Skip to content

Planning correctly for order by queries on time which previously thre… #13965

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

Merged
cheddar merged 4 commits into from
Apr 4, 2023
Merged

Planning correctly for order by queries on time which previously thre… #13965

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
06b576b
Planning correctly for order by queries on time which previously thre…
somu-imply Mar 22, 2023
5907d41
Updating toDruidQueryForExplaining on a query data source if there is…
somu-imply Mar 23, 2023
a6f73eb
Merge remote-tracking branch 'upstream/master' into order_by_planning
somu-imply Mar 23, 2023
a9fdcb8
Fixing an unit test
somu-imply Mar 23, 2023
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
7 changes: 5 additions & 2 deletions sql/src/main/java/org/apache/druid/sql/calcite/rel/DruidOuterQueryRel.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,7 @@
import org.apache.druid.java.util.common.StringUtils;
import org.apache.druid.query.Druids;
import org.apache.druid.query.QueryDataSource;
import org.apache.druid.query.TableDataSource;
import org.apache.druid.segment.column.RowSignature;
import org.apache.druid.sql.calcite.planner.PlannerContext;
import org.apache.druid.sql.calcite.table.RowSignatures;
Expand All @@ -46,7 +47,9 @@
*/
public class DruidOuterQueryRel extends DruidRel<DruidOuterQueryRel>
{
private static final QueryDataSource DUMMY_DATA_SOURCE = new QueryDataSource(
private static final TableDataSource DUMMY_DATA_SOURCE = new TableDataSource("__subquery__");

private static final QueryDataSource DUMMY_QUERY_DATA_SOURCE = new QueryDataSource(
Druids.newScanQueryBuilder().dataSource("__subquery__").eternityInterval().build()
);

Expand Down Expand Up @@ -117,7 +120,7 @@ public DruidQuery toDruidQuery(final boolean finalizeAggregations)
public DruidQuery toDruidQueryForExplaining()
{
return partialQuery.build(
DUMMY_DATA_SOURCE,
partialQuery.getWindow() == null ? DUMMY_DATA_SOURCE : DUMMY_QUERY_DATA_SOURCE,
RowSignatures.fromRelDataType(
sourceRel.getRowType().getFieldNames(),
sourceRel.getRowType()
Expand Down
12 changes: 2 additions & 10 deletions sql/src/main/java/org/apache/druid/sql/calcite/rel/DruidQuery.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1439,7 +1439,8 @@ private ScanQuery toScanQuery()
}

if (!plannerContext.featureAvailable(EngineFeature.SCAN_ORDER_BY_NON_TIME) && !orderByColumns.isEmpty()) {
if (orderByColumns.size() > 1 || !ColumnHolder.TIME_COLUMN_NAME.equals(orderByColumns.get(0).getColumnName())) {
if (orderByColumns.size() > 1 || orderByColumns.stream()
.anyMatch(orderBy -> !orderBy.getColumnName().equals(ColumnHolder.TIME_COLUMN_NAME))) {
// Cannot handle this ordering.
// Scan cannot ORDER BY non-time columns.
plannerContext.setPlanningError(
Expand All @@ -1448,15 +1449,6 @@ private ScanQuery toScanQuery()
);
return null;
}
if (!dataSource.isConcrete()) {
Copy link
Copy Markdown
Contributor Author

@somu-imply somu-imply Mar 22, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This was a change in #12418 that was reverted. Since we are using a table data source and not a query data source now the isConcrete() check is not needed and can be removed

// Cannot handle this ordering.
// Scan cannot ORDER BY non-concrete datasources on _any_ column.
plannerContext.setPlanningError(
"SQL query requires order by on non-concrete datasource [%s], which is not supported.",
dataSource
);
return null;
}
}

