Update vendored DuckDB sources to 51070f2fb6

CMakeLists.txt

@@ -340,46 +340,46 @@ set(DUCKDB_SRC_FILES
   src/duckdb/third_party/zstd/dict/divsufsort.cpp
   src/duckdb/third_party/zstd/dict/fastcover.cpp
   src/duckdb/third_party/zstd/dict/zdict.cpp
+  src/duckdb/extension/core_functions/function_list.cpp
   src/duckdb/extension/core_functions/core_functions_extension.cpp
   src/duckdb/extension/core_functions/lambda_functions.cpp
-  src/duckdb/extension/core_functions/function_list.cpp
-  src/duckdb/ub_extension_core_functions_aggregate_nested.cpp
-  src/duckdb/ub_extension_core_functions_aggregate_holistic.cpp
-  src/duckdb/ub_extension_core_functions_aggregate_regression.cpp
-  src/duckdb/ub_extension_core_functions_aggregate_algebraic.cpp
-  src/duckdb/ub_extension_core_functions_aggregate_distributive.cpp
   src/duckdb/ub_extension_core_functions_scalar_string.cpp
-  src/duckdb/ub_extension_core_functions_scalar_enum.cpp
-  src/duckdb/ub_extension_core_functions_scalar_blob.cpp
-  src/duckdb/ub_extension_core_functions_scalar_map.cpp
-  src/duckdb/ub_extension_core_functions_scalar_array.cpp
-  src/duckdb/ub_extension_core_functions_scalar_bit.cpp
   src/duckdb/ub_extension_core_functions_scalar_union.cpp
+  src/duckdb/ub_extension_core_functions_scalar_blob.cpp
   src/duckdb/ub_extension_core_functions_scalar_debug.cpp
-  src/duckdb/ub_extension_core_functions_scalar_struct.cpp
   src/duckdb/ub_extension_core_functions_scalar_random.cpp
-  src/duckdb/ub_extension_core_functions_scalar_date.cpp
+  src/duckdb/ub_extension_core_functions_scalar_operators.cpp
   src/duckdb/ub_extension_core_functions_scalar_list.cpp
   src/duckdb/ub_extension_core_functions_scalar_math.cpp
-  src/duckdb/ub_extension_core_functions_scalar_operators.cpp
+  src/duckdb/ub_extension_core_functions_scalar_enum.cpp
+  src/duckdb/ub_extension_core_functions_scalar_array.cpp
+  src/duckdb/ub_extension_core_functions_scalar_date.cpp
+  src/duckdb/ub_extension_core_functions_scalar_map.cpp
   src/duckdb/ub_extension_core_functions_scalar_generic.cpp
+  src/duckdb/ub_extension_core_functions_scalar_bit.cpp
+  src/duckdb/ub_extension_core_functions_scalar_struct.cpp
+  src/duckdb/ub_extension_core_functions_aggregate_regression.cpp
+  src/duckdb/ub_extension_core_functions_aggregate_distributive.cpp
+  src/duckdb/ub_extension_core_functions_aggregate_nested.cpp
+  src/duckdb/ub_extension_core_functions_aggregate_holistic.cpp
+  src/duckdb/ub_extension_core_functions_aggregate_algebraic.cpp
   src/duckdb/extension/parquet/parquet_extension.cpp
-  src/duckdb/extension/parquet/parquet_file_metadata_cache.cpp
-  src/duckdb/extension/parquet/serialize_parquet.cpp
-  src/duckdb/extension/parquet/parquet_metadata.cpp
-  src/duckdb/extension/parquet/parquet_crypto.cpp
-  src/duckdb/extension/parquet/parquet_float16.cpp
+  src/duckdb/extension/parquet/zstd_file_system.cpp
+  src/duckdb/extension/parquet/geo_parquet.cpp
   src/duckdb/extension/parquet/parquet_writer.cpp
-  src/duckdb/extension/parquet/parquet_statistics.cpp
-  src/duckdb/extension/parquet/column_reader.cpp
-  src/duckdb/extension/parquet/parquet_reader.cpp
