Fix REUSE compliance failure in gen_hedley_undef_check.cmake

The generated file's embedded banner contains the literal text
'SPDX-License-Identifier: MIT' as part of the *content* being written
to hedley_undef.hpp, not as this .cmake script's own REUSE header (it
is already covered by the blanket 'Files: *' rule in .reuse/dep5).

The reuse tool matched that embedded line as an SPDX tag for the
script itself and failed to parse the trailing 'MIT\n")' as a valid
SPDX License Expression, breaking ci_reuse_compliance. Wrap the
embedded banner in REUSE-IgnoreStart/REUSE-IgnoreEnd comments, as
recommended by the tool's own diagnostic output.

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

Makefile

@@ -1,4 +1,4 @@
-.PHONY: pretty clean ChangeLog.md release
+.PHONY: pretty clean ChangeLog.md release update_hedley update_hedley_undef
 
 ##########################################################################
 # configuration
@@ -41,6 +41,8 @@ all:
 	@echo "fuzz_testing_ubjson - prepare fuzz testing of the UBJSON parser"
 	@echo "pretty - beautify code with Artistic Style"
 	@echo "run_benchmarks - build and run benchmarks"
+	@echo "update_hedley - download Hedley and regenerate hedley.hpp / hedley_undef.hpp"
+	@echo "update_hedley_undef - rebuild hedley_undef.hpp from the JSON_HEDLEY_* #define names in hedley.hpp"
 
 
 ##########################################################################
@@ -241,11 +243,24 @@ update_hedley:
 	rm -f include/nlohmann/thirdparty/hedley/hedley.hpp include/nlohmann/thirdparty/hedley/hedley_undef.hpp
 	curl https://raw.githubusercontent.com/nemequ/hedley/master/hedley.h -o include/nlohmann/thirdparty/hedley/hedley.hpp
 	$(SED) -i 's/HEDLEY_/JSON_HEDLEY_/g' include/nlohmann/thirdparty/hedley/hedley.hpp
-	grep "[[:blank:]]*#[[:blank:]]*undef" include/nlohmann/thirdparty/hedley/hedley.hpp | grep -v "__" | sort | uniq | $(SED) 's/ //g' | $(SED) 's/undef/undef /g' > include/nlohmann/thirdparty/hedley/hedley_undef.hpp
 	$(SED) -i '1s/^/#pragma once\n\n/' include/nlohmann/thirdparty/hedley/hedley.hpp
-	$(SED) -i '1s/^/#pragma once\n\n/' include/nlohmann/thirdparty/hedley/hedley_undef.hpp
+	$(MAKE) update_hedley_undef
 	$(MAKE) amalgamate
 
+# Rebuild hedley_undef.hpp from every JSON_HEDLEY_* name that hedley.hpp
+# #defines. Hedley does not #undef all of its public macros internally (see
+# #5408), so grepping those #undef lines misses names such as
+# JSON_HEDLEY_PRAGMA. cmake/scripts/gen_hedley_undef_check.cmake is the
+# single source of truth for this extraction (tests/CMakeLists.txt uses the
+# same script, in MODE=checks, to generate the matching leak-check test), so
+# the vendored header, the generated #undef list, and the regression test
+# cannot drift apart.
+update_hedley_undef:
+	cmake -DHEDLEY_HPP=include/nlohmann/thirdparty/hedley/hedley.hpp \
+	      -DOUTPUT=include/nlohmann/thirdparty/hedley/hedley_undef.hpp \
+	      -DMODE=undef \
+	      -P cmake/scripts/gen_hedley_undef_check.cmake
+
 ##########################################################################
 # serve_header.py
 ##########################################################################

