Update vendored DuckDB sources to 5bf940c1318

src/duckdb/extension/core_functions/scalar/date/date_part.cpp

@@ -2671,9 +2671,8 @@ ScalarFunctionSet HoursFun::GetFunctions() {
 }
 
 ScalarFunctionSet YearWeekFun::GetFunctions() {
-	auto set = GetDatePartFunction<DatePart::YearWeekOperator>();
-	set.SetUnaryArgProperties(ArgProperties().NonDecreasing());
-	return set;
+	// Not monotone on the full date domain: see YearWeekOperator::YearWeekFromParts.
+	return GetDatePartFunction<DatePart::YearWeekOperator>();
 }
 
 ScalarFunctionSet DayOfMonthFun::GetFunctions() {

src/duckdb/extension/json/json_reader.cpp

@@ -1171,6 +1171,16 @@ void JSONReader::PrepareForReadInternal(JSONReaderScanState &scan_state) {
 	if (!scan_state.read_buffer.IsSet()) {
 		scan_state.read_buffer = scan_state.global_allocator.Allocate(scan_state.buffer_capacity);
 		scan_state.buffer_ptr = char_ptr_cast(scan_state.read_buffer.get());
+	} else if (scan_state.read_buffer.GetSize() < scan_state.buffer_capacity) {
+		// the retained auto-detect buffer only holds what was read from the file - it cannot take a full read
+		auto buffer = scan_state.global_allocator.Allocate(scan_state.buffer_capacity);
+		if (scan_state.file_read_type == JSONFileReadType::SCAN_ENTIRE_FILE) {
+			memcpy(buffer.get(), scan_state.buffer_ptr + scan_state.prev_buffer_offset,
+			       scan_state.prev_buffer_remainder);
+		}
+		scan_state.read_buffer = std::move(buffer);
+		scan_state.buffer_ptr = char_ptr_cast(scan_state.read_buffer.get());
+		return;
 	}
 	if (scan_state.file_read_type == JSONFileReadType::SCAN_ENTIRE_FILE) {
 		// Copy last bit of previous buffer to the beginning if we are doing a single-threaded read
@@ -1256,6 +1266,7 @@ void JSONReader::ReadNextBufferSeek(JSONReaderScanState &scan_state) {
 		}
 
 		// Now read the file lock-free!
+		D_ASSERT(read_offset + scan_state.read_size + YYJSON_PADDING_SIZE <= scan_state.read_buffer.GetSize());
 		file_handle.ReadAtPosition(scan_state.buffer_ptr + read_offset, scan_state.read_size, scan_state.read_position,
 		                           scan_state.thread_local_filehandle);
 	}
@@ -1284,6 +1295,7 @@ bool JSONReader::ReadNextBufferNoSeek(JSONReaderScanState &scan_state) {
 	}
 	scan_state.buffer_index = GetBufferIndex();
 	PrepareForReadInternal(scan_state);
+	D_ASSERT(read_offset + request_size + YYJSON_PADDING_SIZE <= scan_state.read_buffer.GetSize());
 	if (!file_handle.Read(scan_state.buffer_ptr + read_offset, read_size, request_size)) {
 		return false; // Couldn't read anything
 	}

src/duckdb/generated_extension_loader_package_build.cpp

@@ -45,7 +45,7 @@ int32_t duckdb_extension_core_functions_describe(duckdb_extension_descriptor *de
 	descriptor->version = 1;
 	descriptor->name = "core_functions";
 	descriptor->extension_version = EXT_VERSION_CORE_FUNCTIONS;
-	descriptor->api_version = "v2.0.0-alpha48780";
+	descriptor->api_version = "v2.0.0-alpha48877";
 	descriptor->entry_cpp = (void (*)(void))core_functions_duckdb_cpp_init;
 	return 0;
 }
@@ -78,7 +78,7 @@ int32_t duckdb_extension_parquet_describe(duckdb_extension_descriptor *descripto
 	descriptor->version = 1;
 	descriptor->name = "parquet";
 	descriptor->extension_version = EXT_VERSION_PARQUET;
-	descriptor->api_version = "v2.0.0-alpha48780";
+	descriptor->api_version = "v2.0.0-alpha48877";
 	descriptor->entry_cpp = (void (*)(void))parquet_duckdb_cpp_init;
 	return 0;
 }
@@ -111,7 +111,7 @@ int32_t duckdb_extension_icu_describe(duckdb_extension_descriptor *descriptor) {
 	descriptor->version = 1;
 	descriptor->name = "icu";
 	descriptor->extension_version = EXT_VERSION_ICU;
-	descriptor->api_version = "v2.0.0-alpha48780";
+	descriptor->api_version = "v2.0.0-alpha48877";
 	descriptor->entry_cpp = (void (*)(void))icu_duckdb_cpp_init;
 	return 0;
 }
@@ -144,7 +144,7 @@ int32_t duckdb_extension_json_describe(duckdb_extension_descriptor *descriptor)
 	descriptor->version = 1;
 	descriptor->name = "json";
 	descriptor->extension_version = EXT_VERSION_JSON;
-	descriptor->api_version = "v2.0.0-alpha48780";
+	descriptor->api_version = "v2.0.0-alpha48877";
 	descriptor->entry_cpp = (void (*)(void))json_duckdb_cpp_init;
 	return 0;
 }

