sunchao commented on code in PR #35965:
URL: https://github.com/apache/spark/pull/35965#discussion_r863485508
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DataSourceV2ScanExecBase.scala:
##########
@@ -138,6 +138,15 @@ trait DataSourceV2ScanExecBase extends LeafExecNode {
}
}
+ override def outputOrdering: Seq[SortOrder] = scan match {
+ case s: SupportsReportOrdering if this.logicalLink.isDefined &&
+ // when groupedPartitions are applied (see `partitions` and
`outputPartitioning`) and
+ // multiple partitions are grouped together, the order inside individual
partitions gets lost
+ (partitions.length == 1 || groupedPartitions.forall(_.forall(_._2.length
== 1))) =>
Review Comment:
Hmm when `partitions.length == 1` I think partition combining can still
happen? `partitions` is of type `Seq[Seq[InputPartition]]` so even if its
length = 1, the inner `Seq[InputPartition]` can still contain more than one
elements.
##########
sql/core/src/test/java/test/org/apache/spark/sql/connector/JavaOrderAndPartitionAwareDataSource.java:
##########
@@ -0,0 +1,141 @@
+/*
+ * 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 test.org.apache.spark.sql.connector;
+
+import org.apache.spark.sql.connector.catalog.Table;
+import org.apache.spark.sql.connector.expressions.*;
+import org.apache.spark.sql.connector.read.*;
+import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning;
+import org.apache.spark.sql.connector.read.partitioning.Partitioning;
+import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning;
+import org.apache.spark.sql.util.CaseInsensitiveStringMap;
+
+public class JavaOrderAndPartitionAwareDataSource extends
JavaPartitionAwareDataSource {
+
+ static class MyScanBuilder extends JavaSimpleScanBuilder
+ implements SupportsReportPartitioning, SupportsReportOrdering {
+
+ private final Partitioning partitioning;
+ private final SortOrder[] ordering;
+
+ MyScanBuilder(String partitionKeys, String orderKeys) {
+ if (partitionKeys != null) {
+ String[] keys = partitionKeys.split(",");
+ Expression[] clustering = new Transform[keys.length];
+ for (int i=0; i< keys.length; i++) {
+ clustering[i] = Expressions.identity(keys[i]);
+ }
+ this.partitioning = new KeyGroupedPartitioning(clustering, 2);
+ } else {
+ this.partitioning = new UnknownPartitioning(2);
+ }
+
+ if (orderKeys != null) {
+ String[] keys = orderKeys.split(",");
+ this.ordering = new SortOrder[keys.length];
+ for (int i=0; i<keys.length; i++) {
+ this.ordering[i] = new MySortOrder(keys[i]);
+ }
+ } else {
+ this.ordering = new SortOrder[0];
+ }
+ }
+
+ @Override
+ public InputPartition[] planInputPartitions() {
+ InputPartition[] partitions = new InputPartition[2];
+ partitions[0] = new SpecificInputPartition(new int[]{1, 1, 3}, new
int[]{4, 5, 5});
+ partitions[1] = new SpecificInputPartition(new int[]{2, 4, 4}, new
int[]{6, 1, 2});
+ return partitions;
+ }
+
+ @Override
+ public PartitionReaderFactory createReaderFactory() {
+ return new SpecificReaderFactory();
+ }
+
+ @Override
+ public Partitioning outputPartitioning() {
+ return this.partitioning;
+ }
+
+ @Override
+ public SortOrder[] outputOrdering() {
+ return this.ordering;
+ }
+ }
+
+ @Override
+ public Table getTable(CaseInsensitiveStringMap options) {
+ return new JavaSimpleBatchTable() {
+ @Override
+ public Transform[] partitioning() {
+ return new Transform[] { Expressions.identity("i") };
Review Comment:
This may need to be updated since column "i" may not be in the
"partitionKeys".
