vendor: Update vendored sources to duckdb/duckdb@318e783

krlmlr · krlmlr · commit b70c03a7dd9d · 2025-03-07T16:13:03.000+01:00
CSV small code Improvements + initialising boolean variable. (duckdb/duckdb#16454) Adding windows code signing using azure (duckdb/duckdb#16444) Bump delta to working commit (duckdb/duckdb#16442)
diff --git a/src/duckdb/src/execution/operator/csv_scanner/buffer_manager/csv_buffer.cpp b/src/duckdb/src/execution/operator/csv_scanner/buffer_manager/csv_buffer.cpp
@@ -4,7 +4,7 @@
 namespace duckdb {
 
 CSVBuffer::CSVBuffer(ClientContext &context, idx_t buffer_size_p, CSVFileHandle &file_handle,
-                     idx_t &global_csv_current_position, idx_t file_number_p)
+                     const idx_t &global_csv_current_position, idx_t file_number_p)
     : context(context), requested_size(buffer_size_p), file_number(file_number_p), can_seek(file_handle.CanSeek()),
       is_pipe(file_handle.IsPipe()) {
 	AllocateBuffer(buffer_size_p);
@@ -34,7 +34,7 @@ CSVBuffer::CSVBuffer(CSVFileHandle &file_handle, ClientContext &context, idx_t b
 }
 
 shared_ptr<CSVBuffer> CSVBuffer::Next(CSVFileHandle &file_handle, idx_t buffer_size, idx_t file_number_p,
-                                      bool &has_seaked) {
+                                      bool &has_seaked) const {
 	if (has_seaked) {
 		// This means that at some point a reload was done, and we are currently on the incorrect position in our file
 		// handle
diff --git a/src/duckdb/src/execution/operator/csv_scanner/encode/csv_encoder.cpp b/src/duckdb/src/execution/operator/csv_scanner/encode/csv_encoder.cpp
@@ -36,7 +36,7 @@ void CSVEncoderBuffer::Reset() {
 	actual_encoded_buffer_size = 0;
 }
 
-CSVEncoder::CSVEncoder(DBConfig &config, const string &encoding_name_to_find, idx_t buffer_size) {
+CSVEncoder::CSVEncoder(const DBConfig &config, const string &encoding_name_to_find, idx_t buffer_size) {
 	encoding_name = StringUtil::Lower(encoding_name_to_find);
 	auto function = config.GetEncodeFunction(encoding_name_to_find);
 	if (!function) {
diff --git a/src/duckdb/src/execution/operator/csv_scanner/scanner/base_scanner.cpp b/src/duckdb/src/execution/operator/csv_scanner/scanner/base_scanner.cpp
@@ -11,9 +11,10 @@ ScannerResult::ScannerResult(CSVStates &states_p, CSVStateMachine &state_machine
 
 BaseScanner::BaseScanner(shared_ptr<CSVBufferManager> buffer_manager_p, shared_ptr<CSVStateMachine> state_machine_p,
                          shared_ptr<CSVErrorHandler> error_handler_p, bool sniffing_p,
-                         shared_ptr<CSVFileScan> csv_file_scan_p, CSVIterator iterator_p)
+                         shared_ptr<CSVFileScan> csv_file_scan_p, const CSVIterator &iterator_p)
     : csv_file_scan(std::move(csv_file_scan_p)), sniffing(sniffing_p), error_handler(std::move(error_handler_p)),
-      state_machine(std::move(state_machine_p)), buffer_manager(std::move(buffer_manager_p)), iterator(iterator_p) {
+      state_machine(std::move(state_machine_p)), states(), buffer_manager(std::move(buffer_manager_p)),
+      iterator(iterator_p) {
 	D_ASSERT(buffer_manager);
 	D_ASSERT(state_machine);
 	// Initialize current buffer handle
diff --git a/src/duckdb/src/execution/operator/csv_scanner/scanner/csv_schema.cpp b/src/duckdb/src/execution/operator/csv_scanner/scanner/csv_schema.cpp
@@ -76,8 +76,8 @@ void CSVSchema::MergeSchemas(CSVSchema &other, bool null_padding) {
 	}
 }
 
