Skip to content

[FLINK-40214][table] Preserve JSON_OBJECTAGG key when wrapping value #28795

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Au-Miner wants to merge 1 commit into
base: master
Choose a base branch
from
Open

[FLINK-40214][table] Preserve JSON_OBJECTAGG key when wrapping value #28795

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 9 additions & 10 deletions ...a/org/apache/flink/table/planner/plan/rules/logical/WrapJsonAggFunctionArgumentsRule.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,6 @@
import org.apache.calcite.sql.fun.SqlJsonArrayAggAggFunction;
import org.apache.calcite.sql.fun.SqlJsonObjectAggAggFunction;
import org.apache.calcite.tools.RelBuilder;
import org.apache.calcite.util.mapping.MappingType;
import org.apache.calcite.util.mapping.Mappings;
import org.apache.calcite.util.mapping.Mappings.TargetMapping;
import org.immutables.value.Value;

import java.util.ArrayList;
Expand Down Expand Up @@ -126,14 +123,16 @@ private Aggregate wrapJsonAggregate(Aggregate aggregate, RelBuilder relBuilder)
valueIndicesAfterProjection);

List<AggregateCall> newWrappedArgCallList = new ArrayList<>(aggCallList);
final int newInputCount = inputCount + valueIndicesAfterProjection.size();
for (Integer jsonAggCallIndex : wrapIndicesMap.keySet()) {
final TargetMapping argsMapping =
Mappings.create(MappingType.BIJECTION, newInputCount, newInputCount);
Integer valueIndex = wrapIndicesMap.get(jsonAggCallIndex);
argsMapping.set(valueIndex, valueIndicesAfterProjection.get(valueIndex));
final AggregateCall newAggregateCall =
newWrappedArgCallList.get(jsonAggCallIndex).transform(argsMapping);
final AggregateCall aggregateCall = newWrappedArgCallList.get(jsonAggCallIndex);
final List<Integer> newArgList = new ArrayList<>(aggregateCall.getArgList());
// AggregateCall argument positions are zero-based: JSON_OBJECTAGG has (key, value),
// so its value is at position 1; JSON_ARRAYAGG has only (value), at position 0.
final int valueArgPosition =
aggregateCall.getAggregation() instanceof SqlJsonObjectAggAggFunction ? 1 : 0;
final Integer valueIndex = wrapIndicesMap.get(jsonAggCallIndex);
newArgList.set(valueArgPosition, valueIndicesAfterProjection.get(valueIndex));
final AggregateCall newAggregateCall = aggregateCall.withArgList(newArgList);
newWrappedArgCallList.set(jsonAggCallIndex, newAggregateCall);
}