cmake/scripts/gen_hedley_undef_check.cmake

@@ -0,0 +1,112 @@
+# Shared extractor for the JSON_HEDLEY_* macro names defined in hedley.hpp.
+#
+# Every macro that hedley.hpp #defines must be #undef-ed again once json.hpp
+# has been fully processed (see include/nlohmann/detail/macro_unscope.hpp
+# and https://github.com/nlohmann/json/issues/5408). Deriving the macro list
+# straight from hedley.hpp here -- instead of hand-maintaining it in two
+# places -- means hedley_undef.hpp and the regression test that checks for
+# leaked macros can never drift apart, even after a future `make
+# update_hedley` pulls in new macros from upstream Hedley.
+#
+#   MODE=undef  (default): write hedley_undef.hpp (SPDX header, #pragma once,
+#               one #undef per macro name) -- used by `make update_hedley_undef`
+#   MODE=checks:            write one #ifdef/FAIL_CHECK/#endif per macro name,
+#               meant to be #include-d inside a TEST_CASE -- used by
+#               tests/CMakeLists.txt to (re)generate the include for
+#               tests/src/unit-no-macro-leak.cpp
+#
+# Required variables:
+#   HEDLEY_HPP  path to include/nlohmann/thirdparty/hedley/hedley.hpp
+#   OUTPUT      path of the file to (over)write
+# Optional:
+#   MODE        "undef" (default) or "checks"
+
+if(NOT DEFINED HEDLEY_HPP OR NOT DEFINED OUTPUT)
+    message(FATAL_ERROR "HEDLEY_HPP and OUTPUT must be set")
+endif()
+
+if(NOT EXISTS "${HEDLEY_HPP}")
+    message(FATAL_ERROR "Hedley header not found: ${HEDLEY_HPP}")
+endif()
+
+if(NOT DEFINED MODE)
+    set(MODE undef)
+endif()
+
+if(NOT MODE STREQUAL "undef" AND NOT MODE STREQUAL "checks")
+    message(FATAL_ERROR "MODE must be undef or checks, got: ${MODE}")
+endif()
+
+# Line-anchored, like `grep -oE "^[[:blank:]]*#[[:blank:]]*define[[:blank:]]+JSON_HEDLEY_[A-Za-z0-9_]+"`.
+# Unanchored matching would also pick up JSON_HEDLEY_* mentions inside
+# comments or string literals elsewhere in the file, which must not turn
+# into #undef lines.
+file(STRINGS "${HEDLEY_HPP}" hedley_lines)
+set(macro_names)
+foreach(line IN LISTS hedley_lines)
+    if("${line}" MATCHES "^[ \t]*#[ \t]*define[ \t]+(JSON_HEDLEY_[A-Za-z0-9_]+)")
+        list(APPEND macro_names "${CMAKE_MATCH_1}")
+    endif()
+endforeach()
+
+if(NOT macro_names)
+    message(FATAL_ERROR "No JSON_HEDLEY_* macros found in ${HEDLEY_HPP}")
+endif()
+
+list(REMOVE_DUPLICATES macro_names)
+# Lexicographic, locale-independent (ASCII-only names) -- matches `LC_ALL=C sort`.
+list(SORT macro_names COMPARE STRING)
+list(LENGTH macro_names macro_count)
+
+set(generated "")
+if(MODE STREQUAL "undef")
+    # Same banner `make update_hedley_undef` would stamp by hand, so the
+    # recipe is self-contained and its output is byte-stable across reruns.
+    # The embedded SPDX tags below are part of the *generated* file's
+    # content, not a REUSE header for this .cmake script itself (which is
+    # already covered by the blanket "Files: *" rule in .reuse/dep5) -- keep
+    # them wrapped in REUSE-IgnoreStart/End so `reuse lint` does not try to
+    # parse "MIT\n")" as this file's own SPDX-License-Identifier value.
+    # REUSE-IgnoreStart
+    string(APPEND generated "//     __ _____ _____ _____\n")
+    string(APPEND generated "//  __|  |   __|     |   | |  JSON for Modern C++\n")
+    string(APPEND generated "// |  |  |__   |  |  | | | |  version 3.12.0\n")
+    string(APPEND generated "// |_____|_____|_____|_|___|  https://github.com/nlohmann/json\n")
+    string(APPEND generated "//\n")
+    string(APPEND generated "// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>\n")
+    string(APPEND generated "// SPDX-License-Identifier: MIT\n")
+    # REUSE-IgnoreEnd
+    string(APPEND generated "\n")
+    string(APPEND generated "#pragma once\n")
+    string(APPEND generated "\n")
+    foreach(name IN LISTS macro_names)
+        string(APPEND generated "#undef ${name}\n")
+    endforeach()
+else()
+    string(APPEND generated "// This file is generated by cmake/scripts/gen_hedley_undef_check.cmake\n")
+    string(APPEND generated "// from include/nlohmann/thirdparty/hedley/hedley.hpp. Do not edit it by\n")
+    string(APPEND generated "// hand -- it is regenerated on every build. ${macro_count} macros checked.\n\n")
+    foreach(name IN LISTS macro_names)
+        string(APPEND generated "#ifdef ${name}\n")
+        string(APPEND generated "    FAIL_CHECK(\"${name} leaked after including nlohmann/json.hpp\");\n")
+        string(APPEND generated "#endif\n")
+    endforeach()
+endif()
+
+get_filename_component(output_dir "${OUTPUT}" DIRECTORY)
+if(output_dir)
+    file(MAKE_DIRECTORY "${output_dir}")
+endif()
+
+# Avoid rewriting the file (and busting downstream incremental rebuilds)
+# when the content has not actually changed.
+set(write_output TRUE)
+if(EXISTS "${OUTPUT}")
+    file(READ "${OUTPUT}" existing_content)
+    if(existing_content STREQUAL generated)
+        set(write_output FALSE)
+    endif()
+endif()
+if(write_output)
+    file(WRITE "${OUTPUT}" "${generated}")
+endif()

