docs: de-identify reporter usernames repo-wide, add PII rule to skill

Extended the earlier de-identification (jhkwon19) to the other issue #177
reporter (HJcom), who was named in ~20 spots across coordinator.py,
diagnostics.py, subdevices.py, identity.py, capabilities/airconditioner.py,
a fixture's seeds_note, and several test docstrings/function names. Swapped
all of it for "the reporter"/"the Pattern A reporter", matching the
convention already established for the other reporter.

Added a rule to the adding-device-support skill (end of "Lock it in"):
don't put a reporter's name or GitHub username in code comments,
docstrings, seeds_note, or test/function names -- that prose ships and
sits in git history indefinitely, unlike an issue thread or a
release-notes thank-you. Use "the reporter" / "issue #NNN's reporter" /
a pattern label instead.
This commit is contained in:
Marc Billow committed 2026-07-30 03:06:06 +00:00
1 parent ce92d71a92
commit ca27bf7f6e
11 files changed
+121 -97

No files matched your search

@@ -368,6 +368,19 @@ The new fixture is picked up automatically by the corpus-wide checks (the
board token fails the build rather than silently mistyping someone's board token fails the build rather than silently mistyping someone's
appliance. appliance.
**Don't put a reporter's name or GitHub username in code.** Fixture data
gets serials/MACs/other device PII scrubbed per point 1 above — the same
rule applies to the *prose* you write while fixing the issue: comments,
docstrings, `seeds_note`, and test/function names should say "the
reporter," "issue #NNN's reporter," or (when a module already distinguishes
multiple reporters, like `subdevices.py`'s Pattern A/Pattern B) "the
Pattern A reporter," never a real name or handle. That prose ships in the
package and lives in git history indefinitely — unlike an issue thread or a
release-notes thank-you (both fine places to credit someone by name), it's
not somewhere a person would expect to stay named forever. If you're fixing
an issue and about to write `<username>'s board`/`<username>'s dump` in a
comment, stop and swap in a generic reference instead.
## 11. Triage: "one of my subdevices is missing" ## 11. Triage: "one of my subdevices is missing"
For an appliance that exposes several logical indoor subdevices over one IP — For an appliance that exposes several logical indoor subdevices over one IP —
+19 -15
View File
@@ -174,9 +174,10 @@ class LocalThingsCoordinator(DataUpdateCoordinator[dict[str, Any]]):
# it's the *other* subdevices sharing this DTLS session, if any. # it's the *other* subdevices sharing this DTLS session, if any.
self.subdevices: list[Subdevice] = [] self.subdevices: list[Subdevice] = []
# Candidates _run_discovery's gate rejected (an unused SmartThings # Candidates _run_discovery's gate rejected (an unused SmartThings
# slot that still answers its seed, e.g. HJcom's /device/2) -- # slot that still answers its seed, e.g. the issue #177 reporter's
# surfaced in diagnostics alongside the materialized ones so a # /device/2) -- surfaced in diagnostics alongside the materialized
# report shows what was found and why it didn't become an entity. # ones so a report shows what was found and why it didn't become an
# entity.
self._skipped_subdevices: list = [] self._skipped_subdevices: list = []
# Those rejected candidates' raw reps, kept aside for diagnostics # Those rejected candidates' raw reps, kept aside for diagnostics
# only (see _live_subdevice_resources). They are deliberately not in the # only (see _live_subdevice_resources). They are deliberately not in the
@@ -408,9 +409,10 @@ class LocalThingsCoordinator(DataUpdateCoordinator[dict[str, Any]]):
"""GET one subdevice's seed Collection and return its batch, """GET one subdevice's seed Collection and return its batch,
normalized to real hrefs. A sibling failing to answer is a debug normalized to real hrefs. A sibling failing to answer is a debug
log, never a failed poll -- the master must not go unavailable log, never a failed poll -- the master must not go unavailable
because a sibling timed out or dropped off (e.g. HJcom's because a sibling timed out or dropped off (e.g. the issue #177
/device/2, a SmartThings-unused component that may not always reporter's /device/2, a SmartThings-unused component that may not
respond). Blocking -- called from _poll_once, already in executor.""" always respond). Blocking -- called from _poll_once, already in
executor."""
