[SPARK-42040][SQL] SPJ: Introduce a new API for V2 input partition to report partition statistics

sql/catalyst/src/main/java/org/apache/spark/sql/connector/read/HasPartitionSize.java

@@ -0,0 +1,36 @@
+/*
+ * 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.spark.sql.connector.read;
+
+import java.util.OptionalLong;
+
+/**
+ * A mix-in for input partitions whose records are clustered on the same set of partition keys
+ * (provided via {@link SupportsReportPartitioning}, see below). Data sources can opt-in to
+ * implement this interface for the partitions they report to Spark, which will use the info
+ * to decide whether partition grouping should be applied or not.
+ *
+ * @see org.apache.spark.sql.connector.read.SupportsReportPartitioning
+ * @since 4.0.0
+ */
+public interface HasPartitionSize extends InputPartition {
+  /**
+   * Returns the value of the partition size associated to this partition.
+   */
+  OptionalLong partitionSizeInBytes();
+}

sql/catalyst/src/test/scala/org/apache/spark/sql/connector/catalog/InMemoryBaseTable.scala

@@ -605,7 +605,7 @@ object InMemoryBaseTable {
 }
 
 class BufferedRows(val key: Seq[Any] = Seq.empty) extends WriterCommitMessage
-    with InputPartition with HasPartitionKey with Serializable {
+    with InputPartition with HasPartitionKey with HasPartitionSize with Serializable {
   val rows = new mutable.ArrayBuffer[InternalRow]()
   val deletes = new mutable.ArrayBuffer[Int]()
 
@@ -617,7 +617,7 @@ class BufferedRows(val key: Seq[Any] = Seq.empty) extends WriterCommitMessage
   def keyString(): String = key.toArray.mkString("/")
 
   override def partitionKey(): InternalRow = PartitionInternalRow(key.toArray)
-
+  override def partitionSizeInBytes(): OptionalLong = OptionalLong.of(100L)
   def clear(): Unit = rows.clear()
 }
 

sql/core/src/test/scala/org/apache/spark/sql/connector/DataSourceV2SQLSuite.scala

@@ -2928,6 +2928,39 @@ class DataSourceV2SQLSuiteV1Filter
     }
   }
 
+  test("Check HasPartitionSize from InMemoryPartitionTable") {
+    val t = "testpart.tbl"
+    withTable(t) {
+      sql(s"CREATE TABLE $t (id string) USING foo PARTITIONED BY (key int)")
+      val table = catalog("testpart").asTableCatalog
+        .loadTable(Identifier.of(Array(), "tbl"))
+        .asInstanceOf[InMemoryPartitionTable]
+
+      var partSizes = table.data.map(_.partitionSizeInBytes().getAsLong)
+      assert(partSizes.length == 0)
+
+      sql(s"INSERT INTO $t VALUES ('a', 1), ('b', 2), ('c', 3)")
+      partSizes = table.data.map(_.partitionSizeInBytes().getAsLong)
+      assert(partSizes.length == 3)
+      assert(partSizes.toSet == Set(100, 100, 100))
+
+      sql(s"ALTER TABLE $t DROP PARTITION (key=3)")
+      partSizes = table.data.map(_.partitionSizeInBytes().getAsLong)
+      assert(partSizes.length == 2)
+      assert(partSizes.toSet == Set(100, 100))
+
+      sql(s"ALTER TABLE $t ADD PARTITION (key=4)")
+      partSizes = table.data.map(_.partitionSizeInBytes().getAsLong)
+      assert(partSizes.length == 3)
+      assert(partSizes.toSet == Set(100, 100, 100))
+
+      sql(s"INSERT INTO $t VALUES ('c', 3), ('e', 5)")
+      partSizes = table.data.map(_.partitionSizeInBytes().getAsLong)
+      assert(partSizes.length == 5)
+      assert(partSizes.toSet == Set(100, 100, 100, 100, 100))
+    }
+  }
+
   test("time travel") {
     sql("use testcat")
     // The testing in-memory table simply append the version/timestamp to the table name when