@lancedb/lancedb 0.4.15 → 0.4.18

package/dist/connection.d.ts

@@ -1,6 +1,19 @@
-import { Connection as LanceDbConnection } from "./native";
+import { ConnectionOptions, Connection as LanceDbConnection } from "./native";
 import { Table } from "./table";
 import { Table as ArrowTable, Schema } from "apache-arrow";
+/**
+ * Connect to a LanceDB instance at the given URI.
+ *
+ * Accepted formats:
+ *
+ * - `/path/to/database` - local database
+ * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
+ * - `db://host:port` - remote database (LanceDB cloud)
+ * @param {string} uri - The uri of the database. If the database uri starts
+ * with `db://` then it connects to a remote database.
+ * @see {@link ConnectionOptions} for more details on the URI format.
+ */
+export declare function connect(uri: string, opts?: Partial<ConnectionOptions>): Promise<Connection>;
 export interface CreateTableOptions {
     /**
      * The mode to use when creating the table.
@@ -17,6 +30,38 @@ export interface CreateTableOptions {
      * then no error will be raised.
      */
     existOk: boolean;
+    /**
+     * Configuration for object storage.
+     *
+     * Options already set on the connection will be inherited by the table,
+     * but can be overridden here.
+     *
+     * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+     */
+    storageOptions?: Record<string, string>;
+}
+export interface OpenTableOptions {
+    /**
+     * Configuration for object storage.
+     *
+     * Options already set on the connection will be inherited by the table,
+     * but can be overridden here.
+     *
+     * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+     */
+    storageOptions?: Record<string, string>;
+    /**
+     * Set the size of the index cache, specified as a number of entries
+     *
+     * The exact meaning of an "entry" will depend on the type of index:
+     * - IVF: there is one entry for each IVF partition
+     * - BTREE: there is one entry for the entire index
+     *
+     * This cache applies to the entire opened table, across all indices.
+     * Setting this value higher will increase performance on larger datasets
+     * at the expense of more RAM
+     */
+    indexCacheSize?: number;
 }
 export interface TableNamesOptions {
     /**
@@ -75,7 +120,7 @@ export declare class Connection {
      * Open a table in the database.
      * @param {string} name - The name of the table
      */
-    openTable(name: string): Promise<Table>;
+    openTable(name: string, options?: Partial<OpenTableOptions>): Promise<Table>;
     /**
      * Creates a new Table and initialize it with new data.
      * @param {string} name - The name of the table.

