duckdb 0.8.2-dev15.0 → 0.8.2-dev150.0
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.
- package/binding.gyp +7 -7
- package/lib/duckdb.d.ts +59 -0
- package/lib/duckdb.js +21 -0
- package/package.json +1 -1
- package/src/duckdb/extension/icu/icu_extension.cpp +4 -6
- package/src/duckdb/src/common/arrow/arrow_wrapper.cpp +0 -12
- package/src/duckdb/src/common/types/vector.cpp +15 -14
- package/src/duckdb/src/common/vector_operations/is_distinct_from.cpp +6 -4
- package/src/duckdb/src/execution/operator/persistent/physical_batch_insert.cpp +3 -2
- package/src/duckdb/src/execution/physical_operator.cpp +17 -14
- package/src/duckdb/src/function/table/system/test_all_types.cpp +38 -18
- package/src/duckdb/src/function/table/version/pragma_version.cpp +2 -2
- package/src/duckdb/src/include/duckdb/execution/physical_operator.hpp +3 -0
- package/src/duckdb/src/include/duckdb/function/table/system_functions.hpp +1 -1
- package/src/duckdb/src/parallel/pipeline_executor.cpp +7 -6
- package/src/duckdb/src/planner/binder/tableref/bind_pivot.cpp +0 -1
- package/src/duckdb/src/storage/storage_manager.cpp +7 -2
- package/src/duckdb/ub_extension_icu_third_party_icu_i18n.cpp +5 -5
- package/src/duckdb_node.hpp +1 -0
- package/src/statement.cpp +94 -1
- package/test/columns.test.ts +243 -0
package/binding.gyp
CHANGED
@@ -237,18 +237,18 @@
|
|
237
237
|
"src/duckdb/third_party/zstd/compress/zstd_lazy.cpp",
|
238
238
|
"src/duckdb/third_party/zstd/compress/zstd_ldm.cpp",
|
239
239
|
"src/duckdb/third_party/zstd/compress/zstd_opt.cpp",
|
240
|
-
"src/duckdb/extension/icu/./icu_extension.cpp",
|
241
|
-
"src/duckdb/extension/icu/./icu-list-range.cpp",
|
242
|
-
"src/duckdb/extension/icu/./icu-datefunc.cpp",
|
243
240
|
"src/duckdb/extension/icu/./icu-datepart.cpp",
|
244
|
-
"src/duckdb/extension/icu/./icu-
|
241
|
+
"src/duckdb/extension/icu/./icu-timezone.cpp",
|
242
|
+
"src/duckdb/extension/icu/./icu-makedate.cpp",
|
245
243
|
"src/duckdb/extension/icu/./icu-table-range.cpp",
|
244
|
+
"src/duckdb/extension/icu/./icu-datefunc.cpp",
|
245
|
+
"src/duckdb/extension/icu/./icu-list-range.cpp",
|
246
246
|
"src/duckdb/extension/icu/./icu-dateadd.cpp",
|
247
|
-
"src/duckdb/extension/icu/./icu-strptime.cpp",
|
248
247
|
"src/duckdb/extension/icu/./icu-datesub.cpp",
|
249
|
-
"src/duckdb/extension/icu/./icu-makedate.cpp",
|
250
|
-
"src/duckdb/extension/icu/./icu-timezone.cpp",
|
251
248
|
"src/duckdb/extension/icu/./icu-timebucket.cpp",
|
249
|
+
"src/duckdb/extension/icu/./icu-strptime.cpp",
|
250
|
+
"src/duckdb/extension/icu/./icu_extension.cpp",
|
251
|
+
"src/duckdb/extension/icu/./icu-datetrunc.cpp",
|
252
252
|
"src/duckdb/ub_extension_icu_third_party_icu_common.cpp",
|
253
253
|
"src/duckdb/ub_extension_icu_third_party_icu_i18n.cpp",
|
254
254
|
"src/duckdb/extension/icu/third_party/icu/stubdata/stubdata.cpp",
|
package/lib/duckdb.d.ts
CHANGED
@@ -171,6 +171,63 @@ export class Database {
|
|
171
171
|
): Promise<void>;
|
172
172
|
}
|
173
173
|
|
174
|
+
export type GenericTypeInfo = {
|
175
|
+
id: string,
|
176
|
+
sql_type: string,
|
177
|
+
alias?: string,
|
178
|
+
}
|
179
|
+
|
180
|
+
export type StructTypeInfo = {
|
181
|
+
id: "STRUCT",
|
182
|
+
alias?: string,
|
183
|
+
sql_type: string,
|
184
|
+
children: TypeInfoChildren,
|
185
|
+
}
|
186
|
+
|
187
|
+
export type ListTypeInfo = {
|
188
|
+
id: "LIST",
|
189
|
+
alias?: string,
|
190
|
+
sql_type: string,
|
191
|
+
child: TypeInfo,
|
192
|
+
}
|
193
|
+
|
194
|
+
export type MapTypeInfo = {
|
195
|
+
id: "MAP",
|
196
|
+
alias?: string,
|
197
|
+
sql_type: string,
|
198
|
+
key: TypeInfo,
|
199
|
+
value: TypeInfo,
|
200
|
+
}
|
201
|
+
|
202
|
+
export type UnionTypeInfo = {
|
203
|
+
id: "UNION",
|
204
|
+
alias?: string,
|
205
|
+
sql_type: string,
|
206
|
+
children: TypeInfoChildren,
|
207
|
+
}
|
208
|
+
|
209
|
+
export type DecimalTypeInfo = {
|
210
|
+
id: "DECIMAL",
|
211
|
+
alias?: string,
|
212
|
+
sql_type: string,
|
213
|
+
width: number,
|
214
|
+
scale: number,
|
215
|
+
}
|
216
|
+
|
217
|
+
export type EnumTypeInfo = {
|
218
|
+
id: "ENUM",
|
219
|
+
alias?: string,
|
220
|
+
sql_type: string,
|
221
|
+
name: string,
|
222
|
+
values: string[],
|
223
|
+
}
|
224
|
+
|
225
|
+
export type TypeInfoChildren = { name: string, type: TypeInfo }[];
|
226
|
+
|
227
|
+
export type TypeInfo = GenericTypeInfo | StructTypeInfo | ListTypeInfo | MapTypeInfo | UnionTypeInfo | DecimalTypeInfo | EnumTypeInfo;
|
228
|
+
|
229
|
+
export type ColumnInfo = { name: string, type: TypeInfo };
|
230
|
+
|
174
231
|
export class Statement {
|
175
232
|