sess = self._session sess = self._session
if sess is None: if sess is None:
return {} return {}
@@ -543,8 +545,9 @@ class LocalThingsCoordinator(DataUpdateCoordinator[dict[str, Any]]):
_run_discovery sees every candidate's state without a second poll _run_discovery sees every candidate's state without a second poll
round trip. `_run_discovery` is what narrows self.subdevices down to round trip. `_run_discovery` is what narrows self.subdevices down to
the ones that are actually live (see discover_partitioned) -- this the ones that are actually live (see discover_partitioned) -- this
method doesn't know how to tell an unused SmartThings slot (HJcom's method doesn't know how to tell an unused SmartThings slot (the
/device/2) from a real sibling, only that something answered. issue #177 reporter's /device/2) from a real sibling, only that
something answered.
""" """
if self._session is None: if self._session is None:
self._connect_session() self._connect_session()
@@ -622,8 +625,8 @@ class LocalThingsCoordinator(DataUpdateCoordinator[dict[str, Any]]):
# its own /information/vs/0 when it reports one and falling back to # its own /information/vs/0 when it reports one and falling back to
# the master's registry otherwise. See subdevices.discover_partitioned # the master's registry otherwise. See subdevices.discover_partitioned
# -- it also gates each candidate down to whether it actually # -- it also gates each candidate down to whether it actually
# produced live primary state (HJcom's /device/2, an unused # produced live primary state (the issue #177 reporter's /device/2,
# SmartThings slot, answers its seed but never does), so # an unused SmartThings slot, answers its seed but never does), so
# self.subdevices below is narrowed to the ones that passed, not # self.subdevices below is narrowed to the ones that passed, not
# every candidate _enumerate_subdevices_blocking found. For a device # every candidate _enumerate_subdevices_blocking found. For a device
# with no candidates (self.subdevices == []) this is exactly the # with no candidates (self.subdevices == []) this is exactly the
@@ -641,11 +644,12 @@ class LocalThingsCoordinator(DataUpdateCoordinator[dict[str, Any]]):
skip.subdevice.key, skip.subdevice.kind, list(skip.hrefs), skip.subdevice.key, skip.subdevice.kind, list(skip.hrefs),
) )
# Corroborating signal, not a gate (DESIGN-177.md section 4): # Corroborating signal, not a gate (DESIGN-177.md section 4):
# /multidevice/vs/0's numofsubdevice is a plain count HJcom's board # /multidevice/vs/0's numofsubdevice is a plain count the issue
# reports independently of the liveness gate above. Log, don't # #177 reporter's board reports independently of the liveness gate
# raise, on a disagreement -- only this one board family is known to # above. Log, don't raise, on a disagreement -- only this one board
# expose the resource at all, so a mismatch is a "look into this" # family is known to expose the resource at all, so a mismatch is a
# signal for triage, not proof either side is wrong. # "look into this" signal for triage, not proof either side is
# wrong.
numofsubdevice = self._multidevice.get( numofsubdevice = self._multidevice.get(
'x.com.samsung.da.numofsubdevice') 'x.com.samsung.da.numofsubdevice')
if numofsubdevice is not None: if numofsubdevice is not None:
+5 -5
View File
@@ -112,11 +112,11 @@ async def async_get_config_entry_diagnostics(
"subdevices": [_subdevice_diag(su) for su in coordinator.subdevices], "subdevices": [_subdevice_diag(su) for su in coordinator.subdevices],
# Candidates that answered their seed but that discover_partitioned's # Candidates that answered their seed but that discover_partitioned's
# entity-level liveness gate rejected -- an unused SmartThings slot # entity-level liveness gate rejected -- an unused SmartThings slot
# (HJcom's /device/2) that still answers a same-shaped batch, not a # (the issue #177 reporter's /device/2) that still answers a
# real second subdevice. Reported alongside subdevices above so a report # same-shaped batch, not a real second subdevice. Reported alongside
# shows what was found *and* why it didn't become an entity, not # subdevices above so a report shows what was found *and* why it
# just silence where a third climate card might otherwise be # didn't become an entity, not just silence where a third climate
# expected. # card might otherwise be expected.
"subdevices_skipped": [ "subdevices_skipped": [
{ {
"kind": skip.subdevice.kind, "kind": skip.subdevice.kind,
@@ -958,7 +958,7 @@ _AC_IGNORED = [
# Undocumented single int (runningMode: 0 on every dump seen), no # Undocumented single int (runningMode: 0 on every dump seen), no
# supported-values list to interpret it against -- 'don't guess'. # supported-values list to interpret it against -- 'don't guess'.