   src/duckdb/extension/parquet/column_writer.cpp
-  src/duckdb/extension/parquet/geo_parquet.cpp
   src/duckdb/extension/parquet/parquet_timestamp.cpp
+  src/duckdb/extension/parquet/parquet_crypto.cpp
+  src/duckdb/extension/parquet/parquet_file_metadata_cache.cpp
+  src/duckdb/extension/parquet/parquet_reader.cpp
   src/duckdb/extension/parquet/parquet_multi_file_info.cpp
-  src/duckdb/extension/parquet/zstd_file_system.cpp
-  src/duckdb/ub_extension_parquet_decoder.cpp
+  src/duckdb/extension/parquet/column_reader.cpp
+  src/duckdb/extension/parquet/serialize_parquet.cpp
+  src/duckdb/extension/parquet/parquet_float16.cpp
+  src/duckdb/extension/parquet/parquet_metadata.cpp
+  src/duckdb/extension/parquet/parquet_statistics.cpp
   src/duckdb/ub_extension_parquet_writer.cpp
+  src/duckdb/ub_extension_parquet_decoder.cpp
   src/duckdb/ub_extension_parquet_reader.cpp
   src/duckdb/third_party/parquet/parquet_types.cpp
   src/duckdb/third_party/thrift/thrift/protocol/TProtocol.cpp
@@ -419,32 +419,32 @@ set(DUCKDB_SRC_FILES
   src/duckdb/third_party/brotli/enc/metablock.cpp
   src/duckdb/third_party/brotli/enc/static_dict.cpp
   src/duckdb/third_party/brotli/enc/utf8_util.cpp
-  src/duckdb/extension/icu/./icu-datesub.cpp
-  src/duckdb/extension/icu/./icu_extension.cpp
+  src/duckdb/extension/icu/./icu-makedate.cpp
+  src/duckdb/extension/icu/./icu-datefunc.cpp
+  src/duckdb/extension/icu/./icu-timezone.cpp
+  src/duckdb/extension/icu/./icu-current.cpp
   src/duckdb/extension/icu/./icu-list-range.cpp
-  src/duckdb/extension/icu/./icu-strptime.cpp
   src/duckdb/extension/icu/./icu-dateadd.cpp
   src/duckdb/extension/icu/./icu-table-range.cpp
-  src/duckdb/extension/icu/./icu-current.cpp
+  src/duckdb/extension/icu/./icu-datesub.cpp
+  src/duckdb/extension/icu/./icu-datetrunc.cpp
   src/duckdb/extension/icu/./icu-timebucket.cpp
-  src/duckdb/extension/icu/./icu-timezone.cpp
   src/duckdb/extension/icu/./icu-datepart.cpp
-  src/duckdb/extension/icu/./icu-datefunc.cpp
-  src/duckdb/extension/icu/./icu-makedate.cpp
-  src/duckdb/extension/icu/./icu-datetrunc.cpp
+  src/duckdb/extension/icu/./icu-strptime.cpp
+  src/duckdb/extension/icu/./icu_extension.cpp
   src/duckdb/ub_extension_icu_third_party_icu_common.cpp
   src/duckdb/ub_extension_icu_third_party_icu_i18n.cpp
   src/duckdb/extension/icu/third_party/icu/stubdata/stubdata.cpp
-  src/duckdb/extension/json/json_extension.cpp
   src/duckdb/extension/json/json_enums.cpp
-  src/duckdb/extension/json/json_multi_file_info.cpp
-  src/duckdb/extension/json/json_reader.cpp
+  src/duckdb/extension/json/json_scan.cpp
   src/duckdb/extension/json/json_serializer.cpp
   src/duckdb/extension/json/json_common.cpp
-  src/duckdb/extension/json/json_functions.cpp
-  src/duckdb/extension/json/serialize_json.cpp
-  src/duckdb/extension/json/json_scan.cpp
+  src/duckdb/extension/json/json_multi_file_info.cpp
   src/duckdb/extension/json/json_deserializer.cpp
+  src/duckdb/extension/json/json_reader.cpp
+  src/duckdb/extension/json/serialize_json.cpp
+  src/duckdb/extension/json/json_functions.cpp
+  src/duckdb/extension/json/json_extension.cpp
   src/duckdb/ub_extension_json_json_functions.cpp)
 