sql: string;
|
176
233
|
|
@@ -185,6 +242,8 @@ export class Statement {
|
|
185
242
|
finalize(callback?: Callback<void>): void;
|
186
243
|
|
187
244
|
run(...args: [...any, Callback<void>] | any[]): Statement;
|
245
|
+
|
246
|
+
columns(): ColumnInfo[];
|
188
247
|
}
|
189
248
|
|
190
249
|
export const ERROR: number;
|
package/lib/duckdb.js
CHANGED
@@ -689,6 +689,27 @@ Statement.prototype.stream;
|
|
689
689
|
*/
|
690
690
|
Statement.prototype.sql;
|
691
691
|
|
692
|
+
/**
|
693
|
+
* @method
|
694
|
+
* @return {ColumnInfo[]} - Array of column names and types
|
695
|
+
*/
|
696
|
+
Statement.prototype.columns;
|
697
|
+
|
698
|
+
/**
|
699
|
+
* @typedef ColumnInfo
|
700
|
+
* @type {object}
|
701
|
+
* @property {string} name - Column name
|
702
|
+
* @property {TypeInfo} type - Column type
|
703
|
+
*/
|
704
|
+
|
705
|
+
/**
|
706
|
+
* @typedef TypeInfo
|
707
|
+
* @type {object}
|
708
|
+
* @property {string} id - Type ID
|
709
|
+
* @property {string} [alias] - SQL type alias
|
710
|
+
* @property {string} sql_type - SQL type name
|
711
|
+
*/
|
712
|
+
|
692
713
|
/**
|
693
714
|
* @typedef DuckDbError
|
694
715
|
* @type {object}
|
package/package.json
CHANGED
@@ -65,17 +65,15 @@ struct IcuBindData : public FunctionData {
|
|
65
65
|
|
66
66
|
static int32_t ICUGetSortKey(icu::Collator &collator, string_t input, duckdb::unique_ptr<char[]> &buffer,
|
67
67
|
int32_t &buffer_size) {
|
68
|
-
|
69
|
-
|
70
|
-
|
68
|
+
icu::UnicodeString unicode_string =
|
69
|
+
icu::UnicodeString::fromUTF8(icu::StringPiece(input.GetData(), input.GetSize()));
|
70
|
+
int32_t string_size = collator.getSortKey(unicode_string, (uint8_t *)buffer.get(), buffer_size);
|
71
71
|
if (string_size > buffer_size) {
|
72
72
|
// have to resize the buffer
|
73
73
|
buffer_size = string_size;
|
74
74
|
buffer = duckdb::unique_ptr<char[]>(new char[buffer_size]);
|
75
75
|
|
76
|
-
string_size =
|
77
|
-
collator.getSortKey(icu::UnicodeString::fromUTF8(icu::StringPiece(input.GetData(), input.GetSize())),
|
78
|
-
(uint8_t *)buffer.get(), buffer_size);
|
76
|
+
string_size = collator.getSortKey(unicode_string, (uint8_t *)buffer.get(), buffer_size);
|
79
77
|
}
|
80
78
|
return string_size;
|
81
79
|
}
|
@@ -14,12 +14,6 @@ namespace duckdb {
|
|
14
14
|
|
15
15
|
ArrowSchemaWrapper::~ArrowSchemaWrapper() {
|
16
16
|
if (arrow_schema.release) {
|
17
|
-
for (int64_t child_idx = 0; child_idx < arrow_schema.n_children; child_idx++) {
|
18
|
-
auto &child = *arrow_schema.children[child_idx];
|
19
|
-
if (child.release) {
|
20
|
-
child.release(&child);
|
21
|
-
}
|
22
|
-
}
|
23
17
|
arrow_schema.release(&arrow_schema);
|
24
18
|
arrow_schema.release = nullptr;
|
25
19
|
}
|
@@ -27,12 +21,6 @@ ArrowSchemaWrapper::~ArrowSchemaWrapper() {
|
|
27
21
|
|
28
22
|
ArrowArrayWrapper::~ArrowArrayWrapper() {
|
29
23
|
if (arrow_array.release) {
|
30
|
-
for (int64_t child_idx = 0; child_idx < arrow_array.n_children; child_idx++) {
|
31
|
-
auto &child = *arrow_array.children[child_idx];
|
32
|
-
if (child.release) {
|
33
|
-
child.release(&child);
|
34
|
-
}
|
35
|
-
}
|
36
24
|
arrow_array.release(&arrow_array);
|
37
25
|
arrow_array.release = nullptr;
|
38
26
|
}
|
@@ -855,38 +855,39 @@ void Vector::Flatten(const SelectionVector &sel, idx_t count) {
|
|
855
855
|
}
|
856
856
|
}
|
857
857
|
|
858
|
-
void Vector::ToUnifiedFormat(idx_t count, UnifiedVectorFormat &
|
858
|
+
void Vector::ToUnifiedFormat(idx_t count, UnifiedVectorFormat &format) {
|
859
859
|
switch (GetVectorType()) {
|
860
860
|
case VectorType::DICTIONARY_VECTOR: {
|
861
861
|
auto &sel = DictionaryVector::SelVector(*this);
|
862
|
+
format.owned_sel.Initialize(sel);
|
863
|
+
format.sel = &format.owned_sel;
|
864
|
+
|
862
865
|
auto &child = DictionaryVector::Child(*this);
|
863
866
|
if (child.GetVectorType() == VectorType::FLAT_VECTOR) {
|
864
|
-
data
|
865
|
-
|
866
|
-
data.validity = FlatVector::Validity(child);
|
867
|
+
format.data = FlatVector::GetData(child);
|
868
|
+
format.validity = FlatVector::Validity(child);
|
867
869
|
} else {
|
868
|
-
// dictionary with non-flat child: create a new reference to the child and
|
870
|
+
// dictionary with non-flat child: create a new reference to the child and flatten it
|
869
871
|
Vector child_vector(child);
|
870
872
|
child_vector.Flatten(sel, count);
|
871
873
|
auto new_aux = make_buffer<VectorChildBuffer>(std::move(child_vector));
|
872
874
|
|
873
|
-
data
|
874
|
-
|
875
|
-
data.validity = FlatVector::Validity(new_aux->data);
|
875
|
+
format.data = FlatVector::GetData(new_aux->data);
|
876
|
+