-CSVSchema::CSVSchema(vector<string> &names, vector<LogicalType> &types, const string &file_path, idx_t rows_read_p,
-                     const bool empty_p)
+CSVSchema::CSVSchema(const vector<string> &names, const vector<LogicalType> &types, const string &file_path,
+                     idx_t rows_read_p, const bool empty_p)
     : rows_read(rows_read_p), empty(empty_p) {
 	Initialize(names, types, file_path);
 }
diff --git a/src/duckdb/src/execution/operator/csv_scanner/scanner/scanner_boundary.cpp b/src/duckdb/src/execution/operator/csv_scanner/scanner/scanner_boundary.cpp
@@ -13,7 +13,7 @@ CSVBoundary::CSVBoundary(idx_t buffer_idx_p, idx_t buffer_pos_p, idx_t boundary_
 CSVBoundary::CSVBoundary() : buffer_idx(0), buffer_pos(0), boundary_idx(0), end_pos(NumericLimits<idx_t>::Maximum()) {
 }
 
-CSVIterator::CSVIterator() : is_set(false) {
+CSVIterator::CSVIterator() : buffer_size(0), is_set(false) {
 }
 
 void CSVBoundary::Print() const {
diff --git a/src/duckdb/src/execution/operator/csv_scanner/scanner/string_value_scanner.cpp b/src/duckdb/src/execution/operator/csv_scanner/scanner/string_value_scanner.cpp
@@ -970,7 +970,8 @@ StringValueScanner::StringValueScanner(idx_t scanner_idx_p, const shared_ptr<CSV
       result(states, *state_machine, cur_buffer_handle, BufferAllocator::Get(buffer_manager->context), result_size,
              iterator.pos.buffer_pos, *error_handler, iterator,
              buffer_manager->context.client_data->debug_set_max_line_length, csv_file_scan, lines_read, sniffing,
-             buffer_manager->GetFilePath(), scanner_idx_p) {
+             buffer_manager->GetFilePath(), scanner_idx_p),
+      start_pos(0) {
 	iterator.buffer_size = state_machine->options.buffer_size_option.GetValue();
 }
 
@@ -982,7 +983,8 @@ StringValueScanner::StringValueScanner(const shared_ptr<CSVBufferManager> &buffe
       result(states, *state_machine, cur_buffer_handle, Allocator::DefaultAllocator(), result_size,
              iterator.pos.buffer_pos, *error_handler, iterator,
              buffer_manager->context.client_data->debug_set_max_line_length, csv_file_scan, lines_read, sniffing,
-             buffer_manager->GetFilePath(), 0) {
+             buffer_manager->GetFilePath(), 0),
+      start_pos(0) {
 	iterator.buffer_size = state_machine->options.buffer_size_option.GetValue();
 }
 
diff --git a/src/duckdb/src/execution/operator/csv_scanner/sniffer/type_refinement.cpp b/src/duckdb/src/execution/operator/csv_scanner/sniffer/type_refinement.cpp
@@ -2,7 +2,7 @@
 #include "duckdb/execution/operator/csv_scanner/csv_casting.hpp"
 
 namespace duckdb {
-bool CSVSniffer::TryCastVector(Vector &parse_chunk_col, idx_t size, const LogicalType &sql_type) {
+bool CSVSniffer::TryCastVector(Vector &parse_chunk_col, idx_t size, const LogicalType &sql_type) const {
 	auto &sniffing_state_machine = best_candidate->GetStateMachine();
 	// try vector-cast from string to sql_type
 	Vector dummy_result(sql_type, size);
diff --git a/src/duckdb/src/function/table/version/pragma_version.cpp b/src/duckdb/src/function/table/version/pragma_version.cpp
@@ -1,5 +1,5 @@
 #ifndef DUCKDB_PATCH_VERSION
-#define DUCKDB_PATCH_VERSION "1-dev207"
+#define DUCKDB_PATCH_VERSION "1-dev214"
 #endif
 #ifndef DUCKDB_MINOR_VERSION
 #define DUCKDB_MINOR_VERSION 2
@@ -8,10 +8,10 @@
 #define DUCKDB_MAJOR_VERSION 1
 #endif
 #ifndef DUCKDB_VERSION
-#define DUCKDB_VERSION "v1.2.1-dev207"
+#define DUCKDB_VERSION "v1.2.1-dev214"
 #endif
 #ifndef DUCKDB_SOURCE_ID
-#define DUCKDB_SOURCE_ID "5e6f23a220"
+#define DUCKDB_SOURCE_ID "318e783c48"
 #endif
 #include "duckdb/function/table/system_functions.hpp"
 #include "duckdb/main/database.hpp"
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/base_scanner.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/base_scanner.hpp
@@ -116,7 +116,7 @@ class BaseScanner {
 public:
 	explicit BaseScanner(shared_ptr<CSVBufferManager> buffer_manager, shared_ptr<CSVStateMachine> state_machine,
 	                     shared_ptr<CSVErrorHandler> error_handler, bool sniffing = false,
-	                     shared_ptr<CSVFileScan> csv_file_scan = nullptr, CSVIterator iterator = {});
+	                     shared_ptr<CSVFileScan> csv_file_scan = nullptr, const CSVIterator &iterator = {});
 
