427d4936c2
The merges from main kept or added about 45 test files written against the
legacy updater and installer seams that pm-clean replaced. Each failing test
was checked against the current code:
- Deleted, because every test targets code that pm-clean removed or that is
only reachable from dead or frozen old-updater code:
update_orphan/handoff_backend_reap, handoff_desktop_rebuild,
interrupted_recovery (it also launched a real completion against the live
checkout), update_stale_virtualenv, update_venv_health,
verify_core_dependencies, lazy_refresh_venv_repair,
certifi_repair (already dropped in 2f20ceb6f7; the merge resurrected it),
banner_git_state (covered by test_source_check).
- Retargeted where production reads now:
- version identity via version_info.get_code_identity
- update receipts via a ContextVar scope
- pm.extras / pm.installed_package / pm.ensure_import for matrix,
computer_use, fal and tirith
- install hints via pm.install_hint
- the lock pins in pm/lock.json
- probe_root for the critical-module probe
- Fixture repairs:
- git-committed trees for source stamps
- activate's real `--runtime-only` argument
- a semver install stamp for requires_hermes gates
- the Windows gateway restart after a verified update
- snowflake-length ids that cannot collide with the runner uid
- Collection fix: one platforms() marker per test in gateway_proc_fallback.
420 lines
17 KiB
Python
420 lines
17 KiB
Python
"""Tests for ``tools.computer_use.doctor``.
|
|
|
|
The doctor module drives cua-driver's stable ``health_report`` MCP tool over
|
|
stdio JSON-RPC and renders the structured response. Most of the surface is
|
|
about parsing what cua-driver hands back, plus the exit-code contract
|
|
downstream consumers (CI / `hermes update`) rely on:
|
|
|
|
* Exit 0 when overall == "ok"
|
|
* Exit 1 when overall in ("degraded", "failed") — at least one check
|
|
failed but the tool itself ran successfully
|
|
* Exit 2 when the cua-driver binary is missing or the protocol breaks
|
|
|
|
We do NOT spin up a real cua-driver — that lives in the cua-driver
|
|
integration test suite (libs/cua-driver/rust/tests/integration/
|
|
test_health_report_mcp.py). Here we mock the subprocess and assert the
|
|
Hermes-side adapter behaves correctly against the documented response
|
|
shape.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
from io import StringIO
|
|
from pathlib import Path
|
|
from types import SimpleNamespace
|
|
from unittest.mock import MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
|
|
# ── helpers ────────────────────────────────────────────────────────────────
|
|
|
|
|
|
def _pm_driver(binary: str):
|
|
"""PM reports *binary* as the installed cua-driver (the runtime's selection)."""
|
|
return patch("pm.installed_package", return_value=SimpleNamespace(binary=Path(binary)))
|
|
|
|
|
|
def _fake_proc_with_responses(*responses: dict) -> MagicMock:
|
|
"""Build a MagicMock subprocess.Popen handle that yields one JSON-RPC
|
|
response per `readline()` call, then returns "" (EOF)."""
|
|
lines = [json.dumps(r) + "\n" for r in responses] + [""]
|
|
proc = MagicMock()
|
|
proc.stdin = MagicMock()
|
|
proc.stdout = MagicMock()
|
|
proc.stdout.readline = MagicMock(side_effect=lines)
|
|
proc.stderr = MagicMock()
|
|
proc.stderr.read = MagicMock(return_value="")
|
|
proc.wait = MagicMock(return_value=0)
|
|
proc.kill = MagicMock()
|
|
return proc
|
|
|
|
|
|
def _ok_report() -> dict:
|
|
"""Minimal well-formed health_report response."""
|
|
return {
|
|
"schema_version": "1",
|
|
"platform": "darwin",
|
|
"driver_version": "0.5.8",
|
|
"overall": "ok",
|
|
"checks": [
|
|
{"name": "binary_version", "status": "pass", "message": "cua-driver 0.5.8"},
|
|
{"name": "tcc_accessibility", "status": "pass", "message": "Accessibility is granted."},
|
|
],
|
|
}
|
|
|
|
|
|
def _degraded_report() -> dict:
|
|
"""Report with one failing check — overall=degraded."""