include/nlohmann/detail/output/binary_writer.hpp

@@ -1647,20 +1647,6 @@ 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
     */
@@ -1672,16 +1658,6 @@ 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>());
@@ -1691,16 +1667,6 @@ 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'
@@ -1765,60 +1731,6 @@ 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);

include/nlohmann/thirdparty/hedley/hedley_undef.hpp

@@ -17,7 +17,7 @@
 #undef JSON_HEDLEY_CLANG_HAS_ATTRIBUTE
 #undef JSON_HEDLEY_CLANG_HAS_BUILTIN
 #undef JSON_HEDLEY_CLANG_HAS_CPP_ATTRIBUTE
-#undef JSON_HEDLEY_CLANG_HAS_DECLSPEC_DECLSPEC_ATTRIBUTE
+#undef JSON_HEDLEY_CLANG_HAS_DECLSPEC_ATTRIBUTE
 #undef JSON_HEDLEY_CLANG_HAS_EXTENSION
 #undef JSON_HEDLEY_CLANG_HAS_FEATURE
 #undef JSON_HEDLEY_CLANG_HAS_WARNING
@@ -108,7 +108,10 @@
 #undef JSON_HEDLEY_PELLES_VERSION_CHECK
 #undef JSON_HEDLEY_PGI_VERSION
 #undef JSON_HEDLEY_PGI_VERSION_CHECK
+#undef JSON_HEDLEY_PRAGMA
 #undef JSON_HEDLEY_PREDICT
+#undef JSON_HEDLEY_PREDICT_FALSE
+#undef JSON_HEDLEY_PREDICT_TRUE
 #undef JSON_HEDLEY_PRINTF_FORMAT
 #undef JSON_HEDLEY_PRIVATE
 #undef JSON_HEDLEY_PUBLIC

single_include/nlohmann/json.hpp

@@ -18655,20 +18655,6 @@ 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
     */
@@ -18680,16 +18666,6 @@ 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>());
@@ -18699,16 +18675,6 @@ 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'
@@ -18773,60 +18739,6 @@ 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);
@@ -27053,7 +26965,7 @@ struct formatter<nlohmann::NLOHMANN_BASIC_JSON_TPL, char> // NOLINT(cert-dcl58-c
 #undef JSON_HEDLEY_CLANG_HAS_ATTRIBUTE
 #undef JSON_HEDLEY_CLANG_HAS_BUILTIN
 #undef JSON_HEDLEY_CLANG_HAS_CPP_ATTRIBUTE
-#undef JSON_HEDLEY_CLANG_HAS_DECLSPEC_DECLSPEC_ATTRIBUTE
+#undef JSON_HEDLEY_CLANG_HAS_DECLSPEC_ATTRIBUTE
 #undef JSON_HEDLEY_CLANG_HAS_EXTENSION
 #undef JSON_HEDLEY_CLANG_HAS_FEATURE
 #undef JSON_HEDLEY_CLANG_HAS_WARNING
@@ -27144,7 +27056,10 @@ struct formatter<nlohmann::NLOHMANN_BASIC_JSON_TPL, char> // NOLINT(cert-dcl58-c
 #undef JSON_HEDLEY_PELLES_VERSION_CHECK
 #undef JSON_HEDLEY_PGI_VERSION
 #undef JSON_HEDLEY_PGI_VERSION_CHECK
+#undef JSON_HEDLEY_PRAGMA
 #undef JSON_HEDLEY_PREDICT
+#undef JSON_HEDLEY_PREDICT_FALSE
+#undef JSON_HEDLEY_PREDICT_TRUE
 #undef JSON_HEDLEY_PRINTF_FORMAT
 #undef JSON_HEDLEY_PRIVATE
 #undef JSON_HEDLEY_PUBLIC

