Fix 32-bit overflow in huge_string_t BSON length-overflow tests

huge_string_t doubles as basic_json's StringType, so it is used not only
for the JSON string value under test but also for object keys (e.g. "s",
"nested"). Making size() unconditionally lie about being huge therefore
inflated the keys' reported sizes as well, pushing the running totals
computed while walking the BSON document (calc_bson_object_size and
friends in binary_writer.hpp) past what a 32-bit std::size_t can hold.

On 64-bit platforms this happens to still produce a working (if
needlessly large) result, but on 32-bit platforms (e.g. the mingw x86 CI
job) the size_t arithmetic silently wraps around: for the "document" test
this merely surfaces the wrong number in the exception message, but for
the "string" test the wrapped total happens to fall back under
INT32_MAX, so the intended out_of_range.412 guard is skipped entirely and
the code goes on to actually write ~2 GiB worth of characters from the
key's real, tiny buffer - which is what raised the reported
"vector::_M_range_insert" exception instead of a controlled 412.

Make the fake-huge size opt-in via huge_string_t::as_huge() and only
apply it to the string value under test, leaving keys at their real
(small) size. This keeps every intermediate size well within 32-bit
size_t range on any platform, matching how huge_binary_t already avoids
the same trap (it is only ever used as the BSON value type, never as a
key). Expected out_of_range.412 messages are updated accordingly.

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

include/nlohmann/detail/input/json_sax.hpp

@@ -8,11 +8,10 @@
 
 #pragma once
 
-#include <algorithm> // find_if
 #include <cstddef>
 #include <string> // string
 #include <type_traits> // enable_if_t
-#include <utility> // move, pair
+#include <utility> // move
 #include <vector> // vector
 
 #include <nlohmann/detail/exceptions.hpp>
@@ -586,17 +585,7 @@ class json_sax_dom_callback_parser
         // add discarded value at the given key and store the reference for later
         if (keep && ref_stack.back())
         {
-            auto& obj = *ref_stack.back()->m_data.m_value.object;
+            object_element = &(ref_stack.back()->m_data.m_value.object->operator[](val) = discarded);
-            const auto it = obj.find(val);
-            if (it != obj.end())
-            {
-                // this is a duplicate key (legal in JSON); remember its
-                // current value so it can be restored later if the new
-                // value is rejected by the callback, instead of being
-                // erased together with the discarded placeholder
-                duplicate_key_stash.emplace_back(&(it->second), it->second);
-            }
-            object_element = &(obj[val] = discarded);
         }
 
         return true;
@@ -608,11 +597,7 @@ class json_sax_dom_callback_parser
         {
             if (!callback(static_cast<int>(ref_stack.size()) - 1, parse_event_t::object_end, *ref_stack.back()))
             {
-                // discard object, unless this slot holds a duplicate key's
+                // discard object
-                // previous value pending restoration, in which case that
-                // value is restored instead of being discarded
-                if (!resolve_duplicate_key_stash(ref_stack.back(), true))
-                {
                 *ref_stack.back() = discarded;
 
 #if JSON_DIAGNOSTIC_POSITIONS
@@ -620,7 +605,6 @@ class json_sax_dom_callback_parser
                 handle_diagnostic_positions_for_json_value(*ref_stack.back());
 #endif
             }
-            }
             else
             {
 
@@ -633,10 +617,6 @@ class json_sax_dom_callback_parser
 #endif
 
                 ref_stack.back()->set_parents();
-                // this object is finally, definitively kept; drop any
-                // pending duplicate-key stash entry for its slot since it
-                // can no longer be restored
-                resolve_duplicate_key_stash(ref_stack.back(), false);
             }
         }
 
@@ -706,18 +686,10 @@ class json_sax_dom_callback_parser
 #endif
 
                 ref_stack.back()->set_parents();
-                // this array is finally, definitively kept; drop any
-                // pending duplicate-key stash entry for its slot since it
-                // can no longer be restored
-                resolve_duplicate_key_stash(ref_stack.back(), false);
             }
             else
             {
-                // discard array, unless this slot holds a duplicate key's
+                // discard array
-                // previous value pending restoration, in which case that
-                // value is restored instead of being discarded
-                if (!resolve_duplicate_key_stash(ref_stack.back(), true))
-                {
                 *ref_stack.back() = discarded;
 
 #if JSON_DIAGNOSTIC_POSITIONS
@@ -726,7 +698,6 @@ class json_sax_dom_callback_parser
 #endif
             }
         }
-        }
 
         JSON_ASSERT(!ref_stack.empty());
         JSON_ASSERT(!keep_stack.empty());
@@ -838,48 +809,14 @@ class json_sax_dom_callback_parser
     }
 #endif
 
