feat: add missing Hive namespace ops and use deterministic OpenDAL OS classifiers

docs/mkdocs.yml

@@ -121,7 +121,6 @@ nav:
               - CreateTable: spec/impls/rest/create-table.md
           - Directory: spec/impls/dir.md
           - Apache Hive MetaStore: spec/impls/hive.md
-          - Apache Iceberg REST Catalog: spec/impls/iceberg.md
       - Tool Integration: spec/tools.md
       - OpenAPI: https://editor-next.swagger.io/?url=https://raw.githubusercontent.com/lancedb/lance-namespace/refs/heads/main/docs/src/spec/rest.yaml
   - User Guide:

docs/src/spec/impls/glue.md

@@ -3,6 +3,29 @@
 **Lance Glue Namespace** is an implementation using AWS Glue Data Catalog.
 For more details about AWS Glue, please read the [AWS Glue Data Catalog Documentation](https://docs.aws.amazon.com/glue/).
 
+## Configuration
+
+The Lance Glue namespace accepts the following configuration properties:
+
+| Property            | Required | Description                                                     | Default                   | Example                         |
+|---------------------|----------|-----------------------------------------------------------------|---------------------------|---------------------------------|
+| `catalog_id`        | No       | The Catalog ID of the Glue catalog (defaults to AWS account ID) |                           | `123456789012`                  |
+| `endpoint`          | No       | Custom Glue service endpoint for API compatible metastores      |                           | `https://glue.example.com`      |
+| `region`            | No       | AWS region for all Glue operations                              |                           | `us-west-2`                     |
+| `access_key_id`     | No       | AWS access key ID for static credentials                        |                           |                                 |
+| `secret_access_key` | No       | AWS secret access key for static credentials                    |                           |                                 |
+| `session_token`     | No       | AWS session token for temporary credentials                     |                           |                                 |
+| `root`              | No       | Storage root location of the lakehouse on Glue catalog          | Current working directory | `/my/dir`, `s3://bucket/prefix` |
+| `storage.*`         | No       | Additional storage configurations to access table               |                           | `storage.region=us-west-2`      |
+
+### Authentication
+
+The Glue namespace supports multiple authentication methods:
+
+1. **Default AWS credential provider chain**: When no explicit credentials are provided, the client uses the default AWS credential provider chain
+2. **Static credentials**: Set `access_key_id` and `secret_access_key` for basic AWS credentials
+3. **Session credentials**: Additionally provide `session_token` for temporary AWS credentials
+
 ## Namespace Mapping
 
 An AWS Glue Data Catalog can be viewed as the root Lance namespace.

docs/src/spec/impls/hive.md

@@ -2,7 +2,23 @@
 
 **Lance Hive Namespace** is an implementation using Apache Hive MetaStore (HMS).
 For more details about HMS, please read [HMS AdminManual 2.x](https://hive.apache.org/docs/latest/adminmanual-metastore-administration_27362076/) 
-and [HMS AdminManual 3.x](https://hive.apache.org/docs/latest/adminmanual-metastore-3-0-administration_75978150/). 
+and [HMS AdminManual 3.x](https://hive.apache.org/docs/latest/adminmanual-metastore-3-0-administration_75978150/).
+
+## Configuration
+
+The Lance Hive namespace accepts the following configuration properties:
+
+| Property           | Required | Description                                            | Default                   | Example                         |
+|--------------------|----------|--------------------------------------------------------|---------------------------|---------------------------------|
+| `client.pool-size` | No       | Size of the HMS client connection pool                 | 3                         | `5`                             |
+| `root`             | No       | Storage root location of the lakehouse on Hive catalog | Current working directory | `/my/dir`, `s3://bucket/prefix` |
+| `storage.*`        | No       | Additional storage configurations to access table      |                           | `storage.region=us-west-2`      |
+
+### Version-specific Configuration
+
+**Hive 2.x**: Uses the configuration properties listed above.
+
+**Hive 3.x**: Uses the same configuration properties as Hive 2.x, with an additional catalog name constant (`hive`) for catalog-level operations. 
 
 ## Namespace Mapping
 

java/Makefile

@@ -10,7 +10,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-VERSION = 0.0.5
+VERSION = 0.0.6
 
 .PHONY: clean-apache-client
 clean-apache-client:

java/lance-namespace-adapter/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lance-namespace-root</artifactId>
-        <version>0.0.5</version>
+        <version>0.0.6</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/lance-namespace-apache-client/README.md

@@ -51,7 +51,7 @@ Add this dependency to your project's POM:
 <dependency>
   <groupId>com.lancedb</groupId>
   <artifactId>lance-namespace-apache-client</artifactId>
-  <version>0.0.5</version>
+  <version>0.0.6</version>
   <scope>compile</scope>
 </dependency>
 ```
@@ -61,7 +61,7 @@ Add this dependency to your project's POM:
 Add this dependency to your project's build file:
 