src/duckdb/src/common/file_system.cpp

@@ -288,6 +288,16 @@ string FileSystem::GetWorkingDirectory() {
 #endif
 
 string FileSystem::JoinPath(const string &a, const string &b) {
+	// A remote URL may carry a "?query" suffix (object version, requester-pays, an SSE-C key, ...).
+	// Appending a path segment after it would bury the query mid-path and corrupt both the query and
+	// the object path, so join onto the path without the query and keep the query at the very end.
+	auto scheme = a.find("://");
+	if (scheme != string::npos) {
+		auto query = a.find('?', scheme + 3);
+		if (query != string::npos) {
+			return Path::FromString(a.substr(0, query)).Join(b).ToString() + a.substr(query);
+		}
+	}
 	return Path::FromString(a).Join(b).ToString();
 }
 

src/duckdb/src/function/scalar/string/like.cpp

@@ -40,11 +40,8 @@ static bool GlobMatchesBracket(char s, const char *pattern, idx_t plen, idx_t &p
 			break;
 		}
 		bool matches;
-		if (pattern[pidx + 1] == '-') {
+		if (pattern[pidx + 1] == '-' && pidx + 2 < plen && pattern[pidx + 2] != ']') {
 			// range! find the next character in the range
-			if (pidx + 2 == plen) {
-				break;
-			}
 			auto next_char = pattern[pidx + 2];
 			// check if the current character is within the range
 			matches = s >= p && s <= next_char;

src/duckdb/src/function/table/arrow_conversion.cpp

@@ -1459,6 +1459,27 @@ static bool CanContainNull(const ArrowArray &array, const ValidityMask *parent_m
 	return parent_mask->CanHaveNull();
 }
 
+// Arrow's null type converts to a constant vector, and FlatVector::SetNull needs flat vectors along its recursion
+// into STRUCT and ARRAY children, which never reaches LIST contents.
+static void FlattenForSetNull(Vector &vector) {
+	if (vector.GetVectorType() != VectorType::FLAT_VECTOR) {
+		vector.Flatten();
+		return;
+	}
+	switch (vector.GetType().InternalType()) {
+	case PhysicalType::STRUCT:
+		for (auto &entry : StructVector::GetEntries(vector)) {
+			FlattenForSetNull(entry);
+		}
+		break;
+	case PhysicalType::ARRAY:
+		FlattenForSetNull(ArrayVector::GetChildMutable(vector));
+		break;
+	default:
+		break;
+	}
+}
+
 void ArrowToDuckDBConversion::ColumnArrowToDuckDBDictionary(Vector &vector, ArrowArray &array, idx_t chunk_offset,
                                                             ArrowArrayScanState &array_state, idx_t size,
                                                             const ArrowType &arrow_type, int64_t nested_offset,
@@ -1476,7 +1497,6 @@ void ArrowToDuckDBConversion::ColumnArrowToDuckDBDictionary(Vector &vector, Arro
 		auto dict_length = NumericCast<idx_t>(array.dictionary->length);
 		auto base_vector = make_uniq<Vector>(vector.GetType(), dict_length + 1);
 		ArrowToDuckDBConversion::SetValidityMask(*base_vector, *array.dictionary, 0, dict_length, 0, 0, has_nulls);
-		FlatVector::ValidityMutable(*base_vector).SetInvalid(dict_length);
 		auto &dictionary_type = arrow_type.GetDictionary();
 		auto arrow_physical_type = dictionary_type.GetPhysicalType();
 		;
@@ -1502,6 +1522,9 @@ void ArrowToDuckDBConversion::ColumnArrowToDuckDBDictionary(Vector &vector, Arro
 		// dict_length. Re-extend before sizing so the sentinel slot stays in bounds.
 		base_vector->Reserve(dict_length + 1);
 		FlatVector::SetSize(*base_vector, count_t(dict_length + 1));
+		// nested children of the sentinel slot hold uninitialized data unless they are marked NULL as well
+		FlattenForSetNull(*base_vector);
+		FlatVector::SetNull(*base_vector, dict_length, true);
 		array_state.AddDictionary(std::move(base_vector), array.dictionary);
 	}
 	auto offset_type = arrow_type.GetDuckType();

