The 0 W of shutdown and of the Stop button now waits (bounded, 2 s) for a
write already on its way to the same unit, for every driver, GoodWe included.
Once shutdown has begun, cycle, heartbeat and safety-stop writes are dropped
right before the write is created. One shared wait (_await_inflight) serves
the 0 W and SMA-10's _release_last.
Co-Authored-By: Claude <noreply@anthropic.com>
The NL default for a hand-back driver without words (handback._BACK_NL) was
changed to "de eigen sturing" in round 2 at the PR #239 review's request
(comment 2495: the article as in the generic clause). The brief does not name
it, so DECISIONS now says where it came from and a test checks the exact Dutch
sentence. No app code change.
Co-Authored-By: Claude <noreply@anthropic.com>
When control takes the unit again, release_fails, release_retry_at and
release_zero_sent are reset, so the next Stop is a new streak with its own one
0 W, error line and event. Tests: a streak that Start interrupts without a
recovery; a driver that overrides release_fallback(). _BACK_NL "de eigen
sturing" (round 2) is kept.
Co-Authored-By: Claude <noreply@anthropic.com>
When release() fails (False, raises or hangs past RELEASE_BUDGET_S) the unit
gets one bounded write(0) through the new Driver.release_fallback() hook
(default write(0)), then nothing more, so the inverter's own fallback timer is
not reset. The release is retried once per heartbeat, not on every idle
write. One error line and one action-log event per failure streak, one line on
recovery. Dutch default "terug naar de eigen sturing". DECISIONS says what a
unit is left at.
test_messages.py: the RELEASE_BUDGET_S data entry is gone again (the line
that needed it is gone).
Co-Authored-By: Claude <noreply@anthropic.com>
A driver now says whether its release() hands the inverter back
(Driver.hands_back; GoodWe is False, so release() stays write(0) and the
Controller never calls it). command_site_release() sits next to
command_site_zero(): the Stop button and shutdown() (so SIGTERM, restart,
control task death and a late config refusal) hand a hand-back unit back,
once, and zero every other unit as before. Safety stops and refusals still
hold 0 W for every unit. Shadow mode releases a hand-back unit. After a
release nothing is written to the unit (heartbeat and cycle included) until
control starts again. release() is bounded by RELEASE_BUDGET_S (= the write
budget), runs concurrently per unit, and a raising or hanging release on one
unit does not stop the others.
The "set to 0 W" sentences for Stop, restart, shadow mode and the stopped
banner are built from the roster (app/handback.py, EN and NL, words from the
driver); a GoodWe-only roster keeps every old literal. A handed-back unit
shows "Back under <its control>" on Home.
Existing test edited: test_messages.py (two classification entries for the
CRITICAL line that moved into a roster builder, and the new release bound).
Co-Authored-By: Claude <noreply@anthropic.com>
Owner decision 2026-11-07: clients see nothing new until SMA-4. config.yaml,
translations, labels and the page layout are back to 23ff9d0's. The validation
of the keys stays for callers that skip the schema.
Co-Authored-By: Claude <noreply@anthropic.com>
app/drivers.py holds one Driver per roster entry (configured, actuator_id,
limits, read, write, release, probe, link_stale_s, batt_bound, settle_s, brand,
words). GoodWeEsphomeDriver makes the calls the code it replaced made, in the
same order. Inverter gets driver; has_writer, is_live and link_note read it. The
SoC/power read, the write, the device range (run_control, amain, the settings
panel), rule 2 per unit and the SAFETY-28 census go through it.
setpoint_entity is optional per entry beside driver, host, port, unit_id and
maintenance; only goodwe_esphome is selectable, the rest are checked and unused.
Co-Authored-By: Claude <noreply@anthropic.com>
Before the bell ends it replaced today's Home line and the event log with
the kept-for-tonight wording. stand_in() docstring covers the tomorrow case.
Co-Authored-By: Claude <noreply@anthropic.com>
One stale Forecast.Solar tomorrow sensor dropped the whole OPT-17 sum from
17:54 to 20:00. Today readings and the OPT-34 morning rules are unchanged.
Co-Authored-By: Claude <noreply@anthropic.com>
At a 5 s meter the slow meter fit now finds 30 s late first, and the hold leaves
the batteries' fit nothing to measure, so the row says it without a figure.