'/runn/vs/0', '/runn/vs/0',
# 2-in-1/multi-indoor-subdevice systems (issue #177, HJcom's # 2-in-1/multi-indoor-subdevice systems (issue #177, the reporter's
# ARTIK051_DONGLE_FAC_18K): x.com.samsung.da.numofsubdevice, a plain # ARTIK051_DONGLE_FAC_18K): x.com.samsung.da.numofsubdevice, a plain
# corroborating count of indoor subdevices on this connection. Confirmed # corroborating count of indoor subdevices on this connection. Confirmed
# read-only (a write attempt returned CoAP 4.00). Absent from # read-only (a write attempt returned CoAP 4.00). Absent from
@@ -73,7 +73,7 @@ def read_identity(sess, serial: Optional[str]) -> DeviceIdentity:
# physical device -- one IP, one /oic/p -- exposing more than one logical # physical device -- one IP, one /oic/p -- exposing more than one logical
# subdevice, each as its own Collection resource, same rt shape as our own # subdevice, each as its own Collection resource, same rt shape as our own
# /device/0). This is what registry.subdevices.enumerate_subdevices reads # /device/0). This is what registry.subdevices.enumerate_subdevices reads
# to find a board's `/device/<n>` siblings (Pattern A -- HJcom's # to find a board's `/device/<n>` siblings (Pattern A -- the reporter's
# ARTIK051_DONGLE_FAC_18K) -- that probing, plus the /device/1 and # ARTIK051_DONGLE_FAC_18K) -- that probing, plus the /device/1 and
# /device/2 speculative fallback it used to run right here on every # /device/2 speculative fallback it used to run right here on every
# _connect_session (including every reconnect), moved to that module so # _connect_session (including every reconnect), moved to that module so
@@ -4,15 +4,16 @@ more than one logical indoor subdevice -- issue #177.
Two reporters, two different board families, two genuinely different Two reporters, two different board families, two genuinely different
mechanisms for exposing a second indoor subdevice over one IP / one DTLS mechanisms for exposing a second indoor subdevice over one IP / one DTLS
session (see DESIGN-177.md section 1 for the full evidence trail; the two session (see DESIGN-177.md section 1 for the full evidence trail; the two
diagnostics dumps this was built against are HJcom's and issue #177's other diagnostics dumps this was built against come from the Pattern A and
reporter's -- they each filed one of the two reports this module unifies): Pattern B reporters, respectively -- they each filed one of the two
reports this module unifies):
Pattern A -- indexed siblings (`ARTIK051_DONGLE_FAC_18K`, HJcom's board). Pattern A -- indexed siblings (`ARTIK051_DONGLE_FAC_18K`, that reporter's
`/oic/res` lists three complete parallel resource sets whose trailing path board). `/oic/res` lists three complete parallel resource sets whose
segment is the index (`/mode/vs/0`, `/mode/vs/1`, `/mode/vs/2`, ... on both trailing path segment is the index (`/mode/vs/0`, `/mode/vs/1`,
OCF-standard and vendor hrefs), and `/device/0`'s batch carries only the `/mode/vs/2`, ... on both OCF-standard and vendor hrefs), and `/device/0`'s
index-0 hrefs -- the sibling subdevices are reachable only via their own batch carries only the index-0 hrefs -- the sibling subdevices are
`/device/<n>` collection. reachable only via their own `/device/<n>` collection.
Pattern B -- UUID-prefixed tree (`TP2X_FAC_BORA_21K`, that reporter's board). Pattern B -- UUID-prefixed tree (`TP2X_FAC_BORA_21K`, that reporter's board).
`/oic/res` hides the whole appliance tree; `/device/0`'s batch instead `/oic/res` hides the whole appliance tree; `/device/0`'s batch instead
@@ -38,27 +39,29 @@ surface. See `Subdevice.flat_hrefs`.
