Overview

This SDK provides a TypeScript client for interacting with a Weaviate instance. Aim at simplifying operations like creating classes, adding objects, performing searches for data analysis. Please consider using the official Weaviate Node.js SDK.

Getting Started

Step 1: Import & Init

import WeaviateClient from 'weaviate-sdk';
// Initialize Weaviate client
const weaviateURL = 'http://localhost:8080'; 
const client = new WeaviateClient(weaviateURL);

Step 2: Store Data Batch.

• Useful when working with large datasets. Like converting a book or a file into chunks of data.

Use the storeInBatch method to store your data. This method takes care of splitting your data into batches and respecting rate limits.

const articlesTenant1 = [
  {
    title: 'Understanding Multi-tenancy',
    content: 'Multi-tenancy allows multiple users to share the same application while keeping their data isolated.',
    author: 'Alice',
  },
  {
    title: 'Advanced Weaviate Usage',
    content: 'Exploring advanced features of Weaviate for optimized vector search.',
    author: 'Alice',
  },
];

const articlesTenant2 = [
  {
    title: 'Introduction to Weaviate',
    content: 'Weaviate is a cloud-native, modular, real-time vector search engine built to scale your machine learning models.',
    author: 'Bob',
  },
  {
    title: 'Weaviate and UI Integration',
    content: 'Integrating Weaviate with front-end frameworks for seamless user tenant1.',
    author: 'Bob',
  },
];

async function addBatchObjects() {
  try {
    // without tenant
    await client.storeInBatch(articlesTenant1, 'CollectionWithoutTenant');
    // with tenant
    await client.storeInBatch(articlesTenant2, 'CollectionWithTenant', 'tenant9');
    console.log('Batch objects added successfully under specified tenants.');
  } catch (error) {
    console.error('Error adding batch objects:', error);
  }
}

addBatchObjects();

Step 3: Search

// Import WeaviateClient to interact with the Weaviate API
import WeaviateClient from 'weaviate-sdk';
const client = new WeaviateClient('http://localhost:8080');

// Define the search term and the class of data we're searching in
const searchQuery = 'machine learning'; // What we're looking for
const className = 'Articles'; // Where we're searching
// const tenant = 'user_name123'; // If need a seprate space insode the collection/class 

// Optional: If we want to limit the fields retrieved, specify them here (e.g., "title", "content")
// Leaving this empty will retrieve all fields, which may increase response time due to additional API calls
// const fields = ["title", "content"]

// Function to perform the search
async function searchItems() {
  try {
    const results = await client.search(searchQuery, className);
    console.log('Search results:', results);
  } catch (error) {
    console.error('Search failed:', error);
  }
}

// Call the function to execute the search
searchItems();

// Footnotes:
// To retrieve specific fields: search(searchQuery, className, ["title", "content"])
// To retrieve fields with tenant filtering: search(searchQuery, className, ["title", "content"], "tenant1")

License

This project is licensed under the ISC License.