/
statistics.py
450 lines (379 loc) · 15.1 KB
/
statistics.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
"""Statistics helper."""
from __future__ import annotations
from collections import defaultdict
from datetime import datetime, timedelta
from itertools import groupby
import logging
from typing import TYPE_CHECKING, Any, Callable
from sqlalchemy import bindparam
from sqlalchemy.ext import baked
from sqlalchemy.orm.scoping import scoped_session
from homeassistant.const import (
PRESSURE_PA,
TEMP_CELSIUS,
VOLUME_CUBIC_FEET,
VOLUME_CUBIC_METERS,
)
from homeassistant.core import Event, HomeAssistant, callback
from homeassistant.helpers import entity_registry
import homeassistant.util.dt as dt_util
import homeassistant.util.pressure as pressure_util
import homeassistant.util.temperature as temperature_util
from homeassistant.util.unit_system import UnitSystem
import homeassistant.util.volume as volume_util
from .const import DOMAIN
from .models import (
StatisticMetaData,
Statistics,
StatisticsMeta,
StatisticsRuns,
process_timestamp_to_utc_isoformat,
)
from .util import execute, retryable_database_job, session_scope
if TYPE_CHECKING:
from . import Recorder
QUERY_STATISTICS = [
Statistics.metadata_id,
Statistics.start,
Statistics.mean,
Statistics.min,
Statistics.max,
Statistics.last_reset,
Statistics.state,
Statistics.sum,
]
QUERY_STATISTIC_META = [
StatisticsMeta.id,
StatisticsMeta.statistic_id,
StatisticsMeta.unit_of_measurement,
StatisticsMeta.has_mean,
StatisticsMeta.has_sum,
]
QUERY_STATISTIC_META_ID = [
StatisticsMeta.id,
StatisticsMeta.statistic_id,
]
STATISTICS_BAKERY = "recorder_statistics_bakery"
STATISTICS_META_BAKERY = "recorder_statistics_bakery"
# Convert pressure and temperature statistics from the native unit used for statistics
# to the units configured by the user
UNIT_CONVERSIONS = {
PRESSURE_PA: lambda x, units: pressure_util.convert(
x, PRESSURE_PA, units.pressure_unit
)
if x is not None
else None,
TEMP_CELSIUS: lambda x, units: temperature_util.convert(
x, TEMP_CELSIUS, units.temperature_unit
)
if x is not None
else None,
VOLUME_CUBIC_METERS: lambda x, units: volume_util.convert(
x, VOLUME_CUBIC_METERS, _configured_unit(VOLUME_CUBIC_METERS, units)
)
if x is not None
else None,
}
_LOGGER = logging.getLogger(__name__)
def async_setup(hass: HomeAssistant) -> None:
"""Set up the history hooks."""
hass.data[STATISTICS_BAKERY] = baked.bakery()
hass.data[STATISTICS_META_BAKERY] = baked.bakery()
def entity_id_changed(event: Event) -> None:
"""Handle entity_id changed."""
old_entity_id = event.data["old_entity_id"]
entity_id = event.data["entity_id"]
with session_scope(hass=hass) as session:
session.query(StatisticsMeta).filter(
StatisticsMeta.statistic_id == old_entity_id
and StatisticsMeta.source == DOMAIN
).update({StatisticsMeta.statistic_id: entity_id})
@callback
def entity_registry_changed_filter(event: Event) -> bool:
"""Handle entity_id changed filter."""
if event.data["action"] != "update" or "old_entity_id" not in event.data:
return False
return True
if hass.is_running:
hass.bus.async_listen(
entity_registry.EVENT_ENTITY_REGISTRY_UPDATED,
entity_id_changed,
event_filter=entity_registry_changed_filter,
)
def get_start_time() -> datetime:
"""Return start time."""
last_hour = dt_util.utcnow() - timedelta(hours=1)
start = last_hour.replace(minute=0, second=0, microsecond=0)
return start
def _get_metadata_ids(
hass: HomeAssistant, session: scoped_session, statistic_ids: list[str]
) -> list[str]:
"""Resolve metadata_id for a list of statistic_ids."""
baked_query = hass.data[STATISTICS_META_BAKERY](
lambda session: session.query(*QUERY_STATISTIC_META_ID)
)
baked_query += lambda q: q.filter(
StatisticsMeta.statistic_id.in_(bindparam("statistic_ids"))
)
result = execute(baked_query(session).params(statistic_ids=statistic_ids))
return [id for id, _ in result] if result else []
def _update_or_add_metadata(
hass: HomeAssistant,
session: scoped_session,
statistic_id: str,
new_metadata: StatisticMetaData,
) -> str:
"""Get metadata_id for a statistic_id, add if it doesn't exist."""
