cognee 0.5.1__py3-none-any.whl → 0.5.2__py3-none-any.whl

cognee/tests/unit/modules/retrieval/summaries_retriever_test.py

@@ -1,159 +1,188 @@
-import os
 import pytest
-import pathlib
-
-import cognee
-from cognee.low_level import setup
-from cognee.tasks.storage import add_data_points
-from cognee.infrastructure.databases.vector import get_vector_engine
-from cognee.modules.chunking.models import DocumentChunk
-from cognee.tasks.summarization.models import TextSummary
-from cognee.modules.data.processing.document_types import TextDocument
-from cognee.modules.retrieval.exceptions.exceptions import NoDataError
+from unittest.mock import AsyncMock, patch, MagicMock
+from types import SimpleNamespace
+
 from cognee.modules.retrieval.summaries_retriever import SummariesRetriever
+from cognee.modules.retrieval.exceptions.exceptions import NoDataError
+from cognee.infrastructure.databases.vector.exceptions import CollectionNotFoundError
+
+
+@pytest.fixture
+def mock_vector_engine():
+    """Create a mock vector engine."""
+    engine = AsyncMock()
+    engine.search = AsyncMock()
+    return engine
+
+
+@pytest.mark.asyncio
+async def test_get_context_success(mock_vector_engine):
+    """Test successful retrieval of summary context."""
+    mock_result1 = MagicMock()
+    mock_result1.payload = {"text": "S.R.", "made_from": "chunk1"}
+    mock_result2 = MagicMock()
+    mock_result2.payload = {"text": "M.B.", "made_from": "chunk2"}
+
+    mock_vector_engine.search.return_value = [mock_result1, mock_result2]
+
+    retriever = SummariesRetriever(top_k=5)
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+        context = await retriever.get_context_from_objects("test query", objects)
+        completion = await retriever.get_completion_from_context("test query", objects, context)
+
+    assert len(completion) == 2
+    assert completion[0]["text"] == "S.R."
+    assert completion[1]["text"] == "M.B."
+    mock_vector_engine.search.assert_awaited_once_with(
+        "TextSummary_text", "test query", limit=5, include_payload=True
+    )
+
+
+@pytest.mark.asyncio
+async def test_get_context_collection_not_found_error(mock_vector_engine):
+    """Test that CollectionNotFoundError is converted to NoDataError."""
+    mock_vector_engine.search.side_effect = CollectionNotFoundError("Collection not found")
+
+    retriever = SummariesRetriever()
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        with pytest.raises(NoDataError, match="No data found"):
+            await retriever.get_retrieved_objects("test query")
+
+
+@pytest.mark.asyncio
+async def test_get_objects_empty_results(mock_vector_engine):
+    """Test that empty list is returned when no summaries are found."""
+    mock_vector_engine.search.return_value = []
+
+    retriever = SummariesRetriever()
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+
+    assert objects == []
+
+
+@pytest.mark.asyncio
+async def test_get_objects_top_k_limit(mock_vector_engine):
+    """Test that top_k parameter limits the number of results."""
+    mock_results = [MagicMock() for _ in range(3)]
+    for i, result in enumerate(mock_results):
+        result.payload = {"text": f"Summary {i}"}
+
+    mock_vector_engine.search.return_value = mock_results
+
+    retriever = SummariesRetriever(top_k=3)
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+
+    assert len(objects) == 3
+    mock_vector_engine.search.assert_awaited_once_with(
+        "TextSummary_text", "test query", limit=3, include_payload=True
+    )
+
+
+@pytest.mark.asyncio
+async def test_get_context_with_objects(mock_vector_engine):
+    """Test get_completion returns provided context."""
+    retriever = SummariesRetriever()
+
+    provided_context = {"text": "S.R."}
+    sn = SimpleNamespace()
+    sn.payload = provided_context
+    completion = await retriever.get_context_from_objects("test query", retrieved_objects=[sn])
+
+    assert completion == provided_context["text"]
+
+
+@pytest.mark.asyncio
+async def test_get_completion_without_context(mock_vector_engine):
+    """Test get_completion retrieves context when not provided."""
+    mock_result = MagicMock()
+    mock_result.payload = {"text": "S.R."}
+    mock_vector_engine.search.return_value = [mock_result]
+
+    retriever = SummariesRetriever()
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+        context = await retriever.get_context_from_objects("test query", objects)
+        completion = await retriever.get_completion_from_context("test query", objects, context)
+
+    assert len(completion) == 1
+    assert completion[0]["text"] == "S.R."
+
+
+@pytest.mark.asyncio
+async def test_init_defaults():
+    """Test SummariesRetriever initialization with defaults."""
+    retriever = SummariesRetriever()
+
+    assert retriever.top_k == 5
+
+
+@pytest.mark.asyncio
+async def test_init_custom_top_k():
+    """Test SummariesRetriever initialization with custom top_k."""
+    retriever = SummariesRetriever(top_k=10)
+
+    assert retriever.top_k == 10
+
+
+@pytest.mark.asyncio
+async def test_get_objects_empty_payload(mock_vector_engine):
+    """Test get_context handles empty payload."""
+    mock_result = MagicMock()
+    mock_result.payload = {}
+
+    mock_vector_engine.search.return_value = [mock_result]
+
+    retriever = SummariesRetriever()
+
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+
+    assert len(objects) == 1
+    assert objects[0].payload == {}
+
+
+@pytest.mark.asyncio
+async def test_get_completion_with_session_id(mock_vector_engine):
+    """Test get_completion with session_id parameter."""
+    mock_result = MagicMock()
+    mock_result.payload = {"text": "S.R."}
+    mock_vector_engine.search.return_value = [mock_result]
+
+    retriever = SummariesRetriever(session_id="test_session")
 