-    /// if there is a pending duplicate-key stash entry for this exact slot,
+    /// remove the discarded value the callback rejected from its parent
-    /// remove it from the stash; if restore_value is true, the stashed
+    static void remove_discarded_value(BasicJsonType& parent)
-    /// previous value is moved back into the slot first (use this when the
-    /// new value at that slot was rejected); otherwise the stash entry is
-    /// simply dropped (use this when the new value was accepted, so it
-    /// correctly supersedes the old one and no restore should ever happen
-    /// for this slot again)
-    /// @return whether a matching stash entry was found (and processed)
-    bool resolve_duplicate_key_stash(BasicJsonType* slot, bool restore_value)
-    {
-        const auto it = std::find_if(duplicate_key_stash.begin(), duplicate_key_stash.end(),
-                                     [slot](const std::pair<BasicJsonType*, BasicJsonType>& entry)
-        {
-            return entry.first == slot;
-        });
-
-        if (it == duplicate_key_stash.end())
-        {
-            return false;
-        }
-
-        if (restore_value)
-        {
-            *slot = std::move(it->second);
-        }
-        duplicate_key_stash.erase(it);
-        return true;
-    }
-
-    /// remove the discarded value the callback rejected from its parent,
-    /// unless it is a duplicate key's slot with a stashed previous value,
-    /// in which case that previous value is restored instead
-    void remove_discarded_value(BasicJsonType& parent)
     {
         for (auto it = parent.begin(); it != parent.end(); ++it)
         {
             if (it->is_discarded())
-            {
-                if (!resolve_duplicate_key_stash(&(*it), true))
             {
                 parent.erase(it);
-                }
                 break;
             }
         }
@@ -977,16 +914,6 @@ class json_sax_dom_callback_parser
 
         JSON_ASSERT(object_element);
         *object_element = std::move(value);
-        if (!skip_callback)
-        {
-            // this scalar value finally, definitively replaces whatever was
-            // at this slot; drop any pending duplicate-key stash entry for
-            // it since it can no longer be restored (a container value at
-            // this slot is resolved later, in end_object()/end_array(),
-            // since skip_callback is true for the placeholder handling that
-            // happens here for those)
-            resolve_duplicate_key_stash(object_element, false);
-        }
         return {true, object_element};
     }
 
@@ -1000,12 +927,6 @@ class json_sax_dom_callback_parser
     std::vector<bool> key_keep_stack {}; // NOLINT(readability-redundant-member-init)
     /// helper to hold the reference for the next object element
     BasicJsonType* object_element = nullptr;
-    /// stash of (slot pointer, previous value) for object members that
-    /// already existed when key() was called again for the same key
-    /// (duplicate keys); used to restore the previous value if the new
-    /// value is later rejected by the callback, instead of erasing the
-    /// member entirely
-    std::vector<std::pair<BasicJsonType*, BasicJsonType>> duplicate_key_stash {};
     /// whether a syntax error occurred
     bool errored = false;
     /// callback function

include/nlohmann/json.hpp

@@ -3573,7 +3573,6 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
         {
             using std::swap;
             swap(*(m_data.m_value.array), other);
-            set_parents();
         }
         else
         {
@@ -3590,7 +3589,6 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
         {
             using std::swap;
             swap(*(m_data.m_value.object), other);
-            set_parents();
         }
         else
         {
@@ -5159,139 +5157,34 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
 
             case value_t::object:
             {
-                // first pass: record, for every source key, whether it is
+                // first pass: traverse this object's elements
-                // common to both objects (in source's iteration order) or
-                // was deleted (i.e., in source but not in target) -- this is
-                // a by-product of the target.find() call already needed to
-                // tell the two cases apart, so it adds no extra lookups. The
-                // "remove" ops themselves are emitted later, interleaved
-                // with the recursive per-key diffs in the fast path below,
-                // to match source's original iteration order (as the
-                // original, pre-reordering-aware implementation did) instead
-                // of grouping all removes before all recursive diffs.
-                std::vector<typename object_t::key_type> common_keys_source_order;
                 for (auto it = source.cbegin(); it != source.cend(); ++it)
                 {
+                    // escape the key name to be used in a JSON patch
+                    const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
+
                     if (target.find(it.key()) != target.end())
                     {
-                        common_keys_source_order.push_back(it.key());
+                        // recursive call to compare object values at key it
+                        auto temp_diff = diff(it.value(), target[it.key()], path_key);
+                        result.insert(result.end(), temp_diff.begin(), temp_diff.end());
+                    }
+                    else
+                    {
+                        // found a key that is not in o -> remove it
+                        result.push_back(object(
+                        {
+                            {"op", "remove"}, {"path", path_key}
+                        }));
                     }
                 }
 