old_metadata_dict = _get_metadata(hass, session, [statistic_id], None)
if not old_metadata_dict:
unit = new_metadata["unit_of_measurement"]
has_mean = new_metadata["has_mean"]
has_sum = new_metadata["has_sum"]
session.add(
StatisticsMeta.from_meta(DOMAIN, statistic_id, unit, has_mean, has_sum)
)
metadata_ids = _get_metadata_ids(hass, session, [statistic_id])
_LOGGER.debug(
"Added new statistics metadata for %s, new_metadata: %s",
statistic_id,
new_metadata,
)
return metadata_ids[0]
metadata_id, old_metadata = next(iter(old_metadata_dict.items()))
if (
old_metadata["has_mean"] != new_metadata["has_mean"]
or old_metadata["has_sum"] != new_metadata["has_sum"]
or old_metadata["unit_of_measurement"] != new_metadata["unit_of_measurement"]
):
session.query(StatisticsMeta).filter_by(statistic_id=statistic_id).update(
{
StatisticsMeta.has_mean: new_metadata["has_mean"],
StatisticsMeta.has_sum: new_metadata["has_sum"],
StatisticsMeta.unit_of_measurement: new_metadata["unit_of_measurement"],
},
synchronize_session=False,
)
_LOGGER.debug(
"Updated statistics metadata for %s, old_metadata: %s, new_metadata: %s",
statistic_id,
old_metadata,
new_metadata,
)
return metadata_id
@retryable_database_job("statistics")
def compile_statistics(instance: Recorder, start: datetime) -> bool:
"""Compile statistics."""
start = dt_util.as_utc(start)
end = start + timedelta(hours=1)
with session_scope(session=instance.get_session()) as session: # type: ignore
if session.query(StatisticsRuns).filter_by(start=start).first():
_LOGGER.debug("Statistics already compiled for %s-%s", start, end)
return True
_LOGGER.debug("Compiling statistics for %s-%s", start, end)
platform_stats = []
for domain, platform in instance.hass.data[DOMAIN].items():
if not hasattr(platform, "compile_statistics"):
continue
platform_stats.append(platform.compile_statistics(instance.hass, start, end))
_LOGGER.debug(
"Statistics for %s during %s-%s: %s", domain, start, end, platform_stats[-1]
)
with session_scope(session=instance.get_session()) as session: # type: ignore
for stats in platform_stats:
for entity_id, stat in stats.items():
metadata_id = _update_or_add_metadata(
instance.hass, session, entity_id, stat["meta"]
)
session.add(Statistics.from_stats(metadata_id, start, stat["stat"]))
session.add(StatisticsRuns(start=start))
return True
def _get_metadata(
hass: HomeAssistant,
session: scoped_session,
statistic_ids: list[str] | None,
statistic_type: str | None,
) -> dict[str, StatisticMetaData]:
"""Fetch meta data."""
def _meta(metas: list, wanted_metadata_id: str) -> StatisticMetaData | None:
meta: StatisticMetaData | None = None
for metadata_id, statistic_id, unit, has_mean, has_sum in metas:
if metadata_id == wanted_metadata_id:
meta = {
"statistic_id": statistic_id,
"unit_of_measurement": unit,
"has_mean": has_mean,
"has_sum": has_sum,
}
return meta
baked_query = hass.data[STATISTICS_META_BAKERY](
lambda session: session.query(*QUERY_STATISTIC_META)
)
if statistic_ids is not None:
baked_query += lambda q: q.filter(
StatisticsMeta.statistic_id.in_(bindparam("statistic_ids"))
)
if statistic_type == "mean":
baked_query += lambda q: q.filter(StatisticsMeta.has_mean.isnot(False))
elif statistic_type == "sum":
baked_query += lambda q: q.filter(StatisticsMeta.has_sum.isnot(False))
elif statistic_type is not None:
return {}
result = execute(baked_query(session).params(statistic_ids=statistic_ids))
if not result:
return {}
metadata_ids = [metadata[0] for metadata in result]
metadata: dict[str, StatisticMetaData] = {}
for _id in metadata_ids:
meta = _meta(result, _id)
if meta:
metadata[_id] = meta
return metadata
def get_metadata(
hass: HomeAssistant,
statistic_id: str,
) -> StatisticMetaData | None:
"""Return metadata for a statistic_id."""
statistic_ids = [statistic_id]
with session_scope(hass=hass) as session:
metadata_ids = _get_metadata_ids(hass, session, [statistic_id])
if not metadata_ids:
return None
return _get_metadata(hass, session, statistic_ids, None).get(metadata_ids[0])
def _configured_unit(unit: str, units: UnitSystem) -> str:
"""Return the pressure and temperature units configured by the user."""
if unit == PRESSURE_PA:
return units.pressure_unit
if unit == TEMP_CELSIUS:
return units.temperature_unit
if unit == VOLUME_CUBIC_METERS:
if units.is_metric:
return VOLUME_CUBIC_METERS
return VOLUME_CUBIC_FEET
return unit
def list_statistic_ids(
hass: HomeAssistant, statistic_type: str | None = None
) -> list[StatisticMetaData | None]:
"""Return statistic_ids and meta data."""