Both are "the same thing wearing different clothes": a logical subdevice is a Both are "the same thing wearing different clothes": a logical subdevice is a
seed collection path to poll, plus an href transform between the canonical seed collection path to poll, plus an href transform between the canonical
href the registry knows (`/mode/vs/0`) and the actual on-the-wire href. The href the registry knows (`/mode/vs/0`) and the actual on-the-wire href. The
detection signals don't overlap on either captured board (HJcom's has no detection signals don't overlap on either captured board (the Pattern A
`/subdevices/vs/0` at all; the other reporter's has no `/device/1`), so no reporter's has no `/subdevices/vs/0` at all; the Pattern B reporter's has
disambiguation logic is needed -- `enumerate_subdevices` checks both and no `/device/1`), so no disambiguation logic is needed --
materializes any candidate whose seed answers with a non-empty batch. `enumerate_subdevices` checks both and materializes any candidate whose
seed answers with a non-empty batch.
A non-empty seed batch is necessary but not sufficient for the *candidate* A non-empty seed batch is necessary but not sufficient for the *candidate*
to actually be a live second subdevice, though: HJcom's own board also has a to actually be a live second subdevice, though: the Pattern A reporter's
`/device/2` -- a third, unused SmartThings slot -- that answers with the own board also has a `/device/2` -- a third, unused SmartThings slot --
exact same 14-href shape as the real `/device/1` sibling, populated with that answers with the exact same 14-href shape as the real `/device/1`
three constant/echoed/shape-only reps (a region code identical to every sibling, populated with three constant/echoed/shape-only reps (a region
other subdevice's, an /information rep echoing the *same* model string as code identical to every other subdevice's, an /information rep echoing the
subdevice 1, and a /temperatures items[] entry with an id/description but no *same* model string as subdevice 1, and a /temperatures items[] entry with
current/desired/minimum/maximum reading) and nothing resembling live an id/description but no current/desired/minimum/maximum reading) and
climate state. Gating on *resource* shape/hrefs turned out to be the wrong nothing resembling live climate state. Gating on *resource* shape/hrefs
layer -- it would need per-family domain knowledge (which hrefs mean "in turned out to be the wrong layer -- it would need per-family domain
use" for a washer's second drum, a fridge's second compartment, ...) baked knowledge (which hrefs mean "in use" for a washer's second drum, a
into a registry field before any of those families could use this module fridge's second compartment, ...) baked into a registry field before any
at all. `discover_partitioned` instead gates at the *entity* layer, after of those families could use this module at all. `discover_partitioned`
discovery+flattening: a candidate is only kept if it produced at least one instead gates at the *entity* layer, after discovery+flattening: a
*primary* (no `entity_category`) bound entity whose flattened value isn't candidate is only kept if it produced at least one *primary* (no
`None` -- e.g. HJcom's /device/2 does flatten to an `alarm_code` value, but `entity_category`) bound entity whose flattened value isn't `None` -- e.g.
the Pattern A reporter's /device/2 does flatten to an `alarm_code` value, but
that entity is diagnostic-category and derived from an empty /alarms/vs/2, that entity is diagnostic-category and derived from an empty /alarms/vs/2,
so it doesn't count. This reuses the same primary/config/diagnostic so it doesn't count. This reuses the same primary/config/diagnostic
taxonomy every registry already declares (see the adding-device-support taxonomy every registry already declares (see the adding-device-support
@@ -216,11 +219,11 @@ def normalize_seed_batch(subdevice: Subdevice, batch: dict[str, dict]) -> dict[s
normalized so every href actually carries this subdevice's prefix/index. normalized so every href actually carries this subdevice's prefix/index.
Indexed subdevices need no change -- the device echoes the real `/x/<n>` Indexed subdevices need no change -- the device echoes the real `/x/<n>`
href in its own `/device/<n>` batch (confirmed against HJcom's dump). href in its own `/device/<n>` batch (confirmed against the Pattern A
A prefixed subdevice's batch entries may or may not already carry the reporter's dump). A prefixed subdevice's batch entries may or may not
`/<id>` prefix (unconfirmed which -- the Pattern B reporter's board was already carry the `/<id>` prefix (unconfirmed which -- the Pattern B
never probed live before the subdevice id was known), so it's added reporter's board was never probed live before the subdevice id was
when missing. known), so it's added when missing.