##########
sql/core/src/test/java/test/org/apache/spark/sql/connector/JavaOrderAndPartitionAwareDataSource.java:
##########
@@ -0,0 +1,141 @@
+/*
+ * 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 test.org.apache.spark.sql.connector;
+
+import org.apache.spark.sql.connector.catalog.Table;
+import org.apache.spark.sql.connector.expressions.*;
+import org.apache.spark.sql.connector.read.*;
+import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning;
+import org.apache.spark.sql.connector.read.partitioning.Partitioning;
+import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning;
+import org.apache.spark.sql.util.CaseInsensitiveStringMap;
+
+public class JavaOrderAndPartitionAwareDataSource extends
JavaPartitionAwareDataSource {
+
+ static class MyScanBuilder extends JavaSimpleScanBuilder
+ implements SupportsReportPartitioning, SupportsReportOrdering {
+
+ private final Partitioning partitioning;
+ private final SortOrder[] ordering;
+
+ MyScanBuilder(String partitionKeys, String orderKeys) {
+ if (partitionKeys != null) {
+ String[] keys = partitionKeys.split(",");
+ Expression[] clustering = new Transform[keys.length];
+ for (int i=0; i< keys.length; i++) {
+ clustering[i] = Expressions.identity(keys[i]);
+ }
+ this.partitioning = new KeyGroupedPartitioning(clustering, 2);
+ } else {
+ this.partitioning = new UnknownPartitioning(2);
+ }
+
+ if (orderKeys != null) {
+ String[] keys = orderKeys.split(",");
+ this.ordering = new SortOrder[keys.length];
+ for (int i=0; i<keys.length; i++) {
+ this.ordering[i] = new MySortOrder(keys[i]);
+ }
+ } else {
+ this.ordering = new SortOrder[0];
+ }
+ }
+
+ @Override
+ public InputPartition[] planInputPartitions() {
+ InputPartition[] partitions = new InputPartition[2];
+ partitions[0] = new SpecificInputPartition(new int[]{1, 1, 3}, new
int[]{4, 5, 5});
+ partitions[1] = new SpecificInputPartition(new int[]{2, 4, 4}, new
int[]{6, 1, 2});
+ return partitions;
+ }
+
+ @Override
+ public PartitionReaderFactory createReaderFactory() {
+ return new SpecificReaderFactory();
+ }
+
+ @Override
+ public Partitioning outputPartitioning() {
+ return this.partitioning;
+ }
+
+ @Override
+ public SortOrder[] outputOrdering() {
+ return this.ordering;
+ }
+ }
+
+ @Override
+ public Table getTable(CaseInsensitiveStringMap options) {
+ return new JavaSimpleBatchTable() {
+ @Override
+ public Transform[] partitioning() {
+ return new Transform[] { Expressions.identity("i") };
+ }
+
+ @Override
+ public ScanBuilder newScanBuilder(CaseInsensitiveStringMap options) {
+ return new MyScanBuilder(options.get("partitionKeys"),
options.get("orderKeys"));
+ }
+ };
+ }
+
+ static class MySortOrder implements SortOrder {
+ private final Expression expression;
+
+ MySortOrder(String columnName) {
+ this.expression = new MyIdentityTransform(new
MyNamedReference(columnName));
Review Comment:
I don't think you need to define `MyIdentityTransform`. You can just use:
```scala
this.expression = new IdentityTransform(FieldReference.column(columnName));
```
##########
sql/core/src/test/java/test/org/apache/spark/sql/connector/JavaOrderAndPartitionAwareDataSource.java:
##########
@@ -0,0 +1,141 @@
+/*
+ * 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 test.org.apache.spark.sql.connector;
+
+import org.apache.spark.sql.connector.catalog.Table;
+import org.apache.spark.sql.connector.expressions.*;
+import org.apache.spark.sql.connector.read.*;
+import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning;
+import org.apache.spark.sql.connector.read.partitioning.Partitioning;
+import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning;
+import org.apache.spark.sql.util.CaseInsensitiveStringMap;
+
+public class JavaOrderAndPartitionAwareDataSource extends
JavaPartitionAwareDataSource {
+
+ static class MyScanBuilder extends JavaSimpleScanBuilder
+ implements SupportsReportPartitioning, SupportsReportOrdering {
+
+ private final Partitioning partitioning;
+ private final SortOrder[] ordering;
+
+ MyScanBuilder(String partitionKeys, String orderKeys) {
+ if (partitionKeys != null) {
+ String[] keys = partitionKeys.split(",");
+ Expression[] clustering = new Transform[keys.length];
+ for (int i=0; i< keys.length; i++) {
+ clustering[i] = Expressions.identity(keys[i]);
+ }
+ this.partitioning = new KeyGroupedPartitioning(clustering, 2);
+ } else {
+ this.partitioning = new UnknownPartitioning(2);
+ }
+
+ if (orderKeys != null) {
+ String[] keys = orderKeys.split(",");
+ this.ordering = new SortOrder[keys.length];
+ for (int i=0; i<keys.length; i++) {
+ this.ordering[i] = new MySortOrder(keys[i]);
+ }
+ } else {
+ this.ordering = new SortOrder[0];
+ }
+ }
+
+ @Override
+ public InputPartition[] planInputPartitions() {
+ InputPartition[] partitions = new InputPartition[2];
+ partitions[0] = new SpecificInputPartition(new int[]{1, 1, 3}, new
int[]{4, 5, 5});
+ partitions[1] = new SpecificInputPartition(new int[]{2, 4, 4}, new
int[]{6, 1, 2});
+ return partitions;
+ }
+
+ @Override
+ public PartitionReaderFactory createReaderFactory() {
+ return new SpecificReaderFactory();
+ }
+
+ @Override
+ public Partitioning outputPartitioning() {
+ return this.partitioning;
+ }
+
+ @Override
+ public SortOrder[] outputOrdering() {
+ return this.ordering;
+ }
+ }
+
+ @Override
+ public Table getTable(CaseInsensitiveStringMap options) {
+ return new JavaSimpleBatchTable() {
+ @Override
+ public Transform[] partitioning() {
+ return new Transform[] { Expressions.identity("i") };
+ }
+
+ @Override
+ public ScanBuilder newScanBuilder(CaseInsensitiveStringMap options) {
+ return new MyScanBuilder(options.get("partitionKeys"),
options.get("orderKeys"));
+ }
+ };
+ }
+
+ static class MySortOrder implements SortOrder {
Review Comment:
Similarly you can just use:
```scala
SortValue.apply(transform, SortDirection.ASCENDING,
NullOrdering.NULLS_FIRST);
```
instead of defining `MySortOrder`.