package/dist/connection.js

@@ -13,10 +13,30 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.Connection = void 0;
+exports.Connection = exports.connect = void 0;
 const arrow_1 = require("./arrow");
+const native_1 = require("./native");
 const table_1 = require("./table");
 const apache_arrow_1 = require("apache-arrow");
+/**
+ * Connect to a LanceDB instance at the given URI.
+ *
+ * Accepted formats:
+ *
+ * - `/path/to/database` - local database
+ * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
+ * - `db://host:port` - remote database (LanceDB cloud)
+ * @param {string} uri - The uri of the database. If the database uri starts
+ * with `db://` then it connects to a remote database.
+ * @see {@link ConnectionOptions} for more details on the URI format.
+ */
+async function connect(uri, opts) {
+    opts = opts ?? {};
+    opts.storageOptions = cleanseStorageOptions(opts.storageOptions);
+    const nativeConn = await native_1.Connection.new(uri, opts);
+    return new Connection(nativeConn);
+}
+exports.connect = connect;
 /**
  * A LanceDB Connection that allows you to open tables and create new ones.
  *
@@ -72,8 +92,8 @@ class Connection {
      * Open a table in the database.
      * @param {string} name - The name of the table
      */
-    async openTable(name) {
-        const innerTable = await this.inner.openTable(name);
+    async openTable(name, options) {
+        const innerTable = await this.inner.openTable(name, cleanseStorageOptions(options?.storageOptions), options?.indexCacheSize);
         return new table_1.Table(innerTable);
     }
     /**
@@ -96,7 +116,7 @@ class Connection {
             table = (0, arrow_1.makeArrowTable)(data);
         }
         const buf = await (0, arrow_1.fromTableToBuffer)(table);
-        const innerTable = await this.inner.createTable(name, buf, mode);
+        const innerTable = await this.inner.createTable(name, buf, mode, cleanseStorageOptions(options?.storageOptions));
         return new table_1.Table(innerTable);
     }
     /**
@@ -112,7 +132,7 @@ class Connection {
         }
         const table = (0, arrow_1.makeEmptyTable)(schema);
         const buf = await (0, arrow_1.fromTableToBuffer)(table);
-        const innerTable = await this.inner.createEmptyTable(name, buf, mode);
+        const innerTable = await this.inner.createEmptyTable(name, buf, mode, cleanseStorageOptions(options?.storageOptions));
         return new table_1.Table(innerTable);
     }
     /**
@@ -124,3 +144,38 @@ class Connection {
     }
 }
 exports.Connection = Connection;
+/**
+ * Takes storage options and makes all the keys snake case.
+ */
+function cleanseStorageOptions(options) {
+    if (options === undefined) {
+        return undefined;
+    }
+    const result = {};
+    for (const [key, value] of Object.entries(options)) {
+        if (value !== undefined) {
+            const newKey = camelToSnakeCase(key);
+            result[newKey] = value;
+        }
+    }
+    return result;
+}
+/**
+ * Convert a string to snake case. It might already be snake case, in which case it is
+ * returned unchanged.
+ */
+function camelToSnakeCase(camel) {
+    if (camel.includes("_")) {
+        // Assume if there is at least one underscore, it is already snake case
+        return camel;
+    }
+    if (camel.toLocaleUpperCase() === camel) {
+        // Assume if the string is all uppercase, it is already snake case
+        return camel;
+    }
+    let result = camel.replace(/[A-Z]/g, (letter) => `_${letter.toLowerCase()}`);
+    if (result.startsWith("_")) {
+        result = result.slice(1);
+    }
+    return result;
+}

package/dist/index.d.ts

@@ -1,22 +1,7 @@
-import { Connection } from "./connection";
-import { ConnectionOptions } from "./native.js";
 export { WriteOptions, WriteMode, AddColumnsSql, ColumnAlteration, ConnectionOptions, } from "./native.js";
 export { makeArrowTable, MakeArrowTableOptions, Data, VectorColumnOptions, } from "./arrow";
-export { Connection, CreateTableOptions, TableNamesOptions, } from "./connection";
+export { connect, Connection, CreateTableOptions, TableNamesOptions, } from "./connection";
 export { ExecutableQuery, Query, QueryBase, VectorQuery, RecordBatchIterator, } from "./query";
 export { Index, IndexOptions, IvfPqOptions } from "./indices";
 export { Table, AddDataOptions, IndexConfig, UpdateOptions } from "./table";
 export * as embedding from "./embedding";
-/**
- * Connect to a LanceDB instance at the given URI.
- *
- * Accpeted formats:
- *
- * - `/path/to/database` - local database
- * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
- * - `db://host:port` - remote database (LanceDB cloud)
- * @param {string} uri - The uri of the database. If the database uri starts
- * with `db://` then it connects to a remote database.
- * @see {@link ConnectionOptions} for more details on the URI format.
- */
-export declare function connect(uri: string, opts?: Partial<ConnectionOptions>): Promise<Connection>;

