Compare commits

...
Author SHA1 Message Date
Claude 95375e03d5 ci: перепаковка кадров документации без потерь при съёмке
oxipng снимает с набора 19.4%: 2096 КБ становятся 1689 КБ, и все десять кадров
остаются пиксельно идентичными — декодированные RGBA совпадают по sha256. Это
выбор фильтров строки и уровня сжатия, а не квантование: визуального решения
здесь нет.

Внутри съёмки, а не отдельным проходом по закоммиченным файлам: манифест хранит
imageSha256 каждого кадра, поэтому жать их в репозитории руками нельзя —
check-docs покраснеет; а если жать после подсчёта хешей, следующая съёмка вернёт
неоптимизированные байты. Хеш считается после перепаковки.

Версия oxipng попадает в манифест рядом с версией браузера и по той же причине:
байты кадра зависят от того, чем жали. Отсюда же правка шага «Вердикт» — иначе
он объявил бы «тот же браузер, а картинки изменились — изменился продукт», хотя
изменился упаковщик. Теперь эта причина названа отдельной ветвью.

Пин версии и контрольной суммы вместо apt-get: пакет из образа раннера может
пропасть, а падение шага съёмки стоит целого цикла приёмки. Тот же урок, что с
azure-зеркалом Playwright (#175, #206).

Отсутствие инструмента не ошибка: локальная съёмка полезна и без него, приёмка
идёт только из артефакта CI. Но молчать нельзя, поэтому в лог пишется строка.

Проверено исполнением: съёмка прогнана целиком с подставным oxipng на pyoxipng,
десять кадров, 2096 -> 1689 КБ, хеши манифеста совпали с файлами на диске,
check-docs зелёный.

Issue: #345
User-Visible: no
2026-08-28 08:09:41 +03:00
Claude 59ae6b1064 ci: полная история без блобов там, где содержимое старых файлов не нужно
Полный клон — 215 МБ .git, blobless — 26 МБ, история коммитов и теги в обоих
полные (замер в #345). Две job Validate качают историю целиком: preflight и
changes. Первой нужны сообщения коммитов, трейлеры и имена изменённых файлов,
второй — только `git diff --name-only`. Содержимое старых ревизий не читает ни
одна из них ни на одном шаге.

Коммиты и деревья по-прежнему скачиваются полностью, поэтому диапазоны и
merge-base работают как раньше. Единственная догрузка блоба по требованию —
`git show origin/main:.github/workflows/process.yml` в шаге сверки, один файл.

Браузерным job фильтр не нужен: у них глубина по умолчанию, истории они не
касаются вовсе.

Тест закрепляет и обратную сторону: --depth=1 сюда подставлять нельзя, он того
же размера, но без merge-base, а на нём стоят процессный гейт, smoke-select и
каждый диапазон origin/dev..HEAD. Два мутанта проверены руками — снятый фильтр и
подмена на depth=1, — каждый убит.

Issue: #345
User-Visible: no
2026-08-28 08:00:51 +03:00
Claude 80a664bc23 docs: рекомендовать blobless-клон вместо полного
Замерено, не оценено: полный клон — 215 МБ .git, blobless — 26 МБ. История
коммитов и теги в обоих полные, поэтому диапазоны, merge-base и git diff по
истории работают одинаково; diff трёх коммитов в blobless-клоне занимает секунду
и добавляет мегабайт.

Разница в том, что содержимое старых файлов скачивается только если его кто-то
спросит. 32% пака — скриншоты документации, десять PNG, переснятых 196 раз; ещё
заметная доля — закоммиченный бандл, 1.16 МБ на каждую продуктовую правку. Старые
ревизии ни того, ни другого практически никто не читает.

Про --depth=1 сказано отдельно, что он не замена: размер тот же, но merge-base
нет, и процессный гейт вместе со smoke-select перестают работать.

Issue: #345
User-Visible: no
2026-08-28 07:57:34 +03:00
Codex 508945c088 fix: a 0° pair is the shared-wall model, not a duplicate — field revert of #331 §2.2
Red dev caught it ninety minutes after the merge: smoke_plan_drawing_repairs
and smoke_resize_pointer_real_plan went red because the new "a 0° wedge is
always a duplicate" rule refused two ordinary edits — creating a room over
an existing partition ring (#308's legal overlay) and resizing a wall until
it lands on a neighbour's. The premise was wrong at the model level: a
shared wall of two adjacent rooms IS two co-located owner atoms on one line,
so every shared-wall node carries a legitimate 0° pair by construction.
Bisection pinned the exact cut: with only the 0° rule reverted, both smokes
are green again; keys, incidence, the iterative walk and fail-closed stay.

Spec revision 4 records the revert and returns "an exact duplicate wall is
invisible to П1" to the status of a KNOWN LIMITATION — an honest detector
needs owner identity, which is a separate decision for the owner to make.
The zero-wedge mutant is removed with its rule; the .5-tick parity unit now
observes quantisation through valence instead of the retired duplicate
visibility; changelogs drop the over-promise.

Issue: #331
User-Visible: yes
2026-08-28 05:20:45 +03:00
claude[bot] 95fba6dbab docs: review document for #331
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 53s
Проверка (CI) / Предполётные проверки: документация, провенанс, процесс (push) Failing after 1m16s
Проверка (CI) / Классификация изменённых файлов (push) Successful in 1m1s
Проверка (CI) / HACS: валидация репозитория (push) Failing after 17s
Проверка (CI) / Hassfest: манифест интеграции (push) Failing after 16s
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Failing after 7m11s
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Skipped
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
Проверка (CI) / Golden-кадры против принятых эталонов (push) Skipped
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Skipped
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Failing after 9m5s
Issue: #331
User-Visible: no
2026-08-28 02:01:06 +00:00
Codex 550be251a6 test: register the smoke link for the key-quantisation internals (#331)
smoke-select flagged quantizeKeyCoord/INCIDENT_EPS/KEY_FACTOR as symbols no
smoke names — true by design: the smoke proves the verdicts through the
rendered card, never touching the internals that decide which nodes are one
node. The registry entry records that non-textual link (#241 rule).

Issue: #331
User-Visible: no
2026-08-28 04:44:18 +03:00
Codex 0cdf85d9e2 test: the key-precision and dropped-branch mutants actually bite (#331)
Running the mutants exposed two toothless guards before review did:
- reverting the keys to toFixed(6) no longer produced false П4 refusals
  because the new 2e-7 incidence absorbed the debris pair — the REAL harm of
  coarse keys is the opposite direction: nodes 4e-7 apart merged into one
  key and П4 went blind to a genuine near-miss. AC1 now pins that case.
- the `break` patch failed to reproduce the old first-branch-only loss (the
  frontier re-visits the node through the pushed endpoints); the patch now
  truncates the node's candidate list to one entry, which loses forks the
  way `.find` did — both the fork unit and the 10 000-atom run turn red.

Issue: #331
User-Visible: no
2026-08-28 04:42:07 +03:00
Codex 58f3acbbde fix: junction limits are honest at the boundaries (#331)
Six normative cuts, both mirrors symmetric (spec revision 3):

- §2.1 node keys quantise to 1e-7 with the repository's canonicalisation
  formula (sign·floor(|v|·1e7+0.5)/1e7, -0 normalised) — toFixed(6) keys
  split one node into two on floating debris and produced two false П4
  refusals on a legitimate resize (reproduced: -1e-8 vs 0). Node pairs
  within 2e-7 of each other (raw coordinates) are ONE node, and the
  node-to-wall incidence uses the same quantum.
- §2.2 a ~0° wedge IS a violation: two rays leaving a node the same way are
  a duplicated or overlaid wall (a butt joint yields 180°, never 0°) — the
  worst degenerate case was invisible while 0.5° was refused.
- §2.3/§2.4 the wall run is an iterative edge walk over the collinear
  component: no recursion (10 000 atoms answered, not RangeError), no
  silently dropped fork (the old .find lost every branch but the first),
  O(E) by construction, and collinearity is measured against the BASE
  segment's axis so an arc of 0.9°-per-atom pieces cannot pose as one wall.
- §2.5 an exception while judging the CANDIDATE refuses the write with the
  junction.limit_check_failed toast (fail-closed, as the #278 guard); the
  baseline branch stays fail-open by design and the smoke proves the
  asymmetry by breaking only the second call of the deterministic pair.
- §2.6 the python mirror narrows its except on the candidate side only:
  a genuine migration bug (TypeError) surfaces as an honest WS error, while
  a previous-side bug keeps the wide "no baseline" fallback — the two AC6
  cases pin the asymmetry so swapped sides turn a unit red.

Parity fixtures gain the new boundary classes (debris node, duplicate wall,
collinear fork); four new mutants pin the filter, the key precision, the
dropped branch and the fail-open hole.

Issue: #331
User-Visible: yes
2026-08-28 04:39:46 +03:00
claude[bot] c231602a2b docs: review document for #331
Issue: #331
User-Visible: no
2026-08-28 01:33:02 +00:00
Codex 4423d28579 docs: spec #331 revision 3 — AC6 tells the two sides apart
r2 M-r2-1: AC6 now mirrors AC5's structure with two explicit cases — a
candidate-side TypeError is an honest WS error, a previous-side TypeError
falls back to "no baseline" and the unrelated write passes. An
implementation with swapped or missing asymmetry turns at least one of the
two units red. L2: the risk wording follows §2.3's component-sum phrasing.

Issue: #331
User-Visible: no
2026-08-28 04:30:13 +03:00
claude[bot] ebbfe4588f docs: review document for #331
Issue: #331
User-Visible: no
2026-08-28 01:27:57 +00:00
Codex 2d70354345 docs: spec #331 revision 2 — canonical rounding, honest incidence threshold, edge-walk instead of DFS
r1-H1: node keys quantise with the repository's canonicalisation formula
(sign·floor(|v|·1e7+0.5)/1e7) — native Math.round and Python round() part
ways on .5 ticks, the exact parity lesson coordinate-canonicalization
already encodes. r1-M1: the incidence threshold becomes 2e-7 over raw
coordinates, which the spec's own example (1.02e-7) actually satisfies; the
known valence undercount on neighbouring quanta is stated in §3. r1-M2: the
narrow except applies to the candidate side only — a previous-side migration
bug stays a "no baseline" fallback, symmetric with §2.5. r1-M3: the branch
walk is an O(E) edge traversal of the collinear component, not a
combinatorial DFS; AC3 gains a 100-fork case. r1-M4: the USER-GUIDE limits
section documents the new refusal toast. L1: §1 opens with the user
sentence.

Issue: #331
User-Visible: no
2026-08-28 04:20:17 +03:00
claude[bot] 3de2a4a4b5 docs: review document for #331
Issue: #331
User-Visible: no
2026-08-28 01:16:18 +00:00
Codex 60e125e960 docs: spec #331 — boundary precision of the junction limits
Quantised node keys with -0 normalisation and node incidence at the quantum,
zero-degree wedges become visible, an iterative maximal-branch wall run, arc
collinearity measured against the chain base, fail-closed candidate checks,
and a narrow except in the python mirror. Every reproduction in §1 was
verified by execution on current dev after #330.

Issue: #331
User-Visible: no
2026-08-28 04:03:14 +03:00
claude[bot] 2d1ca74a7d docs: review document for #330
Проверка (CI) / Классификация изменённых файлов (push) Successful in 1m30s
Проверка (CI) / Предполётные проверки: документация, провенанс, процесс (push) Failing after 1m50s
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 1m50s
Проверка (CI) / HACS: валидация репозитория (push) Failing after 18s
Проверка (CI) / Hassfest: манифест интеграции (push) Failing after 28s
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Failing after 5m56s
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Skipped
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
Проверка (CI) / Golden-кадры против принятых эталонов (push) Skipped
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Skipped
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Failing after 6m29s
Issue: #330
User-Visible: no
2026-08-28 00:56:30 +00:00
Codex d33aa0a7e5 docs: refresh the screenshot pair after the cache-key fix (#330 r3-H1)
Third time this class bites in one task: any src/** edit staleness the
screenshot source fingerprint mechanically, and I keep forgetting the
capture step after code-only commits. The pair (PNGs + manifest) is
regenerated from one run; check-docs is green on this SHA.

Issue: #330
User-Visible: no
2026-08-28 03:53:04 +03:00
claude[bot] 90a3313d92 docs: review document for #330
Issue: #330
User-Visible: no
2026-08-28 00:49:56 +00:00
Codex 04bb54aef3 fix: the baseline cache keys on geometry, and the smoke tells all three worlds apart (#330 r2-M1)
The reviewer proved my behavioural claim false by running the stale-cache
mutant against the smoke: 11 vs 12 total calls — indistinguishable. Two real
defects hid behind that finding:

1. The cache keyed on _cfgEpoch, which ticks on every ACCEPTED PREVIEW —
   the cache missed on every pointermove and the baseline was recomputed
   ~4 times per gesture (measured). The key is now the document identity
   plus spacePhysicalGeometryFingerprint of its space: content, not a
   counter. A preview overlay leaves the fingerprint alone; an in-place
   structural commit changes it and honestly invalidates.

2. The smoke now counts BASELINE computations only (calls whose document is
   _serverCfg) across two gestures with a commit in between, expecting
   exactly 1 then exactly 2. A disabled cache lands near 20, an eternal
   cache stays at 1 — every mutant class turns the smoke red, and the smoke
   is now the mutant's guard alongside the source-contract unit.

Issue: #330
User-Visible: no
2026-08-28 03:35:40 +03:00
claude[bot] 294d7047f3 docs: review document for #330
Issue: #330
User-Visible: no
2026-08-28 00:30:31 +00:00
Codex 85636a65bb docs: re-capture the screenshot set after the second rebase (#330)
Same pairing rule as before: PNG files and their manifest must come from one
capture run; the rebase over the i18n-registry merge (#62) mixed the sides
again.

Issue: #330
User-Visible: no
2026-08-28 03:08:32 +03:00
Codex ddfca3a865 fix: close code-review 330-r1 — budgets from the slowest machine, the bench in Validate, AC1 through the execution thread (#330)
H2: the benchmark budgets were calibrated on the author's sandbox with a
1.14x margin — the review runner measured tsFullCandidateMs at 169-171 ms
against a 100 ms ceiling. Budgets now keep the spec's 2-3x allowance over
the SLOWEST observed machine, and the benchmark runs as a step of the
Validate perf job on every push (it needs no browser and no bundle), not
only inside the weekly mutation gate.

M1: the promised AC1 backend test exists now and does what AC1 means: it
patches validate_junction_limits with a thread-recording wrapper inside the
real HA harness — on the event loop that would be MainThread — and proves
the verdicts survived the move (a clean write is accepted, a write adding a
spike is refused with junction_limit_angle). Spec revision 4 rewrites AC1
around this invariant instead of a fragile millisecond assertion.

M2: §4.6 equivalence is now behavioural on both sides (three boundary
fixtures each: as-is counts equal through-migration counts, TS and python),
and the parity suite gained the §7 boundary fixtures (exact 15°, exact
20 cm, the thickness-step filler run, exact 5 cm).

H1 was already closed by 7513f93d (the review ran on the previous HEAD):
check-docs is green on this tree — the screenshots and their manifest come
from one capture run.

Issue: #330
User-Visible: no
2026-08-28 03:07:44 +03:00
Codex 4e8e00fa73 docs: regenerate the screenshot set consistently after the rebase (#330)
The rebase resolved docs/images/screenshots.json to the dev side while the
PNG files stayed from this branch's capture — CI correctly refused the
mismatched pair. One local capture regenerates both halves from the same
run, so hashes and the source fingerprint agree again.

Issue: #330
User-Visible: no
2026-08-28 03:07:44 +03:00
Codex 658c955553 docs: spec #330 revision 3 — §4.7, the П5 shared pass the benchmark uncovered
Writing the §5 benchmark honestly exposed a cost the point measurements of
П1-П4 could not see: П5 recomputed the full junction topology and masonry
union PER ROOM — 4.2 s per candidate on the benchmark grid. Revision 3 adds
the shared-pass cut (one topology pass per check, the union only when
multi-wall nodes exist, and the resize path reusing its own preflight
artifact) with the measured numbers. Budgets in §5 already assumed the fix;
they are now achievable and proven by the passing benchmark.

Issue: #330
User-Visible: no
2026-08-28 03:07:37 +03:00
Codex 7aaf5a72b0 test: the baseline-cache contract is pinned against the real method (#330)
The first AC4 unit exercised a re-implementation of the cache algorithm, so
the stale-cache mutant patched houseplan-card.ts and the unit stayed green —
the exact "looks like protection" failure the mutation gate exists to catch,
and it caught mine. The monolith is not compiled into test-build, so the
contract is pinned by source (the #293 technique): the epoch check, the
document-identity key and the §4.6 as-is branch must be present in
_junctionLimitsIntroduced. The behavioural half of AC4 lives in the smoke's
real pointer gesture (resizeBaselineCachedPerGesture).

Issue: #330
User-Visible: no
2026-08-28 03:07:37 +03:00
Codex 5fb4ce328b docs: refresh screenshot source fingerprint (#330)
Issue: #330
User-Visible: no
2026-08-28 03:07:37 +03:00
Codex c90f5bf052 perf: junction limits scale — executor, rev cache, linear П3/П4, shared masonry pass (#330)
Six cuts, zero verdict changes (spec §3; equivalence pinned by units, the
parity suite and the smokes):

- §4.1 the CPU chain of ws_config_set and ws_plan_optimize runs in the
  executor; write_lock still serialises writes, only the HA event loop is
  freed (2.8 s of blocking per 576-atom write before).
- §4.2 the stored document's violation counts are cached on the runtime by
  rev (store.py junction_baseline); a repeated write never re-judges
  `previous`. validate_junction_limits takes baseline_counts and returns the
  candidate's counts to cache after a successful save.
- §4.3 П3 builds its node index once per check in both mirrors
  (289→11 ms TS, 285→~50 ms py).
- §4.5 П4 uses a bucket grid with the threshold as cell size in both
  mirrors (104→19 ms TS, 372→44 ms py); pair enumeration switches to
  lexicographic order — same verdict set, equivalence pinned against a
  brute-force oracle on cell borders.
- §4.6 a document already carrying the current catalogue is judged as-is:
  a no-op re-migration cost 815 ms py / 69 ms TS. Legacy documents migrate
  exactly as before (the #329 H1 test stays green).
- §4.7 П5 shares one junction-topology pass per check and pays the masonry
  union only when multi-wall nodes exist — and the resize path hands over
  the preflight's own artifact, so a pointermove never builds the union
  twice (4.2 s → 88 ms full candidate on the benchmark grid).

The frontend baseline is cached per (document identity, config epoch): ten
pointermoves make N+1 limit computations, not 2N — pinned by the smoke on a
real pointer gesture.

demo/benchmark_junction_limits.mjs (npm run benchmark:junction-limits) pins
the budgets for both mirrors: TS full candidate ≤100 ms (measured 88), py
warm validate ≤250 ms (measured 45), cold legacy ≤3.5 s — that path is
one-off and lives in the executor.

Issue: #330
User-Visible: yes
2026-08-28 03:07:11 +03:00
claude[bot]andCodex 92f476556f docs: review document for #330
Issue: #330
User-Visible: no
2026-08-28 03:06:44 +03:00
Codex 3f73eced95 docs: spec #330 revision 2 — budgets from the profile, П4 bucket, no re-migration for current-version documents
r1-H1 was right twice: the rev cache never touched the candidate's migration,
and П4 is architecturally quadratic. Profiled instead of guessing: the money
is not in deepcopy (3 ms) but in _atomize (663k distance calls), and it runs
even for a document that already carries the current catalogue — 815 ms
python / 69 ms TS for a no-op migration. Two new cuts follow: §4.5 bucket
index for П4 (prototype: 372→44 ms, identical verdicts) and §4.6 current-
version documents are used as-is (an explicit revision of the "both sides
through one migration" wording, guarded by a new parity case: v9 input gives
the same verdict with and without migration).

r1-H2: AC4 now rests on the new benchmark that actually exercises the
junction code; benchmark_safe_resize is named as a non-proof. r1-M1: §9
adds the mandatory i18n/touch/risks/release sections.

Budgets in §5 are recomputed from measured post-fix prototypes with a 2-3x
allowance, including an honest row for the one-off cold legacy case.

Issue: #330
User-Visible: no
2026-08-28 03:06:44 +03:00
claude[bot]andCodex 802604e0c3 docs: review document for #330
Issue: #330
User-Visible: no
2026-08-28 03:06:44 +03:00
Codex ccadb7779e docs: spec #330 — junction limits performance
Four cuts, zero verdict changes: the ws_config_set validator chain moves to
the executor, the previous-document violation counts are cached by
config_rev, П3 builds its node index once per check in both mirrors, and the
frontend baseline is cached per config epoch. A new benchmark with budgets
pins the class of regression (O(n²) returning) in CI.

Measured on dev 2c20f2dc: a 576-atom plan costs 2.8 s in the HA event loop
per config write today; the spec's acceptance bar is ≤50 ms of loop time.

Issue: #330
User-Visible: no
2026-08-28 03:06:44 +03:00
claude[bot] eff786fafa docs: review document for #62
Проверка (CI) / Предполётные проверки: документация, провенанс, процесс (push) Failing after 2m36s
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 1m13s
Проверка (CI) / Классификация изменённых файлов (push) Successful in 1m43s
Проверка (CI) / HACS: валидация репозитория (push) Skipped
Проверка (CI) / Hassfest: манифест интеграции (push) Skipped
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Skipped
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Skipped
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
Проверка (CI) / Golden-кадры против принятых эталонов (push) Skipped
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Skipped
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Skipped
Issue: #62
User-Visible: no
2026-08-28 00:02:11 +00:00
Matysh c755a0efc4 fix: normalize i18n registry lookup keys
Issue: #62
User-Visible: no
2026-08-28 02:55:22 +03:00
claude[bot] d0d6cf57a7 docs: review document for #62
Issue: #62
User-Visible: no
2026-08-27 23:53:02 +00:00
Matysh 6540474ff5 refactor: centralize i18n language registry
Issue: #62
User-Visible: no
2026-08-28 02:44:26 +03:00
claude[bot] 2d4ace1b58 docs: review document for #62
Issue: #62
User-Visible: no
2026-08-27 23:36:15 +00:00
Matysh 0a203649fc docs: address i18n spec review
Issue: #62
User-Visible: no
2026-08-28 02:32:48 +03:00
claude[bot] 60f41a53af docs: review document for #62
Issue: #62
User-Visible: no
2026-08-27 23:31:30 +00:00
Matysh 21d47c52ba docs: update i18n registry specification
Issue: #62
User-Visible: no
2026-08-28 02:24:08 +03:00
Sergey MatyuninandMatysh 5d292831f3 docs: specify scalable i18n registry
Issue: #62
User-Visible: no
2026-08-28 02:20:15 +03:00
claude[bot] fffe9fb1f7 docs: review document for #29
Issue: #29
User-Visible: no
2026-08-27 23:17:53 +00:00
Matyshandclaude[bot] aa9444491d fix: align device catalog with plan rooms
Issue: #29
User-Visible: yes
2026-08-27 23:17:53 +00:00
claude[bot] 361ec47856 docs: review document for #29
Issue: #29
User-Visible: no
2026-08-27 23:17:53 +00:00
Matyshandclaude[bot] cab8d128bf feat: add device lifecycle catalog
Issue: #29
User-Visible: yes
2026-08-27 23:17:52 +00:00
claude[bot] 024866842c docs: review document for #29
Issue: #29
User-Visible: no
2026-08-27 23:17:52 +00:00
Matyshandclaude[bot] e04cf7e573 docs: preserve hidden device ghost mode
Issue: #29
User-Visible: no
2026-08-27 23:17:52 +00:00
claude[bot] ff9ac33149 docs: review document for #29
Issue: #29
User-Visible: no
2026-08-27 23:17:52 +00:00
Matyshandclaude[bot] eeae6752d9 docs: update device inbox lifecycle spec
Issue: #29
User-Visible: no
2026-08-27 23:17:52 +00:00
Claude 1866c0921e ci: ревьюер не перегоняет зелёные гейты, локальный набор перед пушем
Ревьюер гонял tsc, юниты и сборку заново в каждом раунде, хотя Validate на том
же SHA уже зелёный. Промпт прямо это требовал. Теперь шаг `validated` спрашивает
у Validate состояние ровно этого SHA, и доказательство такое же строгое, как у
reuse-маркеров (#208): не «недавно было зелено», а completed success на этом
коммите. После ребейза SHA другой, прогона для него нет — ревьюер честно гоняет
сам, и промпт это говорит.

Что Validate не покрывает, в примечании названо отдельно: смоки по диффу,
golden при правке рендера, инварианты на конкретной конфигурации. Иначе
экономия превратилась бы в «CI зелёный, значит всё проверено».

scripts/pre-push-gate.mjs — локальный набор: tsc, юниты, смоки по диффу
(smoke-select), мутанты по диффу (mutation-gate --changed). Замер на реальном
диапазоне 953f675~1..953f675: 46 секунд на всё вместе с двумя смоками.

Три свойства, без которых набор бесполезен: не останавливается на первом
упавшем; громко перечисляет, чего не проверял; не претендует на полноту. Бандл
не собирает — раскладывает закоммиченный dist, а свежесть проверяет сам продукт
через assertFreshDemoBundle внутри смока.

В хуке выключен по умолчанию: 20-45 секунд на каждый пуш, включая пуш одной
строки документации, — цена осознанная, включается HP_PREPUSH_GATE=1.

Дельта-промпт для spec-ревью (пункт 2) уже существует: блок «объём разбора по
дельте» из #214 покрывает оба этапа и прямо называет «дифф файла ТЗ или тела
issue для spec». Ничего не добавлял.

Issue: #343
User-Visible: no
2026-08-28 02:15:03 +03:00
Claude e63b03a0ee docs: лёгкий трек становится путём по умолчанию
Прежде полный трек был бесплатен, а выбор лёгкого требовал обоснования. Цена —
2.9 ревью-документа на задачу и до шести на одну issue (#329, #316, #290), при
том что Medium-находки всё равно чинятся в той же задаче без отдельного цикла.

Порог не изменился: критерии §5 те же и обязательны все одновременно. Изменилась
сторона доказательства — в S2-analysis называется критерий, который задача НЕ
проходит, если идёт полным треком. «Обычный трек» без названного критерия
обоснованием не является.

Правка идёт и в AGENTS.md: там трек описан как «shortcut для мелкой работы», а
это ровно та формулировка, из-за которой полный трек остаётся умолчанием на
практике. AGENTS.md стоит вторым в порядке доверия, поэтому без него правка
канона поведение не меняет.

