[GreenGrassBlueOcean]

Related: #116 (review 5538969896), #119, #97, #117, OpenWebNet-HA/MyHOME#370, OpenWebNet-HA/MyHOME#466, and OpenWebNet-HA/MyHOME#714.

What is the problem?

Following the merge of #116 (fix(heating,gateway): dimension 19 is_active, plain protection form and device type 200 from a live F461 capture) and #97 (feat/who4-consolidated-heating), post-merge review by @fedem95 (review 5538969896) and follow-up review on this PR (review 5540201753) plus the master merge of #97 identified five items:

1. WHO 13 Dimension 15 code 200 omitted MH201:
   OWNd/message/gateway.py mapped *#13**15*200## to "F454 / MyHomeServer1 / MH202 / F461 / H4890" and described it as "shared by every Linux-based gateway generation". However, physical MH201 gateways (which are catalogued in the firmware oracle on 030644 as a bare-metal image rather than one of the Linux gateways emulated by QEMU) also reply *#13**15*200## on firmware 3.3.6 and 3.6.27. Furthermore, on MH201 (fw 3.6.27), WHO 1013 dimension 1 only distinguishes the model when queried with an empty WHERE (*#1013**1## → *#1013**1*10*15*0*5##, model 10), whereas *#1013*0*1## is NACKed and no unsolicited WHO 1013 frame is pushed after the 200 reply.
2. Dead / redundant self._is_active assignments in WHO 4 Dimension 19 (OWNd/message/heating.py):
   After #116 consolidated self._is_active = bool(self._is_cooling or self._is_heating or self._cooling_fan_on or self._fan_on) at the end of the dimension 19 block, the initial self._is_active = self._is_cooling | self._is_heating assignment and the two self._is_active = True assignments inside the cooling/heating fan-speed branches were unconditionally overwritten and redundant.
3. Stale fixture & oracle path comments in OWNd/message/heating.py and tests/test_f461_live_plant.py:
   Two comments in OWNd/message/heating.py still referenced tests/fixtures/f461_live_2026-10-15/ (removed in ca0519e when the F461 frames were inlined into tests/test_f461_live_plant.py) and the legacy oracle3/out/termo_f461.tsv path (now superseded by the canonical thermo-zone-mode.tsv, ownd-pr82-heating.tsv, and thermo-ts10.tsv suites in own-firmware-oracle), and the module docstring in tests/test_f461_live_plant.py did not yet list MH201 under device type 200.
4. Missing OWNd/message/door_entry.py row in README.md and scripts/sync_documentation.py:
   The coverage table in README.md and COMPONENT_NOTES in scripts/sync_documentation.py were missing an entry for OWNd/message/door_entry.py (WHO 6 & WHO 8).
5. Semantic conflict between PR #116 and PR #97 on WHO 4 bare WHAT 0 / 1 season frames (tests/test_f461_live_plant.py):
   PR #97 (6af9377) aligned bare *4*0*Z## / *4*1*Z## frames with the Legrand WHO 4 specification (season == "conditioning" / "heating", message_type == "hvac_season", mode is None), causing test_season_frames_and_area_zero_lights_parse_as_before (added in #116 asserting .mode == "heat" / "cool") to fail on pull_request merge commits against master.

---

How is it solved?

1. Include MH201 in WHO 13 code 200 and document firmware oracle provenance (OWNd/message/gateway.py):
   • Updated self._device_type for WHO 13 dimension 15 code 200 to "F454 / MyHomeServer1 / MH202 / F461 / H4890 / MH201".
   • Updated the inline comment to clarify that code 200 is shared across modern gateways (F454, MyHomeServer1, MH202, F461, H4890; plus emulated F450/F459/F460 in trace-corpus.tsv) and MH201 (fw 3.3.6, 3.6.27; catalogued on 030644 outside the Linux QEMU fleet), and that WHO 1013 dimension 1 tells them apart (the MH201 on 3.6.27 answers only *#1013**1##, model 10, and NACKs *#1013*0*1##).
   • Updated corresponding assertions and module docstring in tests/test_profiles.py and tests/test_f461_live_plant.py.
2. Remove dead self._is_active assignments (OWNd/message/heating.py):
   • Removed the dead self._is_active = self._is_cooling | self._is_heating assignment and the two redundant self._is_active = True assignments in the cooling and heating fan-coil branches, leaving the single authoritative bool(self._is_cooling or self._is_heating or self._cooling_fan_on or self._fan_on) computation after both sides are parsed.
3. Update F461 capture & firmware oracle suite comments (OWNd/message/heating.py, tests/test_f461_live_plant.py):
   • Pointed both F461 capture comments (dimension 19 active state and standalone protection addressing) to tests/test_f461_live_plant.py.
   • Replaced the legacy oracle3/out/termo_f461.tsv comment in OWNHeatingCommand.set_mode with exact own-firmware-oracle suite references (thermo-zone-mode.tsv, ownd-pr82-heating.tsv, and thermo-ts10.tsv) and SCS bus telegrams (D1 00 03 02 C2 ZZ 00 00 for 311, 93 ZZ 30 04 / 93 ZZ 30 0F for plain zone 302/303, and D1 00 03 02 C2 ZZ 07 00 for #Z central-unit 302).
4. Synchronize door_entry.py documentation (README.md, scripts/sync_documentation.py):
   • Added "OWNd/message/door_entry.py": "WHO 6 & WHO 8: Door entry, electric lock, and video intercom events/commands" to COMPONENT_NOTES in scripts/sync_documentation.py and added the corresponding 100% coverage row to README.md.
5. Rebase onto master (6af9377) and align F461 season assertion (tests/test_f461_live_plant.py):
   • Updated test_season_frames_and_area_zero_lights_parse_as_before to assert OWNEvent.parse("*4*1*2##").season == "heating" and OWNEvent.parse("*4*0*2##").season == "conditioning".

