Package Exports

ai-database
ai-database/dist/index.js
ai-database/dist/index.mjs

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (ai-database) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

ai-database

Direct interface to Payload CMS with database.do compatibility.

Installation

npm install ai-database
# or
yarn add ai-database
# or
pnpm add ai-database

Usage

Node.js Environment

import { getPayload } from 'payload'
import config from '@payload-config'
import { DB } from 'ai-database'

// Initialize with Payload instance
const payload = await getPayload({ config })
const db = DB({ payload })

// Use the same interface as database.do
const posts = await db.posts.find({
  where: { status: 'published' },
  limit: 10,
})

const post = await db.posts.findOne('post-123')

Edge Environment

import { DB } from 'ai-database'

// Initialize with REST API URL
const db = DB({
  apiUrl: 'https://your-payload-api.com/api',
  apiKey: 'your-api-key',
})

// Use the same interface as database.do
const posts = await db.posts.find({
  where: { status: 'published' },
  limit: 10,
})

const post = await db.posts.findOne('post-123')

Generating Embeddings

import { generateEmbedding, calculateSimilarity } from 'ai-database'

// Generate embeddings for text
const result = await generateEmbedding('Text to embed')

if (result.success) {
  console.log('Embedding:', result.embedding)
  console.log('Model used:', result.model)

  // Calculate similarity between two embeddings
  const embedding1 = result.embedding[0]
  const embedding2 = await generateEmbedding('Similar text').then((r) => r.embedding?.[0])

  if (embedding1 && embedding2) {
    const similarity = calculateSimilarity(embedding1, embedding2)
    console.log('Similarity score:', similarity)
  }
}

API

The API is compatible with database.do, providing the same methods for each collection:

find(options?): Find documents in the collection
findOne(id): Find a single document by ID
create(data): Create a new document
update(id, data): Update an existing document
delete(id): Delete a document
search(query, options?): Search for documents

Environment-specific Adapters

For more control over environment-specific initialization:

import { createNodeClient, createEdgeClient } from 'ai-database/adapters'

// Node.js
const nodeDb = createNodeClient({ payload })

// Edge
const edgeDb = createEdgeClient({
  apiUrl: 'https://your-payload-api.com/api',
})

Embedding Functions

generateEmbedding(text, options?): Generate embeddings for text using the AI SDK
calculateSimilarity(embedding1, embedding2): Calculate cosine similarity between two embeddings

License

MIT