Бюджет четырёх циклов, арбитраж владельца, обязательность ТЗ на полном треке и
правило «ревью до мержа» не тронуты.

Issue: #338
User-Visible: no
2026-08-28 02:08:14 +03:00
Claude 2b1964f973 ci: бандл собирается один раз, четыре лёгкие джобы стали одной
Бандл собирался пятью job независимо: три шарда смоков, golden, перф-смок —
каждая гоняла `bundle:sync`, то есть `tsc --noEmit` плюс rollup. Теперь его
собирает `frontend` и выкладывает артефактом, остальные скачивают и раскладывают
`bundle-sync.mjs`. Подмену артефакта отдельной проверкой ловить не нужно:
assertFreshDemoBundle сверяет вшитый в бандл отпечаток с sourceFingerprint
выкачанного дерева, и каждая браузерная job делает это перед первым кадром.

`npm ci` остаётся во всех: браузерным job нужен playwright из node_modules, а не
только бандл. Артефакт node_modules был бы медленнее `npm ci` с тёплым кэшем.

docs, process-workflow-sync, provenance и process-gate стали шагами одной job
`preflight`. Независимость сохранена намеренно: у каждого шага
continue-on-error, вердикт в конце падает и перечисляет всё упавшее сразу.
Прежняя запись «краснеет сам и не роняет остальные» продолжает действовать — на
уровне шагов, с той же гранулярностью в логе.

hacs и hassfest не тронуты: предложение сузить их до dev и тегов уже выполнено
классификатором `changes` — на ветках задач они и так идут только при правке
манифестов, а на dev фильтров нет намеренно (гейт беты требует, чтобы «зелёный
Validate» значил одно и то же).

test/validate-workflow.test.mjs закрепляет то, что в диффе строк не видно:
висячая зависимость `needs` не роняет YAML, а молча пропускает job навсегда.
Три мутанта проверены руками — висячая зависимость, вернувшаяся вторая сборка,
шаг без continue-on-error, — каждый убит.

Issue: #336
User-Visible: no
2026-08-28 02:04:40 +03:00
Claude 4800c44e51 test: локальная съёмка golden допустима по доказательству, а не по доверию
Прежде эталон принимался только из артефакта CI: растеризация шрифтов на другой
машине может отличаться, а доказать обратное было нечем. Цена — два полных
прогона на каждый визуальный фикс, при версии матрицы 48 она платится часто.

Доказательство теперь эмпирическое: среда равна раннеру, если каждая сцена,
которую менять не собирались, совпала со своим эталоном. Расхождение
растеризации спрятать нельзя — оно задевает все сцены с текстом. Ревьюер
объявляет намерение через --expect-change, всё разошедшееся помимо списка
приёмку запрещает. Поэтому неверно угаданный тег образа не может испортить
эталоны: он может только не сработать.

То же правило независимо от среды запрещает «принять всё, чтобы CI позеленел» —
именно так эталон перестаёт быть эталоном, молча и одной командой.

scripts/golden-container.mjs снимает кандидатов в образе Playwright той же
версии, что залочена в package-lock. Хозяйский node_modules прячется анонимным
томом: он собран под Windows, и npm ci внутри контейнера сломал бы дерево.

Обёртка, а не правка demo/golden/accept.mjs, — намеренно. sourceFingerprint
включает ВСЕ .mjs из demo/golden, включая accept.mjs и policy.mjs, которые
исполняются после съёмки и ни одного пикселя изменить не могут. Их правка
объявляет устаревшими бандл и оба манифеста, то есть требует ровно того двойного
цикла, который эта задача убирает. Сужение корпуса отпечатка — отдельная задача:
сам source-fingerprint.mjs в корпусе, и одна пересборка бандла неизбежна.

Issue: #334
User-Visible: no
2026-08-28 01:59:45 +03:00
Codex 2c20f2dc35 perf: mutation-gate builds the bundle only for browser guards, compiles incrementally, shards and diffs (#332)
Four independent cuts into the 2-4 hour full run, none touching the contract
"a mutant must turn its guard red":

- guardNeedsBundle: rollup runs only for guards that open the built bundle
  (demo/ smokes, golden captures, bundle:sync) — 68 of 253 registry entries.
  Unit and backend guards never read dist/ as a build artifact (verified
  against every test that mentions dist/**: they read the git checkout or
  synthetic files), so 185 mutants skip the most expensive step entirely.
- seedTestBuild + incremental tsc: the mutant worktree starts from the main
  tree's warm test-build/ and .tsbuildinfo; tsc compares file hashes, not
  mtimes, so the fresh checkout stays warm and only the mutated delta is
  recompiled. This also speeds up the long guards that run tsc themselves.
- --changed[=range]: run only mutants whose patch files are touched by the
  diff (origin/dev..HEAD by default). An empty selection is an honest success
  with an explicit message — the full registry remains the pre-release
  contract, per the workflow comment.
- --shard=i/n: deterministic interleaved slices; the workflow runs a 4-way
  matrix, and a warm test-build step feeds every shard. Interleaving spreads
  the expensive browser mutants across shards instead of clumping them.

Measured per mutant on this machine: unit 12-13 s (was ~50-70 s), backend
6 s, browser 32 s (unchanged — the bundle is genuinely needed there). Full
run estimate drops to ~70 sequential minutes, ~20 on four shards.

Unit coverage: guard classification on real registry shapes, a floor on both
classes so the split cannot silently collapse, changed-selection semantics,
and shard completeness/disjointness with an anti-clumping bound.

Issue: #332
User-Visible: no
2026-08-28 01:33:36 +03:00
claude[bot] 2c4b505f1a docs: review document for #329
Проверка (CI) / Процесс: process.yml идентичен в main и dev (push) Successful in 2m48s
Проверка (CI) / Провенанс коммитов: трейлеры и эталоны (push) Successful in 2m31s
Проверка (CI) / Процессный гейт: диапазон, трейлеры, статусы issue (push) Successful in 2m30s
Проверка (CI) / Классификация изменённых файлов (push) Successful in 36s
Проверка (CI) / HACS: валидация репозитория (push) Skipped
Проверка (CI) / Hassfest: манифест интеграции (push) Skipped
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Skipped
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 38s
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Skipped
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Skipped
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
Проверка (CI) / Golden-кадры против принятых эталонов (push) Skipped
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Skipped
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Skipped
Проверка (CI) / Документация: гайды, ченджлоги, скриншот-индекс (push) Failing after 47s
Issue: #329
User-Visible: no
2026-08-27 21:49:59 +00:00
Codex 273b0d5ecb test: the backend mutant follows the registry convention (#329 r2-H1)
The reviewer is right twice over. My previous commit fixed the red CI by
relaxing the contract — a guard could name a `.py` file — when the registry
already had a convention for exactly this case: every backend mutant runs
`node scripts/backend-test-guard.mjs <pattern> <file>`, which owns the python
executable choice and the `-k` selection. Bending a rule to fit my one-off is
the worse of the two possible fixes, so the contract goes back to demanding a
`.mjs` guard, and junction-limit-backend-raw-baseline now uses the helper and
targets the one test that proves the migration
(test_legacy_baseline_is_judged_after_the_same_migration).

Re-verified: registry --check clean, the mutant still catches its regression
1/1, npm test 1390 passed / 0 failed.

Issue: #329
User-Visible: no
2026-08-28 00:41:30 +03:00
claude[bot] f2c2084dd4 docs: review document for #329
Issue: #329
User-Visible: no
2026-08-27 21:37:28 +00:00
Codex 26c6e87079 test: a mutant guard may be pytest, not only node (#329)
The registry contract demanded that every guard name a `.mjs` file, which was
true until this task added the first backend mutant —
junction-limit-backend-raw-baseline is guarded by pytest, and the mutation-gate
job already installs it. My mistake: I ran `--check` and the single mutant
after adding it, but not the unit suite that owns the registry contract, so CI
caught what I should have.

The contract keeps its point: a guard must name a file that exists.

Issue: #329
User-Visible: no
2026-08-28 00:28:12 +03:00
Codex 0824e51014 test: prove AC10 — Optimize adds no junction violation (#329 M4)
AC10 was asserted, never shown. Optimize runs alignAllToGrid and
repairNearAxisRoomWalls, which move nodes by fractions of a centimetre, and
none of П1-П5 carries a margin wider than the grid step in general — so
"obviously true by construction" was not available.

Two units, both counting violations the way the write barrier does (each side
through commitWallSegmentModel first):
- the owner's fixture in legacy storage — the inherited apex is there before
  Optimize, and no rule's count grows after;
- the П4 boundary — two rooms exactly 5 cm apart, where snapping could have
  pulled a node under the limit, stay clean.

The first test asserts the baseline actually carries a violation, so it cannot
pass by measuring an empty plan; violationsByRule fails loudly if the space or
its catalogue goes missing, for the same reason. Spec revision 7 records the
proof and the other three review answers.

Issue: #329
User-Visible: no
2026-08-28 00:24:34 +03:00
Codex 574fea98ab docs: one limits section, and Resize does raise a toast (#329 M2, M3)
The Russian guide carried the junction-limits section twice, word for word.
And both guides described Resize as silently stopping, in contrast to a toast
from drawing and Thickness — it stops AND names the rule once per gesture
(resize.limit_stopped, pinned by the smoke). Wording follows the code.

Issue: #329
User-Visible: no
2026-08-28 00:24:34 +03:00
Codex 8f0b6b97e8 refactor: drop the chamfer helpers the honest apex made dead (#329 M1)
002795f7 said "the clip helper and its cap plumbing are gone" while leaving
clipPolygonOutsideCap(), degenerateApexCaps() and the apexCaps ring field in
place — exported, uncalled and untested. They belong to the flat chamfer the
owner rejected; the apex now ends in one point on both faces, so the quads
have no caller and no meaning.

The orphaned JSDoc block that described degenerateApexCaps went with them, and
the wallBodiesGeometry documentation this change had earlier separated from
its function is reattached: the #329 constant and predicate now sit above it.

Golden verify stays green on the whole matrix, which is the evidence the
removed code was indeed dead.

Issue: #329
User-Visible: no
2026-08-28 00:24:34 +03:00
Codex 0bb42caeff fix: the backend judges both sides after the same migration (#329 H1)
The limits read `wall_segments`, so a document older than the catalogue
reports no walls at all — and therefore no violations, whatever its geometry.
Comparing that raw baseline against a candidate the card had already migrated
counted every inherited violation as new, and a legacy plan could not take an
unrelated edit at all: renaming a room was refused with junction_limit_angle.
Spec §3 forbids exactly this, and the frontend had already learned the same
lesson in 4758767e; the backend mirror simply never got the second half.

validate_junction_limits now runs both documents through
commit_wall_segment_model before counting. A document that cannot be migrated
is not this validator's verdict — the wall-model barrier owns that error and
reports it with its own code — so it degrades to "no baseline to inherit".

The regression is pinned twice: a test that asserts the legacy baseline reads
clean raw and carries the apex once migrated, and the mutant
junction-limit-backend-raw-baseline. Both fixtures that exercise the barrier
were rebuilt as real documents (rooms plus walls), because the previous ones
put walls in wall_segments with no rooms and did not survive migration.

Issue: #329
User-Visible: no
2026-08-28 00:23:54 +03:00
claude[bot] e1df015d93 docs: review document for #329
Issue: #329
User-Visible: no
2026-08-27 21:01:48 +00:00
Codex 7017896eb2 docs: refresh screenshot source fingerprint after the rebase (#329)
The branch was rebased onto the extracted resize controller (#264), which
changes the source fingerprint the documentation screenshots are pinned to.
The images themselves are byte-identical — only the recorded fingerprint moves.

Issue: #329
User-Visible: no
2026-08-27 23:36:20 +03:00
Codex ad3f998147 test: golden baseline for the legacy sharp apex (#329)
Reviewed the candidate produced by the Linux CI job of run 33106626544 on
issue/329-junction-limits, where golden failed with exactly one line —
"missing-baseline sharp-apex-legacy-dark" — and accepted only that image. The
nine unrelated baselines whose bytes drifted in the same artifact were
restored to their reviewed versions, so this commit changes one picture.

Baseline-Reviewed: run 33106626544, job 98638432113 (Golden-кадры против принятых эталонов)
Release: v1.68.2
Issue: #329
User-Visible: no
2026-08-27 23:32:07 +03:00
Codex 8945e04fe4 feat: backend mirror of the junction limits (#329 §5, AC9)
custom_components/houseplan/junction_limits.py repeats П1-П4 for the write
barrier in websocket_api, counting per rule so an inherited violation still
round-trips, and raises JunctionLimitError with the stable code
junction_limit_<rule>.

П5 is deliberately not mirrored — it judges the rendered wall bodies, and a
second mitre/inset pipeline in Python would drift more dangerously than the
rule it guards. Optimize stays outside the check for the same reason migration
and import do: it repairs existing geometry.

test_parity_with_the_frontend_checks feeds identical fixtures to the TS
functions and to this module and demands the same verdict, so the two
implementations cannot silently diverge.

Issue: #329
User-Visible: no
2026-08-27 23:32:07 +03:00
Codex 7a74a37c6e test: golden scene for the legacy sharp apex (#329 AC6)
The owner's spike room (≈9.9°, 15 cm walls) is extracted into
test/fixtures/329-sharp-apex.json and rendered on its own so a returning
trident, a flat chamfer or a jagged edge fails the pixel gate. Baseline
follows from the CI candidate, as the process requires.

Issue: #329
User-Visible: no
2026-08-27 23:32:07 +03:00
Codex 2b1f03be2d test: the Resize channel of the junction limits, driven by a real pointer (#329 AC7a)
Measured what Resize itself already forbids: a room cannot be squeezed below
30 cm (two 15 cm walls), so П3 and П5 are unreachable through shrinking and
the gate merely fails closed there. П4 IS reachable on a fine grid, so the
smoke drags a real handle on a 2 cm grid: two rooms 10 cm apart, a 6 cm pull
would leave 4 cm between foreign nodes, the wall stops at 6 cm and exactly one
toast names the 5 cm rule.

Spec revision 6 records both the measurement and the two corrections it forces
on AC7a: a dimmed handle cannot express a per-step limit, and the plan is NOT
byte-unchanged — the allowed part of the gesture is a legitimate edit.

Issue: #329
User-Visible: no
2026-08-27 23:31:42 +03:00
Codex f514fb27fe feat: junction limits refuse the write in every editing surface (#329)
П3 measures the WALL, not the catalogue atom: a short filler segment that
compensates a thickness step (owner's fixture, 5 cm = (30-20)/2) is a legal
continuation of a long same-thickness wall, so the rule walks the maximal
collinear run through the shared nodes before judging the length.

Resize stops at the last allowed position and names the broken rule instead
of the generic "geometry cannot be saved"; the Thickness dialog refuses
through its own toast. Both channels are pinned by demo/smoke_junction_limits
plus three mutants (angle threshold, write barrier, degenerate apex bevel).

Issue: #329
User-Visible: yes
2026-08-27 23:31:42 +03:00
Codex 7d15bd06e9 test: island rooms respect the 20 cm segment limit (#329 П3)
Owner decision (chat, 2026-08-27): keep П3 and fix the smoke. A 10 cm island
room is a column, and columns have their own tool (wall_columns) — that was
the reasoning behind the limit in the first place. The island of the smoke is
now 25 cm, and a new case pins the contract: a 10 cm island is refused.

Issue: #329
User-Visible: no
2026-08-27 23:31:17 +03:00
Codex 002795f7c8 fix: no jags on the edges of a degenerate apex (#329 §4)
Owner report: small serrations remained on the outer edges between the inner
and the outer vertex. Measured on the fixture ring: two ~4 cm steps plus four
micro-vertices at the tip. Their source was the inset contour's two-point
bevel folding into a bow-tie, and the earlier half-plane clip of that fold,
which left a 0.2 cm sliver the boolean union turned into steps. The inset now
ends in ITS own mitre point at a degenerate apex — mirroring the sharp outer
tip — so there is no fold to clip and no sliver to smear: the room ring is
exactly three vertices, every side longer than the half depth. The clip
helper and its cap plumbing are gone. The user-visible wording of this work
already stands in both changelogs from the #329 entry.

Issue: #329
User-Visible: no
2026-08-27 23:31:17 +03:00
Codex 4758767e0c fix: junction limits judge both sides after the same migration (#329)
The baseline for inheritance was the raw previous document, which for a
legacy space carries no wall catalogue at all — so every inherited short
segment of a real plan looked new and the resize smoke's legitimate write was
refused (executed: two 5 cm segments against their own 30 cm thickness).
Both sides now cross commitWallSegmentModel first, and inheritance is counted
per rule rather than per subject, because a structural write re-keys the
carriers it re-atomises.

Issue: #329
User-Visible: no
2026-08-27 23:30:52 +03:00
Codex 214355f424 fix: a degenerate sharp corner renders as a normal sharp apex (#329 §4)
Owner correction (chat, 2026-08-27): no flat chamfer at the tip — a plain
sharp apex. Proven by execution on the issue fixture: the outset contour fell
back to a two-point bevel (the 4·h mitre limit against an 87 cm reach) while
the inset contour folded into a bow-tie, and subtracting that fold carved the
V-notches — together they made the trident. Now a degenerate corner (below 15
degrees, inner faces meeting inside both walls) contributes ONE outset point
at the plan's own vertex — no bevel, no metres-long mitre needle — and its
inset is clipped at the convergence line so no fold is subtracted. Plain and
merely sharp pairs keep the full mitre of #310. The write-side limits of
П1-П5 stop new plans from creating such corners at all.

Issue: #329
User-Visible: yes
2026-08-27 23:30:26 +03:00
Codex 6fc57f2ae8 feat: pure wall-junction limit checks (#329 П1-П5)
The owner's five limits as pure functions: minimum 15 degrees between
neighbouring rays of a node (a straight wall through the node is a 180 pair,
not a violation), at most 6 walls per node, a segment at least
max(20 cm, its own thickness), 5 cm clearance between non-incident nodes and
between a node and a foreign wall (a T-joint sitting exactly on that wall is
incidence, not a near miss), and a room interior of at least 25 cm2 after the
masonry is subtracted. Thresholds are absolute and do not scale with cell_cm.
newViolations() implements the spec's inheritance boundary: only violations
introduced by the write are reported.

Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
claude[bot]andCodex 806f7f784c docs: review document for #329
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
Codex 3120924e26 docs: spec #329 revision 3 — AC7 proves the per-surface refusal channels
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
claude[bot]andCodex 8205897149 docs: review document for #329
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
Codex 73b665ede6 docs: spec #329 revision 2 — per-surface refusal channels, AC5 split, absolute thresholds
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
claude[bot]andCodex d67bb11de0 docs: review document for #329
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
Codex 5266d2f6be docs: spec #329 — wall junction limits and an honest sharp apex
Issue: #329
User-Visible: no
2026-08-27 23:29:49 +03:00
claude[bot] 8161394178 docs: review document for #267
Issue: #267
User-Visible: no
2026-08-27 19:57:33 +00:00
Matysh 118062bb8c test: close device presentation review findings
Issue: #267
User-Visible: no
2026-08-27 22:41:46 +03:00
claude[bot] 8b0ad08db2 docs: review document for #267
Issue: #267
User-Visible: no
2026-08-27 19:31:52 +00:00
Matysh 6efc315ba6 refactor: make device presentation decisions explicit
Issue: #267
User-Visible: no
2026-08-27 22:06:10 +03:00
claude[bot] b87322ed5f docs: review document for #267
Issue: #267
User-Visible: no
2026-08-27 18:26:00 +00:00
Matysh c5e145269b docs: specify device presentation decisions
Issue: #267
User-Visible: no
2026-08-27 21:17:18 +03:00
claude[bot] 2294d46f3f docs: review document for #264
Проверка (CI) / Документация: гайды, ченджлоги, скриншот-индекс (push) Failing after 1m0s
Проверка (CI) / Процесс: process.yml идентичен в main и dev (push) Successful in 42s
Проверка (CI) / Провенанс коммитов: трейлеры и эталоны (push) Successful in 48s
Проверка (CI) / Процессный гейт: диапазон, трейлеры, статусы issue (push) Failing after 57s
Проверка (CI) / Классификация изменённых файлов (push) Successful in 55s
Проверка (CI) / HACS: валидация репозитория (push) Skipped
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 47s
Проверка (CI) / Hassfest: манифест интеграции (push) Skipped
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Skipped
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Successful in 6m44s
Проверка (CI) / Golden-кадры против принятых эталонов (push) Failing after 15m20s
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Failing after 17m46s
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Failing after 20m39s
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Failing after 8m1s
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Failing after 25m38s
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
Issue: #264
User-Visible: no
2026-08-27 18:05:24 +00:00
Matysh 722f26a3c5 docs: refresh screenshot source fingerprint
Issue: #264
User-Visible: no
2026-08-27 20:43:34 +03:00
Matysh 4a833c5e13 test: keep resize preflight mutant effective
Issue: #264
User-Visible: no
2026-08-27 20:32:12 +03:00
Matysh ba66698f55 refactor: extract resize controller
Issue: #264
User-Visible: no
2026-08-27 20:30:07 +03:00
claude[bot] ef32d2326f docs: review document for #264
Issue: #264
User-Visible: no
2026-08-27 16:58:08 +00:00
Matysh 536b750d03 docs: address resize controller spec review
Issue: #264
User-Visible: no
2026-08-27 19:53:48 +03:00
claude[bot] c46c714c58 docs: review document for #264
Issue: #264
User-Visible: no
2026-08-27 16:51:56 +00:00
Matysh 6e1aea3ae7 docs: add resize controller user scenario
Issue: #264
User-Visible: no
2026-08-27 19:43:45 +03:00
Matysh a3e329950a docs: specify resize controller extraction
Issue: #264
User-Visible: no
2026-08-27 19:43:17 +03:00
Codex 78850a8733 build: rebundle after the package.json script addition
Проверка (CI) / Провенанс коммитов: трейлеры и эталоны (push) Failing after 8m3s
Проверка (CI) / Процессный гейт: диапазон, трейлеры, статусы issue (push) Failing after 8m2s
Проверка (CI) / Классификация изменённых файлов (push) Successful in 52s
Проверка (CI) / Переиспользование: это дерево уже проверено (push) Successful in 1m30s
Проверка (CI) / HACS: валидация репозитория (push) Failing after 35s
Проверка (CI) / Процесс: process.yml идентичен в main и dev (push) Successful in 7m32s
Проверка (CI) / Документация: гайды, ченджлоги, скриншот-индекс (push) Successful in 7m53s
Проверка (CI) / Hassfest: манифест интеграции (push) Failing after 23s
Проверка (CI) / Фронтенд: типы, юниты, мутанты, синхрон бандла (push) Successful in 13m14s
Проверка (CI) / Бэкенд: pytest в Home Assistant (push) Failing after 13m21s
Проверка (CI) / Golden-кадры против принятых эталонов (push) Failing after 26m7s
Проверка (CI) / Смоки в браузере (шард 1 из 3) (push) Failing after 31m4s
Проверка (CI) / Смоки в браузере (шард 3 из 3) (push) Failing after 34m5s
Проверка (CI) / Перф-смок: бюджет времени кадра (push) Failing after 12m59s
Проверка (CI) / Смоки в браузере (шард 2 из 3) (push) Failing after 45m46s
Проверка (CI) / Смоки: все шарды зелёные (push) Skipped
The bundle embeds the source fingerprint, which covers package.json; the
release:notes script addition moved it, so the three bundle copies must be
rebuilt in the same change.

Issue: #328
User-Visible: no
2026-08-27 18:59:10 +03:00
Codex 1ac91e43fd build: stable release notes follow the owner's aggregation rules (#328)
A stable body aggregates the changelog since the previous STABLE release,
not since the last beta; in-beta-only bugfixes are excluded by the curator
(the draft lists every candidate with its source section); the small-fixes
filler line is legal only when the range carries user-visible work not
itemised in the body — a single-issue hotfix ships without it, and body
bullets must link their issues so the rule stays checkable.
scripts/release-notes.mjs prints the aggregation draft and verifies
docs/RELEASE-NOTES.md (npm run release:notes -- <tag> [--verify]); the
verifier rejects the v1.68.1-style empty filler by execution. STATUS.md
release mechanics updated in the same commit.

Issue: #328
User-Visible: no
2026-08-27 18:55:41 +03:00
Codex 6a3ac52258 ci: the performance-workflow contract test follows the renamed labels
The test pins the literal workflow name and the release-gate label; both
moved to the Russian names of #327.

Issue: #327
User-Visible: no
2026-08-27 18:50:02 +03:00
Codex 1a8b480355 ci: every workflow and job carries a human-readable Russian name
Owner decision (chat, 2026-08-27): the running check's name must say what it
does, in Russian. Scripts locate workflows by file name (release-gate.mjs ->
validate.yml), so display names are free; job ids and needs are untouched.
The same content is cherry-picked to main because release workflows execute
from the default branch and process.yml must stay identical in main and dev.

Issue: #327
User-Visible: no
2026-08-27 18:46:53 +03:00
Matysh ec957a7c88 merge: reconcile process repair into dev 2026-08-27 17:41:28 +03:00
Matysh 557c44c392 docs: keep cherry-pick provenance trailers terminal
Document the terminal-trailer check required after cherry-pick -x.

Issue: #326
User-Visible: no
2026-08-27 17:41:00 +03:00
Matysh 7181535dec merge: reconcile post-release main into dev 2026-08-27 17:39:55 +03:00
Matysh fb162953a9 chore: repair provenance after HACS revert
Record the published trailer-order violation without rewriting history.

Issue: #326
User-Visible: no
2026-08-27 17:39:34 +03:00
Matysh 244fddeb3b merge: reconcile v1.68.1 main into dev 2026-08-27 17:37:52 +03:00
Codex ad73280b1b fix: restore zip_release — the HACS asset URL was never actually broken
Revert of caf6e6c3. The owner's field report disproved the diagnosis: HACS
2.0.5 downloads zip_release assets fine, because async_download_file strips
the 'tags/' prefix from the composed URL before requesting (hacs/base.py) —
the prefix only survives in the error-log message, which prints the
unnormalized URL. The dacha installed every beta and v1.68.0 through HACS
with zip_release active. The observed 'Failed to download' was a transient
network failure fetching release-assets.githubusercontent.com (five retries
exhausted, one occurrence), not a routing bug.