 	virtual ~BaseScanner() = default;
 
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_buffer.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_buffer.hpp
@@ -45,14 +45,15 @@ class CSVBuffer {
 public:
 	//! Constructor for Initial Buffer
 	CSVBuffer(ClientContext &context, idx_t buffer_size_p, CSVFileHandle &file_handle,
-	          idx_t &global_csv_current_position, idx_t file_number);
+	          const idx_t &global_csv_current_position, idx_t file_number);
 
 	//! Constructor for `Next()` Buffers
 	CSVBuffer(CSVFileHandle &file_handle, ClientContext &context, idx_t buffer_size, idx_t global_csv_current_position,
 	          idx_t file_number_p, idx_t buffer_idx);
 
 	//! Creates a new buffer with the next part of the CSV File
-	shared_ptr<CSVBuffer> Next(CSVFileHandle &file_handle, idx_t buffer_size, idx_t file_number, bool &has_seaked);
+	shared_ptr<CSVBuffer> Next(CSVFileHandle &file_handle, idx_t buffer_size, idx_t file_number,
+	                           bool &has_seaked) const;
 
 	//! Gets the buffer actual size
 	idx_t GetBufferSize() const;
@@ -67,12 +68,12 @@ class CSVBuffer {
 	//! Wrapper for the Pin Function, if it can seek, it means that the buffer might have been destroyed, hence we must
 	//! Scan it from the disk file again.
 	shared_ptr<CSVBufferHandle> Pin(CSVFileHandle &file_handle, bool &has_seeked);
-	//! Wrapper for the unpin
+	//! Wrapper for unpin
 	void Unpin();
 	char *Ptr() {
 		return char_ptr_cast(handle.Ptr());
 	}
-	bool IsUnloaded() {
+	bool IsUnloaded() const {
 		return block->IsUnloaded();
 	}
 
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_option.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_option.hpp
@@ -30,7 +30,7 @@ struct CSVOption { // NOLINT: work-around bug in clang-tidy
 public:
 	CSVOption(T value_p) : value(value_p) { // NOLINT: allow implicit conversion from value
 	}
-	CSVOption(T value_p, bool set_by_user_p) : value(value_p), set_by_user(set_by_user_p) {
+	CSVOption(T value_p, bool set_by_user_p) : set_by_user(set_by_user_p), value(value_p) {
 	}
 
 	CSVOption() {};
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_schema.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_schema.hpp
@@ -14,7 +14,7 @@
 namespace duckdb {
 //! Basic CSV Column Info
 struct CSVColumnInfo {
-	CSVColumnInfo(string &name_p, LogicalType &type_p) : name(name_p), type(type_p) {
+	CSVColumnInfo(const string &name_p, const LogicalType &type_p) : name(name_p), type(type_p) {
 	}
 	string name;
 	LogicalType type;
@@ -25,7 +25,7 @@ struct CSVSchema {
 	explicit CSVSchema(const bool empty = false) : empty(empty) {
 	}
 
-	CSVSchema(vector<string> &names, vector<LogicalType> &types, const string &file_path, idx_t rows_read,
+	CSVSchema(const vector<string> &names, const vector<LogicalType> &types, const string &file_path, idx_t rows_read,
 	          const bool empty = false);
 
