Database Options

BrowserAI provides built-in database support using IndexedDB and SQLite for managing AI data, conversations, and embeddings directly in the browser.

Database Configuration Options

Option	Type	Default	Description
databaseName	string	’BrowserAI-DB’	Name of the database
version	number	1	Database version for schema updates
storeName	string	’BrowserAIStore’	Name of the object store
type	’indexeddb’ \| ‘sqlite'	'indexeddb’	Type of database to use

Database Operations

The database implementation supports the following core operations:

Operation	Description
store	Store a new document with unique ID
get	Retrieve a document by ID
getAll	Retrieve all stored documents
update	Update an existing document
delete	Delete a document by ID
clear	Clear all documents from database

Example Usage

import { DatabaseImpl } from '@browserai/browserai';
 
// Initialize database
const db = new DatabaseImpl();
await db.initialize({ 
  type: 'indexeddb',
  config: {
    databaseName: 'MyApp-DB',
    version: 1,
    storeName: 'documents'
  }
});
 
// Store a document
await db.store({ 
  id: 'doc1',
  text: 'Sample document',
  embedding: [0.1, 0.2, 0.3] // Optional embedding vector
});
 
// Retrieve a document
const doc = await db.get('doc1');
console.log(doc); // { id: 'doc1', text: 'Sample document', ... }
 
// Update a document
await db.update({
  id: 'doc1',
  text: 'Updated document'
});
 
// Delete a document
await db.delete('doc1');
 
// Clear database
await db.clear();
 
// Close database connection
db.close();

Vector Store Support

The database also supports vector storage and similarity search when working with embeddings:

// Initialize vector store
db.setVectorStore('default');
 
// Add vector for a document
await db.addVector('doc1', [0.1, 0.2, 0.3]);
 
// Search similar documents
const similarDocs = await db.searchVector([0.1, 0.2, 0.3], 5);
console.log(similarDocs); // ['doc2', 'doc3', ...]

IndexedDB Demo Implementation

Here’s a complete example showing how to implement a database interface with React:

import { useEffect, useState } from 'react';
import { DatabaseImpl } from '@browserai/browserai';
 
function DatabaseDemo() {
  const [db, setDb] = useState(null);
  
  useEffect(() => {
    const initDb = async () => {
      const database = new DatabaseImpl();
      await database.initialize({
        type: 'indexeddb',
        config: {
          databaseName: 'BrowserAI-DB',
          version: 1,
          stores: { documents: { keyPath: 'id' } }
        }
      });
      setDb(database);
    };
    
    initDb();
    return () => db?.close();
  }, []);
 
  const handleStore = async () => {
    await db.store({
      id: 'doc1',
      text: 'Sample text',
      embedding: [0.1, 0.2, 0.3]
    });
  };
 
  // ... other database operations
}

Best Practices

Initialize Early: Initialize the database when your application starts
Close Properly: Always close database connections when no longer needed
Error Handling: Implement proper error handling for database operations
Version Management: Increment database version when changing schema
Unique IDs: Ensure document IDs are unique within your application

Limitations

IndexedDB storage is limited by available browser storage
Vector store operations require additional memory for similarity computations
SQLite support requires additional setup in some environments

For detailed examples of database implementations, see the Database Demo.

Text to Speech Options BrowserAI Class