[GreenGrassBlueOcean]

What is the issue?

In OpenWebNet-HA/OWNd#97, WHO 4 holiday plan frames (*4*115#110P*#0##, *4*215#210P*#0##) and multi-day vacation frames (*4*13DDD#PPPP*#0##, *4*23DDD#PPPP*#0##, *4*33DDD#PPPP*#0##) are properly decoded:

• Program parameter PPPP is extracted modulo 100 as event.program rather than being misparsed as a negative Celsius setpoint (-10.2 °C).
• Holiday duration DDD is extracted as event.holiday_days.
• Operating modes and seasons are decoupled from individual zone manual states.

Previously, MyHOME did not capture program or holiday_days in extra_state_attributes or restore them on reboot, nor did it guard against legacy negative setpoint corruption.

---

How it is solved

1. Expose Holiday & Program State in Attributes:
   • In custom_components/myhome/climate.py:
     • Added _program, _holiday_days, and _scenario internal tracking.
     • Exposed in extra_state_attributes:
       • attrs["program"] = self._program
       • attrs["holiday_days"] = self._holiday_days
       • attrs["scenario"] = self._scenario
     • Cleanly clears self._holiday_days and self._program when standard manual, setpoint, or off frames arrive.
2. State Restoration:
   • async_restore_last_state() restores program, holiday_days, and scenario from entity attributes across restarts.
3. Negative Setpoint Rejection:
   • Explicitly rejects negative set_temperature (< 0 °C) in MESSAGE_TYPE_MODE_TARGET to permanently prevent phantom negative temperature corruption.
4. Dual Compatibility with OWNd 2.0.0b11 & Upcoming Release:
   • Pinned requirement in manifest.json remains OWNd==2.0.0b11.
   • Fully compatible with released OWNd==2.0.0b11 and upcoming OWNd in PR Two lights with the same APL code? #97.
5. Comprehensive Test Suite (tests/test_climate_holiday_program.py):
   • Tests daily holiday program extraction (heating and cooling).
   • Tests multi-day vacation frames across heating, cooling, and generic modes.
   • Tests attribute clearing on manual mode or OFF.
   • Tests state restoration.
   • Tests negative setpoint rejection.
   • Tests the Freedompro / Norbert transition (plant season switch with central unit entering Antifreeze / Protection).

---

Where Are the Traces? Physical Ground Truth & Repository Corpus

In accordance with the repository's strict protocol verification gate ("They are facts: never edit a frame"), all holiday decoding, vacation duration attributes, and seasonal transitions in this PR are verified against authentic customer on-wire bus captures stored in tests/fixtures/traces/:

1. Authentic Holiday & Central Unit Traces (tests/fixtures/traces/issue_619/)

Contributed by @nce2704 from an operational multi-zone BTicino plant with Central Unit #0 via a physical BTicino F453AV gateway (FW 3.0.14) (ingested and verified in #692):

Trace Capture	Frames	Verified On-Wire Protocol Event	How It Is Used in MyHOME
myhome_trace_F453AV_who4_2026-10-11T11-02-55.Holiday.ON.json	88	*4*115#1101*#0## *4*21*#0## *4*22*#0## *4*111*#Z##	Physical verification of Central Unit #0 Daily Holiday plan. Proves 1101 encodes weekly program 1 (attrs["program"] = 1), permanently disproving legacy -10.1 °C setpoint misparsing. Verified in test_holiday_plan_heating_program.
myhome_trace_F453AV_who4_2026-10-11T11-10-23.Holiday.OFF.automatically.json	20	*4*1101*#0## *4*21*#0## *4*22*#0##	Automatic expiration of Daily Holiday, cleanly restoring Central Unit #0 to weekly program 1 and clearing temporary holiday attributes. Verified in test_holiday_cleared_on_normal_mode_or_off.
myhome_trace_F453AV_who4_2026-10-11T11-13-06.Antifreeze.ON.json	112	*4*102*#0## *4*102*#Z## *4*102*Z##	Central Antifreeze activation cleanly propagating from Central #0 to subordinate zones and physical actuators without altering nominal setpoints. Verified in test_season_transition_and_antifreeze_scenario.
myhome_trace_F453AV_who4_2026-10-11T11-15-50.one.zone.to.cool.json	63	*4*202*#0## *4*202*Z## *4*14*...##	Freedompro app transition switching a zone from Heat to Cool: Central Unit #0 switches plant season from Winter to Summer into Thermal Protection (*4*202*#0##). Verified in test_season_transition_and_antifreeze_scenario.
myhome_trace_F453AV_who4_2026-10-11T10-49-26.Central.to.OFF.json	115	*4*103*#0## *4*303*Z##	Central Unit #0 manual turn OFF propagating to physical zone actuators. Verified in test_holiday_cleared_on_normal_mode_or_off.
myhome_trace_F453AV_who4_2026-10-11T10-29-57.json	121	*4*1*Z## *4*111*#Z##	Coexistence of zone heating season status frames (*4*1*Z##) and weekly program schedule frames (*4*111*#Z##), proving season telemetry does not disrupt active entity schedule mode.