src/duckdb/src/function/table/system/duckdb_settings.cpp

@@ -29,6 +29,12 @@ struct DuckDBSettingsData : public GlobalTableFunctionState {
 struct DuckDBSettingsBindData : public TableFunctionData {
 	bool debug = false;
 	bool deprecated = false;
+	//! Ignore the debug and deprecated filters
+	bool include_all = false;
+
+	bool Includes(bool is_debug, bool is_deprecated) const {
+		return include_all || (debug == is_debug && deprecated == is_deprecated);
+	}
 };
 
 static unique_ptr<FunctionData> DuckDBSettingsBind(ClientContext &context, TableFunctionBindInput &input,
@@ -55,7 +61,7 @@ static unique_ptr<FunctionData> DuckDBSettingsBind(ClientContext &context, Table
 	return_types.emplace_back(LogicalType::VARIANT());
 
 	auto result = make_uniq<DuckDBSettingsBindData>();
-	for (auto name : {"debug", "deprecated"}) {
+	for (auto name : {"debug", "deprecated", "include_all"}) {
 		auto it = input.named_parameters.find(name);
 		if (it != input.named_parameters.end() && it->second.IsNull()) {
 			throw InvalidInputException("Cannot use NULL as argument for %s", name);
@@ -69,6 +75,14 @@ static unique_ptr<FunctionData> DuckDBSettingsBind(ClientContext &context, Table
 		result->deprecated = it->second.GetValue<bool>();
 	}
 
+	if (auto it = input.named_parameters.find("include_all"); it != input.named_parameters.end()) {
+		result->include_all = it->second.GetValue<bool>();
+	}
+	auto &params = input.named_parameters;
+	if (result->include_all && (params.find("debug") != params.end() || params.find("deprecated") != params.end())) {
+		throw InvalidInputException("Cannot combine include_all := true with the debug or deprecated filters");
+	}
+
 	return std::move(result);
 }
 
@@ -87,10 +101,7 @@ unique_ptr<GlobalTableFunctionState> DuckDBSettingsInit(ClientContext &context,
 	for (idx_t i = 0; i < options_count; i++) {
 		auto option = DBConfig::GetOptionByIndex(i);
 		D_ASSERT(option);
-		if (bind_data.debug != option->is_debug) {
-			continue;
-		}
-		if (bind_data.deprecated != option->is_deprecated) {
+		if (!bind_data.Includes(option->is_debug, option->is_deprecated)) {
 			continue;
 		}
 		DuckDBSettingValue value;
@@ -122,10 +133,7 @@ unique_ptr<GlobalTableFunctionState> DuckDBSettingsInit(ClientContext &context,
 		result->settings.push_back(std::move(value));
 	}
 	for (auto &ext_param : config.GetExtensionSettings()) {
-		if (bind_data.debug != ext_param.second.is_debug) {
-			continue;
-		}
-		if (bind_data.deprecated != ext_param.second.is_deprecated) {
+		if (!bind_data.Includes(ext_param.second.is_debug, ext_param.second.is_deprecated)) {
 			continue;
 		}
 		Value setting_val;
@@ -189,7 +197,9 @@ void DuckDBSettingsFunction(ClientContext &context, TableFunctionInput &data_p,
 void DuckDBSettingsFun::RegisterFunction(BuiltinFunctions &set) {
 	TableFunction settings_fun("duckdb_settings", {}, DuckDBSettingsFunction, DuckDBSettingsBind, DuckDBSettingsInit);
 	settings_fun.GetSignature().WithTypedKwargs("options", [](TypedKwargs &options) {
-		options.Add("debug", LogicalType::BOOLEAN).Add("deprecated", LogicalType::BOOLEAN);
+		options.Add("debug", LogicalType::BOOLEAN)
+		    .Add("deprecated", LogicalType::BOOLEAN)
+		    .Add("include_all", LogicalType::BOOLEAN);
 	});
 	set.AddFunction(settings_fun);
 }

src/duckdb/src/function/table/version/pragma_version.cpp

@@ -1,5 +1,5 @@
 #ifndef DUCKDB_PATCH_VERSION
-#define DUCKDB_PATCH_VERSION "0-alpha48780"
+#define DUCKDB_PATCH_VERSION "0-alpha48877"
 #endif
 #ifndef DUCKDB_MINOR_VERSION
 #define DUCKDB_MINOR_VERSION 0
@@ -8,10 +8,10 @@
 #define DUCKDB_MAJOR_VERSION 2
 #endif
 #ifndef DUCKDB_VERSION
-#define DUCKDB_VERSION "v2.0.0-alpha48780"
+#define DUCKDB_VERSION "v2.0.0-alpha48877"
 #endif
 #ifndef DUCKDB_SOURCE_ID
-#define DUCKDB_SOURCE_ID "f8f3f4dcd9"
+#define DUCKDB_SOURCE_ID "5bf940c131"
 #endif
 #include "duckdb/function/table/system_functions.hpp"
 #include "duckdb/main/database.hpp"