duckdb 0.7.2-dev3117.0 → 0.7.2-dev3154.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (24) hide show
  1. package/package.json +1 -1
  2. package/src/duckdb/extension/parquet/include/parquet_writer.hpp +7 -0
  3. package/src/duckdb/extension/parquet/parquet-extension.cpp +42 -0
  4. package/src/duckdb/extension/parquet/parquet_writer.cpp +23 -9
  5. package/src/duckdb/src/common/enums/physical_operator_type.cpp +2 -0
  6. package/src/duckdb/src/common/types/vector.cpp +4 -5
  7. package/src/duckdb/src/common/types/vector_buffer.cpp +1 -1
  8. package/src/duckdb/src/core_functions/function_list.cpp +1 -0
  9. package/src/duckdb/src/core_functions/scalar/map/map_concat.cpp +186 -0
  10. package/src/duckdb/src/execution/operator/persistent/physical_batch_copy_to_file.cpp +65 -21
  11. package/src/duckdb/src/execution/operator/persistent/physical_fixed_batch_copy.cpp +494 -0
  12. package/src/duckdb/src/execution/physical_plan/plan_copy_to_file.cpp +16 -6
  13. package/src/duckdb/src/execution/window_segment_tree.cpp +17 -13
  14. package/src/duckdb/src/function/table/version/pragma_version.cpp +2 -2
  15. package/src/duckdb/src/include/duckdb/common/enums/physical_operator_type.hpp +1 -0
  16. package/src/duckdb/src/include/duckdb/core_functions/scalar/map_functions.hpp +14 -1
  17. package/src/duckdb/src/include/duckdb/execution/operator/persistent/physical_batch_copy_to_file.hpp +13 -0
  18. package/src/duckdb/src/include/duckdb/execution/operator/persistent/physical_fixed_batch_copy.hpp +72 -0
  19. package/src/duckdb/src/include/duckdb/function/copy_function.hpp +3 -1
  20. package/src/duckdb/src/planner/operator/logical_delete.cpp +2 -0
  21. package/src/duckdb/src/planner/operator/logical_update.cpp +2 -0
  22. package/src/duckdb/third_party/utf8proc/include/utf8proc_wrapper.hpp +1 -0
  23. package/src/duckdb/ub_src_core_functions_scalar_map.cpp +2 -0
  24. package/src/duckdb/ub_src_execution_operator_persistent.cpp +2 -0
@@ -0,0 +1,72 @@
1
+ //===----------------------------------------------------------------------===//
2
+ // DuckDB
3
+ //
4
+ // duckdb/execution/operator/persistent/physical_fixed_batch_copy.hpp
5
+ //
6
+ //
7
+ //===----------------------------------------------------------------------===//
8
+
9
+ #pragma once
10
+
11
+ #include "duckdb/execution/physical_operator.hpp"
12
+ #include "duckdb/parser/parsed_data/copy_info.hpp"
13
+ #include "duckdb/function/copy_function.hpp"
14
+ #include "duckdb/common/file_system.hpp"
15
+ #include "duckdb/common/filename_pattern.hpp"
16
+
17
+ namespace duckdb {
18
+
19
+ class PhysicalFixedBatchCopy : public PhysicalOperator {
20
+ public:
21
+ static constexpr const PhysicalOperatorType TYPE = PhysicalOperatorType::FIXED_BATCH_COPY_TO_FILE;
22
+
23
+ public:
24
+ PhysicalFixedBatchCopy(vector<LogicalType> types, CopyFunction function, unique_ptr<FunctionData> bind_data,
25
+ idx_t estimated_cardinality);
26
+
27
+ CopyFunction function;
28
+ unique_ptr<FunctionData> bind_data;
29
+ string file_path;
30
+ bool use_tmp_file;
31
+
32
+ public:
33
+ // Source interface
34
+ SourceResultType GetData(ExecutionContext &context, DataChunk &chunk, OperatorSourceInput &input) const override;
35
+
36
+ bool IsSource() const override {
37
+ return true;
38
+ }
39
+
40
+ public:
41
+ // Sink interface
42
+ SinkResultType Sink(ExecutionContext &context, DataChunk &chunk, OperatorSinkInput &input) const override;
43
+ void Combine(ExecutionContext &context, GlobalSinkState &gstate, LocalSinkState &lstate) const override;
44
+ SinkFinalizeType Finalize(Pipeline &pipeline, Event &event, ClientContext &context,
45
+ GlobalSinkState &gstate) const override;
46
+ unique_ptr<LocalSinkState> GetLocalSinkState(ExecutionContext &context) const override;
47
+ unique_ptr<GlobalSinkState> GetGlobalSinkState(ClientContext &context) const override;
48
+ void NextBatch(ExecutionContext &context, GlobalSinkState &state, LocalSinkState &lstate_p) const override;
49
+
50
+ bool RequiresBatchIndex() const override {
51
+ return true;
52
+ }
53
+
54
+ bool IsSink() const override {
55
+ return true;
56
+ }
57
+
58
+ bool ParallelSink() const override {
59
+ return true;
60
+ }
61
+
62
+ public:
63
+ void AddRawBatchData(ClientContext &context, GlobalSinkState &gstate_p, idx_t batch_index,
64
+ unique_ptr<ColumnDataCollection> collection) const;
65
+ void RepartitionBatches(ClientContext &context, GlobalSinkState &gstate_p, idx_t min_index,
66
+ bool final = false) const;
67
+ void FlushBatchData(ClientContext &context, GlobalSinkState &gstate_p, idx_t min_index) const;
68
+ bool ExecuteTask(ClientContext &context, GlobalSinkState &gstate_p) const;
69
+ void ExecuteTasks(ClientContext &context, GlobalSinkState &gstate_p) const;
70
+ SinkFinalizeType FinalFlush(ClientContext &context, GlobalSinkState &gstate_p) const;
71
+ };
72
+ } // namespace duckdb
@@ -97,6 +97,7 @@ typedef unique_ptr<PreparedBatchData> (*copy_prepare_batch_t)(ClientContext &con
97
97
  unique_ptr<ColumnDataCollection> collection);
98
98
  typedef void (*copy_flush_batch_t)(ClientContext &context, FunctionData &bind_data, GlobalFunctionData &gstate,
99
99
  PreparedBatchData &batch);
100
+ typedef idx_t (*copy_desired_batch_size_t)(ClientContext &context, FunctionData &bind_data);
100
101
 
