* type :
{"uint8", "int8", "uint16", "int16", "uint32", "int32", "uint64", "int64", "char", "byte"
{"uint8", "int8", "uint16", "int16", "uint32", "int32", "uint64", "int64", "char"
})
{
CAPTURE(type);
@@ -2641,6 +2648,14 @@ TEST_CASE("BJData")
CHECK(from_text == json::to_bjdata(json({{"_ArrayType_", type}, {"_ArraySize_", {2, 3}}, {"_ArrayData_", {1, 2, 3, 4, 5, 6}}})));
}
{
const std::string text = R"({"_ArrayType_":"byte","_ArraySize_":[2,3],"_ArrayData_":[1,2,3,4,5,6]})";
const auto from_text = json::to_bjdata(json::parse(text), true, true, json::bjdata_version_t::draft3);
CHECK(from_text.at(0) == '[');
CHECK(from_text == json::to_bjdata(json({{"_ArrayType_", "byte"}, {"_ArraySize_", {2, 3}}, {"_ArrayData_", {1, 2, 3, 4, 5, 6}}}),
true, true, json::bjdata_version_t::draft3));
}
// negative values under a signed type behave the same way
const auto from_neg = json::to_bjdata(json::parse(R"({"_ArrayType_":"int32","_ArraySize_":[2],"_ArrayData_":[-5,7]})"));
CHECK(from_neg.at(0) == '[');
@@ -2731,6 +2746,39 @@ TEST_CASE("BJData")
CHECK(json::from_bjdata(json::to_bjdata(j_size), true, true) == j_size);
}
SECTION("ndarray whose _ArrayType_ is not a string stays as object")
{
// the type name is looked up as a string below the annotation
// check; a non-string _ArrayType_ cannot name a known dtype,
// so calling get<string_t>() on it would throw type_error.302
// instead of falling back like an unrecognized type name
// already does (see GitHub issue #5398)
json const j_number = json({{"_ArrayType_", 1}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 2}}});
const auto out_number = json::to_bjdata(j_number);
CHECK(out_number.at(0) == '{');
CHECK(json::from_bjdata(out_number) == j_number);
json const j_null = json({{"_ArrayType_", nullptr}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 2}}});
const auto out_null = json::to_bjdata(j_null);
CHECK(out_null.at(0) == '{');
CHECK(json::from_bjdata(out_null) == j_null);
json const j_bool = json({{"_ArrayType_", true}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 2}}});
const auto out_bool = json::to_bjdata(j_bool);
CHECK(out_bool.at(0) == '{');
CHECK(json::from_bjdata(out_bool) == j_bool);
json const j_array = json({{"_ArrayType_", {"uint8"}}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 2}}});
const auto out_array = json::to_bjdata(j_array);
CHECK(out_array.at(0) == '{');
CHECK(json::from_bjdata(out_array) == j_array);
json const j_object = json({{"_ArrayType_", {{"a", 1}}}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 2}}});
const auto out_object = json::to_bjdata(j_object);
CHECK(out_object.at(0) == '{');
CHECK(json::from_bjdata(out_object) == j_object);
}
SECTION("ndarray whose dimensions overflow stays as object")
{
// the product of the dimensions wraps around std::size_t to 0
@@ -2776,6 +2824,83 @@ TEST_CASE("BJData")
CHECK(out_num.at(0) == '{');
CHECK(json::from_bjdata(out_num) == j_num);
}
SECTION("ndarray with out-of-range _ArrayData_ elements stays as object")
{
// each element is cast to the (possibly narrower) C++ type
// named by _ArrayType_ before being written; a value that
// does not fit that type would silently wrap instead of
// being reported, so such an object falls back to a plain
// object encoding that still round-trips (see GitHub issue #5403)
// an unsigned element that does not fit uint8
json const j_uint8 = json({{"_ArrayType_", "uint8"}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 256}}});
const auto out_uint8 = json::to_bjdata(j_uint8);
CHECK(out_uint8.at(0) == '{');
CHECK(json::from_bjdata(out_uint8) == j_uint8);
// a signed element that does not fit int8
json const j_int8 = json({{"_ArrayType_", "int8"}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, 200}}});