tests/CMakeLists.txt

@@ -125,6 +125,51 @@ json_test_set_test_options(test-unicode4 TEST_PROPERTIES TIMEOUT 3000)
 # add unit tests
 #############################################################################
 
+# Generate the leak checks for every JSON_HEDLEY_* macro defined in
+# hedley.hpp; tests/src/unit-no-macro-leak.cpp #include-s the result after
+# nlohmann/json.hpp (see issue #5408). Using the shared
+# cmake/scripts/gen_hedley_undef_check.cmake script (also used by `make
+# update_hedley_undef`) instead of a hand-maintained list of macro names
+# means this test can never go stale after a future `make update_hedley`.
+set(hedley_hpp "${PROJECT_SOURCE_DIR}/include/nlohmann/thirdparty/hedley/hedley.hpp")
+set(hedley_undef_check_script "${PROJECT_SOURCE_DIR}/cmake/scripts/gen_hedley_undef_check.cmake")
+set(hedley_undef_checks "${PROJECT_BINARY_DIR}/include/hedley_undef_checks.inc")
+
+# Reconfigure whenever the vendored header or the generator script changes,
+# so a `cmake --build` after `make update_hedley` does not silently keep a
+# stale generated file around.
+set_property(DIRECTORY APPEND PROPERTY CMAKE_CONFIGURE_DEPENDS
+    "${hedley_hpp}"
+    "${hedley_undef_check_script}")
+
+# Generate once at configure time, so the very first build (before any
+# custom-command build step has run) already has an up-to-date file.
+execute_process(
+    COMMAND ${CMAKE_COMMAND}
+        "-DHEDLEY_HPP=${hedley_hpp}"
+        "-DOUTPUT=${hedley_undef_checks}"
+        -DMODE=checks
+        -P "${hedley_undef_check_script}"
+    RESULT_VARIABLE hedley_undef_check_result
+)
+if(NOT hedley_undef_check_result EQUAL 0)
+    message(FATAL_ERROR "Failed to generate ${hedley_undef_checks}")
+endif()
+
+# Also (re)generate as a build step, so an incremental build after editing
+# hedley.hpp without a full reconfigure still picks up the change.
+add_custom_command(
+    OUTPUT "${hedley_undef_checks}"
+    COMMAND ${CMAKE_COMMAND}
+        "-DHEDLEY_HPP=${hedley_hpp}"
+        "-DOUTPUT=${hedley_undef_checks}"
+        -DMODE=checks
+        -P "${hedley_undef_check_script}"
+    DEPENDS "${hedley_hpp}" "${hedley_undef_check_script}"
+    COMMENT "Generating Hedley undef leak checks"
+    VERBATIM)
+add_custom_target(generate_hedley_undef_checks DEPENDS "${hedley_undef_checks}")
+
 if("${JSON_TestStandards}" STREQUAL "")
     set(test_cxx_standards 11 14 17 20 23)
     unset(test_force)
@@ -163,6 +208,14 @@ foreach(file ${files})
     json_test_add_test_for(${file} MAIN test_main CXX_STANDARDS ${test_cxx_standards} ${test_force})
 endforeach()
 
+# tests/src/unit-no-macro-leak.cpp #include-s the generated leak-check file,
+# so its test targets must be built after generate_hedley_undef_checks.
+foreach(cxx_standard ${test_cxx_standards})
+    if(TARGET test-no-macro-leak_cpp${cxx_standard})
+        add_dependencies(test-no-macro-leak_cpp${cxx_standard} generate_hedley_undef_checks)
+    endif()
+endforeach()
+
 if(json_32bit_test_only)
     # Skip all other tests in this file
     return()

tests/src/fuzzer-parse_bjdata.cpp

@@ -21,27 +21,6 @@ array data, it performs the following steps:
 - j4 = from_bjdata(vec3)
 - assert(j1 == j4)
 
