echauchot commented on code in PR #22446:
URL: https://github.com/apache/beam/pull/22446#discussion_r976313556
##########
runners/spark/3/src/main/java/org/apache/beam/runners/spark/structuredstreaming/translation/batch/GroupByKeyTranslatorBatch.java:
##########
@@ -17,74 +17,264 @@
*/
package org.apache.beam.runners.spark.structuredstreaming.translation.batch;
+import static
org.apache.beam.repackaged.core.org.apache.commons.lang3.ArrayUtils.EMPTY_BYTE_ARRAY;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.CoderHelpers.toByteArray;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.EncoderHelpers.collectionEncoder;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.EncoderHelpers.encoderOf;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.EncoderHelpers.kvEncoder;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.EncoderHelpers.windowedValueEncoder;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.concat;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.fun1;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.fun2;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.javaIterator;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.listOf;
+import static
org.apache.beam.runners.spark.structuredstreaming.translation.utils.ScalaInterop.seqOf;
+import static org.apache.beam.sdk.transforms.windowing.PaneInfo.NO_FIRING;
+import static
org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Preconditions.checkState;
+import static org.apache.spark.sql.functions.col;
+import static org.apache.spark.sql.functions.collect_list;
+import static org.apache.spark.sql.functions.explode;
+import static org.apache.spark.sql.functions.max;
+import static org.apache.spark.sql.functions.min;
+import static org.apache.spark.sql.functions.struct;
+
import java.io.Serializable;
import org.apache.beam.runners.core.InMemoryStateInternals;
-import org.apache.beam.runners.core.StateInternals;
+import org.apache.beam.runners.core.ReduceFnRunner;
import org.apache.beam.runners.core.StateInternalsFactory;
import org.apache.beam.runners.core.SystemReduceFn;
-import
org.apache.beam.runners.spark.structuredstreaming.translation.AbstractTranslationContext;
-import
org.apache.beam.runners.spark.structuredstreaming.translation.TransformTranslator;
import
org.apache.beam.runners.spark.structuredstreaming.translation.batch.functions.GroupAlsoByWindowViaOutputBufferFn;
-import
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.EncoderHelpers;
-import
org.apache.beam.runners.spark.structuredstreaming.translation.helpers.KVHelpers;
-import org.apache.beam.sdk.coders.Coder;
-import org.apache.beam.sdk.coders.IterableCoder;
import org.apache.beam.sdk.coders.KvCoder;
-import org.apache.beam.sdk.transforms.PTransform;
+import org.apache.beam.sdk.transforms.GroupByKey;
+import org.apache.beam.sdk.transforms.windowing.BoundedWindow;
+import org.apache.beam.sdk.transforms.windowing.GlobalWindow;
+import org.apache.beam.sdk.transforms.windowing.PaneInfo.PaneInfoCoder;
+import org.apache.beam.sdk.transforms.windowing.TimestampCombiner;
import org.apache.beam.sdk.util.WindowedValue;
import org.apache.beam.sdk.values.KV;
-import org.apache.beam.sdk.values.PCollection;
import org.apache.beam.sdk.values.WindowingStrategy;
+import org.apache.spark.sql.Column;
import org.apache.spark.sql.Dataset;
-import org.apache.spark.sql.KeyValueGroupedDataset;
+import org.apache.spark.sql.Encoder;
+import org.apache.spark.sql.TypedColumn;
+import org.apache.spark.sql.catalyst.expressions.CreateArray;
+import org.apache.spark.sql.catalyst.expressions.CreateNamedStruct;
+import org.apache.spark.sql.catalyst.expressions.Expression;
+import org.apache.spark.sql.catalyst.expressions.Literal;
+import org.apache.spark.sql.catalyst.expressions.Literal$;
+import org.apache.spark.sql.types.DataType;
+import org.apache.spark.sql.types.DataTypes;
+import org.checkerframework.checker.nullness.qual.NonNull;
+import scala.Tuple2;
+import scala.collection.Iterator;
+import scala.collection.Seq;
+import scala.collection.immutable.List;
+/**
+ * Translator for {@link GroupByKey} using {@link Dataset#groupByKey} with the
build-in aggregation
+ * function {@code collect_list} when applicable.
+ *
+ * <p>Note: Using {@code collect_list} isn't any worse than using {@link
ReduceFnRunner}. In the
+ * latter case the entire group (iterator) has to be loaded into memory as
well, risking OOM errors
+ * in both cases. When disabling {@link #useCollectList}, a more memory
sensitive iterable is used
Review Comment:
yes that is what I meant. I took a look at your rephrasing in round 2 it is
a lot clearer, thanks.
--
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]