Co-Authored-By: Claude <noreply@anthropic.com>
The meter's delay fit needed 16 new values in 60 s, so at the owner's 5 s P1
only the batteries' fit ran and 30 s late was found after about 3 minutes
(export to -420 W). A meter that slow is now also fitted over 120 s; it calls
a delay late only over 16 s, on two fits in a row, with the same corroboration
as before, and never ends a verdict, so a slow battery (R2-1) runs as before.
Found about 90 s after the start, export above -200 W. 10 s late at a 5 s
meter is not found (nothing swings; it cannot be told from a 10 s battery).
Co-Authored-By: Claude <noreply@anthropic.com>
The cheap-run roll (cheaproll._Charge) shared the peak room by rating, control
shares it by room left to full (packshare.weights). With unequal levels one
battery's drawn block ended too early (rig F-1). The roll now uses
packshare.weights over the simulated levels, each battery capped at its rating,
a full one dropping out. test_ux77 gains two unequal-level walks and a
by-rating mutant; its walks model control's split by room too.
Co-Authored-By: Claude <noreply@anthropic.com>
The cheap run is sized at the full rating, so under the peak limit a battery is
still short when it ends, and control plans the next run then. Home drew only
the run under way. `cheaproll` rolls the run on the way control does
(`modetrade.energy()` + `priceruns.plan()` at each run end, lifted from
`dayview._levels()`), reading the levels at the charge power the peak limit
leaves (peak aim less the house, else the rating). `homeview.mode_runs()` and
`price_windows()` use it for today: the grid block, the Today line and the
chart's cheap bars run to where control stops, each battery's block to its
own stop. Display only: no control change.
test_ux54: today's grid block of the night-before scenario now follows the roll.
Co-Authored-By: Claude <noreply@anthropic.com>
Brings OPT-36 (#229) and TEST-37 (#231) onto release/1.0, which keeps
SAFETY-128 (#228). No textual conflicts. CHANGELOG: the SAFETY-128 entry
moved out of the released 0.20.0 section into "Not in a release yet" above
## 0.20.1; the 0.20.0 and 0.20.1 sections are identical to master.
Co-Authored-By: Claude <noreply@anthropic.com>
The stale_live scenario stores a latch ending 2026-11-04 07:15 CET. The first
Controller was built before the datetime patch, so Controller.__init__ judged
that latch against the real clock and it expired on 2026-11-04. Build it after
the patch (clock = scenario start), as the restart path already did. Test-only.
Co-Authored-By: Claude <noreply@anthropic.com>
SAFETY-132: the 9.6 kW is the charge the add-on asked for in a test,
not a measured grid figure. SAFETY-133: the quarter also ends over the
3 500 W limit; the guard sees the load late (0.20.1 has no lateness check).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Patch bump: OPT-36 (Leftover solar counts Cheap hours only and Solar first batteries). No option changes.
Co-Authored-By: Claude <noreply@anthropic.com>
Review F1: a slot after B's run counts only if no slot from the expensive
hours on is cheaper (control plans B's charge left in the cheapest window
ahead), and x 0.95 of B's charge is stored. Three-price night: 1ac7fe9 left
the expensive hours 2.61 kWh short; now 0.
Review F2: a Solar first battery's drain until the expensive hours reads each
hour at no less than the night base (the profile is net of the sun). Read net,
a dark morning drained the battery the plan still counted; A's level fell,
the cheap run moved and was lost before it started (+0.40 EUR per 3 days).
The two real-price replays now cost what 0.20.0 does.
F3: the ux54 sweep's peak limit is 15 kW (the option's top). F4: the rest
comes from the grid (EN/NL). F5: CHANGELOG and DECISIONS wording.
Co-Authored-By: Claude <noreply@anthropic.com>
The run is sized at the batteries' ratings, not the peak limit, so a run too
short to fill a Cheap hours battery is planned again later in the night.
Counting that run alone made B's give grow through the night and A's level
fall after A had bought (round 2: 55 -> 69 -> 41 -> 37 %, A to 56 %).
dayplan._cheap_give_kwh() now also counts the later firm slots before the
expensive hours that a run would still be worth charging in (price + the
spread control asks at most the give run's dearest slot); the Controller
passes the slots and the spread.
test_opt34: back to the original house shapes, fixture and expected values.
