From e59bb63efbf3b219dae958bcb56fa393af125b7f Mon Sep 17 00:00:00 2001
From: robot-clickhouse <robot-clickhouse@users.noreply.github.com>
Date: Mon, 7 Oct 2024 06:09:22 +0000
Subject: [PATCH] Backport #70172 to 24.8: Fix crash in JSON column

---
 src/Columns/ColumnObject.cpp                  | 23 +++++++++++++++++++
 src/Columns/ColumnObject.h                    |  2 ++
 .../03247_object_column_copy.reference        |  0
 .../0_stateless/03247_object_column_copy.sql  |  9 ++++++++
 4 files changed, 34 insertions(+)
 create mode 100644 tests/queries/0_stateless/03247_object_column_copy.reference
 create mode 100644 tests/queries/0_stateless/03247_object_column_copy.sql

diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp
index 3577ab1ec82..70eed4788fe 100644
--- a/src/Columns/ColumnObject.cpp
+++ b/src/Columns/ColumnObject.cpp
@@ -87,6 +87,29 @@ ColumnObject::ColumnObject(
     shared_data = ColumnArray::create(ColumnTuple::create(std::move(paths_and_values)));
 }
 
+ColumnObject::ColumnObject(const ColumnObject & other)
+    : COWHelper<IColumnHelper<ColumnObject>, ColumnObject>(other)
+    , typed_paths(other.typed_paths)
+    , dynamic_paths(other.dynamic_paths)
+    , dynamic_paths_ptrs(other.dynamic_paths_ptrs)
+    , shared_data(other.shared_data)
+    , max_dynamic_paths(other.max_dynamic_paths)
+    , global_max_dynamic_paths(other.global_max_dynamic_paths)
+    , max_dynamic_types(other.max_dynamic_types)
+    , statistics(other.statistics)
+{
+    /// We should update string_view in sorted_typed_paths and sorted_dynamic_paths so they
+    /// point to the new strings in typed_paths and dynamic_paths.
+    sorted_typed_paths.clear();
+    for (const auto & [path, _] : typed_paths)
+        sorted_typed_paths.emplace_back(path);
+    std::sort(sorted_typed_paths.begin(), sorted_typed_paths.end());
+
+    sorted_dynamic_paths.clear();
+    for (const auto & [path, _] : dynamic_paths)
+        sorted_dynamic_paths.emplace(path);
+}
+
 ColumnObject::Ptr ColumnObject::create(
     const std::unordered_map<String, ColumnPtr> & typed_paths_,
     const std::unordered_map<String, ColumnPtr> & dynamic_paths_,
diff --git a/src/Columns/ColumnObject.h b/src/Columns/ColumnObject.h
index c7f282d9079..9c5ee4fa089 100644
--- a/src/Columns/ColumnObject.h
+++ b/src/Columns/ColumnObject.h
@@ -54,6 +54,8 @@ private:
         size_t max_dynamic_types_,
         const StatisticsPtr & statistics_ = {});
 
+    ColumnObject(const ColumnObject & other);
+
     /// Use StringHashForHeterogeneousLookup hash for hash maps to be able to use std::string_view in find() method.
     using PathToColumnMap = std::unordered_map<String, WrappedPtr, StringHashForHeterogeneousLookup, StringHashForHeterogeneousLookup::transparent_key_equal>;
     using PathToDynamicColumnPtrMap = std::unordered_map<String, ColumnDynamic *, StringHashForHeterogeneousLookup, StringHashForHeterogeneousLookup::transparent_key_equal>;
diff --git a/tests/queries/0_stateless/03247_object_column_copy.reference b/tests/queries/0_stateless/03247_object_column_copy.reference
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/queries/0_stateless/03247_object_column_copy.sql b/tests/queries/0_stateless/03247_object_column_copy.sql
new file mode 100644
index 00000000000..37c58217b1d
--- /dev/null
+++ b/tests/queries/0_stateless/03247_object_column_copy.sql
@@ -0,0 +1,9 @@
+SET allow_experimental_json_type = 1;
+SET allow_experimental_variant_type = 1;
+DROP TABLE IF EXISTS t0;
+CREATE TABLE t0 (c0 Int) ENGINE = Memory();
+INSERT INTO t0 (c0) VALUES (1);
+ALTER TABLE t0 (ADD COLUMN c1 JSON(c1 Variant(Int,JSON(c1 Int))));
+INSERT INTO t0 (c0, c1) VALUES (2, '{"c1":1}'::JSON);
+SELECT kafkaMurmurHash(c1) FROM t0; -- {serverError NOT_IMPLEMENTED}
+DROP TABLE t0;