-                // second pass: find keys that were added (i.e., in target but
+                // second pass: traverse other object's elements
-                // not in source), and record the keys common to both, in
-                // target's iteration order -- again a by-product of the
-                // source.find() call already needed to detect added keys. At
-                // the same time, determine whether every added key comes
-                // after every common key in target's order (a precondition
-                // for the fast path below, which only ever appends new keys
-                // at the very end): for an object_t whose iteration order is
-                // a pure function of the key set (e.g. the default std::map,
-                // which always iterates in sorted key order), the order
-                // check further below is always true and this whole
-                // mechanism is effectively a no-op; it only matters for a
-                // reorderable object_t such as the one backing `ordered_json`.
-                // patch ops for keys that were added (i.e., in target but not
-                // in source); built here so the fast path below can reuse
-                // them without a second source.find() per target key. Only
-                // used by the fast path -- the slow (reordering) path
-                // rebuilds "add" ops for every key itself.
-                std::vector<typename object_t::key_type> common_keys_target_order;
-                basic_json added_ops(value_t::array);
-                bool new_keys_form_suffix = true;
-                bool seen_new_key = false;
                 for (auto it = target.cbegin(); it != target.cend(); ++it)
                 {
                     if (source.find(it.key()) == source.end())
                     {
-                        seen_new_key = true;
+                        // found a key that is not in this -> add it
-                        const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                        added_ops.push_back(
-                        {
-                            {"op", "add"}, {"path", path_key},
-                            {"value", it.value()}
-                        });
-                    }
-                    else
-                    {
-                        common_keys_target_order.push_back(it.key());
-                        if (seen_new_key)
-                        {
-                            new_keys_form_suffix = false;
-                        }
-                    }
-                }
-
-                if (common_keys_source_order == common_keys_target_order && new_keys_form_suffix)
-                {
-                    // fast path: order of common keys already matches (or the
-                    // object_t's iteration order does not depend on
-                    // insertion history), so a plain per-key recursive diff
-                    // is correct and minimal, as before. common_keys_source_order
-                    // is, by construction, the subsequence of source's keys
-                    // that are common to both objects, in source's iteration
-                    // order -- so it can be walked in lockstep with `source`
-                    // using a cheap key comparison instead of another lookup.
-                    // Deleted keys (those source keys not in common_keys_source_order)
-                    // are interleaved here too, in source's original order, to
-                    // match the historical (pre-reordering-aware) output order.
-                    auto common_it = common_keys_source_order.cbegin();
-                    for (auto it = source.cbegin(); it != source.cend(); ++it)
-                    {
-                        if (common_it != common_keys_source_order.cend() && it.key() == *common_it)
-                        {
-                            const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                            auto temp_diff = diff(it.value(), target[it.key()], path_key);
-                            result.insert(result.end(), temp_diff.begin(), temp_diff.end());
-                            ++common_it;
-                        }
-                        else
-                        {
-                            // found a key that is not in target -> remove it
-                            const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                            result.push_back(object(
-                            {
-                                {"op", "remove"}, {"path", path_key}
-                            }));
-                        }
-                    }
-
-                    // append the "add" ops for brand-new keys collected above
-                    // during the pass over target -- no second source.find()
-                    // per target key needed
-                    result.insert(result.end(), added_ops.begin(), added_ops.end());
-                }
-                else
-                {
-                    // slow path: the common keys are in a different relative
-                    // order in source and target (only possible for a
-                    // reorderable object_t like ordered_map). Building a
-                    // minimal reordering patch is a nontrivial (LCS-like)
-                    // problem; instead, remove every source key -- both
-                    // deleted keys (which must be removed regardless) and
-                    // common keys (removed so they can be re-added in
-                    // target's order) -- and re-add every key that should
-                    // remain, with its final target value, in target's
-                    // order. basic_json::patch()'s "add" operation on an
-                    // object uses operator[], which appends at the end for a
-                    // vector-backed insertion-ordered map when the key does
-                    // not already exist -- so removing a key and then adding
-                    // it moves it to the end, fixing its position.
-                    for (auto it = source.cbegin(); it != source.cend(); ++it)
-                    {
-                        const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                        result.push_back(object(
-                        {
-                            {"op", "remove"}, {"path", path_key}
-                        }));
-                    }
-
-                    // add every key that is either common (just removed
-                    // above) or brand new, in target's iteration order, so
-                    // that the final order after applying the patch matches
-                    // target exactly
-                    for (auto it = target.cbegin(); it != target.cend(); ++it)
-                    {
                         const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
                         result.push_back(
                         {

single_include/nlohmann/json.hpp

@@ -7768,11 +7768,10 @@ NLOHMANN_JSON_NAMESPACE_END
 
 
 
-#include <algorithm> // find_if
 #include <cstddef>
 #include <string> // string
 #include <type_traits> // enable_if_t
-#include <utility> // move, pair
+#include <utility> // move
 #include <vector> // vector
 
 // #include <nlohmann/detail/exceptions.hpp>
@@ -10114,17 +10113,7 @@ class json_sax_dom_callback_parser
         // add discarded value at the given key and store the reference for later
         if (keep && ref_stack.back())
         {
-            auto& obj = *ref_stack.back()->m_data.m_value.object;
+            object_element = &(ref_stack.back()->m_data.m_value.object->operator[](val) = discarded);
-            const auto it = obj.find(val);
-            if (it != obj.end())
-            {
-                // this is a duplicate key (legal in JSON); remember its
-                // current value so it can be restored later if the new
-                // value is rejected by the callback, instead of being
-                // erased together with the discarded placeholder
-                duplicate_key_stash.emplace_back(&(it->second), it->second);
-            }
-            object_element = &(obj[val] = discarded);
         }
 
         return true;
@@ -10136,11 +10125,7 @@ class json_sax_dom_callback_parser
         {
             if (!callback(static_cast<int>(ref_stack.size()) - 1, parse_event_t::object_end, *ref_stack.back()))
             {
-                // discard object, unless this slot holds a duplicate key's
+                // discard object
-                // previous value pending restoration, in which case that
-                // value is restored instead of being discarded
-                if (!resolve_duplicate_key_stash(ref_stack.back(), true))
-                {
                 *ref_stack.back() = discarded;
 
 #if JSON_DIAGNOSTIC_POSITIONS
@@ -10148,7 +10133,6 @@ class json_sax_dom_callback_parser
                 handle_diagnostic_positions_for_json_value(*ref_stack.back());
 #endif
             }
-            }
             else
             {
 
