"""Sensor platform for TaskMate integration."""
from __future__ import annotations
import logging
from datetime import date, timedelta
from types import SimpleNamespace
from homeassistant.components.sensor import (
SensorEntity,
SensorStateClass,
)
from homeassistant.config_entries import ConfigEntry
from homeassistant.core import HomeAssistant, callback
from homeassistant.helpers.entity import DeviceInfo
from homeassistant.helpers.entity_platform import AddEntitiesCallback
from homeassistant.helpers.update_coordinator import CoordinatorEntity
from homeassistant.util import dt as dt_util
from . import images
from .chore_undo import child_undo_metadata, undo_window_seconds
from .const import BOUNTY_RECENT_HOURS, DOMAIN, WISH_MAX_OPEN_PER_CHILD
from .coord_rewards import reward_is_time_locked
from .coordinator import TaskMateCoordinator
from .entity import taskmate_device_info
from .models import Child
from .watch import DEFAULT_STATE_STRINGS, watch_state, watch_summary
_LOGGER = logging.getLogger(__name__)
def _safe_float(value, default: float) -> float:
"""Convert value to float, returning default on failure."""
try:
return float(value) if value is not None else default
except (ValueError, TypeError):
return default
def _safe_int(value, default: int) -> int:
"""Convert value to int, returning default on failure."""
try:
return int(value) if value is not None else default
except (ValueError, TypeError):
return default
# ---------------------------------------------------------------------------
# Shared attribute builders
#
# The overview sensor used to pack every data slice into one 16KB+ attribute
# payload, which Home Assistant's recorder refuses to store. The data is now
# split across five sensors (overview / chores / rewards / activity /
# incentives); the helpers below build each slice once per coordinator update
# and share intermediate lookups so all sensors stay cheap to render.
# ---------------------------------------------------------------------------
_COMMON_CACHE_ATTR = "_taskmate_sensor_common_cache"
def _compute_common(coordinator: TaskMateCoordinator) -> dict:
"""Build the per-update context shared by all TaskMate sensors.
Cached on the coordinator so each sensor re-uses a single pass over
completions / transactions / pool allocations.
"""
data = coordinator.data
data_id = id(data)
cached = getattr(coordinator, _COMMON_CACHE_ATTR, None)
if cached and cached.get("data_id") == data_id:
return cached["common"]
children = data.get("children", [])
chores = data.get("chores", [])
rewards = data.get("rewards", [])
all_completions = data.get("completions", [])
pending_completions = data.get("pending_completions", [])
pending_reward_claim_objs = data.get("pending_reward_claims", [])
pool_alloc_objs = data.get("pool_allocations", [])
child_lookup = {c.id: c for c in children}
chore_lookup = {c.id: c for c in chores}
reward_lookup = {r.id: r for r in rewards}
# Wishlist redemptions (#932) are reward claims with no Reward behind them.
wish_lookup = {w.id: w for w in coordinator.storage.get_wishes()}
# A bounty's completion (#931) carries the bounty id as its chore_id, so
# every "which chore is this?" lookup below falls back to this one.
bounty_lookup = {b.id: b for b in data.get("bounties", []) or []}
# Current-month leaderboard points per child (FEAT-2).
season_points = coordinator.storage.get_season_points(dt_util.now().strftime("%Y-%m"))
# Pending chore points per child (awaiting approval -> points to be earned).
pending_points_by_child: dict[str, int] = {}
for comp in pending_completions:
chore = chore_lookup.get(comp.chore_id) or bounty_lookup.get(comp.chore_id)
if chore:
pending_points_by_child[comp.child_id] = pending_points_by_child.get(comp.child_id, 0) + chore.points
# Committed points per child (reward claims awaiting approval = points reserved).
# Pool-mode pending claims are skipped because their cost was already deducted
# from child.points at allocation time.
committed_points_by_child: dict[str, int] = {}
for rc in pending_reward_claim_objs:
if coordinator.is_pool_mode_claim(rc):
continue
reward = reward_lookup.get(rc.reward_id)
if reward:
committed_points_by_child[rc.child_id] = committed_points_by_child.get(rc.child_id, 0) + reward.cost
# Pool allocation lookups for v3.0 pool mode.
pool_by_child_reward: dict[str, dict[str, int]] = {}
pool_total_by_reward: dict[str, int] = {}
total_allocated_by_child: dict[str, int] = {}
for pa in pool_alloc_objs:
pool_by_child_reward.setdefault(pa.child_id, {})[pa.reward_id] = pa.allocated_points
pool_total_by_reward[pa.reward_id] = pool_total_by_reward.get(pa.reward_id, 0) + pa.allocated_points
total_allocated_by_child[pa.child_id] = total_allocated_by_child.get(pa.child_id, 0) + pa.allocated_points
common = {
"data_id": data_id,
"hass": coordinator.hass,
"data": data,
"children": children,
"chores": chores,
"rewards": rewards,
"child_lookup": child_lookup,
"chore_lookup": chore_lookup,
"reward_lookup": reward_lookup,
"wish_lookup": wish_lookup,
"bounty_lookup": bounty_lookup,
"season_points": season_points,
"all_completions": all_completions,
"chore_undo_seconds": undo_window_seconds(coordinator.storage),
"pending_completions": pending_completions,
"pending_reward_claim_objs": pending_reward_claim_objs,
"pool_alloc_objs": pool_alloc_objs,
"pending_points_by_child": pending_points_by_child,
"committed_points_by_child": committed_points_by_child,
"pool_by_child_reward": pool_by_child_reward,
"pool_total_by_reward": pool_total_by_reward,
"total_allocated_by_child": total_allocated_by_child,
# Sent back to redo by an inspection (#981): not "done" on the card.
"inspection_redo_ids": _inspection_redo_ids(coordinator),
}
setattr(coordinator, _COMMON_CACHE_ATTR, {"data_id": data_id, "common": common})
return common
def _inspection_redo_ids(coordinator) -> set[str]:
getter = getattr(coordinator, "inspection_redo_completion_ids", None)
try:
return set(getter()) if callable(getter) else set()
except Exception: # noqa: BLE001
_LOGGER.debug("inspection redo ids unavailable", exc_info=True)
return set()
def _build_children_summary(coordinator: TaskMateCoordinator, common: dict) -> list[dict]:
"""Build compact per-child summary used on the overview sensor."""
children = common["children"]
pending = common["pending_points_by_child"]
committed = common["committed_points_by_child"]
allocated = common["total_allocated_by_child"]
season = common["season_points"]
freezes_on = _safe_int(((common.get("data") or {}).get("settings") or {}).get("streak_freeze_max"), 2) > 0
summary = []
for c in children:
committed_amount = committed.get(c.id, 0)
lvl = coordinator.level_info(c)
# Weekly targets (#883): {chore_id: done so far this week} for the
# chores that have one, so a card can render "2/3 this week" without
# its own pass over completions. Empty — and so absent — for most.
weekly_progress = {
ch.id: coordinator.weekly_completion_count(ch.id, c.id)
for ch in common["chores"]
if int(getattr(ch, "weekly_target", 0) or 0) > 0
}
summary.append(
{
"level": lvl["level"],
"level_progress": lvl["progress"],
"level_target": lvl["target"],
"id": c.id,
"name": c.name,
"points": c.points,
"pending_points": pending.get(c.id, 0),
# Guest profiles (#690): cards filter these out of competitive views.
**(
{"is_guest": True, "guest_expires_on": getattr(c, "guest_expires_on", "")}
if getattr(c, "is_guest", False)
else {}
),
# Chore roulette (#677): today's pick + spins left, so the card can
# show the result and disable the button once the allowance is used.
**(
{
"roulette": {
**(coordinator.roulette_selection(c.id) or {}),
"spins_left": coordinator.roulette_spins_left(c.id),
}
}
if coordinator.roulette_enabled()
