Use StandardCharsets.UTF_8 for converting String to bytes (apache#5372)

Use StandardCharsets.UTF_8 for converting String to bytes to avoid potential encoding error.
jbampton · Nov 6, 2019 · 0053dbc · 0053dbc
1 parent e0ea6b5
commit 0053dbc
Show file tree

Hide file tree

Showing 12 changed files with 29 additions and 23 deletions.
diff --git a/.../src/main/java/org/apache/spark/streaming/receiver/example/ProducerSparkReceiverData.java b/.../src/main/java/org/apache/spark/streaming/receiver/example/ProducerSparkReceiverData.java
@@ -18,7 +18,8 @@
  */
 package org.apache.spark.streaming.receiver.example;
 
-import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
+
 import org.apache.pulsar.client.api.Producer;
 import org.apache.pulsar.client.api.PulsarClient;
 
@@ -44,7 +45,7 @@ public static void main(String[] args) throws Exception {
     try (PulsarClient client = PulsarClient.builder().serviceUrl(args[0]).build()) {
       try (Producer<byte[]> producer = client.newProducer().topic(args[1]).create()) {
         for (int i = 0; i < 100; i++) {
-          producer.send(("producer spark streaming msg").getBytes(Charset.forName("UTF-8")));
+          producer.send(("producer spark streaming msg").getBytes(StandardCharsets.UTF_8));
         }
       }
     }

diff --git a/...java/org/apache/spark/streaming/receiver/example/SparkStreamingPulsarReceiverExample.java b/...java/org/apache/spark/streaming/receiver/example/SparkStreamingPulsarReceiverExample.java
@@ -18,7 +18,7 @@
  */
 package org.apache.spark.streaming.receiver.example;
 
-import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 import java.util.Arrays;
 import java.util.HashSet;
 import java.util.List;
@@ -74,7 +74,7 @@ public static void main(String[] args) throws Exception {
 
     JavaReceiverInputDStream<byte[]> lineDStream = jsc.receiverStream(pulsarReceiver);
     JavaPairDStream<String, Integer> result = lineDStream.flatMap(x -> {
-        String line = new String(x, Charset.forName("UTF-8"));
+        String line = new String(x, StandardCharsets.UTF_8);
         List<String> list = Arrays.asList(line.split(" "));
         return list.iterator();
       })

diff --git a/pulsar-broker/src/main/java/org/apache/pulsar/utils/CmdUtility.java b/pulsar-broker/src/main/java/org/apache/pulsar/utils/CmdUtility.java
@@ -23,13 +23,14 @@
 import java.io.InputStreamReader;
 import java.io.Writer;
 import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
 public class CmdUtility {
     private static final Logger LOG = LoggerFactory.getLogger(CmdUtility.class);
-    private static final Charset UTF_8 = Charset.forName("UTF-8");
+    private static final Charset UTF_8 = StandardCharsets.UTF_8;
 
     /**
      * Executes the specified string command in a separate process. STDOUT and STDERR output will be buffered to the

diff --git a/pulsar-broker/src/test/java/org/apache/pulsar/utils/SimpleTextOutputStreamTest.java b/pulsar-broker/src/test/java/org/apache/pulsar/utils/SimpleTextOutputStreamTest.java
@@ -20,7 +20,7 @@
 
 import static org.testng.Assert.assertEquals;
 
-import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 
 import org.apache.pulsar.common.util.SimpleTextOutputStream;
 import org.testng.annotations.BeforeMethod;
@@ -113,7 +113,7 @@ public void testString() {
     }
 
     public String str() {
-        String s = buf.toString(Charset.forName("utf-8"));
+        String s = buf.toString(StandardCharsets.UTF_8);
         reset();
         return s;
     }

diff --git a/pulsar-broker/src/test/java/org/apache/pulsar/utils/StatsOutputStreamTest.java b/pulsar-broker/src/test/java/org/apache/pulsar/utils/StatsOutputStreamTest.java
@@ -23,7 +23,7 @@
 import io.netty.buffer.ByteBuf;
 import io.netty.buffer.Unpooled;
 