format.validity = FlatVector::Validity(new_aux->data);
|
876
877
|
this->auxiliary = std::move(new_aux);
|
877
878
|
}
|
878
879
|
break;
|
879
880
|
}
|
880
881
|
case VectorType::CONSTANT_VECTOR:
|
881
|
-
|
882
|
-
|
883
|
-
|
882
|
+
format.sel = ConstantVector::ZeroSelectionVector(count, format.owned_sel);
|
883
|
+
format.data = ConstantVector::GetData(*this);
|
884
|
+
format.validity = ConstantVector::Validity(*this);
|
884
885
|
break;
|
885
886
|
default:
|
886
887
|
Flatten(count);
|
887
|
-
|
888
|
-
|
889
|
-
|
888
|
+
format.sel = FlatVector::IncrementalSelectionVector();
|
889
|
+
format.data = FlatVector::GetData(*this);
|
890
|
+
format.validity = FlatVector::Validity(*this);
|
890
891
|
break;
|
891
892
|
}
|
892
893
|
}
|
@@ -564,10 +564,12 @@ static idx_t DistinctSelectList(Vector &left, Vector &right, idx_t count, const
|
|
564
564
|
SelectionVector lcursor(count);
|
565
565
|
SelectionVector rcursor(count);
|
566
566
|
|
567
|
-
ListVector::GetEntry(left)
|
568
|
-
ListVector::GetEntry(right)
|
569
|
-
|
570
|
-
|
567
|
+
Vector lentry_flattened(ListVector::GetEntry(left));
|
568
|
+
Vector rentry_flattened(ListVector::GetEntry(right));
|
569
|
+
lentry_flattened.Flatten(ListVector::GetListSize(left));
|
570
|
+
rentry_flattened.Flatten(ListVector::GetListSize(right));
|
571
|
+
Vector lchild(lentry_flattened, lcursor, count);
|
572
|
+
Vector rchild(rentry_flattened, rcursor, count);
|
571
573
|
|
572
574
|
// To perform the positional comparison, we use a vectorisation of the following algorithm:
|
573
575
|
// bool CompareLists(T *left, idx_t nleft, T *right, nright) {
|
@@ -327,10 +327,11 @@ SinkResultType PhysicalBatchInsert::Sink(ExecutionContext &context, DataChunk &c
|
|
327
327
|
// no collection yet: create a new one
|
328
328
|
lstate.CreateNewCollection(table, insert_types);
|
329
329
|
lstate.writer = &table.GetStorage().CreateOptimisticWriter(context.client);
|
330
|
-
}
|
330
|
+
}
|
331
|
+
|
332
|
+
if (lstate.current_index != batch_index) {
|
331
333
|
throw InternalException("Current batch differs from batch - but NextBatch was not called!?");
|
332
334
|
}
|
333
|
-
lstate.current_index = batch_index;
|
334
335
|
|
335
336
|
table.GetStorage().VerifyAppendConstraints(table, context.client, lstate.insert_chunk);
|
336
337
|
|
@@ -124,6 +124,22 @@ idx_t PhysicalOperator::GetMaxThreadMemory(ClientContext &context) {
|
|
124
124
|
return (max_memory / num_threads) / 4;
|
125
125
|
}
|
126
126
|
|
127
|
+
bool PhysicalOperator::OperatorCachingAllowed(ExecutionContext &context) {
|
128
|
+
if (!context.client.config.enable_caching_operators) {
|
129
|
+
return false;
|
130
|
+
} else if (!context.pipeline) {
|
131
|
+
return false;
|
132
|
+
} else if (!context.pipeline->GetSink()) {
|
133
|
+
return false;
|
134
|
+
} else if (context.pipeline->GetSink()->RequiresBatchIndex()) {
|
135
|
+
return false;
|
136
|
+
} else if (context.pipeline->IsOrderDependent()) {
|
137
|
+
return false;
|
138
|
+
}
|
139
|
+
|
140
|
+
return true;
|
141
|
+
}
|
142
|
+
|
127
143
|
//===--------------------------------------------------------------------===//
|
128
144
|
// Pipeline Construction
|
129
145
|
//===--------------------------------------------------------------------===//
|
@@ -239,20 +255,7 @@ OperatorResultType CachingPhysicalOperator::Execute(ExecutionContext &context, D
|
|
239
255
|
#if STANDARD_VECTOR_SIZE >= 128
|
240
256
|
if (!state.initialized) {
|
241
257
|
state.initialized = true;
|
242
|
-
state.can_cache_chunk =
|
243
|
-
|
244
|
-
if (!context.client.config.enable_caching_operators) {
|
245
|
-
state.can_cache_chunk = false;
|
246
|
-
} else if (!context.pipeline || !caching_supported) {
|
247
|
-
state.can_cache_chunk = false;
|
248
|
-
} else if (!context.pipeline->GetSink()) {
|
249
|
-
// Disabling for pipelines without Sink, i.e. when pulling
|
250
|
-
state.can_cache_chunk = false;
|
251
|
-
} else if (context.pipeline->GetSink()->RequiresBatchIndex()) {
|
252
|
-
state.can_cache_chunk = false;
|
253
|
-
} else if (context.pipeline->IsOrderDependent()) {
|
254
|
-
state.can_cache_chunk = false;
|
255
|
-
}
|
258
|
+
state.can_cache_chunk = caching_supported && PhysicalOperator::OperatorCachingAllowed(context);
|
256
259
|
}
|
257
260
|
if (!state.can_cache_chunk) {
|
258
261
|
return child_result;
|
@@ -17,7 +17,7 @@ struct TestAllTypesData : public GlobalTableFunctionState {
|
|
17
17
|
idx_t offset;
|
18
18
|
};
|
19
19
|
|
20
|
-
vector<TestType> TestAllTypesFun::GetTestTypes() {
|
20
|
+
vector<TestType> TestAllTypesFun::GetTestTypes(bool use_large_enum) {
|
21
21
|
