This project provides an HTTP interface for the PineconeIndex class, allowing you to interact with your Pinecone vector database through REST API endpoints.

1. Set your environment variables:

   • PINECONE_KEY: Your Pinecone API key
   • OPENAI_KEY: Your OpenAI API key

2. The HTTP server is available at /pinecone_http_server

GET /health/

Returns the server status and configuration check.

POST /query/
Content-Type: application/json

{
  "query": "your search text here"
}

Or simply send a string directly:

POST /query/
Content-Type: application/json

"your search text here"

POST /upsert/
Content-Type: application/json

{
  "texts": ["document 1", "document 2", "document 3"]
}

POST /empty/

Clears all vectors from the index/namespace.

// Query the index
const response = await fetch('https://your-val-url.web.val.run/query/', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({ query: 'machine learning' })
});
const results = await response.json();

// Add documents
await fetch('https://your-val-url.web.val.run/upsert/', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({ 
    texts: ['AI is transforming industries', 'Machine learning enables automation'] 
  })
});

# Health check
curl https://your-val-url.web.val.run/health/

# Query
curl -X POST https://your-val-url.web.val.run/query/ \
  -H "Content-Type: application/json" \
  -d '{"query": "artificial intelligence"}'

# Add documents
curl -X POST https://your-val-url.web.val.run/upsert/ \
  -H "Content-Type: application/json" \
  -d '{"texts": ["Document 1 content", "Document 2 content"]}'

• Automatic routing: The handleRequest method in PineconeIndex automatically routes requests based on URL path
• Error handling: Comprehensive error handling with helpful messages
• Flexible input: Query endpoint accepts both string and object formats
• Environment validation: Clear feedback when API keys are missing
• Health monitoring: Built-in health check endpoint

The enhanced PineconeIndex class now includes:

• handleRequest(req: Request): Main HTTP router
• _handleQuery(req: Request): Process query requests
• _handleUpsert(req: Request): Process document uploads
• _handleEmpty(req: Request): Process index clearing requests

All existing methods (query, upsertRecords, empty, etc.) remain unchanged and work as before.