frankliee commented on code in PR #254:
URL: https://github.com/apache/incubator-uniffle/pull/254#discussion_r1005138924
##########
integration-test/spark-common/src/test/java/org/apache/uniffle/test/RepartitionWithLocalFileRssTest.java:
##########
@@ -53,4 +60,35 @@ public static void setupServers() throws Exception {
@Override
public void updateRssStorage(SparkConf sparkConf) {
}
+
+ /**
+ * Test different compression types with localfile rss mode.
+ * @throws Exception
+ */
+ @Override
+ public void run() throws Exception {
+ String fileName = generateTestFile();
+ SparkConf sparkConf = createSparkConf();
+ Uninterruptibles.sleepUninterruptibly(2, TimeUnit.SECONDS);
+
+ List<Map> results = new ArrayList<>();
+ Map resultWithoutRss = runSparkApp(sparkConf, fileName);
+ results.add(resultWithoutRss);
+
+ updateSparkConfWithRss(sparkConf);
+ updateSparkConfCustomer(sparkConf);
+ for (Codec.Type type :
+ new Codec.Type[]{
+ Codec.Type.NOOP,
+ Codec.Type.ZSTD,
+ Codec.Type.LZ4}) {
+ sparkConf.set("spark." + COMPRESSION_TYPE.key(), type.name());
Review Comment:
It may be better to add lowercase tests.
##########
common/src/main/java/org/apache/uniffle/common/config/RssClientConf.java:
##########
@@ -0,0 +1,38 @@
+/*
+ * 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.uniffle.common.config;
+
+import org.apache.uniffle.common.compression.Codec;
+
+import static org.apache.uniffle.common.compression.Codec.Type.LZ4;
+
+public class RssClientConf {
+
+ public static final ConfigOption<Codec.Type> COMPRESSION_TYPE = ConfigOptions
+ .key("rss.client.io.compression.codec")
+ .enumType(Codec.Type.class)
+ .defaultValue(LZ4)
Review Comment:
It is better to use lowercase as the default string like
`spark.io.compression.codec.`
And this value should be case insensitive.
##########
common/src/main/java/org/apache/uniffle/common/compression/ZstdCodec.java:
##########
@@ -0,0 +1,67 @@
+/*
+ * 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.uniffle.common.compression;
+
+import java.nio.ByteBuffer;
+
+import com.github.luben.zstd.Zstd;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import org.apache.uniffle.common.exception.RssException;
+
+public class ZstdCodec extends Codec {
+ private static final Logger LOGGER =
LoggerFactory.getLogger(ZstdCodec.class);
+
+ private final int compressionLevel;
+
+ public ZstdCodec(int level) {
+ this.compressionLevel = level;
+ LOGGER.info("Initializing zstd compressor.");
+ }
+
+ @Override
+ public void decompress(ByteBuffer src, int uncompressedLen, ByteBuffer dst,
int dstOffset) {
+ if (src.isDirect() && dst.isDirect()) {
+ long size = Zstd.decompressDirectByteBuffer(
+ dst, dstOffset, uncompressedLen,
+ src, src.position(), src.limit() - src.position()
+ );
+ if (size != uncompressedLen) {
+ throw new RssException(
+ "This should not happen that the decompressed data size is not
equals to original size.");
+ }
+ return;
+ }
+
+ if (!src.isDirect() && !dst.isDirect()) {
+ Zstd.decompressByteArray(
+ dst.array(), dstOffset, uncompressedLen,
+ src.array(), src.position(), src.limit() - src.position()
+ );
+ return;
+ }
+
+ throw new RssException("Zstd only supports the same type of bytebuffer
decompression.");
Review Comment:
It may be better to use IllegalStateException when one is direct but another
is not direct.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