Issue: #325
User-Visible: no
(cherry picked from commit 5ad0c1cdc3)
2026-08-27 17:35:07 +03:00
Codex 5ad0c1cdc3 fix: restore zip_release — the HACS asset URL was never actually broken
Revert of caf6e6c3. The owner's field report disproved the diagnosis: HACS
2.0.5 downloads zip_release assets fine, because async_download_file strips
the 'tags/' prefix from the composed URL before requesting (hacs/base.py) —
the prefix only survives in the error-log message, which prints the
unnormalized URL. The dacha installed every beta and v1.68.0 through HACS
with zip_release active. The observed 'Failed to download' was a transient
network failure fetching release-assets.githubusercontent.com (five retries
exhausted, one occurrence), not a routing bug.

Issue: #325
User-Visible: no
2026-08-27 17:35:05 +03:00
Matysh b8976415ac chore: release v1.68.1
Owner-approved emergency stable hotfix; the release commit changes only version fields, generated bundles and release metadata.

Issue: #324
User-Visible: yes
2026-08-27 17:26:26 +03:00
claude[bot]andMatysh 296df95e32 docs: review document for #324
Issue: #324
User-Visible: no
2026-08-27 17:26:26 +03:00
Matysh acc30c2da7 fix: materialize empty wall catalog for new spaces
Issue: #324
User-Visible: yes
2026-08-27 17:26:26 +03:00
Matysh ca383febef chore: release v1.68.1
Owner-approved emergency stable hotfix; the release commit changes only version fields, generated bundles and release metadata.

Issue: #324
User-Visible: yes
2026-08-27 17:17:12 +03:00
Codex 39eda1e136 fix: drop zip_release until HACS builds a valid asset URL for versioned installs
Installing a specific version from the HACS catalog fails: HACS composes the
zip_release asset URL from its internal ref 'tags/<version>' without
stripping the prefix (hacs/integration base.py:932 -> download_zip_files ->
github_release_asset), requesting releases/download/tags/v1.68.0/... which
GitHub cannot serve. A mirror tag with a slash is not servable either.
Without zip_release HACS falls back to the regular tag-tree download that
worked before 2026-08-08. Release assets keep being attached
(release-zip.yml untouched) so re-enabling after the upstream fix is a
one-line revert.

Issue: #325
User-Visible: no
(cherry picked from commit caf6e6c355)
2026-08-27 17:11:21 +03:00
Codex caf6e6c355 fix: drop zip_release until HACS builds a valid asset URL for versioned installs
Installing a specific version from the HACS catalog fails: HACS composes the
zip_release asset URL from its internal ref 'tags/<version>' without
stripping the prefix (hacs/integration base.py:932 -> download_zip_files ->
github_release_asset), requesting releases/download/tags/v1.68.0/... which
GitHub cannot serve. A mirror tag with a slash is not servable either.
Without zip_release HACS falls back to the regular tag-tree download that
worked before 2026-08-08. Release assets keep being attached
(release-zip.yml untouched) so re-enabling after the upstream fix is a
one-line revert.

Issue: #325
User-Visible: no
2026-08-27 17:11:19 +03:00
claude[bot] 4d9542b138 docs: review document for #324
Issue: #324
User-Visible: no
2026-08-27 13:45:07 +00:00
Matysh 69e410fea3 fix: materialize empty wall catalog for new spaces
Issue: #324
User-Visible: yes
2026-08-27 16:33:47 +03:00
Matysh 1edf5cb3a7 perf: reuse masonry after cancelled resize
Issue: #322
User-Visible: no
2026-08-27 15:38:32 +03:00
Matysh 9b7a700fd6 test: keep edit walk behind optimize preflight
Issue: #322
User-Visible: no
2026-08-27 15:10:46 +03:00
Matysh 780b7a6c7a perf: reuse wall topology for light transport
Issue: #322
User-Visible: no
2026-08-27 15:00:30 +03:00
claude[bot] 674607694f docs: review document for #323
Issue: #323
User-Visible: no
2026-08-27 11:12:03 +00:00
Codex fb8fdf479c docs: both user guides install from the HACS default catalog too
CODE-REVIEW-323-r1 M1: the full guides README links to still taught the
custom-repository flow; step 1 now matches the README wording.

Issue: #323
User-Visible: no
2026-08-27 14:06:22 +03:00
claude[bot] 3b915bd077 docs: review document for #323
Issue: #323
User-Visible: no
2026-08-27 11:05:21 +00:00
Codexandclaude[bot] 69facb9918 docs: STATUS and TESTING reflect the default-catalog reality
The HACS row still described an open queue of 835 PRs; the fresh-install
checklist still started from a custom repository.

Issue: #323
User-Visible: no
2026-08-27 11:00:24 +00:00
Codexandclaude[bot] 343af61a61 docs: install House Plan from the HACS default catalog
Since 2026-08-25 the integration is in the HACS default catalog
(hacs/default#9004); the badge and both install sections no longer route
users through Custom repositories — a plain HACS search finds it.

Issue: #323
User-Visible: no
2026-08-27 11:00:24 +00:00
Matysh 59a23de81b perf: reuse validated wall geometry
Issue: #322
User-Visible: no
2026-08-27 13:58:06 +03:00
Matysh 87ff906545 perf: keep room contour cache out of editor previews
Issue: #322
User-Visible: no
2026-08-27 13:11:03 +03:00
Matysh 298030d47e perf: reuse structural room contours
Issue: #322
User-Visible: no
2026-08-27 12:58:02 +03:00
Matysh 0c3aacdd34 test: keep freshness mutant aligned
Issue: #322
User-Visible: no
2026-08-27 12:28:38 +03:00
Matysh 134afb658e docs: record exact-tree performance freshness
Issue: #322
User-Visible: no
2026-08-27 12:24:27 +03:00
Matysh 6a96254956 test: validate comparison bundles against their own trees
Issue: #322
User-Visible: no
2026-08-27 12:23:54 +03:00
Matysh de7554e9e0 docs: record the parallel v1.68.0 performance gate
Issue: #322
User-Visible: no
2026-08-27 12:13:46 +03:00
Matysh 8210af6472 ci: parallelize independent full performance profiles
Issue: #322
User-Visible: no
2026-08-27 12:13:26 +03:00
Matysh 376a2db599 docs: record the v1.68.0 gate repair
Issue: #322
User-Visible: no
2026-08-27 11:06:33 +03:00
Matysh 6b9f5eea96 ci: support legacy performance baselines
Issue: #322
User-Visible: no
2026-08-27 11:04:38 +03:00
Matysh 59bf4665f6 docs: record v1.68.0 final release gates
Issue: #322
User-Visible: no
2026-08-27 10:51:38 +03:00
Matysh 3a562f2bd6 chore: return main process history before v1.68.0
Issue: #322

User-Visible: no
2026-08-27 10:49:01 +03:00
Matysh aaf7611490 build: promote v1.68.0 after beta.4
Issue: #322
User-Visible: yes
2026-08-27 10:29:04 +03:00
Codex 3d4e5580fa build: prepare v1.68.0-beta.4 candidate
Package the two beta.3 write-blocker fixes: the model-upgrade stale-client
guard and the self-resolving zero-wall migration.

Issue: #316
Issue: #319
User-Visible: yes
2026-08-27 08:11:42 +03:00
Codex 1c598a287a fix: the room_wall_ids invariant applies to model v8 and later again (#316, review r4 H1)
The r3-M1 fix was applied as a mechanical substring replacement and flipped
BOTH model_version conditions in the file; the independent room_wall_ids
invariant (#244/#252) silently stopped checking every v9+ document. Only the
opening_host requirement is scoped to model v8 — room_wall_ids is back to
'v8 and every later version', now pinned by its first regression test
(phantom wall_ids reported on v9 and v8; executed red on the broken
comparison, green after the fix).

Issue: #316
User-Visible: no
2026-08-27 00:21:23 +03:00
Codex 05ec0a1de8 fix: the model-invariants CLI accepts the unhosted degraded opening of model v9 (#316, review r3 M1)
Since #316 §3.3 an unhosted contour opening is a valid v9 state — the
migration keeps an opening with no in-place carrier as data. The CLI still
reported it as an opening_host violation for every model_version >= 8; the
requirement now applies to model v8 documents only. The regression test is
proven able to fail on the old comparison (executed red), and the reviewer's
CLI reproduction now finishes clean.

Issue: #316
User-Visible: no
2026-08-27 00:08:15 +03:00
Codex b29472a887 test: re-pin the #316 baselines to the review-fix source
Pixels are untouched — the r1/r2 review fixes change migration data flow,
not rendering; every scenario hash stays byte-identical to the frames of the
reviewed CI run. Only the source fingerprint moves to match the fixed tree.

Issue: #316
User-Visible: no
Release: v1.68.0-beta.4
Baseline-Reviewed: https://github.com/Matysh/houseplan-card/actions/runs/33000824647
2026-08-26 23:41:35 +03:00
Codex 0f36ef55d5 fix: an opening without an in-place carrier migrates unhosted (#316, review r1 H1 + r2 M2)
CODE-REVIEW-316-r1 H1: the §3.3 degraded pool picked an angle-compatible wall
at ANY distance, but the backend geometry-match invariant («wall opening
geometry must match its host») requires the host to agree with the opening's
own x/y — the migrated document was rejected by CONFIG_SCHEMA and the write
wedged again on the schema layer. The pool is removed from both migrations
(TS and the Python mirror): without an in-place eligible carrier the opening
goes straight to the unhosted degraded state, exactly the alternative the
spec's «assumed freely changeable» section reserved; the spec is revision 6.
New tests replay the reviewer's reproduction on both sides, and the frontend
test is proven able to fail by restoring the pool (executed red).

CODE-REVIEW-316-r2 M2: the schema-level host check is shared with #132
partition openings, so its unhosted relaxation is now pinned by a regression
test — a stale writer that keeps a partition-hosted opening but silently
drops its host is still rejected by validate_partition_opening_hosts.

Issue: #316
User-Visible: no
2026-08-26 23:41:35 +03:00
claude[bot]andCodex 1d92193c5d docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 23:41:35 +03:00
claude[bot] dff476ff35 docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 19:35:05 +00:00
Codex ee672dcd1a test: accept the #316 span-over-door baseline
The scene was captured by CI run 33000824647 byte-identical to the locally
reviewed frame: the door keeps its solid jambs inside the former border, the
zero run is dashed on both sides, the leaf swings into the room.

Issue: #316
User-Visible: no
Release: v1.68.0-beta.4
Baseline-Reviewed: https://github.com/Matysh/houseplan-card/actions/runs/33000824647
2026-08-26 22:06:43 +03:00
Codex ffb232398a fix: the initial wall-model migration resolves every opening conflict itself (#316)
Implements spec revision 4 (green r4). §3.1 — a legacy open_spans/open_to cut
never zeroes the atom that carries an existing contour opening: the opening's
edges become atom boundaries, the door keeps its real wall and the zero run
continues on both sides. §3.2 — an ambiguous carrier resolves
deterministically: current host, then distance, thicker cm, smaller id.
§3.3 — an opening with no usable carrier persists unhosted: a valid degraded
v9 state, inert in the physics, rendered by its own x/y, kept by later writes
and re-placeable in the editor (backend schema accepts it). §3.4 — the
initial migration never throws over an opening; a post-v9 write that LOST its
carrier keeps the fail-closed opening-host refusal. The Python migration
mirror implements the same rules with byte-identical output (verified on the
span+door fixture including the segment id).

The new smoke replays #316 end to end: a conflicted space no longer blocks
drawing on an empty plan. The golden scene span-over-door-migrated-dark
renders the migrated fixture pinned byte-for-byte to the real writer; two
gate mutants revert §3.1 and §3.4 and are red by execution.

Issue: #316
User-Visible: yes
2026-08-26 22:06:43 +03:00
claude[bot]andCodex 33a5b46990 docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
Codex f353be3d76 docs: spec #316 revision 4 — the golden expectation rests on the render/write boundary, not on masonry kind
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
claude[bot]andCodex 295548f0fd docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
Codex 0f6583b51c docs: spec #316 revision 3 — correct compatibility reference, honest golden expectation
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
claude[bot]andCodex 5886069c3a docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
Codex 6205c8b2d7 docs: spec #316 revision 2 — registry/compatibility/goldens sections, reachable AC3 fixture
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
claude[bot]andCodex b6de84e528 docs: review document for #316
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
Codex 5b7d37e911 docs: spec #316 — migration auto-resolves opening-host conflicts
Issue: #316
User-Visible: no
2026-08-26 21:54:19 +03:00
claude[bot] f3fe86371c docs: review document for #319
Issue: #319
User-Visible: no
2026-08-26 18:05:38 +00:00
Codex 3ab6fa9f8a fix: the first write of a newer wall model is not an outdated client (#319)
A stale client can only echo the stored model_version, never raise it. The
'unchanged wall catalogue' refusal now applies only when the submitted model
is not above the stored one; the first v9 write over a v8 document with an
orphan open_span/open_to legitimately drops the legacy projection without
touching the catalogue and passes. The regression pair fixture is produced
by the real writers (stored: v1.68.0-beta.2, sent: current migration) and is
pinned on the frontend byte-for-byte so it cannot drift.

Issue: #319
User-Visible: yes
2026-08-26 18:20:15 +03:00
claude[bot] 3d31d65735 docs: review document for #319
Issue: #319
User-Visible: no
2026-08-26 15:11:38 +00:00
Matysh 9ff464fe09 test: accept v1.68.0-beta.3 golden baselines
Accept the complete reviewed Linux matrix after the zero-thickness wall UI and migration changes.

Issue: #306
Release: v1.68.0-beta.3
Baseline-Reviewed: https://github.com/Matysh/houseplan-card/actions/runs/32963513340
User-Visible: no
2026-08-26 14:35:08 +03:00
Matysh 8bfd56eb59 build: prepare v1.68.0-beta.3 candidate
Package the reviewed zero-thickness wall model and legacy virtual-wall migration.

Issue: #306
User-Visible: yes
2026-08-26 14:28:35 +03:00
claude[bot] 26ad7651fb docs: review document for #306
Validate / process-workflow-sync (push) Successful in 1m7s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / provenance (push) Successful in 1m29s
Validate / frontend (push) Skipped
Validate / process-gate (push) Successful in 1m27s
Validate / changes (push) Successful in 1m15s
Validate / docs (push) Failing after 35s
Validate / reuse (push) Successful in 1m10s
Validate / smoke (1) (push) Skipped
Validate / smoke (2) (push) Skipped
Validate / smoke (3) (push) Skipped
Validate / smoke_done (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Issue: #306
User-Visible: no
2026-08-26 11:14:32 +00:00
Matysh 9134d86266 docs: accept issue 306 screenshots
Source: https://github.com/Matysh/houseplan-card/actions/runs/32961287574

Issue: #306
User-Visible: no
2026-08-26 14:06:00 +03:00
Matysh f15bdc678b fix: address zero-wall review findings
Issue: #306
User-Visible: yes
2026-08-26 14:01:58 +03:00
claude[bot] db4e0ef999 docs: review document for #306
Issue: #306
User-Visible: no
2026-08-26 10:56:25 +00:00
Matysh a66fb7ef53 feat: unify zero-thickness walls
Issue: #306
User-Visible: yes
2026-08-26 13:39:52 +03:00
claude[bot]andMatysh 5e35612bdd docs: review document for #306
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
Matysh fa9d9b05e9 docs: align zero-wall spec limits
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
claude[bot]andMatysh 916a42e33a docs: review document for #306
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
Matysh 799289e73b docs: adapt zero-wall spec to model v8
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
claude[bot]andMatysh 73cf8ed01a docs: review document for #306
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
Matysh 1857cb69cf docs: address zero-wall spec review
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
claude[bot]andMatysh 1fcbb0de6d docs: review document for #306
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
Matysh b8a9f9b289 docs: specify zero-thickness wall migration
Issue: #306
User-Visible: no
2026-08-26 12:55:27 +03:00
Codex c821dad436 build: prepare v1.68.0-beta.2 candidate
Package self-explaining preflight refusals, the universal Thickness tool
and reliable model-v8 room drawing.

Issue: #269
Issue: #295
Issue: #313
Issue: #314
User-Visible: yes
2026-08-26 11:20:47 +03:00
claude[bot] 1f6aefd2a7 docs: review document for #295
Issue: #295
User-Visible: no
2026-08-26 08:12:05 +00:00
Codex 7273a3ad9a test: re-pin the #295 baselines to the review-fix source
Pixels are untouched — the r1 fixes change diagnostics data and dialog state,
not rendering. Every scenario hash is byte-identical to the frames CI
captured and I reviewed for run 32940625718; only the source fingerprint
moves to match the fixed tree.

Issue: #295
User-Visible: no
Release: v1.68.0-beta.2
Baseline-Reviewed: https://github.com/Matysh/houseplan-card/actions/runs/32940625718
2026-08-26 10:52:47 +03:00
Codex b542f44bd6 fix: preflight diagnostics hash the judged candidate and the fallback dies with its dialog (#295)
CODE-REVIEW-295-r1, both Medium findings:

M1 — _preflightDiagnostics hashed this._serverCfg, the saved config a space
export would reproduce anyway. The hash now comes from the candidate the
preflight actually judged: the report path passes r.config / d.config and the
copy button reads the dialog's own candidate. The smoke no longer masks the
difference — its dialog carries a candidate whose geometry differs from the
saved config, and swapping the two must change the reported hash.

M2 — the inline clipboard fallback survived dialog close and reopen, so a
later refusal could hand the previous refusal's JSON to a bug report. The
field now dies with its dialog: reset on open (_previewAlignDialog) and on
every close path (escape, mode reset, successful apply, hp-close, cancel).

Both regressions are pinned by execution: reverting either fix turns the
extended smoke red (fingerprintTracksCandidate / fallbackClearedOnClose),
and two new gate mutants keep it that way.

Issue: #295
User-Visible: no
2026-08-26 10:52:47 +03:00
claude[bot] 43f91a214b docs: review document for #295
Issue: #295
User-Visible: no
2026-08-26 07:47:10 +00:00
Codex 822e6223f7 test: accept the #295 preflight-dialog baselines
Both scenes were captured by CI run 32940625718 (byte-identical to run
32939996348), visually reviewed: failure reasons render as rows, the
ghost copy button is borderless per dialogs.styles.

Issue: #295
User-Visible: no
Release: v1.68.0-beta.2
Baseline-Reviewed: https://github.com/Matysh/houseplan-card/actions/runs/32940625718
2026-08-26 10:28:29 +03:00
Codex 1cdd4ed6de fix: a refused geometry preflight names its reason and hands over diagnostics (#295)
Диалог «Оптимизировать» при отказе перечисляет причину по каждому
пространству (7 значений OptimizeGeometryFailureReason получили RU/EN
строки), даёт «Скопировать диагностику» — JSON-блок с origin: runtime,
версией карточки, отпечатками и классами исключений (граница приватности
checkOptimizeGeometry; privacy-тесты дополнены позитивной проверкой) — и
пишет одну структурированную запись в dev-лог (дедупликация по fingerprint).
При недоступном clipboard блок раскрывается прямо в диалоге.

Совет «обновите House Plan» больше не безусловный: websocket
houseplan/config/get теперь возвращает integration_version (бэкенд-тест),
и подсказка показывается только при реальном расхождении с версией карточки;
старый бэкенд без поля — подсказки нет.

Три новых мутанта (потеря причины в диалоге, блок без reason, отключённый
dev-лог) — краснота каждого проверена исполнением; смок
smoke_preflight_diagnostics на dev падает.

Issue: #295
User-Visible: yes
2026-08-26 10:28:29 +03:00
claude[bot]andCodex a5abfe8bd6 docs: review document for #295
Issue: #295
User-Visible: no
2026-08-26 10:28:29 +03:00
Codex 83ec6fb879 docs: spec #295 revision 2 — keep the #199 privacy boundary, real integration-version channel
Issue: #295
User-Visible: no
2026-08-26 10:28:29 +03:00
claude[bot]andCodex 2710dca7d1 docs: review document for #295
Issue: #295
User-Visible: no
2026-08-26 10:28:29 +03:00
Codex 114dcec27c docs: spec for #295 preflight failure diagnostics
Issue: #295
User-Visible: no
2026-08-26 10:28:28 +03:00
Codex 93054823c3 fix: the github-range runner throws so an orphaned BEFORE_SHA can fall back to dev
resolveValidationRange already knows how to replace a force-push-orphaned
BEFORE_SHA with origin/dev, but the CLI handed it a runner that killed the
process with exit 2 on the first cat-file instead of throwing into
gitObjectExists' catch. Every push after a mandatory issue-branch rebase
therefore painted process-gate red (runs 32939996348, 32940625718,
32942113142). The regression test drives the real CLI in a throwaway repo
with a BEFORE_SHA that no longer exists.

Issue: #315
User-Visible: no
2026-08-26 10:28:04 +03:00
claude[bot] 2364d6dd43 docs: review document for #314
Issue: #314
User-Visible: no
2026-08-26 07:09:22 +00:00
Matysh 64e9bf8f6b Cover dense v8 draft invariants
Issue: #314
User-Visible: no
2026-08-26 09:56:35 +03:00
claude[bot] 069c4867ea docs: review document for #314
Issue: #314
User-Visible: no
2026-08-26 06:43:41 +00:00
Matysh ce18012fb9 Finalize issue 314 specification status
Issue: #314
User-Visible: no
2026-08-26 09:20:39 +03:00
Matysh 0e18038c1b Fix documentation screenshot provenance
Issue: #314
User-Visible: no
2026-08-26 09:19:49 +03:00
Matysh 2347e8df88 Fix atomic model-v8 draft writes
Issue: #314
User-Visible: yes
2026-08-26 09:13:50 +03:00
claude[bot]andMatysh 5da7812099 docs: review document for #314
Issue: #314
User-Visible: no
2026-08-26 09:12:08 +03:00
Matysh 7a73bcbc2b docs: specify atomic v8 draft writes
Issue: #314
User-Visible: no
2026-08-26 09:12:08 +03:00
claude[bot] 98a0a24bd4 docs: review document for #269
Issue: #269
User-Visible: no
2026-08-26 06:10:29 +00:00
Codex 676610c0cd fix: align marker.show_entities with the user guide wording (#269)
USER-GUIDE.ru.md — канон формулировок интерфейса (AGENTS.md): флаг диалога
привязки называется «Показывать сущности», как в трёх местах гайда и в
преобладающей форме флагов самого словаря. en.json не меняется.

Issue: #269
User-Visible: yes
2026-08-26 09:04:07 +03:00
Codex 9726f2dfdb docs: spec #266 revision 4 — post-merge numbers per CODE-REVIEW-266-r2
Medium жёлтого r2 (случайно перезапущенное ревью на смерженной задаче):
AC6a приведён к факту — 12 reduced-motion обёрток (10 исходных, две
смешанные разрезаны фиксом каскада 0b782ee по зонам), сохранность обёртки
каждого правила держит scope-ключ styles-diff; AC2 — принятые wc -l числа.

Issue: #266
User-Visible: no
2026-08-26 08:52:42 +03:00
claude[bot] ccdd47a6bd docs: review document for #313
Issue: #313
User-Visible: no
2026-08-26 05:43:07 +00:00
Codex ed126d6a43 fix: keep a stored zero draft thickness and guard both thickness writers (#313)
По находкам CODE-REVIEW-313-r1:

High — резолвер больше не превращает сохранённый 0 сегмента драфта в 15:
ноль — легитимное значение (docs/WALL-THICKNESS.md §6), к дефолту 15 падает
только ОТСУТСТВУЮЩАЯ запись. Смок дополнен: hit нулевого сегмента несёт 0,
диалог показывает пустое поле, Apply без правки отказывает и не портит
данные.

Medium — гвард #278 усилен: паттерн допускает перенос строки после скобки,
и счётчик требует РОВНО ДВЕ точки коммита wall_thickness — мутация любой из
них (включая новую независимую) красит юнит; краснота второго патча мутанта
проверена изолированным исполнением.

Issue: #313
User-Visible: no
2026-08-26 08:35:58 +03:00
claude[bot] d34e56ba78 docs: review document for #313
Issue: #313
User-Visible: no
2026-08-26 05:28:07 +00:00
Codexandclaude[bot] 5e5dad277c fix: the Thickness tool serves standalone walls and saved drafts (#313)
Кандидаты _wallThickHit расширены: интервалы комнат ∪ перегородки ∪ сегменты
сохранённых драфтов (активная цепочка исключена, как в снап-геометрии); при
точном наложении побеждает независимая кладка — она владеет хит-зоной и
рисует видимое тело (решение владельца, согласовано с select и кейсом #308).
Диалог для независимой кладки без кнопки «на всю комнату»; запись — в
partition.cm / draft.segments[i].cm той же физической транзакцией с одним
Undo. Ноль/пусто для независимой кладки отклоняется существующим тостом
диапазона; switch записи — единственный шов, куда #306 повесит ветку
«ноль превращает перегородку в виртуальную стену».

Мутант wall-thickness-writer-bypasses-common-barrier расширен вторым патчем
на новую точку коммита (#278-гвард), краснота обоих проверена исполнением.
Смок smoke_wallthick_standalone: hit/диалог/запись/отказ нуля/приоритет
наложения/hover — на dev падает.

Issue: #313
User-Visible: yes
2026-08-26 05:09:48 +00:00
claude[bot] ffe0f0c788 docs: review document for #313
Issue: #313
User-Visible: no
2026-08-26 05:06:03 +00:00
Matysh a78deb7ce7 build: prepare v1.68.0-beta.1 candidate
Validate / smoke (1) (push) Failing after 25m57s
Validate / smoke (3) (push) Failing after 20m46s
Validate / docs (push) Failing after 40s
Validate / process-workflow-sync (push) Successful in 39s
Validate / golden (push) Failing after 19m23s
Validate / smoke (2) (push) Failing after 34m44s
Validate / changes (push) Successful in 1m2s
Validate / process-gate (push) Failing after 1m5s
Validate / smoke_done (push) Skipped
Validate / hacs (push) Failing after 19s
Validate / hassfest (push) Failing after 18s
Validate / performance_smoke (push) Failing after 21m41s
Validate / provenance (push) Successful in 1m10s
Validate / frontend (push) Successful in 9m51s
Validate / reuse (push) Successful in 56s
Validate / backend (push) Failing after 10m9s
Package stable wall identity, wall junction polish and active-chain visibility.