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/DataSourceV2ScanExecBase.scala:
##########
@@ -138,6 +138,15 @@ trait DataSourceV2ScanExecBase extends LeafExecNode {
}
}
+ override def outputOrdering: Seq[SortOrder] = scan match {
+ case s: SupportsReportOrdering if this.logicalLink.isDefined &&
Review Comment:
I'm not sure if `logicalLink` is the idiomatic way of processing a logical
plan here (don't see it used much elsewhere) - normally we do in an optimizer
rule like `V2ScanPartitioning` or `V2Writes`.
@cloud-fan @viirya do you see any issue with this?
##########
sql/core/src/test/java/test/org/apache/spark/sql/connector/JavaOrderAndPartitionAwareDataSource.java:
##########
@@ -0,0 +1,141 @@
+/*
+ * 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 test.org.apache.spark.sql.connector;
+
+import org.apache.spark.sql.connector.catalog.Table;
+import org.apache.spark.sql.connector.expressions.*;
+import org.apache.spark.sql.connector.read.*;
+import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning;
+import org.apache.spark.sql.connector.read.partitioning.Partitioning;
+import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning;
+import org.apache.spark.sql.util.CaseInsensitiveStringMap;
+
+public class JavaOrderAndPartitionAwareDataSource extends
JavaPartitionAwareDataSource {
+
+ static class MyScanBuilder extends JavaSimpleScanBuilder
+ implements SupportsReportPartitioning, SupportsReportOrdering {
+
+ private final Partitioning partitioning;
+ private final SortOrder[] ordering;
+
+ MyScanBuilder(String partitionKeys, String orderKeys) {
+ if (partitionKeys != null) {
+ String[] keys = partitionKeys.split(",");
+ Expression[] clustering = new Transform[keys.length];
+ for (int i=0; i< keys.length; i++) {
Review Comment:
nit: spaces
##########
sql/core/src/test/java/test/org/apache/spark/sql/connector/JavaOrderAndPartitionAwareDataSource.java:
##########
@@ -0,0 +1,141 @@
+/*
+ * 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 test.org.apache.spark.sql.connector;
+
+import org.apache.spark.sql.connector.catalog.Table;
+import org.apache.spark.sql.connector.expressions.*;
+import org.apache.spark.sql.connector.read.*;
+import org.apache.spark.sql.connector.read.partitioning.KeyGroupedPartitioning;
+import org.apache.spark.sql.connector.read.partitioning.Partitioning;
+import org.apache.spark.sql.connector.read.partitioning.UnknownPartitioning;
+import org.apache.spark.sql.util.CaseInsensitiveStringMap;
+
+public class JavaOrderAndPartitionAwareDataSource extends
JavaPartitionAwareDataSource {
+
+ static class MyScanBuilder extends JavaSimpleScanBuilder
+ implements SupportsReportPartitioning, SupportsReportOrdering {
+
+ private final Partitioning partitioning;
+ private final SortOrder[] ordering;
+
+ MyScanBuilder(String partitionKeys, String orderKeys) {
+ if (partitionKeys != null) {
+ String[] keys = partitionKeys.split(",");
+ Expression[] clustering = new Transform[keys.length];
+ for (int i=0; i< keys.length; i++) {
+ clustering[i] = Expressions.identity(keys[i]);
+ }
+ this.partitioning = new KeyGroupedPartitioning(clustering, 2);
+ } else {
+ this.partitioning = new UnknownPartitioning(2);
+ }
+
+ if (orderKeys != null) {
+ String[] keys = orderKeys.split(",");
+ this.ordering = new SortOrder[keys.length];
+ for (int i=0; i<keys.length; i++) {
+ this.ordering[i] = new MySortOrder(keys[i]);
+ }
+ } else {
+ this.ordering = new SortOrder[0];
+ }
+ }
+
+ @Override
+ public InputPartition[] planInputPartitions() {
Review Comment:
You can extend the same class from the parent class:
```scala
static class MyScanBuilder extends
JavaPartitionAwareDataSource.MyScanBuilder
implements SupportsReportOrdering
```
and then you don't need to redefine this and `createReaderFactory`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
