manual_slop/tests/test_rag_engine.py

import pytest
import os
from unittest.mock import MagicMock, patch
from src import models
from src.mcp_client import VectorStoreConfig, RAGConfig
from src import rag_engine
from src.rag_engine import RAGEngine, BaseEmbeddingProvider, LocalEmbeddingProvider, GeminiEmbeddingProvider

class MockEmbeddingProvider(BaseEmbeddingProvider):
 def embed(self, texts):
  return [[0.1] * 384 for _ in texts]

@pytest.fixture
def mock_rag_config():
 vs_config = VectorStoreConfig(provider='mock', collection_name='test')
 return RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='gemini')

def test_rag_engine_init_mock(mock_rag_config):
 engine = RAGEngine(mock_rag_config)
 assert engine.config.enabled is True
 assert engine.collection == "mock"

def test_local_embedding_provider_missing_dependency_has_install_hint():
 with patch.object(rag_engine, "_SENTENCE_TRANSFORMERS", None):
  with patch.dict("sys.modules", {"sentence_transformers": None}):
   with pytest.raises(ImportError, match=r"manual_slop\[local-rag\]"):
    LocalEmbeddingProvider()

@patch('src.rag_engine.LocalEmbeddingProvider.embed')
@patch('src.rag_engine._get_chromadb')
def test_rag_engine_chroma(mock_get_chroma, mock_embed):
 mock_chroma = MagicMock()
 mock_settings = MagicMock()
 mock_get_chroma.return_value = (mock_chroma, mock_settings)

 mock_embed.return_value = [[0.1, 0.2, 0.3]]
 mock_collection = MagicMock()
 mock_client = MagicMock()
 mock_client.get_or_create_collection.return_value = mock_collection
 mock_chroma.PersistentClient.return_value = mock_client

 vs_config = VectorStoreConfig(provider='chroma', collection_name='test')
 config = RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='local')

 with patch('src.rag_engine._get_sentence_transformers') as mock_st:
  mock_st.return_value = MagicMock()
  engine = RAGEngine(config)
  assert engine.collection == mock_collection

  engine.add_documents(["doc1"], ["hello world"])
  mock_collection.upsert.assert_called_once()
  
  mock_collection.query.return_value = {
   "ids": [["doc1"]],
   "documents": [["hello world"]],
   "metadatas": [[{}]],
   "distances": [[0.0]]
  }
  
  results = engine.search("hello", top_k=1)
  assert len(results) == 1
  assert results[0].id == "doc1"
  engine.delete_documents(["doc1"])
  mock_collection.delete.assert_called_once_with(ids=["doc1"])

@patch('src.rag_engine.LocalEmbeddingProvider.embed')
@patch('src.rag_engine._get_chromadb')
def test_rag_collection_dim_mismatch_recreates_collection(mock_get_chroma, mock_embed):
 """
 Regression test for the live_gui_test_hardening_v2 followup
 (RAG dimension-mismatch flake in test_rag_phase4_stress).

 Scenario: a ChromaDB collection exists on disk with vectors from a
 previous embedding provider (e.g. Gemini, 3072-dim), but the current
 config uses a different provider (e.g. local SentenceTransformers,
 384-dim). Without the dim check, upsert silently corrupts the
 collection and search() later fails with
 "Collection expecting embedding with dimension of 3072, got 384".

 Expected: RAGEngine.__init__ detects the mismatch, deletes the
 mismatched collection via client.delete_collection, and recreates it
 empty so subsequent indexing uses the correct dim.
 """
 mock_chroma = MagicMock()
 mock_settings = MagicMock()
 mock_get_chroma.return_value = (mock_chroma, mock_settings)

 mock_embed.return_value = [[0.1] * 384]
 mock_collection = MagicMock()
 mock_collection.get.return_value = {
  "embeddings": [[0.1] * 3072],
  "metadatas":   [{}],
  "ids":         ["stale_doc_1"],
 }
 mock_collection.name = "test"

 mock_client = MagicMock()
 mock_client.get_or_create_collection.return_value = mock_collection
 mock_chroma.PersistentClient.return_value = mock_client

 vs_config = VectorStoreConfig(provider='chroma', collection_name='test')
 config = RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='local')

 with patch('src.rag_engine._get_sentence_transformers') as mock_st:
  mock_st.return_value = MagicMock()
  engine = RAGEngine(config)
  assert engine.collection == mock_collection
  # On dim mismatch, _validate_collection_dim_result rmtree's the collection
  # directory (WinError 32 safe) then creates a new PersistentClient +
  # collection. The first get_or_create_collection call was in
  # _init_vector_store_result. The old implementation called
  # client.delete_collection(name); the new implementation uses
  # shutil.rmtree + new PersistentClient for better Windows file-lock
  # robustness (per fix_rag_test_phase4_final_verify_20260627).
  assert mock_client.get_or_create_collection.call_count == 2
  mock_client.delete_collection.assert_not_called()

@patch('src.rag_engine.LocalEmbeddingProvider.embed')
@patch('src.rag_engine._get_chromadb')
def test_rag_collection_dim_match_preserves_collection(mock_get_chroma, mock_embed):
 """
 Companion test: when the collection's existing dim matches the current
 provider's dim, the engine must NOT delete the collection (which would
 discard indexed data).
 """
 mock_chroma = MagicMock()
 mock_settings = MagicMock()
 mock_get_chroma.return_value = (mock_chroma, mock_settings)

 mock_embed.return_value = [[0.1] * 384]
 mock_collection = MagicMock()
 mock_collection.get.return_value = {
  "embeddings": [[0.1] * 384],
  "metadatas":   [{"path": "file_25.txt"}],
  "ids":         ["doc_25_0"],
 }
 mock_collection.name = "test"

 mock_client = MagicMock()
 mock_client.get_or_create_collection.return_value = mock_collection
 mock_chroma.PersistentClient.return_value = mock_client

 vs_config = VectorStoreConfig(provider='chroma', collection_name='test')
 config = RAGConfig(enabled=True, vector_store=vs_config, embedding_provider='local')

 with patch('src.rag_engine._get_sentence_transformers') as mock_st:
  mock_st.return_value = MagicMock()
  engine = RAGEngine(config)
  assert engine.collection == mock_collection
  mock_client.delete_collection.assert_not_called()
  assert mock_client.get_or_create_collection.call_count == 1

  engine.delete_documents(["doc1"])
  mock_collection.delete.assert_called_once_with(ids=["doc1"])