@@ -10161,10 +10145,6 @@ class json_sax_dom_callback_parser
 #endif
 
                 ref_stack.back()->set_parents();
-                // this object is finally, definitively kept; drop any
-                // pending duplicate-key stash entry for its slot since it
-                // can no longer be restored
-                resolve_duplicate_key_stash(ref_stack.back(), false);
             }
         }
 
@@ -10234,18 +10214,10 @@ class json_sax_dom_callback_parser
 #endif
 
                 ref_stack.back()->set_parents();
-                // this array is finally, definitively kept; drop any
-                // pending duplicate-key stash entry for its slot since it
-                // can no longer be restored
-                resolve_duplicate_key_stash(ref_stack.back(), false);
             }
             else
             {
-                // discard array, unless this slot holds a duplicate key's
+                // discard array
-                // previous value pending restoration, in which case that
-                // value is restored instead of being discarded
-                if (!resolve_duplicate_key_stash(ref_stack.back(), true))
-                {
                 *ref_stack.back() = discarded;
 
 #if JSON_DIAGNOSTIC_POSITIONS
@@ -10254,7 +10226,6 @@ class json_sax_dom_callback_parser
 #endif
             }
         }
-        }
 
         JSON_ASSERT(!ref_stack.empty());
         JSON_ASSERT(!keep_stack.empty());
@@ -10366,48 +10337,14 @@ class json_sax_dom_callback_parser
     }
 #endif
 
-    /// if there is a pending duplicate-key stash entry for this exact slot,
+    /// remove the discarded value the callback rejected from its parent
-    /// remove it from the stash; if restore_value is true, the stashed
+    static void remove_discarded_value(BasicJsonType& parent)
-    /// previous value is moved back into the slot first (use this when the
-    /// new value at that slot was rejected); otherwise the stash entry is
-    /// simply dropped (use this when the new value was accepted, so it
-    /// correctly supersedes the old one and no restore should ever happen
-    /// for this slot again)
-    /// @return whether a matching stash entry was found (and processed)
-    bool resolve_duplicate_key_stash(BasicJsonType* slot, bool restore_value)
-    {
-        const auto it = std::find_if(duplicate_key_stash.begin(), duplicate_key_stash.end(),
-                                     [slot](const std::pair<BasicJsonType*, BasicJsonType>& entry)
-        {
-            return entry.first == slot;
-        });
-
-        if (it == duplicate_key_stash.end())
-        {
-            return false;
-        }
-
-        if (restore_value)
-        {
-            *slot = std::move(it->second);
-        }
-        duplicate_key_stash.erase(it);
-        return true;
-    }
-
-    /// remove the discarded value the callback rejected from its parent,
-    /// unless it is a duplicate key's slot with a stashed previous value,
-    /// in which case that previous value is restored instead
-    void remove_discarded_value(BasicJsonType& parent)
     {
         for (auto it = parent.begin(); it != parent.end(); ++it)
         {
             if (it->is_discarded())
-            {
-                if (!resolve_duplicate_key_stash(&(*it), true))
             {
                 parent.erase(it);
-                }
                 break;
             }
         }
@@ -10505,16 +10442,6 @@ class json_sax_dom_callback_parser
 
         JSON_ASSERT(object_element);
         *object_element = std::move(value);
-        if (!skip_callback)
-        {
-            // this scalar value finally, definitively replaces whatever was
-            // at this slot; drop any pending duplicate-key stash entry for
-            // it since it can no longer be restored (a container value at
-            // this slot is resolved later, in end_object()/end_array(),
-            // since skip_callback is true for the placeholder handling that
-            // happens here for those)
-            resolve_duplicate_key_stash(object_element, false);
-        }
         return {true, object_element};
     }
 
@@ -10528,12 +10455,6 @@ class json_sax_dom_callback_parser
     std::vector<bool> key_keep_stack {}; // NOLINT(readability-redundant-member-init)
     /// helper to hold the reference for the next object element
     BasicJsonType* object_element = nullptr;
-    /// stash of (slot pointer, previous value) for object members that
-    /// already existed when key() was called again for the same key
-    /// (duplicate keys); used to restore the previous value if the new
-    /// value is later rejected by the callback, instead of erasing the
-    /// member entirely
-    std::vector<std::pair<BasicJsonType*, BasicJsonType>> duplicate_key_stash {};
     /// whether a syntax error occurred
     bool errored = false;
     /// callback function
@@ -25080,7 +25001,6 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
         {
             using std::swap;
             swap(*(m_data.m_value.array), other);
-            set_parents();
         }
         else
         {
@@ -25097,7 +25017,6 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
         {
             using std::swap;
             swap(*(m_data.m_value.object), other);
-            set_parents();
         }
         else
         {
@@ -26666,139 +26585,34 @@ class basic_json // NOLINT(cppcoreguidelines-special-member-functions,hicpp-spec
 
             case value_t::object:
             {
-                // first pass: record, for every source key, whether it is
+                // first pass: traverse this object's elements
-                // common to both objects (in source's iteration order) or
-                // was deleted (i.e., in source but not in target) -- this is
-                // a by-product of the target.find() call already needed to
-                // tell the two cases apart, so it adds no extra lookups. The
-                // "remove" ops themselves are emitted later, interleaved
-                // with the recursive per-key diffs in the fast path below,
-                // to match source's original iteration order (as the
-                // original, pre-reordering-aware implementation did) instead
-                // of grouping all removes before all recursive diffs.
-                std::vector<typename object_t::key_type> common_keys_source_order;
                 for (auto it = source.cbegin(); it != source.cend(); ++it)
                 {
+                    // escape the key name to be used in a JSON patch
+                    const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
+
                     if (target.find(it.key()) != target.end())
                     {
-                        common_keys_source_order.push_back(it.key());
+                        // recursive call to compare object values at key it
+                        auto temp_diff = diff(it.value(), target[it.key()], path_key);
+                        result.insert(result.end(), temp_diff.begin(), temp_diff.end());
+                    }
+                    else
+                    {
+                        // found a key that is not in o -> remove it
+                        result.push_back(object(
+                        {
+                            {"op", "remove"}, {"path", path_key}
+                        }));
                     }
                 }
 
