Github user sameeragarwal commented on a diff in the pull request:
https://github.com/apache/spark/pull/14174#discussion_r71253283
--- Diff:
sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/RowBasedKeyValueBatch.java
---
@@ -0,0 +1,322 @@
+/*
+ * 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.catalyst.expressions;
+
+import java.io.IOException;
+
+import org.apache.spark.memory.MemoryConsumer;
+import org.apache.spark.memory.TaskMemoryManager;
+import org.apache.spark.sql.types.*;
+import org.apache.spark.unsafe.memory.MemoryBlock;
+import org.apache.spark.unsafe.Platform;
+
+
+/**
+ * RowBasedKeyValueBatch stores key value pairs in contiguous memory
region.
+ *
+ * Each key or value is stored as a single UnsafeRow. The format for each
record looks like this:
+ * [4 bytes total size = (klen + vlen + 4)] [4 bytes key size = klen]
+ * [UnsafeRow for key of length klen] [UnsafeRow for Value of length vlen]
+ * [8 bytes pointer to next]
+ * Thus, record length = 8 + klen + vlen + 8
+ *
+ * RowBasedKeyValueBatch will automatically acquire new pages
(MemoryBlock) when the current page
+ * is used up.
+ *
+ * TODO: making each entry more compact, e.g., combine key and value into
a single UnsafeRow
+ */
+public final class RowBasedKeyValueBatch extends MemoryConsumer{
+ private static final int DEFAULT_CAPACITY = 1 << 16;
+ private static final long DEFAULT_PAGE_SIZE = 64 * 1024 * 1024;
+
+ private final StructType keySchema;
+ private final StructType valueSchema;
+ private final int capacity;
+ private int numRows = 0;
+
+ // Staging row returned from getRow.
+ final UnsafeRow keyRow;
+ final UnsafeRow valueRow;
+
+ // ids for current key row and value row being retrieved
+ private int keyRowId = -1;
+
+ // full addresses for key rows and value rows
+ private long[] keyOffsets;
+
+ // if all data types in the schema are fixed length
+ private boolean allFixedLength;
+ private int klen;
+ private int vlen;
+ private int recordLength;
+
+ private MemoryBlock currentAndOnlyPage = null;
+ private Object currentAndOnlyBase = null;
+ private long recordStartOffset;
+ private long pageCursor = 0;
+
+ public static RowBasedKeyValueBatch allocate(StructType keySchema,
StructType valueSchema,
+ TaskMemoryManager manager) {
+ return new RowBasedKeyValueBatch(keySchema, valueSchema,
DEFAULT_CAPACITY, manager);
+ }
+
+ public static RowBasedKeyValueBatch allocate(StructType keySchema,
StructType valueSchema,
+ TaskMemoryManager manager,
int maxRows) {
+ return new RowBasedKeyValueBatch(keySchema, valueSchema, maxRows,
manager);
+ }
+
+ public int numRows() { return numRows; }
+
+ public void close() {
+ if (currentAndOnlyPage != null) {
+ freePage(currentAndOnlyPage);
+ currentAndOnlyPage = null;
+ }
+ }
+
+ private boolean acquireNewPage(long required) {
+ try {
+ currentAndOnlyPage = allocatePage(required);
+ } catch (OutOfMemoryError e) {
+ return false;
+ }
+ currentAndOnlyBase = currentAndOnlyPage.getBaseObject();
+ Platform.putInt(currentAndOnlyBase,
currentAndOnlyPage.getBaseOffset(), 0);
+ pageCursor = 4;
+ recordStartOffset = pageCursor + currentAndOnlyPage.getBaseOffset();
+
+ return true;
+ }
+
+ private long getKeyOffsetForFixedLengthRecords(int rowId) {
+ return recordStartOffset + rowId * recordLength + 8;
--- End diff --
Let's explicitly cast `recordLength` as long to prevent inadvertant
overflows while multiplying 2 integers
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
