Purpose

This pull request adds unit tests for the vector_db.py file, which contains the main implementation of the vector database functionality.

Description

The changes include:

1. Setting up the test environment with pytest fixtures for mocking external dependencies like OpenAI and Redis.
2. Writing test cases for key functions in vector_db.py, including:
   • cosine_similarity
   • chunk
   • normalize_l2
   • embed_text_array
   • openai_call_embedding with caching

The tests use mocking to simulate the behavior of external dependencies and ensure the functions work as expected.

Summary

• Added sweepai/core/vector_db_test.py to contain the unit tests
• Wrote tests for the following functions:
  • cosine_similarity
  • chunk
  • normalize_l2
  • embed_text_array
  • openai_call_embedding with caching
• Used pytest fixtures to mock external dependencies like OpenAI and Redis

Suggested changes from Sweep Chat by @kevinlu1248. Continue chatting at http://localhost:3000/c/7f605494-06b7-43db-8a28-a50cbb8c92a2.