-                // second pass: find keys that were added (i.e., in target but
+                // second pass: traverse other object's elements
-                // not in source), and record the keys common to both, in
-                // target's iteration order -- again a by-product of the
-                // source.find() call already needed to detect added keys. At
-                // the same time, determine whether every added key comes
-                // after every common key in target's order (a precondition
-                // for the fast path below, which only ever appends new keys
-                // at the very end): for an object_t whose iteration order is
-                // a pure function of the key set (e.g. the default std::map,
-                // which always iterates in sorted key order), the order
-                // check further below is always true and this whole
-                // mechanism is effectively a no-op; it only matters for a
-                // reorderable object_t such as the one backing `ordered_json`.
-                // patch ops for keys that were added (i.e., in target but not
-                // in source); built here so the fast path below can reuse
-                // them without a second source.find() per target key. Only
-                // used by the fast path -- the slow (reordering) path
-                // rebuilds "add" ops for every key itself.
-                std::vector<typename object_t::key_type> common_keys_target_order;
-                basic_json added_ops(value_t::array);
-                bool new_keys_form_suffix = true;
-                bool seen_new_key = false;
                 for (auto it = target.cbegin(); it != target.cend(); ++it)
                 {
                     if (source.find(it.key()) == source.end())
                     {
-                        seen_new_key = true;
+                        // found a key that is not in this -> add it
-                        const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                        added_ops.push_back(
-                        {
-                            {"op", "add"}, {"path", path_key},
-                            {"value", it.value()}
-                        });
-                    }
-                    else
-                    {
-                        common_keys_target_order.push_back(it.key());
-                        if (seen_new_key)
-                        {
-                            new_keys_form_suffix = false;
-                        }
-                    }
-                }
-
-                if (common_keys_source_order == common_keys_target_order && new_keys_form_suffix)
-                {
-                    // fast path: order of common keys already matches (or the
-                    // object_t's iteration order does not depend on
-                    // insertion history), so a plain per-key recursive diff
-                    // is correct and minimal, as before. common_keys_source_order
-                    // is, by construction, the subsequence of source's keys
-                    // that are common to both objects, in source's iteration
-                    // order -- so it can be walked in lockstep with `source`
-                    // using a cheap key comparison instead of another lookup.
-                    // Deleted keys (those source keys not in common_keys_source_order)
-                    // are interleaved here too, in source's original order, to
-                    // match the historical (pre-reordering-aware) output order.
-                    auto common_it = common_keys_source_order.cbegin();
-                    for (auto it = source.cbegin(); it != source.cend(); ++it)
-                    {
-                        if (common_it != common_keys_source_order.cend() && it.key() == *common_it)
-                        {
-                            const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                            auto temp_diff = diff(it.value(), target[it.key()], path_key);
-                            result.insert(result.end(), temp_diff.begin(), temp_diff.end());
-                            ++common_it;
-                        }
-                        else
-                        {
-                            // found a key that is not in target -> remove it
-                            const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                            result.push_back(object(
-                            {
-                                {"op", "remove"}, {"path", path_key}
-                            }));
-                        }
-                    }
-
-                    // append the "add" ops for brand-new keys collected above
-                    // during the pass over target -- no second source.find()
-                    // per target key needed
-                    result.insert(result.end(), added_ops.begin(), added_ops.end());
-                }
-                else
-                {
-                    // slow path: the common keys are in a different relative
-                    // order in source and target (only possible for a
-                    // reorderable object_t like ordered_map). Building a
-                    // minimal reordering patch is a nontrivial (LCS-like)
-                    // problem; instead, remove every source key -- both
-                    // deleted keys (which must be removed regardless) and
-                    // common keys (removed so they can be re-added in
-                    // target's order) -- and re-add every key that should
-                    // remain, with its final target value, in target's
-                    // order. basic_json::patch()'s "add" operation on an
-                    // object uses operator[], which appends at the end for a
-                    // vector-backed insertion-ordered map when the key does
-                    // not already exist -- so removing a key and then adding
-                    // it moves it to the end, fixing its position.
-                    for (auto it = source.cbegin(); it != source.cend(); ++it)
-                    {
-                        const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
-                        result.push_back(object(
-                        {
-                            {"op", "remove"}, {"path", path_key}
-                        }));
-                    }
-
-                    // add every key that is either common (just removed
-                    // above) or brand new, in target's iteration order, so
-                    // that the final order after applying the patch matches
-                    // target exactly
-                    for (auto it = target.cbegin(); it != target.cend(); ++it)
-                    {
                         const auto path_key = detail::concat<string_t>(path, '/', detail::escape(it.key()));
                         result.push_back(
                         {