Issue: #282
Issue: #307
Issue: #309
Issue: #310
User-Visible: yes
2026-08-26 03:27:57 +03:00
Codex 05622d155b fix: merge only the reviewed SHA in the review pipeline (#312)
Validate / docs (push) Successful in 4m20s
Validate / changes (push) Successful in 4m32s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / process-gate (push) Failing after 4m43s
Validate / provenance (push) Successful in 4m48s
Validate / reuse (push) Successful in 1m4s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 2h21m6s
Синхронизация process.yml с dev: шаг слияния сверяет вершину ветки с SHA
материала ревью (допустим ровно один doc-коммит публикации поверх) и при
расхождении отменяет слияние с возвратом в S6-in-progress. Конвейер
исполняется из ветки по умолчанию — правка обязана жить в обеих ветках
(process-workflow-sync).

Issue: #312
User-Visible: no
2026-08-26 03:22:47 +03:00
Codex b4cccfdf63 fix: pin DoR to the commit era and merge only the reviewed SHA (#311, #312)
Правило 10 гейта (#311): DoR сверяется с моментом НАПИСАНИЯ кода — authorDate
коммита класса A не может предшествовать первому labeled-событию S5-ready+
из timeline issue; продвижение метки больше не прячет нарушение, ребейзы
конвейера его не смывают (authorDate переживает их). Проверка вторичная к
правилу 8: недоступный timeline — warn, правило 8 остаётся fail-closed.
LOG_FORMAT несёт authorDate третьим полем (append-совместимо).

Шаг слияния конвейера (#312): сливается только проверенный SHA — вершина
ветки сверяется с материалом ревью (допустим ровно один doc-коммит публикации
с диффом только docs/reviews/ поверх); расхождение отменяет слияние с
возвратом в S6-in-progress тем же путём, что конфликт (инвариант «метка
меняется всегда» сохранён). PROCESS.md §2.7 фиксирует правило «вердикт
привязан к SHA» и для ревьюера.

Issue: #311
Issue: #312
User-Visible: no
2026-08-26 03:15:57 +03:00
Claude 2f9348adf1 ci(process): ask the reviewer about the wall key invariant
Validate / docs (push) Failing after 26s
Validate / provenance (push) Successful in 1m40s
Validate / changes (push) Successful in 1m27s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / process-gate (push) Failing after 1m46s
Validate / reuse (push) Successful in 52s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 1h57m25s
Issue: #259
User-Visible: no
2026-08-23 13:51:31 +03:00
Claude 793a6486d8 ci(process): привести ветку к dev до код-ревью, а не после
Ревью шло по ветке как есть, слияние делало ребейз: проверенный SHA и
слитый SHA были разными коммитами. Текстовое расхождение ловил конфликт,
смысловое git склеивал молча — так пришёл регресс #234. Заодно конфликт
обнаруживался после сорока минут работы ревьюера, хотя виден до них.

Новый шаг для этапа code, сразу после выбора ветки: потомок dev —
ничего; отстала и ребейзится — ребейз, push с --force-with-lease, ревью
приведённого состояния и запись о ребейзе в промпт (§7.2 требует полного
разбора); конфликт — возврат в S6-in-progress без запуска ревью.

Issue: #257
User-Visible: no
2026-08-23 10:55:32 +03:00
Matysh b6c3c5ced3 ci: drop the hardcoded smoke count from the prompt
Issue: #254
User-Visible: no
2026-08-23 09:55:28 +03:00
Matysh d27c3f023e ci: ask the reviewer about numbers seen twice
Issue: #254
User-Visible: no
2026-08-23 09:50:35 +03:00
Matysh 735b98c8ca ci: name the model invariants in the review gate set
Issue: #254
User-Visible: no
2026-08-23 09:41:12 +03:00
Matysh 944a36ee67 ci: make the screenshot verdict compare browsers, not counts
Validate / docs (push) Failing after 26s
Validate / provenance (push) Successful in 1m18s
Validate / process-gate (push) Failing after 1m26s
Validate / changes (push) Successful in 1m3s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / reuse (push) Successful in 55s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 2h2m5s
Issue: #246
User-Visible: no
2026-08-22 19:31:01 +03:00
Matysh 839e3316c3 ci: capture documentation screenshots in one place
Issue: #246
User-Visible: no
2026-08-22 19:18:39 +03:00
Matysh 61c401b810 ci: derive the browser-smoke shortlist from the diff
Issue: #241
User-Visible: no
2026-08-22 18:33:45 +03:00
Matysh e246d32580 ci: name check-docs in the review gate set
Validate / docs (push) Failing after 23s
Validate / provenance (push) Successful in 2m2s
Validate / process-gate (push) Failing after 2m11s
Validate / changes (push) Successful in 2m0s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / reuse (push) Successful in 6m17s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 2h14m0s
Issue: #237
User-Visible: no
2026-08-22 11:44:31 +03:00
Matysh 35bb2b58b7 docs: finish the sentence the reviewer prompt lost
Validate / docs (push) Failing after 39s
Validate / provenance (push) Successful in 1m27s
Validate / process-gate (push) Failing after 1m46s
Validate / changes (push) Successful in 1m11s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / reuse (push) Successful in 1m1s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 11m13s
The previous change replaced the document path with REVIEW_DOC and left a
parenthetical hanging: the prompt jumped from "create no files in the
repository" straight into "(SPEC for the spec stage, CODE for code): scope,
how it was checked…", with the sentence that introduced the document
structure gone. The prompt now says plainly that the publish step derives
the name in docs/reviews, and the content requirements start a paragraph of
their own.

Issue: #220
User-Visible: no
2026-08-21 10:37:54 +03:00
Matysh 5fdedf39d6 fix: keep the review document outside the tree the reviewer mutates
Three code-review rounds on #220 published a verdict and then failed the
run: the document never reached the branch, so the #171 guard refused
before the label step and neither the merge nor S8-merged happened. The
cause was structural. The document lived as an untracked file inside the
very checkout the reviewer edits while proving that a test can fail, and
restoring that tree — git checkout, git clean — deletes an untracked file.
Spec rounds survived only because they never mutate anything.

The reviewer now writes to REVIEW_DOC under RUNNER_TEMP, outside the
repository, and the publish step copies it into docs/reviews before
committing. Tree cleanup can no longer destroy the artefact, and the
reviewer no longer needs to touch docs/reviews at all.

Verified against a local git fixture on five paths: document outside the
repo with a mutated tree (published, and the code mutation does not leak),
nothing anywhere (loud failure, exit 1), document only in the working copy
(still published — the clean exclusion stays for exactly this), document
already committed by the reviewer (recognised, no duplicate), and a branch
that moved during the review (rebased, both commits kept).

Issue: #220
User-Visible: no
2026-08-21 10:32:21 +03:00
Matysh 41325a852f fix: spend the review budget on blocking verdicts only
Validate / docs (push) Failing after 18s
Validate / process-gate (push) Failing after 1m17s
Validate / provenance (push) Successful in 1m21s
Validate / changes (push) Successful in 1m4s
Validate / hacs (push) Skipped
Validate / hassfest (push) Skipped
Validate / frontend (push) Skipped
Validate / reuse (push) Successful in 49s
Validate / smoke (push) Skipped
Validate / golden (push) Skipped
Validate / performance_smoke (push) Skipped
Validate / backend (push) Skipped
Full Performance / performance (push) Failing after 2h9m46s
The pipeline punished what it prescribed: after a failed merge it tells the
author to rebase and restore S7-code-review, and that attempt finished the
budget. On #225 (light track, limit 2) the sequence yellow, green, rebase
produced review-4 on a task whose code review was green and whose CI was
green, with no product change after the verdict — the owner had to
arbitrate work that was already accepted.

A cycle under section 4 is a verdict with blocking findings followed by a
return to the author, so only yellow and red verdicts spend the budget now.
A green verdict returned nothing and consumes nothing, which also removes
any need to mark rebase re-runs specially.

Attempts and cycles are now separate quantities. The attempt number keeps
naming the document, because two runs sharing a number would overwrite each
other's review artefact, while the limit compares blocking cycles only. The
exhaustion comment lists the verdicts it counted, and the guard no longer
strips review-4 — it reports the recount and leaves the decision with the
owner.

Rule 7 of the process gate follows: its document threshold rises above the
cycle limit, because legitimate attempts can exceed cycles and a threshold
equal to the limit would refuse the very rebase the pipeline demands.