---

How is it tested?

• Unit & Live-Capture Tests:
  • tests/test_f461_live_plant.py: All 17 F461 live-plant tests pass, including all 9 WHO 4 dimension 19 actuator/fan-coil combinations (5*1, 5*0, 8*0, 1*5, 5*5, 0*6, 0*1, 1*0, 0*0), test_device_type_200_is_not_a_single_model (asserting "MH201" in msg.device_type), and test_season_frames_and_area_zero_lights_parse_as_before.
  • tests/test_profiles.py: test_owngateway_event_properties verifies OWNGatewayEvent("*#13**15*200##").device_type == "F454 / MyHomeServer1 / MH202 / F461 / H4890 / MH201" and device_type_code == "200".
• Full Test Suite & Coverage:
  • pytest: 1,488 passed, 1 skipped on rebased master.
  • Statement coverage: 100.0% (4,808 / 4,808 statements across all 25 modules verified by scripts/verify_coverage.py).
• Static Analysis & Documentation Checks:
  • ruff check, mypy, scripts/verify_library_standards.py, and scripts/sync_documentation.py --check: 0 violations.

---

Where are the traces (real and firmware oracle)?

1. Real Hardware Traces:

   • BTicino F461 (fw 2.0.11, captured 2026-10-15):
     • Inlined in OWNd: tests/test_f461_live_plant.py (GATEWAY_FRAMES, ACKED_FRAMES, NACKED_FRAMES, EVENTS_AFTER), including *#13**15*200## followed by *#1013**1*134*15*5*0##, WHO 4 dimension 19 frames (*#4*2*19*5*1##, *#4*2*19*5*0##, *#4*2*19*8*0##), season frames (*4*1*2##, *4*0*2##), and standalone protection mode transitions (*4*302*2## → *4*202*2##).
     • Full multi-session capture archive: Stored in the MyHOME trace corpus at tests/fixtures/traces/f461_live_2026-10-15/ (OpenWebNet-HA/MyHOME#714).
   • BTicino MH201 (fw 3.6.27 and 3.3.6):
     • fw 3.6.27 (captured 2026-10-14 by @fedem95): Reports *#13**15*200##, NACKs *#1013*0*1##, and answers *#1013**1## with *#1013**1*10*15*0*5## (model 10); plant fixture archived in OpenWebNet-HA/MyHOME at tests/fixtures/plants/mh201_physical_plant/.
     • fw 3.3.6: Reports *#13**15*200## in OpenWebNet-HA/MyHOME at tests/fixtures/traces/issue_466/myhome_sweep_MH201_all_2026-09-25T19-11-35.json (OpenWebNet-HA/MyHOME#466).
   • F454 / MyHomeServer1 / MH202 / H4890:
     • Captured in OpenWebNet-HA/MyHOME issues #370 and #466 and catalogued in tests/golden/frames/who13_gateway.yaml.