package/dist/index.js

@@ -13,15 +13,14 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.connect = exports.embedding = exports.Table = exports.Index = exports.RecordBatchIterator = exports.VectorQuery = exports.QueryBase = exports.Query = exports.Connection = exports.VectorColumnOptions = exports.MakeArrowTableOptions = exports.makeArrowTable = void 0;
-const connection_1 = require("./connection");
-const native_js_1 = require("./native.js");
+exports.embedding = exports.Table = exports.Index = exports.RecordBatchIterator = exports.VectorQuery = exports.QueryBase = exports.Query = exports.Connection = exports.connect = exports.VectorColumnOptions = exports.MakeArrowTableOptions = exports.makeArrowTable = void 0;
 var arrow_1 = require("./arrow");
 Object.defineProperty(exports, "makeArrowTable", { enumerable: true, get: function () { return arrow_1.makeArrowTable; } });
 Object.defineProperty(exports, "MakeArrowTableOptions", { enumerable: true, get: function () { return arrow_1.MakeArrowTableOptions; } });
 Object.defineProperty(exports, "VectorColumnOptions", { enumerable: true, get: function () { return arrow_1.VectorColumnOptions; } });
-var connection_2 = require("./connection");
-Object.defineProperty(exports, "Connection", { enumerable: true, get: function () { return connection_2.Connection; } });
+var connection_1 = require("./connection");
+Object.defineProperty(exports, "connect", { enumerable: true, get: function () { return connection_1.connect; } });
+Object.defineProperty(exports, "Connection", { enumerable: true, get: function () { return connection_1.Connection; } });
 var query_1 = require("./query");
 Object.defineProperty(exports, "Query", { enumerable: true, get: function () { return query_1.Query; } });
 Object.defineProperty(exports, "QueryBase", { enumerable: true, get: function () { return query_1.QueryBase; } });
@@ -32,21 +31,3 @@ Object.defineProperty(exports, "Index", { enumerable: true, get: function () { r
 var table_1 = require("./table");
 Object.defineProperty(exports, "Table", { enumerable: true, get: function () { return table_1.Table; } });
 exports.embedding = require("./embedding");
-/**
- * Connect to a LanceDB instance at the given URI.
- *
- * Accpeted formats:
- *
- * - `/path/to/database` - local database
- * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
- * - `db://host:port` - remote database (LanceDB cloud)
- * @param {string} uri - The uri of the database. If the database uri starts
- * with `db://` then it connects to a remote database.
- * @see {@link ConnectionOptions} for more details on the URI format.
- */
-async function connect(uri, opts) {
-    opts = opts ?? {};
-    const nativeConn = await native_js_1.Connection.new(uri, opts);
-    return new connection_1.Connection(nativeConn);
-}
-exports.connect = connect;

package/dist/native.d.ts

@@ -62,6 +62,12 @@ export interface ConnectionOptions {
    * always consistent.
    */
   readConsistencyInterval?: number
+  /**
+   * (For LanceDB OSS only): configuration for object storage.
+   *
+   * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+   */
+  storageOptions?: Record<string, string>
 }
 /** Write mode for writing a table. */
 export const enum WriteMode {
@@ -73,7 +79,9 @@ export const enum WriteMode {
 export interface WriteOptions {
   mode?: WriteMode
 }
-export function connect(uri: string, options: ConnectionOptions): Promise<Connection>
+export interface OpenTableOptions {
+  storageOptions?: Record<string, string>
+}
 export class Connection {
   /** Create a new Connection instance from the given URI. */
   static new(uri: string, options: ConnectionOptions): Promise<Connection>
@@ -90,9 +98,9 @@ export class Connection {
    * - buf: The buffer containing the IPC file.
    *
    */
-  createTable(name: string, buf: Buffer, mode: string): Promise<Table>
-  createEmptyTable(name: string, schemaBuf: Buffer, mode: string): Promise<Table>
-  openTable(name: string): Promise<Table>
+  createTable(name: string, buf: Buffer, mode: string, storageOptions?: Record<string, string> | undefined | null): Promise<Table>
+  createEmptyTable(name: string, schemaBuf: Buffer, mode: string, storageOptions?: Record<string, string> | undefined | null): Promise<Table>
+  openTable(name: string, storageOptions?: Record<string, string> | undefined | null, indexCacheSize?: number | undefined | null): Promise<Table>
   /** Drop table with the name. Or raise an error if the table does not exist. */
   dropTable(name: string): Promise<void>
 }

package/dist/native.js

@@ -303,7 +303,7 @@ if (!nativeBinding) {
     }
     throw new Error(`Failed to load native binding`);
 }
-const { Connection, Index, RecordBatchIterator, Query, VectorQuery, Table, WriteMode, connect } = nativeBinding;
+const { Connection, Index, RecordBatchIterator, Query, VectorQuery, Table, WriteMode } = nativeBinding;
 module.exports.Connection = Connection;
 module.exports.Index = Index;
 module.exports.RecordBatchIterator = RecordBatchIterator;
@@ -311,4 +311,3 @@ module.exports.Query = Query;
 module.exports.VectorQuery = VectorQuery;
 module.exports.Table = Table;
 module.exports.WriteMode = WriteMode;
-module.exports.connect = connect;

package/lancedb/connection.ts

@@ -13,10 +13,32 @@
 // limitations under the License.
 
