HBASE-25770 Http InfoServers should honor gzip encoding when requested (#3159)

Signed-off-by: Duo Zhang <zhangduo@apache.org>
Signed-off-by: Josh Elser <elserj@apache.org>

hbase-http/pom.xml

@@ -220,6 +220,11 @@
       <artifactId>junit</artifactId>
       <scope>test</scope>
     </dependency>
+    <dependency>
+      <groupId>org.hamcrest</groupId>
+      <artifactId>hamcrest-library</artifactId>
+      <scope>test</scope>
+    </dependency>
     <dependency>
       <groupId>org.mockito</groupId>
       <artifactId>mockito-core</artifactId>

hbase-http/src/main/java/org/apache/hadoop/hbase/http/HttpServer.java

@@ -66,7 +66,6 @@ import org.apache.yetus.audience.InterfaceAudience;
 import org.apache.yetus.audience.InterfaceStability;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
-
 import org.apache.hbase.thirdparty.com.google.common.base.Preconditions;
 import org.apache.hbase.thirdparty.com.google.common.collect.Lists;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.http.HttpVersion;
@@ -81,6 +80,7 @@ import org.apache.hbase.thirdparty.org.eclipse.jetty.server.SslConnectionFactory
 import org.apache.hbase.thirdparty.org.eclipse.jetty.server.handler.ContextHandlerCollection;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.server.handler.HandlerCollection;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.server.handler.RequestLogHandler;
+import org.apache.hbase.thirdparty.org.eclipse.jetty.server.handler.gzip.GzipHandler;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.servlet.DefaultServlet;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.servlet.FilterHolder;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.servlet.FilterMapping;
@@ -575,6 +575,7 @@ public class HttpServer implements FilterContainer {
     this.findPort = b.findPort;
     this.authenticationEnabled = b.securityEnabled;
     initializeWebServer(b.name, b.hostName, b.conf, b.pathSpecs, b);
+    this.webServer.setHandler(buildGzipHandler(this.webServer.getHandler()));
   }
 
   private void initializeWebServer(String name, String hostName,
@@ -662,6 +663,23 @@ public class HttpServer implements FilterContainer {
     return ctx;
   }
 
+  /**
+   * Construct and configure an instance of {@link GzipHandler}. With complex
+   * multi-{@link WebAppContext} configurations, it's easiest to apply this handler directly to the
+   * instance of {@link Server} near the end of its configuration, something like
+   * <pre>
+   *    Server server = new Server();
+   *    //...
+   *    server.setHandler(buildGzipHandler(server.getHandler()));
+   *    server.start();
+   * </pre>
+   */
+  public static GzipHandler buildGzipHandler(final Handler wrapped) {
+    final GzipHandler gzipHandler = new GzipHandler();
+    gzipHandler.setHandler(wrapped);
+    return gzipHandler;
+  }
+
   private static void addNoCacheFilter(WebAppContext ctxt) {
     defineFilter(ctxt, NO_CACHE_FILTER, NoCacheFilter.class.getName(),
         Collections.<String, String> emptyMap(), new String[] { "/*" });

hbase-http/src/test/java/org/apache/hadoop/hbase/http/TestHttpServer.java

@@ -1,4 +1,4 @@
-/**
+/*
  * 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
@@ -17,11 +17,16 @@
  */
 package org.apache.hadoop.hbase.http;
 
+import static org.hamcrest.Matchers.greaterThan;
+import java.io.BufferedReader;
 import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
 import java.io.PrintWriter;
 import java.net.HttpURLConnection;
 import java.net.URI;
 import java.net.URL;
+import java.nio.CharBuffer;
 import java.util.Arrays;
 import java.util.Collections;
 import java.util.Enumeration;
@@ -56,6 +61,13 @@ import org.apache.hadoop.security.Groups;
 import org.apache.hadoop.security.ShellBasedUnixGroupsMapping;
 import org.apache.hadoop.security.UserGroupInformation;
 import org.apache.hadoop.security.authorize.AccessControlList;
+import org.apache.http.HttpEntity;
+import org.apache.http.HttpHeaders;
+import org.apache.http.client.methods.CloseableHttpResponse;
+import org.apache.http.client.methods.HttpGet;
+import org.apache.http.impl.client.CloseableHttpClient;
+import org.apache.http.impl.client.HttpClients;
+import org.hamcrest.MatcherAssert;
 import org.junit.AfterClass;
 import org.junit.Assert;
 import org.junit.BeforeClass;
@@ -66,7 +78,6 @@ import org.junit.experimental.categories.Category;
 import org.mockito.Mockito;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
-
 import org.apache.hbase.thirdparty.org.eclipse.jetty.server.ServerConnector;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.util.ajax.JSON;
 
@@ -269,6 +280,60 @@ public class TestHttpServer extends HttpServerFunctionalTest {
     // assertEquals("text/html; charset=utf-8", conn.getContentType());
   }
 