Expand Down
30 changes: 30 additions & 0 deletions ...rc/test/java/org/apache/flink/table/planner/functions/JsonAggregationFunctionsITCase.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,6 +60,15 @@ public Stream<TestSpec> getTestCaseSpecs() {
ROW(VARCHAR(2000).notNull()),
ROW(STRING().notNull()),
Collections.singletonList(Row.of("{\"A\":1,\"B\":null,\"C\":3}"))),
TestSpec.forFunction(BuiltInFunctionDefinitions.JSON_OBJECTAGG_NULL_ON_NULL)
.withDescription("Aggregation With Same Key And Value")
.withSource(
ROW(STRING()),
Arrays.asList(Row.ofKind(INSERT, "A"), Row.ofKind(INSERT, "B")))
.testSqlResult(
source -> "SELECT JSON_OBJECTAGG(f0 VALUE f0) FROM " + source,
ROW(VARCHAR(2000).notNull()),
Collections.singletonList(Row.of("{\"A\":\"A\",\"B\":\"B\"}"))),
TestSpec.forFunction(BuiltInFunctionDefinitions.JSON_OBJECTAGG_ABSENT_ON_NULL)
.withDescription("Omits NULLs")
.withSource(
Expand Down Expand Up @@ -273,6 +282,27 @@ public Stream<TestSpec> getTestCaseSpecs() {
+ " GROUP BY TUMBLE(f2, INTERVAL '5' SECOND)",
ROW(VARCHAR(2000).notNull()),
Arrays.asList(Row.of("{\"A\":1,\"B\":2}"), Row.of("{\"C\":3}"))),
TestSpec.forFunction(BuiltInFunctionDefinitions.JSON_OBJECTAGG_NULL_ON_NULL)
.withDescription("Window Aggregation With Same Key And Value")
.withSource(
ROW(STRING(), TIMESTAMP(3)),
Arrays.asList(
Row.ofKind(
INSERT,
"A",
LocalDateTime.parse("2020-01-01T00:00:01")),
Row.ofKind(
INSERT,
"B",
LocalDateTime.parse("2020-01-01T00:00:02"))))
.withWatermark("f1", "f1 - INTERVAL '1' SECOND")
.testSqlResult(
source ->
"SELECT JSON_OBJECTAGG(f0 VALUE f0) FROM "
+ source
+ " GROUP BY TUMBLE(f1, INTERVAL '5' SECOND)",
ROW(VARCHAR(2000).notNull()),
Collections.singletonList(Row.of("{\"A\":\"A\",\"B\":\"B\"}"))),
TestSpec.forFunction(BuiltInFunctionDefinitions.JSON_OBJECTAGG_NULL_ON_NULL)
.withDescription("Window Group Aggregation With Other Aggs")
.withSource(
Expand Down
20 changes: 10 additions & 10 deletions ...rg/apache/flink/table/planner/plan/rules/logical/WrapJsonAggFunctionArgumentsRuleTest.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -718,7 +718,7 @@ LogicalAggregate(group=[{}], EXPR$0=[COUNT()], EXPR$1=[JSON_OBJECTAGG_NULL_ON_NU
</Resource>
<Resource name="optimized rel plan">
<![CDATA[
GroupAggregate(select=[COUNT(*) AS EXPR$0, JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$1])
GroupAggregate(select=[COUNT(*) AS EXPR$0, JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$1])
+- Exchange(distribution=[single])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[f1], metadata=[]]], fields=[f1])
Expand All @@ -740,7 +740,7 @@ LogicalProject(EXPR$0=[$1], EXPR$1=[$2])
</Resource>
<Resource name="optimized rel plan">
<![CDATA[
GroupWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[COUNT(*) AS EXPR$0, JSON_OBJECTAGG_NULL_ON_NULL($f2, $f2) AS EXPR$1])
GroupWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[COUNT(*) AS EXPR$0, JSON_OBJECTAGG_NULL_ON_NULL(f1, $f2) AS EXPR$1])
+- Exchange(distribution=[single])
+- Calc(select=[rt, f1, JSON_STRING(f1) AS $f2])
+- WatermarkAssigner(rowtime=[rt], watermark=[-(rt, 1000:INTERVAL SECOND)])
Expand All @@ -763,7 +763,7 @@ LogicalAggregate(group=[{}], EXPR$0=[COUNT()], EXPR$1=[JSON_OBJECTAGG_NULL_ON_NU
<![CDATA[
SortAggregate(isMerge=[true], select=[Final_COUNT(count1$0) AS EXPR$0, Final_JSON_OBJECTAGG_NULL_ON_NULL(EXPR$1) AS EXPR$1])
+- Exchange(distribution=[single])
+- LocalSortAggregate(select=[Partial_COUNT(*) AS count1$0, Partial_JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$1])
+- LocalSortAggregate(select=[Partial_COUNT(*) AS count1$0, Partial_JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$1])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[f1], metadata=[]]], fields=[f1])
]]>
Expand All @@ -786,7 +786,7 @@ LogicalProject(EXPR$0=[$1], EXPR$1=[$2])
SortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Final_COUNT(count1$0) AS EXPR$0, Final_JSON_OBJECTAGG_NULL_ON_NULL(EXPR$1) AS EXPR$1])