""" """
if subdevice.kind != 'prefixed': if subdevice.kind != 'prefixed':
return batch return batch
@@ -283,9 +286,9 @@ def _get_batch(sess, path_segs: tuple[str, ...]) -> dict[str, dict]:
def _get_property(sess, path_segs: tuple[str, ...]) -> dict: def _get_property(sess, path_segs: tuple[str, ...]) -> dict:
"""GET a plain OCF Property-map resource (a bare dict, not a Collection """GET a plain OCF Property-map resource (a bare dict, not a Collection
batch). Used for `/multidevice/vs/0` (issue #177 follow-up): listed in batch). Used for `/multidevice/vs/0` (issue #177 follow-up): listed in
`/oic/res` on HJcom's board but absent from `/device/0`'s batch, so it `/oic/res` on the Pattern A reporter's board but absent from
needs its own RETRIEVE, and it answers a single Property map, not a `/device/0`'s batch, so it needs its own RETRIEVE, and it answers a
[devcol-rep, ...] list.""" single Property map, not a [devcol-rep, ...] list."""
body = _get_raw(sess, path_segs) body = _get_raw(sess, path_segs)
return body if isinstance(body, dict) else {} return body if isinstance(body, dict) else {}
@@ -314,9 +317,10 @@ def enumerate_subdevices(
Every candidate whose seed answers with a non-empty batch is returned Every candidate whose seed answers with a non-empty batch is returned
here -- this function has no way to tell a real sibling from an unused here -- this function has no way to tell a real sibling from an unused
SmartThings slot that merely answers the same shape (HJcom's SmartThings slot that merely answers the same shape (the Pattern A
`/device/2`); that requires discovering+flattening the candidate's own reporter's `/device/2`); that requires discovering+flattening the
entities first, which is `discover_partitioned`'s job, not this one's. candidate's own entities first, which is `discover_partitioned`'s job,
not this one's.
See this module's docstring. See this module's docstring.
""" """
subdevices: list[Subdevice] = [] subdevices: list[Subdevice] = []
@@ -410,9 +414,9 @@ def enumerate_subdevices(
fetched.update(batch) # already real /x/<n> hrefs, no normalization needed fetched.update(batch) # already real /x/<n> hrefs, no normalization needed
subdevices.append(subdevice) subdevices.append(subdevice)
# /multidevice/vs/0 (issue #177 follow-up): HJcom's board lists it in # /multidevice/vs/0 (issue #177 follow-up): the Pattern A reporter's
# /oic/res but it never appears in /device/0's batch, so it needs its # board lists it in /oic/res but it never appears in /device/0's batch,
# own RETRIEVE. It's a plain corroborating count # so it needs its own RETRIEVE. It's a plain corroborating count
# (x.com.samsung.da.numofsubdevice), confirmed read-only (a write # (x.com.samsung.da.numofsubdevice), confirmed read-only (a write
# attempt returned CoAP 4.00) -- captured for diagnostics only, folded # attempt returned CoAP 4.00) -- captured for diagnostics only, folded
# into the merged resources dict like any other href (see # into the merged resources dict like any other href (see
@@ -436,9 +440,9 @@ def enumerate_subdevices(
class SkippedSubdevice: class SkippedSubdevice:
"""A candidate `enumerate_subdevices` found whose seed answered, but that """A candidate `enumerate_subdevices` found whose seed answered, but that
`discover_partitioned`'s entity-level liveness gate rejected -- an `discover_partitioned`'s entity-level liveness gate rejected -- an
unused SmartThings slot (HJcom's `/device/2`), not a real second subdevice. unused SmartThings slot (the Pattern A reporter's `/device/2`), not a
Kept around (rather than silently dropped) so a caller can log/report real second subdevice. Kept around (rather than silently dropped) so a
what was skipped and why.""" caller can log/report what was skipped and why."""
subdevice: Subdevice subdevice: Subdevice
hrefs: tuple[str, ...] hrefs: tuple[str, ...]