 ```groovy
-compile "com.lancedb:lance-namespace-apache-client:0.0.5"
+compile "com.lancedb:lance-namespace-apache-client:0.0.6"
 ```
 
 ### Others
@@ -74,7 +74,7 @@ mvn clean package
 
 Then manually install the following JARs:
 
-- `target/lance-namespace-apache-client-0.0.5.jar`
+- `target/lance-namespace-apache-client-0.0.6.jar`
 - `target/lib/*.jar`
 
 ## Getting Started

java/lance-namespace-apache-client/pom.xml

@@ -5,7 +5,7 @@
     <artifactId>lance-namespace-apache-client</artifactId>
     <packaging>jar</packaging>
     <name>lance-namespace-apache-client</name>
-    <version>0.0.5</version>
+    <version>0.0.6</version>
     <url>https://github.com/openapitools/openapi-generator</url>
     <description>OpenAPI Java</description>
     <scm>
@@ -16,7 +16,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lance-namespace-root</artifactId>
-        <version>0.0.5</version>
+        <version>0.0.6</version>
     </parent>
 
     <licenses>

java/lance-namespace-apache-client/src/main/java/com/lancedb/lance/namespace/client/apache/ApiClient.java

@@ -157,7 +157,7 @@ public ApiClient(CloseableHttpClient httpClient) {
     dateFormat = ApiClient.buildDefaultDateFormat();
 
     // Set default User-Agent.
-    setUserAgent("OpenAPI-Generator/0.0.5/java");
+    setUserAgent("OpenAPI-Generator/0.0.6/java");
 
     // Setup authentications (key: authentication name, value: authentication).
     authentications = new HashMap<String, Authentication>();

java/lance-namespace-apache-client/src/main/java/com/lancedb/lance/namespace/client/apache/Configuration.java

@@ -17,7 +17,7 @@
     value = "org.openapitools.codegen.languages.JavaClientCodegen",
     comments = "Generator version: 7.12.0")
 public class Configuration {
-  public static final String VERSION = "0.0.5";
+  public static final String VERSION = "0.0.6";
 
   private static volatile ApiClient defaultApiClient = new ApiClient();
 

java/lance-namespace-core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lance-namespace-root</artifactId>
-        <version>0.0.5</version>
+        <version>0.0.6</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 
@@ -37,7 +37,19 @@
             <groupId>org.apache.opendal</groupId>
             <artifactId>opendal</artifactId>
             <version>0.48.0</version>
-            <classifier>${os.detected.classifier}</classifier>
+            <classifier>linux-aarch_64</classifier>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.opendal</groupId>
+            <artifactId>opendal</artifactId>
+            <version>0.48.0</version>
+            <classifier>linux-x86_64</classifier>
+        </dependency>
+        <dependency>
+            <groupId>org.apache.opendal</groupId>
+            <artifactId>opendal</artifactId>
+            <version>0.48.0</version>
+            <classifier>osx-aarch_64</classifier>
         </dependency>
         <dependency>
             <groupId>org.apache.arrow</groupId>
@@ -103,13 +115,6 @@
                 </executions>
             </plugin>
         </plugins>
-        <extensions>
-            <extension>
-                <groupId>kr.motd.maven</groupId>
-                <artifactId>os-maven-plugin</artifactId>
-                <version>1.7.0</version>
-            </extension>
-        </extensions>
     </build>
     <profiles>
     </profiles>

java/lance-namespace-core/src/main/java/com/lancedb/lance/namespace/LanceNamespaceException.java

@@ -54,7 +54,7 @@ public LanceNamespaceException(ApiException e) {
   }
 
   public LanceNamespaceException(ErrorResponse errorResponse) {
-    super(errorResponse.getError());
+    super(errorResponse.toString());
     this.code = errorResponse.getCode() == null ? UNKNOWN_ERROR_CODE : errorResponse.getCode();
     this.errorResponse = Optional.of(errorResponse);
   }

java/lance-namespace-core/src/main/java/com/lancedb/lance/namespace/dir/DirectoryNamespace.java

@@ -34,6 +34,7 @@
 import com.lancedb.lance.namespace.model.ListTablesRequest;
 import com.lancedb.lance.namespace.model.ListTablesResponse;
 import com.lancedb.lance.namespace.model.NamespaceExistsRequest;
+import com.lancedb.lance.namespace.model.TableExistsRequest;
 import com.lancedb.lance.namespace.util.JsonArrowSchemaConverter;
 import com.lancedb.lance.namespace.util.OpenDalUtil;
 import com.lancedb.lance.namespace.util.ValidationUtil;
@@ -64,7 +65,8 @@ public class DirectoryNamespace implements LanceNamespace, Closeable {
   public void initialize(Map<String, String> configProperties, BufferAllocator allocator) {
     this.config = new DirectoryNamespaceConfig(configProperties);
     this.allocator = allocator;
-    this.operator = OpenDalUtil.initializeOperator(this.config.getRoot());
+    this.operator =
+        OpenDalUtil.initializeOperator(this.config.getRoot(), this.config.getStorageOptions());