-import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 
 import org.testng.annotations.BeforeMethod;
 import org.testng.annotations.Test;
@@ -145,7 +145,7 @@ public void testNestedObjects() {
     }
 
     public String str() {
-        String s = buf.toString(Charset.forName("utf-8"));
+        String s = buf.toString(StandardCharsets.UTF_8);
         reset();
         return s;
     }

diff --git a/..._8/src/test/java/org/apache/pulsar/client/kafka/test/KafkaProducerSimpleConsumerTest.java b/..._8/src/test/java/org/apache/pulsar/client/kafka/test/KafkaProducerSimpleConsumerTest.java
@@ -23,6 +23,7 @@
 
 import java.io.Serializable;
 import java.nio.ByteBuffer;
+import java.nio.charset.StandardCharsets;
 import java.util.Collections;
 import java.util.List;
 import java.util.Map;
@@ -170,7 +171,7 @@ public void testPulsarKafkaProducerWithSerializer(int partitions) throws Excepti
 
             byte[] bytes = new byte[payload.limit()];
             payload.get(bytes);
-            received.add(new String(bytes, "UTF-8"));
+            received.add(new String(bytes, StandardCharsets.UTF_8));
         }
         lastOffset -= 1;
 

diff --git a/pulsar-client-tools-test/src/test/java/org/apache/pulsar/admin/cli/utils/IOUtilsTest.java b/pulsar-client-tools-test/src/test/java/org/apache/pulsar/admin/cli/utils/IOUtilsTest.java
@@ -23,6 +23,7 @@
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.PrintStream;
+import java.nio.charset.StandardCharsets;
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Executors;
 import java.util.concurrent.Future;
@@ -127,7 +128,7 @@ public void test6() {
             String data = "\n";
             ByteArrayOutputStream baos = new ByteArrayOutputStream();
             System.setOut(new PrintStream(baos));
-            System.setIn(new ByteArrayInputStream(data.getBytes("UTF-8")));
+            System.setIn(new ByteArrayInputStream(data.getBytes(StandardCharsets.UTF_8)));
             ExecutorService executor = Executors.newSingleThreadExecutor();
             @SuppressWarnings("unchecked")
             Future<Void> future = (Future<Void>) executor.submit(() -> {

diff --git a/pulsar-io/kinesis/src/main/java/org/apache/pulsar/io/kinesis/KinesisRecord.java b/pulsar-io/kinesis/src/main/java/org/apache/pulsar/io/kinesis/KinesisRecord.java
@@ -19,8 +19,8 @@
 package org.apache.pulsar.io.kinesis;
 
 import java.nio.charset.CharacterCodingException;
-import java.nio.charset.Charset;
 import java.nio.charset.CharsetDecoder;
+import java.nio.charset.StandardCharsets;
 import java.util.HashMap;
 import java.util.Map;
 import java.util.Optional;
@@ -36,7 +36,7 @@ public class KinesisRecord implements Record<byte[]> {
     public static final String PARTITION_KEY = "";
     public static final String SEQUENCE_NUMBER = "";
 
-    private static final CharsetDecoder decoder = Charset.forName("UTF-8").newDecoder();
+    private static final CharsetDecoder decoder = StandardCharsets.UTF_8.newDecoder();
     private final Optional<String> key;
     private final byte[] value;
     private final HashMap<String, String> userProperties = new HashMap<String, String> ();

diff --git a/pulsar-io/mongo/src/main/java/org/apache/pulsar/io/mongodb/MongoSink.java b/pulsar-io/mongo/src/main/java/org/apache/pulsar/io/mongodb/MongoSink.java
@@ -34,7 +34,7 @@
 import org.bson.Document;
 import org.bson.json.JsonParseException;
 
-import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 import java.util.ArrayList;
 import java.util.Iterator;
 import java.util.List;
@@ -105,7 +105,7 @@ public void open(Map<String, Object> config, SinkContext sinkContext) throws Exc
 