else {}
),
"committed_points": committed_amount,
"allocated_points": allocated.get(c.id, 0),
# Allocations were deducted from child.points already, so spendable
# only needs to account for pending-claim commitments.
"spendable_balance": max(0, c.points - committed_amount),
"chore_order": c.chore_order,
"current_streak": getattr(c, "current_streak", 0) or 0,
"best_streak": getattr(c, "best_streak", 0) or 0,
"season_points": int(season.get(c.id, 0)),
"total_points_earned": getattr(c, "total_points_earned", 0) or 0,
"total_chores_completed": getattr(c, "total_chores_completed", 0) or 0,
"avatar": getattr(c, "avatar", "mdi:account-circle") or "mdi:account-circle",
"last_completion_date": getattr(c, "last_completion_date", None),
"streak_paused": getattr(c, "streak_paused", False),
"on_vacation": coordinator._is_child_on_vacation(c),
# Birthday mode (#924): only on the day, so it costs nothing otherwise.
**({"birthday": bday} if (bday := coordinator.birthday_summary(c)) else {}),
"streak_milestones_achieved": getattr(c, "streak_milestones_achieved", None) or [],
# Streak freeze tokens (#925): only while the feature is on, so
# the cards can tell "none left" apart from "not in use".
**({"streak_freezes": getattr(c, "streak_freezes", 0) or 0} if freezes_on else {}),
"awarded_perfect_weeks": getattr(c, "awarded_perfect_weeks", None) or [],
"career_score": getattr(c, "career_score", 0) or 0,
"total_penalties_received": getattr(c, "total_penalties_received", 0) or 0,
"quests": coordinator.quest_progress_for_child(c.id),
"avatar_options": coordinator.avatar_options_for_child(c),
"challenges": coordinator.challenge_progress_for_child(c.id),
**({"weekly_chore_progress": weekly_progress} if weekly_progress else {}),
# Reject reasons (#976): why a chore / claim was just sent back,
# only while there's one to show.
**({"rejections": rej} if (rej := coordinator.rejections_for_child(c.id)) else {}),
# Surprise inspections (#981): today's banner(s), only when any.
**({"inspections": insp} if (insp := coordinator.inspections_for_child(c.id)) else {}),
}
)
return summary
def _build_chores_list(coordinator: TaskMateCoordinator, common: dict) -> list[dict]:
"""Build the chores list, omitting rarely-used / empty fields.
Fields dropped entirely (no frontend consumer):
- publish_calendar_entities, last_completed_at,
first_occurrence_mode, assignment_rotation_anchor.
Optional fields are only emitted when they hold a non-default value so
small / simple chores produce compact records and the slice stays under
the 16KB recorder limit.
Per-child availability is split into its own attribute (`chore_availability`)
on the chore-availability sensor to keep this record compact.
"""
chores = common["chores"]
chores_list = []
today = dt_util.as_local(dt_util.now()).date()
# Calendar moves older than the calendar card's week view are history only.
moved_floor = (today - timedelta(days=7)).isoformat()
for c in chores:
assigned_to = c.assigned_to if isinstance(c.assigned_to, list) else []
depends_on = c.depends_on if isinstance(getattr(c, "depends_on", None), list) else []
record: dict = {
"id": c.id,
"name": c.name,
"points": c.points,
"time_category": c.time_category,
"assigned_to": assigned_to,
"depends_on": depends_on,
"schedule_mode": getattr(c, "schedule_mode", "specific_days"),
"enabled": getattr(c, "enabled", True),
"assignment_mode": getattr(c, "assignment_mode", "everyone"),
}
# Difficulty tier + the points it actually awards. Emitted only when
# non-default (medium / ×1.0) so simple chores stay compact.
difficulty = getattr(c, "difficulty", "medium") or "medium"
if difficulty != "medium":
record["difficulty"] = difficulty
# A teamwork chore (#928) pays each participant their share, so that
# is the figure the cards quote as "+N".
team_size = coordinator.teamwork_size(c) if int(getattr(c, "team_size", 0) or 0) >= 2 else 0
effective_points = coordinator.team_share_points(c) if team_size else coordinator.effective_chore_points(c)
if effective_points != c.points:
record["effective_points"] = effective_points
# Teamwork (#928): only on teamwork chores, and only the non-default
# parts — the joined list is the day's progress the card draws.
if team_size:
team: dict = {"size": team_size}
joined = coordinator.team_joined_ids(c)
if joined:
team["joined"] = joined
if getattr(c, "team_points_mode", "each") == "split":
team["split"] = True
record["team"] = team
# Optional fields — emit only when non-default to save bytes.
description = getattr(c, "description", "") or ""
if description:
record["description"] = description
daily_limit = getattr(c, "daily_limit", 1)
if daily_limit != 1:
record["daily_limit"] = daily_limit
# Weekly target (#883) — off for almost every chore, so it only goes
# into the record when a family actually uses one.
weekly_target = int(getattr(c, "weekly_target", 0) or 0)
if weekly_target:
record["weekly_target"] = weekly_target
claim_allowance_minutes = getattr(c, "claim_allowance_minutes", 0) or 0
if claim_allowance_minutes:
record["claim_allowance_minutes"] = claim_allowance_minutes
due_days = getattr(c, "due_days", []) or []
if due_days:
record["due_days"] = due_days
# Occurrences moved or removed from the HA calendar (#977). Only chores
# that have one carry these: `occ_today` settles today for the cards'
# weekday filters, the map lets the calendar card place other days.
moved = getattr(c, "moved_occurrences", None) or {}
if moved:
override = coordinator.occurrence_override(c, today)
if override is not None:
record["occ_today"] = override
recent = {src: dst for src, dst in moved.items() if src >= moved_floor or (dst and dst >= moved_floor)}
if recent:
record["moved_occurrences"] = recent
# A one-off belongs to its own date only (#992) — a calendar-added one
# for later in the week, or yesterday's before the midnight sweep. Same
# flag, so every card's today filter drops it as the backend does.
if record["schedule_mode"] == "one_shot" and record["enabled"] and _one_off_not_today(c, today):
record["occ_today"] = False
# Won at auction (#982). Only chores with a won occurrence this week
# carry these: `auction` settles today for the cards (the winner alone,
# at the winning price, which is also what the chore pays today); the
# map lets the calendar card place the other days.
wins = coordinator.auction_wins_for_chore(c, today, 7)
if isinstance(wins, dict) and wins:
record["auction_wins"] = {day: w["child_id"] for day, w in wins.items()}
# What each won day pays, so a future one shows it too (#998).
record["auction_prices"] = {day: w["points"] for day, w in wins.items()}
today_win = wins.get(today.isoformat())
if today_win:
record["auction"] = today_win
record["effective_points"] = today_win["points"]
requires_approval = getattr(c, "requires_approval", True)
if not requires_approval:
record["requires_approval"] = False
# Mandatory chores (#532): emit only when set so the child card can
# show the mandatory styling/badge. penalty rides along when non-zero.
if getattr(c, "mandatory", False):
record["mandatory"] = True
penalty = getattr(c, "mandatory_penalty_points", 0) or 0
if penalty:
record["mandatory_penalty_points"] = penalty
if getattr(c, "require_photo", False):
record["require_photo"] = True
# Open-ended chores (#832): the child card asks for a description and a
# suggested point value instead of completing on a single tap.
if getattr(c, "open_ended", False):
record["open_ended"] = True
recurrence = getattr(c, "recurrence", "weekly")
if recurrence != "weekly":
record["recurrence"] = recurrence
recurrence_day = getattr(c, "recurrence_day", "")
if recurrence_day:
record["recurrence_day"] = recurrence_day
recurrence_start = getattr(c, "recurrence_start", "")
if recurrence_start:
record["recurrence_start"] = recurrence_start
visibility_entity = getattr(c, "visibility_entity", "")
if visibility_entity:
record["visibility_entity"] = visibility_entity
record["visibility_operator"] = getattr(c, "visibility_operator", "equals")
record["visibility_state"] = getattr(c, "visibility_state", "on")
weather_entity = getattr(c, "weather_entity", "")
if weather_entity:
record["weather_entity"] = weather_entity
record["weather_block_conditions"] = list(getattr(c, "weather_block_conditions", []) or [])
for limit in ("weather_temp_min", "weather_temp_max", "weather_wind_max"):
value = getattr(c, limit, None)
if value is not None:
record[limit] = value
# Why it's hidden, so the panel can say "rained off" instead of
# just dropping the chore out of the list.
reason = coordinator.weather_block_reason(c)
if reason:
record["weather_blocked"] = reason
deadline_at = getattr(c, "deadline_at", "")
if deadline_at:
record["deadline_at"] = deadline_at
speed_bonus = getattr(c, "speed_bonus_points", 0)
if speed_bonus:
record["speed_bonus_points"] = speed_bonus
disabled_for = getattr(c, "disabled_for", [])
if disabled_for:
record["disabled_for"] = disabled_for
created_date = getattr(c, "created_date", "")
if created_date:
record["created_date"] = created_date
assignment_current_child_id = getattr(c, "assignment_current_child_id", "")
if assignment_current_child_id:
record["assignment_current_child_id"] = assignment_current_child_id
icon = getattr(c, "icon", "")
if icon:
record["icon"] = icon
# Signed so the card's ![]()
loads; emitted only when set, matching
# `icon` above, to keep records under the 16KB recorder limit.
image_url = getattr(c, "image_url", "")
if image_url:
# Same trade-off as custom sounds, and unlike evidence photos
# (which deliberately emit a bare path for the card to sign per
# viewer): a chore picture is decorative and renders on first
# paint, so signing it here keeps that render synchronous. It does
# mean the signed URL is readable in this attribute — fine for a
# chore picture, not fine for anything personal.
record["image_url"] = images.sign_image_url(common["hass"], image_url)
completion_sound = getattr(c, "completion_sound", "coin")
if completion_sound and completion_sound != "coin":
record["completion_sound"] = completion_sound
task_type = getattr(c, "task_type", "standard")
if task_type == "timed":
record["task_type"] = "timed"
record["timed_rate_points"] = getattr(c, "timed_rate_points", 10)
record["timed_rate_minutes"] = getattr(c, "timed_rate_minutes", 5)
record["timed_max_daily_minutes"] = getattr(c, "timed_max_daily_minutes", 0)
bonus_subtasks = getattr(c, "bonus_subtasks", [])
if bonus_subtasks:
record["bonus_subtasks"] = [{"id": b.id, "name": b.name, "points": b.points} for b in bonus_subtasks]
chores_list.append(record)
return chores_list
def _one_off_not_today(chore, today: date) -> bool:
"""True when a one-off chore is dated for a day other than ``today``.
Mirrors the one_shot branch of is_chore_available_for_child: no date, or
one that doesn't parse, counts as today.
"""
created_date = getattr(chore, "created_date", "") or ""
if not created_date:
return False
try:
return date.fromisoformat(created_date) != today
except ValueError:
return False
def _build_chore_availability(coordinator: TaskMateCoordinator, common: dict) -> dict:
"""Build the chore availability matrix: {chore_id: {child_id: bool}}.
Extracted from the chores sensor so `sensor.taskmate_chores` carries only
definitions and stays compact even for families with many chores/kids.
"""
children = common["children"]
chores = common["chores"]
availability: dict[str, dict[str, bool]] = {}
with coordinator.availability_build_scope():
for c in chores:
per_child = {}
for child in children:
per_child[child.id] = coordinator.is_chore_available_for_child(c, child.id)
availability[c.id] = per_child
return availability
def _completion_award(completion, recalculated: int) -> int:
"""What a completion is worth, recorded rather than re-derived.
Approved: what the child was actually paid, weekend multiplier included.
Re-deriving it from the chore reported the chore's *current* points, so
re-pricing a chore rewrote every past completion of it.
Pending: what the child was promised at submission, including a speed
bonus or roulette multiplier that may no longer apply — the same figure
approval will pay.
Neither recorded (a completion from before those fields existed): fall
back to the caller's calculation.
"""
if completion.approved:
return completion.points_awarded
if completion.submitted_points is not None:
return completion.submitted_points
return recalculated
def _bounty_title(common: dict, comp) -> str:
"""The bounty's title for a bounty completion (#931), else ""."""
bounty = common.get("bounty_lookup", {}).get(getattr(comp, "bounty_id", "") or "")
return bounty.title if bounty else ""
def _build_todays_completions(common: dict) -> list[dict]:
"""Build today's completions (both approved and pending)."""
now = dt_util.now()
today = now.date()
child_lookup = common["child_lookup"]
chore_lookup = common["chore_lookup"]
redo_ids = common.get("inspection_redo_ids") or set()
out = []
for comp in common["all_completions"]:
# An inspection sent this one back to redo (#981): the chore is owed
# again today, so the card must not see it as done.
if comp.id in redo_ids:
continue
comp_dt = comp.completed_at
if hasattr(comp_dt, "astimezone"):
comp_dt = dt_util.as_local(comp_dt)
comp_date = comp_dt.date() if hasattr(comp_dt, "date") else comp_dt
if comp_date != today:
continue
matched_chore = chore_lookup.get(comp.chore_id)
bonus_subtask_id = getattr(comp, "bonus_subtask_id", "") or ""
if bonus_subtask_id and matched_chore:
subtask = next((b for b in matched_chore.bonus_subtasks if b.id == bonus_subtask_id), None)
display_name = f"{matched_chore.name} › {subtask.name}" if subtask else matched_chore.name
display_points = subtask.points if subtask else 0
else:
display_name = matched_chore.name if matched_chore else _bounty_title(common, comp)
display_points = matched_chore.points if matched_chore else 0
timed_secs = getattr(comp, "timed_duration_seconds", 0) or 0
if timed_secs > 0 and matched_chore and getattr(matched_chore, "task_type", "") == "timed":
rate_seconds = matched_chore.timed_rate_minutes * 60
if rate_seconds > 0:
display_points = (timed_secs // rate_seconds) * matched_chore.timed_rate_points
rec = {
"completion_id": comp.id,
"chore_id": comp.chore_id,
"child_id": comp.child_id,
"child_name": "Parent"
if comp.child_id == "__parent__"
else (child_lookup[comp.child_id].name if comp.child_id in child_lookup else ""),
"chore_name": display_name,
"points": _completion_award(comp, display_points),
"approved": comp.approved,
"completed_at": comp.completed_at.isoformat()
if hasattr(comp.completed_at, "isoformat")
else str(comp.completed_at),
"bonus_subtask_id": bonus_subtask_id,
}
if timed_secs > 0:
rec["timed_duration_seconds"] = timed_secs
# Child undo (#918): child_undo_pending / child_undo_until, present only
# on a completion the child may still take back, so the attribute
# stays small (and empty whenever the window is 0).
rec.update(child_undo_metadata(comp, common["all_completions"], common.get("chore_undo_seconds", 0)))
# Emit the bare (unsigned) photo path. A card's ![]()
carries no bearer
# token, so the card signs each path per-viewer via auth/sign_path before
# rendering — this keeps a self-authenticating URL out of this
# world-readable attribute (a signed URL here would be redeemable by
# anyone who could read the state).
photo = getattr(comp, "photo_url", "") or ""
if photo:
rec["photo_url"] = photo
out.append(rec)
return out
def _build_active_timed_sessions(coordinator: TaskMateCoordinator) -> list[dict]:
"""Build active timed sessions for frontend consumption."""
sessions = coordinator.data.get("timed_sessions", [])
out = []
for s in sessions:
if hasattr(s, "state"):
state = s.state
if state not in ("running", "paused"):
continue
segments = s.segments or []
current_segment_start = ""
if state == "running" and segments:
last_seg = segments[-1]
if isinstance(last_seg, dict) and last_seg.get("end") is None:
current_segment_start = last_seg.get("start", "")
out.append(
{
"chore_id": s.chore_id,
"child_id": s.child_id,
"state": state,
"total_seconds_today": s.total_seconds_today,
"current_segment_start": current_segment_start,
}
)
else:
state = s.get("state", "stopped") if isinstance(s, dict) else "stopped"
if state not in ("running", "paused"):
continue
segments = s.get("segments", []) if isinstance(s, dict) else []
current_segment_start = ""
if state == "running" and segments:
last_seg = segments[-1]
if isinstance(last_seg, dict) and last_seg.get("end") is None:
current_segment_start = last_seg.get("start", "")
out.append(
{
"chore_id": s.get("chore_id", "") if isinstance(s, dict) else "",
"child_id": s.get("child_id", "") if isinstance(s, dict) else "",
"state": state,
"total_seconds_today": s.get("total_seconds_today", 0) if isinstance(s, dict) else 0,
"current_segment_start": current_segment_start,
}
)
return out
def _build_rewards_list(common: dict) -> list[dict]:
"""Build the rewards list with calculated_costs and pool allocations."""