@@ -450,7 +454,7 @@ def _has_live_primary_entity(bound, state: dict) -> bool:
tier (see the adding-device-support skill's entity-taxonomy section). tier (see the adding-device-support skill's entity-taxonomy section).
This is the materialization gate itself (see this module's docstring): This is the materialization gate itself (see this module's docstring):
HJcom's `/device/2` does flatten to one non-`None` value the Pattern A reporter's `/device/2` does flatten to one non-`None` value
(`alarm_code`), but that entity is `diagnostic`-category and derived (`alarm_code`), but that entity is `diagnostic`-category and derived
from an empty `/alarms/vs/2` -- a config/diagnostic entity reading from an empty `/alarms/vs/2` -- a config/diagnostic entity reading
"something" proves nothing about whether a physical subdevice is actually "something" proves nothing about whether a physical subdevice is actually
@@ -2029,5 +2029,5 @@
"x.com.samsung.da.numofsubdevice": "2" "x.com.samsung.da.numofsubdevice": "2"
} }
}, },
"seeds_note": "ALL REAL CAPTURED DATA -- issue #177, HJcom's ARTIK051_DONGLE_FAC_18K (Samsung 2-in-1: floor-standing master + wall-mounted second indoor subdevice). device0, oic_res and both /device/<n> seeds come verbatim from their v0.16.0 diagnostics download, which is the first release that probes /device/1 and /device/2. Nothing here is reconstructed. The reporter masked a handful of values in their own dump before posting (serialNum, and the airflow `direction` / vendor `humidity` readings); those are normalized onto this repo's usual **REDACTED** marker. /device/1 is the real second subdevice -- populated power/mode/temperature and its own /information/vs/1 reporting ARTIK051_DONGLE_FAC_RAC_18K (RAC = the wall-mounted subdevice) against the master's ARTIK051_DONGLE_FAC_18K. /device/2 answers with the same 14-href shape but every state rep is empty {} -- it is the unused third slot SmartThings shows disabled, and it is why subdevice materialization gates on populated climate state rather than on a seed merely answering. `probes` is not part of any batch response: /multidevice/vs/0 is listed in oic_res but absent from /device/0's batch, and the reporter read it by hand through the debug panel (a write attempt returned CoAP 4.00, i.e. read-only). Its numofsubdevice=2 independently corroborates 2 real subdevices, not 3." "seeds_note": "ALL REAL CAPTURED DATA -- issue #177, the reporter's ARTIK051_DONGLE_FAC_18K (Samsung 2-in-1: floor-standing master + wall-mounted second indoor subdevice). device0, oic_res and both /device/<n> seeds come verbatim from their v0.16.0 diagnostics download, which is the first release that probes /device/1 and /device/2. Nothing here is reconstructed. The reporter masked a handful of values in their own dump before posting (serialNum, and the airflow `direction` / vendor `humidity` readings); those are normalized onto this repo's usual **REDACTED** marker. /device/1 is the real second subdevice -- populated power/mode/temperature and its own /information/vs/1 reporting ARTIK051_DONGLE_FAC_RAC_18K (RAC = the wall-mounted subdevice) against the master's ARTIK051_DONGLE_FAC_18K. /device/2 answers with the same 14-href shape but every state rep is empty {} -- it is the unused third slot SmartThings shows disabled, and it is why subdevice materialization gates on populated climate state rather than on a seed merely answering. `probes` is not part of any batch response: /multidevice/vs/0 is listed in oic_res but absent from /device/0's batch, and the reporter read it by hand through the debug panel (a write attempt returned CoAP 4.00, i.e. read-only). Its numofsubdevice=2 independently corroborates 2 real subdevices, not 3."
} }
+15 -13
View File
@@ -3,10 +3,11 @@
two -- and that the legacy-board test (is_legacy_board/_legacy_airflow) is two -- and that the legacy-board test (is_legacy_board/_legacy_airflow) is
evaluated per subdevice rather than once globally. evaluated per subdevice rather than once globally.