 set(JEMALLOC_SRC_FILES 

src/duckdb/extension/core_functions/scalar/random/random.cpp

@@ -29,7 +29,9 @@ struct ExtractTimestampUuidOperator {
 		}
 
 		// UUID v7 begins with a 48 bit big-endian Unix Epoch timestamp with millisecond granularity.
-		const int64_t upper = input.upper;
+		int64_t upper = input.upper;
+		// flip the top byte
+		upper ^= NumericLimits<int64_t>::Minimum();
 		int64_t unix_ts_milli = upper;
 		unix_ts_milli = unix_ts_milli >> 16;
 

src/duckdb/extension/icu/icu-timezone.cpp

@@ -247,6 +247,10 @@ struct ICUToNaiveTimestamp : public ICUDateFunc {
 		if (!input.context) {
 			throw InternalException("Missing context for TIMESTAMPTZ to TIMESTAMP cast.");
 		}
+		if (input.context->config.disable_timestamptz_casts) {
+			throw BinderException("Casting from TIMESTAMP WITH TIME ZONE to TIMESTAMP without an explicit time zone "
+			                      "has been disabled  - use \"AT TIME ZONE ...\"");
+		}
 
 		auto cast_data = make_uniq<CastData>(make_uniq<BindData>(*input.context));
 

src/duckdb/extension/parquet/include/writer/templated_column_writer.hpp

@@ -403,7 +403,7 @@ class StandardColumnWriter : public PrimitiveColumnWriter {
 			break;
 		}
 		case duckdb_parquet::Encoding::BYTE_STREAM_SPLIT: {
-			if (page_state.bss_initialized) {
+			if (!page_state.bss_initialized) {
 				page_state.bss_encoder.BeginWrite(BufferAllocator::Get(writer.GetContext()));
 				page_state.bss_initialized = true;
 			}

src/duckdb/extension/parquet/parquet_float16.cpp

@@ -11,7 +11,9 @@ float Float16ToFloat32(const uint16_t &float16_value) {
 	uint32_t sign = float16_value >> 15;
 	uint32_t exponent = (float16_value >> 10) & 0x1F;
 	uint32_t fraction = (float16_value & 0x3FF);
-	uint32_t float32_value;
+	// Avoid strict aliasing issues and compiler warnings
+	uint32_t float32_value = 0;
+
 	if (exponent == 0) {
 		if (fraction == 0) {
 			// zero
@@ -39,7 +41,7 @@ float Float16ToFloat32(const uint16_t &float16_value) {
 		float32_value = (sign << 31) | ((exponent + (127 - 15)) << 23) | (fraction << 13);
 	}
 
-	return *reinterpret_cast<float *>(&float32_value);
+	return Load<float>(const_data_ptr_cast(&float32_value));
 }
 
 } // namespace duckdb

src/duckdb/extension/parquet/parquet_reader.cpp

@@ -1116,7 +1116,7 @@ void ParquetReader::InitializeScan(ClientContext &context, ParquetReaderScanStat
 			state.prefetch_mode = false;
 		}
 
-		state.file_handle = fs.OpenFile(file_handle->GetPath(), flags);
+		state.file_handle = fs.OpenFile(file, flags);
 	}
 	state.adaptive_filter.reset();
 	state.scan_filters.clear();

src/duckdb/extension/parquet/reader/struct_column_reader.cpp

@@ -118,12 +118,21 @@ static bool TypeHasExactRowCount(const LogicalType &type) {
 }
 
 idx_t StructColumnReader::GroupRowsAvailable() {
-	for (idx_t i = 0; i < child_readers.size(); i++) {
-		if (TypeHasExactRowCount(child_readers[i]->Type())) {
-			return child_readers[i]->GroupRowsAvailable();
+	for (auto &child : child_readers) {
+		if (!child) {
+			continue;
+		}
+		if (TypeHasExactRowCount(child->Type())) {
+			return child->GroupRowsAvailable();
+		}
+	}
+	for (auto &child : child_readers) {
+		if (!child) {
+			continue;
 		}
+		return child->GroupRowsAvailable();
 	}
-	return child_readers[0]->GroupRowsAvailable();
+	throw InternalException("No projected columns in struct?");
 }
 
 } // namespace duckdb

src/duckdb/src/common/local_file_system.cpp

@@ -838,21 +838,21 @@ static string AdditionalLockInfo(const std::wstring path) {
 
 	status = RmStartSession(&session, 0, session_key);
 	if (status != ERROR_SUCCESS) {
-		return "";
+		return string();
 	}
 