units = hass.config.units
statistic_ids = {}
with session_scope(hass=hass) as session:
metadata = _get_metadata(hass, session, None, statistic_type)
for meta in metadata.values():
unit = meta["unit_of_measurement"]
if unit is not None:
unit = _configured_unit(unit, units)
meta["unit_of_measurement"] = unit
statistic_ids = {
meta["statistic_id"]: meta["unit_of_measurement"]
for meta in metadata.values()
}
for platform in hass.data[DOMAIN].values():
if not hasattr(platform, "list_statistic_ids"):
continue
platform_statistic_ids = platform.list_statistic_ids(hass, statistic_type)
for statistic_id, unit in platform_statistic_ids.items():
if unit is not None:
unit = _configured_unit(unit, units)
platform_statistic_ids[statistic_id] = unit
statistic_ids = {**statistic_ids, **platform_statistic_ids}
return [
{"statistic_id": _id, "unit_of_measurement": unit}
for _id, unit in statistic_ids.items()
]
def statistics_during_period(
hass: HomeAssistant,
start_time: datetime,
end_time: datetime | None = None,
statistic_ids: list[str] | None = None,
) -> dict[str, list[dict[str, str]]]:
"""Return states changes during UTC period start_time - end_time."""
metadata = None
with session_scope(hass=hass) as session:
metadata = _get_metadata(hass, session, statistic_ids, None)
if not metadata:
return {}
baked_query = hass.data[STATISTICS_BAKERY](
lambda session: session.query(*QUERY_STATISTICS)
)
baked_query += lambda q: q.filter(Statistics.start >= bindparam("start_time"))
if end_time is not None:
baked_query += lambda q: q.filter(Statistics.start < bindparam("end_time"))
metadata_ids = None
if statistic_ids is not None:
baked_query += lambda q: q.filter(
Statistics.metadata_id.in_(bindparam("metadata_ids"))
)
metadata_ids = list(metadata.keys())
baked_query += lambda q: q.order_by(Statistics.metadata_id, Statistics.start)
stats = execute(
baked_query(session).params(
start_time=start_time, end_time=end_time, metadata_ids=metadata_ids
)
)
if not stats:
return {}
return _sorted_statistics_to_dict(hass, stats, statistic_ids, metadata)
def get_last_statistics(
hass: HomeAssistant, number_of_stats: int, statistic_id: str
) -> dict[str, list[dict]]:
"""Return the last number_of_stats statistics for a statistic_id."""
statistic_ids = [statistic_id]
with session_scope(hass=hass) as session:
metadata = _get_metadata(hass, session, statistic_ids, None)
if not metadata:
return {}
baked_query = hass.data[STATISTICS_BAKERY](
lambda session: session.query(*QUERY_STATISTICS)
)
baked_query += lambda q: q.filter_by(metadata_id=bindparam("metadata_id"))
metadata_id = next(iter(metadata.keys()))
baked_query += lambda q: q.order_by(
Statistics.metadata_id, Statistics.start.desc()
)
baked_query += lambda q: q.limit(bindparam("number_of_stats"))
stats = execute(
baked_query(session).params(
number_of_stats=number_of_stats, metadata_id=metadata_id
)
)
if not stats:
return {}
return _sorted_statistics_to_dict(hass, stats, statistic_ids, metadata)
def _sorted_statistics_to_dict(
hass: HomeAssistant,
stats: list,
statistic_ids: list[str] | None,
metadata: dict[str, StatisticMetaData],
) -> dict[str, list[dict]]:
"""Convert SQL results into JSON friendly data structure."""
result: dict = defaultdict(list)
units = hass.config.units
# Set all statistic IDs to empty lists in result set to maintain the order
if statistic_ids is not None:
for stat_id in statistic_ids:
result[stat_id] = []
# Called in a tight loop so cache the function here
_process_timestamp_to_utc_isoformat = process_timestamp_to_utc_isoformat
# Append all statistic entries, and do unit conversion
for meta_id, group in groupby(stats, lambda stat: stat.metadata_id): # type: ignore
unit = metadata[meta_id]["unit_of_measurement"]
statistic_id = metadata[meta_id]["statistic_id"]
convert: Callable[[Any, Any], float | None] = UNIT_CONVERSIONS.get(
unit, lambda x, units: x # type: ignore
)
ent_results = result[meta_id]
ent_results.extend(
{
"statistic_id": statistic_id,
"start": _process_timestamp_to_utc_isoformat(db_state.start),
"mean": convert(db_state.mean, units),
"min": convert(db_state.min, units),
"max": convert(db_state.max, units),
"last_reset": _process_timestamp_to_utc_isoformat(db_state.last_reset),
"state": convert(db_state.state, units),
"sum": convert(db_state.sum, units),
}
for db_state in group
)
# Filter out the empty lists if some states had 0 results.
return {metadata[key]["statistic_id"]: val for key, val in result.items() if val}