vector<TestType> result;
|
22
22
|
// scalar types/numerics
|
23
23
|
result.emplace_back(LogicalType::BOOLEAN, "bool");
|
@@ -79,13 +79,21 @@ vector<TestType> TestAllTypesFun::GetTestTypes() {
|
|
79
79
|
}
|
80
80
|
result.emplace_back(LogicalType::ENUM("medium_enum", medium_enum, 300), "medium_enum");
|
81
81
|
|
82
|
-
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
82
|
+
if (use_large_enum) {
|
83
|
+
// this is a big one... not sure if we should push this one here, but it's required for completeness
|
84
|
+
Vector large_enum(LogicalType::VARCHAR, 70000);
|
85
|
+
auto large_enum_ptr = FlatVector::GetData<string_t>(large_enum);
|
86
|
+
for (idx_t i = 0; i < 70000; i++) {
|
87
|
+
large_enum_ptr[i] = StringVector::AddStringOrBlob(large_enum, string("enum_") + to_string(i));
|
88
|
+
}
|
89
|
+
result.emplace_back(LogicalType::ENUM("large_enum", large_enum, 70000), "large_enum");
|
90
|
+
} else {
|
91
|
+
Vector large_enum(LogicalType::VARCHAR, 2);
|
92
|
+
auto large_enum_ptr = FlatVector::GetData<string_t>(large_enum);
|
93
|
+
large_enum_ptr[0] = StringVector::AddStringOrBlob(large_enum, string("enum_") + to_string(0));
|
94
|
+
large_enum_ptr[1] = StringVector::AddStringOrBlob(large_enum, string("enum_") + to_string(69999));
|
95
|
+
result.emplace_back(LogicalType::ENUM("large_enum", large_enum, 2), "large_enum");
|
87
96
|
}
|
88
|
-
result.emplace_back(LogicalType::ENUM("large_enum", large_enum, 70000), "large_enum");
|
89
97
|
|
90
98
|
// arrays
|
91
99
|
auto int_list_type = LogicalType::LIST(LogicalType::INTEGER);
|
@@ -199,26 +207,36 @@ vector<TestType> TestAllTypesFun::GetTestTypes() {
|
|
199
207
|
return result;
|
200
208
|
}
|
201
209
|
|
210
|
+
struct TestAllTypesBindData : public TableFunctionData {
|
211
|
+
vector<TestType> test_types;
|
212
|
+
};
|
213
|
+
|
202
214
|
static unique_ptr<FunctionData> TestAllTypesBind(ClientContext &context, TableFunctionBindInput &input,
|
203
215
|
vector<LogicalType> &return_types, vector<string> &names) {
|
204
|
-
auto
|
205
|
-
|
206
|
-
|
207
|
-
|
216
|
+
auto result = make_uniq<TestAllTypesBindData>();
|
217
|
+
bool use_large_enum = false;
|
218
|
+
auto entry = input.named_parameters.find("use_large_enum");
|
219
|
+
if (entry != input.named_parameters.end()) {
|
220
|
+
use_large_enum = BooleanValue::Get(entry->second);
|
208
221
|
}
|
209
|
-
|
222
|
+
result->test_types = TestAllTypesFun::GetTestTypes(use_large_enum);
|
223
|
+
for (auto &test_type : result->test_types) {
|
224
|
+
return_types.push_back(test_type.type);
|
225
|
+
names.push_back(test_type.name);
|
226
|
+
}
|
227
|
+
return std::move(result);
|
210
228
|
}
|
211
229
|
|
212
230
|
unique_ptr<GlobalTableFunctionState> TestAllTypesInit(ClientContext &context, TableFunctionInitInput &input) {
|
231
|
+
auto &bind_data = input.bind_data->Cast<TestAllTypesBindData>();
|
213
232
|
auto result = make_uniq<TestAllTypesData>();
|
214
|
-
auto test_types = TestAllTypesFun::GetTestTypes();
|
215
233
|
// 3 rows: min, max and NULL
|
216
234
|
result->entries.resize(3);
|
217
235
|
// initialize the values
|
218
|
-
for (auto &test_type : test_types) {
|
219
|
-
result->entries[0].push_back(
|
220
|
-
result->entries[1].push_back(
|
221
|
-
result->entries[2].emplace_back(
|
236
|
+
for (auto &test_type : bind_data.test_types) {
|
237
|
+
result->entries[0].push_back(test_type.min_value);
|
238
|
+
result->entries[1].push_back(test_type.max_value);
|
239
|
+
result->entries[2].emplace_back(test_type.type);
|
222
240
|
}
|
223
241
|
return std::move(result);
|
224
242
|
}
|
@@ -243,7 +261,9 @@ void TestAllTypesFunction(ClientContext &context, TableFunctionInput &data_p, Da
|
|
243
261
|
}
|
244
262
|
|
245
263
|
void TestAllTypesFun::RegisterFunction(BuiltinFunctions &set) {
|
246
|
-
|
264
|
+
TableFunction test_all_types("test_all_types", {}, TestAllTypesFunction, TestAllTypesBind, TestAllTypesInit);
|
265
|
+
test_all_types.named_parameters["use_large_enum"] = LogicalType::BOOLEAN;
|
266
|
+
set.AddFunction(test_all_types);
|
247
267
|
}
|
248
268
|
|
249
269
|
} // namespace duckdb
|
@@ -1,8 +1,8 @@
|
|
1
1
|
#ifndef DUCKDB_VERSION
|
2
|
-
#define DUCKDB_VERSION "0.8.2-
|
2
|
+
#define DUCKDB_VERSION "0.8.2-dev150"
|
3
3
|
#endif
|
4
4
|
#ifndef DUCKDB_SOURCE_ID
|
5
|
-
#define DUCKDB_SOURCE_ID "
|
5
|
+
#define DUCKDB_SOURCE_ID "6b06ebe561"
|
6
6
|
#endif
|
7
7
|
#include "duckdb/function/table/system_functions.hpp"
|
8
8
|
#include "duckdb/main/database.hpp"
|
@@ -156,6 +156,9 @@ public:
|
|
156
156
|
//! The maximum amount of memory the operator should use per thread.