const auto out_int8 = json::to_bjdata(j_int8);
CHECK(out_int8.at(0) == '{');
CHECK(json::from_bjdata(out_int8) == j_int8);
// a negative element is likewise out of range for an
// unsigned _ArrayType_
json const j_uint16_neg = json({{"_ArrayType_", "uint16"}, {"_ArraySize_", {2}}, {"_ArrayData_", {1, -1}}});
const auto out_uint16_neg = json::to_bjdata(j_uint16_neg);
CHECK(out_uint16_neg.at(0) == '{');
CHECK(json::from_bjdata(out_uint16_neg) == j_uint16_neg);
// a double element that overflows to infinity when narrowed
// to the "single" (float) precision named by _ArrayType_
json const j_single = json({{"_ArrayType_", "single"}, {"_ArraySize_", {2}}, {"_ArrayData_", {1.5, 1e40}}});
const auto out_single = json::to_bjdata(j_single);
CHECK(out_single.at(0) == '{');
CHECK(json::from_bjdata(out_single) == j_single);
// in-range boundary values still use the compact ndarray encoding
json const j_uint8_ok = json({{"_ArrayType_", "uint8"}, {"_ArraySize_", {2}}, {"_ArrayData_", {0, 255}}});
CHECK(json::to_bjdata(j_uint8_ok) == std::vector<uint8_t>({'[', '$', 'U', '#', '[', 'i', 2, ']', 0, 255}));
json const j_int8_ok = json({{"_ArrayType_", "int8"}, {"_ArraySize_", {2}}, {"_ArrayData_", {-128, 127}}});
CHECK(json::to_bjdata(j_int8_ok) == std::vector<uint8_t>({'[', '$', 'i', '#', '[', 'i', 2, ']', 0x80, 0x7F}));
json const j_single_ok = json({{"_ArrayType_", "single"}, {"_ArraySize_", {1}}, {"_ArrayData_", {1.5}}});
const auto out_single_ok = json::to_bjdata(j_single_ok);
CHECK(out_single_ok.at(0) == '[');
CHECK(json::from_bjdata(out_single_ok) == json({1.5f}));
}
SECTION("ndarray with _ArrayType_ \"byte\" is gated by the BJData draft version")
{
// the 'B' (byte) marker used by _ArrayType_ "byte" is only defined
// by BJData Draft 3; Draft 2 (the default) has no such marker, so
// emitting it unconditionally produced a stream that a Draft 2
// reader could not parse as intended (see GitHub issue #5404).
// Two dimensions are used so that a successfully written ndarray
// round-trips back into the annotated object (a single dimension
// is, by the BJData ndarray convention, read back as a plain
// binary value rather than the annotated object, same as every
// other single-dimension ndarray of a non-"byte" type is read
// back as a plain array instead of the annotated object).
json const j_byte = json({{"_ArrayType_", "byte"}, {"_ArraySize_", {2, 3}}, {"_ArrayData_", {1, 2, 3, 4, 5, 6}}});
// default (Draft 2): falls back to a plain object and round-trips
const auto out_draft2 = json::to_bjdata(j_byte);
CHECK(out_draft2.at(0) == '{');
CHECK(json::from_bjdata(out_draft2) == j_byte);
// explicit Draft 2: same as the default
const auto out_draft2_explicit = json::to_bjdata(j_byte, true, true, json::bjdata_version_t::draft2);
CHECK(out_draft2_explicit.at(0) == '{');
CHECK(json::from_bjdata(out_draft2_explicit) == j_byte);
// Draft 3 explicitly selected: still uses the compact 'B' ndarray encoding
const auto out_draft3 = json::to_bjdata(j_byte, true, true, json::bjdata_version_t::draft3);
CHECK(out_draft3 == std::vector<uint8_t>({'[', '$', 'B', '#', '[', '$', 'i', '#', 'i', 2, 2, 3, 1, 2, 3, 4, 5, 6}));
CHECK(json::from_bjdata(out_draft3) == j_byte);
}
}
}
-107
View File
@@ -14,40 +14,6 @@ using nlohmann::json;
#include <array>
#include <sstream>
#include <iomanip>
#include <cstdlib>
#include <new>
namespace
{
// heap allocation counter used by the regression test for issue #5413
// (https://github.com/nlohmann/json/issues/5413); disabled (and thus a
// no-op besides the counting) unless explicitly toggled on
bool count_heap_allocations = false; // NOLINT(cppcoreguidelines-avoid-non-const-global-variables)