Issue: #227
User-Visible: no
2026-08-20 23:17:16 +03:00
349 changed files with 32669 additions and 6021 deletions
+10
View File
@@ -31,6 +31,16 @@ fi
status=0
# Локальный набор гейтов (#343). Выключен по умолчанию намеренно: 20-45 секунд на
# каждый пуш, включая пуши одной строки документации, — цена, которую стоит
# платить осознанно. Документация: docs/TESTING.md.
if [ "${HP_PREPUSH_GATE:-}" = "1" ] && [ -f "$repo_root/scripts/pre-push-gate.mjs" ]; then
echo "pre-push-gate: HP_PREPUSH_GATE=1, прогоняю локальный набор" >&2
if ! node "$repo_root/scripts/pre-push-gate.mjs" >&2; then
status=1
fi
fi
while read -r local_ref local_sha remote_ref remote_sha; do
# Deleting a remote branch pushes nothing to examine.
if [ "$local_sha" = "$zero" ]; then
+2 -1
View File
@@ -1,4 +1,4 @@
name: Announce release
name: Анонс релиза
# Telegram notifications for t.me/ha_houseplan (owner request, 2026-08-07).
# Stable releases are announced; prereleases are deliberately silent.
# workflow_dispatch exists purely as a connectivity test button and therefore
@@ -36,6 +36,7 @@ permissions:
contents: read
jobs:
telegram:
name: Оповещение в Telegram (только стабильные)
if: ${{ github.event_name == 'workflow_dispatch' || (github.event_name == 'release' && github.event.release.prerelease == false) || (github.event_name == 'workflow_call' && inputs.prerelease == false) }}
runs-on: ubuntu-latest
steps:
+44 -6
View File
@@ -9,7 +9,7 @@
# локально через `npm run docs:accept -- --reviewed --from=<распакованный>`.
# Та же конструкция, что у golden-эталонов, и по той же причине: картинки
# попадают в репозиторий через явное решение, а не через бота.
name: Docs screenshots
name: Скриншоты документации
on:
workflow_dispatch:
@@ -24,6 +24,7 @@ permissions:
jobs:
capture:
name: Съёмка и сверка скриншот-индекса
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
@@ -45,6 +46,29 @@ jobs:
- name: Install pinned Chromium
if: steps.pw.outputs.cache-hit != 'true'
run: npx playwright install chromium
# oxipng без потерь снимает с набора ~19% (замер в #345 на pyoxipng 9.1.1;
# у 10.2.0 пресеты уровней перебалансированы, точная доля может отличаться,
# но кадры остаются пиксельно идентичными в любой версии — это перепаковка).
#
# Пин версии и контрольной суммы намеренно, а не `apt-get install oxipng`:
# пакет из образа раннера может пропасть или переехать, а падение шага
# съёмки стоит целого цикла приёмки. Тот же урок, что с azure-зеркалом
# Playwright (#175, #206).
- name: Установить oxipng
env:
OXIPNG_VERSION: 10.2.0
OXIPNG_SHA256: b33f84c73d42cb592bea5d84c431030b1e97784817693380dfcec7d9575f871e
run: |
set -euo pipefail
asset="oxipng-${OXIPNG_VERSION}-x86_64-unknown-linux-gnu.tar.gz"
curl -fsSL -o "$asset" \
"https://github.com/oxipng/oxipng/releases/download/v${OXIPNG_VERSION}/${asset}"
echo "${OXIPNG_SHA256} ${asset}" | sha256sum -c -
mkdir -p "$HOME/.local/bin"
tar -xzf "$asset" --strip-components=1 -C "$HOME/.local/bin" \
"oxipng-${OXIPNG_VERSION}-x86_64-unknown-linux-gnu/oxipng"
echo "$HOME/.local/bin" >> "$GITHUB_PATH"
"$HOME/.local/bin/oxipng" --version
- name: Build the bundle the screenshots must come from
run: npm run build
- name: Capture
@@ -57,16 +81,30 @@ jobs:
# разницу рендеринга.
- name: Вердикт
run: |
# Поле манифеста «до» — из закоммиченного состояния, «после» — из
# свежего. Читается одинаково для браузера и для упаковщика: оба
# переписывают все кадры сразу, и различить их причины обязан вердикт,
# а не человек по памяти (#345).
field() {
node -e "let s='';process.stdin.on('data',d=>s+=d).on('end',()=>{try{console.log(JSON.parse(s).$1||'')}catch{console.log('')}})"
}
git status --porcelain docs/images
changed=$(git diff --name-only docs/images | grep -c png || true)
before=$(git show HEAD:docs/images/screenshots.json | node -e \
"let s='';process.stdin.on('data',d=>s+=d).on('end',()=>{try{console.log(JSON.parse(s).chromium||'')}catch{console.log('')}})")
before=$(git show HEAD:docs/images/screenshots.json | field chromium)
after=$(node -e "console.log(require('./docs/images/screenshots.json').chromium)")
packer_before=$(git show HEAD:docs/images/screenshots.json | field oxipng)
packer_after=$(node -e "console.log(require('./docs/images/screenshots.json').oxipng || '')")
echo "--- изменившихся PNG: $changed"
echo "--- Chromium: было «${before:-не записан}», стало «$after»"
if [ "$before" = "$after" ] && [ "$changed" -gt 0 ]; then
echo "ВЕРДИКТ: тот же браузер, а картинки изменились — изменился продукт."
echo "Смотрите на кадры: если изменение ожидаемое, принимайте."
echo "--- oxipng: было «${packer_before:-не записан}», стало «${packer_after:-нет}»"
if [ "$packer_before" != "$packer_after" ] && [ "$changed" -gt 0 ]; then
echo "ВЕРДИКТ: изменился упаковщик, поэтому переписаны все кадры сразу."
echo "Это перепаковка без потерь: пиксели те же, размер меньше на ~19%."
echo "Ожидаемо один раз — при включении oxipng либо при смене его версии."
echo "Проверить можно сравнением декодированных кадров, а не байтов."
elif [ "$before" = "$after" ] && [ "$changed" -gt 0 ]; then
echo "ВЕРДИКТ: тот же браузер и тот же упаковщик, а картинки изменились —"
echo "изменился продукт. Смотрите на кадры: если изменение ожидаемое, принимайте."
elif [ "$before" != "$after" ]; then
echo "ВЕРДИКТ: браузер другой, поэтому переписаны все кадры сразу."
echo "Это ожидаемо один раз — при переходе на канонический прогон."
+23 -9
View File
@@ -1,15 +1,21 @@
name: Mutation gate
name: Мутационный гейт
# Реестр известных поломок (issue #85): каждый мутант ломает продуктовый код
# известным способом, и объявленный тест ОБЯЗАН на этом покраснеть. Тест,
# оставшийся зелёным на сломанном коде, ничего не защищает — он лишь выглядит
# защитой, и это хуже его отсутствия.
#
# Прогон дорогой: пересборка бандла на каждого мутанта. Поэтому он не входит в
# Validate и не идёт на каждый push. Его место — перед стабильным релизом
# (PROCESS.md §8) и раз в неделю по расписанию, чтобы дрейф тестов не копился
# до релиза. Дешёвая половина — «якоря патчей живы, guard-файлы существуют» —
# идёт с обычными юнитами: test/mutation-gate.test.mjs.
# Прогон дорогой, поэтому он не входит в Validate и не идёт на каждый push.
# Его место — перед стабильным релизом (PROCESS.md §8) и раз в неделю по
# расписанию, чтобы дрейф тестов не копился до релиза. Дешёвая половина —
# «якоря патчей живы, guard-файлы существуют» — идёт с обычными юнитами:
# test/mutation-gate.test.mjs.
#
# #332: бандл собирается только мутантам с браузерным гвардом (guardNeedsBundle),
# компиляция тестов в worktree стартует с тёплого test-build (инкрементальный
# tsc), а реестр режется на четыре чересполосных шарда — полный прогон
# укладывается в десятки минут вместо часов. Локальный дифф-режим:
# node scripts/mutation-gate.mjs --changed origin/dev..HEAD
on:
workflow_dispatch:
@@ -31,9 +37,14 @@ concurrency:
jobs:
mutants:
name: "Мутанты: каждый обязан красить тесты (шард ${{ matrix.shard }} из 4)"
runs-on: ubuntu-latest
# Все мутанты × (сборка + свой guard) — это десятки минут, и это нормально:
# гейт предрелизный. Час — потолок против зависшего Chromium.
strategy:
fail-fast: false
matrix:
shard: [1, 2, 3, 4]
# Шард ~64 мутантов × свой guard; бандл собирают только браузерные гварды.
# Час — потолок против зависшего Chromium.
timeout-minutes: 60
steps:
- uses: actions/checkout@v7
@@ -69,5 +80,8 @@ jobs:
- name: Реестр применим к текущему коду
run: node scripts/mutation-gate.mjs --check
- name: Тёплый test-build для инкрементальной компиляции мутантов
run: npx tsc -p tsconfig.test.json && node scripts/fix-test-build.mjs
- name: Каждый тест ловит свою поломку
run: node scripts/mutation-gate.mjs
run: node scripts/mutation-gate.mjs --shard=${{ matrix.shard }}/4
+80 -29
View File
@@ -1,4 +1,4 @@
name: Full Performance
name: Полные бенчмарки производительности
on:
# Every main promotion is a stable-release candidate and must have an
@@ -24,10 +24,21 @@ concurrency:
jobs:
performance:
# Base and candidate stay sequential on one hosted runner. Splitting them
# across runners would turn machine variance into a false regression.
name: Бенчмарки рендера и геометрии
# Every profile keeps base and candidate sequential on one hosted runner.
# Independent profile pairs run in parallel: cross-profile timing is never
# compared, while serialising all five pairs cannot fit the job timeout.
runs-on: ubuntu-latest
timeout-minutes: 60
strategy:
fail-fast: false
matrix:
profile:
- large-house
- isometric
- plan-snap
- blend
- overlay
steps:
- name: Check out candidate
uses: actions/checkout@v7
@@ -149,38 +160,78 @@ jobs:
npm --prefix candidate run build
(cd candidate && node scripts/bundle-sync.mjs)
npm --prefix baseline run build
(cd baseline && node scripts/bundle-sync.mjs)
- name: Capture base and candidate profiles
working-directory: candidate
run: |
npm run benchmark:large-house -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/baseline.json
npm run benchmark:large-house -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/candidate.json
npm run benchmark:large-house-isometric -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/isometric-baseline.json
npm run benchmark:large-house-isometric -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/isometric-candidate.json
npm run benchmark:large-house-plan-snap -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/plan-snap-baseline.json
npm run benchmark:large-house-plan-snap -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/plan-snap-candidate.json
npm run benchmark:glow -- --profile=large-light-blend-v1 --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/blend-baseline.json
npm run benchmark:glow -- --profile=large-light-blend-v1 --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/blend-candidate.json
npm run benchmark:glow -- --profile=large-house-glow-overlay-v1 --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/overlay-baseline.json
npm run benchmark:glow -- --profile=large-house-glow-overlay-v1 --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/overlay-candidate.json
if ! grep -q "glow_enabled" ../baseline/src/logic.ts; then
echo "Base predates independent Glow; bootstrap relative overlay baseline, keep absolute gate"
cp ../artifacts/performance/overlay-candidate.json ../artifacts/performance/overlay-baseline.json
if [ -f baseline/scripts/bundle-sync.mjs ]; then
(cd baseline && node scripts/bundle-sync.mjs)
else
echo "::notice::The comparison base predates bundle-sync.mjs; materializing its freshly built bundle with the legacy copy path."
mkdir -p baseline/custom_components/houseplan/frontend baseline/demo/srv/assets
cp baseline/dist/houseplan-card.js baseline/custom_components/houseplan/frontend/houseplan-card.js
cp baseline/dist/houseplan-card.js baseline/demo/srv/assets/houseplan-card.js
fi
- name: Enforce relative and absolute performance budgets
- name: Capture base and candidate profile
working-directory: candidate
env:
PROFILE: ${{ matrix.profile }}
run: |
npm run benchmark:compare -- --baseline=../artifacts/performance/baseline.json --candidate=../artifacts/performance/candidate.json --output=../artifacts/performance/comparison.json
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-isometric.json --baseline=../artifacts/performance/isometric-baseline.json --candidate=../artifacts/performance/isometric-candidate.json --output=../artifacts/performance/isometric-comparison.json
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-plan-snap.json --baseline=../artifacts/performance/plan-snap-baseline.json --candidate=../artifacts/performance/plan-snap-candidate.json --output=../artifacts/performance/plan-snap-comparison.json
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-light-blend.json --baseline=../artifacts/performance/blend-baseline.json --candidate=../artifacts/performance/blend-candidate.json --output=../artifacts/performance/blend-comparison.json
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-glow-overlay.json --baseline=../artifacts/performance/overlay-baseline.json --candidate=../artifacts/performance/overlay-candidate.json --output=../artifacts/performance/overlay-comparison.json
mkdir -p ../artifacts/performance
case "$PROFILE" in
large-house)
npm run benchmark:large-house -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/baseline.json
npm run benchmark:large-house -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/candidate.json
;;
isometric)
npm run benchmark:large-house-isometric -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/isometric-baseline.json
npm run benchmark:large-house-isometric -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/isometric-candidate.json
;;
plan-snap)
npm run benchmark:large-house-plan-snap -- --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/plan-snap-baseline.json
npm run benchmark:large-house-plan-snap -- --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/plan-snap-candidate.json
;;
blend)
npm run benchmark:glow -- --profile=large-light-blend-v1 --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/blend-baseline.json
npm run benchmark:glow -- --profile=large-light-blend-v1 --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/blend-candidate.json
;;
overlay)
npm run benchmark:glow -- --profile=large-house-glow-overlay-v1 --target-root=../baseline --samples=7 --warmups=1 --output=../artifacts/performance/overlay-baseline.json
npm run benchmark:glow -- --profile=large-house-glow-overlay-v1 --target-root=. --samples=7 --warmups=1 --output=../artifacts/performance/overlay-candidate.json
if ! grep -q "glow_enabled" ../baseline/src/logic.ts; then
echo "Base predates independent Glow; bootstrap relative overlay baseline, keep absolute gate"
cp ../artifacts/performance/overlay-candidate.json ../artifacts/performance/overlay-baseline.json
fi
;;
*)
echo "::error::Unknown performance profile: $PROFILE"
exit 1
;;
esac
- name: Upload full performance reports
- name: Enforce relative and absolute performance budget
working-directory: candidate
env:
PROFILE: ${{ matrix.profile }}
run: |
case "$PROFILE" in
large-house)
npm run benchmark:compare -- --baseline=../artifacts/performance/baseline.json --candidate=../artifacts/performance/candidate.json --output=../artifacts/performance/comparison.json
;;
isometric)
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-isometric.json --baseline=../artifacts/performance/isometric-baseline.json --candidate=../artifacts/performance/isometric-candidate.json --output=../artifacts/performance/isometric-comparison.json
;;
plan-snap)
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-plan-snap.json --baseline=../artifacts/performance/plan-snap-baseline.json --candidate=../artifacts/performance/plan-snap-candidate.json --output=../artifacts/performance/plan-snap-comparison.json
;;
blend)
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-light-blend.json --baseline=../artifacts/performance/blend-baseline.json --candidate=../artifacts/performance/blend-candidate.json --output=../artifacts/performance/blend-comparison.json
;;
overlay)
npm run benchmark:compare -- --budgets=demo/performance/budgets-large-house-glow-overlay.json --baseline=../artifacts/performance/overlay-baseline.json --candidate=../artifacts/performance/overlay-candidate.json --output=../artifacts/performance/overlay-comparison.json
;;
esac
- name: Upload full performance report
if: always()
uses: actions/upload-artifact@v7
with:
name: full-performance
name: full-performance-${{ matrix.profile }}
path: artifacts/performance
+82 -5
View File
@@ -1,4 +1,4 @@
name: Process
name: Ревью-конвейер
# Событийный конвейер процесса (PROCESS.md). Смена статусной метки — это
# сообщение: она порождает событие, событие запускает следующий шаг.
@@ -35,6 +35,7 @@ permissions:
jobs:
guard:
name: "Страж: ребейз на dev и предпосылки ревью"
runs-on: ubuntu-latest
outputs:
stage: ${{ steps.decide.outputs.stage }}
@@ -162,6 +163,7 @@ jobs:
echo "limit=$limit" >> "$GITHUB_OUTPUT"
review:
name: "Ревью (Claude): вердикт в issue"
needs: guard
if: needs.guard.outputs.stage != ''
runs-on: ubuntu-latest
@@ -287,6 +289,16 @@ jobs:
echo "note=Ветка приведена к dev конвейером до ревью: поверх легло $behind коммит(ов) dev, $short_before -> $short_after. После ребейза это другой код (§7.2) — разбор полный, а не по дельте." >> "$GITHUB_OUTPUT"
echo "ветка $BRANCH приведена к dev: $short_before -> $short_after"
# Материал ревью — конкретный SHA (#312). Вердикт применим только к
# нему: если во время ревью в ветку прилетит коммит, шаг слияния обязан
# это заметить и отказаться, а не молча увезти в dev непроверенный код.
- name: Зафиксировать SHA материала ревью
id: material
if: steps.rebase.outputs.conflict != 'true'
run: |
echo "sha=$(git rev-parse HEAD)" >> "$GITHUB_OUTPUT"
echo "материал ревью: $(git rev-parse --short HEAD)"
# Конфликт возвращает задачу автору ДО ревью. Инвариант «после прогона
# метка меняется всегда» при этом держится: возврат в S6-in-progress —
# тоже смена метки, и автор не ждёт впустую.
@@ -315,6 +327,41 @@ jobs:
--add-label S6-in-progress --remove-label S7-code-review
echo "S7-code-review -> S6-in-progress (ревью не запускалось)"
# Ревьюер перегонял tsc, юниты и сборку заново в каждом раунде, хотя
# Validate на том же SHA уже зелёный (#343). Это не тщательность: бюджет
# ревью тратится на повторение CI вместо чтения кода.
#
# Доказательство здесь такое же строгое, как у reuse-маркеров (#208): не
# «недавно было зелено», а «completed success ровно на этом SHA». После
# ребейза SHA другой, прогона для него нет — и ревьюер честно гоняет сам.
- name: Зелёные гейты на этом SHA
id: validated
if: steps.rebase.outputs.conflict != 'true'
env:
GH_TOKEN: ${{ secrets.HP_PROCESS_TOKEN }}
run: |
sha=$(git rev-parse HEAD)
short=$(git rev-parse --short HEAD)
row=$(gh run list --repo "${{ github.repository }}" --workflow validate.yml \
--commit "$sha" --limit 5 \
--json status,conclusion,url \
--jq '[.[] | select(.status=="completed" and .conclusion=="success")][0] // empty')
{
echo 'note<<EOF_NOTE'
if [ -n "$row" ]; then
url=$(printf '%s' "$row" | node -e 'let s="";process.stdin.on("data",d=>s+=d).on("end",()=>process.stdout.write(JSON.parse(s).url||""))')
echo "**Дешёвые гейты на этом SHA уже подтверждены** (#343). Validate на \`$short\` завершился success: $url"
echo ""
echo "Значит \`npx tsc --noEmit\`, \`npm test\` и \`npm run build\` со сверкой копий бандла перегонять не нужно — сошлись на этом прогоне, назвав его ссылкой. Бюджет раунда тратится на чтение кода."
echo ""
echo "Что Validate НЕ покрывает и остаётся за тобой: смоки, выбранные по диффу; golden, если diff трогает рендер; инварианты модели на конкретной конфигурации; и любой гейт, который требуют AC задачи."
else
echo "**Зелёного Validate на этом SHA (\`$short\`) нет** — прогон не найден, не завершён либо не success. Дешёвые гейты прогоняешь сам и называешь результат."
fi
echo 'EOF_NOTE'
} >> "$GITHUB_OUTPUT"
if [ -n "$row" ]; then echo "Validate на $short: зелёный"; else echo "Validate на $short: зелёного нет"; fi
# Зависимости ставятся ПОСЛЕ переключения на ветку задачи: lockfile мог
# измениться именно в ней, и установка по копии из dev дала бы не то дерево.
- name: Установить зависимости
@@ -441,8 +488,10 @@ jobs:
правке — не тщательность, а потеря времени: полные наборы это
предрелизный гейт (PROCESS.md §8), а не гейт ревью.
Всегда, они дешёвые, и в повторном раунде тоже: код изменился,
а стоят они минуты:
${{ steps.validated.outputs.note }}
Если зелёного прогона на этом SHA нет — прогоняешь сам, они дешёвые,
и в повторном раунде тоже: код изменился, а стоят они минуты:
`npx tsc --noEmit`, `npm test`, `npm run build` со сверкой трёх
копий бандла. Плюс `node scripts/check-docs.mjs`, если diff трогает
`src/**`: отпечаток скриншотов документации считается по всему
@@ -701,14 +750,42 @@ jobs:
GH_TOKEN: ${{ secrets.HP_PROCESS_TOKEN }}
BRANCH: ${{ steps.branch.outputs.name }}
NUM: ${{ github.event.issue.number }}
MATERIAL_SHA: ${{ steps.material.outputs.sha }}
run: |
if [ -z "$BRANCH" ]; then
echo "::error::ветки задачи нет — сливать нечего"
echo "merged=false" >> "$GITHUB_OUTPUT"
exit 0
fi
git fetch -q origin dev
git checkout -q -B merge-into-dev "origin/$BRANCH"
git fetch -q origin dev "$BRANCH"
# #312: сливается только проверенный код. Допустимые вершины ветки:
# сам SHA материала либо он же плюс ровно один коммит публикации
# документа ревью (дифф только docs/reviews/). Любой другой коммит —
# ветка уехала после ревью, вердикт к ней не применим: возврат в
# S6-in-progress через merged=false, как при конфликте.
actual=$(git rev-parse "origin/$BRANCH")
reviewed="$MATERIAL_SHA"
fresh=false
if [ "$actual" = "$reviewed" ]; then
fresh=true
elif [ "$(git rev-parse "$actual^" 2>/dev/null)" = "$reviewed" ] \
&& [ -z "$(git diff --name-only "$reviewed" "$actual" -- . ':!docs/reviews')" ]; then
fresh=true
fi
if [ "$fresh" != true ]; then
echo "merged=false" >> "$GITHUB_OUTPUT"
echo "::warning::ветка $BRANCH уехала после проверенного SHA $reviewed (сейчас $actual) — слияние отменено (#312)"
cat > /tmp/stale-verdict.md <<EOF
**Слияние отменено: ветка изменилась после проверенного материала (#312).**
Ревью выполнялось на \\`$(git rev-parse --short "$reviewed")\\`, а вершина ветки сейчас \\`$(git rev-parse --short "$actual")\\` — в ней есть коммиты, которых вердикт не покрывает. Зелёный вердикт остаётся в силе только для проверенного SHA.
Задача переведена в \\`S6-in-progress\\`. Дальше: убедиться, что вершина ветки — именно то, что должно ехать в dev, и вернуть метку \\`S7-code-review\\` — новый заход ревью проверит актуальный код.
EOF
gh issue comment "$NUM" --repo "${{ github.repository }}" --body-file /tmp/stale-verdict.md
exit 0
fi
git checkout -q -B merge-into-dev "$actual"
if ! git -c user.name="claude[bot]" \
-c user.email="209825114+claude[bot]@users.noreply.github.com" \
rebase origin/dev; then
+5 -1
View File
@@ -1,4 +1,4 @@
name: Publish prerelease
name: Публикация пре-релиза (ручная)
run-name: Publish ${{ inputs.tag }}
on:
@@ -19,6 +19,7 @@ concurrency:
jobs:
gate:
name: "Гейт: зелёная Проверка и релизный контракт"
runs-on: ubuntu-latest
outputs:
sha: ${{ steps.candidate.outputs.sha }}
@@ -61,6 +62,7 @@ jobs:
run: node scripts/release-gate.mjs "$SHA"
publish:
name: Публикация тега и релиза
needs: gate
runs-on: ubuntu-latest
outputs:
@@ -198,6 +200,7 @@ jobs:
# A manual step after a successful release is the worst kind: by the time it is
# due, the work already looks finished, which is exactly why it gets forgotten.
close-merged:
name: Закрытие вошедших issue
needs: [gate, publish]
if: ${{ needs.publish.outputs.newly_published == 'true' }}
runs-on: ubuntu-latest
@@ -245,6 +248,7 @@ jobs:
}
announce:
name: Комментарий о публикации
needs: [gate, publish]
if: ${{ needs.publish.outputs.newly_published == 'true' }}
uses: ./.github/workflows/announce.yml
+2 -1
View File
@@ -1,4 +1,4 @@
name: Attach HACS zip to release
name: HACS-zip к релизу
# hacs.json declares zip_release + filename=houseplan.zip, so every release
# (prereleases included) must carry the asset — HACS installs from it and
# GitHub's public download counter becomes a free per-version install metric
@@ -19,6 +19,7 @@ permissions:
contents: write
jobs:
zip:
name: Собрать houseplan.zip и приложить к релизу
runs-on: ubuntu-latest
steps:
- name: Resolve tag
+5 -2
View File
@@ -1,4 +1,4 @@
name: Release
name: "Релиз: ассеты после зелёной проверки"
on:
release:
types: [published]
@@ -14,6 +14,7 @@ jobs:
# Needs a push with a token that has the `workflow` scope (the ordinary
# Personal Access Token used for `git push` refuses workflow file updates).
gate:
name: "Гейт: зелёная Проверка точного SHA тега"
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
@@ -42,8 +43,9 @@ jobs:
run: |
set -euo pipefail
SHA=$(git rev-parse HEAD)
node scripts/release-gate.mjs "$SHA" --workflow=performance.yml --label="Full Performance"
node scripts/release-gate.mjs "$SHA" --workflow=performance.yml --label="Полные бенчмарки производительности"
build:
name: Сборка бандла и загрузка ассетов
needs: gate
runs-on: ubuntu-latest
steps:
@@ -71,6 +73,7 @@ jobs:
with:
files: dist/houseplan-card.js
hacs-discovery:
name: HACS-видимость пре-релиза (порядок бет)
# HACS 2.0.x takes the first prerelease in GitHub's response instead of
# sorting SemVer. A valid asset can therefore be invisible to beta users
# (beta.10 appeared after beta.9). Keep the release asset, but
+129 -41
View File
@@ -1,4 +1,4 @@
name: Validate
name: Проверка (CI)
on:
push:
@@ -21,25 +21,42 @@ concurrency:
cancel-in-progress: true
jobs:
docs:
# Четыре секундных скрипта жили четырьмя job, каждая со своим checkout и
# setup-node ради нескольких секунд работы (#336). Здесь они шаги одной job.
#
# Независимость при этом сохранена намеренно: у каждого шага
# `continue-on-error`, поэтому упавший провенанс не прячет результат
# процессного гейта, а вердикт в конце падает, если упал хоть один. Прежняя
# запись «job независимый: краснеет сам и не роняет остальные» продолжает
# действовать — на уровне шагов, с той же гранулярностью в логе.
preflight:
name: "Предполётные проверки: документация, провенанс, процесс"
runs-on: ubuntu-latest
steps:
# `blob:none` при полной истории (#345): этой job нужны сообщения
# коммитов, трейлеры и ИМЕНА изменённых файлов, а не содержимое старых
# ревизий. Полный клон стоит 215 МБ против 26 МБ, и 32% этого веса —
# скриншоты документации, переснятые 196 раз.
#
# Коммиты и деревья скачиваются целиком, поэтому диапазоны и `merge-base`
# работают как раньше. Единственная догрузка по требованию здесь —
# `git show origin/main:.github/workflows/process.yml`: один блоб.
- uses: actions/checkout@v7
with: { fetch-depth: 0, filter: 'blob:none' }
- uses: actions/setup-node@v7
with: { node-version: 22 }
- name: Validate public documentation
- name: "Документация: гайды, ченджлоги, скриншот-индекс"
id: docs
continue-on-error: true
run: node scripts/check-docs.mjs --external
# Конвейер читает `process.yml` из ветки по умолчанию, поэтому файл обязан
# совпадать в `main` и `dev`. До этой проверки совпадение держалось на
# дисциплине: каждая правка требовала двух пушей и ручной сверки.
process-workflow-sync:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
with:
fetch-depth: 0
- name: process.yml идентичен в main и dev
# Конвейер читает `process.yml` из ветки по умолчанию, поэтому файл обязан
# совпадать в `main` и `dev`. До этой проверки совпадение держалось на
# дисциплине: каждая правка требовала двух пушей и ручной сверки.
- name: "Процесс: process.yml идентичен в main и dev"
id: workflow_sync
continue-on-error: true
run: |
git fetch --quiet origin main dev
if diff <(git show origin/main:.github/workflows/process.yml) \
@@ -52,14 +69,9 @@ jobs:
exit 1
fi
provenance:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
with: { fetch-depth: 0 }
- uses: actions/setup-node@v7
with: { node-version: 22 }
- name: Validate commit trailers and hook mode
- name: "Провенанс коммитов: трейлеры и эталоны"
id: provenance
continue-on-error: true
env:
EVENT_NAME: ${{ github.event_name }}
BEFORE_SHA: ${{ github.event.before }}
@@ -70,19 +82,14 @@ jobs:
git fetch -q origin "refs/heads/$DEVELOPMENT_BRANCH:refs/remotes/origin/$DEVELOPMENT_BRANCH"
node scripts/validate-commit-provenance.mjs --check-hook-mode --github-range
# Догоняющая проверка процесса (PROCESS.md §10.3). Хуки ловят нарушение на
# машине автора, но их можно обойти `--no-verify`, а коммиты идут прямо в dev
# без PR — GitHub на своей стороне не блокирует ничего. Это последнее место,
# где нарушение правила №1 ловится машиной. Job независимый: краснеет сам и
# не роняет остальные, откат — удалить его отсюда, скрипт остаётся рабочим.
process-gate:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v7
with: { fetch-depth: 0 }
- uses: actions/setup-node@v7
with: { node-version: 22 }
- name: Process gate
# Догоняющая проверка процесса (PROCESS.md §10.3). Хуки ловят нарушение на
# машине автора, но их можно обойти `--no-verify`, а коммиты идут прямо в
# dev без PR — GitHub на своей стороне не блокирует ничего. Это последнее
# место, где нарушение правила №1 ловится машиной. Откат — удалить шаг,
# скрипт остаётся рабочим.
- name: "Процессный гейт: диапазон, трейлеры, статусы issue"
id: process_gate
continue-on-error: true
env:
EVENT_NAME: ${{ github.event_name }}
BEFORE_SHA: ${{ github.event.before }}
@@ -96,6 +103,33 @@ jobs:
git fetch -q origin "refs/heads/$DEVELOPMENT_BRANCH:refs/remotes/origin/$DEVELOPMENT_BRANCH"
node scripts/process-gate.mjs --github-range --issues
# Вердикт обязан быть громким и перечислять всё упавшее сразу: иначе
# слияние job стоило бы автору лишнего круга «починил одно — узнал о
# втором».
- name: Вердикт предполётных проверок
env:
DOCS: ${{ steps.docs.outcome }}
WORKFLOW_SYNC: ${{ steps.workflow_sync.outcome }}
PROVENANCE: ${{ steps.provenance.outcome }}
PROCESS_GATE: ${{ steps.process_gate.outcome }}
run: |
fail=0
check() {
if [ "$2" = "success" ]; then
echo "ok $1"
else
echo "FAIL $1 ($2)"
echo "- **$1**: $2" >> "$GITHUB_STEP_SUMMARY"
fail=1
fi
}
echo "### Предполётные проверки" >> "$GITHUB_STEP_SUMMARY"
check "документация" "$DOCS"
check "process.yml в main и dev" "$WORKFLOW_SYNC"
check "провенанс коммитов" "$PROVENANCE"
check "процессный гейт" "$PROCESS_GATE"
exit $fail
# Классификация изменённых путей: тяжёлые job идут только там, где менялось
# относящееся к ним. НА DEV ФИЛЬТРОВ НЕТ: гейт беты принимает «зелёный Validate
# на точном SHA», и если объём прогона зависит от diff, «зелёный» перестаёт
@@ -104,14 +138,17 @@ jobs:
# экономят на ветках задач, где Validate — ранний сигнал: настоящую приёмку
# там делает код-ревью, которое гоняет гейты само (#127).
changes:
name: Классификация изменённых файлов
runs-on: ubuntu-latest
outputs:
frontend: ${{ steps.classify.outputs.frontend }}
backend: ${{ steps.classify.outputs.backend }}
integration: ${{ steps.classify.outputs.integration }}
steps:
# `git diff --name-only` содержимого файлов не читает вовсе, поэтому
# блобы истории этой job не нужны ни на одном шаге (#345).
- uses: actions/checkout@v7
with: { fetch-depth: 0 }
with: { fetch-depth: 0, filter: 'blob:none' }
- id: classify
env:
EVENT_NAME: ${{ github.event_name }}
@@ -158,6 +195,7 @@ jobs:
# CARD_VERSION и package.json входят в фингерпринт, поэтому ключи кандидата
# заведомо новые и полный набор гейтов перед бетой и релизом идёт всегда.
reuse:
name: "Переиспользование: это дерево уже проверено"
runs-on: ubuntu-latest
outputs:
smoke: ${{ steps.probe.outputs.smoke }}
@@ -237,6 +275,7 @@ jobs:
waive backend "$BACKEND"
hacs:
name: "HACS: валидация репозитория"
needs: changes
if: needs.changes.outputs.integration == 'true'
runs-on: ubuntu-latest
@@ -248,6 +287,7 @@ jobs:
category: integration
hassfest:
name: "Hassfest: манифест интеграции"
needs: changes
if: needs.changes.outputs.integration == 'true'
runs-on: ubuntu-latest
@@ -257,6 +297,7 @@ jobs:
uses: home-assistant/actions/hassfest@master
frontend:
name: "Фронтенд: типы, юниты, мутанты, синхрон бандла"
needs: changes
if: needs.changes.outputs.frontend == 'true'
runs-on: ubuntu-latest
@@ -276,8 +317,23 @@ jobs:
# Копия стенда больше не коммитится (#255): сверяются две обязательные.
- name: Card bundle snapshots in sync
run: cmp dist/houseplan-card.js custom_components/houseplan/frontend/houseplan-card.js
# Бандл собирается один раз и уезжает артефактом (#336). Прежде каждая из
# пяти браузерных job собирала его заново: `bundle:sync` — это `tsc
# --noEmit` плюс rollup, около минуты на job.
#
# Подмену артефакта ловить не нужно отдельной проверкой: `assertFreshDemoBundle`
# сверяет вшитый в бандл отпечаток с `sourceFingerprint` выкачанного
# дерева, и каждая браузерная job делает это перед первым кадром.
- name: Выложить бандл для браузерных job
uses: actions/upload-artifact@v7
with:
name: card-bundle
path: dist/houseplan-card.js
retention-days: 1
if-no-files-found: error
smoke:
name: Смоки в браузере (шард ${{ matrix.shard }} из 3)
# Gated on `frontend` so a typecheck failure does not burn browser minutes.
needs: [frontend, reuse]
if: needs.reuse.outputs.smoke != 'true'
@@ -310,8 +366,16 @@ jobs:
- name: Install pinned Chromium
if: steps.pw.outputs.cache-hit != 'true'
run: npx playwright install chromium
- name: Build a fresh bundle for the smokes
run: npm run bundle:sync
# Бандл приходит артефактом из job `frontend` (#336): та же сборка, но
# один раз на прогон. Свежесть проверяет сам продукт — `assertFreshDemoBundle`
# сверяет вшитый отпечаток с исходниками этого дерева.
- name: Забрать собранный бандл
uses: actions/download-artifact@v7
with:
name: card-bundle
path: dist
- name: Разложить бандл по копиям
run: node scripts/bundle-sync.mjs
- name: Smoke suite (шард ${{ matrix.shard }} из 3)
env:
SHARD: ${{ matrix.shard }}
@@ -350,6 +414,7 @@ jobs:
# Маркер переиспользования пишется ОДИН раз и только когда прошли все шарды:
# частично прогнанная матрица не имеет права выглядеть как выполненная работа.
smoke_done:
name: "Смоки: все шарды зелёные"
needs: [smoke, reuse]
if: needs.reuse.outputs.smoke != 'true'
runs-on: ubuntu-latest
@@ -370,6 +435,7 @@ jobs:
key: reuse-smoke-${{ needs.reuse.outputs.smoke_key }}
golden:
name: Golden-кадры против принятых эталонов
# Deterministic visual correctness stays in every prerelease gate: it is
# inexpensive and catches a different class of regressions than timings.
needs: [frontend, reuse]
@@ -396,8 +462,16 @@ jobs:
- name: Install pinned Chromium
if: steps.pw.outputs.cache-hit != 'true'
run: npx playwright install chromium
- name: Build the exact source under review
run: npm run bundle:sync
# Бандл приходит артефактом из job `frontend` (#336): та же сборка, но
# один раз на прогон. Свежесть проверяет сам продукт — `assertFreshDemoBundle`
# сверяет вшитый отпечаток с исходниками этого дерева.
- name: Забрать собранный бандл
uses: actions/download-artifact@v7
with:
name: card-bundle
path: dist
- name: Разложить бандл по копиям
run: node scripts/bundle-sync.mjs
- name: Capture or verify golden matrix
id: golden
run: |
@@ -432,6 +506,7 @@ jobs:
key: reuse-golden-${{ needs.reuse.outputs.golden_key }}
performance_smoke:
name: "Перф-смок: бюджет времени кадра"
# Candidate-only catastrophic-regression guard for ordinary pushes and
# prereleases. The expensive same-runner comparison lives in performance.yml.
needs: [frontend, reuse]
@@ -462,14 +537,26 @@ jobs:
- name: Install pinned Chromium
if: steps.pw.outputs.cache-hit != 'true'
run: npx playwright install chromium
- name: Build the exact candidate source
run: npm run bundle:sync
# Бандл приходит артефактом из job `frontend` (#336): та же сборка, но
# один раз на прогон. Свежесть проверяет сам продукт — `assertFreshDemoBundle`
# сверяет вшитый отпечаток с исходниками этого дерева.
- name: Забрать собранный бандл
uses: actions/download-artifact@v7
with:
name: card-bundle
path: dist
- name: Разложить бандл по копиям
run: node scripts/bundle-sync.mjs
- name: Capture the heaviest Glow state
run: |
npm run benchmark:glow -- --profile=large-house-glow-overlay-v1 --variants=60 --samples=3 --warmups=1 --output=artifacts/performance-smoke/candidate.json
- name: Enforce absolute smoke ceilings
run: |
npm run benchmark:compare -- --absolute-only --budgets=demo/performance/budgets-glow-smoke.json --candidate=artifacts/performance-smoke/candidate.json --output=artifacts/performance-smoke/comparison.json
# #330 AC7: перф-контракт ограничений стыков — быстрый (без браузера),
# ловит возврат квадратичных путей в оба зеркала прямо на пуше.
- name: Бюджеты ограничений стыков (#330)
run: npm run benchmark:junction-limits
- name: Upload performance smoke report
if: always()
uses: actions/upload-artifact@v7
@@ -494,6 +581,7 @@ jobs:
key: reuse-performance_smoke-${{ needs.reuse.outputs.performance_smoke_key }}
backend:
name: "Бэкенд: pytest в Home Assistant"
needs: [changes, reuse]
if: needs.changes.outputs.backend == 'true' && needs.reuse.outputs.backend != 'true'
runs-on: ubuntu-latest
+13 -5
View File
@@ -40,10 +40,16 @@ task records: problem, scope, acceptance criteria and discussion.
of a status and `rejected` on a closed issue. Exactly one `S*` label per open
issue. Labels are the whole of it: GitHub Projects is no longer used.