rewards = common["rewards"]
children = common["children"]
pool_by_child_reward = common["pool_by_child_reward"]
pool_total_by_reward = common["pool_total_by_reward"]
now = dt_util.now()
today = now.date()
out = []
for r in rewards:
assigned = r.assigned_to if isinstance(r.assigned_to, list) and r.assigned_to else [c.id for c in children]
calculated_costs = dict.fromkeys(assigned, r.cost)
reward_pool_allocations = {cid: pool_by_child_reward.get(cid, {}).get(r.id, 0) for cid in assigned}
jackpot_pool_total = pool_total_by_reward.get(r.id, 0) if getattr(r, "is_jackpot", False) else None
quantity = getattr(r, "quantity", None)
expires_at = getattr(r, "expires_at", None)
is_sold_out = quantity is not None and quantity <= 0
is_expired = False
days_until_expiry: int | None = None
if expires_at:
try:
deadline = date.fromisoformat(expires_at)
is_expired = deadline <= today
days_until_expiry = (deadline - today).days
except (TypeError, ValueError):
pass
is_time_locked = reward_is_time_locked(r, now)
out.append(
{
"id": r.id,
"name": r.name,
"cost": r.cost,
"description": getattr(r, "description", ""),
"icon": r.icon,
"assigned_to": r.assigned_to if isinstance(r.assigned_to, list) else [],
"is_jackpot": getattr(r, "is_jackpot", False),
"pool_enabled": getattr(r, "pool_enabled", False),
"calculated_costs": calculated_costs,
"pool_allocations": reward_pool_allocations,
"jackpot_pool_total": jackpot_pool_total,
"quantity": quantity,
"expires_at": expires_at,
"is_sold_out": is_sold_out,
"is_expired": is_expired,
"is_available": not (is_sold_out or is_expired or is_time_locked),
"days_until_expiry": days_until_expiry,
}
)
# Time lock (#857): only carried for rewards that actually use it. This
# attribute slice is capped at the recorder's 16 KB limit, so unused
# feature fields must not cost every reward bytes.
# Streak freeze (#925): flag only the rewards that are one.
if getattr(r, "streak_freeze", False):
out[-1]["streak_freeze"] = True
if getattr(r, "time_lock_enabled", False):
out[-1]["time_lock"] = {
"days": list(getattr(r, "available_days", []) or []),
"from": getattr(r, "available_from", "") or "",
"until": getattr(r, "available_until", "") or "",
}
if is_time_locked:
out[-1]["is_time_locked"] = True
return out
def _wish_as_reward(wish) -> SimpleNamespace | None:
"""Present a wish's redemption claim (#932) the way a reward claim reads.
Approval cards and the activity feed look a claim's reward up by id; a
wish has no Reward record, so they get its name, a heart and its target.
"""
if wish is None:
return None
return SimpleNamespace(id=wish.id, name=wish.name, icon="mdi:heart", cost=wish.target)
def _build_pending_reward_claims(common: dict) -> list[dict]:
"""Build enriched pending reward claims list."""
reward_lookup = common["reward_lookup"]
child_lookup = common["child_lookup"]
out = []
wish_lookup = common.get("wish_lookup", {})
for rc in common["pending_reward_claim_objs"]:
reward = reward_lookup.get(rc.reward_id) or _wish_as_reward(wish_lookup.get(rc.wish_id))
child = child_lookup.get(rc.child_id)
if not reward or not child:
continue
out.append(
{
"claim_id": rc.id,
"reward_id": rc.reward_id,
"child_id": rc.child_id,
"child_name": child.name,
"child_avatar": getattr(child, "avatar", "mdi:account-circle") or "mdi:account-circle",
"reward_name": reward.name,
"reward_icon": reward.icon or "mdi:gift",
"cost": reward.cost,
"claimed_at": rc.claimed_at.isoformat() if hasattr(rc.claimed_at, "isoformat") else str(rc.claimed_at),
}
)
return out
def _build_recent_completions(common: dict, limit: int = 35) -> list[dict]:
"""Return the most recent completions, capped to `limit`.
Lowered from 200 to 35 so the activity sensor's attribute payload stays
under the 16KB recorder limit even for families with heavy chore activity.
"""
child_lookup = common["child_lookup"]
chore_lookup = common["chore_lookup"]
recent = sorted(common["all_completions"], key=lambda c: c.completed_at, reverse=True)[:limit]
out = []
for comp in recent:
matched_chore = chore_lookup.get(comp.chore_id)
display_points = matched_chore.points if matched_chore else 0
timed_secs = getattr(comp, "timed_duration_seconds", 0) or 0
if timed_secs > 0 and matched_chore and getattr(matched_chore, "task_type", "") == "timed":
rate_seconds = matched_chore.timed_rate_minutes * 60
if rate_seconds > 0:
display_points = (timed_secs // rate_seconds) * matched_chore.timed_rate_points
out.append(
{
"completion_id": comp.id,
"chore_id": comp.chore_id,
"child_id": comp.child_id,
"child_name": "Parent"
if comp.child_id == "__parent__"
else (child_lookup[comp.child_id].name if comp.child_id in child_lookup else ""),
"chore_name": matched_chore.name if matched_chore else _bounty_title(common, comp),
"points": _completion_award(comp, display_points),
"approved": comp.approved,
"completed_at": comp.completed_at.isoformat()
if hasattr(comp.completed_at, "isoformat")
else str(comp.completed_at),
# Star rating the parent gave at approval (#927) — only when rated.
**({"rating": comp.quality_rating} if getattr(comp, "quality_rating", 0) else {}),
}
)
return out
def _build_photo_gallery(common: dict, limit: int = 40) -> list[dict]:
"""Recent completions that carry an evidence photo, newest first (FEAT-13).
Returns the raw (unsigned) ``/api/taskmate/photo/`` path; the gallery
card signs each via ``auth/sign_path`` before rendering (a card's ![]()
can't
send a bearer token).
"""
child_lookup = common["child_lookup"]
chore_lookup = common["chore_lookup"]
with_photos = [c for c in common["all_completions"] if getattr(c, "photo_url", "")]
recent = sorted(with_photos, key=lambda c: c.completed_at, reverse=True)[:limit]
out = []
for comp in recent:
chore = chore_lookup.get(comp.chore_id)
out.append(
{
"completion_id": comp.id,
"child_name": child_lookup[comp.child_id].name if comp.child_id in child_lookup else "",
"chore_name": chore.name if chore else _bounty_title(common, comp),
"approved": comp.approved,
"completed_at": comp.completed_at.isoformat()
if hasattr(comp.completed_at, "isoformat")
else str(comp.completed_at),
"photo_url": comp.photo_url,
}
)
return out
_FEED_REJECTIONS_MAX = 8
_FEED_INSPECTIONS_MAX = 8
def _build_recent_transactions(
common: dict,
limit: int = 20,
rejections: list[dict] | None = None,
inspections: list[dict] | None = None,
) -> list[dict]:
"""Unified activity feed of manual point adjustments and reward claims.
Capped at 20 so the combined activity slice (completions + transactions)
stays under the 16KB recorder limit.
