Skip to content

Commit

Permalink
SQL: UNION ALL operator. (#6314)
Browse files Browse the repository at this point in the history
* SQL: UNION ALL operator.

* Remove unused import.
  • Loading branch information
gianm authored and fjy committed Sep 10, 2018
1 parent e6e068c commit 4669f08
Show file tree
Hide file tree
Showing 10 changed files with 478 additions and 31 deletions.
4 changes: 4 additions & 0 deletions docs/content/querying/sql.md
Original file line number Diff line number Diff line change
Expand Up @@ -44,6 +44,7 @@ FROM table
[ HAVING expr ]
[ ORDER BY expr [ ASC | DESC ], expr [ ASC | DESC ], ... ]
[ LIMIT limit ]
[ UNION ALL <another query> ]
```

The FROM clause refers to either a Druid datasource, like `druid.foo`, an [INFORMATION_SCHEMA table](#retrieving-metadata), a
Expand Down Expand Up @@ -74,6 +75,9 @@ versions of Druid will support pushing down limits using the native GroupBy quer
adding a limit doesn't change performance very much, then it's likely that Druid didn't push down the limit for your
query.

The "UNION ALL" operator can be used to fuse multiple queries together. Their results will be concatenated, and each
query will run separately, back to back (not in parallel). Druid does not currently support "UNION" without "ALL".

Add "EXPLAIN PLAN FOR" to the beginning of any query to see how it would be run as a native Druid query. In this case,
the query will not actually be executed.

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -68,6 +68,8 @@ public class DruidConvertletTable implements SqlRexConvertletTable
.add(SqlStdOperatorTable.SYMMETRIC_NOT_BETWEEN)
.add(SqlStdOperatorTable.ITEM)
.add(SqlStdOperatorTable.TIMESTAMP_ADD)
.add(SqlStdOperatorTable.UNION)
.add(SqlStdOperatorTable.UNION_ALL)
.build();

private final Map<SqlOperator, SqlRexConvertlet> table;
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -20,15 +20,6 @@
package org.apache.druid.sql.calcite.planner;

import com.google.common.collect.ImmutableList;
import org.apache.druid.sql.calcite.rel.QueryMaker;
import org.apache.druid.sql.calcite.rule.CaseFilteredAggregatorRule;
import org.apache.druid.sql.calcite.rule.DruidRelToBindableRule;
import org.apache.druid.sql.calcite.rule.DruidRelToDruidRule;
import org.apache.druid.sql.calcite.rule.DruidRules;
import org.apache.druid.sql.calcite.rule.DruidSemiJoinRule;
import org.apache.druid.sql.calcite.rule.DruidTableScanRule;
import org.apache.druid.sql.calcite.rule.ProjectAggregatePruneUnusedCallRule;
import org.apache.druid.sql.calcite.rule.SortCollapseRule;
import org.apache.calcite.interpreter.Bindables;
import org.apache.calcite.plan.RelOptLattice;
import org.apache.calcite.plan.RelOptMaterialization;
Expand Down Expand Up @@ -78,6 +69,15 @@
import org.apache.calcite.tools.Program;
import org.apache.calcite.tools.Programs;
import org.apache.calcite.tools.RelBuilder;
import org.apache.druid.sql.calcite.rel.QueryMaker;
import org.apache.druid.sql.calcite.rule.CaseFilteredAggregatorRule;
import org.apache.druid.sql.calcite.rule.DruidRelToBindableRule;
import org.apache.druid.sql.calcite.rule.DruidRelToDruidRule;
import org.apache.druid.sql.calcite.rule.DruidRules;
import org.apache.druid.sql.calcite.rule.DruidSemiJoinRule;
import org.apache.druid.sql.calcite.rule.DruidTableScanRule;
import org.apache.druid.sql.calcite.rule.ProjectAggregatePruneUnusedCallRule;
import org.apache.druid.sql.calcite.rule.SortCollapseRule;

import java.util.List;

Expand Down
11 changes: 8 additions & 3 deletions sql/src/main/java/org/apache/druid/sql/calcite/rel/DruidRel.java
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,6 @@
package org.apache.druid.sql.calcite.rel;

import com.google.common.base.Throwables;
import org.apache.druid.java.util.common.guava.Accumulator;
import org.apache.druid.java.util.common.guava.Sequence;
import org.apache.druid.sql.calcite.planner.PlannerContext;
import org.apache.calcite.DataContext;
import org.apache.calcite.interpreter.BindableRel;
import org.apache.calcite.interpreter.Node;
Expand All @@ -32,6 +29,9 @@
import org.apache.calcite.plan.RelOptCluster;
import org.apache.calcite.plan.RelTraitSet;
import org.apache.calcite.rel.AbstractRelNode;
import org.apache.druid.java.util.common.guava.Accumulator;
import org.apache.druid.java.util.common.guava.Sequence;
import org.apache.druid.sql.calcite.planner.PlannerContext;

import javax.annotation.Nullable;
import java.util.List;
Expand All @@ -46,6 +46,11 @@ protected DruidRel(RelOptCluster cluster, RelTraitSet traitSet, QueryMaker query
this.queryMaker = queryMaker;
}

/**
* Returns the PartialDruidQuery associated with this DruidRel, and which can be built on top of. Returns null