tests/src/unit-bson.cpp

@@ -38,6 +38,54 @@ class huge_binary_t : public std::vector<std::uint8_t>
 using huge_binary_json = nlohmann::basic_json <
                          std::map, std::vector, std::string, bool, std::int64_t, std::uint64_t,
                          double, std::allocator, nlohmann::adl_serializer, huge_binary_t, void >;
+
+// a string type that can be made to report a size beyond INT32_MAX without
+// allocating that much memory, so BSON length overflow can be tested for
+// strings and (embedded) documents as well, following the same idea as
+// huge_binary_t.
+//
+// Unlike huge_binary_t (which is only ever used as the BSON *value* type),
+// this type doubles as basic_json's StringType and is therefore also used
+// for *object keys* (e.g. "s" or "nested" below). Only the designated test
+// value is meant to lie about its size - if every huge_string_t (including
+// keys) reported a huge size, the running totals computed while walking the
+// BSON document (see calc_bson_object_size & friends in binary_writer.hpp)
+// would need more than 32 bits, and on platforms where std::size_t is only
+// 32 bits wide that arithmetic would silently wrap around, producing wrong
+// (or even unguarded) lengths. The fake size is therefore opt-in via
+// as_huge(), and plain strings - in particular object keys - keep reporting
+// their real, small size.
+class huge_string_t : public std::string
+{
+  public:
+    using std::string::string;
+    huge_string_t(const std::string& s) : std::string(s) {} // NOLINT(google-explicit-constructor,hicpp-explicit-conversions)
+
+    // returns a copy of @a s whose size() pretends to be huge
+    static huge_string_t as_huge(const std::string& s)
+    {
+        huge_string_t result(s);
+        result.pretend_huge = true;
+        return result;
+    }
+
+    size_type size() const noexcept
+    {
+        if (pretend_huge)
+        {
+            // one byte more than the BSON length field can represent
+            return static_cast<size_type>((std::numeric_limits<std::int32_t>::max)()) + 1;
+        }
+        return std::string::size();
+    }
+
+  private:
+    bool pretend_huge = false;
+};
+
+using huge_string_json = nlohmann::basic_json <
+                         std::map, std::vector, huge_string_t, bool, std::int64_t, std::uint64_t,
+                         double, std::allocator, nlohmann::adl_serializer, std::vector<std::uint8_t>, void >;
 } // namespace
 
 TEST_CASE("BSON")
@@ -104,6 +152,11 @@ TEST_CASE("BSON")
     }
 
     SECTION("lengths exceeding INT32_MAX cannot be serialized to BSON")
+    {
+        // out_of_range.412 is thrown from a single shared helper
+        // (to_bson_length) that guards the BSON length fields of binary
+        // values, strings, and (embedded) documents alike
+        SECTION("binary")
         {
             huge_binary_json j;
             j["b"] = huge_binary_json::binary(huge_binary_t{});
@@ -111,6 +164,27 @@ TEST_CASE("BSON")
             CHECK_THROWS_WITH_AS(huge_binary_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 2147483661 exceeds maximum of 2147483647", huge_binary_json::out_of_range&);
         }
 
+        SECTION("string")
+        {
+            huge_string_json j;
+            j["s"] = huge_string_t::as_huge("value");
+
+            CHECK_THROWS_WITH_AS(huge_string_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 2147483661 exceeds maximum of 2147483647", huge_string_json::out_of_range&);
+        }
+
+        SECTION("document")
+        {
+            // an oversized string nested one level deep makes the
+            // *embedded* document's own length exceed INT32_MAX as well
+            huge_string_json nested;
+            nested["s"] = huge_string_t::as_huge("value");
+            huge_string_json j;
+            j["nested"] = nested;
+
+            CHECK_THROWS_WITH_AS(huge_string_json::to_bson(j), "[json.exception.out_of_range.412] BSON length 2147483674 exceeds maximum of 2147483647", huge_string_json::out_of_range&);
+        }
+    }
+
     SECTION("string length must be at least 1")
     {
         // from https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=11175
@@ -193,6 +267,23 @@ TEST_CASE("BSON")
             CHECK(json::from_bson(result, true, false) == j);
         }
 
+        SECTION("non-empty object with bool from a non-0/1 byte (lenient parsing)")
+        {
+            // documented lenient behavior (see gh-5333): any non-zero byte
+            // is accepted as `true`, not just 0x01
+            std::vector<std::uint8_t> const input =
+            {
+                0x0D, 0x00, 0x00, 0x00, // size (little endian)
+                0x08,               // entry: boolean
+                'e', 'n', 't', 'r', 'y', '\x00',
+                0x02,           // value = 0x02 (neither 0x00 nor 0x01)
+                0x00                    // end marker
+            };
+
+            const json expected = { { "entry", true } };
+            CHECK(json::from_bson(input) == expected);
+        }
+
         SECTION("non-empty object with double")
         {
             json const j =
@@ -499,6 +590,29 @@ TEST_CASE("BSON")
             CHECK(json::from_bson(result, true, false) == j);
         }
 