 	//! Initializes the schema based on names and types
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/encode/csv_encoder.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/encode/csv_encoder.hpp
@@ -46,7 +46,7 @@ struct CSVEncoderBuffer {
 class CSVEncoder {
 public:
 	//! Constructor, basically takes an encoding and the output buffer size
-	CSVEncoder(DBConfig &config, const string &encoding_name, idx_t buffer_size);
+	CSVEncoder(const DBConfig &config, const string &encoding_name, idx_t buffer_size);
 	//! Main encode function, it reads the file into an encoded buffer and converts it to the output buffer
 	idx_t Encode(FileHandle &file_handle_input, char *output_buffer, const idx_t decoded_buffer_size);
 	string encoding_name;
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/sniffer/csv_sniffer.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/sniffer/csv_sniffer.hpp
@@ -222,7 +222,7 @@ class CSVSniffer {
 	//! ------------------ Type Refinement ------------------ //
 	//! ------------------------------------------------------//
 	void RefineTypes();
-	bool TryCastVector(Vector &parse_chunk_col, idx_t size, const LogicalType &sql_type);
+	bool TryCastVector(Vector &parse_chunk_col, idx_t size, const LogicalType &sql_type) const;
 	vector<LogicalType> detected_types;
 	//! If when finding a SQLNULL type in type detection we default it to varchar
 	const bool default_null_to_varchar;
diff --git a/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/string_value_scanner.hpp b/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/string_value_scanner.hpp
@@ -121,8 +121,8 @@ class LineError {
 };
 
 struct ParseTypeInfo {
-	ParseTypeInfo() {};
-	ParseTypeInfo(const LogicalType &type, bool validate_utf_8_p) : validate_utf8(validate_utf_8_p) {
+	ParseTypeInfo() : validate_utf8(false), type_id(), internal_type(), scale(0), width(0) {};
+	ParseTypeInfo(const LogicalType &type, const bool validate_utf_8_p) : validate_utf8(validate_utf_8_p) {
 		type_id = type.id();
 		internal_type = type.InternalType();
 		if (type.id() == LogicalTypeId::DECIMAL) {

Original file line number	Diff line number	Diff line change
`@@ -36,7 +36,7 @@ void CSVEncoderBuffer::Reset() {`
`36`	`36`	`actual_encoded_buffer_size = 0;`
`37`	`37`	`}`
`38`	`38`
`39`		`-CSVEncoder::CSVEncoder(DBConfig &config, const string &encoding_name_to_find, idx_t buffer_size) {`
	`39`	`+CSVEncoder::CSVEncoder(const DBConfig &config, const string &encoding_name_to_find, idx_t buffer_size) {`
`40`	`40`	`encoding_name = StringUtil::Lower(encoding_name_to_find);`
`41`	`41`	`auto function = config.GetEncodeFunction(encoding_name_to_find);`
`42`	`42`	`if (!function) {`
Original file line number	Diff line number	Diff line change
`@@ -76,8 +76,8 @@ void CSVSchema::MergeSchemas(CSVSchema &other, bool null_padding) {`
`76`	`76`	`}`
`77`	`77`	`}`
`78`	`78`
`79`		`-CSVSchema::CSVSchema(vector<string> &names, vector<LogicalType> &types, const string &file_path, idx_t rows_read_p,`
`80`		`- const bool empty_p)`
	`79`	`+CSVSchema::CSVSchema(const vector<string> &names, const vector<LogicalType> &types, const string &file_path,`
	`80`	`+ idx_t rows_read_p, const bool empty_p)`
`81`	`81`	`: rows_read(rows_read_p), empty(empty_p) {`
`82`	`82`	`Initialize(names, types, file_path);`
`83`	`83`	`}`
Original file line number	Diff line number	Diff line change
`@@ -13,7 +13,7 @@ CSVBoundary::CSVBoundary(idx_t buffer_idx_p, idx_t buffer_pos_p, idx_t boundary_`
`13`	`13`	`CSVBoundary::CSVBoundary() : buffer_idx(0), buffer_pos(0), boundary_idx(0), end_pos(NumericLimits<idx_t>::Maximum()) {`
`14`	`14`	`}`
`15`	`15`
`16`		`-CSVIterator::CSVIterator() : is_set(false) {`
	`16`	`+CSVIterator::CSVIterator() : buffer_size(0), is_set(false) {`
`17`	`17`	`}`
`18`	`18`
`19`	`19`	`void CSVBoundary::Print() const {`