+  @Test
+  public void testNegotiatesEncodingGzip() throws IOException {
+    final InputStream stream = ClassLoader.getSystemResourceAsStream("webapps/static/test.css");
+    assertNotNull(stream);
+    final String sourceContent = readFully(stream);
+
+    try (final CloseableHttpClient client = HttpClients.createMinimal()) {
+      final HttpGet request = new HttpGet(new URL(baseUrl, "/static/test.css").toString());
+
+      request.setHeader(HttpHeaders.ACCEPT_ENCODING, null);
+      final long unencodedContentLength;
+      try (final CloseableHttpResponse response = client.execute(request)) {
+        final HttpEntity entity = response.getEntity();
+        assertNotNull(entity);
+        assertNull(entity.getContentEncoding());
+        unencodedContentLength = entity.getContentLength();
+        MatcherAssert.assertThat(unencodedContentLength, greaterThan(0L));
+        final String unencodedEntityBody = readFully(entity.getContent());
+        assertEquals(sourceContent, unencodedEntityBody);
+      }
+
+      request.setHeader(HttpHeaders.ACCEPT_ENCODING, "gzip");
+      final long encodedContentLength;
+      try (final CloseableHttpResponse response = client.execute(request)) {
+        final HttpEntity entity = response.getEntity();
+        assertNotNull(entity);
+        assertNotNull(entity.getContentEncoding());
+        assertEquals("gzip", entity.getContentEncoding().getValue());
+        encodedContentLength = entity.getContentLength();
+        MatcherAssert.assertThat(encodedContentLength, greaterThan(0L));
+        final String encodedEntityBody = readFully(entity.getContent());
+        // the encoding/decoding process, as implemented in this specific combination of dependency
+        // versions, does not perfectly preserve trailing whitespace. thus, `trim()`.
+        assertEquals(sourceContent.trim(), encodedEntityBody.trim());
+      }
+      MatcherAssert.assertThat(unencodedContentLength, greaterThan(encodedContentLength));
+    }
+  }
+
+  private static String readFully(final InputStream input) throws IOException {
+    // TODO: when the time comes, delete me and replace with a JDK11 IO helper API.
+    try (final BufferedReader reader = new BufferedReader(new InputStreamReader(input))) {
+      final StringBuilder sb = new StringBuilder();
+      final CharBuffer buffer = CharBuffer.allocate(1024 * 2);
+      while (reader.read(buffer) > 0) {
+        sb.append(buffer);
+        buffer.clear();
+      }
+      return sb.toString();
+    } finally {
+      input.close();
+    }
+  }
+
   /**
    * Dummy filter that mimics as an authentication filter. Obtains user identity
    * from the request parameter user.name. Wraps around the request so that

hbase-server/src/main/java/org/apache/hadoop/hbase/master/HMaster.java

@@ -21,7 +21,6 @@ import static org.apache.hadoop.hbase.HConstants.DEFAULT_HBASE_SPLIT_COORDINATED
 import static org.apache.hadoop.hbase.HConstants.HBASE_MASTER_LOGCLEANER_PLUGINS;
 import static org.apache.hadoop.hbase.HConstants.HBASE_SPLIT_WAL_COORDINATED_BY_ZK;
 import static org.apache.hadoop.hbase.util.DNS.MASTER_HOSTNAME_KEY;
-
 import java.io.IOException;
 import java.io.InterruptedIOException;
 import java.lang.reflect.Constructor;
@@ -88,9 +87,9 @@ import org.apache.hadoop.hbase.client.TableDescriptorBuilder;
 import org.apache.hadoop.hbase.client.TableState;
 import org.apache.hadoop.hbase.coprocessor.CoprocessorHost;
 import org.apache.hadoop.hbase.exceptions.MasterStoppedException;
-import org.apache.hadoop.hbase.executor.ExecutorService.ExecutorConfig;
 import org.apache.hadoop.hbase.executor.ExecutorType;
 import org.apache.hadoop.hbase.favored.FavoredNodesManager;
+import org.apache.hadoop.hbase.http.HttpServer;
 import org.apache.hadoop.hbase.ipc.CoprocessorRpcUtils;
 import org.apache.hadoop.hbase.ipc.RpcServer;
 import org.apache.hadoop.hbase.ipc.ServerNotRunningYetException;
@@ -216,7 +215,6 @@ import org.apache.yetus.audience.InterfaceAudience;
 import org.apache.zookeeper.KeeperException;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
-
 import org.apache.hbase.thirdparty.com.google.common.collect.Lists;
 import org.apache.hbase.thirdparty.com.google.common.collect.Maps;
 import org.apache.hbase.thirdparty.com.google.common.collect.Sets;
@@ -227,7 +225,6 @@ import org.apache.hbase.thirdparty.org.eclipse.jetty.server.Server;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.server.ServerConnector;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.servlet.ServletHolder;
 import org.apache.hbase.thirdparty.org.eclipse.jetty.webapp.WebAppContext;
-
 import org.apache.hadoop.hbase.shaded.protobuf.RequestConverter;
 import org.apache.hadoop.hbase.shaded.protobuf.generated.AdminProtos.GetRegionInfoResponse;
 import org.apache.hadoop.hbase.shaded.protobuf.generated.SnapshotProtos.SnapshotDescription;
@@ -552,7 +549,8 @@ public class HMaster extends HRegionServer implements MasterServices {
     if (infoPort < 0 || infoServer == null) {
       return -1;
     }
-    if(infoPort == infoServer.getPort()) {
+    if (infoPort == infoServer.getPort()) {
+      // server is already running
       return infoPort;
     }
     final String addr = conf.get("hbase.master.info.bindAddress", "0.0.0.0");
@@ -574,6 +572,7 @@ public class HMaster extends HRegionServer implements MasterServices {
     connector.setPort(infoPort);
     masterJettyServer.addConnector(connector);
     masterJettyServer.setStopAtShutdown(true);
+    masterJettyServer.setHandler(HttpServer.buildGzipHandler(masterJettyServer.getHandler()));
 
     final String redirectHostname =
         StringUtils.isBlank(useThisHostnameInstead) ? null : useThisHostnameInstead;