 import { fromTableToBuffer, makeArrowTable, makeEmptyTable } from "./arrow";
-import { Connection as LanceDbConnection } from "./native";
+import { ConnectionOptions, Connection as LanceDbConnection } from "./native";
 import { Table } from "./table";
 import { Table as ArrowTable, Schema } from "apache-arrow";
 
+/**
+ * Connect to a LanceDB instance at the given URI.
+ *
+ * Accepted formats:
+ *
+ * - `/path/to/database` - local database
+ * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
+ * - `db://host:port` - remote database (LanceDB cloud)
+ * @param {string} uri - The uri of the database. If the database uri starts
+ * with `db://` then it connects to a remote database.
+ * @see {@link ConnectionOptions} for more details on the URI format.
+ */
+export async function connect(
+  uri: string,
+  opts?: Partial<ConnectionOptions>,
+): Promise<Connection> {
+  opts = opts ?? {};
+  opts.storageOptions = cleanseStorageOptions(opts.storageOptions);
+  const nativeConn = await LanceDbConnection.new(uri, opts);
+  return new Connection(nativeConn);
+}
+
 export interface CreateTableOptions {
   /**
    * The mode to use when creating the table.
@@ -33,6 +55,40 @@ export interface CreateTableOptions {
    * then no error will be raised.
    */
   existOk: boolean;
+
+  /**
+   * Configuration for object storage.
+   *
+   * Options already set on the connection will be inherited by the table,
+   * but can be overridden here.
+   *
+   * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+   */
+  storageOptions?: Record<string, string>;
+}
+
+export interface OpenTableOptions {
+  /**
+   * Configuration for object storage.
+   *
+   * Options already set on the connection will be inherited by the table,
+   * but can be overridden here.
+   *
+   * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+   */
+  storageOptions?: Record<string, string>;
+  /**
+   * Set the size of the index cache, specified as a number of entries
+   *
+   * The exact meaning of an "entry" will depend on the type of index:
+   * - IVF: there is one entry for each IVF partition
+   * - BTREE: there is one entry for the entire index
+   *
+   * This cache applies to the entire opened table, across all indices.
+   * Setting this value higher will increase performance on larger datasets
+   * at the expense of more RAM
+   */
+  indexCacheSize?: number;
 }
 
 export interface TableNamesOptions {
@@ -109,8 +165,15 @@ export class Connection {
    * Open a table in the database.
    * @param {string} name - The name of the table
    */
-  async openTable(name: string): Promise<Table> {
-    const innerTable = await this.inner.openTable(name);
+  async openTable(
+    name: string,
+    options?: Partial<OpenTableOptions>,
+  ): Promise<Table> {
+    const innerTable = await this.inner.openTable(
+      name,
+      cleanseStorageOptions(options?.storageOptions),
+      options?.indexCacheSize,
+    );
     return new Table(innerTable);
   }
 
@@ -139,7 +202,12 @@ export class Connection {
       table = makeArrowTable(data);
     }
     const buf = await fromTableToBuffer(table);
-    const innerTable = await this.inner.createTable(name, buf, mode);
+    const innerTable = await this.inner.createTable(
+      name,
+      buf,
+      mode,
+      cleanseStorageOptions(options?.storageOptions),
+    );
     return new Table(innerTable);
   }
 
@@ -162,7 +230,12 @@ export class Connection {
 
     const table = makeEmptyTable(schema);
     const buf = await fromTableToBuffer(table);
-    const innerTable = await this.inner.createEmptyTable(name, buf, mode);
+    const innerTable = await this.inner.createEmptyTable(
+      name,
+      buf,
+      mode,
+      cleanseStorageOptions(options?.storageOptions),
+    );
     return new Table(innerTable);
   }
 