|
|
return {
|
|
"schema_version": "1",
|
|
"platform": "darwin",
|
|
"driver_version": "0.5.8",
|
|
"overall": "degraded",
|
|
"checks": [
|
|
{"name": "binary_version", "status": "pass", "message": "cua-driver 0.5.8"},
|
|
{
|
|
"name": "bundle_identity",
|
|
"status": "fail",
|
|
"message": "Process has no CFBundleIdentifier.",
|
|
"hint": "Run inside CuaDriver.app",
|
|
"data": {"executable_path": "/tmp/cua-driver"},
|
|
},
|
|
],
|
|
}
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _default_cli_version_matches_report(monkeypatch):
|
|
"""Existing tests mock only the MCP Popen handshake. ``subprocess.run``
|
|
(used for ``--version``) goes through Popen too, so without this the
|
|
mock breaks version probing. Default to a CLI version that matches
|
|
``_ok_report`` / ``_degraded_report`` (0.5.8); identity tests override.
|
|
"""
|
|
from tools.computer_use import doctor
|
|
|
|
monkeypatch.setattr(
|
|
doctor,
|
|
"_read_cli_version",
|
|
lambda binary, timeout=5.0: "cua-driver 0.5.8",
|
|
)
|
|
|
|
|
|
# ── exit codes ─────────────────────────────────────────────────────────────
|
|
|
|
|
|
class TestDoctorExitCodes:
|
|
def test_ok_exits_0(self):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _ok_report()}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO):
|
|
code = doctor.run_doctor()
|
|
assert code == 0
|
|
|
|
def test_degraded_exits_1(self):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _degraded_report()}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO):
|
|
code = doctor.run_doctor()
|
|
assert code == 1
|
|
|
|
def test_failed_overall_exits_1(self):
|
|
"""`failed` overall (every check failed) is also exit 1, not 2 —
|
|
the tool ran successfully; the diagnosis was bad."""
|
|
from tools.computer_use import doctor
|
|
|
|
report = _degraded_report()
|
|
report["overall"] = "failed"
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": report}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO):
|
|
code = doctor.run_doctor()
|
|
assert code == 1
|
|
|
|
def test_spawn_denied_exits_2_with_diagnosis(self, capsys):
|
|
"""The runtime interpreter cannot execute the resolved binary (Windows WinError 5 on a
|
|
`WindowsApps` install): a diagnosis + exit 2, never a raw traceback."""
|
|
from tools.computer_use import doctor
|
|
|
|
with _pm_driver("/protected/cua-driver"), \
|
|
patch("subprocess.Popen", side_effect=PermissionError(13, "Access is denied")):
|
|
code = doctor.run_doctor()
|
|
assert code == 2
|
|
err = capsys.readouterr().err
|
|
assert "Access is denied" in err and "HERMES_CUA_DRIVER_CMD" in err
|
|
|
|
def test_protocol_error_exits_2(self, capsys):
|
|
"""An empty stdout response (driver crashed during handshake) is a
|
|
protocol failure → exit 2."""
|
|
from tools.computer_use import doctor
|
|
|
|
proc = MagicMock()
|
|
proc.stdin = MagicMock()
|
|
proc.stdout = MagicMock()
|
|
proc.stdout.readline = MagicMock(return_value="") # EOF on initialize
|
|
proc.stderr = MagicMock()
|
|
proc.stderr.read = MagicMock(return_value="boom\n")
|
|
proc.wait = MagicMock(return_value=0)
|
|
proc.kill = MagicMock()
|
|
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc):
|
|
code = doctor.run_doctor()
|
|
assert code == 2
|
|
# stderr should mention the failure
|
|
captured = capsys.readouterr()
|
|
assert "cua-driver" in captured.err.lower() or "health_report" in captured.err.lower()
|
|
|
|
|
|
# ── response-shape parsing ─────────────────────────────────────────────────
|
|
|
|
|
|
class TestResponseShapeParsing:
|
|
|
|
|
|
def test_jsonrpc_error_response_exits_2(self, capsys):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "error": {"code": -32601, "message": "method not found"}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc):
|
|
code = doctor.run_doctor()
|
|
assert code == 2
|
|
assert "method not found" in capsys.readouterr().err
|
|
|
|
|
|
# ── args / arg passthrough ─────────────────────────────────────────────────
|
|
|
|
|
|
class TestArgPassthrough:
|
|
def test_include_passed_through_to_tools_call(self):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _ok_report()}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO):
|
|
doctor.run_doctor(include=["binary_version", "tcc_accessibility"])