101
102
  class CopyFunction : public Function {
102
103
  public:
@@ -104,7 +105,7 @@ public:
104
105
  : Function(name), plan(nullptr), copy_to_bind(nullptr), copy_to_initialize_local(nullptr),
105
106
  copy_to_initialize_global(nullptr), copy_to_sink(nullptr), copy_to_combine(nullptr),
106
107
  copy_to_finalize(nullptr), execution_mode(nullptr), prepare_batch(nullptr), flush_batch(nullptr),
107
- serialize(nullptr), deserialize(nullptr), copy_from_bind(nullptr) {
108
+ desired_batch_size(nullptr), serialize(nullptr), deserialize(nullptr), copy_from_bind(nullptr) {
108
109
  }
109
110
 
110
111
  //! Plan rewrite copy function
@@ -120,6 +121,7 @@ public:
120
121
 
121
122
  copy_prepare_batch_t prepare_batch;
122
123
  copy_flush_batch_t flush_batch;
124
+ copy_desired_batch_size_t desired_batch_size;
123
125
 
124
126
  copy_to_serialize_t serialize;
125
127
  copy_to_deserialize_t deserialize;
@@ -15,6 +15,7 @@ void LogicalDelete::Serialize(FieldWriter &writer) const {
15
15
  table.Serialize(writer.GetSerializer());
16
16
  writer.WriteField(table_index);
17
17
  writer.WriteField(return_chunk);
18
+ writer.WriteSerializableList(this->expressions);
18
19
  }
19
20
 
20
21
  unique_ptr<LogicalOperator> LogicalDelete::Deserialize(LogicalDeserializationState &state, FieldReader &reader) {
@@ -26,6 +27,7 @@ unique_ptr<LogicalOperator> LogicalDelete::Deserialize(LogicalDeserializationSta
26
27
  auto table_index = reader.ReadRequired<idx_t>();
27
28
  auto result = make_uniq<LogicalDelete>(table_catalog_entry, table_index);
28
29
  result->return_chunk = reader.ReadRequired<bool>();
30
+ result->expressions = reader.ReadRequiredSerializableList<duckdb::Expression>(state.gstate);
29
31
  return std::move(result);
30
32
  }
31
33
 
@@ -17,6 +17,7 @@ void LogicalUpdate::Serialize(FieldWriter &writer) const {
17
17
  writer.WriteIndexList<PhysicalIndex>(columns);
18
18
  writer.WriteSerializableList(bound_defaults);
19
19
  writer.WriteField(update_is_del_and_insert);
20
+ writer.WriteSerializableList(this->expressions);
20
21
  }
21
22
 
22
23
  unique_ptr<LogicalOperator> LogicalUpdate::Deserialize(LogicalDeserializationState &state, FieldReader &reader) {
@@ -31,6 +32,7 @@ unique_ptr<LogicalOperator> LogicalUpdate::Deserialize(LogicalDeserializationSta
31
32
  result->columns = reader.ReadRequiredIndexList<PhysicalIndex>();
32
33
  result->bound_defaults = reader.ReadRequiredSerializableList<Expression>(state.gstate);
33
34
  result->update_is_del_and_insert = reader.ReadRequired<bool>();
35
+ result->expressions = reader.ReadRequiredSerializableList<duckdb::Expression>(state.gstate);
34
36
  return std::move(result);
35
37
  }
36
38
 
@@ -3,6 +3,7 @@
3
3
  #include <string>
4
4
  #include <cassert>
5
5
  #include <cstring>
6
+ #include <cstdint>
6
7
 
7
8
  namespace duckdb {
8
9
 
@@ -6,6 +6,8 @@
6
6
 
7
7
  #include "src/core_functions/scalar/map/map_entries.cpp"
8
8
 
9
+ #include "src/core_functions/scalar/map/map_concat.cpp"
10
+
9
11
  #include "src/core_functions/scalar/map/map_keys_values.cpp"
10
12
 
11
13
  #include "src/core_functions/scalar/map/cardinality.cpp"
@@ -8,6 +8,8 @@
8
8
 
9
9
  #include "src/execution/operator/persistent/csv_reader_options.cpp"
10
10
 
11
+ #include "src/execution/operator/persistent/physical_fixed_batch_copy.cpp"
12
+
11
13
  #include "src/execution/operator/persistent/physical_batch_copy_to_file.cpp"
12
14
 
13
15
  #include "src/execution/operator/persistent/physical_batch_insert.cpp"