"""
data = common["data"]
child_lookup = common["child_lookup"]
reward_lookup = common["reward_lookup"]
points_transactions = data.get("points_transactions", [])
all_reward_claims = data.get("reward_claims", [])
events = []
for t in points_transactions:
child = child_lookup.get(t.child_id)
if not child:
continue
events.append(
{
"transaction_id": t.id,
"type": "points_added" if t.points > 0 else "points_removed",
"child_id": t.child_id,
"child_name": child.name,
"points": t.points,
"reason": t.reason or "",
"created_at": t.created_at.isoformat() if hasattr(t.created_at, "isoformat") else str(t.created_at),
}
)
wish_lookup = common.get("wish_lookup", {})
for rc in all_reward_claims:
child = child_lookup.get(rc.child_id)
reward = reward_lookup.get(rc.reward_id) or _wish_as_reward(wish_lookup.get(getattr(rc, "wish_id", "")))
if not child or not reward:
continue
event_type = "reward_approved" if rc.approved else "reward_claimed"
timestamp = rc.approved_at if rc.approved and rc.approved_at else rc.claimed_at
events.append(
{
"transaction_id": rc.id,
"type": event_type,
"child_id": rc.child_id,
"child_name": child.name,
"reward_id": rc.reward_id,
"reward_name": reward.name,
"reward_icon": reward.icon or "mdi:gift",
# An approved purchase is worth what it cost that day, not what
# the reward has been re-priced to since.
"points": -(rc.approved_cost if rc.approved and rc.approved_cost is not None else reward.cost),
"approved": rc.approved,
"created_at": timestamp.isoformat() if hasattr(timestamp, "isoformat") else str(timestamp),
}
)
# Reasoned rejections (#976): the completion / claim itself is gone, so
# these come from the rejection log rather than the records above. A
# reason can run to 200 characters, so only the newest few make the feed
# (the 16 KB cap).
for r in (rejections or [])[-_FEED_REJECTIONS_MAX:]:
child = child_lookup.get(r.get("child_id"))
if not child:
continue
kind = "reward" if r.get("kind") == "reward" else "chore"
events.append(
{
"transaction_id": r.get("id", ""),
"type": f"{kind}_rejected",
"child_id": child.id,
"child_name": child.name,
f"{kind}_id": r.get("item_id", ""),
f"{kind}_name": r.get("item_name", ""),
"reason": r.get("reason", ""),
"created_at": str(r.get("rejected_at", "")),
}
)
# Surprise inspections (#981): started / didn't pass / sent back (a pass
# with a bonus is already here as its "Inspection passed:" transaction).
for e in (inspections or [])[-_FEED_INSPECTIONS_MAX:]:
child = child_lookup.get(e.get("child_id"))
if not child:
continue
row = {k: v for k, v in e.items() if k != "id"}
events.append({**row, "transaction_id": e.get("id", ""), "child_name": child.name})
events.sort(key=lambda e: e["created_at"], reverse=True)
return events[:limit]
def _build_penalties_list(common: dict) -> list[dict]:
return [
{
"id": p.id,
"name": p.name,
"points": p.points,
"description": p.description,
"icon": p.icon,
"assigned_to": p.assigned_to or [],
}
for p in common["data"].get("penalties", [])
]
def _build_bonuses_list(common: dict) -> list[dict]:
return [
{
"id": b.id,
"name": b.name,
"points": b.points,
"description": b.description,
"icon": b.icon,
"assigned_to": b.assigned_to or [],
}
for b in common["data"].get("bonuses", [])
]
async def async_setup_entry(
hass: HomeAssistant,
entry: ConfigEntry,
async_add_entities: AddEntitiesCallback,
) -> None:
"""Set up TaskMate sensors."""
coordinator: TaskMateCoordinator = hass.data[DOMAIN][entry.entry_id]
entities: list[SensorEntity] = []
# Track child IDs that have sensors created
tracked_child_ids: set[str] = set()
# Add global sensors. The overview sensor is kept for backward compatibility
# with existing Lovelace dashboards; its heavy attributes have been split
# across four companion sensors to stay below the 16KB recorder limit.
entities.append(TaskMateOverallStatsSensor(coordinator, entry))
entities.append(TaskMateChoresSensor(coordinator, entry))
entities.append(TaskMateChoreAvailabilitySensor(coordinator, entry))
entities.append(TaskMateRewardsSensor(coordinator, entry))
entities.append(TaskMateActivitySensor(coordinator, entry))
entities.append(TaskMateIncentivesSensor(coordinator, entry))
entities.append(TaskMateBountiesSensor(coordinator, entry))
entities.append(TaskMateAuctionsSensor(coordinator, entry))
# Add sensors for each child
for child in coordinator.data.get("children", []):
entities.append(ChildPointsSensor(coordinator, entry, child))
entities.append(ChildStatsSensor(coordinator, entry, child))
entities.append(ChildBadgesSensor(coordinator, entry, child))
entities.append(ChildWishlistSensor(coordinator, entry, child))
entities.append(ChildWatchSensor(coordinator, entry, child))
tracked_child_ids.add(child.id)
# Add pending approvals sensor
entities.append(PendingApprovalsSensor(coordinator, entry))
async_add_entities(entities)
# Set up listener for new children
@callback
def async_add_child_sensors() -> None:
"""Add sensors for newly added children."""
new_entities: list[SensorEntity] = []
for child in coordinator.data.get("children", []):
if child.id not in tracked_child_ids:
new_entities.append(ChildPointsSensor(coordinator, entry, child))
new_entities.append(ChildStatsSensor(coordinator, entry, child))
new_entities.append(ChildBadgesSensor(coordinator, entry, child))
new_entities.append(ChildWishlistSensor(coordinator, entry, child))
new_entities.append(ChildWatchSensor(coordinator, entry, child))
tracked_child_ids.add(child.id)
if new_entities:
async_add_entities(new_entities)
coordinator.async_add_listener(async_add_child_sensors)
class TaskMateBaseSensor(CoordinatorEntity, SensorEntity):
"""Base class for TaskMate sensors."""
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator)
self._entry = entry
@property
def device_info(self) -> DeviceInfo:
"""Return device info."""
return taskmate_device_info(self._entry.entry_id)
class _CachedAttrsSensor(TaskMateBaseSensor):
"""Base for sensors whose attributes are expensive to build."""
def __init__(self, coordinator: TaskMateCoordinator, entry: ConfigEntry) -> None:
super().__init__(coordinator, entry)
self._cached_attrs: dict | None = None
self._cached_key: tuple[int, int] | None = None
@property
def extra_state_attributes(self) -> dict:
# The coordinator reuses one snapshot object while storage.data_version
# is unchanged, so id(data) alone would pin these attributes forever
# when only an external entity moved — stranding the visibility and
# weather gates on stale state. external_state_version covers that.
key = (id(self.coordinator.data), getattr(self.coordinator, "external_state_version", 0))
if self._cached_key == key and self._cached_attrs is not None:
return self._cached_attrs
attrs = self._build_attributes()
self._cached_attrs = attrs
self._cached_key = key
return attrs
def _build_attributes(self) -> dict: # pragma: no cover - abstract
raise NotImplementedError
class TaskMateOverallStatsSensor(_CachedAttrsSensor):
"""Overview sensor — scalars plus the compact per-child summary.
Heavy lists (chores/rewards/activity/incentives) live on companion
sensors so this entity stays well under the 16KB recorder limit.
"""
# The per-child summary grows with the family; the scalars stay recorded
# so history/statistics on them keep working (#817).
_unrecorded_attributes = frozenset({"children", "vacation_periods", "season_champions"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_overall_stats"
self._attr_name = "TaskMate Overview"
@property
def native_value(self) -> int:
"""Return the total number of children."""
return len(self.coordinator.data.get("children", []))
@property
def icon(self) -> str:
return "mdi:clipboard-check-multiple"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
data = common["data"]
children = common["children"]
chores = common["chores"]
rewards = common["rewards"]
total_points = sum(c.points for c in children)
total_chores_completed = sum(c.total_chores_completed for c in children)
today_dow = dt_util.now().strftime("%A").lower()
settings = data.get("settings", {})
time_periods = self.coordinator.get_time_periods()
period_by_id = {p["id"]: p for p in time_periods}
active_vacation = self.coordinator.active_vacation()
# Legacy shape kept for cards that still read the four fixed keys.
legacy_defaults = {
"morning": ("06:00", "12:00"),
"afternoon": ("12:00", "17:00"),
"evening": ("17:00", "21:00"),
"night": ("21:00", "23:59"),
}
time_boundaries = {}
for cat, (def_start, def_end) in legacy_defaults.items():
period = period_by_id.get(cat)
time_boundaries[f"{cat}_start"] = period["start"] if period else def_start
time_boundaries[f"{cat}_end"] = period["end"] if period else def_end
return {
"today_day_of_week": today_dow,
"streak_reset_mode": settings.get("streak_reset_mode", "reset"),
"weekend_multiplier": _safe_float(settings.get("weekend_multiplier"), 2.0),
"streak_milestones_enabled": settings.get("streak_milestones_enabled", "true") == "true",
"streak_milestones": settings.get("streak_milestones", "3:5, 7:10, 14:20, 30:50, 60:100, 100:200"),
"perfect_week_enabled": settings.get("perfect_week_enabled", "true") == "true",
"perfect_week_bonus": _safe_int(settings.get("perfect_week_bonus"), 50),
"streak_requires_all_chores": settings.get("streak_requires_all_chores", "false") in (True, "true"),
"streak_freeze_max": max(0, _safe_int(settings.get("streak_freeze_max"), 2)),
"streak_freeze_earn_every": max(0, _safe_int(settings.get("streak_freeze_earn_every"), 7)),
"perfect_week_requires_all_chores": settings.get("perfect_week_requires_all_chores", "false")