@@ -174,3 +247,43 @@ export class Connection {
     return this.inner.dropTable(name);
   }
 }
+
+/**
+ * Takes storage options and makes all the keys snake case.
+ */
+function cleanseStorageOptions(
+  options?: Record<string, string>,
+): Record<string, string> | undefined {
+  if (options === undefined) {
+    return undefined;
+  }
+  const result: Record<string, string> = {};
+  for (const [key, value] of Object.entries(options)) {
+    if (value !== undefined) {
+      const newKey = camelToSnakeCase(key);
+      result[newKey] = value;
+    }
+  }
+  return result;
+}
+
+/**
+ * Convert a string to snake case. It might already be snake case, in which case it is
+ * returned unchanged.
+ */
+function camelToSnakeCase(camel: string): string {
+  if (camel.includes("_")) {
+    // Assume if there is at least one underscore, it is already snake case
+    return camel;
+  }
+  if (camel.toLocaleUpperCase() === camel) {
+    // Assume if the string is all uppercase, it is already snake case
+    return camel;
+  }
+
+  let result = camel.replace(/[A-Z]/g, (letter) => `_${letter.toLowerCase()}`);
+  if (result.startsWith("_")) {
+    result = result.slice(1);
+  }
+  return result;
+}

package/lancedb/index.ts

@@ -12,12 +12,6 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-import { Connection } from "./connection";
-import {
-  Connection as LanceDbConnection,
-  ConnectionOptions,
-} from "./native.js";
-
 export {
   WriteOptions,
   WriteMode,
@@ -32,6 +26,7 @@ export {
   VectorColumnOptions,
 } from "./arrow";
 export {
+  connect,
   Connection,
   CreateTableOptions,
   TableNamesOptions,
@@ -46,24 +41,3 @@ export {
 export { Index, IndexOptions, IvfPqOptions } from "./indices";
 export { Table, AddDataOptions, IndexConfig, UpdateOptions } from "./table";
 export * as embedding from "./embedding";
-
-/**
- * Connect to a LanceDB instance at the given URI.
- *
- * Accpeted formats:
- *
- * - `/path/to/database` - local database
- * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
- * - `db://host:port` - remote database (LanceDB cloud)
- * @param {string} uri - The uri of the database. If the database uri starts
- * with `db://` then it connects to a remote database.
- * @see {@link ConnectionOptions} for more details on the URI format.
- */
-export async function connect(
-  uri: string,
-  opts?: Partial<ConnectionOptions>,
-): Promise<Connection> {
-  opts = opts ?? {};
-  const nativeConn = await LanceDbConnection.new(uri, opts);
-  return new Connection(nativeConn);
-}

package/nodejs-artifacts/connection.d.ts

@@ -1,6 +1,19 @@
-import { Connection as LanceDbConnection } from "./native";
+import { ConnectionOptions, Connection as LanceDbConnection } from "./native";
 import { Table } from "./table";
 import { Table as ArrowTable, Schema } from "apache-arrow";
+/**
+ * Connect to a LanceDB instance at the given URI.
+ *
+ * Accepted formats:
+ *
+ * - `/path/to/database` - local database
+ * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
+ * - `db://host:port` - remote database (LanceDB cloud)
+ * @param {string} uri - The uri of the database. If the database uri starts
+ * with `db://` then it connects to a remote database.
+ * @see {@link ConnectionOptions} for more details on the URI format.
+ */
+export declare function connect(uri: string, opts?: Partial<ConnectionOptions>): Promise<Connection>;
 export interface CreateTableOptions {
     /**
      * The mode to use when creating the table.
@@ -17,6 +30,38 @@ export interface CreateTableOptions {
      * then no error will be raised.
      */
     existOk: boolean;
+    /**
+     * Configuration for object storage.
+     *
+     * Options already set on the connection will be inherited by the table,
+     * but can be overridden here.
+     *
+     * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+     */
+    storageOptions?: Record<string, string>;
+}
+export interface OpenTableOptions {
+    /**
+     * Configuration for object storage.
+     *
+     * Options already set on the connection will be inherited by the table,
+     * but can be overridden here.
+     *
+     * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+     */
+    storageOptions?: Record<string, string>;
+    /**
+     * Set the size of the index cache, specified as a number of entries
+     *
+     * The exact meaning of an "entry" will depend on the type of index:
+     * - IVF: there is one entry for each IVF partition
+     * - BTREE: there is one entry for the entire index
+     *
+     * This cache applies to the entire opened table, across all indices.
+     * Setting this value higher will increase performance on larger datasets
+     * at the expense of more RAM
+     */
+    indexCacheSize?: number;
 }
 export interface TableNamesOptions {
     /**
@@ -75,7 +120,7 @@ export declare class Connection {
      * Open a table in the database.
      * @param {string} name - The name of the table
      */
-    openTable(name: string): Promise<Table>;
+    openTable(name: string, options?: Partial<OpenTableOptions>): Promise<Table>;
     /**
      * Creates a new Table and initialize it with new data.
      * @param {string} name - The name of the table.