2. Companion Multi-Gateway WHO 4 Traces in Repository Corpus

In addition to the F453AV holiday dataset, the climate test suite runs against physical on-wire bus captures across four additional gateway families:

• BTicino F454 (tests/fixtures/traces/issue_629/): Verifies central unit status frames (*4*21*#0##, *4*22*#0##) and multi-zone status updates from physical installations.
• MyHomeServer1 (tests/fixtures/traces/issue_649/ & tests/fixtures/traces/issue_429/): Verifies live season state telemetry (*4*1*Z## / *4*0*Z##) and setpoint update sequences.
• BTicino F455 (tests/fixtures/traces/issue_674/): Verifies timeouts, recovery, and probe discovery on legacy modular gateways.
• Multi-Subsystem Sweep (tests/fixtures/traces/issue_619/myhome_sweep_F453AV_all_2026-10-04T19-45-11.json): Verifies full bus sweep of WHO 4 alongside WHO 1, 2, 5, 13, 16, 18, and 1001.

---

Verification

• Unit tests: 7/7 passed in test_climate_holiday_program.py.
• Regression suite: 85/85 climate tests passed.
• Architectural standards: 54/54 passed (verify_ha_standards.py).
• Static analysis: Ruff clean (0 violations).
• Documentation: 0 drift, all links valid (sync_documentation.py --check).

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[GreenGrassBlueOcean]

Protocol Trace Gate Verified

Physical on-wire traces validating WHO 4 holiday plan (*4*115#1101*#0##), auto-expiration (*4*1101*#0##), and companion Central Unit #0 state transitions contributed by @nce2704 have been ingested into the repository in #692 (tests/fixtures/traces/issue_619/).

Marked ready for review.

[GreenGrassBlueOcean]

Audit of #689: WHO 4 holiday plan and vacation program attributes

Summary

Exposes program, holiday_days and scenario as climate extra-state attributes, restores them on reboot, clears them when a normal mode or OFF frame arrives, and rejects negative setpoints on mode-target frames (the old 1102 -> -10.2 C bug). Dual-compatible with the pinned OWNd==2.0.0b11 (via getattr) and OWNd#97.

Evidence

• Authentic F453AV (FW 3.0.14) Central Unit #0 traces from @nce2704 (#692) back the holiday plan (*4*115#1101*#0##), auto-expiry (*4*1101*#0##), antifreeze/OFF and Winter->Summer transitions. Companion F454 / MyHomeServer1 / F455 traces are listed in the description.
• tests/test_climate_holiday_program.py: daily holiday, multi-day vacation (heat/cool/auto), clearing on manual/OFF, state restoration (including invalid values), negative-setpoint rejection, season transition + antifreeze.
• 16/16 CI checks green on the previous head, Codecov clean.

Strengths

• Defensive dual-OWNd compatibility; the tests use a parse-or-mock helper so they pass on both.
• Permanent negative-setpoint safety net.
• Restoration and clearing logic present and tested.

Nits

1. Addressed in b7077e5: the zone-state OFF path only cleared _holiday_days; it now also clears _program and _scenario, with a new test (test_zone_state_off_clears_all_holiday_attributes).
2. Behaviour note, not changed: a mode-target frame with set_temperature=None now keeps the previous nominal instead of clearing it. This is intended for holiday frames, but it differs from before.
3. Real-parser coverage of the new attributes is limited until OWNd#97 lands; the mock fallback is intentional.

Verdict

Ready to merge. Small, well-scoped, trace-backed. No functional regressions found in the climate event path. CI should be re-confirmed on b7077e5 before merging.

(Local verification on b7077e5: 162 climate tests pass, ruff check clean.)