Files
hotspring/custom_components/hotspring/coordinator.py
T
dfritzandClaude Fable 5.2 298df5864a Make spa entities unavailable when the spa is not reporting
A module that is online but cut off from the control box keeps serving and
publishing a full document of blank and default values. Read literally that
is an idle spa: no temperatures, salt level 0, lights and jets off, and
controls that accept a value and snap back.

- The spa counts as reporting only while telemetry is fresh (something
  arrived in the last five minutes) and carries control-box data. Otherwise
  every entity is unavailable, and the reason is logged once per change.
- The entry still loads in that state, so entities return on their own when
  the spa reports again.
- A watchdog catches silence between updates, and reloads the entry when the
  broker connection stays down for three minutes. The broker is the only
  control path and its credentials rotate; paho reconnects with the old ones,
  so only a fresh sign-in recovers it.
- Version 0.2.0.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-14 21:16:20 -06:00

237 lines
9.0 KiB
Python

"""Coordinator: owns the cloud session + MQTT connection for one spa and exposes
the latest telemetry to entities.
Read path:
* Default — subscribe to the spa's MQTT topics (cloud push).
* If ``local_status_url`` is configured — poll the dongle's local ``/status``
endpoint instead, keeping monitoring on-LAN and off the vendor cloud. The MQTT
connection is still opened (control is cloud-only) but not subscribed.
* If that local endpoint does not answer (the dongle's DHCP address changed, or
it is off the LAN) — read from the cloud until it does. A dead local URL must
never take the whole integration, and with it control, down.
"""
from __future__ import annotations
import logging
import os
import time
from datetime import timedelta
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant, callback
from homeassistant.helpers.event import async_track_time_interval
from homeassistant.helpers.update_coordinator import DataUpdateCoordinator
from .api import HotSpringApiError, HotSpringCloud, HotSpringSpa, Spa, fetch_local_status
from .constants import (
BROKER_DOWN_TICKS_BEFORE_RELOAD,
CA_FILE,
CERT_FILE,
CONF_CERT_DIR,
CONF_EMAIL,
CONF_LOCAL_STATUS_URL,
CONF_PASSWORD,
DEFAULT_CERT_DIR,
DOMAIN,
KEY_FILE,
LOCAL_FAILURES_BEFORE_CLOUD,
LOCAL_POLL_INTERVAL,
STALE_AFTER,
WATCHDOG_INTERVAL,
)
_LOGGER = logging.getLogger(__name__)
class HotSpringCoordinator(DataUpdateCoordinator[dict]):
def __init__(self, hass: HomeAssistant, entry: ConfigEntry):
super().__init__(hass, _LOGGER, name=DOMAIN)
self.entry = entry
self.cloud: HotSpringCloud | None = None
self.spa: HotSpringSpa | None = None
self.spa_info: Spa | None = None
self._local_url: str | None = entry.data.get(CONF_LOCAL_STATUS_URL) or None
self._unsub_poll = None
# True while telemetry comes from the local URL; False while the cloud
# subscription is standing in for it.
self._local_ok = False
self._local_failures = 0
self._unsub_watchdog = None
self._last_status: float | None = None # monotonic time of the last telemetry
self._reporting: bool | None = None # last health state that was logged
self._broker_down_ticks = 0
def _cert_paths(self) -> tuple[str, str, str]:
d = self.entry.data.get(CONF_CERT_DIR, DEFAULT_CERT_DIR)
return (
os.path.join(d, CERT_FILE),
os.path.join(d, KEY_FILE),
os.path.join(d, CA_FILE),
)
async def async_setup(self) -> None:
cert, key, ca = self._cert_paths()
cloud = HotSpringCloud(
self.entry.data[CONF_EMAIL], self.entry.data[CONF_PASSWORD], cert, key, ca
)
local = self._local_url
def _blocking_connect() -> HotSpringSpa:
cloud.login()
spas = cloud.get_spas()
if not spas:
raise RuntimeError("account has no spas")
self.spa_info = spas[0]
creds = cloud.fetch_mqtt_credentials()
spa = HotSpringSpa(self.spa_info.root_topic, creds, on_status=self._on_status)
status = None
if local:
try:
status = fetch_local_status(local)
except HotSpringApiError as err:
_LOGGER.warning(
"Spa local status URL %s does not answer (%s); reading "
"telemetry from the cloud until it does",
local,
err,
)
self._local_ok = status is not None
# Local read source -> connect for control only (no MQTT subscribe).
spa.connect(subscribe=status is None)