package/nodejs-artifacts/connection.js

@@ -13,10 +13,30 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.Connection = void 0;
+exports.Connection = exports.connect = void 0;
 const arrow_1 = require("./arrow");
+const native_1 = require("./native");
 const table_1 = require("./table");
 const apache_arrow_1 = require("apache-arrow");
+/**
+ * Connect to a LanceDB instance at the given URI.
+ *
+ * Accepted formats:
+ *
+ * - `/path/to/database` - local database
+ * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
+ * - `db://host:port` - remote database (LanceDB cloud)
+ * @param {string} uri - The uri of the database. If the database uri starts
+ * with `db://` then it connects to a remote database.
+ * @see {@link ConnectionOptions} for more details on the URI format.
+ */
+async function connect(uri, opts) {
+    opts = opts ?? {};
+    opts.storageOptions = cleanseStorageOptions(opts.storageOptions);
+    const nativeConn = await native_1.Connection.new(uri, opts);
+    return new Connection(nativeConn);
+}
+exports.connect = connect;
 /**
  * A LanceDB Connection that allows you to open tables and create new ones.
  *
@@ -72,8 +92,8 @@ class Connection {
      * Open a table in the database.
      * @param {string} name - The name of the table
      */
-    async openTable(name) {
-        const innerTable = await this.inner.openTable(name);
+    async openTable(name, options) {
+        const innerTable = await this.inner.openTable(name, cleanseStorageOptions(options?.storageOptions), options?.indexCacheSize);
         return new table_1.Table(innerTable);
     }
     /**
@@ -96,7 +116,7 @@ class Connection {
             table = (0, arrow_1.makeArrowTable)(data);
         }
         const buf = await (0, arrow_1.fromTableToBuffer)(table);
-        const innerTable = await this.inner.createTable(name, buf, mode);
+        const innerTable = await this.inner.createTable(name, buf, mode, cleanseStorageOptions(options?.storageOptions));
         return new table_1.Table(innerTable);
     }
     /**
@@ -112,7 +132,7 @@ class Connection {
         }
         const table = (0, arrow_1.makeEmptyTable)(schema);
         const buf = await (0, arrow_1.fromTableToBuffer)(table);
-        const innerTable = await this.inner.createEmptyTable(name, buf, mode);
+        const innerTable = await this.inner.createEmptyTable(name, buf, mode, cleanseStorageOptions(options?.storageOptions));
         return new table_1.Table(innerTable);
     }
     /**
@@ -124,3 +144,38 @@ class Connection {
     }
 }
 exports.Connection = Connection;
+/**
+ * Takes storage options and makes all the keys snake case.
+ */
+function cleanseStorageOptions(options) {
+    if (options === undefined) {
+        return undefined;
+    }
+    const result = {};
+    for (const [key, value] of Object.entries(options)) {
+        if (value !== undefined) {
+            const newKey = camelToSnakeCase(key);
+            result[newKey] = value;
+        }
+    }
+    return result;
+}
+/**
+ * Convert a string to snake case. It might already be snake case, in which case it is
+ * returned unchanged.
+ */
+function camelToSnakeCase(camel) {
+    if (camel.includes("_")) {
+        // Assume if there is at least one underscore, it is already snake case
+        return camel;
+    }
+    if (camel.toLocaleUpperCase() === camel) {
+        // Assume if the string is all uppercase, it is already snake case
+        return camel;
+    }
+    let result = camel.replace(/[A-Z]/g, (letter) => `_${letter.toLowerCase()}`);
+    if (result.startsWith("_")) {
+        result = result.slice(1);
+    }
+    return result;
+}

