mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
39975613b1
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
125 lines
4 KiB
Python
125 lines
4 KiB
Python
"""Tests for `hermes checkpoints prune`'s orphan confirmation flow.
|
|
|
|
Covers the P1 raised on PR #69141: the confirmation preview must cover
|
|
BOTH v2 projects (`store_status()["projects"]`) and pre-v2 shadow repos
|
|
(`store_status()["pre_v2_projects"]`), since `prune_checkpoints()` deletes
|
|
orphans from both layouts. Exercises decline / accept / --force across
|
|
pre-v2-only and mixed (v2 + pre-v2) stores.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
|
|
import pytest
|
|
|
|
|
|
def _ns(**kwargs) -> argparse.Namespace:
|
|
defaults = {"retention_days": 7, "max_size_mb": 500, "keep_orphans": False, "force": False}
|
|
defaults.update(kwargs)
|
|
return argparse.Namespace(**defaults)
|
|
|
|
|
|
def _prune_result(**kwargs) -> dict:
|
|
result = {"scanned": 0, "deleted_orphan": 0, "deleted_stale": 0, "errors": 0, "bytes_freed": 0}
|
|
result.update(kwargs)
|
|
return result
|
|
|
|
|
|
_V2_ORPHAN_ONLY_STATUS = {
|
|
"projects": [],
|
|
"pre_v2_projects": [],
|
|
}
|
|
|
|
_PRE_V2_ONLY_STATUS = {
|
|
"projects": [],
|
|
"pre_v2_projects": [
|
|
{"path": "/home/user/.hermes/checkpoints/deadbeefcafebabe", "workdir": None, "exists": False},
|
|
],
|
|
}
|
|
|
|
_MIXED_STATUS = {
|
|
"projects": [
|
|
{"hash": "abc123", "workdir": "/gone/v2-project", "exists": False, "commits": 4},
|
|
],
|
|
"pre_v2_projects": [
|
|
{"path": "/home/user/.hermes/checkpoints/deadbeefcafebabe", "workdir": "/gone/pre-v2-project", "exists": False},
|
|
],
|
|
}
|
|
|
|
|
|
def _patch_checkpoint_manager(monkeypatch, status: dict, prune_calls: list):
|
|
import tools.checkpoint_manager as ckpt_mgr
|
|
|
|
monkeypatch.setattr(ckpt_mgr, "store_status", lambda *a, **k: status)
|
|
|
|
def _fake_prune(**kwargs):
|
|
prune_calls.append(kwargs)
|
|
return _prune_result(
|
|
deleted_orphan=len(status["projects"]) + len(status["pre_v2_projects"]),
|
|
)
|
|
|
|
monkeypatch.setattr(ckpt_mgr, "prune_checkpoints", _fake_prune)
|
|
|
|
|
|
# ─── pre-v2-only store ──────────────────────────────────────────────────────
|
|
|
|
|
|
|
|
|
|
# ─── mixed store (v2 + pre-v2) ──────────────────────────────────────────────
|
|
|
|
|
|
|
|
|
|
# ─── --keep-orphans skips the prompt entirely, on either layout ───────────
|
|
|
|
|
|
@pytest.mark.parametrize("status", [_PRE_V2_ONLY_STATUS, _MIXED_STATUS], ids=["pre_v2_only", "mixed"])
|
|
def test_keep_orphans_skips_prompt(monkeypatch, capsys, status):
|
|
import hermes_cli.checkpoints as checkpoints_cli
|
|
|
|
prune_calls: list = []
|
|
_patch_checkpoint_manager(monkeypatch, status, prune_calls)
|
|
|
|
def _unexpected_input(_prompt):
|
|
raise AssertionError("input() must not be called when --keep-orphans is passed")
|
|
|
|
monkeypatch.setattr("builtins.input", _unexpected_input)
|
|
|
|
rc = checkpoints_cli.cmd_prune(_ns(keep_orphans=True))
|
|
|
|
assert rc == 0
|
|
assert len(prune_calls) == 1
|
|
assert prune_calls[0]["delete_orphans"] is False
|
|
|
|
|
|
# ─── no orphans present: never prompts even without --force ───────────────
|
|
|
|
|
|
# ─── allowlist binding: preview set == deletion set, even when empty ───────
|
|
|
|
|
|
def test_empty_preview_binds_empty_allowlist(monkeypatch, capsys):
|
|
"""Zero-orphan-preview timing regression (PR #69141 review).
|
|
|
|
When the non-force preview shows zero orphans, no prompt runs — but the
|
|
later rescan inside prune_checkpoints() may discover a project that
|
|
became orphaned *after* the preview. That undisplayed, unconfirmed orphan
|
|
must not be deletable: the allowlist passed down must be the exact
|
|
(empty) displayed set, never the unrestricted None sentinel.
|
|
"""
|
|
import hermes_cli.checkpoints as checkpoints_cli
|
|
|
|
prune_calls: list = []
|
|
_patch_checkpoint_manager(monkeypatch, _V2_ORPHAN_ONLY_STATUS, prune_calls)
|
|
|
|
rc = checkpoints_cli.cmd_prune(_ns())
|
|
|
|
assert rc == 0
|
|
assert len(prune_calls) == 1
|
|
assert prune_calls[0]["orphan_allowlist"] == set()
|
|
|
|
|
|
|
|
|