|
157
157
|
static idx_t GetMaxThreadMemory(ClientContext &context);
|
158
158
|
|
159
|
+
//! Whether operator caching is allowed in the current execution context
|
160
|
+
static bool OperatorCachingAllowed(ExecutionContext &context);
|
161
|
+
|
159
162
|
virtual bool IsSink() const {
|
160
163
|
return false;
|
161
164
|
}
|
@@ -118,7 +118,7 @@ struct TestType {
|
|
118
118
|
|
119
119
|
struct TestAllTypesFun {
|
120
120
|
static void RegisterFunction(BuiltinFunctions &set);
|
121
|
-
static vector<TestType> GetTestTypes();
|
121
|
+
static vector<TestType> GetTestTypes(bool large_enum = false);
|
122
122
|
};
|
123
123
|
|
124
124
|
struct TestVectorTypesFun {
|
@@ -17,11 +17,10 @@ PipelineExecutor::PipelineExecutor(ClientContext &context_p, Pipeline &pipeline_
|
|
17
17
|
requires_batch_index = pipeline.sink->RequiresBatchIndex() && pipeline.source->SupportsBatchIndex();
|
18
18
|
if (requires_batch_index) {
|
19
19
|
auto &partition_info = local_sink_state->partition_info;
|
20
|
-
|
21
|
-
|
22
|
-
|
23
|
-
|
24
|
-
}
|
20
|
+
D_ASSERT(!partition_info.batch_index.IsValid());
|
21
|
+
// batch index is not set yet - initialize before fetching anything
|
22
|
+
partition_info.batch_index = pipeline.RegisterNewBatchIndex();
|
23
|
+
partition_info.min_batch_index = partition_info.batch_index;
|
25
24
|
}
|
26
25
|
}
|
27
26
|
local_source_state = pipeline.source->GetLocalSourceState(context, *pipeline.source_state);
|
@@ -79,6 +78,7 @@ bool PipelineExecutor::TryFlushCachingOperators() {
|
|
79
78
|
OperatorResultType push_result;
|
80
79
|
|
81
80
|
if (in_process_operators.empty()) {
|
81
|
+
curr_chunk.Reset();
|
82
82
|
StartOperator(current_operator);
|
83
83
|
finalize_result = current_operator.FinalExecute(context, curr_chunk, *current_operator.op_state,
|
84
84
|
*intermediate_states[flushing_idx]);
|
@@ -477,7 +477,8 @@ SourceResultType PipelineExecutor::FetchFromSource(DataChunk &result) {
|
|
477
477
|
} else {
|
478
478
|
next_batch_index =
|
479
479
|
pipeline.source->GetBatchIndex(context, result, *pipeline.source_state, *local_source_state);
|
480
|
-
|
480
|
+
// we start with the base_batch_index as a valid starting value. Make sure that next batch is called below
|
481
|
+
next_batch_index += pipeline.base_batch_index + 1;
|
481
482
|
}
|
482
483
|
auto &partition_info = local_sink_state->partition_info;
|
483
484
|
if (next_batch_index != partition_info.batch_index.GetIndex()) {
|
@@ -315,7 +315,6 @@ unique_ptr<SelectNode> Binder::BindPivot(PivotRef &ref, vector<unique_ptr<Parsed
|
|
315
315
|
}
|
316
316
|
ExtractPivotExpressions(*aggr, handled_columns);
|
317
317
|
}
|
318
|
-
value_set_t pivots;
|
319
318
|
|
320
319
|
// first add all pivots to the set of handled columns, and check for duplicates
|
321
320
|
idx_t total_pivots = 1;
|
@@ -94,8 +94,13 @@ void SingleFileStorageManager::LoadDatabase() {
|
|
94
94
|
table_io_manager = make_uniq<SingleFileTableIOManager>(*block_manager);
|
95
95
|
return;
|
96
96
|
}
|
97
|
-
|
98
|
-
|
97
|
+
std::size_t question_mark_pos = path.find('?');
|
98
|
+
auto wal_path = path;
|
99
|
+
if (question_mark_pos != std::string::npos) {
|
100
|
+
wal_path.insert(question_mark_pos, ".wal");
|
101
|
+
} else {
|
102
|
+
wal_path += ".wal";
|
103
|
+
}
|
99
104
|
auto &fs = FileSystem::Get(db);
|
100
105
|
auto &config = DBConfig::Get(db);
|
101
106
|
bool truncate_wal = false;
|
@@ -348,17 +348,17 @@
|
|
348
348
|
|
349
349
|
#include "extension/icu/third_party/icu/i18n/wintzimpl.cpp"
|
350
350
|
|
351
|
-
#include "extension/icu/third_party/icu/i18n/double-conversion-fast-dtoa.cpp"
|
352
|
-
|
353
351
|
#include "extension/icu/third_party/icu/i18n/double-conversion-double-to-string.cpp"
|
354
352
|
|
355
|
-
#include "extension/icu/third_party/icu/i18n/double-conversion-
|
353
|
+
#include "extension/icu/third_party/icu/i18n/double-conversion-string-to-double.cpp"
|
356
354
|
|
357
355
|
#include "extension/icu/third_party/icu/i18n/double-conversion-cached-powers.cpp"
|
358
356
|
|
359
|
-
#include "extension/icu/third_party/icu/i18n/double-conversion-
|
357
|
+
#include "extension/icu/third_party/icu/i18n/double-conversion-fast-dtoa.cpp"
|
358
|
+
|
359
|
+
#include "extension/icu/third_party/icu/i18n/double-conversion-bignum.cpp"
|
360
360
|
|
361
361