// Compute the list of columns to select, sorted and deduped.
Expand Down
2 changes: 1 addition & 1 deletion sql/src/test/java/org/apache/druid/sql/calcite/CalciteExplainQueryTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,7 +109,7 @@ public void testExplainExactCountDistinctOfSemiJoinResult()
+ " )\n"
+ ")";
final String legacyExplanation =
"DruidOuterQueryRel(query=[{\"queryType\":\"groupBy\",\"dataSource\":{\"type\":\"query\",\"query\":{\"queryType\":\"scan\",\"dataSource\":{\"type\":\"table\",\"name\":\"__subquery__\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"resultFormat\":\"list\",\"granularity\":{\"type\":\"all\"}}},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"granularity\":{\"type\":\"all\"},\"dimensions\":[],\"aggregations\":[{\"type\":\"count\",\"name\":\"a0\"}],\"limitSpec\":{\"type\":\"NoopLimitSpec\"},\"context\":{\"defaultTimeout\":300000,\"maxScatterGatherBytes\":9223372036854775807,\"sqlCurrentTimestamp\":\"2000-01-01T00:00:00Z\",\"sqlQueryId\":\"dummy\",\"vectorize\":\"false\",\"vectorizeVirtualColumns\":\"false\"}}], signature=[{a0:LONG}])\n"
"DruidOuterQueryRel(query=[{\"queryType\":\"timeseries\",\"dataSource\":{\"type\":\"table\",\"name\":\"__subquery__\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"granularity\":{\"type\":\"all\"},\"aggregations\":[{\"type\":\"count\",\"name\":\"a0\"}],\"context\":{\"defaultTimeout\":300000,\"maxScatterGatherBytes\":9223372036854775807,\"sqlCurrentTimestamp\":\"2000-01-01T00:00:00Z\",\"sqlQueryId\":\"dummy\",\"vectorize\":\"false\",\"vectorizeVirtualColumns\":\"false\"}}], signature=[{a0:LONG}])\n"
+ " DruidJoinQueryRel(condition=[=(SUBSTRING($2, 1, 1), $8)], joinType=[inner], query=[{\"queryType\":\"groupBy\",\"dataSource\":{\"type\":\"table\",\"name\":\"__join__\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"granularity\":{\"type\":\"all\"},\"dimensions\":[{\"type\":\"default\",\"dimension\":\"dim2\",\"outputName\":\"d0\",\"outputType\":\"STRING\"}],\"limitSpec\":{\"type\":\"NoopLimitSpec\"},\"context\":{\"defaultTimeout\":300000,\"maxScatterGatherBytes\":9223372036854775807,\"sqlCurrentTimestamp\":\"2000-01-01T00:00:00Z\",\"sqlQueryId\":\"dummy\",\"vectorize\":\"false\",\"vectorizeVirtualColumns\":\"false\"}}], signature=[{d0:STRING}])\n"
+ " DruidQueryRel(query=[{\"queryType\":\"scan\",\"dataSource\":{\"type\":\"table\",\"name\":\"foo\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"resultFormat\":\"compactedList\",\"columns\":[\"__time\",\"cnt\",\"dim1\",\"dim2\",\"dim3\",\"m1\",\"m2\",\"unique_dim1\"],\"legacy\":false,\"context\":{\"defaultTimeout\":300000,\"maxScatterGatherBytes\":9223372036854775807,\"sqlCurrentTimestamp\":\"2000-01-01T00:00:00Z\",\"sqlQueryId\":\"dummy\",\"vectorize\":\"false\",\"vectorizeVirtualColumns\":\"false\"},\"granularity\":{\"type\":\"all\"}}], signature=[{__time:LONG, dim1:STRING, dim2:STRING, dim3:STRING, cnt:LONG, m1:FLOAT, m2:DOUBLE, unique_dim1:COMPLEX<hyperUnique>}])\n"
+ " DruidQueryRel(query=[{\"queryType\":\"groupBy\",\"dataSource\":{\"type\":\"table\",\"name\":\"foo\"},\"intervals\":{\"type\":\"intervals\",\"intervals\":[\"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z\"]},\"filter\":{\"type\":\"not\",\"field\":{\"type\":\"selector\",\"dimension\":\"dim1\",\"value\":null}},\"granularity\":{\"type\":\"all\"},\"dimensions\":[{\"type\":\"extraction\",\"dimension\":\"dim1\",\"outputName\":\"d0\",\"outputType\":\"STRING\",\"extractionFn\":{\"type\":\"substring\",\"index\":0,\"length\":1}}],\"limitSpec\":{\"type\":\"NoopLimitSpec\"},\"context\":{\"defaultTimeout\":300000,\"maxScatterGatherBytes\":9223372036854775807,\"sqlCurrentTimestamp\":\"2000-01-01T00:00:00Z\",\"sqlQueryId\":\"dummy\",\"vectorize\":\"false\",\"vectorizeVirtualColumns\":\"false\"}}], signature=[{d0:STRING}])\n";
Expand Down
58 changes: 58 additions & 0 deletions sql/src/test/java/org/apache/druid/sql/calcite/CalciteJoinQueryTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5264,4 +5264,62 @@ public void testRegressionFilteredAggregatorsSubqueryJoins(Map<String, Object> q
)
);
}