package/nodejs-artifacts/index.d.ts

@@ -1,22 +1,7 @@
-import { Connection } from "./connection";
-import { ConnectionOptions } from "./native.js";
 export { WriteOptions, WriteMode, AddColumnsSql, ColumnAlteration, ConnectionOptions, } from "./native.js";
 export { makeArrowTable, MakeArrowTableOptions, Data, VectorColumnOptions, } from "./arrow";
-export { Connection, CreateTableOptions, TableNamesOptions, } from "./connection";
+export { connect, Connection, CreateTableOptions, TableNamesOptions, } from "./connection";
 export { ExecutableQuery, Query, QueryBase, VectorQuery, RecordBatchIterator, } from "./query";
 export { Index, IndexOptions, IvfPqOptions } from "./indices";
 export { Table, AddDataOptions, IndexConfig, UpdateOptions } from "./table";
 export * as embedding from "./embedding";
-/**
- * Connect to a LanceDB instance at the given URI.
- *
- * Accpeted formats:
- *
- * - `/path/to/database` - local database
- * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
- * - `db://host:port` - remote database (LanceDB cloud)
- * @param {string} uri - The uri of the database. If the database uri starts
- * with `db://` then it connects to a remote database.
- * @see {@link ConnectionOptions} for more details on the URI format.
- */
-export declare function connect(uri: string, opts?: Partial<ConnectionOptions>): Promise<Connection>;

package/nodejs-artifacts/index.js

@@ -13,15 +13,14 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.connect = exports.embedding = exports.Table = exports.Index = exports.RecordBatchIterator = exports.VectorQuery = exports.QueryBase = exports.Query = exports.Connection = exports.VectorColumnOptions = exports.MakeArrowTableOptions = exports.makeArrowTable = void 0;
-const connection_1 = require("./connection");
-const native_js_1 = require("./native.js");
+exports.embedding = exports.Table = exports.Index = exports.RecordBatchIterator = exports.VectorQuery = exports.QueryBase = exports.Query = exports.Connection = exports.connect = exports.VectorColumnOptions = exports.MakeArrowTableOptions = exports.makeArrowTable = void 0;
 var arrow_1 = require("./arrow");
 Object.defineProperty(exports, "makeArrowTable", { enumerable: true, get: function () { return arrow_1.makeArrowTable; } });
 Object.defineProperty(exports, "MakeArrowTableOptions", { enumerable: true, get: function () { return arrow_1.MakeArrowTableOptions; } });
 Object.defineProperty(exports, "VectorColumnOptions", { enumerable: true, get: function () { return arrow_1.VectorColumnOptions; } });
-var connection_2 = require("./connection");
-Object.defineProperty(exports, "Connection", { enumerable: true, get: function () { return connection_2.Connection; } });
+var connection_1 = require("./connection");
+Object.defineProperty(exports, "connect", { enumerable: true, get: function () { return connection_1.connect; } });
+Object.defineProperty(exports, "Connection", { enumerable: true, get: function () { return connection_1.Connection; } });
 var query_1 = require("./query");
 Object.defineProperty(exports, "Query", { enumerable: true, get: function () { return query_1.Query; } });
 Object.defineProperty(exports, "QueryBase", { enumerable: true, get: function () { return query_1.QueryBase; } });
@@ -32,21 +31,3 @@ Object.defineProperty(exports, "Index", { enumerable: true, get: function () { r
 var table_1 = require("./table");
 Object.defineProperty(exports, "Table", { enumerable: true, get: function () { return table_1.Table; } });
 exports.embedding = require("./embedding");
-/**
- * Connect to a LanceDB instance at the given URI.
- *
- * Accpeted formats:
- *
- * - `/path/to/database` - local database
- * - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud storage
- * - `db://host:port` - remote database (LanceDB cloud)
- * @param {string} uri - The uri of the database. If the database uri starts
- * with `db://` then it connects to a remote database.
- * @see {@link ConnectionOptions} for more details on the URI format.
- */
-async function connect(uri, opts) {
-    opts = opts ?? {};
-    const nativeConn = await native_js_1.Connection.new(uri, opts);
-    return new connection_1.Connection(nativeConn);
-}
-exports.connect = connect;