 	PCWSTR path_ptr = path.c_str();
 	status = RmRegisterResources(session, 1, &path_ptr, 0, NULL, 0, NULL);
 	if (status != ERROR_SUCCESS) {
-		return "";
+		return string();
 	}
 	UINT process_info_size_needed, process_info_size;
 
 	// we first call with nProcInfo = 0 to find out how much to allocate
 	process_info_size = 0;
 	status = RmGetList(session, &process_info_size_needed, &process_info_size, NULL, &reason);
 	if (status != ERROR_MORE_DATA || process_info_size_needed == 0) {
-		return "";
+		return string();
 	}
 
 	// allocate
@@ -866,8 +866,7 @@ static string AdditionalLockInfo(const std::wstring path) {
 		return "";
 	}
 
-	string conflict_string = "File is already open in ";
-
+	string conflict_string;
 	for (UINT process_idx = 0; process_idx < process_info_size; process_idx++) {
 		string process_name = WindowsUtil::UnicodeToUTF8(process_info[process_idx].strAppName);
 		auto pid = process_info[process_idx].Process.dwProcessId;
@@ -886,7 +885,10 @@ static string AdditionalLockInfo(const std::wstring path) {
 	}
 
 	RmEndSession(session);
-	return conflict_string;
+	if (conflict_string.empty()) {
+		return string();
+	}
+	return "File is already open in " + conflict_string;
 }
 
 bool LocalFileSystem::IsPrivateFile(const string &path_p, FileOpener *opener) {
@@ -950,12 +952,11 @@ unique_ptr<FileHandle> LocalFileSystem::OpenFile(const string &path_p, FileOpenF
 		}
 		auto error = LocalFileSystem::GetLastErrorAsString();
 
-		auto better_error = AdditionalLockInfo(unicode_path);
-		if (!better_error.empty()) {
-			throw IOException(better_error);
-		} else {
-			throw IOException("Cannot open file \"%s\": %s", path.c_str(), error);
+		auto extended_error = AdditionalLockInfo(unicode_path);
+		if (!extended_error.empty()) {
+			extended_error = "\n" + extended_error;
 		}
+		throw IOException("Cannot open file \"%s\": %s%s", path.c_str(), error, extended_error);
 	}
 	auto handle = make_uniq<WindowsFileHandle>(*this, path.c_str(), hFile, flags);
 	if (flags.OpenForAppending()) {

src/duckdb/src/common/multi_file/multi_file_column_mapper.cpp

@@ -297,6 +297,19 @@ ColumnMapResult MapColumnList(ClientContext &context, const MultiFileColumnDefin
 			result.column_map = Value::STRUCT(std::move(child_list));
 		}
 	}
+	if (is_selected && child_map.default_value) {
+		// we have default values at a previous level wrap it in a "list"
+		child_list_t<LogicalType> default_type_list;
+		default_type_list.emplace_back("list", child_map.default_value->return_type);
+		vector<unique_ptr<Expression>> default_expressions;
+		child_map.default_value->alias = "list";
+		default_expressions.push_back(std::move(child_map.default_value));
+		auto default_type = LogicalType::STRUCT(std::move(default_type_list));
+		auto struct_pack_fun = StructPackFun::GetFunction();
+		auto bind_data = make_uniq<VariableReturnBindData>(default_type);
+		result.default_value = make_uniq<BoundFunctionExpression>(std::move(default_type), std::move(struct_pack_fun),
+		                                                          std::move(default_expressions), std::move(bind_data));
+	}
 	result.column_index = make_uniq<ColumnIndex>(local_id.GetId(), std::move(child_indexes));
 	result.mapping = std::move(mapping);
 	return result;
@@ -347,7 +360,7 @@ ColumnMapResult MapColumnMap(ClientContext &context, const MultiFileColumnDefini
 
 	auto nested_mapper = mapper.Create(local_key_value.children);
 	child_list_t<Value> column_mapping;
-	unique_ptr<Expression> default_expression;
+	vector<unique_ptr<Expression>> default_expressions;
 	unordered_map<idx_t, const_reference<ColumnIndex>> selected_children;
 	if (global_index.HasChildren()) {
 		//! FIXME: is this expected for maps??
@@ -378,6 +391,10 @@ ColumnMapResult MapColumnMap(ClientContext &context, const MultiFileColumnDefini
 			// found a column mapping for the component - emplace it
 			column_mapping.emplace_back(name, std::move(map_result.column_map));
 		}
+		if (map_result.default_value) {
+			map_result.default_value->alias = name;
+			default_expressions.push_back(std::move(map_result.default_value));
+		}
 	}
 
