add new query endpoint for getting document count

[Rashmil-1999]

Looking at the diff, I can provide a clear explanation of this performance optimization:

Pull Request Summary

Performance Optimization: Document Count API Endpoint

Performance Gain: ~90% faster (3724ms → 386ms)

This PR introduces a dedicated lightweight endpoint for fetching document counts, resulting in a 89.6% reduction in response time.

Changes Made:

1. New Endpoint: Created /api/materialsTable/fetchDocumentCount.ts

   • Implements an optimized query that checks for document existence using .limit(1) instead of counting all documents
   • Returns a binary result: 1 if documents exist, 0 if none exist
   • Includes proper authorization via withCourseAccessFromRequest

2. Frontend Update: Modified src/pages/[course_name]/chat.tsx

   • Switched from calling fetchProjectMaterials (which fetches full document data with pagination)
   • Now calls the new dedicated fetchDocumentCount endpoint

Performance Impact:

• Before: 3,724 ms (fetching full project materials with pagination)
• After: 386 ms (simple existence check query)
• Improvement: 3,338 ms saved per request (89.6% faster)
• Speed multiplier: ~9.6x faster

Technical Approach:

Instead of fetching and paginating through all documents, the new endpoint uses a simple SQL LIMIT 1 query to check if any documents exist for the course. This dramatically reduces database load and query execution time while providing the same functional result for the UI.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Review	Updated (UTC)
uiuc-chat-frontend	Ready	Preview, Comment	Jan 12, 2026 9:27pm

[rohan-uiuc]

Let's change this to fetch a single id from documents table where course_name is a match as that will be way faster that getting document count for projects with large number of documents. Additionally, we don't need filters in this API as they are not used anywhere.

[Rashmil-1999]

@rohan-uiuc I guess it is ready for another peek