B is Off in the nights that test A's level (a Cheap hours B now covers
these hours); B's checks run in new `_b` nights; the 03:00 check runs in a
night from midnight; tail starts A at 60 %; late_fresh reads the kept 6.0.
test_opt36: slot cases, a whole night on one figure (level never moves),
F-1 with B counted (raised level reached and held). test_ux54: the run end
is read from the plan and asserted (02:30); the F1/F2 sweep uses a peak
limit that does not bind and reads a level shown as the target as at it.
Co-Authored-By: Claude <noreply@anthropic.com>
The silent and fresh nights diverged only because the test's fresh sensors
kept reading the evening's 0.4 kWh 'left today' after midnight, while the
kept figure is the 4th's 6.0 kWh (solarfc.Hold.stand_in, as_today): two
different shortfalls (16.5 and 12.7 kWh) that A's ceiling hid until B was
counted. Fresh sensors now read 6.0 after midnight, so fresh and kept are
the same figure; the 0.3 tolerance is back, the other exact values are
recomputed. test_ux54 literals are recomputed too.
Co-Authored-By: Claude <noreply@anthropic.com>
Review B1: the meter fit took repeating house loads (ovens, a heat pump)
for late readings on healthy 1-2 s sites, about 9 % of the recorded
house. B2: at a 5 s meter it never ran. M1: no hysteresis. M2: an
unbounded check.
lateness.Delay now fits how long each battery's reading takes to show
its own request, which the house does not move, from the 1 s reads
(read_inputs -> Answer.watch). The meter fit stays for speed at 1-3 s
meters, but counts only when an ask was slow to start and no battery
fit said on time. Late from 9.5 s after 2 fits, on time under 6.5 s,
dropped after 120 s without a fit. Fits are bounded (a few ms).
Readings on time: no verdict in any reviewer scenario (22 shapes at
1/2/5 s meters, 7 harm runs) nor in 2 days of the recorded house at
1, 2 and 5 s ticks; digests equal to 77fe31b. Late: found 36-68 s after
the start at 1-2 s, about 3 min at 5 s. The row gives the batteries'
delay, or no figure while only the meter has shown it.
Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bdqr7PiaHcBnzPituu2yTC
With Cheap hours B counted, A buys only what B cannot reach. The OPT-34
night scenarios use a 1.6 x larger house profile, and the ux54 tab scenarios
a 4.8 kW evening (ux54_scenarios.make(evening=)), so A still charges; the
hard-coded levels become relations.
Co-Authored-By: Claude <noreply@anthropic.com>
Live 0.20.0: A (Leftover solar, 60 %) charged to 95 % while B (Cheap hours,
94 %) already covered the expensive hours. dayplan.plan() now takes off the
shortfall what Cheap hours batteries will hold when the hours start (the
ceiling or what their cheap run reaches under the peak aim) and what Solar
first batteries will still hold after the night's use. Home reads a give
block after the planned charge, plan_cover() is the plan's own figure, and
the Why line says other batteries cover.
Co-Authored-By: Claude <noreply@anthropic.com>
Readings that Home Assistant delivers late but still changing passed
every staleness gate, so the loop corrected again each cycle an error
its last command had already fixed (10 s late: the give run swung and
exported to -1.9 kW), and the peak guard pushed to the inverter limit
on the same late figures (OPT-34 F-2).
app/lateness.py fits how long after a power request the meter shows
it. Over 8 s the readings are late: a request the battery reading has
not answered holds the loop (it may not lead the reading by more than
100 W) and caps the peak guard at what stands; unanswered for 23 s it
holds, then 0 W after stale_input_s. An amber row and one log line per
episode say so. With readings on time nothing changes (digests equal).
DECISIONS/SAFETY-128-late-readings.md; test_safety128.py.
Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bdqr7PiaHcBnzPituu2yTC
- The OPT-23 headline said a daily-total forecast is now counted right by
day; OPT-30 (known issue) still over-counts by day. Say what changed.
- CI-15: 16 min was one run; runs took 16 to 24 min.
- UX-75: quote the page text as it reads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
admin
Claude