 	ColumnMapResult result;
@@ -393,6 +410,18 @@ ColumnMapResult MapColumnMap(ClientContext &context, const MultiFileColumnDefini
 			result.column_map = Value::STRUCT(std::move(child_list));
 		}
 	}
+	if (!default_expressions.empty()) {
+		// we have default values at a previous level wrap it in a "list"
+		child_list_t<LogicalType> default_type_list;
+		for (auto &expr : default_expressions) {
+			default_type_list.emplace_back(expr->GetAlias(), expr->return_type);
+		}
+		auto default_type = LogicalType::STRUCT(std::move(default_type_list));
+		auto struct_pack_fun = StructPackFun::GetFunction();
+		auto bind_data = make_uniq<VariableReturnBindData>(default_type);
+		result.default_value = make_uniq<BoundFunctionExpression>(std::move(default_type), std::move(struct_pack_fun),
+		                                                          std::move(default_expressions), std::move(bind_data));
+	}
 	vector<ColumnIndex> map_indexes;
 	map_indexes.emplace_back(0, std::move(child_indexes));
 
@@ -563,11 +592,6 @@ unique_ptr<Expression> ConstructMapExpression(ClientContext &context, idx_t loca
 		}
 		return expr;
 	}
-	// struct column - generate a remap_struct - but only if we have any columns to remap
-	if (mapping.column_map.IsNull()) {
-		// no columns to map - emit the default value directly
-		return std::move(mapping.default_value);
-	}
 	// generate the remap_struct function call
 	vector<unique_ptr<Expression>> children;
 	children.push_back(std::move(expr));

src/duckdb/src/common/types.cpp

@@ -648,6 +648,34 @@ bool LogicalType::IsIntegral() const {
 	}
 }
 
+bool LogicalType::IsSigned() const {
+	switch (id_) {
+	case LogicalTypeId::TINYINT:
+	case LogicalTypeId::SMALLINT:
+	case LogicalTypeId::INTEGER:
+	case LogicalTypeId::BIGINT:
+	case LogicalTypeId::HUGEINT:
+		return true;
+	default:
+		break;
+	}
+	return false;
+}
+
+bool LogicalType::IsUnsigned() const {
+	switch (id_) {
+	case LogicalTypeId::UTINYINT:
+	case LogicalTypeId::USMALLINT:
+	case LogicalTypeId::UINTEGER:
+	case LogicalTypeId::UBIGINT:
+	case LogicalTypeId::UHUGEINT:
+		return true;
+	default:
+		break;
+	}
+	return false;
+}
+
 bool LogicalType::IsFloating() const {
 	switch (id_) {
 	case LogicalTypeId::FLOAT:

src/duckdb/src/common/types/row/tuple_data_segment.cpp

@@ -26,7 +26,7 @@ static inline void SwapTupleDataChunk(TupleDataChunk &a, TupleDataChunk &b) noex
 	std::swap(a.lock, b.lock);
 }
 
-TupleDataChunk::TupleDataChunk(TupleDataChunk &&other) noexcept {
+TupleDataChunk::TupleDataChunk(TupleDataChunk &&other) noexcept : count(0) {
 	SwapTupleDataChunk(*this, other);
 }
 

src/duckdb/src/common/types/uuid.cpp

@@ -210,7 +210,10 @@ hugeint_t UUIDv7::GenerateRandomUUID(RandomEngine &engine) {
 	// Fill in variant field.
 	bytes[8] = (bytes[8] & 0x3f) | 0x80;
 
-	return Convert(bytes);
+	// Flip the top byte
+	auto result = Convert(bytes);
+	result.upper ^= NumericLimits<int64_t>::Minimum();
+	return result;
 }
 
 hugeint_t UUIDv7::GenerateRandomUUID() {