ed/manual_slop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

test(stabilization): Resolve run_linear API drift and implement vlogger high-signal reporting

Browse Source
This commit is contained in:
Edward R. Gonzalez
2026-02-28 20:18:05 -05:00
parent 2a2675e386
commit ece46f922c
5 changed files with 174 additions and 143 deletions
Download Patch File Download Diff File Expand all files Collapse all files

63
tests/test_conductor_engine.py
View File

@@ -1,7 +1,8 @@
import pytest
import pytest
from unittest.mock import MagicMock, patch
from models import Ticket, Track, WorkerContext
import ai_client
import multi_agent_conductor
# These tests define the expected interface for multi_agent_conductor.py
# which will be implemented in the next phase of TDD.
@@ -12,19 +13,24 @@ def test_conductor_engine_initialization() -> None:
"""
track = Track(id="test_track", description="Test Track")
from multi_agent_conductor import ConductorEngine
engine = ConductorEngine(track=track)
engine = ConductorEngine(track=track, auto_queue=True)
assert engine.track == track
@pytest.mark.asyncio
async def test_conductor_engine_run_linear_executes_tickets_in_order(monkeypatch: pytest.MonkeyPatch) -> None:
async def test_conductor_engine_run_executes_tickets_in_order(monkeypatch: pytest.MonkeyPatch, vlogger) -> None:
"""
Test that run_linear iterates through executable tickets and calls the worker lifecycle.
Test that run iterates through executable tickets and calls the worker lifecycle.
"""
ticket1 = Ticket(id="T1", description="Task 1", status="todo", assigned_to="worker1")
ticket2 = Ticket(id="T2", description="Task 2", status="todo", assigned_to="worker2", depends_on=["T1"])
track = Track(id="track1", description="Track 1", tickets=[ticket1, ticket2])
from multi_agent_conductor import ConductorEngine
engine = ConductorEngine(track=track)
engine = ConductorEngine(track=track, auto_queue=True)
vlogger.log_state("Ticket Count", 0, 2)
vlogger.log_state("T1 Status", "todo", "todo")
vlogger.log_state("T2 Status", "todo", "todo")
# Mock ai_client.send using monkeypatch
mock_send = MagicMock()
monkeypatch.setattr(ai_client, 'send', mock_send)
@@ -36,7 +42,11 @@ async def test_conductor_engine_run_linear_executes_tickets_in_order(monkeypatch
ticket.mark_complete()
return "Success"
mock_lifecycle.side_effect = side_effect
await engine.run_linear()
await engine.run()
vlogger.log_state("T1 Status Final", "todo", ticket1.status)
vlogger.log_state("T2 Status Final", "todo", ticket2.status)
# Track.get_executable_tickets() should be called repeatedly until all are done
# T1 should run first, then T2.
assert mock_lifecycle.call_count == 2
@@ -46,6 +56,7 @@ async def test_conductor_engine_run_linear_executes_tickets_in_order(monkeypatch
calls = mock_lifecycle.call_args_list
assert calls[0][0][0].id == "T1"
assert calls[1][0][0].id == "T2"
vlogger.finalize("Verify dependency execution order", "PASS", "T1 executed before T2")
@pytest.mark.asyncio
async def test_run_worker_lifecycle_calls_ai_client_send(monkeypatch: pytest.MonkeyPatch) -> None:
@@ -144,8 +155,12 @@ async def test_run_worker_lifecycle_step_mode_confirmation(monkeypatch: pytest.M
# Mock ai_client.send using monkeypatch
mock_send = MagicMock()
monkeypatch.setattr(ai_client, 'send', mock_send)
with patch("multi_agent_conductor.confirm_execution") as mock_confirm:
# We simulate ai_client.send by making it call the pre_tool_callback it received
# Important: confirm_spawn is called first if event_queue is present!
with patch("multi_agent_conductor.confirm_spawn") as mock_spawn, \
patch("multi_agent_conductor.confirm_execution") as mock_confirm:
mock_spawn.return_value = (True, "mock prompt", "mock context")
mock_confirm.return_value = True
def mock_send_side_effect(md_content, user_message, **kwargs):
callback = kwargs.get("pre_tool_callback")
@@ -154,9 +169,12 @@ async def test_run_worker_lifecycle_step_mode_confirmation(monkeypatch: pytest.M
callback('{"tool": "read_file", "args": {"path": "test.txt"}}')
return "Success"
mock_send.side_effect = mock_send_side_effect
mock_confirm.return_value = True
mock_event_queue = MagicMock()
run_worker_lifecycle(ticket, context, event_queue=mock_event_queue)
# Verify confirm_spawn was called because event_queue was present
mock_spawn.assert_called_once()
# Verify confirm_execution was called
mock_confirm.assert_called_once()
assert ticket.status == "completed"
@@ -173,25 +191,28 @@ async def test_run_worker_lifecycle_step_mode_rejection(monkeypatch: pytest.Monk
# Mock ai_client.send using monkeypatch
mock_send = MagicMock()
monkeypatch.setattr(ai_client, 'send', mock_send)
with patch("multi_agent_conductor.confirm_execution") as mock_confirm:
with patch("multi_agent_conductor.confirm_spawn") as mock_spawn, \
patch("multi_agent_conductor.confirm_execution") as mock_confirm:
mock_spawn.return_value = (True, "mock prompt", "mock context")
mock_confirm.return_value = False
mock_send.return_value = "Task failed because tool execution was rejected."
run_worker_lifecycle(ticket, context)
mock_event_queue = MagicMock()
run_worker_lifecycle(ticket, context, event_queue=mock_event_queue)
# Verify it was passed to send
args, kwargs = mock_send.call_args
assert kwargs["pre_tool_callback"] is not None
# Since we've already tested ai_client's implementation of pre_tool_callback (mentally or via other tests),
# here we just verify the wiring.
@pytest.mark.asyncio
async def test_conductor_engine_dynamic_parsing_and_execution(monkeypatch: pytest.MonkeyPatch) -> None:
async def test_conductor_engine_dynamic_parsing_and_execution(monkeypatch: pytest.MonkeyPatch, vlogger) -> None:
"""
Test that parse_json_tickets correctly populates the track and run_linear executes them in dependency order.
Test that parse_json_tickets correctly populates the track and run executes them in dependency order.
"""
import json
from multi_agent_conductor import ConductorEngine
track = Track(id="dynamic_track", description="Dynamic Track")
engine = ConductorEngine(track=track)
engine = ConductorEngine(track=track, auto_queue=True)
tickets_json = json.dumps([
{
"id": "T1",
@@ -216,6 +237,8 @@ async def test_conductor_engine_dynamic_parsing_and_execution(monkeypatch: pytes
}
])
engine.parse_json_tickets(tickets_json)
vlogger.log_state("Parsed Ticket Count", 0, len(engine.track.tickets))
assert len(engine.track.tickets) == 3
assert engine.track.tickets[0].id == "T1"
assert engine.track.tickets[1].id == "T2"
@@ -229,12 +252,18 @@ async def test_conductor_engine_dynamic_parsing_and_execution(monkeypatch: pytes
ticket.mark_complete()
return "Success"
mock_lifecycle.side_effect = side_effect
await engine.run_linear()
await engine.run()
assert mock_lifecycle.call_count == 3
# Verify dependency order: T1 must be called before T2
calls = [call[0][0].id for call in mock_lifecycle.call_args_list]
t1_idx = calls.index("T1")
t2_idx = calls.index("T2")
vlogger.log_state("T1 Sequence Index", "N/A", t1_idx)
vlogger.log_state("T2 Sequence Index", "N/A", t2_idx)
assert t1_idx < t2_idx
# T3 can be anywhere relative to T1 and T2, but T1 < T2 is mandatory
assert "T3" in calls
vlogger.finalize("Dynamic track parsing and dependency execution", "PASS", "Dependency chain T1 -> T2 honored.")