in (True, "true"),
"total_children": len(children),
"total_chores": len(chores),
"total_rewards": len(rewards),
"total_points_available": total_points,
"total_chores_completed": total_chores_completed,
"total_completions_all_time": len(common["all_completions"]),
"total_pending_completions": len(common["pending_completions"]),
"points_name": data.get("points_name", "Stars"),
"points_icon": data.get("points_icon", "mdi:star"),
# Global default card-design style; cards read this when no per-card override (#design).
"card_design": settings.get("card_design", "classic"),
# Non-admin parent role (#661): cards unlock parent controls for
# these HA users. Deliberately still published in the clear, and
# deliberately never trusted: every privileged path re-resolves the
# role server-side (`_async_require_parent` / `authz`), so this list
# only decides which buttons a card draws. Any authenticated user
# can read it, which tells them which accounts hold the role — worth
# knowing, but it grants nothing, and the alternatives (digests, or
# a per-connection lookup) make the check asynchronous and race the
# first paint in every card that gates controls on it.
"parent_user_ids": self.coordinator.storage.get_parent_user_ids(),
"children": _build_children_summary(self.coordinator, common),
"time_boundaries": time_boundaries,
"time_periods": time_periods,
"vacation_active": active_vacation is not None,
"vacation_name": active_vacation.get("name", "") if active_vacation else "",
"vacation_end": active_vacation.get("end", "") if active_vacation else "",
"vacation_periods": self.coordinator.get_vacation_periods(),
# Leaderboard seasons (FEAT-2): current-month standings + champion history.
"season_month": dt_util.now().strftime("%Y-%m"),
"season_champions": list(reversed(self.coordinator.storage.get_season_champions()))[:12],
# Family co-op goal (FEAT-4).
"family_goal": {
"enabled": settings.get("family_goal_enabled", False) in (True, "true"),
"name": settings.get("family_goal_name", "") or "",
"target": _safe_int(settings.get("family_goal_target"), 0),
"reward": settings.get("family_goal_reward", "") or "",
"progress": sum(c.points for c in children),
"achieved": settings.get("family_goal_achieved", False) in (True, "true"),
},
}
class TaskMateChoresSensor(_CachedAttrsSensor):
"""Chores catalog + today's completions."""
# custom_sounds carries a freshly signed URL per sound, so it changes on every
# refresh — recording it would churn the DB for no history value (#856).
_unrecorded_attributes = frozenset(
{"chores", "todays_completions", "task_groups", "active_timed_sessions", "custom_sounds"}
)
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_chores"
self._attr_name = "TaskMate Chores"
@property
def native_value(self) -> int:
return len(self.coordinator.data.get("chores", []))
@property
def icon(self) -> str:
return "mdi:format-list-checks"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
task_groups = [
{
"id": g.id,
"name": g.name,
"policy": g.policy,
"chore_ids": list(g.chore_ids or []),
}
for g in self.coordinator.storage.get_task_groups()
]
active_vacation = self.coordinator.active_vacation()
return {
"chores": _build_chores_list(self.coordinator, common),
"todays_completions": _build_todays_completions(common),
# Lookup table for chores whose completion_sound is "custom:" (#856).
"custom_sounds": self.coordinator.custom_sounds_state(),
"task_groups": task_groups,
"active_timed_sessions": _build_active_timed_sessions(self.coordinator),
"vacation_active": active_vacation is not None,
"vacation_name": active_vacation.get("name", "") if active_vacation else "",
"vacation_end": active_vacation.get("end", "") if active_vacation else "",
}
class TaskMateChoreAvailabilitySensor(_CachedAttrsSensor):
"""Per-chore × per-child availability matrix.
Split off from the chores sensor so `sensor.taskmate_chores` can stay
under the 16 KB recorder limit even for families with lots of chores.
The map is `{chore_id: {child_id: bool}}`.
"""
_unrecorded_attributes = frozenset({"chore_availability"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_chore_availability"
self._attr_name = "TaskMate Chore Availability"
@property
def native_value(self) -> int:
# Read the matrix back off the attribute cache rather than rebuilding
# it: HA writes the state and the attributes together, so building it
# here too doubled the cost of every write (#823).
availability = self.extra_state_attributes.get("chore_availability", {})
return sum(sum(1 for v in per_child.values() if v) for per_child in availability.values())
@property
def icon(self) -> str:
return "mdi:calendar-check"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
return {
"chore_availability": _build_chore_availability(self.coordinator, common),
}
class TaskMateRewardsSensor(_CachedAttrsSensor):
"""Rewards catalog + pending claims + pool allocations."""
_unrecorded_attributes = frozenset({"rewards", "pending_reward_claims", "pool_allocations"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_rewards"
self._attr_name = "TaskMate Rewards"
@property
def native_value(self) -> int:
return len(self.coordinator.data.get("rewards", []))
@property
def icon(self) -> str:
return "mdi:gift-outline"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
return {
"rewards": _build_rewards_list(common),
"pending_reward_claims": _build_pending_reward_claims(common),
"pool_allocations": [pa.to_dict() for pa in common["pool_alloc_objs"]],
}
class TaskMateActivitySensor(_CachedAttrsSensor):
"""Recent completions + recent points/reward transactions."""
_unrecorded_attributes = frozenset(
{"recent_completions", "recent_transactions", "career_score_history", "photo_gallery"}
)
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_activity"
self._attr_name = "TaskMate Activity"
@property
def native_value(self) -> int:
return len(self.coordinator.data.get("completions", []))
@property
def icon(self) -> str:
return "mdi:history"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
children = common["children"]
career_history = {}
for c in children:
career_history[c.id] = self.coordinator.storage.get_career_score_history(c.id)
return {
"recent_completions": _build_recent_completions(common),
"recent_transactions": _build_recent_transactions(
common,
rejections=self.coordinator.recent_rejections(),
inspections=self.coordinator.recent_inspection_events(),
),
"career_score_history": career_history,
"photo_gallery": _build_photo_gallery(common),
}
class TaskMateIncentivesSensor(_CachedAttrsSensor):
"""Penalties + bonuses catalogue."""
_unrecorded_attributes = frozenset({"penalties", "bonuses"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_incentives"
self._attr_name = "TaskMate Incentives"
@property
def native_value(self) -> int:
data = self.coordinator.data
return len(data.get("penalties", [])) + len(data.get("bonuses", []))
@property
def icon(self) -> str:
return "mdi:scale-balance"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
return {
"penalties": _build_penalties_list(common),
"bonuses": _build_bonuses_list(common),
}
def _build_bounties_list(coordinator: TaskMateCoordinator, common: dict) -> list[dict]:
"""The bounty board for the cards (#931): live bounties plus the last day's
completed ones. Expired and older history stays in storage for the panel.
Compact: optional fields only appear when they carry something, so a board
of a handful of jobs costs a few hundred bytes.
"""
now = dt_util.now()
recent_cutoff = now - timedelta(hours=BOUNTY_RECENT_HOURS)
completions = {c.id: c for c in common["all_completions"]}
window = common.get("chore_undo_seconds", 0)
out = []
for b in common["data"].get("bounties", []) or []:
if b.status == "expired":
continue
if b.status == "completed" and (b.closed_at is None or b.closed_at < recent_cutoff):
continue
rec = {
"id": b.id,
"title": b.title,
"points": b.points,
"icon": b.icon,
"status": b.status,
"claim_hours": b.claim_hours,
}
if b.description:
rec["description"] = b.description
if b.eligible_child_ids:
rec["eligible"] = list(b.eligible_child_ids)
if b.expires_at is not None:
rec["expires_at"] = b.expires_at.isoformat()
if b.require_photo:
rec["require_photo"] = True
if b.status in ("claimed", "pending", "completed") and b.claimed_by:
rec["claimed_by"] = b.claimed_by
if b.status == "claimed":
if b.claimed_at is not None:
rec["claimed_at"] = b.claimed_at.isoformat()
if b.claim_until is not None:
rec["claim_until"] = b.claim_until.isoformat()
if b.status == "pending" and b.completion_id:
rec["completion_id"] = b.completion_id
comp = completions.get(b.completion_id)
if comp is not None and child_undo_metadata(comp, common["all_completions"], window).get(
"child_undo_pending"
):
rec["undo"] = True
if b.status == "completed":
rec["points_awarded"] = b.points_awarded
rec["completed_at"] = b.closed_at.isoformat()
out.append(rec)
return out
class TaskMateBountiesSensor(_CachedAttrsSensor):
"""The bounty board (#931): open bounties as the state, the board as an
attribute for the bounty card."""
