Skip to content
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

[CDAP-20657] fixes & additional support #15405

Merged
merged 1 commit into from
Nov 4, 2023
Merged

[CDAP-20657] fixes & additional support #15405

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
10 changes: 5 additions & 5 deletions ...rator-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/DelegatingSparkCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ public SparkCollection<RecordInfo<Object>> transform(
@Override
public SparkCollection<RecordInfo<Object>> multiOutputTransform(StageSpec stageSpec,
StageStatisticsCollector collector) {
return getDelegate().transform(stageSpec, collector);
return getDelegate().multiOutputTransform(stageSpec, collector);
}

@Override
Expand Down Expand Up @@ -94,8 +94,8 @@ public SparkCollection union(SparkCollection other) {

@Override
public Runnable createStoreTask(StageSpec stageSpec,
SparkSink sink) throws Exception {
return getDelegate().createStoreTask(stageSpec, sink);
SparkSink sink) {
return () -> getDelegate().createStoreTask(stageSpec, sink).run();
}

@Override
Expand Down Expand Up @@ -135,13 +135,13 @@ public <U> SparkCollection<U> compute(StageSpec stageSpec, SparkCompute<T, U> co
@Override
public Runnable createStoreTask(StageSpec stageSpec,
PairFlatMapFunction<T, Object, Object> sinkFunction) {
return getDelegate().createStoreTask(stageSpec, sinkFunction);
return () -> getDelegate().createStoreTask(stageSpec, sinkFunction).run();
}

@Override
public Runnable createMultiStoreTask(PhaseSpec phaseSpec, Set<String> group, Set<String> sinks,
Map<String, StageStatisticsCollector> collectors) {
return getDelegate().createMultiStoreTask(phaseSpec, group, sinks, collectors);
return () -> getDelegate().createMultiStoreTask(phaseSpec, group, sinks, collectors).run();
}

@Override
Expand Down
4 changes: 3 additions & 1 deletion ...ap-etl/hydrator-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/SparkCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,8 @@
import io.cdap.cdap.etl.common.RecordInfo;
import io.cdap.cdap.etl.common.StageStatisticsCollector;
import io.cdap.cdap.etl.proto.v2.spec.StageSpec;
import io.cdap.cdap.etl.spark.function.PluginFunctionContext;
import io.cdap.cdap.etl.spark.function.TransformFunction;
import io.cdap.cdap.etl.spark.join.JoinExpressionRequest;
import io.cdap.cdap.etl.spark.join.JoinRequest;
import org.apache.spark.api.java.function.FlatMapFunction;
Expand Down Expand Up @@ -81,7 +83,7 @@ SparkCollection<RecordInfo<Object>> reduceAggregate(StageSpec stageSpec, @Nullab
Runnable createMultiStoreTask(PhaseSpec phaseSpec, Set<String> group, Set<String> sinks,
Map<String, StageStatisticsCollector> collectors);

Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink) throws Exception;
Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink);

void publishAlerts(StageSpec stageSpec, StageStatisticsCollector collector) throws Exception;

Expand Down
23 changes: 17 additions & 6 deletions ...ydrator-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/batch/DatasetCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,25 +17,21 @@
package io.cdap.cdap.etl.spark.batch;