|
|
|
|
# Inspect the second write to stdin — the tools/call payload.
|
|
writes = [call.args[0] for call in proc.stdin.write.call_args_list]
|
|
call_payload = next(json.loads(w) for w in writes if "tools/call" in w)
|
|
assert call_payload["params"]["arguments"]["include"] == [
|
|
"binary_version", "tcc_accessibility",
|
|
]
|
|
|
|
|
|
# ── json output ────────────────────────────────────────────────────────────
|
|
|
|
|
|
class TestJsonOutput:
|
|
def test_json_output_is_parseable_round_trip(self):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _ok_report()}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO) as out:
|
|
doctor.run_doctor(json_output=True)
|
|
# Verify the captured text round-trips through json.loads. Upstream
|
|
# health_report keys are preserved; Hermes adds hermes_identity.
|
|
parsed = json.loads(out.getvalue())
|
|
report = _ok_report()
|
|
for key, value in report.items():
|
|
assert parsed[key] == value
|
|
assert "hermes_identity" in parsed
|
|
assert parsed["hermes_identity"]["resolved_binary"]
|
|
|
|
|
|
# ── HERMES_CUA_DRIVER_CMD resolution ───────────────────────────────────────
|
|
|
|
|
|
class TestDriverCmdResolution:
|
|
@staticmethod
|
|
def _executable(path: Path) -> Path:
|
|
path.parent.mkdir(parents=True, exist_ok=True)
|
|
path.write_text("#!/bin/sh\nexit 0\n")
|
|
path.chmod(0o755)
|
|
return path
|
|
|
|
def _inspected_binary(self, **kwargs) -> Path:
|
|
from tools.computer_use import doctor
|
|
|
|
with _pm_driver("/pm/store/cua-driver"), \
|
|
patch("tools.computer_use.doctor._drive_health_report", return_value=_ok_report()) as health, \
|
|
patch("sys.stdout", new_callable=StringIO):
|
|
assert doctor.run_doctor(**kwargs) == 0
|
|
return Path(health.call_args.args[0])
|
|
|
|
def test_explicit_driver_cmd_arg_wins(self, tmp_path, monkeypatch):
|
|
explicit = self._executable(tmp_path / "custom" / "cua-driver")
|
|
monkeypatch.setenv("HERMES_CUA_DRIVER_CMD", str(self._executable(tmp_path / "env" / "cua-driver")))
|
|
|
|
assert self._inspected_binary(driver_cmd=str(explicit)) == explicit
|
|
|
|
def test_env_var_used_when_no_arg_given(self, tmp_path, monkeypatch):
|
|
from_env = self._executable(tmp_path / "env" / "cua-driver")
|
|
monkeypatch.setenv("HERMES_CUA_DRIVER_CMD", str(from_env))
|
|
|
|
assert self._inspected_binary() == from_env
|
|
|
|
def test_doctor_inspects_the_pm_selected_driver_not_path(self, tmp_path, monkeypatch):
|
|
"""Doctor must diagnose the driver the runtime invokes: PM's pin, not a PATH copy."""
|
|
monkeypatch.delenv("HERMES_CUA_DRIVER_CMD", raising=False)
|
|
monkeypatch.setenv("PATH", str(self._executable(tmp_path / "bin" / "cua-driver").parent))
|
|
|
|
assert self._inspected_binary() == Path("/pm/store/cua-driver")
|
|
|
|
|
|
# ── cua-driver 0.10 unclassified health_report fallback ────────────────────
|
|
|
|
|
|
def _unclassified_health_result() -> dict:
|
|
"""MCP tools/call result shape from cua-driver 0.10.x denial."""
|
|
return {
|
|
"isError": True,
|
|
"content": [
|
|
{
|
|
"type": "text",
|
|
"text": (
|
|
"Permission denied: tool 'health_report' has no "
|
|
"reviewed risk classification"
|
|
),
|
|
}
|
|
],
|
|
"structuredContent": {"exit_code": 1},
|
|
}
|
|
|
|
|
|
def _perms_ok_result() -> dict:
|
|
return {
|
|
"isError": False,
|
|
"content": [{"type": "text", "text": "ok"}],
|
|
"structuredContent": {
|
|
"accessibility": True,
|
|
"screen_recording": True,
|
|
"screen_recording_capturable": True,
|
|
},
|
|
}
|
|
|
|
|
|
def _list_apps_ok_result() -> dict:
|
|
return {
|
|
"isError": False,
|
|
"content": [{"type": "text", "text": "Found 1 app"}],
|
|
"structuredContent": {
|
|
"apps": [{"name": "Finder", "pid": 1, "running": True}],
|
|
},
|
|
}
|
|
|
|
|
|
class TestHealthReportFallback:
|
|
"""cua-driver 0.10 marks health_report risk-unclassified → isError.