Two shortcuts exist for small work. `small` — the light track: the spec lives in
the issue body and its review is a comment. `trivial` — the short track: no spec
stage at all, `S2-analysis` straight to `S5-ready`, with the AC written into the
issue body first. `trivial` requires a bug confined to one surface with no new UX
**The light track is the default, not a shortcut** (owner's decision 2026-08-27,
issue #338). `small` — the spec lives in the issue body and its review is a
comment. Analysis names the `small` criterion the task *fails* when it takes the
full track; "ordinary track" without a named criterion is not a justification.
The threshold itself did not move — only which side carries the proof. The full
track stays what it was for geometry, config migrations and public contracts,
where a criterion is broken plainly and saying which one is easy.
`trivial` — the short track: no spec stage at all, `S2-analysis` straight to
`S5-ready`, with the AC written into the issue body first. `trivial` requires a bug confined to one surface with no new UX
contract, no migration, no i18n, no perf or touch impact, at most three checkable
AC, **and expected behaviour already on record** — nothing left to decide. Code
review is never skipped on either track; it is what stands in for testing.
@@ -274,7 +280,9 @@ the owner splits the task, rejects it, or arbitrates.
On the light track (`small`: complexity ≤3, one surface, no config migration, no
new UX contract, no perf or touch impact — all at once) the spec lives in the issue
body and the spec review is a comment. Code review is never skipped.
body and the spec review is a comment. Code review is never skipped. This track is
the default: taking the full one means naming the criterion above that the task
does not meet.
## Specs
+50 -3
View File
@@ -9,6 +9,35 @@ User-visible changes go into **both** changelogs in the same commit:
`docs/CHANGELOG.md` (English) and `docs/CHANGELOG.ru.md` (Russian). Entries
older than v1.42.0 exist only in the English file — no need to backfill them.
## Translations
A shipped UI language has three matching parts:
1. `src/i18n/<code>.json` for the card;
2. `custom_components/houseplan/translations/<code>.json` for the Home Assistant
integration;
3. one static entry (dictionary import, code and native label) in
`src/i18n/registry.ts`.
Use the canonical Home Assistant/BCP 47 language tag as `<code>` (for example,
`fr` or `pt-BR`) and use that exact spelling for both JSON filenames. Lookup is
case-insensitive and also accepts `_` from legacy locale sources.
The registry drives language resolution, the visual-editor selector and parity
tests. The tests reject missing or extra locale files; frontend dictionaries
also fail on mismatched keys, empty values and changed placeholders.
Placeholders such as `{name}` and `{n}` are a contract: do not translate, add
or remove them.
The current `subst()` helper does not implement plural rules. Phrase strings so
their grammar does not depend on the numeric value (for example, use a neutral
label followed by `{n}` rather than an English singular/plural pair).
Adding a UI locale does not automatically create another full documentation
set; maintain the existing English and Russian documentation according to the
project's normal rules. Right-to-left layout is a separate product project,
because the plan canvas and editors cannot be mirrored by translations alone.
## Where to ask
Not sure whether something is a bug, or just want to discuss an idea before
@@ -28,7 +57,7 @@ issue, but they do not replace it or maintain a separate checklist.
## Five-minute setup
```bash
git clone https://github.com/Matysh/houseplan-card && cd houseplan-card
git clone --filter=blob:none https://github.com/Matysh/houseplan-card && cd houseplan-card
npm ci # frontend toolchain
npm run typecheck # tsc --noEmit (strict)
npm test # node:test — pure logic, i18n parity, tap-action security
@@ -37,6 +66,24 @@ pip install pytest voluptuous && python -m pytest tests_backend -q # pure back
npm install # also installs .githooks through the prepare script
```
### Why `--filter=blob:none` (#345)
A full clone is **215 MB of `.git`**; a blobless one is **26 MB** — measured, not
estimated. Both carry all 1611 commits and all 182 tags, so ranges, `merge-base`
and `git diff` across history work identically; `git diff origin/dev~3..origin/dev`
in a blobless clone takes about a second and grows `.git` by one megabyte.
The difference is that historical *file contents* are fetched only if something
actually asks for them. That matters here because 32% of the pack is documentation
screenshots — ten PNGs re-captured 196 times — and another sizeable share is the
committed bundle, one 1.16 MB file per product change. Almost nobody ever reads an
old revision of either.
Drop the flag if you work offline with history, or need `git log -p` over the whole
tree repeatedly. Do **not** replace it with `--depth=1`: a shallow clone is about
the same size but has no `merge-base`, so the process gate, `smoke-select` and
every `origin/dev..HEAD` range stop working.
The HA-harness backend tests (`tests_backend/test_ha_*.py`) need Python ≥3.13 and
`pytest-homeassistant-custom-component home-assistant-frontend`; CI runs them on
every push — locally they are skipped when `homeassistant` is not importable.
@@ -45,8 +92,8 @@ every push — locally they are skipped when `homeassistant` is not importable.
- **Docs in the same commit**: CHANGELOG entry for user-visible changes;
`docs/STATUS.md` for state changes; `docs/DEVELOPMENT.md` for new gotchas.
- Every UI string goes through `src/i18n/<lang>.json` (tests enforce en/ru key parity).
Adding a language = adding one JSON file + registering it in `src/i18n.ts`.
- Every UI string goes through `src/i18n/<lang>.json`; follow the
[Translations](#translations) flow for registry and backend parity.
- The built card must be committed in sync: `cp dist/houseplan-card.js
custom_components/houseplan/frontend/` (CI compares them byte-for-byte).
- Tap actions have a security model (locks/alarms never toggle from the plan) —
+31 -4
View File
@@ -101,7 +101,9 @@ S1-new → S2-analysis → S3-spec → S4-spec-review ⟲ → S5-ready →
5. приоритет **P1/P2/P3**;
6. тип: баг / фича / техдолг;
7. затронутые поверхности (модули, диалоги, бэкенд, i18n);
8. трек: обычный / `small` / `trivial` по критериям §5 и §5.1.
8. трек — **по умолчанию `small`** (§5). Если задача идёт полным треком,
называется критерий §5, который она не проходит: «обычный трек» без
названного критерия обоснованием не является.
- **Оценки и приоритет ставятся метками сразу, согласие не запрашивается.**
Комментарий аналитики — уведомление, а не запрос: **молчание владельца —
согласие**, несогласие он выражает правкой меток или комментарием, и это не
@@ -194,6 +196,12 @@ S1-new → S2-analysis → S3-spec → S4-spec-review ⟲ → S5-ready →
- **High блокируют.** Medium **в скоупе задачи** чинится в текущем issue:
без High это жёлтый вердикт и возврат автору, фикс проходит повторный цикл.
Medium **вне скоупа** — отдельный issue (#202).
- **Вердикт привязан к SHA (#312).** Все числа и факты отчёта сверяются с
`git rev-parse HEAD` непосредственно перед подведением итогов, а не с SHA,
зафиксированным в начале разбора: во время ревью в ветку может прилететь
fix-up. Серверный стопор — шаг слияния конвейера сверяет вершину ветки с
SHA материала ревью (допустим только собственный doc-коммит публикации
поверх) и при расхождении отменяет слияние с возвратом в `S6-in-progress`.
- **Выход:** очередь на пре-релиз либо возврат в «В разработке», не более
4 циклов (§4). Второй и последующие циклы разбираются по дельте (§2.10).
@@ -281,7 +289,9 @@ S1-new → S2-analysis → S3-spec → S4-spec-review ⟲ → S5-ready →
правка. Блокирующая находка отправляет текущий issue в «Заблокировано».
10. **Каждый коммит класса A и B несёт трейлер `Issue: #NN`**, ветка называется
`issue/NN-slug`, а `User-Visible: yes` требует правок в **обоих** changelog в
том же коммите.
том же коммите. После `cherry-pick -x` служебная строка `(cherry picked
from ...)` должна оставаться выше финального блока трейлеров: перед push
проверяем порядок через `git show -s --format=full HEAD`.
11. **Документация — в том же коммите, что поведение.** Отдельным «допишу потом»
коммитом документация не бывает.
12. **Сгенерированное не коммитится само по себе.** Только релизный промоушен или
@@ -340,9 +350,26 @@ S1-new → S2-analysis → S3-spec → S4-spec-review ⟲ → S5-ready →
---
## 5. Лёгкий трек (метка `small`)
## 5. Лёгкий трек (метка `small`) — путь по умолчанию
**Критерии — все одновременно:**
Умолчание изменено решением владельца 2026-08-27, issue #338. Прежде полный трек
был бесплатен, а выбор лёгкого требовал обоснования. Фактическая цена: **2.9
ревью-документа на задачу** в среднем и до шести на одну issue (#329, #316,
#290) — при том что Medium-находки всё равно чинятся в той же задаче, без
отдельного цикла.
**Порог не изменился.** Критерии ниже те же и по-прежнему обязательны все
одновременно. Изменилась сторона доказательства: теперь обосновывается не выбор
лёгкого трека, а отказ от него — в `S2-analysis` называется критерий, который
задача не проходит. Полный трек остаётся тем, чем был, для геометрии, миграций
конфига и публичных контрактов: там критерии нарушаются сами, и назвать
нарушенный несложно.
Инверсия умолчания не отменяет ничего из §5 ниже и ничего из §4: бюджет четырёх
циклов, арбитраж владельца, обязательность ТЗ на полном треке и правило «ревью
до мержа» остаются как были. Меняется только стоимость пути по умолчанию.
**Критерии — все одновременно; нарушенный называется явно:**
- сложность и риск ≤ 3;
- одна поверхность (один диалог, один модуль, один эндпоинт);
+6 -5
View File
@@ -1,6 +1,6 @@
# 🏠 House Plan — a live home map for Home Assistant
[![HACS Custom](https://img.shields.io/badge/HACS-Custom-41BDF5.svg)](https://github.com/hacs/integration)
[![HACS Default](https://img.shields.io/badge/HACS-Default-41BDF5.svg)](https://github.com/hacs/integration)
[![GitHub release](https://img.shields.io/github/v/release/Matysh/houseplan-card)](https://github.com/Matysh/houseplan-card/releases)
[![CI](https://github.com/Matysh/houseplan-card/actions/workflows/validate.yml/badge.svg)](https://github.com/Matysh/houseplan-card/actions)
[![Live demo](https://img.shields.io/badge/demo-try_it_live-00c853?logo=homeassistant&logoColor=white)](https://demo.houseplan.tech)
@@ -88,10 +88,11 @@ The [Background editor contract](docs/DECOR-EDITOR.md) and
[![Open the repository in HACS](https://my.home-assistant.io/badges/hacs_repository.svg)](https://my.home-assistant.io/redirect/hacs_repository/?owner=Matysh&repository=houseplan-card&category=integration)
1. In HACS open **⋮ → Custom repositories**.
2. Add `https://github.com/Matysh/houseplan-card` as an **Integration**.
3. Install House Plan and restart Home Assistant.
4. Open **Settings → Devices & services → Add integration → House Plan**.
House Plan is in the HACS default catalog — no custom repository needed.
1. In HACS search for **House Plan** and install it.
2. Restart Home Assistant.
3. Open **Settings → Devices & services → Add integration → House Plan**.
The card is registered automatically. If you manage Lovelace resources
manually, use the URL served by the integration:
+7 -5
View File
@@ -1,6 +1,6 @@
# 🏠 House Plan — живой план дома для Home Assistant
[![HACS Custom](https://img.shields.io/badge/HACS-Custom-41BDF5.svg)](https://github.com/hacs/integration)
[![HACS Default](https://img.shields.io/badge/HACS-Default-41BDF5.svg)](https://github.com/hacs/integration)
[![GitHub release](https://img.shields.io/github/v/release/Matysh/houseplan-card)](https://github.com/Matysh/houseplan-card/releases)
[![CI](https://github.com/Matysh/houseplan-card/actions/workflows/validate.yml/badge.svg)](https://github.com/Matysh/houseplan-card/actions)
[![Live demo](https://img.shields.io/badge/демо-попробовать-00c853?logo=homeassistant&logoColor=white)](https://demo.houseplan.tech)
@@ -91,10 +91,12 @@ Assistant — и устройства появятся на плане авто
[![Открыть репозиторий в HACS](https://my.home-assistant.io/badges/hacs_repository.svg)](https://my.home-assistant.io/redirect/hacs_repository/?owner=Matysh&repository=houseplan-card&category=integration)
1. В HACS откройте **⋮ → Пользовательские репозитории**.
2. Добавьте `https://github.com/Matysh/houseplan-card` с типом **Интеграция**.
3. Установите House Plan и перезапустите Home Assistant.
4. Откройте **Настройки → Устройства и службы → Добавить интеграцию → House Plan**.
House Plan входит в основной каталог HACS — пользовательский репозиторий
добавлять не нужно.
1. Найдите **House Plan** в поиске HACS и установите.
2. Перезапустите Home Assistant.
3. Откройте **Настройки → Устройства и службы → Добавить интеграцию → House Plan**.
Карточка регистрируется автоматически. Если ресурсы Lovelace управляются вручную,
добавьте именно URL, который публикует интеграция:
+2 -2
View File
@@ -46,12 +46,12 @@ PLAN_ORPHAN_TTL_S = 3600
SCHEDULED_GRACE_S = 30 * 24 * 3600
FILES_DIR = "houseplan/files"
CONF_ADMIN_ONLY = "admin_only"
VERSION = "1.67.0-rc.3"
VERSION = "1.68.1"
# Portable backup format. This is deliberately independent from the Home
# Assistant Store version above: storage migrations and files exported by a
# user have different compatibility lifecycles.
PLAN_MODEL_VERSION = 8
PLAN_MODEL_VERSION = 9
EXPORT_VERSION = 1
MAX_EXPORT_BYTES = 8 * 1024 * 1024
IMPORT_PREVIEW_TTL_S = 10 * 60
File diff suppressed because one or more lines are too long
@@ -77,6 +77,7 @@ _REPORT_EXAMPLE_LIMIT = 24
_SPACE_PLAN_FIELDS = (
"id", "title", "cell_cm", "plan_url", "plan_aspect", "plan_x", "plan_y",
"plan_scale", "plan_scale_x", "plan_scale_y", "plan_angle", "view_box",
"zero_wall_style",
)
_SPACE_DISPLAY_FIELDS = (
"show_borders", "show_names", "room_color", "bg_color", "room_opacity",
@@ -0,0 +1,428 @@
"""Issue #329 — wall junction limits, Python mirror of src/junction-limits.ts.
The card refuses a WRITE that would ADD a junction violation; this module is
the backend half of that contract, so a stale or hostile client cannot post
what the editor refuses. It mirrors П1–П4 exactly, including the two legal
incidences of П4 (a shared node and a T-joint into the middle of a foreign
wall) and the П3 rule that measures the collinear same-thickness WALL RUN
rather than one catalogue atom.
П5 (the room keeps at least 25 cm² of interior) is deliberately NOT mirrored:
it is a statement about the rendered wall bodies, and reproducing the mitre /
inset pipeline in Python would be a second geometry implementation whose drift
is a larger risk than the rule it guards. A document that violates only П5 is
ugly, not corrupt — see docs/specs/329-junction-limits.md §5.
"""
from __future__ import annotations
import math
import logging
from .wall_segment_model import (
WALL_SEGMENT_MODEL_VERSION, WallSegmentMigrationError, commit_wall_segment_model,
)
_LOGGER = logging.getLogger(__name__)
MIN_JUNCTION_ANGLE_DEG = 15.0
MAX_JUNCTION_VALENCE = 6
MIN_SEGMENT_LENGTH_CM = 20.0
MIN_NODE_DISTANCE_CM = 5.0
GRID_STEP_N = 1 / 240
_EPS = 1e-9
# #331 §2.1: below this two points are ONE node / a node is ON the wall —
# floating debris of pre-canonicalisation arithmetic, not a near miss.
_INCIDENT_EPS = 2e-7
_KEY_FACTOR = 1e7
class JunctionLimitError(ValueError):
"""A write introduced a wall-junction violation (#329)."""
def __init__(self, space_id: str, rule: str, subject: str,
actual: float, limit: float) -> None:
self.space_id = space_id
self.rule = rule
self.subject = subject
self.actual = actual
self.limit = limit
self.code = f"junction_limit_{rule}"
super().__init__(
f"space={space_id}; rule={rule}; subject={subject}; "
f"actual={actual:.12g}; limit={limit:.12g}"
)
def _finite_point(point: object) -> bool:
return (
isinstance(point, (list, tuple))
and len(point) >= 2
and all(isinstance(value, (int, float)) and math.isfinite(value)
for value in point[:2])
)
def _quantize_key_coord(value: float) -> float:
"""#331 §2.1: the canonicalisation formula — sign*floor(|v|*f+0.5)/f.
Native round() is banker's rounding and parts ways with JS Math-style
rounding on .5 ticks; coordinate canonicalisation already encodes this
parity lesson, and node keys must follow it. -0 normalises to 0.
"""
rounded = math.copysign(
math.floor(abs(value) * _KEY_FACTOR + 0.5), value,
) / _KEY_FACTOR
return 0.0 if rounded == 0 else rounded
def _key(point) -> str:
return f"{_quantize_key_coord(point[0])},{_quantize_key_coord(point[1])}"
def _length(a, b) -> float:
return math.hypot(b[0] - a[0], b[1] - a[1])
def cm_to_units(cm: float, cell_cm: float, grid_pitch: float = GRID_STEP_N) -> float:
return (cm / (cell_cm or 1)) * grid_pitch
def limit_segments(space: dict) -> list[dict]:
"""Every wall the limits judge: contour atoms, partitions, draft segments."""
segments: list[dict] = []
for segment in space.get("wall_segments") or []:
if _finite_point(segment.get("a")) and _finite_point(segment.get("b")):
segments.append({
"id": str(segment.get("id") or ""),
"a": segment["a"], "b": segment["b"],
"cm": float(segment.get("cm") or 0),
})
for partition in space.get("partitions") or []:
if _finite_point(partition.get("a")) and _finite_point(partition.get("b")):
segments.append({
"id": str(partition.get("id") or ""),
"a": partition["a"], "b": partition["b"],
"cm": float(partition.get("cm") or 0),
})
for draft in space.get("room_drafts") or []:
points = draft.get("points") or []
drafted = draft.get("segments") or []
for index in range(max(len(points) - 1, 0)):
if not (_finite_point(points[index]) and _finite_point(points[index + 1])):
continue
piece = drafted[index] if index < len(drafted) else {}
segments.append({
"id": str((piece or {}).get("id")
or f"{draft.get('id') or 'draft'}-{index}"),
"a": points[index], "b": points[index + 1],
"cm": float((piece or {}).get("cm") or 0),
})
return [segment for segment in segments
if _length(segment["a"], segment["b"]) > _EPS]
def check_nodes(segments: list[dict]) -> list[tuple[str, str, float, float]]:
"""П1 + П2: valence of a node and the narrowest wedge in it."""
rays: dict[str, list[float]] = {}
for segment in segments:
for start, end in ((segment["a"], segment["b"]), (segment["b"], segment["a"])):
rays.setdefault(_key(start), []).append(
math.atan2(end[1] - start[1], end[0] - start[0])
)
violations: list[tuple[str, str, float, float]] = []
for node, angles in rays.items():
if len(angles) > MAX_JUNCTION_VALENCE:
violations.append(
("valence", node, float(len(angles)), float(MAX_JUNCTION_VALENCE))
)
if len(angles) < 2:
continue
ordered = sorted(angles)
smallest = math.inf
for index, value in enumerate(ordered):
delta = ordered[(index + 1) % len(ordered)] - value
if index == len(ordered) - 1:
delta += math.pi * 2
degrees = (delta * 180) / math.pi
# A ~0° pair is NOT a violation — and cannot be (#331 revision 4,
# learned in the field): a shared wall of two adjacent rooms is
# two co-located owner atoms ON ONE LINE, so every shared-wall
# node carries a legitimate 0° pair by construction. An exact
# duplicate stays a KNOWN LIMITATION of П1 at this level.
if _EPS < degrees < smallest:
smallest = degrees
if smallest < MIN_JUNCTION_ANGLE_DEG - 1e-9:
violations.append(("angle", node, smallest, MIN_JUNCTION_ANGLE_DEG))
return violations
def _axis_degrees(segment: dict) -> float:
degrees = math.degrees(math.atan2(
segment["b"][1] - segment["a"][1], segment["b"][0] - segment["a"][0]
))
return (degrees % 180 + 180) % 180
def _collinear(left: dict, right: dict, tolerance_deg: float = 1.0) -> bool:
delta = abs(_axis_degrees(left) - _axis_degrees(right))
return min(delta, 180 - delta) <= tolerance_deg
def _build_node_index(segments: list[dict]) -> dict[str, list[dict]]:
by_node: dict[str, list[dict]] = {}
for item in segments:
for point in (item["a"], item["b"]):
by_node.setdefault(_key(point), []).append(item)
return by_node
def collinear_run_length_units(
segment: dict, segments: list[dict],
by_node_index: dict[str, list[dict]] | None = None,
) -> float:
"""Length of the WALL a segment belongs to, not of the atom.
#331 §2.3/§2.4: an iterative edge walk over the collinear component — no
recursion (a 10 000-atom chain must answer, not overflow), no
combinatorial DFS (every atom joins at most once, O(E)), no silently
dropped fork. Collinearity is measured against the BASE segment's axis,
so an arc of small per-atom turns cannot pose as one straight wall.
#330 §4.3: the per-check caller builds the node index once.
"""
by_node = by_node_index if by_node_index is not None else _build_node_index(segments)
visited = {id(segment)}
total = _length(segment["a"], segment["b"])
frontier = [segment["a"], segment["b"]]
base_cm = float(segment.get("cm") or 0)
while frontier:
node = frontier.pop()
for candidate in by_node.get(_key(node), ()):
if id(candidate) in visited:
continue
if not _collinear(candidate, segment):
continue
if float(candidate.get("cm") or 0) != base_cm:
continue
visited.add(id(candidate))
total += _length(candidate["a"], candidate["b"])
frontier.append(candidate["a"])
frontier.append(candidate["b"])
return total
def check_segment_lengths(
segments: list[dict], cell_cm: float, grid_pitch: float = GRID_STEP_N,
) -> list[tuple[str, str, float, float]]:
"""П3: a wall is at least 20 cm and never shorter than its own thickness."""
violations: list[tuple[str, str, float, float]] = []
by_node = _build_node_index(segments)
for segment in segments:
units = collinear_run_length_units(segment, segments, by_node)
cm = (units / grid_pitch) * (cell_cm or 1)
thickness = float(segment.get("cm") or 0)
limit = max(MIN_SEGMENT_LENGTH_CM, thickness if thickness > 0 else 0.0)
if cm < limit - 1e-9:
violations.append((
"length", str(segment.get("id") or _key(segment["a"])), cm, limit,
))
return violations
def _distance_to_segment(point, a, b) -> float:
dx, dy = b[0] - a[0], b[1] - a[1]
length_sq = dx * dx + dy * dy
t = 0.0 if length_sq <= _EPS else max(0.0, min(1.0, (
(point[0] - a[0]) * dx + (point[1] - a[1]) * dy
) / length_sq))
return math.hypot(point[0] - (a[0] + dx * t), point[1] - (a[1] + dy * t))
def check_node_distances(
segments: list[dict], cell_cm: float, grid_pitch: float = GRID_STEP_N,
) -> list[tuple[str, str, float, float]]:
"""П4: non-incident nodes and node-to-foreign-wall clearance.
#330 §4.5: the all-pairs form cost 372 ms on 576 atoms. Nodes and padded
segment bounding boxes are hashed into a grid with the threshold as cell
size; each node is compared only against its 9-cell neighbourhood.
Verdicts are identical — pinned by the equivalence unit and the parity
suite. Pair order inside the subject follows the lexicographic key order,
which replaces the i<j of the all-pairs loop.
"""
nodes: dict[str, list] = {}
for segment in segments:
nodes[_key(segment["a"])] = segment["a"]
nodes[_key(segment["b"])] = segment["b"]
min_units = cm_to_units(MIN_NODE_DISTANCE_CM, cell_cm, grid_pitch)
size = min_units if min_units > _EPS else 1.0
node_grid: dict[tuple[int, int], list] = {}
for node_key, point in nodes.items():
node_grid.setdefault(
(int(point[0] // size), int(point[1] // size)), []
).append((node_key, point))
segment_grid: dict[tuple[int, int], list] = {}
for segment in segments:
x0 = min(segment["a"][0], segment["b"][0]) - min_units
x1 = max(segment["a"][0], segment["b"][0]) + min_units
y0 = min(segment["a"][1], segment["b"][1]) - min_units
y1 = max(segment["a"][1], segment["b"][1]) + min_units
for cx in range(int(x0 // size), int(x1 // size) + 1):
for cy in range(int(y0 // size), int(y1 // size) + 1):
segment_grid.setdefault((cx, cy), []).append(segment)
violations: list[tuple[str, str, float, float]] = []
for node_key, point in nodes.items():
cx, cy = int(point[0] // size), int(point[1] // size)
for dx in (-1, 0, 1):
for dy in (-1, 0, 1):
for other_key, other in node_grid.get((cx + dx, cy + dy), ()):
if node_key >= other_key:
continue
distance = _length(point, other)
# #331 §2.1: raw debris within the incidence quantum is ONE
# node on two neighbouring keys — never a near miss.
if distance <= _INCIDENT_EPS:
continue
if distance < min_units - 1e-9:
violations.append((
"distance", f"{node_key} ↔ {other_key}",
(distance / grid_pitch) * (cell_cm or 1),
MIN_NODE_DISTANCE_CM,
))
for segment in segment_grid.get((cx, cy), ()):
# A node that belongs to the wall is a legal corner or T-joint.
if _key(segment["a"]) == node_key or _key(segment["b"]) == node_key:
continue
distance = _distance_to_segment(point, segment["a"], segment["b"])
# Sitting exactly ON the wall is the other legal incidence.
if distance <= _INCIDENT_EPS:
continue
if distance < min_units - 1e-9:
violations.append((
"distance",
f"{node_key} → {segment.get('id') or _key(segment['a'])}",
(distance / grid_pitch) * (cell_cm or 1),
MIN_NODE_DISTANCE_CM,
))
return violations
def space_violations(space: dict) -> list[tuple[str, str, float, float]]:
"""П1–П4 over one space, in the frontend's order."""
segments = limit_segments(space)
cell_cm = float(space.get("cell_cm") or 1)
return [
*check_nodes(segments),
*check_segment_lengths(segments, cell_cm),
*check_node_distances(segments, cell_cm),
]
def _migrated_spaces(config: dict | None, *, side: str = "previous") -> dict[str, dict]:
"""Spaces of one document AFTER the wall-segment migration, by id.
The limits read `wall_segments`, so a document that predates the catalogue
reports NO walls at all — a legacy space would answer "no violations"
regardless of its geometry. Comparing such a baseline against a candidate
the client already migrated counts every inherited violation as new and
refuses an unrelated edit (spec §3 forbids exactly that). Both sides are
therefore judged after the SAME migration, mirroring the frontend barrier.
A document that cannot be migrated is not a reason to refuse the write:
the wall-model barrier owns that verdict and reports it with its own code.
Here it simply means there is no baseline to inherit from.
"""
if not isinstance(config, dict):
return {}
# #330 §4.6: a document that already carries the current catalogue is
# used as-is. The migration's purpose here is to never compare a
# catalogue-less legacy baseline against a migrated candidate; a
# current-version document has the catalogue, so the purpose is met with
# zero work — while a no-op re-migration costs 0.8 s of _atomize on a
# 576-atom plan. Equivalence "as-is == migrated" for current-version
# documents is pinned by a parity case.
if int(config.get("model_version") or 0) >= WALL_SEGMENT_MODEL_VERSION:
migrated = config
else:
try:
migrated, _ = commit_wall_segment_model(config)
except (WallSegmentMigrationError, ValueError) as error:
# The wall-model barrier owns this verdict and reports it with its
# own code; here it merely means "no catalogue to read".
_LOGGER.debug("junction limits: %s migration fell back: %s", side, error)
migrated = config
except Exception:
# #331 §2.6 (AC6): a genuine migration bug on the CANDIDATE side
# must surface as an honest WS error, not a silent "no
# violations". The PREVIOUS side keeps the wide fallback — an
# unprovable inheritance is no reason to block an unrelated write
# (the exact P1 symptom this task fixes).
if side == "candidate":
raise
_LOGGER.debug("junction limits: previous-side migration bug swallowed by design")
migrated = config
return {
str(space.get("id")): space
for space in (migrated or {}).get("spaces") or []
if isinstance(space, dict)
}
def space_violation_counts(spaces: dict[str, dict]) -> dict[str, dict[str, int]]:
"""Violation counts per space per rule — the shape the barrier consumes
and the rev cache (#330 §4.2) stores. Documents are NOT retained."""
result: dict[str, dict[str, int]] = {}
for space_id, space in spaces.items():
counts: dict[str, int] = {}
for rule, _subject, _actual, _limit in space_violations(space):
counts[rule] = counts.get(rule, 0) + 1
result[space_id] = counts
return result
def validate_junction_limits(
config: dict, previous: dict | None = None, *,
baseline_counts: dict[str, dict[str, int]] | None = None,
) -> dict[str, dict[str, int]]:
"""Refuse a write that ADDS a junction violation; inherit the rest.
Counted per RULE, not per subject: a structural write re-atomises and
re-keys contour segments, so subject identity is not stable across the
barrier and matching by it would report an inherited violation as new
(the mistake that refused legitimate resizes on the frontend). Spec §3:
a write may keep existing violations, it may never add one.
A LEGACY document still goes through `commit_wall_segment_model` first,
so a catalogue-less baseline is compared in the same terms as the
candidate; a current-version document is used as-is (#330 §4.6).
`baseline_counts` (#330 §4.2) replaces the previous document entirely:
the caller caches the counts of the stored document by config_rev, so a
repeated write does not even touch `previous`. Returns the candidate's
counts — the value to cache after a successful write.
"""
if baseline_counts is not None:
old_counts = baseline_counts
else:
old_counts = space_violation_counts(_migrated_spaces(previous))
new_spaces = _migrated_spaces(config, side="candidate")
candidate_counts: dict[str, dict[str, int]] = {}
for space_id, space in new_spaces.items():
# A brand-new space has nothing to inherit from — but neither is a
# first write allowed to arrive already broken.
before = old_counts.get(space_id) or {}
after: dict[str, list[tuple[str, str, float, float]]] = {}
for violation in space_violations(space):
after.setdefault(violation[0], []).append(violation)
candidate_counts[space_id] = {
rule: len(items) for rule, items in after.items()
}
for rule, items in after.items():
if len(items) > before.get(rule, 0):
_, subject, actual, limit = items[0]
raise JunctionLimitError(space_id, rule, subject, actual, limit)
return candidate_counts
+1 -1
View File
@@ -16,5 +16,5 @@
"issue_tracker": "https://github.com/Matysh/houseplan-card/issues",
"requirements": [],
"single_config_entry": true,
"version": "1.67.0-rc.3"
"version": "1.68.1"
}
+5
View File
@@ -101,6 +101,11 @@ class HouseplanData:
# Parsed import candidates are short-lived, user-bound and memory-only.
# dict keeps insertion order, which lets the preview service evict oldest.
import_previews: dict[str, dict[str, Any]] = field(default_factory=dict)
# #330 §4.2: junction-limit violation counts of the STORED document,
# keyed by its rev — (rev, {space_id: {rule: count}}). One slot,
# memory-only, invalidated by a rev mismatch; the previous document is
# not re-judged on every write. Counts only — documents are never held.
junction_baseline: tuple[int, dict[str, dict[str, int]]] | None = None
HouseplanConfigEntry = ConfigEntry[HouseplanData]
+43 -37
View File
@@ -102,8 +102,13 @@ def _legacy_wall_model_projection(config: dict) -> dict:
return projected
def _legacy_wall_geometry_projection(config: dict) -> list[dict]:
"""Only geometry whose edit requires a matching v8 identity update."""
def _catalog_coupled_wall_geometry_projection(config: dict) -> list[dict]:
"""Legacy geometry whose edit must also update the v8 contour catalogue.
Drafts, partitions and columns carry their own ids. Openings carry an
explicit v8 host and are checked against it by CONFIG_SCHEMA. None of those
independent objects requires a different contour wall_segments catalogue.
"""
result: list[dict] = []
for space in config.get("spaces") or []:
rooms = []
@@ -111,28 +116,11 @@ def _legacy_wall_geometry_projection(config: dict) -> list[dict]:
rooms.append({key: copy.deepcopy(room.get(key)) for key in (
"id", "poly", "x", "y", "w", "h", "open_to"
) if key in room})
openings = []
for opening in space.get("openings") or []:
openings.append({key: copy.deepcopy(opening.get(key)) for key in (
"id", "type", "x", "y", "angle", "length"
) if key in opening})
drafts = []
for draft in space.get("room_drafts") or []:
drafts.append({
"id": draft.get("id"),
"points": copy.deepcopy(draft.get("points")),
"segments": [
{"cm": segment.get("cm")} for segment in draft.get("segments") or []
],
})
result.append({
"id": space.get("id"),
"rooms": rooms,
"walls": copy.deepcopy(space.get("walls") or []),
"open_spans": copy.deepcopy(space.get("open_spans") or []),
"room_drafts": drafts,
"partitions": copy.deepcopy(space.get("partitions") or []),
"openings": openings,
})
return result
@@ -186,12 +174,12 @@ def validate_wall_model_transition(config: dict, previous: dict | None) -> None:
except (TypeError, ValueError):
return # CONFIG_SCHEMA owns malformed values.
previous = previous or {}
geometry_changed = (
_legacy_wall_geometry_projection(config)
!= _legacy_wall_geometry_projection(previous)
contour_geometry_changed = (
_catalog_coupled_wall_geometry_projection(config)
!= _catalog_coupled_wall_geometry_projection(previous)
)
if old_model >= 8 and new_model < 8:
if not geometry_changed:
if not contour_geometry_changed:
_restore_wall_model_fields(config, previous or {})
# Re-run semantic parity after hydration; previous identity is
# accepted only when it still matches the submitted projections.
@@ -202,12 +190,19 @@ def validate_wall_model_transition(config: dict, previous: dict | None) -> None:
raise WallModelClientOutdatedError(
f"stored model={old_model}; submitted model={new_model}"
)
if old_model >= 8 and new_model >= 8 and geometry_changed:
if old_model >= 8 and 8 <= new_model <= old_model and contour_geometry_changed:
# The realistic stale-client case echoes model_version and the unknown
# catalogue verbatim while changing rooms/walls/openings. Let the
# catalogue verbatim while changing room/contour geometry. Let the
# frontend show the dedicated reload guidance instead of a generic
# schema error. A current writer necessarily changes the catalogue in
# the same transaction.
# schema error. Independent drafts, partitions, columns and explicitly
# hosted openings are validated by CONFIG_SCHEMA without requiring a
# contour catalogue change (#314).
#
# A submitted model ABOVE the stored one is excluded on purpose (#319):
# a stale client can only echo the stored version, never raise it. The
# first write of a newer client legitimately drops legacy projection
# fields (an orphan open_span/open_to) without touching the catalogue;
# treating that as "outdated" wedged every structural write forever.
if _wall_catalog_projection(config) == _wall_catalog_projection(previous):
raise WallModelClientOutdatedError(
f"stored model={old_model}; unchanged wall catalogue"
@@ -1433,7 +1428,7 @@ ROOM_DRAFT_SCHEMA = vol.All(
vol.Required("segments"): vol.All(
[vol.Schema({
vol.Optional("id"): vol.All(str, vol.Length(min=1, max=64)),
vol.Required("cm"): vol.All(_finite, vol.Range(min=1, max=100)),
vol.Required("cm"): vol.All(_finite, vol.Range(min=0, max=100)),
},
extra=vol.ALLOW_EXTRA)],
vol.Length(min=1, max=499),
@@ -1456,7 +1451,7 @@ PARTITION_SCHEMA = vol.All(
vol.Required("id"): vol.All(str, vol.Length(min=1, max=64)),
vol.Required("a"): POINT,
vol.Required("b"): POINT,
vol.Required("cm"): vol.All(_finite, vol.Range(min=1, max=100)),
vol.Required("cm"): vol.All(_finite, vol.Range(min=0, max=100)),
},
extra=vol.ALLOW_EXTRA,
),
@@ -1636,6 +1631,7 @@ SPACE_SCHEMA = vol.All(vol.Schema(
vol.Optional("wall_segments"): vol.All(
[WALL_SEGMENT_SCHEMA], vol.Length(max=MAX_WALL_SEGMENTS)
),
vol.Optional("zero_wall_style"): vol.In(["dashed", "solid"]),
vol.Optional("room_drafts"): vol.All(
[ROOM_DRAFT_SCHEMA], vol.Length(max=MAX_ROOM_DRAFTS)
),
@@ -1774,7 +1770,7 @@ def _canonical_segment_key(a: list, b: list) -> tuple:
def _config_wall_segment_invariants(value: dict) -> dict:
"""Fail closed when a v8 writer sends stale ids or a stale walls projection."""
"""Fail closed when a v8+ writer sends stale ids or wall projections."""
try:
model = int(value.get("model_version", 0))
except (TypeError, ValueError):
@@ -1783,9 +1779,12 @@ def _config_wall_segment_invariants(value: dict) -> dict:
return value
for space in value.get("spaces", []):
if model >= 9:
if "open_spans" in space or any("open_to" in room for room in space.get("rooms", [])):
raise vol.Invalid("v9 config must not contain legacy open boundaries")
segments = space.get("wall_segments")
if segments is None:
raise vol.Invalid("v8 space requires wall_segments")