+    with patch(
+        "cognee.modules.retrieval.summaries_retriever.get_vector_engine",
+        return_value=mock_vector_engine,
+    ):
+        objects = await retriever.get_retrieved_objects("test query")
+        context = await retriever.get_context_from_objects("test query", objects)
+        completion = await retriever.get_completion_from_context("test query", objects, context)
 
-class TestSummariesRetriever:
-    @pytest.mark.asyncio
-    async def test_chunk_context(self):
-        system_directory_path = os.path.join(
-            pathlib.Path(__file__).parent, ".cognee_system/test_chunk_context"
-        )
-        cognee.config.system_root_directory(system_directory_path)
-        data_directory_path = os.path.join(
-            pathlib.Path(__file__).parent, ".data_storage/test_chunk_context"
-        )
-        cognee.config.data_root_directory(data_directory_path)
-
-        await cognee.prune.prune_data()
-        await cognee.prune.prune_system(metadata=True)
-        await setup()
-
-        document1 = TextDocument(
-            name="Employee List",
-            raw_data_location="somewhere",
-            external_metadata="",
-            mime_type="text/plain",
-        )
-
-        document2 = TextDocument(
-            name="Car List",
-            raw_data_location="somewhere",
-            external_metadata="",
-            mime_type="text/plain",
-        )
-
-        chunk1 = DocumentChunk(
-            text="Steve Rodger",
-            chunk_size=2,
-            chunk_index=0,
-            cut_type="sentence_end",
-            is_part_of=document1,
-            contains=[],
-        )
-        chunk1_summary = TextSummary(
-            text="S.R.",
-            made_from=chunk1,
-        )
-        chunk2 = DocumentChunk(
-            text="Mike Broski",
-            chunk_size=2,
-            chunk_index=1,
-            cut_type="sentence_end",
-            is_part_of=document1,
-            contains=[],
-        )
-        chunk2_summary = TextSummary(
-            text="M.B.",
-            made_from=chunk2,
-        )
-        chunk3 = DocumentChunk(
-            text="Christina Mayer",
-            chunk_size=2,
-            chunk_index=2,
-            cut_type="sentence_end",
-            is_part_of=document1,
-            contains=[],
-        )
-        chunk3_summary = TextSummary(
-            text="C.M.",
-            made_from=chunk3,
-        )
-        chunk4 = DocumentChunk(
-            text="Range Rover",
-            chunk_size=2,
-            chunk_index=0,
-            cut_type="sentence_end",
-            is_part_of=document2,
-            contains=[],
-        )
-        chunk4_summary = TextSummary(
-            text="R.R.",
-            made_from=chunk4,
-        )
-        chunk5 = DocumentChunk(
-            text="Hyundai",
-            chunk_size=2,
-            chunk_index=1,
-            cut_type="sentence_end",
-            is_part_of=document2,
-            contains=[],
-        )
-        chunk5_summary = TextSummary(
-            text="H.Y.",
-            made_from=chunk5,
-        )
-        chunk6 = DocumentChunk(
-            text="Chrysler",
-            chunk_size=2,
-            chunk_index=2,
-            cut_type="sentence_end",
-            is_part_of=document2,
-            contains=[],
-        )
-        chunk6_summary = TextSummary(
-            text="C.H.",
-            made_from=chunk6,
-        )
-
-        entities = [
-            chunk1_summary,
-            chunk2_summary,
-            chunk3_summary,
-            chunk4_summary,
-            chunk5_summary,
-            chunk6_summary,
-        ]
-
-        await add_data_points(entities)
-
-        retriever = SummariesRetriever(top_k=20)
-
-        context = await retriever.get_context("Christina")
-
-        assert context[0]["text"] == "C.M.", "Failed to get Christina Mayer"
-
-    @pytest.mark.asyncio
-    async def test_chunk_context_on_empty_graph(self):
-        system_directory_path = os.path.join(
-            pathlib.Path(__file__).parent, ".cognee_system/test_chunk_context_on_empty_graph"
-        )
-        cognee.config.system_root_directory(system_directory_path)
-        data_directory_path = os.path.join(
-            pathlib.Path(__file__).parent, ".data_storage/test_chunk_context_on_empty_graph"
-        )
-        cognee.config.data_root_directory(data_directory_path)
-
-        await cognee.prune.prune_data()
-        await cognee.prune.prune_system(metadata=True)
-
-        retriever = SummariesRetriever()
-
-        with pytest.raises(NoDataError):
-            await retriever.get_context("Christina Mayer")
-
-        vector_engine = get_vector_engine()
-        await vector_engine.create_collection("TextSummary_text", payload_schema=TextSummary)
-
-        context = await retriever.get_context("Christina Mayer")
-        assert context == [], "Returned context should be empty on an empty graph"
+    assert len(completion) == 1
+    assert completion[0]["text"] == "S.R."