     @Override
     public void write(Record<byte[]> record) {
-        final String recordValue = new String(record.getValue(), Charset.forName("UTF-8"));
+        final String recordValue = new String(record.getValue(), StandardCharsets.UTF_8);
 
         if (log.isDebugEnabled()) {
             log.debug("Received record: " + recordValue);
@@ -143,7 +143,7 @@ private void flush() {
 
             try {
                 final byte[] docAsBytes = record.getValue();
-                final Document doc = Document.parse(new String(docAsBytes, Charset.forName("UTF-8")));
+                final Document doc = Document.parse(new String(docAsBytes, StandardCharsets.UTF_8));
                 docsToInsert.add(doc);
             }
             catch (JsonParseException | BSONException e) {

diff --git a/pulsar-io/mongo/src/main/java/org/apache/pulsar/io/mongodb/MongoSource.java b/pulsar-io/mongo/src/main/java/org/apache/pulsar/io/mongodb/MongoSource.java
@@ -35,7 +35,7 @@
 import org.reactivestreams.Subscriber;
 import org.reactivestreams.Subscription;
 
-import java.io.UnsupportedEncodingException;
+import java.nio.charset.StandardCharsets;
 import java.util.HashMap;
 import java.util.Map;
 import java.util.Optional;
@@ -132,9 +132,9 @@ public void onNext(ChangeStreamDocument<Document> doc) {
 
                     consume(new DocRecord(
                             Optional.of(doc.getDocumentKey().toJson()),
-                            mapper.writeValueAsString(recordValue).getBytes("UTF-8")));
+                            mapper.writeValueAsString(recordValue).getBytes(StandardCharsets.UTF_8)));
 
-                } catch (UnsupportedEncodingException | JsonProcessingException e) {
+                } catch (JsonProcessingException e) {
                     log.error("Processing doc from mongo", e);
                 }
             }

diff --git a/...zookeeper-utils/src/main/java/org/apache/pulsar/zookeeper/ZookeeperClientFactoryImpl.java b/...zookeeper-utils/src/main/java/org/apache/pulsar/zookeeper/ZookeeperClientFactoryImpl.java
@@ -22,6 +22,7 @@
 
 import java.io.IOException;
 import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 import java.util.concurrent.CompletableFuture;
 
 import org.apache.zookeeper.KeeperException;
@@ -31,7 +32,7 @@
 import org.slf4j.LoggerFactory;
 
 public class ZookeeperClientFactoryImpl implements ZooKeeperClientFactory {
-    public static final Charset ENCODING_SCHEME = Charset.forName("UTF-8");
+    public static final Charset ENCODING_SCHEME = StandardCharsets.UTF_8;
 
     @Override
     public CompletableFuture<ZooKeeper> create(String serverList, SessionType sessionType, int zkSessionTimeoutMillis) {

diff --git a/pulsar-zookeeper-utils/src/test/java/org/apache/pulsar/zookeeper/DeserializersTest.java b/pulsar-zookeeper-utils/src/test/java/org/apache/pulsar/zookeeper/DeserializersTest.java
@@ -21,11 +21,12 @@
 import static org.testng.Assert.assertEquals;
 
 import org.apache.pulsar.common.policies.data.Policies;
-import org.apache.pulsar.zookeeper.Deserializers;
 import org.testng.annotations.AfterMethod;
 import org.testng.annotations.BeforeMethod;
 import org.testng.annotations.Test;
 
+import java.nio.charset.StandardCharsets;
+
 @Test
 public class DeserializersTest {
 
@@ -40,7 +41,7 @@ void teardown() throws Exception {
     @Test
     void testSimpleStringDeserialize() throws Exception {
         String key = "test_key";
-        byte[] content = "test_content".getBytes("UTF-8");
+        byte[] content = "test_content".getBytes(StandardCharsets.UTF_8);
         String result = Deserializers.STRING_DESERIALIZER.deserialize(key, content);
         assertEquals(result, "test_content");
     }
@@ -50,7 +51,7 @@ void testSimplePolicyDeserialize() throws Exception {
         String key = "test_key";
         String jsonPolicy = "{\"auth_policies\":{\"namespace_auth\":{},\"destination_auth\":{}},\"replication_clusters\":[],"
                 + "\"bundles_activated\":true,\"backlog_quota_map\":{},\"persistence\":null,\"latency_stats_sample_rate\":{},\"message_ttl_in_seconds\":0}";
-        byte[] content = jsonPolicy.getBytes("UTF-8");
+        byte[] content = jsonPolicy.getBytes(StandardCharsets.UTF_8);
         Policies result = Deserializers.POLICY_DESERIALIZER.deserialize(key, content);
         assertEquals(result, new Policies());
     }