Uses HJcom's ARTIK051_DONGLE_FAC_18K fixture deliberately: it's a legacy Uses the issue #177 reporter's ARTIK051_DONGLE_FAC_18K fixture deliberately:
`/airflow/vs/<n>` board (no `/wind/*` at all) on *both* the master and its it's a legacy `/airflow/vs/<n>` board (no `/wind/*` at all) on *both* the
materialized sibling, which is exactly the shape climate.py's own comments master and its materialized sibling, which is exactly the shape climate.py's
warn is easy to get wrong if the canonical view leaks between subdevices. own comments warn is easy to get wrong if the canonical view leaks between
subdevices.
""" """
from __future__ import annotations from __future__ import annotations
@@ -71,15 +72,16 @@ async def test_subdevice_climate_reads_its_own_power_state(climates):
async def test_legacy_board_test_is_evaluated_per_subdevice(climates): async def test_legacy_board_test_is_evaluated_per_subdevice(climates):
"""HJcom's board has no /wind/* resources at all on *either* subdevice -- """The reporter's board has no /wind/* resources at all on *either*
is_legacy_board(self._resources) must independently evaluate True for subdevice -- is_legacy_board(self._resources) must independently
the master's own canonical view and for the subdevice's own canonical evaluate True for the master's own canonical view and for the
view. If is_legacy_board were fed the raw, unpartitioned snapshot (or subdevice's own canonical view. If is_legacy_board were fed the raw,
if canonical_view leaked one subdevice's hrefs into the other's), this unpartitioned snapshot (or if canonical_view leaked one subdevice's
wouldn't distinguish "this subdevice is legacy" from "some subdevice on hrefs into the other's), this wouldn't distinguish "this subdevice is
this connection is legacy" -- and a future board with one legacy + one legacy" from "some subdevice on this connection is legacy" -- and a
modern subdevice sharing a connection would silently read the wrong future board with one legacy + one modern subdevice sharing a
fan/swing channel on one side.""" connection would silently read the wrong fan/swing channel on one
side."""
main, sub1 = climates[None], climates['1'] main, sub1 = climates[None], climates['1']
assert main._legacy_airflow() != {} assert main._legacy_airflow() != {}
assert sub1._legacy_airflow() != {} assert sub1._legacy_airflow() != {}
+1 -1
View File
@@ -903,7 +903,7 @@ def _new_subdevice_aware_state_keys(name):
def test_registry_reproduces_golden_state_keys_for_airconditioner_artik051_dongle_fac_18k(): def test_registry_reproduces_golden_state_keys_for_airconditioner_artik051_dongle_fac_18k():
"""HJcom's ARTIK051_DONGLE_FAC_18K (issue #177, Pattern A -- indexed """The reporter's ARTIK051_DONGLE_FAC_18K (issue #177, Pattern A -- indexed
siblings): a real v0.16.0 dump with a genuine second indoor subdevice at siblings): a real v0.16.0 dump with a genuine second indoor subdevice at
`/device/1` (subdevice1_-prefixed keys below) and an unused SmartThings `/device/1` (subdevice1_-prefixed keys below) and an unused SmartThings
slot at `/device/2` that answers its seed but never produces a slot at `/device/2` that answers its seed but never produces a
+7 -7
View File
@@ -87,10 +87,10 @@ def _climate_bound(coordinator, subdevice_key: str):
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
# HJcom -- ARTIK051_DONGLE_FAC_18K, Pattern A (indexed siblings) # Pattern A reporter -- ARTIK051_DONGLE_FAC_18K, indexed siblings
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
async def test_hjcom_materializes_master_and_bedroom_subdevice(hass: HomeAssistant): async def test_pattern_a_materializes_master_and_bedroom_subdevice(hass: HomeAssistant):
coordinator = _coordinator(hass) coordinator = _coordinator(hass)
await _discover(coordinator, 'airconditioner_artik051_dongle_fac_18k') await _discover(coordinator, 'airconditioner_artik051_dongle_fac_18k')
@@ -104,11 +104,11 @@ async def test_hjcom_materializes_master_and_bedroom_subdevice(hass: HomeAssista
assert sub1_climate.href == '/mode/vs/1' assert sub1_climate.href == '/mode/vs/1'
async def test_hjcom_device_2_produces_no_entities_at_all(hass: HomeAssistant): async def test_pattern_a_device_2_produces_no_entities_at_all(hass: HomeAssistant):
"""HJcom's /device/2 is the unused SmartThings slot (DESIGN-177.md """The reporter's /device/2 is the unused SmartThings slot
section 4): it answers its seed with a full-shaped batch, but every (DESIGN-177.md section 4): it answers its seed with a full-shaped
climate-state rep on it is empty. It must be recorded as skipped, not batch, but every climate-state rep on it is empty. It must be recorded
materialized, and must contribute zero bound entities.""" as skipped, not materialized, and must contribute zero bound entities."""