_unrecorded_attributes = frozenset({"bounties"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_bounties"
self._attr_name = "TaskMate Bounties"
@property
def native_value(self) -> int:
return sum(1 for b in self.coordinator.data.get("bounties", []) or [] if b.status == "open")
@property
def icon(self) -> str:
return "mdi:flag-checkered"
def _build_attributes(self) -> dict:
common = _compute_common(self.coordinator)
return {"bounties": _build_bounties_list(self.coordinator, common)}
class TaskMateAuctionsSensor(_CachedAttrsSensor):
"""Chore auctions (#982): live auctions as the state, and a public digest
the auction card watches to know when to re-fetch its view.
Bids are secret, so the digest carries only how many are in — never who
bid or how much. The card fetches its own bid over the
``taskmate/auctions/list`` WebSocket command, which enforces that.
"""
_unrecorded_attributes = frozenset({"auctions"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_auctions"
self._attr_name = "TaskMate Auctions"
@property
def native_value(self) -> int:
return sum(1 for a in self.coordinator.storage.get_auctions() if a.status == "open")
@property
def icon(self) -> str:
return "mdi:gavel"
def _build_attributes(self) -> dict:
return {"auctions": self.coordinator.auctions_public_state()}
class ChildPointsSensor(TaskMateBaseSensor):
"""Sensor for a child's points."""
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
child: Child,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator, entry)
self.child_id = child.id
self._attr_unique_id = f"{entry.entry_id}_{child.id}_points"
self._attr_name = f"{child.name} Points"
self._attr_state_class = SensorStateClass.TOTAL
@property
def native_value(self) -> int:
"""Return the child's current points."""
child = self.coordinator.get_child(self.child_id)
return child.points if child else 0
@property
def native_unit_of_measurement(self) -> str:
"""Return the unit of measurement."""
return self.coordinator.data.get("points_name", "Stars")
@property
def icon(self) -> str:
"""Return the icon."""
return self.coordinator.data.get("points_icon", "mdi:star")
@property
def extra_state_attributes(self) -> dict:
"""Return additional attributes."""
child = self.coordinator.get_child(self.child_id)
if not child:
return {}
return {
"child_id": child.id,
"child_name": child.name,
"avatar": child.avatar,
"total_points_earned": child.total_points_earned,
"total_chores_completed": child.total_chores_completed,
"current_streak": child.current_streak,
"best_streak": child.best_streak,
"career_score": child.career_score,
"total_penalties_received": child.total_penalties_received,
}
class ChildStatsSensor(TaskMateBaseSensor):
"""Sensor for a child's statistics."""
_attr_translation_key = "child_stats"
_unrecorded_attributes = frozenset({"assigned_chores", "chore_order"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
child: Child,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator, entry)
self.child_id = child.id
self._attr_unique_id = f"{entry.entry_id}_{child.id}_stats"
self._attr_name = f"{child.name} Stats"
self._attr_state_class = SensorStateClass.TOTAL
@property
def native_value(self) -> int:
"""Return the child's total chores completed."""
child = self.coordinator.get_child(self.child_id)
return child.total_chores_completed if child else 0
@property
def icon(self) -> str:
"""Return the icon."""
child = self.coordinator.get_child(self.child_id)
return child.avatar if child else "mdi:account-circle"
@property
def extra_state_attributes(self) -> dict:
"""Return additional attributes."""
child = self.coordinator.get_child(self.child_id)
if not child:
return {}
# Get chores assigned to this child. For alternating/random/balanced
# assignment, only include a chore when this child is the active one
# today. first_come is competitive: the whole pool sees it (no single
# active child) until one member fills the shared quota. Either way,
# drop it once any pool member has completed it today (so a parent
# crediting the off-rotation child clears the chore for everyone).
chores = self.coordinator.data.get("chores", [])
def _included(c):
if not (child.id in c.assigned_to or not c.assigned_to):
return False
# Won at auction (#982): today's occurrence is the winner's alone.
winner = self.coordinator.auction_winner(c)
if isinstance(winner, str) and winner:
return winner == child.id
mode = getattr(c, "assignment_mode", "everyone")
if mode not in ("everyone", "first_come") and getattr(c, "assignment_current_child_id", "") != child.id:
return False
if mode != "everyone" and self.coordinator._is_rotation_done_today(c):
return False
return True
assigned_chores = [c for c in chores if _included(c)]
return {
"child_id": child.id,
"child_name": child.name,
"avatar": child.avatar,
"points": child.points,
"total_points_earned": child.total_points_earned,
"total_chores_completed": child.total_chores_completed,
"current_streak": child.current_streak,
"best_streak": child.best_streak,
"career_score": child.career_score,
"total_penalties_received": child.total_penalties_received,
"assigned_chores": [
{"id": c.id, "name": c.name, "points": c.points, "time_category": c.time_category}
for c in assigned_chores
],
"chore_order": child.chore_order,
}
class ChildWishlistSensor(TaskMateBaseSensor):
"""A child's open wishes (#932), for the wishlist card.
One per child rather than one for the family: a wish can carry a name,
a link, a picture and a row of pledge chips, and a family's worth of
those would not fit under the recorder's 16 KB attribute limit. The card
finds this sensor by ``wishlist_child_id``. The redeemed history is left
to the admin panel.
"""
_attr_icon = "mdi:heart-outline"
_unrecorded_attributes = frozenset({"wishes"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
child: Child,
) -> None:
super().__init__(coordinator, entry)
self.child_id = child.id
self._attr_unique_id = f"{entry.entry_id}_{child.id}_wishlist"
self._attr_name = f"{child.name} Wishlist"
@property
def native_value(self) -> int:
"""Wishes on the go: waiting, saving, or waiting to be handed over."""
return sum(
1
for w in self.coordinator.storage.get_wishes()
if w.child_id == self.child_id and w.status in ("pending", "active", "redeem_requested")
)
@property
def extra_state_attributes(self) -> dict:
return {
"wishlist_child_id": self.child_id,
"wishes": self.coordinator.wishlist_sensor_rows(self.child_id),
"wish_max_open": WISH_MAX_OPEN_PER_CHILD,
}
class ChildWatchSensor(TaskMateBaseSensor):
"""A child's day in a few characters, for watch complications (#978).
The state is a short, already-translated phrase ("3 left", "All done")
because the companion apps render complication templates server-side and
show the raw state — the frontend's state translation never reaches the
wrist. The attributes are a handful of scalars so templates can build
gauges and rings (``progress`` is 0–1) without touching the heavy sensors.
"""
_attr_icon = "mdi:watch"
_attr_translation_key = "child_watch"
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
child: Child,
) -> None:
super().__init__(coordinator, entry)
self.child_id = child.id
self._attr_unique_id = f"{entry.entry_id}_{child.id}_watch"
self._attr_name = f"{child.name} Watch"
self._state_strings: dict[str, str] = dict(DEFAULT_STATE_STRINGS)
self._summary_key: tuple | None = None
self._summary: dict | None = None
def _current_summary(self) -> dict | None:
"""The summary once per coordinator update — state and attributes share it."""
key = (
id(self.coordinator.data),
getattr(self.coordinator, "external_state_version", 0),
dt_util.as_local(dt_util.now()).date(),
)
if key != self._summary_key:
self._summary = watch_summary(self.coordinator, self.child_id)
self._summary_key = key
return self._summary
async def async_added_to_hass(self) -> None:
"""Load the state phrases in the server's language."""
await super().async_added_to_hass()
self._state_strings = await _async_watch_state_strings(self.hass)
@property
def native_value(self) -> str | None:
return watch_state(self._current_summary(), self._state_strings)
@property
def extra_state_attributes(self) -> dict:
return self._current_summary() or {}
async def _async_watch_state_strings(hass: HomeAssistant) -> dict[str, str]:
"""The watch sensor's state phrases from TaskMate's entity translations.