   }
 
   @Override
@@ -123,6 +125,7 @@ public CreateTableResponse createTable(CreateTableRequest request, byte[] reques
     CreateTableResponse response = new CreateTableResponse();
     response.setLocation(tablePath);
     response.setVersion(1L);
+    response.setStorageOptions(config.getStorageOptions());
     return response;
   }
 
@@ -180,6 +183,24 @@ public ListTablesResponse listTables(ListTablesRequest request) {
     return response;
   }
 
+  @Override
+  public void tableExists(TableExistsRequest request) {
+    String tableName = tableNameFromId(request.getId());
+
+    LOG.debug("Checking if table {} exists", tableName);
+
+    String versionsPath = tableVersionsPath(tableName);
+    List<Entry> versionEntries =
+        operator.list(versionsPath, ListOptions.builder().limit(1).build());
+    if (versionEntries.isEmpty()) {
+      throw LanceNamespaceException.notFound(
+          "Table does not exist: " + tableName,
+          "TABLE_NOT_FOUND",
+          tableName,
+          "The requested table was not found in the namespace");
+    }
+  }
+
   @Override
   public DescribeTableResponse describeTable(DescribeTableRequest request) {
     String tableName = tableNameFromId(request.getId());
@@ -208,6 +229,7 @@ public DescribeTableResponse describeTable(DescribeTableRequest request) {
 
     DescribeTableResponse response = new DescribeTableResponse();
     response.setLocation(tablePath);
+    response.setStorageOptions(config.getStorageOptions());
     return response;
   }
 

java/lance-namespace-core/src/main/java/com/lancedb/lance/namespace/dir/DirectoryNamespaceConfig.java

@@ -13,6 +13,7 @@
  */
 package com.lancedb.lance.namespace.dir;
 
+import com.lancedb.lance.namespace.util.OpenDalUtil;
 import com.lancedb.lance.namespace.util.PropertyUtil;
 
 import java.io.Serializable;
@@ -29,7 +30,9 @@ public class DirectoryNamespaceConfig implements Serializable {
   private final Map<String, String> storageOptions;
 
   public DirectoryNamespaceConfig(Map<String, String> properties) {
-    this.root = PropertyUtil.propertyAsString(properties, ROOT, ROOT_DEFAULT);
+    this.root =
+        OpenDalUtil.stripTrailingSlash(
+            PropertyUtil.propertyAsString(properties, ROOT, ROOT_DEFAULT));
     this.storageOptions = PropertyUtil.propertiesWithPrefix(properties, STORAGE_OPTIONS_PREFIX);
   }
 

java/lance-namespace-core/src/main/java/com/lancedb/lance/namespace/util/OpenDalUtil.java

@@ -13,9 +13,11 @@
  */
 package com.lancedb.lance.namespace.util;
 
-import com.google.common.collect.ImmutableMap;
 import org.apache.opendal.Operator;
 
+import java.util.HashMap;
+import java.util.Map;
+
 /** Utility class for OpenDAL operations and configurations. */
 public class OpenDalUtil {
 
@@ -24,33 +26,53 @@ private OpenDalUtil() {
   }
 
   /**
-   * Initialize an OpenDAL operator based on the root path.
+   * Initialize an OpenDAL operator based on the root path and storage options.
    *
    * @param root the root path/URI
+   * @param storageOptions additional storage configuration options (credentials, endpoints, etc.)
    * @return configured OpenDAL operator
    */
-  public static Operator initializeOperator(String root) {
+  public static Operator initializeOperator(String root, Map<String, String> storageOptions) {
     String[] schemeSplit = root.split("://", -1);
 
     // local file system path
     if (schemeSplit.length < 2) {
-      return Operator.of("fs", ImmutableMap.of("root", root));
+      Map<String, String> config = new HashMap<>();
+      config.put("root", root);
+      if (storageOptions != null) {
+        config.putAll(storageOptions);
+      }
+      return Operator.of("fs", config);
     }
 
     String scheme = normalizeScheme(schemeSplit[0]);
     String[] authoritySplit = schemeSplit[1].split("/", 2);
     String authority = authoritySplit[0];
     String path = authoritySplit.length > 1 ? authoritySplit[1] : "";
 
+    Map<String, String> config = new HashMap<>();
+
     switch (scheme) {
       case "s3":
       case "gcs":
-        return Operator.of(scheme, ImmutableMap.of("root", path, "bucket", authority));
+        config.put("root", path);
+        config.put("bucket", authority);
+        break;
       case "azblob":
-        return Operator.of(scheme, ImmutableMap.of("root", path, "CONTAINER", authority));
+        config.put("root", path);
+        config.put("CONTAINER", authority);
+        break;
       default:
-        return Operator.of(scheme, ImmutableMap.of("root", schemeSplit[1]));
+        config.put("root", schemeSplit[1]);
+        break;
     }
+
+    // Add storage options
+    if (storageOptions != null) {
+      config.putAll(storageOptions);
+    }
+
+    return Operator.of(scheme, config);
   }
 
   /**