Files
Fenris/tests/test_day_aggregates.py
T
xavierk 7d219c4697 feat(hour/day derivation): hour classification, monitoring periods, day aggregates, pruning
Hour classification (PR-4):
- Powered-off: POH delta < 90% of wall-clock span
- Active: DUW delta >= 256 MiB
- Idle: powered on + sampled + below active threshold
- Unknown: unsampled without POH evidence
- Four splits sum to exactly wall_clock_seconds
- Disabled time is never an hour state

Monitoring periods (FL-8):
- ensure_period_open: opens period at run moment if none exists
- close_period: closes with end cause
- is_inside_period: checks timestamp against period bounds
- Never backdated; wall-clock outside periods excluded from denominator

Day aggregates (ST-4, PR-5):
- Derived monotonically from hour rows
- UTC-bounded; no 23/25-hour days
- Coverage: known seconds / period wall-clock
- Gap hours inside periods contribute unknown seconds
- Hours outside periods excluded entirely
- No absent hour interpolated/estimated/fabricated (FL-3)

Raw sample pruning (ST-5):
- Prunes samples older than 14 days
- Hour observations and day aggregates retained indefinitely

Closes #22
2026-09-01 22:58:03 +05:30

235 lines
9.2 KiB
Python

"""Day aggregate tests.
From spec §5.4, §3.3, ST-4, ST-5:
- One row per UTC day, derived monotonically from hour rows
- No 23/25-hour days (UTC-bounded, DST never applies)
- Coverage: share of wall-clock seconds inside monitoring periods
whose classification is known
- No absent hour ever interpolated/estimated/fabricated (FL-3)
"""
import sqlite3
import sys
from datetime import datetime, timezone, timedelta
from pathlib import Path
import pytest
sys.path.insert(0, str(Path(__file__).parent.parent / "src"))
from fenris.store import init_store
from fenris.monitoring_periods import ensure_period_open, close_period
from fenris.day_aggregate import derive_day, derive_all_days, DayAggregate
from fenris.hour_classify import HourSplit, ACTIVE_THRESHOLD_BYTES
@pytest.fixture
def store_conn(tmp_path: Path):
db_path = tmp_path / "test.db"
conn = init_store(db_path)
yield conn
conn.close()
def _insert_hour(conn, hour_iso, split, bytes_written_delta=0, bytes_read_delta=0,
sample_count=1, coverage=1.0):
conn.execute(
"INSERT INTO hour_observations "
"(hour, active_seconds, idle_seconds, powered_off_seconds, unknown_seconds, "
" bytes_written_delta, bytes_read_delta, sample_count, coverage) "
"VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)",
(
hour_iso,
split.seconds_active,
split.seconds_idle,
split.seconds_powered_off,
split.seconds_unknown,
bytes_written_delta,
bytes_read_delta,
sample_count,
coverage,
),
)
conn.commit()
class TestDeriveDay:
"""Spec §5.4: One row per UTC day, derived monotonically from hour rows."""
def test_single_hour_day(self, store_conn):
t_start = datetime(2026, 9, 1, 12, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 13, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
hour = "2026-09-01T12:00:00+00:00"
split = HourSplit(seconds_active=3600, seconds_idle=0, seconds_powered_off=0, seconds_unknown=0)
_insert_hour(store_conn, hour, split, bytes_written_delta=1024*1024*100)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.day == "2026-09-01"
assert day.seconds_active == 3600
assert day.seconds_idle == 0
assert day.seconds_powered_off == 0
assert day.bytes_written_delta == 1024*1024*100
assert day.sample_count == 1
def test_multiple_hours(self, store_conn):
t_start = datetime(2026, 9, 1, 10, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 13, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
hours = [
("2026-09-01T10:00:00+00:00", HourSplit(1800, 1800, 0, 0), 50*1024*1024),
("2026-09-01T11:00:00+00:00", HourSplit(3600, 0, 0, 0), 200*1024*1024),
("2026-09-01T12:00:00+00:00", HourSplit(0, 0, 3600, 0), 0),
]
for h, s, bw in hours:
_insert_hour(store_conn, h, s, bytes_written_delta=bw)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.seconds_active == 1800 + 3600 + 0
assert day.seconds_idle == 1800 + 0 + 0
assert day.seconds_powered_off == 0 + 0 + 3600
assert day.bytes_written_delta == 50*1024*1024 + 200*1024*1024 + 0
assert day.sample_count == 3
def test_no_hours_returns_none(self, store_conn):
day = derive_day(store_conn, "2026-09-01")
assert day is None
class TestUtcBounded:
"""Spec §3.3, ST-4: Hours and days are UTC-bounded. No 23/25-hour days."""
def test_day_keys_are_utc_date_strings(self, store_conn):
t_start = datetime(2026, 9, 1, 12, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 13, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
hour = "2026-09-01T12:00:00+00:00"
split = HourSplit(0, 0, 3600, 0)
_insert_hour(store_conn, hour, split)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.day == "2026-09-01"
def test_cross_midnight_hours_produce_two_days(self, store_conn):
# Period spans both days
t_start = datetime(2026, 9, 1, 23, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 2, 1, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
hours = [
("2026-09-01T23:00:00+00:00", HourSplit(3600, 0, 0, 0), 100),
("2026-09-02T00:00:00+00:00", HourSplit(0, 0, 3600, 0), 0),
]
for h, s, bw in hours:
_insert_hour(store_conn, h, s, bytes_written_delta=bw)
days = derive_all_days(store_conn)
day_keys = [d.day for d in days]
assert "2026-09-01" in day_keys
assert "2026-09-02" in day_keys
assert len(days) == 2
class TestCoverage:
"""Spec §5.2, §5.3, PR-5: Coverage is known share of wall-clock seconds
inside monitoring periods."""
def test_full_coverage_all_known(self, store_conn):
"""All hours in period classified -> coverage = 1.0."""
t_start = datetime(2026, 9, 1, 0, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 3, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
hours = [
("2026-09-01T00:00:00+00:00", HourSplit(3600, 0, 0, 0), 100),
("2026-09-01T01:00:00+00:00", HourSplit(0, 3600, 0, 0), 50),
("2026-09-01T02:00:00+00:00", HourSplit(0, 0, 3600, 0), 0),
]
for h, s, bw in hours:
_insert_hour(store_conn, h, s, bytes_written_delta=bw)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.coverage == pytest.approx(1.0)
def test_gap_reduces_coverage(self, store_conn):
"""Missing hour -> unknown seconds reduce coverage."""
t_start = datetime(2026, 9, 1, 0, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 3, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
_insert_hour(store_conn, "2026-09-01T00:00:00+00:00",
HourSplit(3600, 0, 0, 0), bytes_written_delta=100)
_insert_hour(store_conn, "2026-09-01T02:00:00+00:00",
HourSplit(0, 3600, 0, 0), bytes_written_delta=50)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.coverage == pytest.approx(7200 / 10800)
assert day.seconds_unknown == 3600
def test_outside_period_excluded(self, store_conn):
"""Hours outside any monitoring period excluded from denominator."""
t_start = datetime(2026, 9, 1, 1, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 2, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
# Hour 01 is inside the period
_insert_hour(store_conn, "2026-09-01T01:00:00+00:00",
HourSplit(3600, 0, 0, 0), bytes_written_delta=100)
# Hour 02 is outside (period ends at 02:00)
_insert_hour(store_conn, "2026-09-01T02:00:00+00:00",
HourSplit(0, 3600, 0, 0), bytes_written_delta=50)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.coverage == pytest.approx(1.0)
def test_unknown_in_period_reduces_coverage(self, store_conn):
"""Unknown seconds inside period count in denominator but not numerator."""
t_start = datetime(2026, 9, 1, 0, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 1, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
_insert_hour(store_conn, "2026-09-01T00:00:00+00:00",
HourSplit(1800, 0, 0, 1800), bytes_written_delta=100)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.coverage == pytest.approx(0.5)
assert day.seconds_unknown == 1800
class TestNoFabrication:
"""Spec §5.3, FL-3: No absent hour is ever interpolated/estimated/fabricated."""
def test_missing_hours_stay_unknown(self, store_conn):
"""Gap hours are never filled in — they remain as unknown seconds."""
t_start = datetime(2026, 9, 1, 0, 0, 0, tzinfo=timezone.utc)
t_end = datetime(2026, 9, 1, 3, 0, 0, tzinfo=timezone.utc)
ensure_period_open(store_conn, t_start)
close_period(store_conn, t_end, "user_disabled")
# Only hour 02 — hours 00 and 01 are gaps
_insert_hour(store_conn, "2026-09-01T02:00:00+00:00",
HourSplit(3600, 0, 0, 0), bytes_written_delta=100)
day = derive_day(store_conn, "2026-09-01")
assert day is not None
assert day.seconds_unknown == 2 * 3600
assert day.seconds_active == 3600
assert day.sample_count == 1