feat(community): Remove required param from LanceDB integration

examples/src/indexes/vector_stores/lancedb/fromDocs.ts

@@ -11,6 +11,23 @@ const loader = new TextLoader("src/document_loaders/example_data/example.txt");
 const docs = await loader.load();
 
 export const run = async () => {
+  const vectorStore = await LanceDB.fromDocuments(
+    docs,
+    new OpenAIEmbeddings(),
+  );
+
+  const resultOne = await vectorStore.similaritySearch("hello world", 1);
+  console.log(resultOne);
+
+  // [
+  //   Document {
+  //     pageContent: 'Foo\nBar\nBaz\n\n',
+  //     metadata: { source: 'src/document_loaders/example_data/example.txt' }
+  //   }
+  // ]
+};
+
+export const run_with_existing_table = async () => {
   const dir = await fs.mkdtemp(path.join(os.tmpdir(), "lancedb-"));
   const db = await connect(dir);
   const table = await db.createTable("vectors", [

examples/src/indexes/vector_stores/lancedb/fromTexts.ts

@@ -6,6 +6,18 @@ import * as path from "node:path";
 import os from "node:os";
 
 export const run = async () => {
+  const vectorStore = await LanceDB.fromTexts(
+    ["Hello world", "Bye bye", "hello nice world"],
+    [{ id: 2 }, { id: 1 }, { id: 3 }],
+    new OpenAIEmbeddings(),
+  );
+
+  const resultOne = await vectorStore.similaritySearch("hello world", 1);
+  console.log(resultOne);
+  // [ Document { pageContent: 'hello nice world', metadata: { id: 3 } } ]
+};
+
+export const run_with_existing_table = async () => {
   const dir = await fs.mkdtemp(path.join(os.tmpdir(), "lancedb-"));
   const db = await connect(dir);
   const table = await db.createTable("vectors", [

libs/langchain-community/package.json

@@ -205,7 +205,7 @@
     "typescript": "~5.1.6",
     "typesense": "^1.5.3",
     "usearch": "^1.1.1",
-    "vectordb": "^0.1.4",
+    "vectordb": "^0.9.0",
     "voy-search": "0.6.2",
     "weaviate-ts-client": "^1.4.0",
     "web-auth-library": "^1.0.3",

libs/langchain-community/src/vectorstores/lancedb.ts

@@ -1,4 +1,4 @@
-import { Table } from "vectordb";
+import { connect, Table, Connection, WriteMode } from "vectordb";
 import type { EmbeddingsInterface } from "@langchain/core/embeddings";
 import { VectorStore } from "@langchain/core/vectorstores";
 import { Document } from "@langchain/core/documents";
@@ -8,8 +8,11 @@ import { Document } from "@langchain/core/documents";
  * table and an optional textKey.
  */
 export type LanceDBArgs = {
-  table: Table;
+  table?: Table;
   textKey?: string;
+  uri?: string;
+  table_name?: string;
+  mode?: string;
 };
 
 /**
@@ -18,15 +21,24 @@ export type LanceDBArgs = {
  * embeddings.
  */
 export class LanceDB extends VectorStore {
-  private table: Table;
+  private table?: Table;
 
   private textKey: string;
 
-  constructor(embeddings: EmbeddingsInterface, args: LanceDBArgs) {
-    super(embeddings, args);
-    this.table = args.table;
+  private uri: string;
+
+  private table_name: string;
+
+  private mode: string;
+
+  constructor(embeddings: EmbeddingsInterface, args?: LanceDBArgs) {
+    super(embeddings, args || {});
+    this.table = args?.table || undefined;
     this.embeddings = embeddings;
-    this.textKey = args.textKey || "text";
+    this.textKey = args?.textKey || "text";
+    this.uri = args?.uri || "~/lancedb";
+    this.table_name = args?.table_name || "langchain";
+    this.mode = args?.mode || WriteMode.Overwrite;
   }
 
   /**
@@ -71,6 +83,14 @@ export class LanceDB extends VectorStore {
       });
       data.push(record);
     }
+    if (!this.table) {
+      const db: Connection = await connect(this.uri);
+      this.table = await db.createTable(this.table_name, data, {
+        writeMode: this.mode,
+      });
+
+      return;
+    }
     await this.table.add(data);
   }
 
@@ -85,6 +105,11 @@ export class LanceDB extends VectorStore {
     query: number[],
     k: number
   ): Promise<[Document, number][]> {
+    if (!this.table) {
+      throw new Error(
+        "Table not found. Please add vectors to the table first."
+      );
+    }
     const results = await this.table.search(query).limit(k).execute();
 
     const docsAndScore: [Document, number][] = [];
@@ -119,7 +144,7 @@ export class LanceDB extends VectorStore {
     texts: string[],
     metadatas: object[] | object,
     embeddings: EmbeddingsInterface,
-    dbConfig: LanceDBArgs
+    dbConfig?: LanceDBArgs
   ): Promise<LanceDB> {
     const docs: Document[] = [];
     for (let i = 0; i < texts.length; i += 1) {
@@ -143,7 +168,7 @@ export class LanceDB extends VectorStore {
   static async fromDocuments(
     docs: Document[],
     embeddings: EmbeddingsInterface,
-    dbConfig: LanceDBArgs
+    dbConfig?: LanceDBArgs
   ): Promise<LanceDB> {
     const instance = new this(embeddings, dbConfig);
     await instance.addDocuments(docs);

libs/langchain-community/src/vectorstores/tests/lancedb.int.test.ts

@@ -45,3 +45,27 @@ describe("LanceDB", () => {
     expect(resultsTwo.length).toBe(5);
   });
 });
+
+describe("LanceDB empty schema", () => {
+  test("Test fromTexts + addDocuments", async () => {
+    const embeddings = new OpenAIEmbeddings();
+    const vectorStore = await LanceDB.fromTexts(
+      ["hello bye", "hello world", "bye bye"],
+      [{ id: 1 }, { id: 2 }, { id: 3 }],
+      embeddings
+    );
+
+    const results = await vectorStore.similaritySearch("hello bye", 10);
+    expect(results.length).toBe(3);
+
+    await vectorStore.addDocuments([
+      new Document({
+        pageContent: "a new world",
+        metadata: { id: 4 },
+      }),
+    ]);
+
+    const resultsTwo = await vectorStore.similaritySearch("hello bye", 10);
+    expect(resultsTwo.length).toBe(4);
+  });
+});