|
#include "extension/icu/third_party/icu/i18n/double-conversion-bignum-dtoa.cpp"
|
362
362
|
|
363
|
-
#include "extension/icu/third_party/icu/i18n/double-conversion-
|
363
|
+
#include "extension/icu/third_party/icu/i18n/double-conversion-strtod.cpp"
|
364
364
|
|
package/src/duckdb_node.hpp
CHANGED
@@ -165,6 +165,7 @@ public:
|
|
165
165
|
Napi::Value Run(const Napi::CallbackInfo &info);
|
166
166
|
Napi::Value Finish(const Napi::CallbackInfo &info);
|
167
167
|
Napi::Value Stream(const Napi::CallbackInfo &info);
|
168
|
+
Napi::Value Columns(const Napi::CallbackInfo &info);
|
168
169
|
|
169
170
|
public:
|
170
171
|
static Napi::FunctionReference constructor;
|
package/src/statement.cpp
CHANGED
@@ -10,6 +10,7 @@
|
|
10
10
|
|
11
11
|
#include "duckdb/common/helper.hpp"
|
12
12
|
#include "duckdb/common/vector.hpp"
|
13
|
+
#include "duckdb/common/types.hpp"
|
13
14
|
|
14
15
|
using duckdb::unique_ptr;
|
15
16
|
using duckdb::vector;
|
@@ -25,7 +26,8 @@ Napi::Object Statement::Init(Napi::Env env, Napi::Object exports) {
|
|
25
26
|
DefineClass(env, "Statement",
|
26
27
|
{InstanceMethod("run", &Statement::Run), InstanceMethod("all", &Statement::All),
|
27
28
|
InstanceMethod("arrowIPCAll", &Statement::ArrowIPCAll), InstanceMethod("each", &Statement::Each),
|
28
|
-
InstanceMethod("finalize", &Statement::Finish), InstanceMethod("stream", &Statement::Stream)
|
29
|
+
InstanceMethod("finalize", &Statement::Finish), InstanceMethod("stream", &Statement::Stream),
|
30
|
+
InstanceMethod("columns", &Statement::Columns)});
|
29
31
|
|
30
32
|
constructor = Napi::Persistent(t);
|
31
33
|
constructor.SuppressDestruct();
|
@@ -503,6 +505,97 @@ Napi::Value Statement::Stream(const Napi::CallbackInfo &info) {
|
|
503
505
|
return deferred.Promise();
|
504
506
|
}
|
505
507
|
|
508
|
+
static Napi::Value TypeToObject(Napi::Env &env, const duckdb::LogicalType &type) {
|
509
|
+
auto obj = Napi::Object::New(env);
|
510
|
+
|
511
|
+
auto id = duckdb::LogicalTypeIdToString(type.id());
|
512
|
+
obj.Set("id", id);
|
513
|
+
obj.Set("sql_type", type.ToString());
|
514
|
+
|
515
|
+
if (type.HasAlias()) {
|
516
|
+
obj.Set("alias", type.GetAlias());
|
517
|
+
}
|
518
|
+
|
519
|
+
switch (type.id()) {
|
520
|
+
case duckdb::LogicalTypeId::STRUCT: {
|
521
|
+
auto &child_types = duckdb::StructType::GetChildTypes(type);
|
522
|
+
auto arr = Napi::Array::New(env, child_types.size());
|
523
|
+
for (size_t i = 0; i < child_types.size(); i++) {
|
524
|
+
auto child_name = child_types[i].first;
|
525
|
+
auto child_type = child_types[i].second;
|
526
|
+
auto child_obj = Napi::Object::New(env);
|
527
|
+
child_obj.Set("name", child_name);
|
528
|
+
child_obj.Set("type", TypeToObject(env, child_type));
|
529
|
+
arr.Set(i, child_obj);
|
530
|
+
}
|
531
|
+
obj.Set("children", arr);
|
532
|
+
} break;
|
533
|
+
case duckdb::LogicalTypeId::LIST: {
|
534
|
+
auto &child_type = duckdb::ListType::GetChildType(type);
|
535
|
+
obj.Set("child", TypeToObject(env, child_type));
|
536
|
+
} break;
|
537
|
+
case duckdb::LogicalTypeId::MAP: {
|
538
|
+
auto &key_type = duckdb::MapType::KeyType(type);
|
539
|
+
auto &value_type = duckdb::MapType::ValueType(type);
|
540
|
+
obj.Set("key", TypeToObject(env, key_type));
|
541
|
+
obj.Set("value", TypeToObject(env, value_type));
|
542
|
+
} break;
|
543
|
+
case duckdb::LogicalTypeId::ENUM: {
|
544
|
+
auto name = duckdb::EnumType::GetTypeName(type);
|
545
|
+
auto &values_vec = duckdb::EnumType::GetValuesInsertOrder(type);
|
546
|
+
auto enum_size = duckdb::EnumType::GetSize(type);
|
547
|
+
auto arr = Napi::Array::New(env, enum_size);
|
548
|
+
for (size_t i = 0; i < enum_size; i++) {
|
549
|
+
auto child_name = values_vec.GetValue(i).GetValue<duckdb::string>();
|
550
|
+
arr.Set(i, child_name);
|
551
|
+
}
|
552
|
+
obj.Set("name", name);
|
553
|
+
obj.Set("values", arr);
|
554
|
+
} break;
|
555
|
+
case duckdb::LogicalTypeId::UNION: {
|
556
|
+
auto child_count = duckdb::UnionType::GetMemberCount(type);
|
557
|
+
auto arr = Napi::Array::New(env, child_count);
|
558
|
+
for (size_t i = 0; i < child_count; i++) {
|
559
|
+
auto &child_name = duckdb::UnionType::GetMemberName(type, i);
|
560
|
+
auto &child_type = duckdb::UnionType::GetMemberType(type, i);
|
561
|
+
auto child_obj = Napi::Object::New(env);