raise vol.Invalid("v8+ space requires wall_segments")
by_id = {segment["id"]: segment for segment in segments}
if len(by_id) != len(segments):
raise vol.Invalid("wall segment ids must be unique")
@@ -1795,7 +1794,7 @@ def _config_wall_segment_invariants(value: dict) -> dict:
poly = room.get("poly")
wall_ids = room.get("wall_ids")
if not poly or not wall_ids or len(poly) != len(wall_ids):
raise vol.Invalid("v8 room wall_ids must match poly edges")
raise vol.Invalid("v8+ room wall_ids must match poly edges")
room_id = room["id"]
for index, segment_id in enumerate(wall_ids):
segment = by_id.get(segment_id)
@@ -1815,31 +1814,38 @@ def _config_wall_segment_invariants(value: dict) -> dict:
actual_walls: dict[tuple, float] = {}
for wall in space.get("walls", []):
if "a" not in wall or "b" not in wall:
raise vol.Invalid("v8 compatibility walls require exact endpoints")
raise vol.Invalid("wall compatibility projection requires exact endpoints")
key = _canonical_segment_key(wall["a"], wall["b"])
if key in actual_walls:
raise vol.Invalid("v8 compatibility walls must be unique")
raise vol.Invalid("wall compatibility projection must be unique")
actual_walls[key] = float(wall["cm"])
if actual_walls != expected_walls:
raise vol.Invalid("v8 compatibility walls must match wall_segments")
raise vol.Invalid("wall compatibility projection must match wall_segments")
for draft in space.get("room_drafts", []):
if any(not segment.get("id") for segment in draft.get("segments", [])):
raise vol.Invalid("v8 draft wall segments require ids")
raise vol.Invalid("v8+ draft wall segments require ids")
partitions = {item["id"]: item for item in space.get("partitions", [])}
wall_opening_intervals: dict[str, list[tuple[float, float]]] = {}
for opening in space.get("openings", []):
host = opening.get("host")
if host is None:
raise vol.Invalid("v8 opening requires an explicit host")
# #316 §3.3: a degraded migration may leave a contour opening
# unhosted. It is inert (no body/tunnel/cut) and excluded from
# the interval checks; a later edit can re-place it.
continue
if host["kind"] == "partition":
if host["id"] not in partitions:
raise vol.Invalid("partition opening host must exist in the same space")
if float(partitions[host["id"]]["cm"]) <= 0:
raise vol.Invalid("opening host must have positive thickness")
continue
segment = by_id.get(host["id"])
if segment is None:
raise vol.Invalid("wall opening host must exist in the same space")
if float(segment["cm"]) <= 0:
raise vol.Invalid("opening host must have positive thickness")
t = float(host["t"])
x = float(segment["a"][0]) + (float(segment["b"][0]) - float(segment["a"][0])) * t
y = float(segment["a"][1]) + (float(segment["b"][1]) - float(segment["a"][1])) * t
+152 -12
View File
@@ -1,4 +1,4 @@
"""Deterministic persisted contour-wall identity (model v8, issue #282).
"""Deterministic persisted contour-wall identity (model v9, issues #282/#306).
This is the backend twin of ``src/wall-segment-model.ts``. Import preview is
a server-side structural writer, so it cannot depend on a browser being open
@@ -16,7 +16,7 @@ from typing import Any
from .coordinate_canonicalization import canonicalize_config_geometry
WALL_SEGMENT_MODEL_VERSION = 8
WALL_SEGMENT_MODEL_VERSION = 9
GRID_STEP_N = 1 / 240
EPS = 1e-9
@@ -182,9 +182,96 @@ def _atomize(
rooms = space.get("rooms") or []
room_polys = {str(room.get("id", "")): _room_poly(room) for room in rooms}
global_breaks: list[list[float]] = []
for span in space.get("open_spans") or []:
explicit_spans = space.get("open_spans") or []
legacy_segments: list[list[float]] = []
for span in explicit_spans:
if isinstance(span, dict) and isinstance(span.get("a"), list) and isinstance(span.get("b"), list):
global_breaks.extend((span["a"], span["b"]))
legacy_segments.append([
float(span["a"][0]), float(span["a"][1]),
float(span["b"][0]), float(span["b"][1]),
])
if not legacy_segments:
linked = lambda first, second: (
str(second.get("id", "")) in (first.get("open_to") or [])
or str(first.get("id", "")) in (second.get("open_to") or [])
)
for first_index, first in enumerate(rooms):
for second in rooms[first_index + 1:]:
if not linked(first, second):
continue
for shared in _shared_boundaries(
room_polys.get(str(first.get("id", "")), []),
room_polys.get(str(second.get("id", "")), []),
):
legacy_segments.append(shared)
global_breaks.extend(([shared[0], shared[1]], [shared[2], shared[3]]))
canonical_zero_segments: list[list[float]] = []
for room in rooms:
poly = room_polys.get(str(room.get("id", ""))) or []
wall_ids = room.get("wall_ids") or []
if len(wall_ids) != len(poly):
continue
for index, segment_id in enumerate(wall_ids):
previous = old.get(str(segment_id)) if isinstance(segment_id, str) else None
if previous is None or float(previous.get("cm", 0)) != 0:
continue
following = (index + 1) % len(poly)
canonical_zero_segments.append([
float(poly[index][0]), float(poly[index][1]),
float(poly[following][0]), float(poly[following][1]),
])
for segment in canonical_zero_segments:
global_breaks.extend((segment[:2], segment[2:]))
zero_segments = canonical_zero_segments + legacy_segments
# #316 §3.1: a legacy open_spans/open_to cut never zeroes the atom that
# carries an existing contour opening; its edges become atom boundaries so
# the zero run continues on both sides of the opening.
legacy_era_openings: list[dict[str, Any]] = []
if legacy_segments:
for opening in space.get("openings") or []:
if not isinstance(opening, dict):
continue
host = opening.get("host")
if isinstance(host, dict) and host.get("kind") == "partition":
continue
try:
centre = [float(opening["x"]), float(opening["y"])]
angle = float(opening["angle"])
half = float(opening["length"]) / 2
except (KeyError, TypeError, ValueError):
continue
if half <= 0:
continue
# Only an opening that actually stands on a legacy cut changes the
# atomization; unrelated openings must not churn the catalogue.
if not any(
_distance_to_segment(centre, cut[:2], cut[2:]) <= GRID_STEP_N * 0.04
for cut in legacy_segments
):
continue
legacy_era_openings.append(
{"centre": centre, "angle": angle, "half": half}
)
direction = (math.cos(math.radians(angle)), math.sin(math.radians(angle)))
global_breaks.extend((
[centre[0] - direction[0] * half, centre[1] - direction[1] * half],
[centre[0] + direction[0] * half, centre[1] + direction[1] * half],
))
def _atom_carries_opening(a: list[float], b: list[float]) -> bool:
for entry in legacy_era_openings:
if not _angle_matches(a, b, entry["angle"]):
continue
if _distance_to_segment(entry["centre"], a, b) > GRID_STEP_N * 0.02:
continue
span = _length(a, b)
tc = _project_t(entry["centre"], a, b) * span
if tc + entry["half"] >= -EPS and tc - entry["half"] <= span + EPS:
return True
return False
for wall in space.get("walls") or []:
if isinstance(wall, dict) and isinstance(wall.get("a"), list) and isinstance(wall.get("b"), list):
global_breaks.extend((wall["a"], wall["b"]))
@@ -253,13 +340,23 @@ def _atomize(
"key": key, "a": _canonical_span(a, b)[0], "b": _canonical_span(a, b)[1],
"owners": set(), "preferred": set(), "positional": set(),
"preferred_carriers": {},
"parent_keys": set(),
"parent_keys": set(), "zero_wall": False,
})
atom["owners"].add(room_id)
if len(atom["owners"]) > 2:
raise WallSegmentMigrationError("third-owner", key)
parent_index = parents[index]
atom["parent_keys"].add(_wall_key(original[parent_index], original[(parent_index + 1) % len(original)]))
midpoint = [(a[0] + b[0]) / 2, (a[1] + b[1]) / 2]
covered_by = lambda cuts: any(
_distance_to_segment(midpoint, cut[:2], cut[2:]) <= GRID_STEP_N * 0.04
for cut in cuts
)
# Canonical cm:0 atoms stay zero; a LEGACY cut spares the atom
# that carries an opening (#316 §3.1).
atom["zero_wall"] = atom["zero_wall"] or covered_by(canonical_zero_segments) or (
covered_by(legacy_segments) and not _atom_carries_opening(a, b)
)
if indexed_lineage and isinstance(old_ids[parent_index], str) and old_ids[parent_index]:
previous = old.get(old_ids[parent_index])
if rigid_indexed_lineage or previous is None or _collinear_overlap(
@@ -281,6 +378,8 @@ def _atomize(
def _thickness(space: dict[str, Any], atom: dict[str, Any], previous: dict[str, Any] | None) -> float:
if atom.get("zero_wall"):
return 0.0
candidates: list[float] = []
query_key = _wall_key(atom["a"], atom["b"])
query_length = _length(atom["a"], atom["b"])
@@ -431,7 +530,9 @@ def _angle_matches(a: list[float], b: list[float], angle: float) -> bool:
return difference <= 8
def _host_openings(space: dict[str, Any], segments: list[dict[str, Any]]) -> None:
def _host_openings(
space: dict[str, Any], segments: list[dict[str, Any]], initial_migration: bool,
) -> None:
for opening in space.get("openings") or []:
host = opening.get("host")
if isinstance(host, dict) and host.get("kind") == "partition":
@@ -440,9 +541,15 @@ def _host_openings(space: dict[str, Any], segments: list[dict[str, Any]]) -> Non
centre = [float(opening["x"]), float(opening["y"])]
angle, half = float(opening["angle"]), float(opening["length"]) / 2
except (KeyError, TypeError, ValueError):
if initial_migration:
# #316 §3.4: the initial migration never throws over an opening.
opening.pop("host", None)
continue
raise WallSegmentMigrationError("opening-host", str(opening.get("id", ""))) from None
def eligible(segment: dict[str, Any]) -> bool:
if float(segment.get("cm", 0)) <= 0:
return False
t = _project_t(centre, segment["a"], segment["b"])
span = _length(segment["a"], segment["b"])
return (-EPS <= t <= 1 + EPS
@@ -452,19 +559,49 @@ def _host_openings(space: dict[str, Any], segments: list[dict[str, Any]]) -> Non
and half >= 0 and t * span - half >= -EPS
and t * span + half <= span + EPS)
def materialize(carrier: dict[str, Any]) -> None:
opening["host"] = {
"kind": "wall", "id": carrier["id"],
"t": max(0.0, min(1.0, _project_t(centre, carrier["a"], carrier["b"]))),
}
current = None
if isinstance(host, dict) and host.get("kind") == "wall":
current = next((segment for segment in segments if segment["id"] == host.get("id")), None)
candidates = [current] if current is not None and eligible(current) else [
segment for segment in segments if eligible(segment)
]
if len(candidates) != 1:
if len(candidates) == 1:
materialize(candidates[0])
continue
# #316 §3.3: an unhosted opening is a valid degraded v9 state. A later
# write keeps it, may self-heal it, and never fails over it.
if host is None and not initial_migration:
continue
if not initial_migration:
raise WallSegmentMigrationError("opening-host", str(opening.get("id", "")))
carrier = candidates[0]
opening["host"] = {
"kind": "wall", "id": carrier["id"],
"t": max(0.0, min(1.0, _project_t(centre, carrier["a"], carrier["b"]))),
}
def pick(pool: list[dict[str, Any]]) -> dict[str, Any] | None:
if not pool:
return None
if current is not None and any(item is current for item in pool):
return current
return sorted(pool, key=lambda segment: (
_distance_to_segment(centre, segment["a"], segment["b"]),
-float(segment.get("cm", 0)),
str(segment.get("id", "")),
))[0]
# #316 §3.2 tie-break. No distant fallback pool (CODE-REVIEW-316-r1
# H1): a host away from the opening's own x/y would violate the
# geometry-match invariant of CONFIG_SCHEMA and wedge the write on the
# schema layer; without an in-place carrier the opening goes straight
# to the unhosted degraded state.
carrier = pick([segment for segment in segments if eligible(segment)])
if carrier is not None:
materialize(carrier)
else:
opening.pop("host", None)
def _migrate_space(space: dict[str, Any], initial_migration: bool) -> int:
@@ -498,6 +635,9 @@ def _migrate_space(space: dict[str, Any], initial_migration: bool) -> int:
space["walls"] = walls
else:
space.pop("walls", None)
space.pop("open_spans", None)
for room in space.get("rooms") or []:
room.pop("open_to", None)
draft_used = {
str(item["id"])
for name in ("rooms", "openings", "decor", "room_drafts", "partitions",
@@ -527,7 +667,7 @@ def _migrate_space(space: dict[str, Any], initial_migration: bool) -> int:
else:
segment["id"] = _fresh_wall_segment_id(draft_used)
draft_used.add(segment["id"])
_host_openings(space, segments)
_host_openings(space, segments, initial_migration)
return sum(1 for segment in segments if segment["id"] not in old)
+50 -17
View File
@@ -24,6 +24,7 @@ from .const import (
CONTENT_URL, FILES_DIR, MAX_PLANS_BYTES, MAX_PLANS_FILES, MAX_PLANS_LISTED,
MAX_SIGN_PATHS,
PLANS_DIR, PLANS_URL,
VERSION,
)
from .coordinate_canonicalization import (
canonicalize_config_geometry,
@@ -65,6 +66,7 @@ from .wall_segment_model import (
WallSegmentMigrationError,
commit_wall_segment_model,
)
from .junction_limits import JunctionLimitError, validate_junction_limits
from .validation import (
CONFIG_SCHEMA, LAYOUT_SCHEMA, MAX_CONFIG_BYTES, MAX_PLAN_BYTES,
PLAN_EXTENSIONS, POS_SCHEMA, MarkerControlError, OpeningPassageError,
@@ -1157,6 +1159,9 @@ async def ws_config_get(hass: HomeAssistant, connection, msg: dict[str, Any]) ->
"can_write": may_write(hass, getattr(connection, "user", None)),
"can_optimize_undo": _optimizer_backup_is_current(data, layout_data),
"undo_kind": _undo_kind(data, layout_data),
# #295: the card compares this against its own version to decide
# whether the «update House Plan» preflight hint can actually help.
"integration_version": VERSION,
},
)
@@ -1316,19 +1321,35 @@ async def ws_config_set(hass: HomeAssistant, connection, msg: dict[str, Any]) ->
# Marker-to-marker light links are a semantic graph layered on top of
# the lossless controls array. Validate only edges introduced by this
# write so an unrelated edit can still round-trip a legacy broken ref.
try:
# #330 §4.1/§4.2: the CPU part of validation runs in the executor —
# schema + semantic validators cost seconds on a large plan and the
# event loop must not carry that (HA guideline: >50 ms). write_lock
# is still held across the await: writes stay serialised, only the
# loop is freed. The junction baseline of the STORED document is
# cached by rev, so a repeated write does not re-judge `previous`.
baseline = rt.junction_baseline
baseline_counts = baseline[1] if baseline and baseline[0] == current_rev else None
def _validate_config_cpu():
validate_wall_model_transition(msg["config"], data.get("config"))
validated_config = CONFIG_SCHEMA(msg["config"])
checked = CONFIG_SCHEMA(msg["config"])
msg["config"].clear()
msg["config"].update(validated_config)
msg["config"].update(checked)
validate_marker_controls(msg["config"], data.get("config"))
validate_marker_light_entities(msg["config"], data.get("config"))
validate_marker_value_badges(msg["config"], data.get("config"))
validate_opening_passages(msg["config"], data.get("config"))
validate_partition_opening_hosts(msg["config"], data.get("config"))
return validate_junction_limits(
msg["config"], data.get("config"),
baseline_counts=baseline_counts,
)
try:
candidate_counts = await hass.async_add_executor_job(_validate_config_cpu)
except (
MarkerControlError, OpeningPassageError, PartitionOpeningHostError,
PartitionOpeningJambMarginError, WallModelClientOutdatedError,
JunctionLimitError, MarkerControlError, OpeningPassageError,
PartitionOpeningHostError, PartitionOpeningJambMarginError,
WallModelClientOutdatedError,
) as err:
connection.send_error(msg["id"], err.code, str(err))
return
@@ -1370,6 +1391,9 @@ async def ws_config_set(hass: HomeAssistant, connection, msg: dict[str, Any]) ->
new_rev,
previous_rev=int(current_rev),
)
# The candidate just became the stored document: its counts are the
# next write's baseline (#330 §4.2).
rt.junction_baseline = (int(new_rev), candidate_counts)
try:
await _discard_optimizer_snapshot(rt)
except Exception: # noqa: BLE001 — stale backup cleanup is best-effort
@@ -1636,7 +1660,10 @@ async def ws_plan_optimize(hass: HomeAssistant, connection, msg: dict[str, Any])
# Optimization is a normal configuration write with an additional
# layout transaction. It must enforce the same marker-link semantics
# as config/set; otherwise a crafted client can persist a new cycle.
try:
# #330 §4.1: the same executor treatment as config/set — Optimize
# carries schema + a possible full migration, the costliest CPU path
# of all writers, and it used to run on the event loop.
def _validate_optimize_cpu():
validate_wall_model_transition(msg["config"], config_data.get("config"))
try:
submitted_model = int(msg["config"].get("model_version", 0) or 0)
@@ -1655,17 +1682,12 @@ async def ws_plan_optimize(hass: HomeAssistant, connection, msg: dict[str, Any])
validate_opening_passages(candidate_config, config_data.get("config"))
if submitted_model < WALL_SEGMENT_MODEL_VERSION:
candidate_config, _ = commit_wall_segment_model(candidate_config)
validated_config = CONFIG_SCHEMA(candidate_config)
migrated_size = len(json.dumps(validated_config, separators=(",", ":")))
checked = CONFIG_SCHEMA(candidate_config)
migrated_size = len(json.dumps(checked, separators=(",", ":")))
if migrated_size > MAX_CONFIG_BYTES:
connection.send_error(
msg["id"], "too_large",
f"Configuration is {migrated_size // 1024} KB, "
f"the limit is {MAX_CONFIG_BYTES // 1024} KB",
)
return
return migrated_size
msg["config"].clear()
msg["config"].update(validated_config)
msg["config"].update(checked)
validate_marker_controls(msg["config"], config_data.get("config"))
validate_marker_light_entities(msg["config"], config_data.get("config"))
validate_marker_value_badges(msg["config"], config_data.get("config"))
@@ -1674,9 +1696,20 @@ async def ws_plan_optimize(hass: HomeAssistant, connection, msg: dict[str, Any])
msg["config"], config_data.get("config"),
allow_optimize_rehost=True,
)
return None
try:
oversize = await hass.async_add_executor_job(_validate_optimize_cpu)
if oversize is not None:
connection.send_error(
msg["id"], "too_large",
f"Configuration is {oversize // 1024} KB, "
f"the limit is {MAX_CONFIG_BYTES // 1024} KB",
)
return
except (
MarkerControlError, OpeningPassageError, PartitionOpeningHostError,
PartitionOpeningJambMarginError, WallModelClientOutdatedError,
JunctionLimitError, MarkerControlError, OpeningPassageError,
PartitionOpeningHostError, PartitionOpeningJambMarginError,
WallModelClientOutdatedError,
WallSegmentMigrationError,
) as err:
connection.send_error(msg["id"], err.code, str(err))
+1 -1
View File
@@ -87,7 +87,7 @@ const viewport = { width: 1280, height: 900 };
const { page, browser } = await launch(
viewport, 1,
['--enable-precise-memory-info', '--js-flags=--expose-gc'],
{}, resolve(targetRoot, 'demo/srv'),
{}, resolve(targetRoot, 'demo/srv'), targetRoot,
);
await page.addScriptTag({
content: `window.__hpAssertCardContract = ${assertCardContract.toString()};`,
+169
View File
@@ -0,0 +1,169 @@
/**
* #330 §5: the performance contract of the junction limits (#329).
*
* Guards the CLASS of regression — a quadratic path returning — not runner
* noise: budgets are 2-3x the measured post-fix numbers. Measures the exact
* code the write barrier runs: the pure П1-П5 checks over a 576-atom plan
* (the S2 reproduction grid of #330), for both mirrors. The python half runs
* through a child process with the real custom_components module.
*
* Runs on test-build/ like benchmark:wall-model — no browser, no bundle.
*/
import { spawnSync } from 'node:child_process';
import { fileURLToPath } from 'node:url';
import { dirname, join } from 'node:path';
import {
checkNodes, checkSegmentLengths, checkNodeDistances, checkRoomClearance,
cmToUnits,
} from '../test-build/junction-limits.js';
import { commitWallSegmentModel } from '../test-build/wall-segment-model.js';
import { GRID_STEP_N } from '../test-build/space-geometry.js';
import {
innerContourForRoom, multiWallNodesForGeometry, wallBodiesGeometry,
} from '../test-build/wall-thickness.js';
const REPO_ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
const CELL = 5;
const GRID_N = 12; // 576 contour atoms — the #330 S2 grid
const WARMUPS = 2;
const SAMPLES = 5;
// Budgets are calibrated from the SLOWEST machine observed, not the
// author's: the CI review runner measured tsFullCandidateMs at 169-171 ms
// where the dev sandbox saw 88 (code-review 330-r1 H2). Each budget keeps
// the spec's 2-3x allowance over that worst observation, so the bench turns
// red for the O(n²) class (which costs seconds), not for a slower runner.
const BUDGETS = {
tsSegmentLengthsMs: 60,
tsNodeDistancesMs: 80,
tsFullCandidateMs: 400,
pyWarmValidateMs: 300,
pyColdValidateMs: 5000,
};
const u = (cm) => cmToUnits(cm, CELL, GRID_STEP_N);
const rooms = []; const walls = []; let k = 0;
for (let i = 0; i < GRID_N; i++) {
for (let j = 0; j < GRID_N; j++) {
const x0 = u(310) * i; const y0 = u(310) * j;
const poly = [[x0, y0], [x0 + u(300), y0], [x0 + u(300), y0 + u(300)], [x0, y0 + u(300)]];
rooms.push({ id: `r${i}-${j}`, name: `r${i}${j}`, area: null, poly });
for (let e = 0; e < 4; e++) {
walls.push({ key: `w${k++}`, a: poly[e], b: poly[(e + 1) % 4], cm: 15 });
}
}
}
const legacy = {
spaces: [{
id: 's', title: 's', cell_cm: CELL, view_box: [0, 0, 1, 1],
rooms, walls, openings: [], room_drafts: [], partitions: [], wall_columns: [],
}],
markers: [], settings: {},
};
const { config: v9 } = commitWallSegmentModel(JSON.parse(JSON.stringify(legacy)));
const space = v9.spaces[0];
const segments = (space.wall_segments || []).map((item) => ({
id: item.id, a: item.a, b: item.b, cm: Number(item.cm),
}));
/** The same П1-П5 set the card's write barrier computes for one candidate. */
const fullCandidate = () => {
const violations = [
...checkNodes(segments),
...checkSegmentLengths(segments, CELL, GRID_STEP_N),
...checkNodeDistances(segments, CELL, GRID_STEP_N),
];
// #330 §4.7: shared passes exactly as the card computes them.
const sharedNodes = multiWallNodesForGeometry(
space.rooms || [], space.walls || [], [], GRID_STEP_N, CELL, GRID_STEP_N, 1,
);
let sharedRoomGeometry;
if (sharedNodes.nodes.length) {
const geometry = wallBodiesGeometry(
space.rooms || [], space.walls || [], [], [], GRID_STEP_N, CELL, GRID_STEP_N, 1,
);
sharedRoomGeometry = geometry?.status === 'ok' || geometry?.status === 'degraded-extra'
? geometry.roomGeom : undefined;
}
for (const room of space.rooms || []) {
let inner = null;
try {
inner = innerContourForRoom(
space.rooms || [], String(room.id), space.walls || [], [],
GRID_STEP_N, CELL, GRID_STEP_N, 1, sharedRoomGeometry, sharedNodes,
);
} catch { inner = null; }
violations.push(...checkRoomClearance(String(room.id), inner, CELL, GRID_STEP_N));
}
return violations.length;
};
const median = (fn) => {
for (let i = 0; i < WARMUPS; i++) fn();
const times = [];
for (let i = 0; i < SAMPLES; i++) {
const start = performance.now();
fn();
times.push(performance.now() - start);
}
times.sort((left, right) => left - right);
return times[Math.floor(times.length / 2)];
};
const tsSegmentLengthsMs = median(() => checkSegmentLengths(segments, CELL, GRID_STEP_N));
const tsNodeDistancesMs = median(() => checkNodeDistances(segments, CELL, GRID_STEP_N));
const tsFullCandidateMs = median(fullCandidate);
// Python mirror: cold (legacy both sides) and warm (v9 + baseline counts).
const pyScript = `
import importlib.util, sys, types, os, json, time
root = ${JSON.stringify(REPO_ROOT)}
pr = os.path.join(root, 'custom_components'); hr = os.path.join(pr, 'houseplan')
pkg = types.ModuleType('custom_components'); pkg.__path__ = [pr]; sys.modules['custom_components'] = pkg
p2 = types.ModuleType('custom_components.houseplan'); p2.__path__ = [hr]; sys.modules['custom_components.houseplan'] = p2
spec = importlib.util.spec_from_file_location('custom_components.houseplan.junction_limits', os.path.join(hr, 'junction_limits.py'))
jl = importlib.util.module_from_spec(spec); sys.modules[spec.name] = jl; spec.loader.exec_module(jl)
legacy = json.load(sys.stdin)
from custom_components.houseplan.wall_segment_model import commit_wall_segment_model
v9, _ = commit_wall_segment_model(json.loads(json.dumps(legacy)))
prev = json.loads(json.dumps(v9)); cand = json.loads(json.dumps(v9))
cand['spaces'][0]['rooms'][0]['name'] = 'x'
t0 = time.perf_counter()
jl.validate_junction_limits(json.loads(json.dumps(legacy)), json.loads(json.dumps(legacy)))
cold = 1000 * (time.perf_counter() - t0)
counts = jl.validate_junction_limits(json.loads(json.dumps(cand)), json.loads(json.dumps(prev)))
best = None
for _ in range(3):
t0 = time.perf_counter()
jl.validate_junction_limits(json.loads(json.dumps(cand)), None, baseline_counts=counts)
warm = 1000 * (time.perf_counter() - t0)
best = warm if best is None else min(best, warm)
print(json.dumps({'pyColdValidateMs': cold, 'pyWarmValidateMs': best}))
`;
const python = process.env.PYTHON || (process.platform === 'win32' ? 'python' : 'python3');
const pyRun = spawnSync(python, ['-c', pyScript], {
input: JSON.stringify(legacy), encoding: 'utf8', maxBuffer: 64 * 1024 * 1024,
});
if (pyRun.status !== 0) {
console.error('python mirror failed:', (pyRun.stderr || '').slice(-2000));
process.exit(1);
}
const py = JSON.parse(pyRun.stdout.trim().split('\n').pop());
const results = {
tsSegmentLengthsMs, tsNodeDistancesMs, tsFullCandidateMs,
pyWarmValidateMs: py.pyWarmValidateMs, pyColdValidateMs: py.pyColdValidateMs,
};
const pass = Object.entries(BUDGETS).every(([name, budget]) => results[name] <= budget);
console.log(JSON.stringify({
issue: 330,
fixture: { atoms: segments.length, grid: `${GRID_N}x${GRID_N}` },
warmups: WARMUPS, samples: SAMPLES,
results: Object.fromEntries(Object.entries(results).map(
([name, value]) => [name, Math.round(value * 10) / 10],
)),
budgets: BUDGETS,
pass,
}, null, 2));
if (!pass) process.exitCode = 1;
+2 -1
View File
@@ -43,6 +43,7 @@ const { page, browser } = await launch(
['--enable-precise-memory-info', '--js-flags=--expose-gc'],
{},
resolve(targetRoot, 'demo/srv'),
targetRoot,
);
await page.emulateMedia({ reducedMotion: 'reduce' });
await page.addStyleTag({
@@ -345,7 +346,7 @@ try {
target: null,
};
card._rszEdgeDown(quietEvent, room.id, 1);
const plan = card._rszDrag?.plan;
const plan = card._resize?.plan;
if (!plan) throw new Error('large-house resize plan was not created');
const target = [
plan.a[0] + plan.n[0] * card._gridPitch,
+50 -2
View File
@@ -1,5 +1,6 @@
#!/usr/bin/env node
import { createHash } from 'node:crypto';
import { spawnSync } from 'node:child_process';
import { copyFileSync, mkdirSync, readFileSync, writeFileSync } from 'node:fs';
import { dirname, resolve } from 'node:path';
import { fileURLToPath } from 'node:url';
@@ -17,6 +18,47 @@ const INTEGRATION_BUNDLE = resolve(ROOT, 'custom_components/houseplan/frontend/h
const SCRIPT = fileURLToPath(import.meta.url);
const sha256 = (value) => createHash('sha256').update(value).digest('hex');
/**
* Перепаковка кадра без потерь (#345).
*
* Замер на этом наборе: 2096 КБ превращаются в 1689 КБ, минус 19.4%, и все
* десять кадров остаются ПИКСЕЛЬНО идентичными — декодированные RGBA совпадают
* по sha256. Это выбор фильтров строки и уровня сжатия, а не квантование:
* визуального решения здесь нет вовсе.
*
* Почему внутри съёмки, а не отдельным проходом по закоммиченным файлам.
* Манифест хранит `imageSha256` каждого кадра, поэтому оптимизировать файлы в
* репозитории руками нельзя — `check-docs` покраснеет; а если жать после
* подсчёта хешей, следующая же съёмка вернёт неоптимизированные байты.
*
* Отсутствие инструмента не ошибка: локальная съёмка и без него полезна для
* глаз, а приёмка всё равно идёт только из артефакта CI, где `oxipng` стоит
* пином (`.github/workflows/docs-screenshots.yml`). Но молчать об этом нельзя —
* байты кадра зависят от того, был ли инструмент, поэтому его версия попадает
* в манифест рядом с версией браузера, по той же причине.
*/
const oxipngVersion = (() => {
const probe = spawnSync('oxipng', ['--version'], { encoding: 'utf8' });
if (probe.status !== 0) {
console.log('oxipng не найден: кадры пишутся как есть, без перепаковки');
return null;
}
return String(probe.stdout || '').trim().split('\n')[0];
})();
/** Пожать файл на месте и вернуть его новые байты. */
const shrinkPng = (path, before) => {
if (!oxipngVersion) return before;
const run = spawnSync('oxipng', ['-o', '4', '--strip', 'safe', '--quiet', path]);
if (run.status !== 0) {
throw new Error(`oxipng не смог обработать ${path}: код ${run.status}`
+ `${run.stderr ? ` · ${run.stderr}` : ''}`);
}
const after = readFileSync(path);
console.log(` ${(before.length / 1024).toFixed(0)} КБ -> ${(after.length / 1024).toFixed(0)} КБ`);
return after;
};
const roomCardClip = (page) => page.evaluate(() => {
const card = window.__goldenCard;
@@ -129,14 +171,18 @@ try {
const image = await page.screenshot({
...(clip ? { clip } : {}), animations: 'disabled', caret: 'hide', scale: 'css',
});
writeFileSync(resolve(OUTPUT, scenario.file), image);
const path = resolve(OUTPUT, scenario.file);
writeFileSync(path, image);
// Хеш считается ПОСЛЕ перепаковки: манифест обязан описывать те байты,
// которые лежат на диске, иначе приёмка отвергнет свой же кандидат.
const stored = shrinkPng(path, image);
scenarios[scenario.id] = {
file: scenario.file,
viewport: scenario.viewport,
theme: scenario.theme,
language: scenario.language,
sourceSha256: fingerprint,
imageSha256: sha256(image),
imageSha256: sha256(stored),
};
console.log(`captured ${scenario.id} -> docs/images/${scenario.file}`);
}
@@ -147,6 +193,8 @@ try {
// Кто снимал. Смена браузера переписывает все картинки без содержательных
// изменений (#246), поэтому окружение съёмки — часть доказательства.
chromium: browser.version(),
// Чем жали — тоже часть доказательства: без инструмента байты другие.
oxipng: oxipngVersion,
sourceFingerprint: fingerprint,
captureScriptSha256: sha256(readFileSync(SCRIPT)),
command: 'npm run build && node demo/docs/capture.mjs',
+86
View File
@@ -0,0 +1,86 @@
/**
* #314 anonymised dense-plan fixture.
*
* It preserves the structural shape of the owner's report without names,
* markers, layout, or original coordinates: 13 rooms, 44 catalogue walls,
* 24 intentional partitions, and one pre-existing two-point draft. The last
* object is known debt and must remain the only invariant finding before and
* after drawing another room.
*/
import { WALL_KEY_PITCH } from './wall-key.mjs';
export function makeV8DraftRegressionFixture() {
const step = WALL_KEY_PITCH;
const point = (x, y) => [x * step, y * step];
const segmentIds = new Map();
const wallSegments = [];
const keyOf = (a, b) => {
const left = `${a[0]},${a[1]}`;
const right = `${b[0]},${b[1]}`;
return left < right ? `${left}|${right}` : `${right}|${left}`;
};
const wallId = (a, b) => {
const key = keyOf(a, b);
let id = segmentIds.get(key);
if (!id) {
id = `wall-314-${String(wallSegments.length + 1).padStart(3, '0')}`;
segmentIds.set(key, id);
wallSegments.push({ id, a: [...a], b: [...b], cm: 15 });
}
return id;
};
const rooms = Array.from({ length: 13 }, (_, index) => {
const x0 = 4 + index * 12;
const x1 = x0 + 12;
const y0 = 4;
const y1 = 16;
const poly = [point(x0, y0), point(x1, y0), point(x1, y1)];
// Four split top edges make the catalogue population exactly 44 while
// keeping all rooms valid and non-overlapping.
if (index < 4) poly.push(point(x0 + 6, y1));
poly.push(point(x0, y1));
return {
id: `room-314-${String(index + 1).padStart(2, '0')}`,
name: `Room ${index + 1}`,
poly,
wall_ids: poly.map((a, edge) => wallId(a, poly[(edge + 1) % poly.length])),
};
});
const partitions = Array.from({ length: 24 }, (_, index) => {
const row = Math.floor(index / 8);
const x = 4 + (index % 8) * 18;
const y = 28 + row * 8;
return {
id: `partition-314-${String(index + 1).padStart(2, '0')}`,
a: point(x, y),
b: point(x + 8, y + 2),
cm: 15 + (index % 3) * 5,
};
});
if (wallSegments.length !== 44) {
throw new Error(`fixture catalogue drifted: expected 44, got ${wallSegments.length}`);
}
return {
model_version: 8,
spaces: [{
id: 'space-314-dense',
title: 'Dense regression fixture',
cell_cm: 5,
view_box: [0, 0, 400, 400],
rooms,
wall_segments: wallSegments,
partitions,
room_drafts: [{
id: 'draft-314-known-debt',
points: [point(4, 56), point(14, 64)],
segments: [{ id: 'wall-314-known-draft', cm: 15 }],
}],
}],
markers: [],
settings: {},
};
}
+51 -5
View File
@@ -28,6 +28,12 @@ radial spokes visible instead of hiding them under a translucent room fill.
- `accept` requires `--reviewed`, a complete candidate report and current
source fingerprint. It validates the whole set before copying anything and
is the only command allowed to update baselines.
- `scripts/golden-accept.mjs` wraps `accept` and additionally requires the
reviewer to declare intent: every scenario whose capture differs from its
accepted baseline must be named in `--expect-change=<id,id>`. Anything else
that differs refuses the whole acceptance, before a single file is copied.
This is what makes a local capture admissible (see below) and it
simultaneously blocks the one-command "accept everything so CI turns green".
## Workflow
@@ -43,17 +49,57 @@ Review `artifacts/golden/actual/` and, when existing references are present,
`artifacts/golden/diff/`. If every image is intentional:
```bash
npm run golden:accept -- --reviewed
node scripts/golden-accept.mjs --reviewed --expect-change=wall-junctions-plan-t-dark
npm run golden:verify
```
Never accept images merely to make CI green. A matrix/framing change increments
`GOLDEN_MATRIX_VERSION`; a normal rendering fix does not. The first canonical
Linux baseline was reviewed and accepted during the v1.60.3-beta.1 gate.
Future updates must still use the `golden-images` artifact produced by the Linux
CI job as the review set: desktop font rasterisation can differ from the CI
environment even with the same pinned Chromium. Pass its unpacked root via
`--from=...` when accepting it locally.
## Capturing candidates without a second CI round trip (#334)
Accepting from the `golden-images` CI artifact still works and is still the
safest route: unpack it and pass `--from=...`. It costs two full CI runs per
visual fix, though — one to produce the artifact and one to verify the accepted
baseline — and at matrix version 48 that toll is paid often.
A local capture is admissible instead, because admissibility is now *proved*
rather than assumed. Desktop font rasterisation can differ from the runner, but
it cannot differ quietly: it would move every text-bearing scenario, not only
the ones under edit. So the rule is simply that the capture must reproduce every
accepted baseline the reviewer did not intend to change:
```bash
node scripts/golden-container.mjs # capture in the pinned image
node scripts/golden-accept.mjs --reviewed --expect-change=<the scenarios you changed>
```
If the environment is not pixel-equivalent, unrelated scenarios come out
`different`, the wrapper names them and refuses. A wrong container tag or a
mismatched font set therefore cannot corrupt baselines — it can only fail.
`scripts/golden-container.mjs` derives the image tag from the `playwright`
version locked in `package-lock.json`, so the container Chromium equals the
runner's; `--image=` overrides it when a distro-specific tag is needed
(`...:v1.62.0-jammy`). The host `node_modules` is shadowed by an anonymous
volume: the repository copy may be built for Windows, and `npm ci` inside the
container would otherwise replace it with Linux binaries. The run does write
`dist/` and the three bundle copies, exactly as a local `npm run bundle:sync`
would.
Docker is not a requirement of the rule, only a convenience: any Linux
environment that satisfies the parity condition qualifies, WSL included. The
`chromium` string recorded in the manifest keeps the browser build itself
pinned, and `golden:verify` rejects a manifest captured by a different build.
`scripts/golden-accept.mjs` deliberately wraps `demo/golden/accept.mjs` instead
of replacing its checks: every `.mjs` under `demo/golden` belongs to
`sourceFingerprint`, so editing the acceptance tool itself would declare the
committed bundle, the documentation screenshot manifest and the baseline
manifest stale — the very double round trip this change removes. Narrowing that
corpus is worthwhile but separate: `scripts/source-fingerprint.mjs` is itself a
build input, so any change to it forces one bundle rebuild.
Scenarios may also declare a semantic pixel region (for example, a receiving
room that must contain warm light). `golden:capture` and `golden:verify` reject
Binary file not shown.