+        SECTION("array elements with non-conforming keys (lenient parsing)")
+        {
+            // documented lenient behavior (see gh-5333): BSON array element
+            // keys are not checked against the required decimal sequence
+            // "0", "1", "2", ... - elements are taken in encoded order
+            std::vector<std::uint8_t> const input =
+            {
+                0x26, 0x00, 0x00, 0x00, // size (little endian)
+                0x04, 'e', 'n', 't', 'r', 'y', '\x00', // entry: embedded array
+
+                0x1A, 0x00, 0x00, 0x00, // size (little endian)
+                0x10, '5', 0x00, 0x0A, 0x00, 0x00, 0x00, // key "5" (bogus)      -> 10
+                0x10, 'x', 0x00, 0x14, 0x00, 0x00, 0x00, // key "x" (non-numeric) -> 20
+                0x10, '1', 0x00, 0x1E, 0x00, 0x00, 0x00, // key "1" (out of order) -> 30
+                0x00, // end marker (embedded array)
+
+                0x00 // end marker
+            };
+
+            const json expected = { { "entry", json::array({10, 20, 30}) } };
+            CHECK(json::from_bson(input) == expected);
+        }
+
         SECTION("non-empty object with binary member")
         {
             const size_t N = 10;
@@ -594,6 +708,31 @@ TEST_CASE("BSON")
             CHECK(json::from_bson(result, true, false) == j);
         }
 
+        SECTION("binary member with subtype 0x02 (old binary) keeps its inner length prefix (lenient parsing)")
+        {
+            // documented lenient behavior (see gh-5333): the payload for
+            // binary subtype 0x02 ("old binary") is returned as-is,
+            // including its own inner 4-byte length prefix; it is not
+            // stripped or reinterpreted
+            std::vector<std::uint8_t> const input =
+            {
+                0x17, 0x00, 0x00, 0x00, // size (little endian)
+                0x05, 'e', 'n', 't', 'r', 'y', '\x00', // entry: binary
+
+                0x06, 0x00, 0x00, 0x00, // size of binary (little endian)
+                0x02, // "old binary" subtype
+                0x02, 0x00, 0x00, 0x00, // inner length prefix (part of the old-binary payload)
+                0x68, 0x69, // payload ('h', 'i')
+
+                0x00 // end marker
+            };
+
+            // the inner length prefix is part of the (unmodified) payload
+            const std::vector<std::uint8_t> expected_payload = {0x02, 0x00, 0x00, 0x00, 0x68, 0x69};
+            const json expected = { { "entry", json::binary(expected_payload, 0x02) } };
+            CHECK(json::from_bson(input) == expected);
+        }
+
         SECTION("Some more complex document")
         {
             json const j =

tests/src/unit-diagnostics.cpp

@@ -273,36 +273,5 @@ TEST_CASE("Regression tests for extended diagnostics")
         CHECK(j1["numbers"]["two"] == 2);
         CHECK(j1["string"] == "t");
     }
-
-    SECTION("Regression test - swap(array_t&)/swap(object_t&) must update JSON_DIAGNOSTICS parent pointers")
-    {
-        // swap(array_t&)
-        {
-            json j = json::array();
-            json::array_t arr = {json::array({1})};
-            j.swap(arr);
-
-            // parent pointers of the moved-in elements must point into j, not
-            // into the now-defunct free-standing array_t
-            CHECK_THROWS_WITH_AS(j[0][0].get<std::string>(), "[json.exception.type_error.302] (/0/0) type must be string, but is number", json::type_error);
-
-            // must not trigger assert_invariant() in a debug/assert-enabled build
-            json const k = j;
-            CHECK(k == j);
-        }
-
-        // swap(object_t&)
-        {
-            json o = json::object();
-            json::object_t obj = {{"a", json::array({1})}};
-            o.swap(obj);
-
-            CHECK_THROWS_WITH_AS(o["a"][0].get<std::string>(), "[json.exception.type_error.302] (/a/0) type must be string, but is number", json::type_error);
-
-            // must not trigger assert_invariant() in a debug/assert-enabled build
-            json const p = o;
-            CHECK(p == o);
-        }
-    }
 }
 