-Re-serializing j2/j3/j4 with the same use_size/use_type settings is checked
-for value-stability rather than byte-exact stability: from_bjdata(to_bjdata(j2))
-must equal j2 (and likewise for j3, j4). Byte-exact stability does not hold in
-general, because a BJData value can lose type fidelity across a round trip
-(e.g. a binary_t value serialized without the optimized "$U#" array header is
-parsed back as a plain array of numbers, see #5398 and the discussion on
-PR #5494) - the numeric value is preserved, but the writer's smallest-type
-selection for the now-plain numbers may legitimately pick a different, but
-equally valid, single-byte type marker than the dedicated binary-data writer
-would have. Both encodings are valid BJData and both decode to the same
-value, so this is not treated as a round-trip failure here.
-
-"Value-stable" is checked by comparing dump()s rather than with operator==
-directly: a BJData/UBJSON payload can decode to a non-finite double (NaN or
-+-Infinity), and IEEE 754 NaN is never equal to itself, so operator== would
-report two structurally-identical trees as different whenever a NaN is
-involved -- not a round-trip bug, just NaN's ordinary (non-)reflexivity.
-dump() serializes any non-finite double the same deterministic way (as JSON
-`null`, since JSON itself cannot represent NaN/Infinity), so comparing
-dumps is stable under exactly the same values that break operator==.
-
 The provided function `LLVMFuzzerTestOneInput` can be used in different fuzzer
 drivers.
 */
@@ -52,13 +31,6 @@ drivers.
 
 using json = nlohmann::json;
 
-// value-stable comparison for the round-trip checks below; see the note
-// above on why this compares dump()s rather than the json values directly
-static bool is_value_stable(const json& lhs, const json& rhs)
-{
-    return lhs.dump() == rhs.dump();
-}
-
 // see http://llvm.org/docs/LibFuzzer.html
 extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size)
 {
@@ -84,12 +56,10 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size)
             json const j3 = json::from_bjdata(vec3);
             json const j4 = json::from_bjdata(vec4);
 
-            // re-serializing must be value-stable (see the notes above on
-            // why byte-exact stability is not guaranteed in general, and
-            // why this compares dump()s rather than the values directly)
-            assert(is_value_stable(json::from_bjdata(json::to_bjdata(j2, false, false)), j2));
-            assert(is_value_stable(json::from_bjdata(json::to_bjdata(j3, true, false)), j3));
-            assert(is_value_stable(json::from_bjdata(json::to_bjdata(j4, true, true)), j4));
+            // serializations must match
+            assert(json::to_bjdata(j2, false, false) == vec2);
+            assert(json::to_bjdata(j3, true, false) == vec3);
+            assert(json::to_bjdata(j4, true, true) == vec4);
         }
         catch (const json::parse_error&)
         {

tests/src/unit-bjdata.cpp

@@ -2586,12 +2586,7 @@ 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);
-                // 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);
+                CHECK(json::to_bjdata(json::from_bjdata(v_B), true, true) == v_B);
             }
 