2. Firmware Oracle Traces & Disassembly (OpenWebNet-HA/own-firmware-oracle, synced in tests/golden/firmware_oracle.json, tests/oracle/who4_facts.json, & #117):

   • WHO 13 Dimension 15 (*#13**15##) & WHO 1013 Dimension 1 (*#1013*0*1##) (trace-corpus suite, results/<PRODUCT>/<VERSION>/oracle/full/trace-corpus.tsv):
     • Across the 11 QEMU-emulated gateways (F450 020010, F453AV 030014, F454 020051, F459 020105, F460 020012, F461 020011, H4684 020054, MH200 020102, MH200N 010108, MH202 010024, MyHomeServer1 028206):
       • F450, F454, F459, F460, F461, MH202, and MyHomeServer1 all reply *#13**15*200## (while H4684 replies *#13**15*13##, MH200 replies *#13**15*4##, and F453AV / MH200N reply *#13**15*0##).
       • MH201 (030644, SHA-256 8972720b01d1...) is catalogued in results/mcp_index.json (status: "catalogued"), not among the Linux QEMU-emulated targets.
       • All 11 emulated gateways accept *#1013*0*1## with ACK (H4684 emits *#1013**1*19##, MH200 emits *#1013**1*4##, and the unprovisioned Linux images emit *#1013**1*0*0*0*0## when AS_ITEM_SYSTEM is unpopulated in the clean QEMU rootfs), whereas on physical hardware MH201 (fw 3.6.27) NACKs *#1013*0*1## and answers only *#1013**1## (*#1013**1*10*15*0*5##).
   • WHO 4 Mode & Protection Addressing (ownd-pr82-heating, thermo-zone-mode, thermo-central-mode, thermo-ts10 suites):
     • Automatic mode (311): Plain *4*311*1## and *4*311*99## are NACKed by all 11 emulated gateways with zero bus frames emitted, whereas #Z (*4*311*#1##, *4*311*#99##, *4*311*#23#1##) is translated and dispatched onto the SCS bus (24 30 36 44 31 30 30 30 33 30 32 43 32 30 31 30 30 30 30 0d → D1 00 03 02 C2 01 00 00).
     • Standalone vs Central-Unit Zone Mode (302 / 102 / 202 / 303): Emulated bt_termo translates plain zone commands into direct 4-byte SCS zone telegrams (e.g., *4*302*23## → 93 17 30 04; *4*303*1## → 24 30 34 39 33 30 31 33 30 30 46 0d / 93 01 30 0F in thermo-zone-mode.tsv), whereas #Z commands (*4*302*#23##, *4*302*#23#1##, *4*102*#0##, *4*202*#0##) are translated into extended D1 central-unit frames (D1 00 03 02 C2 17 07 00 / 24 30 36 44 31 30 31 30 33 30 32 43 32 31 37 30 37 30 30 0d in thermo-ts10.tsv).
   • WHO 4 Dimension 19 & Season (thermo-valve suite & tests/oracle/who4_facts.json):
     • *#4*1*19## is replayed in results/<PRODUCT>/<VERSION>/oracle/full/thermo-valve.tsv (no virtual fan-coil/valve actuator is attached on the emulated bus to emit *#4*Z*19*COOL*HEAT## responses, so the 9 actuator/fan-coil combinations are verified via the live F461 capture in tests/test_f461_live_plant.py), and tests/oracle/who4_facts.json (1,745 decoded firmware facts) verifies the WHO 4 season contract (*4*0*Z## / *4*1*Z##).

[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!

[fedem95]

Thanks for picking these up. Tested 6869de1: 1354 passed locally (test_metadata deselected, local env), CI green. The fan-branch _is_active = True removals don't change behaviour, since the final bool(...) already covers cooling_fan_on/fan_on, and the nine dimension 19 cases in test_f461_live_plant.py still pass. Approving.

One nit on the new comment, non-blocking: model 10 comes from my MH201 on 3.6.27 only; I haven't seen a *#1013 reply from a 3.3.6. And the other gateways in the traces were asked *#1013*0*1##, not *#1013**1##, so "*#1013**1## tells them apart" is only shown for the MH201 so far. Maybe: "WHO 1013 dimension 1 tells them apart (the MH201 on 3.6.27 answers only *#1013**1##, model 10)".

(The README also gains the missing door_entry.py coverage row, which isn't in the summary.)

[GreenGrassBlueOcean]

Thanks @fedem95! Addressed both points in 34e23e3:

1. Updated the WHO 13 code 200 comment in OWNd/message/gateway.py to your exact phrasing (WHO 1013 dimension 1 tells them apart (the MH201 on 3.6.27 answers only *#1013**1##, model 10)).
2. Added OWNd/message/door_entry.py to COMPONENT_NOTES in scripts/sync_documentation.py (so README.md has the proper WHO 6 & WHO 8 description instead of the generic fallback) and documented it in the PR summary.

[fedem95]

Reviewed the latest commit 13dc38f. The follow-ups are addressed, and CI and the MyHOME conformance checks are green. Approving.