open-swe/tests/analyzer/test_eval_jobs.py
Adam Moussa ae1f883b4c
refactor: move tests into tests/<domain>/ layout
Applies the plan's C5 step: git mv every test per the domain-reorg
move-map (movemap-m50.txt) into tests/{agent,analyzer,auth,dashboard,
github,middleware,models,reviewer,sandbox,slack,tools,webhooks}/, plus
the 13 fork-only placements from the scoping report §2c (Atlassian
webhook tests -> tests/webhooks/, test_atlassian_connect.py and
test_auth_error_leak.py -> tests/auth/, jira/confluence util tests ->
tests/tools/, test_repo_binding_isolation.py -> tests/sandbox/,
bot-identity/autofix tests -> tests/github/).

Path-only move: the only content edits are parents[1] -> parents[2]
fixes in test_e2b_integration.py and test_daytona_integration.py,
required because their __file__-relative ROOT path gained one more
directory level in the move.

Monkeypatch retargets for these files were already completed in C4;
none remained outstanding here.
2026-07-17 14:42:45 -04:00

68 lines
2.7 KiB
Python

from __future__ import annotations
from datetime import UTC, datetime, timedelta
from unittest.mock import AsyncMock, patch
import pytest
from agent.dashboard import eval_jobs
@pytest.mark.asyncio
async def test_get_status_returns_idle_when_no_record() -> None:
with patch.object(eval_jobs, "_get_record", new=AsyncMock(return_value=None)):
status = await eval_jobs.get_reviewer_eval_status()
assert status["status"] == "idle"
assert status["name"] == eval_jobs.REVIEWER_EVAL_KEY
@pytest.mark.asyncio
async def test_get_status_returns_terminal_record_unchanged() -> None:
record = {"name": "reviewer", "status": "completed", "experiment_url": "https://x"}
with (
patch.object(eval_jobs, "_get_record", new=AsyncMock(return_value=record)),
patch.object(eval_jobs, "_put_record", new=AsyncMock(side_effect=lambda r: r)) as put,
):
status = await eval_jobs.get_reviewer_eval_status()
assert status is record
put.assert_not_awaited()
@pytest.mark.asyncio
async def test_get_status_reconciles_stale_running() -> None:
stale = (datetime.now(UTC) - timedelta(seconds=300)).isoformat()
record = {"name": "reviewer", "status": "running", "heartbeat": stale}
with (
patch.object(eval_jobs, "_get_record", new=AsyncMock(return_value=record)),
patch.object(eval_jobs, "_put_record", new=AsyncMock(side_effect=lambda r: r)) as put,
):
status = await eval_jobs.get_reviewer_eval_status()
assert status["status"] == "failed"
assert "no longer tracked" in status["error"]
put.assert_awaited_once()
@pytest.mark.asyncio
async def test_get_status_keeps_running_with_fresh_heartbeat() -> None:
"""A poll must not kill a run whose Action is still heartbeating."""
fresh = datetime.now(UTC).isoformat()
record = {"name": "reviewer", "status": "running", "heartbeat": fresh}
with (
patch.object(eval_jobs, "_get_record", new=AsyncMock(return_value=record)),
patch.object(eval_jobs, "_put_record", new=AsyncMock(side_effect=lambda r: r)) as put,
):
status = await eval_jobs.get_reviewer_eval_status()
assert status["status"] == "running"
put.assert_not_awaited()
@pytest.mark.asyncio
async def test_get_status_fails_running_without_heartbeat() -> None:
"""A running record with no heartbeat at all is treated as stale."""
record = {"name": "reviewer", "status": "running"}
with (
patch.object(eval_jobs, "_get_record", new=AsyncMock(return_value=record)),
patch.object(eval_jobs, "_put_record", new=AsyncMock(side_effect=lambda r: r)),
):
status = await eval_jobs.get_reviewer_eval_status()
assert status["status"] == "failed"