diet-guard/diet_guard/tests/conftest.py
Krzysztof kuhy Rudnicki 8871408b5b perf(cli): publish off-thread so logging a meal returns immediately
`diet-guard log ...` blocked for ~15.5s after every meal. The publish itself
has to happen -- until it lands the phone still believes the slot is unlogged
and nags for it -- but only ~2.5s of that is the 293KB push the phone needs;
the rest is pulling peers' banks, which publishing does not require.

The local write is already durable before the publish starts, so the worst
case is a late publish, which is the trade `_sync_events` already documents
for the event-driven model. Moving it to a worker thread lets the caller
print and return at once. Measured: 15.5s -> 2ms before the prompt comes back.

The thread is deliberately non-daemon: the CLI's output is already flushed,
so the process lingers only as long as the push, and the publish is never
killed halfway. A failure still surfaces, via a callback the caller supplies,
just after its own output rather than in the middle of it.

Deliberately NOT applied to the gate. `diet-guard-gate.service` is
`Type=oneshot`, so systemd reaps the unit the moment the main thread exits
and a background publish would be killed mid-push. The gate keeps calling
`publish_after_log` and waiting, which is correct there -- its publish
already runs after the window is gone, where nobody is waiting on it.

`conftest._isolate_state` follows the rename to
`publish_after_log_detached`; without that every meal-logging test would
reach the network through a newly unpatched seam. Verified with
`scripts/check_patch_targets.py` and by timing the real command against a
temporary data dir, never the live log.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LJ5YnA3CERitQ57HjMu2Qz
2026-08-21 13:22:26 +02:00

169 lines
5.6 KiB
Python

"""Shared fixtures for diet_guard tests.
Three safety nets run for every test:
* ``_isolate_state`` redirects the food log, sealed budget, gate lock, and
sync token into ``tmp_path`` so a test can never read or clobber the real
``~/.local/share`` or ``~/.config/diet_guard``.
* ``_block_real_tk`` swaps ``tk`` and the ``GateRoot`` window class inside
``_gatelock`` for mocks, so no test can open a real fullscreen window or grab
the keyboard even if it forgets to.
* ``_block_real_vt`` makes ``gatelock``'s VT-switch disable a no-op, so a
prod-mode (``demo_mode=False``) gate built in a test never runs a real
``setxkbmap`` against the live X session.
The ``gate`` fixture and its supporting fakes (``FakeEntry``, ``_FAKE_TK``, ...)
build a demo :class:`~diet_guard._gatelock.MealGate` whose widgets
are functional in-memory stand-ins, shared by ``test_gatelock.py`` and
``test_gatelock_mealflow.py``.
"""
from __future__ import annotations
from typing import TYPE_CHECKING
from unittest.mock import patch
import pytest
from diet_guard._estimator import Nutrition
# Importing these fixtures IS their registration: pytest discovers fixtures
# defined *or imported* in a conftest, and ``pytest_plugins`` is an error in a
# non-root conftest. They look unused and are not -- do not "tidy" them away.
from diet_guard.tests._gate_fixtures import (
_GATE_TK_MODULES,
FAKE_OUTPUTS,
TWO_OUTPUTS,
_block_real_tk,
_block_real_vt,
_hermetic_gatelock,
_hmac_key,
dual_output,
fake_tk,
gate,
)
from diet_guard.tests._tk_fakes import (
_FAKE_TK,
_FAKE_TTK,
FakeCanvas,
FakeEntry,
FakeListbox,
FakeNotebook,
FakeRadiobutton,
FakeScrollbar,
FakeStyle,
FakeText,
FakeVar,
FakeWidget,
_FakeTclError,
)
# Re-exported: the fake widgets moved into the package so this file stays
# under the 250-line cap, but tests import them from conftest by name.
__all__ = [
"FAKE_OUTPUTS",
"TWO_OUTPUTS",
"_FAKE_TK",
"_FAKE_TTK",
# The autouse fixtures: importing them here is what registers them with
# pytest, and naming them here is what stops a linter deleting the import.
"_GATE_TK_MODULES",
"FakeCanvas",
"FakeEntry",
"FakeListbox",
"FakeNotebook",
"FakeRadiobutton",
"FakeScrollbar",
"FakeStyle",
"FakeText",
"FakeVar",
"FakeWidget",
"_FakeTclError",
"_block_real_tk",
"_block_real_vt",
"_hermetic_gatelock",
"_hmac_key",
"dual_output",
"fake_tk",
"gate",
]
if TYPE_CHECKING:
from collections.abc import Iterator
from pathlib import Path
@pytest.fixture(autouse=True)
def _isolate_state(tmp_path: Path) -> Iterator[None]:
"""Redirect all on-disk diet_guard state into a temp dir."""
with (
patch(
"diet_guard._budget.BUDGET_FILE",
tmp_path / ".budget",
),
patch(
"diet_guard._budget_history.BUDGET_HISTORY_FILE",
tmp_path / ".budget_history",
),
patch(
"diet_guard._meal_schedule_store.MEAL_SCHEDULE_FILE",
tmp_path / ".meal_schedule",
),
patch(
"diet_guard._state.FOOD_LOG_FILE",
tmp_path / "food_log.json",
),
patch(
"diet_guard._foodbank.FOOD_BANK_FILE",
tmp_path / "food_bank.json",
),
patch(
"diet_guard._foodbank_manual.MANUAL_BANK_FILE",
tmp_path / "food_bank_manual.json",
),
patch(
"diet_guard._gatelock_lockfile.GATE_LOCK_FILE",
tmp_path / ".gate.lock",
),
patch(
"diet_guard._sync_client.SYNC_TOKEN_FILE",
tmp_path / "sync_token",
),
patch(
"diet_guard._sync.SYNC_STATE_FILE",
tmp_path / "sync_state.json",
),
# Without this a test that syncs mints a uuid into the REAL
# ~/.local/share/diet_guard/.device_id, and this machine's live sync
# identity is decided by whichever test happened to run first.
patch(
"diet_guard._device.SYNC_DEVICE_ID_FILE",
tmp_path / ".device_id",
),
# `run_sync` reads this to decide whether to build a Firebase-primary
# mirror. On a developer machine the real file exists, so without this
# every sync test would sign in and push to the live database.
patch(
"diet_guard._sync_client.CONFIG_FILE",
tmp_path / "nonexistent-firebase.json",
),
# Logging a meal now publishes immediately (`_sync_events`), so every
# test that logs one would otherwise open a real connection through a
# path that has nothing to do with what it is asserting. Patched at the
# *call sites* rather than on `_sync_events`, so a test that imports
# `publish_after_log` directly still exercises the real helper.
patch("diet_guard._cli_log.publish_after_log_detached", return_value=None),
patch("diet_guard._cli_gate.publish_after_log", return_value=None),
patch("diet_guard._mcp.publish_after_log", return_value=None),
# The gate's pre-lock refresh is a *second*, independent network
# entry point (narrow peer-log pull, not the full tick). Without
# this every test that reaches `_should_lock` hits the real remote.
patch("diet_guard._cli_gate.pull_peer_logs", return_value=None),
):
yield
def _nutrition(kcal: float = 100, grams: float = 100) -> Nutrition:
"""A simple reference nutrition for driving the gate form."""
return Nutrition(kcal, 10, 20, 5, grams, "food bank")