@Test
public void testJoinWithAliasAndOrderByNoGroupBy()
{
minTopNThreshold = 1;
Map<String, Object> context = new HashMap<>(QUERY_CONTEXT_DEFAULT);
context.put(PlannerConfig.CTX_KEY_USE_APPROXIMATE_TOPN, false);
testQuery(
"select t1.__time from druid.foo as t1 join\n"
+ " druid.numfoo as t2 on t1.dim2 = t2.dim2\n"
+ " order by t1.__time ASC ",
context, // turn on exact topN
ImmutableList.of(
newScanQueryBuilder()
.intervals(querySegmentSpec(Filtration.eternity()))
.dataSource(
JoinDataSource.create(
new TableDataSource(CalciteTests.DATASOURCE1),
new QueryDataSource(
newScanQueryBuilder()
.dataSource(CalciteTests.DATASOURCE3)
.intervals(querySegmentSpec(Intervals.of(
"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z")))
.columns("dim2")
.context(context)
.build()
),
"j0.",
"(\"dim2\" == \"j0.dim2\")",
JoinType.INNER,
null,
ExprMacroTable.nil(),
CalciteTests.createJoinableFactoryWrapper()
)
)
.columns("__time")
.order(ScanQuery.Order.ASCENDING)
.context(context)
.build()
),
NullHandling.sqlCompatible()
? ImmutableList.of(
new Object[]{946684800000L},
new Object[]{946684800000L},
new Object[]{946857600000L},
new Object[]{978307200000L},
new Object[]{978307200000L},
new Object[]{978393600000L}
)
: ImmutableList.of(
new Object[]{946684800000L},
new Object[]{946684800000L},
new Object[]{978307200000L},
new Object[]{978307200000L},
new Object[]{978393600000L}
)
);
}
}
103 changes: 96 additions & 7 deletions sql/src/test/java/org/apache/druid/sql/calcite/CalciteQueryTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7306,21 +7306,26 @@ public void testMinMaxAvgDailyCountWithLimit()
)
)
.setInterval(querySegmentSpec(Filtration.eternity()))
.setLimitSpec(
new DefaultLimitSpec(
ImmutableList.of(),
1
)
)
.setGranularity(Granularities.ALL)
.setAggregatorSpecs(
useDefault ?
aggregators(
new LongMaxAggregatorFactory("_a0", "a0"),
new LongMinAggregatorFactory("_a1", "a0"),
new LongSumAggregatorFactory("_a2:sum", "a0"),
new CountAggregatorFactory("_a2:count"),
new LongMaxAggregatorFactory("_a3", "d0"),
new CountAggregatorFactory("_a4")
) : aggregators(
new LongMaxAggregatorFactory("_a0", "a0"),
new LongMinAggregatorFactory("_a1", "a0"),
new LongSumAggregatorFactory("_a2:sum", "a0"),
new FilteredAggregatorFactory(
new CountAggregatorFactory("_a2:count"),
not(selector("a0", null, null))
),
new LongMaxAggregatorFactory("_a3", "d0"),
new CountAggregatorFactory("_a4")
Comment on lines +7319 to +7328
Copy link
Copy Markdown
Contributor

@imply-cheddar imply-cheddar Mar 24, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why does this change generate this adjustment to the plan?