std::size_t heap_allocations = 0; // NOLINT(cppcoreguidelines-avoid-non-const-global-variables)
} // namespace
void* operator new (std::size_t size) // NOLINT(cppcoreguidelines-owning-memory,misc-new-delete-overloads)
{
if (count_heap_allocations)
{
++heap_allocations;
}
if (void* ptr = std::malloc(size)) // NOLINT(cppcoreguidelines-no-malloc,cppcoreguidelines-owning-memory)
{
return ptr;
}
throw std::bad_alloc(); // NOLINT(hicpp-exception-baseclass)
}
void operator delete (void* ptr) noexcept // NOLINT(cppcoreguidelines-owning-memory,misc-new-delete-overloads)
{
std::free(ptr); // NOLINT(cppcoreguidelines-no-malloc,cppcoreguidelines-owning-memory)
}
void operator delete (void* ptr, std::size_t /*size*/) noexcept // NOLINT(cppcoreguidelines-owning-memory,misc-new-delete-overloads)
{
std::free(ptr); // NOLINT(cppcoreguidelines-no-malloc,cppcoreguidelines-owning-memory)
}
TEST_CASE("serialization")
{
@@ -416,76 +382,3 @@ TEST_CASE("dump for basic_json with long double number_float_t")
check_same(100.0L, 100.0);
}
}
TEST_CASE("regression test for issue #5413 - lazily allocated indent_string")
{
// the serializer used to unconditionally allocate a 512-byte
// indent_string in its constructor, even though it is only ever read
// inside the pretty_print branches of dump(). This wasted a heap
// allocation (and its matching deallocation) on every single compact
// (i.e. non-pretty, the default) dump() call. indent_string is now
// allocated lazily, the first time a pretty-print branch actually
// needs it -- so a compact dump() must perform strictly fewer heap
// allocations than a pretty dump() of the same value.
const json j = {{"level", "info"}, {"msg", "hello world"}, {"id", 12345}};
// warm up anything unrelated to indentation (e.g., one-time locale
// lookups) that might otherwise allocate on first use regardless of
// pretty-printing, so it does not skew the counts measured below
const auto warmup = j.dump();
const auto warmup_pretty = j.dump(4);
CHECK(!warmup.empty());
CHECK(!warmup_pretty.empty());
SECTION("compact dump() has a stable, minimal allocation count")
{
count_heap_allocations = true;
heap_allocations = 0;
const auto compact1 = j.dump();
const auto allocs_compact1 = heap_allocations;
heap_allocations = 0;
const auto compact2 = j.dump(-1);
const auto allocs_compact2 = heap_allocations;
count_heap_allocations = false;
CHECK(compact1 == compact2);
// dump() and dump(-1) both take the compact code path and must
// never touch indent_string, so they allocate identically often
CHECK(allocs_compact1 == allocs_compact2);
}
SECTION("first pretty dump() allocates more than a compact dump()")
{
// use a tiny value whose compact ({"a":1}, 7 bytes) and pretty
// ({"a": 1} with 1-space indent, 11 bytes) serializations both stay
// well inside every common std::string small-string-optimization
// buffer (>= 15 bytes on libstdc++/MSVC STL, >= 22 on libc++), so
// building the result string itself causes no heap allocation
// either way -- isolating indent_string as the only thing that can
// possibly account for a difference in allocation count
const json tiny = {{"a", 1}};
count_heap_allocations = true;
heap_allocations = 0;
const auto compact = tiny.dump();
const auto allocs_compact = heap_allocations;
heap_allocations = 0;
const auto pretty = tiny.dump(1);
const auto allocs_pretty = heap_allocations;
count_heap_allocations = false;
CHECK(compact == "{\"a\":1}");
CHECK(pretty == "{\n \"a\": 1\n}");
// a fresh serializer is created per dump() call; the pretty branch
// lazily allocates indent_string on its first use, so it must
// allocate at least once more than the compact branch, which never
// touches indent_string at all
CHECK(allocs_pretty > allocs_compact);
}
}