import io.cdap.cdap.api.data.DatasetContext;
import io.cdap.cdap.api.data.format.StructuredRecord;
import io.cdap.cdap.api.data.schema.Schema;
import io.cdap.cdap.api.spark.JavaSparkExecutionContext;
import io.cdap.cdap.etl.api.batch.SparkCompute;
import io.cdap.cdap.etl.api.batch.SparkExecutionPluginContext;
import io.cdap.cdap.etl.common.Constants;
import io.cdap.cdap.etl.common.PipelineRuntime;
import io.cdap.cdap.etl.common.RecordInfo;
import io.cdap.cdap.etl.common.StageStatisticsCollector;
import io.cdap.cdap.etl.proto.v2.spec.StageSpec;
import io.cdap.cdap.etl.spark.DelegatingSparkCollection;
import io.cdap.cdap.etl.spark.SparkCollection;
import io.cdap.cdap.etl.spark.SparkPipelineRuntime;
import io.cdap.cdap.etl.spark.function.DatasetAggregationAccumulator;
import io.cdap.cdap.etl.spark.function.DatasetAggregationFinalizeFunction;
import io.cdap.cdap.etl.spark.function.DatasetAggregationGetKeyFunction;
import io.cdap.cdap.etl.spark.function.DatasetAggregationReduceFunction;
import io.cdap.cdap.etl.spark.function.FunctionCache;
import io.cdap.cdap.etl.spark.function.MultiOutputTransformFunction;
import io.cdap.cdap.etl.spark.function.PluginFunctionContext;
import io.cdap.cdap.etl.spark.function.TransformFunction;
import javax.annotation.Nullable;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaRDD;
Expand Down Expand Up @@ -176,6 +172,21 @@ public SparkCollection<RecordInfo<Object>> reduceAggregate(StageSpec stageSpec,
return reduceDatasetAggregate(stageSpec, partitions, collector);
}

@Override
public SparkCollection<RecordInfo<Object>> transform(StageSpec stageSpec, StageStatisticsCollector collector) {
PluginFunctionContext pluginFunctionContext = new PluginFunctionContext(stageSpec, sec, collector);
return flatMap(stageSpec, new TransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache()));
}

@Override
public SparkCollection<RecordInfo<Object>> multiOutputTransform(StageSpec stageSpec,
StageStatisticsCollector collector) {
PluginFunctionContext pluginFunctionContext = new PluginFunctionContext(stageSpec, sec, collector);
return flatMap(stageSpec,new MultiOutputTransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache()));
}

/**
* Performs reduce aggregate using Dataset API. This allows SPARK to perform various optimizations that
* are not available when working on the RDD level.
Expand Down
10 changes: 5 additions & 5 deletions ...tl/hydrator-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/batch/RDDCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,16 +144,16 @@ public SparkCollection<T> union(SparkCollection<T> other) {
@Override
public SparkCollection<RecordInfo<Object>> transform(StageSpec stageSpec, StageStatisticsCollector collector) {
PluginFunctionContext pluginFunctionContext = new PluginFunctionContext(stageSpec, sec, collector);
return wrap(rdd.flatMap(new TransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache())));
return flatMap(stageSpec, new TransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache()));
}

@Override
public SparkCollection<RecordInfo<Object>> multiOutputTransform(StageSpec stageSpec,
StageStatisticsCollector collector) {
PluginFunctionContext pluginFunctionContext = new PluginFunctionContext(stageSpec, sec, collector);
return wrap(rdd.flatMap(new MultiOutputTransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache())));
return flatMap(stageSpec,new MultiOutputTransformFunction<T>(
pluginFunctionContext, functionCacheFactory.newCache()));
}

@Override
Expand Down Expand Up @@ -305,7 +305,7 @@ private void recordLineage(String name) {
}

@Override
public Runnable createStoreTask(final StageSpec stageSpec, final SparkSink<T> sink) throws Exception {
public Runnable createStoreTask(final StageSpec stageSpec, final SparkSink<T> sink) {
return new Runnable() {
@Override
public void run() {
Expand Down
2 changes: 1 addition & 1 deletion ...rator-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/batch/SQLEngineCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -290,7 +290,7 @@ public Runnable createMultiStoreTask(PhaseSpec phaseSpec, Set<String> group, Set
}

@Override
public Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink) throws Exception {
public Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink) {
return () -> {
try {
pull().createStoreTask(stageSpec, sink).run();
Expand Down
2 changes: 1 addition & 1 deletion ...tor-spark-core-base/src/main/java/io/cdap/cdap/etl/spark/streaming/DStreamCollection.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -205,7 +205,7 @@ public void run() {
}

@Override
public Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink) throws Exception {
public Runnable createStoreTask(StageSpec stageSpec, SparkSink<T> sink) {
return new Runnable() {
@Override
public void run() {
Expand Down