* if this rel cannot be built on top of.
*/
@Nullable
public abstract PartialDruidQuery getPartialDruidQuery();

/**
Expand Down
219 changes: 219 additions & 0 deletions sql/src/main/java/org/apache/druid/sql/calcite/rel/DruidUnionRel.java
Original file line number Diff line number Diff line change
@@ -0,0 +1,219 @@
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/

package org.apache.druid.sql.calcite.rel;

import com.google.common.base.Preconditions;
import com.google.common.collect.FluentIterable;
import org.apache.calcite.interpreter.BindableConvention;
import org.apache.calcite.plan.RelOptCluster;
import org.apache.calcite.plan.RelOptCost;
import org.apache.calcite.plan.RelOptPlanner;
import org.apache.calcite.plan.RelOptRule;
import org.apache.calcite.plan.RelTraitSet;
import org.apache.calcite.rel.RelNode;
import org.apache.calcite.rel.RelWriter;
import org.apache.calcite.rel.metadata.RelMetadataQuery;
import org.apache.calcite.rel.type.RelDataType;
import org.apache.druid.java.util.common.StringUtils;
import org.apache.druid.java.util.common.guava.Sequence;
import org.apache.druid.java.util.common.guava.Sequences;

import javax.annotation.Nullable;
import java.util.ArrayList;
import java.util.List;
import java.util.stream.Collectors;

public class DruidUnionRel extends DruidRel<DruidUnionRel>
{
private final RelDataType rowType;
private final List<RelNode> rels;
private final int limit;

private DruidUnionRel(
final RelOptCluster cluster,
final RelTraitSet traitSet,
final QueryMaker queryMaker,
final RelDataType rowType,
final List<RelNode> rels,
final int limit
)
{
super(cluster, traitSet, queryMaker);
this.rowType = rowType;
this.rels = rels;
this.limit = limit;
}

public static DruidUnionRel create(
final QueryMaker queryMaker,
final RelDataType rowType,
final List<RelNode> rels,
final int limit
)
{
Preconditions.checkState(rels.size() > 0, "rels must be nonempty");

return new DruidUnionRel(
rels.get(0).getCluster(),
rels.get(0).getTraitSet(),
queryMaker,
rowType,
new ArrayList<>(rels),
limit
);
}

@Override
@Nullable
public PartialDruidQuery getPartialDruidQuery()
{
return null;
}

@Override
public int getQueryCount()
{
return rels.stream().mapToInt(rel -> ((DruidRel) rel).getQueryCount()).sum();
}

@Override
@SuppressWarnings("unchecked")
public Sequence<Object[]> runQuery()
{
// Lazy: run each query in sequence, not all at once.
if (limit == 0) {
return Sequences.empty();
} else {
final Sequence baseSequence = Sequences.concat(
FluentIterable.from(rels).transform(rel -> ((DruidRel) rel).runQuery())
);

return limit > 0 ? baseSequence.limit(limit) : baseSequence;
}
}

@Override
public DruidUnionRel withPartialQuery(final PartialDruidQuery newQueryBuilder)
{
throw new UnsupportedOperationException();
}

@Nullable
@Override
public DruidQuery toDruidQuery(final boolean finalizeAggregations)
{
throw new UnsupportedOperationException();
}

@Override
public DruidQuery toDruidQueryForExplaining()
{
throw new UnsupportedOperationException();
}

@Override
public DruidUnionRel asBindable()
{
return new DruidUnionRel(
getCluster(),
getTraitSet().replace(BindableConvention.INSTANCE),
getQueryMaker(),
rowType,
rels.stream().map(rel -> RelOptRule.convert(rel, BindableConvention.INSTANCE)).collect(Collectors.toList()),
limit
);
}

@Override
public DruidUnionRel asDruidConvention()
{
return new DruidUnionRel(
getCluster(),
getTraitSet().replace(DruidConvention.instance()),
getQueryMaker(),
rowType,
rels.stream().map(rel -> RelOptRule.convert(rel, DruidConvention.instance())).collect(Collectors.toList()),
limit
);
}

@Override
public List<RelNode> getInputs()
{
return rels;
}

@Override
public void replaceInput(int ordinalInParent, RelNode p)
{
rels.set(ordinalInParent, p);
}

@Override
public RelNode copy(final RelTraitSet traitSet, final List<RelNode> inputs)
{
return new DruidUnionRel(
getCluster(),
traitSet,
getQueryMaker(),
rowType,
inputs,
limit
);
}

@Override
public List<String> getDatasourceNames()
{
return rels.stream()
.flatMap(rel -> ((DruidRel<?>) rel).getDatasourceNames().stream())
.distinct()
.collect(Collectors.toList());
}

@Override
public RelWriter explainTerms(RelWriter pw)
{
super.explainTerms(pw);

for (int i = 0; i < rels.size(); i++) {
pw.input(StringUtils.format("input#%d", i), rels.get(i));
}

return pw.item("limit", limit);
}

@Override
protected RelDataType deriveRowType()
{
return rowType;
}

@Override
public RelOptCost computeSelfCost(final RelOptPlanner planner, final RelMetadataQuery mq)
{
return planner.getCostFactory().makeCost(rels.stream().mapToDouble(mq::getRowCount).sum(), 0, 0);
}

public int getLimit()
{
return limit;
}
}
Loading

0 comments on commit 4669f08

Please sign in to comment.