From d46bd55cb4e008eeb8b8e44f6ffe8f0f535c117d Mon Sep 17 00:00:00 2001
From: Eduardo Rios <eduardo.rios@aignosi.com.br>
Date: Thu, 17 Sep 2026 09:39:35 -0300
Subject: [PATCH] test(e2e): add batch-size cap coverage for SIENTIAPDE-2013

Also fixes the test_activities fixture, which built mongodb_config by
hand and was missing the new required notification_queue_batch_size key
-- every e2e test would have hit a KeyError without this.

New test seeds 5 notifications with a batch size of 2 and confirms three
successive runs return them oldest-first in order (n1,n2 / n3,n4 / n5),
proving the cursor only advances to what was actually returned rather
than skipping documents a capped query didn't fetch. Verified against
real testcontainers Mongo/Redis/Temporal, not mocks.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
---
 e2e/conftest.py                               |  1 +
 e2e/scenarios.md                              |  1 +
 ...t_subworkflow_load_notification_package.py | 63 +++++++++++++++++++
 3 files changed, 65 insertions(+)

diff --git a/e2e/conftest.py b/e2e/conftest.py
index 520d231..ab81973 100644
--- a/e2e/conftest.py
+++ b/e2e/conftest.py
@@ -330,6 +330,7 @@ async def test_activities(
             'connection_string': f'mongodb://localhost:{mongo_port}',
             'database_name': E2E_DATABASE,
             'ttl_index_seconds': 3600,
+            'notification_queue_batch_size': 0,
         },
         email_config={
             'sender_email': 'e2e@example.com',
diff --git a/e2e/scenarios.md b/e2e/scenarios.md
index 1bf1eb5..f5bb353 100644
--- a/e2e/scenarios.md
+++ b/e2e/scenarios.md
@@ -110,6 +110,7 @@ No notifications → no SMTP, no Postgres row.
 - Prior timestamp → only newer notifications returned.
 - Query-time filter (SIENTIAPDE-2011) → union of active groups' `levels`, intersection of their `ignore_models_id`; a model ignored by only one of several active groups is not excluded from the fetch.
 - Empty Mongo → no Redis timestamp write.
+- Batch size cap (SIENTIAPDE-2013) → with 5 pending notifications and a batch size of 2, each run returns exactly the next-oldest 2 (sorted ascending by timestamp), and the Redis cursor advances only to what was actually returned, so three successive runs drain the backlog in order instead of skipping anything.
 
 ### ProcessNotifications — `e2e/test_subworkflow_process_notifications.py`
 
diff --git a/e2e/test_subworkflow_load_notification_package.py b/e2e/test_subworkflow_load_notification_package.py
index 9ba3f61..b860524 100644
--- a/e2e/test_subworkflow_load_notification_package.py
+++ b/e2e/test_subworkflow_load_notification_package.py
@@ -228,3 +228,66 @@ async def test_load_package_empty_mongo_no_redis_write(
     )
 
     assert redis_client.get('notification_last_timestamp:Alerts') is None
+
+
+@pytest.mark.e2e
+@pytest.mark.asyncio
+async def test_load_package_batch_size_caps_and_returns_oldest_first(
+    temporal_env: WorkflowEnvironment,
+    orchestrator_worker,
+    stub_workers,
+    mongo_uri: str,
+    redis_client: Redis,
+    test_activities,
+):
+    """
+    SIENTIAPDE-2013: a capped run returns the oldest pending documents first,
+    and the incremental cursor picks up the remainder on the next run.
+
+    Five notifications exist, but the batch size is capped at 2. Each run must
+    return exactly the next-oldest 2 (never an arbitrary subset), and the
+    Redis cursor must advance only to the max timestamp of what was actually
+    returned -- not the true max across all 5 -- or the skipped documents
+    would never be fetched again.
+    """
+    test_activities.notification_queue_batch_size = 2
+
+    seed_receiver_groups(mongo_uri, E2E_DATABASE, [default_receiver_group()])
+    seed_notifications(
+        mongo_uri,
+        E2E_DATABASE,
+        [
+            default_notification(
+                notification_id=f'n{i}',
+                timestamp=datetime(2024, 6, 1, 12, i, 0, tzinfo=UTC),
+            )
+            for i in range(1, 6)
+        ],
+    )
+
+    input_data = load_scenario_input('subworkflow_load_notification_package.json')
+    _build_metadata(input_data)
+
+    first_run = await start_and_await_workflow(
+        temporal_env.client,
+        LoadNotificationPackage.run,
+        input_data,
+        make_workflow_id('load-pkg-batch-1'),
+    )
+    assert [n['notification_id'] for n in first_run['notification_package']] == ['n1', 'n2']
+
+    second_run = await start_and_await_workflow(
+        temporal_env.client,
+        LoadNotificationPackage.run,
+        input_data,
+        make_workflow_id('load-pkg-batch-2'),
+    )
+    assert [n['notification_id'] for n in second_run['notification_package']] == ['n3', 'n4']
+
+    third_run = await start_and_await_workflow(
+        temporal_env.client,
+        LoadNotificationPackage.run,
+        input_data,
+        make_workflow_id('load-pkg-batch-3'),
+    )
+    assert [n['notification_id'] for n in third_run['notification_package']] == ['n5']