|
562
|
+
child_obj.Set("name", child_name);
|
563
|
+
child_obj.Set("type", TypeToObject(env, child_type));
|
564
|
+
arr.Set(i, child_obj);
|
565
|
+
}
|
566
|
+
obj.Set("children", arr);
|
567
|
+
} break;
|
568
|
+
case duckdb::LogicalTypeId::DECIMAL: {
|
569
|
+
auto width = duckdb::DecimalType::GetWidth(type);
|
570
|
+
auto scale = duckdb::DecimalType::GetScale(type);
|
571
|
+
obj.Set("width", width);
|
572
|
+
obj.Set("scale", scale);
|
573
|
+
} break;
|
574
|
+
default:
|
575
|
+
break;
|
576
|
+
}
|
577
|
+
return obj;
|
578
|
+
}
|
579
|
+
|
580
|
+
Napi::Value Statement::Columns(const Napi::CallbackInfo &info) {
|
581
|
+
Napi::Env env = info.Env();
|
582
|
+
|
583
|
+
if (!statement) {
|
584
|
+
return env.Null();
|
585
|
+
}
|
586
|
+
|
587
|
+
auto &names = statement->GetNames();
|
588
|
+
auto &types = statement->GetTypes();
|
589
|
+
auto arr = Napi::Array::New(env, names.size());
|
590
|
+
for (size_t i = 0; i < names.size(); i++) {
|
591
|
+
auto obj = Napi::Object::New(env);
|
592
|
+
obj.Set("name", Napi::String::New(env, names[i]));
|
593
|
+
obj.Set("type", TypeToObject(env, types[i]));
|
594
|
+
arr.Set(i, obj);
|
595
|
+
}
|
596
|
+
return arr;
|
597
|
+
}
|
598
|
+
|
506
599
|
struct FinishTask : public Task {
|
507
600
|
FinishTask(Statement &statement, Napi::Function callback) : Task(statement, callback) {
|
508
601
|
}
|
@@ -0,0 +1,243 @@
|
|
1
|
+
import * as duckdb from '..';
|
2
|
+
import * as assert from 'assert';
|
3
|
+
|
4
|
+
describe('Column Types', function() {
|
5
|
+
let db: duckdb.Database;
|
6
|
+
before(function(done) { db = new duckdb.Database(':memory:', done); });
|
7
|
+
|
8
|
+
it('should prepare a statement and return the columns and their types', function(done) {
|
9
|
+
// we dont include the large_enum and small_enum since they are huge and test the same code path as the small_enum
|
10
|
+
var stmt = db.prepare("SELECT * EXCLUDE(medium_enum, large_enum) FROM test_all_types()", function(err: null | Error) {
|
11
|
+
if (err) throw err;
|
12
|
+
|
13
|
+
let cols = stmt.columns();
|
14
|
+
|
15
|
+
assert.equal(cols.length, 41);
|
16
|
+
|
17
|
+
var expected = [
|
18
|
+
{ name: 'bool', type: { id: 'BOOLEAN', sql_type: 'BOOLEAN' } },
|
19
|
+
{ name: 'tinyint', type: { id: 'TINYINT', sql_type: 'TINYINT' } },
|
20
|
+
{ name: 'smallint', type: { id: 'SMALLINT', sql_type: 'SMALLINT' } },
|
21
|
+
{ name: 'int', type: { id: 'INTEGER', sql_type: 'INTEGER' } },
|
22
|
+
{ name: 'bigint', type: { id: 'BIGINT', sql_type: 'BIGINT' } },
|
23
|
+
{ name: 'hugeint', type: { id: 'HUGEINT', sql_type: 'HUGEINT' } },
|
24
|
+
{ name: 'utinyint', type: { id: 'UTINYINT', sql_type: 'UTINYINT' } },
|
25
|
+
{ name: 'usmallint', type: { id: 'USMALLINT', sql_type: 'USMALLINT' } },
|
26
|
+
{ name: 'uint', type: { id: 'UINTEGER', sql_type: 'UINTEGER' } },
|
27
|
+
{ name: 'ubigint', type: { id: 'UBIGINT', sql_type: 'UBIGINT' } },
|
28
|
+
{ name: 'date', type: { id: 'DATE', sql_type: 'DATE' } },
|
29
|
+
{ name: 'time', type: { id: 'TIME', sql_type: 'TIME' } },
|
30
|
+
{ name: 'timestamp', type: { id: 'TIMESTAMP', sql_type: 'TIMESTAMP' } },
|
31
|
+
{ name: 'timestamp_s', type: { id: 'TIMESTAMP_S', sql_type: 'TIMESTAMP_S' } },
|
32
|
+
{ name: 'timestamp_ms', type: { id: 'TIMESTAMP_MS', sql_type: 'TIMESTAMP_MS' } },
|
33
|
+
{ name: 'timestamp_ns', type: { id: 'TIMESTAMP_NS', sql_type: 'TIMESTAMP_NS' } },
|
34
|
+
{ name: 'time_tz', type: { id: 'TIME WITH TIME ZONE', sql_type: 'TIME WITH TIME ZONE' } },
|
35
|
+
{ name: 'timestamp_tz', type: { id: 'TIMESTAMP WITH TIME ZONE', sql_type: 'TIMESTAMP WITH TIME ZONE' } },
|
36
|
+
{ name: 'float', type: { id: 'FLOAT', sql_type: 'FLOAT' } },
|
37
|
+
{ name: 'double', type: { id: 'DOUBLE', sql_type: 'DOUBLE' } },
|
38
|
+
{ name: 'dec_4_1', type: { id: 'DECIMAL', sql_type: 'DECIMAL(4,1)', width: 4, scale: 1 } },
|
39
|
+
{ name: 'dec_9_4', type: { id: 'DECIMAL', sql_type: 'DECIMAL(9,4)', width: 9, scale: 4 } },
|
40
|
+
{ name: 'dec_18_6', type: { id: 'DECIMAL', sql_type: 'DECIMAL(18,6)', width: 18, scale: 6 } },
|
41
|
+
{ name: 'dec38_10', type: { id: 'DECIMAL', sql_type: 'DECIMAL(38,10)', width: 38, scale: 10 } },
|
42
|
+
{ name: 'uuid', type: { id: 'UUID', sql_type: 'UUID' } },
|
43
|
+
{ name: 'interval', type: { id: 'INTERVAL', sql_type: 'INTERVAL' } },
