Use session scoped fixtures instead of static variables (#599)

Author: cbornet

libs/colbert/pyproject.toml

@@ -17,10 +17,14 @@ torch = "2.2.1"
 cassio = "~0.1.7"
 pydantic = "^2.7.1"
 
+# Workaround for https://github.com/pytorch/pytorch/pull/127921
+# Remove when we upgrade to pytorch 2.4
+setuptools = { version = ">=70", python = ">=3.12" }
+
+
 [tool.poetry.group.test.dependencies]
 ragstack-ai-tests-utils = { path = "../tests-utils", develop = true }
 pytest-asyncio = "^0.23.6"
 
-[tool.poetry.group.dev.dependencies]
-setuptools = "70.0.0"
-
+[tool.pytest.ini_options]
+asyncio_mode = "auto"

libs/colbert/tests/integration_tests/conftest.py

@@ -1,28 +1,24 @@
 import pytest
+from cassandra.cluster import Session
 from ragstack_tests_utils import AstraDBTestStore, LocalCassandraTestStore
 
-status = {
-    "local_cassandra_test_store": None,
-    "astradb_test_store": None,
-}
 
+@pytest.fixture(scope="session")
+def cassandra() -> LocalCassandraTestStore:
+    store = LocalCassandraTestStore()
+    yield store
+    if store.docker_container:
+        store.docker_container.stop()
 
-def get_local_cassandra_test_store():
-    if not status["local_cassandra_test_store"]:
-        status["local_cassandra_test_store"] = LocalCassandraTestStore()
-    return status["local_cassandra_test_store"]
 
+@pytest.fixture(scope="session")
+def astra_db() -> AstraDBTestStore:
+    return AstraDBTestStore()
 
-def get_astradb_test_store():
-    if not status["astradb_test_store"]:
-        status["astradb_test_store"] = AstraDBTestStore()
-    return status["astradb_test_store"]
 
-
-@pytest.hookimpl()
-def pytest_sessionfinish():
-    if (
-        status["local_cassandra_test_store"]
-        and status["local_cassandra_test_store"].docker_container
-    ):
-        status["local_cassandra_test_store"].docker_container.stop()
+@pytest.fixture()
+def session(request) -> Session:
+    test_store = request.getfixturevalue(request.param)
+    session = test_store.create_cassandra_session()
+    session.default_timeout = 180
+    return session

libs/colbert/tests/integration_tests/test_database.py

@@ -1,27 +1,11 @@
 import pytest
+from cassandra.cluster import Session
 from ragstack_colbert import CassandraDatabase, Chunk
 from ragstack_tests_utils import TestData
 
-from tests.integration_tests.conftest import (
-    get_astradb_test_store,
-    get_local_cassandra_test_store,
-)
-
-
-@pytest.fixture()
-def cassandra():
-    return get_local_cassandra_test_store()
-
-
-@pytest.fixture()
-def astra_db():
-    return get_astradb_test_store()
-
-
-@pytest.mark.parametrize("vector_store", ["cassandra", "astra_db"])
-def test_database_sync(request, vector_store: str):
-    vector_store = request.getfixturevalue(vector_store)
 
+@pytest.mark.parametrize("session", ["cassandra", "astra_db"], indirect=["session"])
+def test_database_sync(session: Session):
     doc_id = "earth_doc_id"
 
     chunk_0 = Chunk(
@@ -40,9 +24,6 @@ def test_database_sync(request, vector_store: str):
         embedding=TestData.renewable_energy_embedding(),
     )
 
-    session = vector_store.create_cassandra_session()
-    session.default_timeout = 180
-
     database = CassandraDatabase.from_session(
         keyspace="default_keyspace",
         table_name="test_database_sync",
@@ -61,11 +42,8 @@ def test_database_sync(request, vector_store: str):
     assert result
 
 
-@pytest.mark.parametrize("vector_store", ["cassandra", "astra_db"])
-@pytest.mark.asyncio()
-async def test_database_async(request, vector_store: str):
-    vector_store = request.getfixturevalue(vector_store)
-
+@pytest.mark.parametrize("session", ["cassandra", "astra_db"], indirect=["session"])
+async def test_database_async(session: Session):
     doc_id = "earth_doc_id"
 
     chunk_0 = Chunk(
@@ -84,9 +62,6 @@ async def test_database_async(request, vector_store: str):
         embedding=TestData.renewable_energy_embedding(),
     )
 
