Update tests for vector_store_code

tests/test_vector_store.py

@@ -0,0 +1,149 @@
+import os
+import tempfile
+from pathlib import Path
+from unittest.mock import MagicMock, Mock, patch
+
+import pytest
+from langchain_core.documents.base import Document
+from langchain_core.vectorstores import VectorStoreRetriever
+
+from reviewllama.utilities import is_ollama_available
+from reviewllama.vector_store import (create_retriever,
+                                      documents_from_path_list,
+                                      get_context_from_store)
+
+
+@pytest.fixture
+def temp_files():
+    """Create temporary test files"""
+    temp_dir = tempfile.mkdtemp()
+    files = []
+
+    # Create test files with different content
+    file_contents = [
+        "This is the first test file with Python code examples.",
+        "The second file contains JavaScript functions and classes.",
+        "File three has documentation about testing best practices.",
+        "Final file includes configuration settings and deployment info.",
+    ]
+
+    for i, content in enumerate(file_contents):
+        file_path = Path(temp_dir) / f"test_file_{i}.txt"
+        file_path.write_text(content)
+        files.append(file_path)
+
+    yield files
+
+    # Cleanup
+    for file_path in files:
+        file_path.unlink()
+    os.rmdir(temp_dir)
+
+
+def test_load_documents(temp_files):
+    docs = documents_from_path_list(temp_files)
+
+    assert len(docs) == 4
+    assert all(isinstance(doc, Document) for doc in docs)
+    assert all(doc.page_content for doc in docs)
+    assert all(doc.metadata.get("source") for doc in docs)
+
+
+@patch("reviewllama.vector_store.OllamaEmbeddings")
+@patch("reviewllama.vector_store.FAISS")
+@patch("reviewllama.vector_store.documents_from_path_list")
+def test_create_retriever(mock_docs_from_list, mock_faiss, mock_embeddings):
+    """Test successful retriever creation"""
+    # Setup mocks
+    mock_docs = [Document(page_content="test", metadata={"source": "test.txt"})]
+    mock_docs_from_list.return_value = mock_docs
+
+    mock_embedding_instance = Mock()
+    mock_embeddings.return_value = mock_embedding_instance
+
+    mock_vectorstore = Mock()
+    mock_retriever = Mock(spec=VectorStoreRetriever)
+    mock_vectorstore.as_retriever.return_value = mock_retriever
+    mock_faiss.from_documents.return_value = mock_vectorstore
+
+    # Test
+    result = create_retriever(["test.txt"], "test-embedding-model")
+
+    # Assertions
+    assert result == mock_retriever
+    mock_embeddings.assert_called_once_with(model="test-embedding-model")
+    mock_docs_from_list.assert_called_once_with(["test.txt"])
+    mock_faiss.from_documents.assert_called_once_with(
+        mock_docs, mock_embedding_instance
+    )
+    mock_vectorstore.as_retriever.assert_called_once()
+
+def test_get_context_from_store_success():
+    """Test successful context retrieval"""
+    mock_retriever = Mock(spec=VectorStoreRetriever)
+    mock_docs = [
+        Document(page_content="First relevant document", metadata={}),
+        Document(page_content="Second relevant document", metadata={}),
+        Document(page_content="Third relevant document", metadata={}),
+    ]
+    mock_retriever.get_relevant_documents.return_value = mock_docs
+    
+    result = get_context_from_store("test query", mock_retriever)
+    
+    expected = "First relevant document\n\nSecond relevant document\n\nThird relevant document"
+    assert result == expected
+    mock_retriever.get_relevant_documents.assert_called_once_with("test query")
+
+@patch('reviewllama.vector_store.OllamaEmbeddings')
+@patch('reviewllama.vector_store.FAISS')
+def test_full_pipeline_mock(mock_faiss, mock_embeddings, temp_files):
+    """Test the full pipeline with mocked external dependencies"""
+    # Setup mocks
+    mock_embedding_instance = Mock()
+    mock_embeddings.return_value = mock_embedding_instance
+    
+    mock_vectorstore = Mock()
+    mock_retriever = Mock(spec=VectorStoreRetriever)
+    mock_retriever.get_relevant_documents.return_value = [
+        Document(page_content="Relevant test content", metadata={})
+    ]
+    mock_vectorstore.as_retriever.return_value = mock_retriever
+    mock_faiss.from_documents.return_value = mock_vectorstore
+    
+    # Test full pipeline
+    retriever = create_retriever(temp_files[:2], "test-model")
+    context = get_context_from_store("test query", retriever)
+    
+    assert context == "Relevant test content"
+    mock_embeddings.assert_called_once_with(model="test-model")
+    mock_retriever.get_relevant_documents.assert_called_once_with("test query")
+
+def test_documents_from_list_content_verification(temp_files):
+    """Test that documents contain expected content"""
+    docs = documents_from_path_list(temp_files)
+    
+    contents = [doc.page_content for doc in docs]
+    
+    # Check that we have the expected content
+    assert any("Python code examples" in content for content in contents)
+    assert any("JavaScript functions" in content for content in contents)
+    assert any("testing best practices" in content for content in contents)
+    assert any("deployment info" in content for content in contents)
+
+# Optional: Integration test that requires actual Ollama server
+def test_create_retriever_with_real_ollama(temp_files, ollama_config):
+    """Integration test with real Ollama (requires server running)"""
+    if not is_ollama_available(ollama_config):
+        pytest.skip("Local Ollama server is not available")
+    try:
+        # This test would use a real embedding model
+        # Skip by default unless explicitly testing integration
+        retriever = create_retriever(temp_files[:2], "nomic-embed-text")
+        assert retriever is not None
+        
+        # Test actual retrieval
+        context = get_context_from_store("Python code", retriever)
+        assert isinstance(context, str)
+        
+    except Exception as e:
+        pytest.skip(f"Ollama server not available or model not found: {e}")