package/nodejs-artifacts/native.d.ts

@@ -62,6 +62,12 @@ export interface ConnectionOptions {
    * always consistent.
    */
   readConsistencyInterval?: number
+  /**
+   * (For LanceDB OSS only): configuration for object storage.
+   *
+   * The available options are described at https://lancedb.github.io/lancedb/guides/storage/
+   */
+  storageOptions?: Record<string, string>
 }
 /** Write mode for writing a table. */
 export const enum WriteMode {
@@ -73,7 +79,9 @@ export const enum WriteMode {
 export interface WriteOptions {
   mode?: WriteMode
 }
-export function connect(uri: string, options: ConnectionOptions): Promise<Connection>
+export interface OpenTableOptions {
+  storageOptions?: Record<string, string>
+}
 export class Connection {
   /** Create a new Connection instance from the given URI. */
   static new(uri: string, options: ConnectionOptions): Promise<Connection>
@@ -90,9 +98,9 @@ export class Connection {
    * - buf: The buffer containing the IPC file.
    *
    */
-  createTable(name: string, buf: Buffer, mode: string): Promise<Table>
-  createEmptyTable(name: string, schemaBuf: Buffer, mode: string): Promise<Table>
-  openTable(name: string): Promise<Table>
+  createTable(name: string, buf: Buffer, mode: string, storageOptions?: Record<string, string> | undefined | null): Promise<Table>
+  createEmptyTable(name: string, schemaBuf: Buffer, mode: string, storageOptions?: Record<string, string> | undefined | null): Promise<Table>
+  openTable(name: string, storageOptions?: Record<string, string> | undefined | null, indexCacheSize?: number | undefined | null): Promise<Table>
   /** Drop table with the name. Or raise an error if the table does not exist. */
   dropTable(name: string): Promise<void>
 }

package/nodejs-artifacts/native.js

@@ -303,7 +303,7 @@ if (!nativeBinding) {
     }
     throw new Error(`Failed to load native binding`);
 }
-const { Connection, Index, RecordBatchIterator, Query, VectorQuery, Table, WriteMode, connect } = nativeBinding;
+const { Connection, Index, RecordBatchIterator, Query, VectorQuery, Table, WriteMode } = nativeBinding;
 module.exports.Connection = Connection;
 module.exports.Index = Index;
 module.exports.RecordBatchIterator = RecordBatchIterator;
@@ -311,4 +311,3 @@ module.exports.Query = Query;
 module.exports.VectorQuery = VectorQuery;
 module.exports.Table = Table;
 module.exports.WriteMode = WriteMode;
-module.exports.connect = connect;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.4.15",
+  "version": "0.4.18",
   "main": "./dist/index.js",
   "types": "./dist/index.d.ts",
   "napi": {
@@ -18,6 +18,8 @@
   },
   "license": "Apache 2.0",
   "devDependencies": {
+    "@aws-sdk/client-s3": "^3.33.0",
+    "@aws-sdk/client-kms": "^3.33.0",
     "@napi-rs/cli": "^2.18.0",
     "@types/jest": "^29.1.2",
     "@types/tmp": "^0.2.6",
@@ -63,18 +65,19 @@
     "lint": "eslint lancedb && eslint __test__",
     "prepublishOnly": "napi prepublish -t npm",
     "test": "npm run build && jest --verbose",
+    "integration": "S3_TEST=1 npm run test",
     "universal": "napi universal",
     "version": "napi version"
   },
-  "optionalDependencies": {
-    "@lancedb/lancedb-darwin-arm64": "0.4.15",
-    "@lancedb/lancedb-linux-arm64-gnu": "0.4.15",
-    "@lancedb/lancedb-darwin-x64": "0.4.15",
-    "@lancedb/lancedb-linux-x64-gnu": "0.4.15",
-    "@lancedb/lancedb-win32-x64-msvc": "0.4.15"
-  },
   "dependencies": {
     "openai": "^4.29.2",
     "apache-arrow": "^15.0.0"
+  },
+  "optionalDependencies": {
+    "@lancedb/lancedb-darwin-arm64": "0.4.18",
+    "@lancedb/lancedb-linux-arm64-gnu": "0.4.18",
+    "@lancedb/lancedb-darwin-x64": "0.4.18",
+    "@lancedb/lancedb-linux-x64-gnu": "0.4.18",
+    "@lancedb/lancedb-win32-x64-msvc": "0.4.18"
   }
 }