             SECTION("ndarray with data not matching _ArrayType_ is written as an object")
@@ -2634,10 +2629,8 @@ 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"
+                        {"uint8", "int8", "uint16", "int16", "uint32", "int32", "uint64", "int64", "char", "byte"
                         })
                 {
                     CAPTURE(type);
@@ -2648,14 +2641,6 @@ 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) == '[');
@@ -2746,83 +2731,6 @@ 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("re-serializing a value containing a plain-array-of-bytes is value-stable but not byte-stable")
-            {
-                // OSS-Fuzz found this input (an array whose first element is a
-                // binary_t byte, followed by an object whose _ArrayType_ is
-                // not a string) while exercising the fix for #5398 above: once
-                // the fix stops to_bjdata() from throwing type_error.302 for
-                // the third element, serialization proceeds far enough to
-                // reach a pre-existing, unrelated round-trip quirk in how a
-                // single-byte binary_t value is re-encoded.
-                std::vector<std::uint8_t> const input
-                {
-                    0x5b, 0x5b, 0x24, 0x42, 0x23, 0x5b, 0x69, 0x01, 0x5d, 0x5b, 0x5b, 0x5d, 0x7b, 0x55, 0x0b,
-                    0x5f, 0x41, 0x72, 0x72, 0x61, 0x79, 0x44, 0x61, 0x74, 0x61, 0x5f, 0x54, 0x55, 0x0b, 0x5f,
-                    0x41, 0x72, 0x72, 0x61, 0x79, 0x53, 0x69, 0x7a, 0x65, 0x5f, 0x5a, 0x55, 0x0b, 0x5f, 0x41,
-                    0x72, 0x72, 0x61, 0x79, 0x54, 0x79, 0x70, 0x65, 0x5f, 0x54, 0x7d, 0x5d
-                };
-                json const j1 = json::from_bjdata(input);
-
-                // to_bjdata() must not throw (this is what #5398 fixes)
-                std::vector<std::uint8_t> vec2;
-                CHECK_NOTHROW(vec2 = json::to_bjdata(j1, false, false));
-
-                // parsing back a plain (non-optimized) array of bytes cannot
-                // recover that it used to be a binary_t: from_bjdata() has no
-                // way to distinguish "array of uint8 numbers" from "array of
-                // bytes" unless the compact "$U#" array header is used, so
-                // the binary_t collapses into a plain JSON array
-                json const j2 = json::from_bjdata(vec2);
-                CHECK(j1 != j2);
-                CHECK(j2 == json({{91}, json::array(), {{"_ArrayData_", true}, {"_ArraySize_", nullptr}, {"_ArrayType_", true}}}));
-
-                // re-serializing j2 no longer goes through the dedicated
-                // binary_t writer (which always uses the 'U' marker for raw
-                // bytes); the now-plain number 91 goes through the generic
-                // smallest-type writer instead, which - like the rest of the
-                // UBJSON/BJData writer, and unchanged by this fix - prefers
-                // the 'i' (int8) marker over 'U' (uint8) for values that fit
-                // both. Both markers are valid BJData and both decode back to
-                // 91, so this is not byte-for-byte identical to vec2, but it
-                // is value-stable: parsing it again reproduces j2 exactly.
-                std::vector<std::uint8_t> const vec3 = json::to_bjdata(j2, false, false);
-                CHECK(json::from_bjdata(vec3) == j2);
-            }
-
             SECTION("ndarray whose dimensions overflow stays as object")
             {
                 // the product of the dimensions wraps around std::size_t to 0
@@ -2868,83 +2776,6 @@ 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);
-            }
         }
     }
 

tests/src/unit-no-macro-leak.cpp

@@ -0,0 +1,34 @@
+//     __ _____ _____ _____
+//  __|  |   __|     |   | |  JSON for Modern C++
+// |  |  |__   |  |  | | | |  version 3.12.0
+// |_____|_____|_____|_|___|  https://github.com/nlohmann/json
+//
+// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
+// SPDX-License-Identifier: MIT
+
+// This file makes sure that none of the internal JSON_HEDLEY_* macros (vendored
+// from https://nemequ.github.io/hedley/, see
+// include/nlohmann/thirdparty/hedley/hedley.hpp) leak into the including
+// translation unit. include/nlohmann/detail/macro_unscope.hpp is supposed to
+// #undef every JSON_HEDLEY_* macro (via hedley_undef.hpp) once json.hpp has
+// been fully processed. See https://github.com/nlohmann/json/issues/5408,
+// where JSON_HEDLEY_PRAGMA, JSON_HEDLEY_PREDICT_TRUE, JSON_HEDLEY_PREDICT_FALSE,
+// and JSON_HEDLEY_CLANG_HAS_DECLSPEC_ATTRIBUTE escaped this cleanup because
+// hedley_undef.hpp had no matching #undef for them.
+//
+// hedley_undef_checks.inc (included below) is generated at CMake configure/
+// build time by cmake/scripts/gen_hedley_undef_check.cmake, which derives the
+// full list of JSON_HEDLEY_* macro names directly from hedley.hpp. That way
+// this test covers every macro Hedley actually defines -- not a hardcoded
+// snapshot that would silently go stale the next time `make update_hedley`
+// runs -- and can never drift from the vendored header.
+
+#include "doctest_compatibility.h"
+
+#include <nlohmann/json.hpp>
+
+TEST_CASE("JSON_HEDLEY macros do not leak after including json.hpp")
+{
+#include "hedley_undef_checks.inc"
+    CHECK(true); // keep an assertion when nothing leaked
+}