|
|
|
|
Doctor must NOT treat structuredContent={exit_code:1} as a real report
|
|
(that produced '• cua-driver ? on ? — ?'). It synthesizes schema_version=1
|
|
via check_permissions / list_apps / CLI --version instead.
|
|
"""
|
|
|
|
|
|
def test_extract_raises_health_report_unavailable_on_isError(self):
|
|
from tools.computer_use import doctor
|
|
|
|
with __import__("pytest").raises(doctor.HealthReportUnavailable) as ei:
|
|
doctor._extract_health_report_from_result(_unclassified_health_result())
|
|
assert "Permission denied" in str(ei.value) or "unclassified" in str(ei.value).lower() or "risk" in str(ei.value).lower()
|
|
|
|
|
|
# ── binary identity (CLI --version vs health_report) ───────────────────────
|
|
|
|
|
|
class TestDoctorVersionIdentity:
|
|
def test_header_prefers_cli_version_on_mismatch(self):
|
|
"""Windows has been observed reporting 0.8.3 via health_report while
|
|
the resolved binary is 0.12.6 — doctor must surface the real version."""
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _ok_report()}},
|
|
)
|
|
# _ok_report claims 0.5.8; CLI says 0.12.6
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch.object(doctor, "_read_cli_version", return_value="cua-driver 0.12.6"), \
|
|
patch("sys.stdout", new_callable=StringIO) as out:
|
|
code = doctor.run_doctor()
|
|
assert code == 0
|
|
text = out.getvalue()
|
|
assert "0.12.6" in text
|
|
assert "version mismatch" in text.lower()
|
|
assert "0.5.8" in text # health_report value still shown
|
|
|
|
|
|
def test_matching_versions_no_mismatch_flag(self):
|
|
from tools.computer_use import doctor
|
|
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": _ok_report()}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), \
|
|
patch("subprocess.Popen", return_value=proc), \
|
|
patch.object(doctor, "_read_cli_version", return_value="cua-driver 0.5.8"), \
|
|
patch("sys.stdout", new_callable=StringIO) as out:
|
|
code = doctor.run_doctor(json_output=True)
|
|
assert code == 0
|
|
payload = json.loads(out.getvalue())
|
|
assert payload["hermes_identity"]["version_mismatch"] is False
|
|
|
|
|
|
def test_failed_tcc_row_from_health_report_names_the_stale_row_reset_for_that_service():
|
|
"""A failed ``tcc_*`` row from the driver's own health_report (0.22+, not only the 0.10 fallback probes) must
|
|
carry the stale-grant recovery for its own TCC service, because System Settings can show the toggle ON while
|
|
the daemon is denied (trycua/cua#3170)."""
|
|
from tools.computer_use import doctor
|
|
|
|
report = _ok_report()
|
|
report["checks"] = [{"name": "tcc_screen_recording", "status": "fail", "message": "Screen Recording is not granted.",
|
|
"hint": "Grant it in System Settings."},
|
|
{"name": "tcc_accessibility", "status": "pass", "message": "granted"}]
|
|
proc = _fake_proc_with_responses(
|
|
{"jsonrpc": "2.0", "id": 1, "result": {}},
|
|
{"jsonrpc": "2.0", "id": 2, "result": {"structuredContent": report}},
|
|
)
|
|
with _pm_driver("/fake/cua-driver"), patch("subprocess.Popen", return_value=proc), \
|
|
patch("sys.stdout", new_callable=StringIO) as out:
|
|
doctor.run_doctor(json_output=True)
|
|
checks = {c["name"]: c for c in json.loads(out.getvalue())["checks"]}
|
|
|
|
assert checks["tcc_screen_recording"]["hint"].startswith("Grant it in System Settings.")
|
|
assert "tccutil reset ScreenCapture com.trycua.driver" in checks["tcc_screen_recording"]["hint"]
|
|
assert "reset Accessibility" not in checks["tcc_screen_recording"]["hint"]
|
|
assert "tccutil" not in checks["tcc_accessibility"].get("hint", "")
|