|
44
|
+
{ name: 'varchar', type: { id: 'VARCHAR', sql_type: 'VARCHAR' } },
|
45
|
+
{ name: 'blob', type: { id: 'BLOB', sql_type: 'BLOB' } },
|
46
|
+
{ name: 'bit', type: { id: 'BIT', sql_type: 'BIT' } },
|
47
|
+
{
|
48
|
+
name: 'small_enum',
|
49
|
+
type: {
|
50
|
+
id: 'ENUM',
|
51
|
+
sql_type: 'small_enum',
|
52
|
+
name: 'small_enum',
|
53
|
+
values: [
|
54
|
+
"DUCK_DUCK_ENUM",
|
55
|
+
"GOOSE"
|
56
|
+
]
|
57
|
+
}
|
58
|
+
},
|
59
|
+
{
|
60
|
+
name: 'int_array',
|
61
|
+
type: {
|
62
|
+
id: 'LIST',
|
63
|
+
sql_type: 'INTEGER[]',
|
64
|
+
child: {
|
65
|
+
id: 'INTEGER',
|
66
|
+
sql_type: 'INTEGER'
|
67
|
+
}
|
68
|
+
}
|
69
|
+
},
|
70
|
+
{
|
71
|
+
name: 'double_array',
|
72
|
+
type: {
|
73
|
+
id: 'LIST',
|
74
|
+
sql_type: 'DOUBLE[]',
|
75
|
+
child: {
|
76
|
+
id: 'DOUBLE',
|
77
|
+
sql_type: 'DOUBLE'
|
78
|
+
}
|
79
|
+
}
|
80
|
+
},
|
81
|
+
{
|
82
|
+
name: 'date_array',
|
83
|
+
type: {
|
84
|
+
id: 'LIST',
|
85
|
+
sql_type: 'DATE[]',
|
86
|
+
child: {
|
87
|
+
id: 'DATE',
|
88
|
+
sql_type: 'DATE'
|
89
|
+
}
|
90
|
+
}
|
91
|
+
},
|
92
|
+
{
|
93
|
+
name: 'timestamp_array',
|
94
|
+
type: {
|
95
|
+
id: 'LIST',
|
96
|
+
sql_type: 'TIMESTAMP[]',
|
97
|
+
child: {
|
98
|
+
id: 'TIMESTAMP',
|
99
|
+
sql_type: 'TIMESTAMP'
|
100
|
+
}
|
101
|
+
}
|
102
|
+
},
|
103
|
+
{
|
104
|
+
name: 'timestamptz_array',
|
105
|
+
type: {
|
106
|
+
id: 'LIST',
|
107
|
+
sql_type: 'TIMESTAMP WITH TIME ZONE[]',
|
108
|
+
child: {
|
109
|
+
id: 'TIMESTAMP WITH TIME ZONE',
|
110
|
+
sql_type: 'TIMESTAMP WITH TIME ZONE'
|
111
|
+
}
|
112
|
+
}
|
113
|
+
},
|
114
|
+
{
|
115
|
+
name: 'varchar_array',
|
116
|
+
type: {
|
117
|
+
id: 'LIST',
|
118
|
+
sql_type: 'VARCHAR[]',
|
119
|
+
child: {
|
120
|
+
id: 'VARCHAR',
|
121
|
+
sql_type: 'VARCHAR'
|
122
|
+
}
|
123
|
+
}
|
124
|
+
},
|
125
|
+
{
|
126
|
+
name: 'nested_int_array',
|
127
|
+
type: {
|
128
|
+
id: 'LIST',
|
129
|
+
sql_type: 'INTEGER[][]',
|
130
|
+
child: {
|
131
|
+
id: 'LIST',
|
132
|
+
sql_type: 'INTEGER[]',
|
133
|
+
child: {
|
134
|
+
id: 'INTEGER',
|
135
|
+
sql_type: 'INTEGER'
|
136
|
+
}
|
137
|
+
}
|
138
|
+
}
|
139
|
+
},
|
140
|
+
{
|
141
|
+
name: 'struct',
|
142
|
+
type: {
|
143
|
+
id: 'STRUCT',
|
144
|
+
sql_type: 'STRUCT(a INTEGER, b VARCHAR)',
|
145
|
+
children: [
|
146
|
+
{
|
147
|
+
name: 'a',
|
148
|
+
type: {
|
149
|
+
id: 'INTEGER',
|
150
|
+
sql_type: 'INTEGER'
|
151
|
+
}
|
152
|
+
},
|
153
|
+
{
|
154
|
+
name: 'b',
|
155
|
+
type: {
|
156
|
+
id: 'VARCHAR',
|
157
|
+
sql_type: 'VARCHAR'
|
158
|
+
}
|
159
|
+
}
|
160
|
+
]
|
161
|
+
}
|
162
|
+
},
|
163
|
+
{
|
164
|
+
name: 'struct_of_arrays',
|
165
|
+
type: {
|
166
|
+
id: 'STRUCT',
|
167
|
+
sql_type: 'STRUCT(a INTEGER[], b VARCHAR[])',
|
168
|
+
children: [
|
169
|
+
{
|
170
|
+
name: 'a',
|
171
|
+
type: {
|
172
|
+
id: 'LIST',
|
173
|
+
sql_type: 'INTEGER[]',
|
174
|
+
child: {
|
175
|
+
id: 'INTEGER',
|
176
|
+
sql_type: 'INTEGER'
|
177
|
+
}
|
178
|
+
}
|
179
|
+
},
|
180
|
+
{
|
181
|
+
name: 'b',
|
182
|
+
type: {
|
183
|
+
id: 'LIST',
|
184
|
+
sql_type: 'VARCHAR[]',
|
185
|
+
child: {
|
186
|
+
id: 'VARCHAR',
|
187
|
+
sql_type: 'VARCHAR'
|
188
|
+
}
|
189
|
+
}
|
190
|
+
}
|
191
|
+
]
|
192
|
+
}
|
193
|
+
},
|
194
|
+
{
|
195
|
+
name: 'array_of_structs',
|
196
|
+
type: {
|
197
|
+
id: 'LIST',
|
198
|
+
sql_type: 'STRUCT(a INTEGER, b VARCHAR)[]',
|
199
|
+
child: {
|
200
|
+
id: 'STRUCT',
|
201
|
+
sql_type: 'STRUCT(a INTEGER, b VARCHAR)',
|
202
|
+
children: [
|
203
|
+
{
|
204
|
+
name: 'a',
|
205
|
+
type: {
|
206
|
+
id: 'INTEGER',
|
207
|
+
sql_type: 'INTEGER'
|
208
|
+
}
|
209
|
+
},
|
210
|
+
{
|
211
|
+
name: 'b',
|
212
|
+
type: {
|
213
|
+
id: 'VARCHAR',
|
214
|
+
sql_type: 'VARCHAR'
|
215
|
+
}
|
216
|
+
}
|
217
|
+
]
|
218
|
+
}
|
219
|
+
}
|
220
|
+
},
|
221
|
+
{
|
222
|
+
name: 'map',
|
223
|
+
type: {
|
224
|
+
id: 'MAP',
|
225
|
+
sql_type: 'MAP(VARCHAR, VARCHAR)',
|
226
|
+
key: {
|
227
|
+
id: 'VARCHAR',
|
228
|
+
sql_type: 'VARCHAR'
|
229
|
+
},
|
230
|
+
value: {
|
231
|
+
id: 'VARCHAR',
|
232
|
+
sql_type: 'VARCHAR'
|
233
|
+
}
|
234
|
+
}
|
235
|
+
}
|
236
|
+
]
|
237
|
+
|
238
|
+
assert.deepEqual(cols, expected);
|
239
|
+
|
240
|
+
});
|
241
|
+
stmt.finalize(done);
|
242
|
+
});
|
243
|
+
});
|