coordinator = _coordinator(hass) coordinator = _coordinator(hass)
await _discover(coordinator, 'airconditioner_artik051_dongle_fac_18k') await _discover(coordinator, 'airconditioner_artik051_dongle_fac_18k')
+9 -8
View File
@@ -383,10 +383,10 @@ def test_enumerate_no_subdevices_resource_at_all():
def test_enumerate_indexed_materializes_candidate_regardless_of_content(): def test_enumerate_indexed_materializes_candidate_regardless_of_content():
"""enumerate_subdevices itself has no way to tell a real sibling from an """enumerate_subdevices itself has no way to tell a real sibling from an
unused slot that merely answers the same shape (HJcom's /device/2) -- unused slot that merely answers the same shape (the reporter's
that's discover_partitioned's job (see its own tests below). A /device/2) -- that's discover_partitioned's job (see its own tests
same-shaped batch of otherwise-empty reps is still returned as a below). A same-shaped batch of otherwise-empty reps is still returned
candidate here.""" as a candidate here."""
oic_res = [{'di': 'a', 'links': [{'href': '/device/2'}]}] oic_res = [{'di': 'a', 'links': [{'href': '/device/2'}]}]
sess = _FakeSession({ sess = _FakeSession({
('device', '2'): [_DEVCOL_REP, {'href': '/power/vs/2', 'rep': {}}, ('device', '2'): [_DEVCOL_REP, {'href': '/power/vs/2', 'rep': {}},
@@ -598,13 +598,14 @@ def test_discover_partitioned_no_subdevices_matches_plain_discover():
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
# discover_partitioned's materialization gate: a candidate is only kept if # discover_partitioned's materialization gate: a candidate is only kept if
# it produced at least one *primary* (no entity_category) bound entity whose # it produced at least one *primary* (no entity_category) bound entity whose
# flattened value isn't None. This is what tells HJcom's real /device/1 # flattened value isn't None. This is what tells the reporter's real
# sibling apart from the unused /device/2 slot that answers the same shape. # /device/1 sibling apart from the unused /device/2 slot that answers the
# same shape.
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
def test_discover_partitioned_skips_candidate_with_no_live_primary_entity(): def test_discover_partitioned_skips_candidate_with_no_live_primary_entity():
"""A candidate whose only populated entity is diagnostic-category """A candidate whose only populated entity is diagnostic-category
doesn't count -- exactly HJcom's /device/2 shape (an alarm_code doesn't count -- exactly the reporter's /device/2 shape (an alarm_code
sensor reading something even though the subdevice itself is empty).""" sensor reading something even though the subdevice itself is empty)."""
diag_cap = Capability( diag_cap = Capability(
href='/alarms/vs/0', href='/alarms/vs/0',
@@ -632,7 +633,7 @@ def test_discover_partitioned_skips_candidate_with_no_live_primary_entity():
def test_discover_partitioned_materializes_candidate_with_live_primary_entity(): def test_discover_partitioned_materializes_candidate_with_live_primary_entity():
"""A candidate with a populated *primary* (no entity_category) entity """A candidate with a populated *primary* (no entity_category) entity
is materialized, even alongside an all-empty diagnostic sibling href -- is materialized, even alongside an all-empty diagnostic sibling href --
exactly HJcom's /device/1 shape.""" exactly the reporter's /device/1 shape."""
climate_cap = Capability( climate_cap = Capability(
href='/mode/vs/0', href='/mode/vs/0',
entities=(BinarySensorDesc(key='mode', field='m'),), # no entity_category -> primary entities=(BinarySensorDesc(key='mode', field='m'),), # no entity_category -> primary