+- Sort(orderBy=[assignedWindow$ ASC])
+- Exchange(distribution=[single])
+- LocalSortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Partial_COUNT(*) AS count1$0, Partial_JSON_OBJECTAGG_NULL_ON_NULL($f2, $f2) AS EXPR$1])
+- LocalSortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Partial_COUNT(*) AS count1$0, Partial_JSON_OBJECTAGG_NULL_ON_NULL(f1, $f2) AS EXPR$1])
+- Calc(select=[rt, f1, JSON_STRING(f1) AS $f2])
+- Sort(orderBy=[rt ASC])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[rt, f1], metadata=[]]], fields=[rt, f1])
Expand All @@ -809,7 +809,7 @@ LogicalAggregate(group=[{}], EXPR$0=[JSON_OBJECTAGG_NULL_ON_NULL($0, $0)])
</Resource>
<Resource name="optimized rel plan">
<![CDATA[
GroupAggregate(select=[JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$0])
GroupAggregate(select=[JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$0])
+- Exchange(distribution=[single])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- WindowTableFunction(window=[TUMBLE(time_col=[rt], size=[5 s])])
Expand All @@ -835,7 +835,7 @@ LogicalAggregate(group=[{}], EXPR$0=[JSON_OBJECTAGG_NULL_ON_NULL($0, $0)])
<![CDATA[
SortAggregate(isMerge=[true], select=[Final_JSON_OBJECTAGG_NULL_ON_NULL(EXPR$0) AS EXPR$0])
+- Exchange(distribution=[single])
+- LocalSortAggregate(select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$0])
+- LocalSortAggregate(select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$0])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- WindowTableFunction(window=[TUMBLE(time_col=[rt], size=[5 s])])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[f1, rt], metadata=[]]], fields=[f1, rt])
Expand All @@ -855,7 +855,7 @@ LogicalAggregate(group=[{}], EXPR$0=[JSON_OBJECTAGG_NULL_ON_NULL($0, $0)])
</Resource>
<Resource name="optimized rel plan">
<![CDATA[
GroupAggregate(select=[JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$0])
GroupAggregate(select=[JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$0])
+- Exchange(distribution=[single])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[f1], metadata=[]]], fields=[f1])
Expand All @@ -877,7 +877,7 @@ LogicalProject(EXPR$0=[$1])
</Resource>
<Resource name="optimized rel plan">
<![CDATA[
GroupWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[JSON_OBJECTAGG_NULL_ON_NULL($f2, $f2) AS EXPR$0])
GroupWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[JSON_OBJECTAGG_NULL_ON_NULL(f1, $f2) AS EXPR$0])
+- Exchange(distribution=[single])
+- Calc(select=[rt, f1, JSON_STRING(f1) AS $f2])
+- WatermarkAssigner(rowtime=[rt], watermark=[-(rt, 1000:INTERVAL SECOND)])
Expand All @@ -900,7 +900,7 @@ LogicalAggregate(group=[{}], EXPR$0=[JSON_OBJECTAGG_NULL_ON_NULL($0, $0)])
<![CDATA[
SortAggregate(isMerge=[true], select=[Final_JSON_OBJECTAGG_NULL_ON_NULL(EXPR$0) AS EXPR$0])
+- Exchange(distribution=[single])
+- LocalSortAggregate(select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL($f1, $f1) AS EXPR$0])
+- LocalSortAggregate(select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL(f1, $f1) AS EXPR$0])
+- Calc(select=[f1, JSON_STRING(f1) AS $f1])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[f1], metadata=[]]], fields=[f1])
]]>
Expand All @@ -923,7 +923,7 @@ LogicalProject(EXPR$0=[$1])
SortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Final_JSON_OBJECTAGG_NULL_ON_NULL(EXPR$0) AS EXPR$0])
+- Sort(orderBy=[assignedWindow$ ASC])
+- Exchange(distribution=[single])
+- LocalSortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL($f2, $f2) AS EXPR$0])
+- LocalSortWindowAggregate(window=[TumblingGroupWindow('w$, rt, 5000)], select=[Partial_JSON_OBJECTAGG_NULL_ON_NULL(f1, $f2) AS EXPR$0])
+- Calc(select=[rt, f1, JSON_STRING(f1) AS $f2])
+- Sort(orderBy=[rt ASC])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[rt, f1], metadata=[]]], fields=[rt, f1])
Expand Down