HADOOP-19681: Fix S3A failing to initialize S3 buckets having namespace with dot followed by number

Author: shameersss1

hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/AbstractFileSystem.java

@@ -335,7 +335,7 @@ private URI getUri(URI uri, String supportedScheme,
     int port = uri.getPort();
     port = (port == -1 ? defaultPort : port);
     if (port == -1) { // no port supplied and default port is not specified
-      return new URI(supportedScheme, authority, "/", null);
+      return URI.create(supportedScheme + "://" + authority);
     }
     return new URI(supportedScheme + "://" + uri.getHost() + ":" + port);
   }

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AFileSystem.java

@@ -573,8 +573,11 @@ private static void addDeprecatedKeys() {
    */
   public void initialize(URI name, Configuration originalConf)
       throws IOException {
-    // get the host; this is guaranteed to be non-null, non-empty
+    // get the host; fallback to authority if getHost() returns null
     bucket = name.getHost();
+    if (bucket == null) {
+      bucket = name.getAuthority();
+    }
     AuditSpan span = null;
     // track initialization duration; will only be set after
     // statistics are set up.

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/tools/BucketTool.java

@@ -172,7 +172,7 @@ public int run(final String[] args, final PrintStream out)
     final String bucketPath = parsedArgs.get(0);
     final Path source = new Path(bucketPath);
     URI fsURI = source.toUri();
-    String bucket = fsURI.getHost();
+    String bucket = fsURI.getHost() != null ? fsURI.getHost() : fsURI.getAuthority();
 
     println(out, "Filesystem %s", fsURI);
     if (!"s3a".equals(fsURI.getScheme())) {

hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3native/S3xLoginHelper.java

@@ -56,8 +56,12 @@ public static URI buildFSURI(URI uri) {
     // look for login secrets and fail if they are present.
     Objects.requireNonNull(uri, "null uri");
     Objects.requireNonNull(uri.getScheme(), "null uri.getScheme()");
-    Objects.requireNonNull(uri.getHost(), "null uri host.");
-    return URI.create(uri.getScheme() + "://" + uri.getHost());
+    String host = uri.getHost();
+    if (host == null) {
+      host = uri.getAuthority();
+    }
+    Objects.requireNonNull(host, "null uri host.");
+    return URI.create(uri.getScheme() + "://" + host);
   }
 
   /**

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/ITestS3AConfiguration.java

@@ -48,6 +48,7 @@
 import org.apache.commons.lang3.reflect.FieldUtils;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.FileStatus;
+import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.LocalDirAllocator;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.fs.contract.ContractTestUtils;
@@ -594,13 +595,28 @@ private static <T> T getField(Object target, Class<T> fieldType,
   public void testConfOptionPropagationToFS() throws Exception {
     Configuration config = new Configuration();
     String testFSName = config.getTrimmed(TEST_FS_S3A_NAME, "");
-    String bucket = new URI(testFSName).getHost();
+    URI uri = new URI(testFSName);
+    String bucket = uri.getHost();
+    if (bucket == null) {
+      bucket = uri.getAuthority();
+    }
     setBucketOption(config, bucket, "propagation", "propagated");
     fs = S3ATestUtils.createTestFileSystem(config);
     Configuration updated = fs.getConf();
     assertOptionEquals(updated, "fs.s3a.propagation", "propagated");
   }
 
+  @Test
+  public void testBucketNameWithDotAndNumber() throws Exception {
+    Configuration config = new Configuration();
+    Path path = new Path("s3a://test-bucket-v1.1");
+    try (FileSystem fs = path.getFileSystem(config)) {
+      assertThat(fs instanceof S3AFileSystem)
+          .describedAs("FileSystem should be S3AFileSystem instance")
+          .isTrue();
+    }
+  }
+
   @Test
   @Timeout(10)
   public void testS3SpecificSignerOverride() throws Exception {

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/S3ATestUtils.java

@@ -883,7 +883,11 @@ public static <T extends Writable> T roundTrip(
   public static String getTestBucketName(final Configuration conf) {
     String bucket = checkNotNull(conf.get(TEST_FS_S3A_NAME),
         "No test bucket");
-    return URI.create(bucket).getHost();
+    URI uri = URI.create(bucket);
+    if (uri.getHost() != null) {
+      return uri.getHost();
+    }
+    return uri.getAuthority();
   }
 
   /**

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/TestBucketConfiguration.java

@@ -29,6 +29,7 @@
 
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.s3a.auth.delegation.EncryptionSecrets;
+import org.apache.hadoop.fs.s3native.S3xLoginHelper;
 import org.apache.hadoop.security.ProviderUtils;
 import org.apache.hadoop.security.alias.CredentialProvider;
 import org.apache.hadoop.security.alias.CredentialProviderFactory;
@@ -75,6 +76,24 @@ public void setup() throws Exception {
     S3AFileSystem.initializeClass();
   }
 
+  @Test
+  public void testS3xLoginHelperWithDotInBucketName() throws Throwable {
+    // Test buildFSURI with bucket name containing dot followed by number
+    URI uri = URI.create("s3a://bucket-v1.1/path");
+    URI result = S3xLoginHelper.buildFSURI(uri);
+    assertEquals("s3a://bucket-v1.1", result.toString());
+
+    // Test with normal bucket name
+    URI normalUri = URI.create("s3a://normal-bucket/path");
+    URI normalResult = S3xLoginHelper.buildFSURI(normalUri);
+    assertEquals("s3a://normal-bucket", normalResult.toString());
+
+    // Test edge case with multiple dots
+    URI multiDotUri = URI.create("s3a://bucket.v1.2.test/path");
+    URI multiDotResult = S3xLoginHelper.buildFSURI(multiDotUri);
+    assertEquals("s3a://bucket.v1.2.test", multiDotResult.toString());
+  }
+
   @Test
   public void testBucketConfigurationPropagation() throws Throwable {
     Configuration config = new Configuration(false);

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/auth/ITestCustomSigner.java

@@ -283,6 +283,17 @@ public SdkHttpFullRequest sign(SdkHttpFullRequest request,
 
       String host = request.host();
       String bucketName = parseBucketFromHost(host);
+      // If host-based parsing fails (path-style requests), extract from path
+      if (bucketName.equals("s3")) {
+        String path = request.encodedPath();
+        if (path != null && path.startsWith("/") && path.length() > 1) {
+          String[] pathParts = path.substring(1).split("/", 2);
+          if (pathParts.length > 0 && !pathParts[0].isEmpty()) {
+            bucketName = pathParts[0];
+          }
+        }
+      }
+
       try {
         lastStoreValue = CustomSignerInitializer
             .getStoreValue(bucketName, UserGroupInformation.getCurrentUser());

hadoop-tools/hadoop-aws/src/test/java/org/apache/hadoop/fs/s3a/auth/ITestJceksIO.java

@@ -187,8 +187,12 @@ private String run(CredentialShell cs, String expected, String... args)
    */
   private String toJceksProvider(Path keystore) {
     final URI uri = keystore.toUri();
+    String bucket = uri.getHost();
+    if (bucket == null) {
+      bucket = uri.getAuthority();
+    }
     return String.format("jceks://%s@%s%s",
-        uri.getScheme(), uri.getHost(), uri.getPath());
+        uri.getScheme(), bucket, uri.getPath());
   }
 
 }