Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/6444#discussion_r34118539
--- Diff:
core/src/main/java/org/apache/spark/util/collection/unsafe/sort/UnsafeSorterSpillReader.java
---
@@ -0,0 +1,94 @@
+/*
+ * 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.util.collection.unsafe.sort;
+
+import java.io.*;
+
+import com.google.common.io.ByteStreams;
+
+import org.apache.spark.storage.BlockId;
+import org.apache.spark.storage.BlockManager;
+import org.apache.spark.unsafe.PlatformDependent;
+
+/**
+ * Reads spill files written by {@link UnsafeSorterSpillWriter} (see that
class for a description
+ * of the file format).
+ */
+final class UnsafeSorterSpillReader extends UnsafeSorterIterator {
+
+ private InputStream in;
+ private DataInputStream din;
+
+ // Variables that change with every record read:
+ private int recordLength;
+ private long keyPrefix;
+ private int numRecordsRemaining;
+
+ private final byte[] arr = new byte[1024 * 1024]; // TODO: tune this
(maybe grow dynamically)?
+ private final Object baseObject = arr;
+ private final long baseOffset = PlatformDependent.BYTE_ARRAY_OFFSET;
+
+ public UnsafeSorterSpillReader(
+ BlockManager blockManager,
+ File file,
+ BlockId blockId) throws IOException {
+ assert (file.length() > 0);
+ final BufferedInputStream bs = new BufferedInputStream(new
FileInputStream(file));
+ this.in = blockManager.wrapForCompression(blockId, bs);
+ this.din = new DataInputStream(this.in);
+ numRecordsRemaining = din.readInt();
+ }
+
+ @Override
+ public boolean hasNext() {
+ return (numRecordsRemaining > 0);
+ }
+
+ @Override
+ public void loadNext() throws IOException {
+ recordLength = din.readInt();
+ keyPrefix = din.readLong();
+ ByteStreams.readFully(in, arr, 0, recordLength);
+ numRecordsRemaining--;
+ if (numRecordsRemaining == 0) {
+ in.close();
--- End diff --
This reminds me: this kind of cleanup is actually a problem in general for
this sort code, since we also need to free memory pages, etc. in the `order by
xxx limit xxx` case. Until we implement a proper internal iterator interface,
a stopgap measure might be to define our own Scala iterator subclass which
implements a close method and to update operators like `Limit` to call
`close()` if it's available.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