-    session = vector_store.create_cassandra_session()
-    session.default_timeout = 180
-
     database = CassandraDatabase.from_session(
         keyspace="default_keyspace",
         table_name="test_database_async",

libs/colbert/tests/integration_tests/test_embedding_retrieval.py

@@ -1,32 +1,17 @@
 import logging
 
 import pytest
+from cassandra.cluster import Session
 from ragstack_colbert import (
     CassandraDatabase,
     ColbertEmbeddingModel,
     ColbertVectorStore,
 )
 from ragstack_tests_utils import TestData
 
-from tests.integration_tests.conftest import (
-    get_astradb_test_store,
-    get_local_cassandra_test_store,
-)
-
-
-@pytest.fixture()
-def cassandra():
-    return get_local_cassandra_test_store()
-
 
-@pytest.fixture()
-def astra_db():
-    return get_astradb_test_store()
-
-
-@pytest.mark.parametrize("vector_store", ["cassandra", "astra_db"])
-def test_embedding_cassandra_retriever(request, vector_store: str):
-    vector_store = request.getfixturevalue(vector_store)
+@pytest.mark.parametrize("session", ["cassandra", "astra_db"], indirect=["session"])
+def test_embedding_cassandra_retriever(session: Session):
     narrative = TestData.marine_animals_text()
 
     # Define the desired chunk size and overlap size
@@ -53,9 +38,6 @@ def chunk_texts(text, chunk_size, overlap_size):
 
     doc_id = "marine_animals"
 
-    session = vector_store.create_cassandra_session()
-    session.default_timeout = 180
-
     database = CassandraDatabase.from_session(
         keyspace="default_keyspace",
         table_name="test_embedding_cassandra_retriever",

libs/knowledge-store/pyproject.toml

@@ -20,7 +20,6 @@ mypy = "^1.10.0"
 pytest-asyncio = "^0.23.6"
 ipykernel = "^6.29.4"
 testcontainers = "~3.7.1"
-setuptools = "^70.0.0"
 python-dotenv = "^1.0.1"
 
 # Resolve numpy version for 3.8 to 3.12+

libs/knowledge-store/tests/integration_tests/test_graph_store.py

@@ -1,6 +1,6 @@
 import math
 import secrets
-from typing import Callable, Iterable, Iterator, List
+from typing import Iterable, Iterator, List
 
 import numpy as np
 import pytest
@@ -89,26 +89,25 @@ def cassandra() -> Iterator[LocalCassandraTestStore]:
 
 
 @pytest.fixture()
-def graph_store_factory(
+def graph_store(
     cassandra: LocalCassandraTestStore,
-) -> Iterator[Callable[[], GraphStore]]:
+) -> Iterator[GraphStore]:
     session = cassandra.create_cassandra_session()
     session.set_keyspace(KEYSPACE)
 
     embedding = SimpleEmbeddingModel()
 
-    def _make_graph_store() -> GraphStore:
-        name = secrets.token_hex(8)
+    name = secrets.token_hex(8)
 
-        node_table = f"nodes_{name}"
-        return GraphStore(
-            embedding,
-            session=session,
-            keyspace=KEYSPACE,
-            node_table=node_table,
-        )
+    node_table = f"nodes_{name}"
+    store = GraphStore(
+        embedding,
+        session=session,
+        keyspace=KEYSPACE,
+        node_table=node_table,
+    )
 
-    yield _make_graph_store
+    yield store
 
     session.shutdown()
 
@@ -117,15 +116,7 @@ def _result_ids(nodes: Iterable[Node]) -> List[str]:
     return [n.id for n in nodes if n.id is not None]
 
 
-def test_graph_store_creation(graph_store_factory: Callable[[], GraphStore]) -> None:
-    """Test that a graph store can be created.
-
-    This verifies the schema can be applied and the queries prepared.
-    """
-    graph_store_factory()
-
-
-def test_mmr_traversal(graph_store_factory: Callable[[], GraphStore]) -> None:
+def test_mmr_traversal(graph_store: GraphStore) -> None:
     """
     Test end to end construction and MMR search.
     The embedding function used here ensures `texts` become
@@ -145,8 +136,6 @@ def test_mmr_traversal(graph_store_factory: Callable[[], GraphStore]) -> None:
     Both v2 and v3 are reachable via edges from v0, so once it is
     selected, those are both considered.
     """
-    gs = graph_store_factory()
-
     v0 = Node(
         id="v0",
         text="-0.124",
@@ -166,32 +155,30 @@ def test_mmr_traversal(graph_store_factory: Callable[[], GraphStore]) -> None:
         text="+1.0",
         links={Link(direction="in", kind="explicit", tag="link")},
     )
-    gs.add_nodes([v0, v1, v2, v3])
+    graph_store.add_nodes([v0, v1, v2, v3])
 