Falls back to English per phrase, so a missing or broken translation
never leaves the sensor without a state.
"""
strings = dict(DEFAULT_STATE_STRINGS)
try:
from homeassistant.helpers.translation import async_get_translations
translations = await async_get_translations(hass, hass.config.language, "entity", {DOMAIN})
except Exception: # noqa: BLE001 - translations are cosmetic; never block the sensor
_LOGGER.debug("Could not load TaskMate watch translations", exc_info=True)
return strings
prefix = f"component.{DOMAIN}.entity.sensor.child_watch.state."
for key in strings:
value = translations.get(prefix + key) if isinstance(translations, dict) else None
if isinstance(value, str) and value:
strings[key] = value
return strings
class ChildBadgesSensor(TaskMateBaseSensor):
"""Sensor exposing a child's earned and available badges."""
_attr_icon = "mdi:trophy-award"
_attr_translation_key = "child_badges"
_unrecorded_attributes = frozenset({"earned", "available"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
child: Child,
) -> None:
super().__init__(coordinator, entry)
self.child_id = child.id
self._attr_unique_id = f"{entry.entry_id}_{child.id}_badges"
self._attr_name = f"{child.name} Badges"
@property
def native_value(self) -> int:
"""Number of badges earned by this child."""
return len(self.coordinator.storage.get_awarded_badges_for_child(self.child_id))
@property
def extra_state_attributes(self) -> dict:
from .coord_badges import resolve_metric
storage = self.coordinator.storage
child = storage.get_child(self.child_id)
if child is None:
return {"earned": [], "available": [], "total_badges": 0}
all_badges = [b for b in storage.get_badges() if b.enabled]
applicable = [b for b in all_badges if not b.assigned_to or self.child_id in b.assigned_to]
awarded_records = storage.get_awarded_badges_for_child(self.child_id)
record_by_id = {a.badge_id: a for a in awarded_records}
earned: list[dict] = []
available: list[dict] = []
for b in applicable:
if b.id in record_by_id:
rec = record_by_id[b.id]
earned.append(
{
"badge_id": b.id,
"name": b.name,
"icon": b.icon,
"tier": b.tier,
"earned_at": rec.earned_at.isoformat() if rec.earned_at else None,
"manually_awarded": rec.manually_awarded,
"silent": rec.silent,
}
)
else:
# The criterion that actually gates the award: for an AND badge
# that is the weakest one, for an OR badge the strongest. Using
# min() for both understated OR badges (and could report <100%
# for one already earned) — see coord_badges.BadgeEvaluator.
combinator = (getattr(b, "combinator", "AND") or "AND").upper()
scored = []
for c in b.criteria:
cur = resolve_metric(c.metric, child, storage)
pct = min(100, int(100 * cur / max(c.value, 1)))
scored.append((pct, c.metric, cur, c.value))
if not scored:
progress_pct = 0
closest = None
else:
pick = max(scored) if combinator == "OR" else min(scored)
progress_pct = pick[0]
closest = {"metric": pick[1], "current": pick[2], "target": pick[3]}
available.append(
{
"badge_id": b.id,
"name": b.name,
"icon": b.icon,
"tier": b.tier,
"progress_pct": progress_pct,
"closest_criterion": closest,
"criteria_summary": ", ".join(f"{c.metric} >= {c.value}" for c in b.criteria),
}
)
earned.sort(key=lambda e: e.get("earned_at") or "", reverse=True)
return {
"earned": earned,
"available": available,
"total_badges": len(applicable),
}
class PendingApprovalsSensor(TaskMateBaseSensor):
"""Sensor for pending approvals."""
_unrecorded_attributes = frozenset({"chore_completions", "reward_claims", "mandatory_misses"})
def __init__(
self,
coordinator: TaskMateCoordinator,
entry: ConfigEntry,
) -> None:
"""Initialize the sensor."""
super().__init__(coordinator, entry)
self._attr_unique_id = f"{entry.entry_id}_pending_approvals"
self._attr_name = "Pending Approvals"
self._attr_state_class = SensorStateClass.MEASUREMENT
@property
def native_value(self) -> int:
"""Return the number of pending approvals."""
pending_completions = self.coordinator.data.get("pending_completions", [])
pending_rewards = self.coordinator.data.get("pending_reward_claims", [])
return len(pending_completions) + len(pending_rewards)
@property
def icon(self) -> str:
"""Return the icon."""
return "mdi:clipboard-clock"
@property
def extra_state_attributes(self) -> dict:
"""Return additional attributes."""
pending_completions = self.coordinator.data.get("pending_completions", [])
pending_rewards = self.coordinator.data.get("pending_reward_claims", [])
completion_details = []
for comp in pending_completions:
child = self.coordinator.get_child(comp.child_id)
# A bounty's completion (#931) is approved in the same queue.
chore = self.coordinator.get_chore(comp.chore_id) or self.coordinator.bounty_completion_chore(comp)
if child and chore:
# Bonus sub-task completions carry the sub-task's own name and
# points; mirror _build_todays_completions so cards reading this
# list render them identically.
bonus_subtask_id = getattr(comp, "bonus_subtask_id", "") or ""
if bonus_subtask_id:
subtask = next((b for b in chore.bonus_subtasks if b.id == bonus_subtask_id), None)
chore_name = f"{chore.name} › {subtask.name}" if subtask else chore.name
pts = subtask.points if subtask else 0
else:
chore_name = chore.name
pts = chore.points
timed_secs = getattr(comp, "timed_duration_seconds", 0) or 0
if timed_secs > 0 and getattr(chore, "task_type", "") == "timed":
rate_seconds = chore.timed_rate_minutes * 60
if rate_seconds > 0:
pts = (timed_secs // rate_seconds) * chore.timed_rate_points
detail = {
"completion_id": comp.id,
"type": "chore",
"child_name": child.name,
"child_id": child.id,
"chore_name": chore_name,
"chore_id": chore.id,
"points": _completion_award(comp, pts),
"time_category": chore.time_category,
"completed_at": comp.completed_at.isoformat(),
"bonus_subtask_id": bonus_subtask_id,
}
if timed_secs > 0:
detail["timed_duration_seconds"] = timed_secs
if getattr(comp, "bounty_id", ""):
detail["bounty_id"] = comp.bounty_id
# Bare (unsigned) path; the card signs per-viewer via
# auth/sign_path so no self-authenticating URL lands in this
# world-readable attribute.
photo = getattr(comp, "photo_url", "") or ""
if photo:
detail["photo_url"] = photo
# The child's description and their own estimate (#832). Both
# are emitted only when set — an ordinary tap-to-complete
# carries neither, and this slice is capped at 16KB.
note = getattr(comp, "note", "") or ""
if note:
detail["note"] = note
suggested = getattr(comp, "suggested_points", 0) or 0
if suggested:
detail["suggested_points"] = suggested
completion_details.append(detail)
reward_details = []
for claim in pending_rewards:
child = self.coordinator.get_child(claim.child_id)
reward = self.coordinator.get_reward(claim.reward_id)
if reward is None and getattr(claim, "wish_id", ""):
reward = _wish_as_reward(self.coordinator.get_wish(claim.wish_id))
if child and reward:
reward_details.append(
{
"claim_id": claim.id,
"type": "reward",
"child_name": child.name,
"child_id": child.id,
"reward_name": reward.name,
"reward_id": reward.id,
"cost": reward.cost,
"claimed_at": claim.claimed_at.isoformat(),
}
)
mandatory_misses = self.coordinator.mandatory_misses_state()
attrs = {
"pending_chore_completions": len(pending_completions),
"pending_reward_claims": len(pending_rewards),
"pending_mandatory_misses": len(mandatory_misses),
"chore_completions": completion_details,
"reward_claims": reward_details,
"mandatory_misses": mandatory_misses,
}
# Quality rating (#927): tells approval cards to draw the star picker,
# and what each star pays. Absent entirely while the feature is off.
if self.coordinator.quality_rating_enabled():
attrs["quality_rating"] = {
"multipliers": [round(m, 3) for _, m in sorted(self.coordinator.quality_rating_multipliers().items())]
}
return attrs