if status is not None:
spa.ingest_status(status)
else:
spa.wait_for_state()
return spa
self.cloud = cloud
self.spa = await self.hass.async_add_executor_job(_blocking_connect)
self.async_set_updated_data(self.spa.merged)
self._unsub_watchdog = async_track_time_interval(
self.hass, self._async_watchdog, timedelta(seconds=WATCHDOG_INTERVAL)
)
if local:
self._unsub_poll = async_track_time_interval(
self.hass, self._async_poll_local, timedelta(seconds=LOCAL_POLL_INTERVAL)
)
async def _async_poll_local(self, _now) -> None:
if self.spa is None or not self._local_url:
return
try:
status = await self.hass.async_add_executor_job(
fetch_local_status, self._local_url
)
except HotSpringApiError as err:
_LOGGER.debug("local /status poll failed: %s", err)
self._local_failures += 1
if self._local_ok and self._local_failures >= LOCAL_FAILURES_BEFORE_CLOUD:
self._local_ok = False
_LOGGER.warning(
"Spa local status URL %s stopped answering (%s); reading "
"telemetry from the cloud until it does",
self._local_url,
err,
)
await self.hass.async_add_executor_job(self.spa.subscribe)
return
self._local_failures = 0
if not self._local_ok:
self._local_ok = True
_LOGGER.info("Spa local status URL %s answers again", self._local_url)
await self.hass.async_add_executor_job(self.spa.unsubscribe)
self.spa.ingest_status(status)
self.async_set_updated_data(self.spa.merged)
@property
def spa_reporting(self) -> bool:
"""True while fresh telemetry carries control-box data.
Entities are unavailable whenever this is False: the module is cut off
from the control box, or nothing has arrived from either read path.
"""
if self.spa is None or self._last_status is None:
return False
if time.monotonic() - self._last_status > STALE_AFTER:
return False
return self.spa.reporting()
@callback
def _note_reporting(self) -> bool:
"""Log a health change once; return True when it changed."""
reporting = self.spa_reporting
if reporting == self._reporting:
return False
self._reporting = reporting
if reporting:
_LOGGER.info("Spa is reporting")
elif self.spa is not None and self.spa.merged and not self.spa.reporting():
_LOGGER.warning(
"Spa is not reporting: the module answers but has no data from "
"the control box"
)
else:
_LOGGER.warning("Spa is not reporting: no telemetry is arriving")
return True
@callback
def async_set_updated_data(self, data: dict) -> None:
self._note_reporting()
super().async_set_updated_data(data)
@callback
def _async_watchdog(self, _now) -> None:
"""Catch what no telemetry update can: silence, and a dead control path."""
if self._note_reporting():
self.async_update_listeners()
if self.spa is None or self.spa.connected:
self._broker_down_ticks = 0
return
self._broker_down_ticks += 1
if self._broker_down_ticks >= BROKER_DOWN_TICKS_BEFORE_RELOAD:
_LOGGER.warning(
"Spa broker connection has been down for %s s; reloading to sign "
"in again and fetch current broker credentials",
self._broker_down_ticks * WATCHDOG_INTERVAL,
)
self._broker_down_ticks = 0
self.hass.config_entries.async_schedule_reload(self.entry.entry_id)
@callback
def _on_status(self, suffix: str, payload: dict) -> None:
"""Runs on paho's thread (MQTT) or the executor (local); hop to the loop."""
self._last_status = time.monotonic()
if self.spa is None:
return
self.hass.loop.call_soon_threadsafe(self.async_set_updated_data, self.spa.merged)
async def _run(self, method_name: str, *args) -> None:
if self.spa is None:
raise RuntimeError("spa not connected")
await self.hass.async_add_executor_job(getattr(self.spa, method_name), *args)
async def async_set_temperature(self, value: int) -> None:
await self._run("set_temperature", value)
async def async_set_salt_level(self, value: int) -> None:
await self._run("set_water_care_level", value)
async def async_set_lights(self, on: bool) -> None:
await self._run("set_lights", on)
async def async_set_jets(self, on: bool) -> None:
await self._run("set_jets", on)
async def async_shutdown(self) -> None:
if self._unsub_watchdog is not None:
self._unsub_watchdog()
self._unsub_watchdog = None
if self._unsub_poll is not None:
self._unsub_poll()
self._unsub_poll = None
if self.spa is not None:
await self.hass.async_add_executor_job(self.spa.disconnect)
await super().async_shutdown()