tests/src/unit-ordered_json.cpp

@@ -81,84 +81,3 @@ TEST_CASE("regression test for issue #3732 - iteration_proxy_value<iter_impl<ord
     };
     static_cast<void>(fn);
 }
-
-TEST_CASE("regression test - diff() must account for ordered_json member order")
-{
-    SECTION("pure reorder, no value changes")
-    {
-        ordered_json a = {{"a", 1}, {"b", 2}};
-        ordered_json b = {{"b", 2}, {"a", 1}};
-        CHECK(a != b); // order-sensitive equality
-        CHECK(a.patch(ordered_json::diff(a, b)) == b);
-    }
-
-    SECTION("new key must land at the front")
-    {
-        ordered_json c = {{"b", 2}};
-        ordered_json e = {{"a", 1}, {"b", 2}};
-        CHECK(c.patch(ordered_json::diff(c, e)) == e);
-    }
-
-    SECTION("reorder plus a value change on one of the reordered keys")
-    {
-        ordered_json a = {{"a", 1}, {"b", 2}};
-        ordered_json b = {{"b", 20}, {"a", 1}};
-        CHECK(a != b);
-        CHECK(a.patch(ordered_json::diff(a, b)) == b);
-    }
-
-    SECTION("reorder plus a deleted key")
-    {
-        ordered_json a = {{"a", 1}, {"b", 2}, {"c", 3}};
-        ordered_json b = {{"b", 2}, {"a", 1}};
-        CHECK(a != b);
-        CHECK(a.patch(ordered_json::diff(a, b)) == b);
-    }
-
-    SECTION("reorder plus a nested value that itself needs a recursive diff")
-    {
-        ordered_json a = {{"a", {{"x", 1}, {"y", 2}}}, {"b", 2}};
-        ordered_json b = {{"b", 2}, {"a", {{"x", 1}, {"y", 99}}}};
-        CHECK(a != b);
-        CHECK(a.patch(ordered_json::diff(a, b)) == b);
-    }
-
-    SECTION("three or more keys shuffled into a different order")
-    {
-        ordered_json a = {{"a", 1}, {"b", 2}, {"c", 3}, {"d", 4}};
-        ordered_json b = {{"d", 4}, {"b", 2}, {"a", 1}, {"c", 3}};
-        CHECK(a != b);
-        CHECK(a.patch(ordered_json::diff(a, b)) == b);
-    }
-
-    SECTION("matching order still produces a minimal patch (fast path unaffected)")
-    {
-        ordered_json a = {{"a", 1}, {"b", 2}, {"c", 3}};
-        ordered_json b = {{"a", 1}, {"b", 20}, {"c", 3}};
-        auto p = ordered_json::diff(a, b);
-        // only the changed value should be touched, not a wholesale remove+add
-        CHECK(p.size() == 1);
-        CHECK(p[0]["op"] == "replace");
-        CHECK(p[0]["path"] == "/b");
-        CHECK(a.patch(p) == b);
-    }
-
-    SECTION("plain json (std::map-backed) is unaffected by same-key-different-insertion-order")
-    {
-        json a;
-        a["b"] = 2;
-        a["a"] = 1;
-
-        json b;
-        b["a"] = 1;
-        b["b"] = 2;
-
-        // std::map iteration is always sorted by key, so a == b regardless of
-        // insertion order, and diff() must still produce the same minimal
-        // (empty) result as before this fix
-        CHECK(a == b);
-        auto p = json::diff(a, b);
-        CHECK(p.empty());
-        CHECK(a.patch(p) == b);
-    }
-}

tests/src/unit-regression2.cpp

@@ -1566,75 +1566,4 @@ TEST_CASE("issue #5402 - update(merge_objects=true) overwrites a primitive with
     CHECK(mixed == json({{"keep", {{"a", 1}, {"b", 2}}}, {"replace", {{"x", 2}}}}));
 }
 
-TEST_CASE("regression test - parser callback must not lose a duplicate key's prior value")
-{
-    // a callback that rejects only the scalar value 2
-    const json::parser_callback_t drop_value_2 = [](int /*depth*/, json::parse_event_t ev, json & v) noexcept
-    {
-        return !(ev == json::parse_event_t::value && v == 2);
-    };
-
-    SECTION("duplicate key, second (scalar) value rejected - prior value is restored")
-    {
-        const json j = json::parse(R"({"a":1,"a":2})", drop_value_2);
-        CHECK(j.dump() == "{\"a\":1}");
-    }
-
-    SECTION("duplicate key, second value is an object rejected at object_end - prior value is restored")
-    {
-        const json j = json::parse(R"({"a":1,"a":{"x":2}})",
-                                   [](int depth, json::parse_event_t ev, json& /*parsed*/) noexcept
-        {
-            return !(ev == json::parse_event_t::object_end && depth == 1);
-        });
-        CHECK(j.dump() == "{\"a\":1}");
-    }
-
-    SECTION("duplicate key, second value is an array rejected at array_end - prior value is restored")
-    {
-        const json j = json::parse(R"({"a":1,"a":[9,9]})",
-                                   [](int depth, json::parse_event_t ev, json& /*parsed*/) noexcept
-        {
-            return !(ev == json::parse_event_t::array_end && depth == 1);
-        });
-        CHECK(j.dump() == "{\"a\":1}");
-    }
-
-    SECTION("duplicate key, second value accepted (scalar) - last value wins")
-    {
-        const json j = json::parse(R"({"a":1,"a":2})", [](int, json::parse_event_t, json&) noexcept
-        {
-            return true;
-        });
-        CHECK(j.dump() == "{\"a\":2}");
-    }
-
-    SECTION("duplicate key, second value accepted (object) - last value wins")
-    {
-        const json j = json::parse(R"({"a":1,"a":{"x":2}})", [](int, json::parse_event_t, json&) noexcept
-        {
-            return true;
-        });
-        CHECK(j.dump() == "{\"a\":{\"x\":2}}");
-    }
-
-    SECTION("brand new (non-duplicate) key, value rejected - member is fully absent")
-    {
-        const json j = json::parse(R"({"a":1,"b":2})", drop_value_2);
-        CHECK(j.dump() == "{\"a\":1}");
-    }
-
-    SECTION("duplicate key nested two levels deep")
-    {
-        const json j = json::parse(R"({"outer":{"a":1,"a":2}})", drop_value_2);
-        CHECK(j.dump() == "{\"outer\":{\"a\":1}}");
-    }
-
-    SECTION("three occurrences of the same key - middle rejected, last accepted")
-    {
-        const json j = json::parse(R"({"k":1,"k":2,"k":3})", drop_value_2);
-        CHECK(j.dump() == "{\"k\":3}");
-    }
-}
-
 DOCTEST_CLANG_SUPPRESS_WARNING_POP