-    results = gs.mmr_traversal_search("0.0", k=2, fetch_k=2)
+    results = graph_store.mmr_traversal_search("0.0", k=2, fetch_k=2)
     assert _result_ids(results) == ["v0", "v2"]
 
     # With max depth 0, no edges are traversed, so this doesn't reach v2 or v3.
     # So it ends up picking "v1" even though it's similar to "v0".
-    results = gs.mmr_traversal_search("0.0", k=2, fetch_k=2, depth=0)
+    results = graph_store.mmr_traversal_search("0.0", k=2, fetch_k=2, depth=0)
     assert _result_ids(results) == ["v0", "v1"]
 
     # With max depth 0 but higher `fetch_k`, we encounter v2
-    results = gs.mmr_traversal_search("0.0", k=2, fetch_k=3, depth=0)
+    results = graph_store.mmr_traversal_search("0.0", k=2, fetch_k=3, depth=0)
     assert _result_ids(results) == ["v0", "v2"]
 
     # v0 score is .46, v2 score is 0.16 so it won't be chosen.
-    results = gs.mmr_traversal_search("0.0", k=2, score_threshold=0.2)
+    results = graph_store.mmr_traversal_search("0.0", k=2, score_threshold=0.2)
     assert _result_ids(results) == ["v0"]
 
     # with k=4 we should get all of the documents.
-    results = gs.mmr_traversal_search("0.0", k=4)
+    results = graph_store.mmr_traversal_search("0.0", k=4)
     assert _result_ids(results) == ["v0", "v2", "v1", "v3"]
 
 
-def test_write_retrieve_keywords(graph_store_factory: Callable[[], GraphStore]) -> None:
-    gs = graph_store_factory()
-
+def test_write_retrieve_keywords(graph_store: GraphStore) -> None:
     greetings = Node(
         id="greetings",
         text="Typical Greetings",
@@ -218,36 +205,34 @@ def test_write_retrieve_keywords(graph_store_factory: Callable[[], GraphStore])
         },
     )
 
-    gs.add_nodes([greetings, doc1, doc2])
+    graph_store.add_nodes([greetings, doc1, doc2])
 
     # Doc2 is more similar, but World and Earth are similar enough that doc1 also shows
     # up.
-    results = gs.similarity_search(text_to_embedding("Earth"), k=2)
+    results = graph_store.similarity_search(text_to_embedding("Earth"), k=2)
     assert _result_ids(results) == ["doc2", "doc1"]
 
-    results = gs.similarity_search(text_to_embedding("Earth"), k=1)
+    results = graph_store.similarity_search(text_to_embedding("Earth"), k=1)
     assert _result_ids(results) == ["doc2"]
 
-    results = gs.traversal_search("Earth", k=2, depth=0)
+    results = graph_store.traversal_search("Earth", k=2, depth=0)
     assert _result_ids(results) == ["doc2", "doc1"]
 
-    results = gs.traversal_search("Earth", k=2, depth=1)
+    results = graph_store.traversal_search("Earth", k=2, depth=1)
     assert _result_ids(results) == ["doc2", "doc1", "greetings"]
 
     # K=1 only pulls in doc2 (Hello Earth)
-    results = gs.traversal_search("Earth", k=1, depth=0)
+    results = graph_store.traversal_search("Earth", k=1, depth=0)
     assert _result_ids(results) == ["doc2"]
 
     # K=1 only pulls in doc2 (Hello Earth). Depth=1 traverses to parent and via keyword
     # edge.
-    results = gs.traversal_search("Earth", k=1, depth=1)
+    results = graph_store.traversal_search("Earth", k=1, depth=1)
     assert set(_result_ids(results)) == {"doc2", "doc1", "greetings"}
 
 
-def test_metadata(graph_store_factory: Callable[[], GraphStore]) -> None:
-    gs = graph_store_factory()
-
-    gs.add_nodes(
+def test_metadata(graph_store: GraphStore) -> None:
+    graph_store.add_nodes(
         [
             Node(
                 id="a",
@@ -260,7 +245,7 @@ def test_metadata(graph_store_factory: Callable[[], GraphStore]) -> None:
             )
         ]
     )
-    results = list(gs.similarity_search(text_to_embedding("A")))
+    results = list(graph_store.similarity_search(text_to_embedding("A")))
     assert len(results) == 1
     assert results[0].id == "a"
     assert results[0].metadata["other"] == "some other field"

libs/langchain/pyproject.toml

@@ -42,6 +42,5 @@ pytest-asyncio = "^0.23.6"
 keybert = "^0.8.5"
 gliner = "^0.2.5"
 
-[tool.poetry.group.dev.dependencies]
-setuptools = "^70.0.0"
-
+[tool.pytest.ini_options]
+asyncio_mode = "auto"