Copy link
Copy Markdown
Contributor Author

@somu-imply somu-imply Mar 25, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No clue but in the case of the useDefault=false it uses the same thing but with a check of not null so thought it was kind of correct

)
)
.setPostAggregatorSpecs(
Expand All @@ -7333,7 +7338,7 @@ public void testMinMaxAvgDailyCountWithLimit()
new FieldAccessPostAggregator(null, "_a2:count")
)
),
expressionPostAgg("s0", "timestamp_extract(\"_a3\",'EPOCH','UTC')")
expressionPostAgg("p0", "timestamp_extract(\"_a3\",'EPOCH','UTC')")
)
)
.setContext(QUERY_CONTEXT_DEFAULT)
Expand Down Expand Up @@ -14604,4 +14609,88 @@ public void testComplexDecodeAgg()
)
);
}

@Test
public void testOrderByAlongWithInternalScanQuery()
{
testQuery(
"select __time as t, m1 from druid.foo where (m1 in (select distinct m1 from druid.foo)) order by 1 limit 1",
ImmutableList.of(
newScanQueryBuilder()
.intervals(querySegmentSpec(Filtration.eternity()))
.dataSource(
JoinDataSource.create(
new TableDataSource(CalciteTests.DATASOURCE1),
new QueryDataSource(
GroupByQuery.builder()
.setGranularity(Granularities.ALL)
.setDataSource(CalciteTests.DATASOURCE1)
.setInterval(querySegmentSpec(Intervals.of(
"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z")))
.setDimensions(new DefaultDimensionSpec("m1", "d0", ColumnType.FLOAT))
.build()
),
"j0.",
"(\"m1\" == \"j0.d0\")",
JoinType.INNER,
null,
ExprMacroTable.nil(),
CalciteTests.createJoinableFactoryWrapper()
)
)
.context(QUERY_CONTEXT_DEFAULT)
.intervals(querySegmentSpec(Intervals.of(
"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z")))
.limit(1)
.columns(ImmutableList.of("__time", "m1"))
.order(ScanQuery.Order.ASCENDING)
.build()
),
ImmutableList.of(
new Object[]{946684800000L, 1.0f}
)
);
}

@Test
public void testOrderByAlongWithInternalScanQueryNoDistinct()
{
testQuery(
"select __time, m1 from druid.foo where (m1 in (select m1 from druid.foo)) order by __time DESC limit 1",
ImmutableList.of(
newScanQueryBuilder()
.intervals(querySegmentSpec(Filtration.eternity()))
.dataSource(
JoinDataSource.create(
new TableDataSource(CalciteTests.DATASOURCE1),
new QueryDataSource(
GroupByQuery.builder()
.setGranularity(Granularities.ALL)
.setDataSource(CalciteTests.DATASOURCE1)
.setInterval(querySegmentSpec(Intervals.of(
"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z")))
.setDimensions(new DefaultDimensionSpec("m1", "d0", ColumnType.FLOAT))
.build()
),
"j0.",
"(\"m1\" == \"j0.d0\")",
JoinType.INNER,
null,
ExprMacroTable.nil(),
CalciteTests.createJoinableFactoryWrapper()
)
)
.context(QUERY_CONTEXT_DEFAULT)
.intervals(querySegmentSpec(Intervals.of(
"-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z")))
.limit(1)
.columns(ImmutableList.of("__time", "m1"))
.order(ScanQuery.Order.DESCENDING)
.build()
),
ImmutableList.of(
new Object[]{978480000000L, 6.0f}
)
);
}
}