Before

Width:  |  Height:  |  Size: 104 KiB

After

Width:  |  Height:  |  Size: 104 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 94 KiB

After

Width:  |  Height:  |  Size: 94 KiB

+104 -102
View File
@@ -1,138 +1,140 @@
{
"schema": 1,
"matrixVersion": 46,
"acceptedAt": "2026-08-25T20:12:50.448Z",
"sourceFingerprint": "e28606975682347f90f12971053825ce474085cbddd56082e867370b1039d14d",
"matrixVersion": 48,
"acceptedAt": "2026-08-27T19:14:12.216Z",
"sourceFingerprint": "d3735b5ee1f41b4fbf52f95046ab81f8a4ca9e15b02b8b9ca4ff8e7eff3bb179",
"chromium": "151.0.7922.34",
"scenarios": {
"hidden-wall-diagnostics-plan-light": "6470ed8b09b21144ec1399389e9a57774e977a6473b8eaa7de7e9b5df3273f21",
"hidden-wall-diagnostics-plan-dark": "da6e725ddfad7c807408a710b78090dc222b56dcee484a29baa68b5452fc44e5",
"hidden-wall-diagnostics-plan-light": "37ba151aa6b1ea7b1b270d461ccca348444c5eeb14a5592bf045a0e69b0fb83e",
"hidden-wall-diagnostics-plan-dark": "565d5481f744704a2a6f18658b9b6fa65e7cdbfcd11b50a2bedbc0daae2d8488",
"coincident-partition-before-dark": "19626db704fccb7a42b519ffb20b7bad82c726e8845c752ab451f731d264561c",
"coincident-partition-thin-dark": "3afede1f99b86454bf4e50d9628196a2240243e034644592c5c3efa1b4561930",
"coincident-partition-thick-dark": "57960bd09e6cbbeb073561692690ca68a72d6dbf8242198ef73b8d4ae4d8fbcc",
"coincident-partition-virtual-dark": "da7adbf931ad95f4e620d9d088177f1da20d88a21b9e85b6ff2675925d5557b4",
"span-over-door-migrated-dark": "a817f5baacf5ff80d1f236ffaa176cb6331b75423393dae5b16e54f67f37c7e0",
"split-corner-wall-before-dark": "3176dc67f54d5309f87c94e1077b4f69eb1db9f660469fbf97953038323430f3",
"split-corner-wall-thin-dark": "17e49394c300887bb5349bf477292571e96e209c9515db6853348883d95a1422",
"split-corner-wall-thick-dark": "f7864a6b2ab58586540f836183c9e59f09f9f57165ec0e0b9cd8444ada71153c",
"split-corner-wall-thin-dark": "48062a49e12728c377a7c4352148381b1be2a5606b6792d754ae055b04c22be1",
"split-corner-wall-thick-dark": "a5d795eec4561e3a1aa9c37843fadddcb04d888d5a49e6ffacf29c17b7798568",
"split-zero-divider-taper-dark": "3de38befb41f15ef4047da1390e061b1b5756340142dd9204d011287ff39be5b",
"isometric-geometry-view-dark": "43ad006844a78b502b36aeeb82d82669532e06d2c798de928a5207acd24bd5f9",
"isometric-geometry-view-light": "0fcb10ffdc0baa9006d1c2f0406ebf19e71ba9feda180979a51b74166d7200d2",
"isometric-live-layers-dark": "8aaa914fc45780f67dd5c008a106696d21235a9acee7402f9cc8ae51a26660f3",
"isometric-no-borders-dark": "c105382d883bbdae8562b756ccbf47ac1fcf46675c6aa7654289856223a3ebf0",
"isometric-touch-kiosk-dark": "be122ceec8114374ba384f51d5d9287e498a97354865c5a9f91d505ca18d8640",
"isometric-large-warm-remount-dark": "d76e3334bf80bff265b1e8aa6770c5eff5df6fbfebeb4703b04d048c4252eb68",
"geometry-view-dark-fit": "1b80d2d0fe84cfe15f1fa29ccc2a2dcc66b4c5e62d1722c06c00442c78979be0",
"geometry-view-light-fit": "91caf6621af00eb0cd16589e2fc79976cdca1e8467d0e86151ecbc46948612d9",
"room-label-parity-view-dark": "7f1da43466a13f82a7595dddabe6938ed974debfb14c8f40e729e273a53ac4ff",
"room-label-parity-plan-dark": "5f976d5743d68a27b594e1a5ddf45e5ad3708f5c703409e5cefb7c96d3048096",
"room-label-parity-view-light": "ee3d303fb2b329c356a4354d355c29694130a38d4812ffbfefbe861f5d3f9860",
"room-label-parity-plan-light": "d32b538499a031d4310b243cc9ef43610f6cd67aabfde7cd01b59c88b0363ec5",
"isometric-geometry-view-dark": "0d8427299c77e1e3bb86f37f53128a6640b6666e0b682e5f96085c4f07a657ec",
"isometric-geometry-view-light": "a3e592448ad3ad4ff65266f1fd6f31259c01a9baf47392f6b8be9811d74f8dce",
"isometric-live-layers-dark": "90fa1c0b64a12626e481c5f34e515c53b3e06cd657f5976bb8a72127aec3e1e9",
"isometric-no-borders-dark": "a0abf9fdc31092980de3bcbf8598844b6e3d152363694e9bd070d45af73c26d9",
"isometric-touch-kiosk-dark": "d3ae363165c4f0aba65bcbac9bd6bdca79f29aca03bba997f3abec1860eb1e55",
"isometric-large-warm-remount-dark": "de7733e6aeef7cf7b05cca9c8f6bb4aa627fd16ab2d4198659e24ff9fa3b5bda",
"geometry-view-dark-fit": "2d96e776ea9fc8a136896b58df30abb13a8a53e6c40e5b4f4e12c6dba522380f",
"geometry-view-light-fit": "60e8a908fc1c3be3ca5c9b82bf432120d808d913f743c04683b4a2c501d84659",
"room-label-parity-view-dark": "5ff9dff2da5c747cfab83ade2a929afadcf2eb9340182862da38ab36a35cd3c1",
"room-label-parity-plan-dark": "6528f1edf07e51e470b1c05dcf5292888a14ef6e44d04f7119cfb11b06155c2b",
"room-label-parity-view-light": "789f8f3b6211d35e70021094f2063dc9c6ac57599d9d64743a33b7ae9dd868c2",
"room-label-parity-plan-light": "272dc35d85b1f07efc4c2a05c31f6adbf7d10c648e3c678e7235e35167f83a7d",
"washer-active-cycle-dark": "d74a646bc71b8d6d0d5b09e6862b4846b0e369da2520afa76d5b3d601c832a15",
"washer-idle-cycle-dark": "f7fae5c2f6856a3df5bd3f3028fcf829ef0039c7b43c91c27ac86d25626e8d66",
"day-cycle-dawn-dark": "3f32bc1ed042475516f7e239aedd7fee11717b00fa2103ae9f1231e006d64015",
"day-cycle-day-dark": "1a0f53c948ae669c7ab4e1dcb931c11e7c878974862637dd1f596011876a7ed4",
"day-cycle-dusk-dark": "ec95e99ac3c7e4d6affa2f7944759727251dc4f1b62e2855e421a7e48f7280eb",
"day-cycle-night-dark": "cc49dc53dabb63b7dd800eaa65c8748972fcf3c1abd948b34d1e0f973e63f0d2",
"geometry-plan-editor-dark": "fd8d067daa3b2a4e572dea9644486e6324318e8ea4e5f43e3df2069a05389a88",
"safe-resize-handles-clamp-light": "ed705040357cad043095baf12d02cc69abdc3db27ff40066a6979be85d035f77",
"safe-resize-handles-clamp-dark": "9d81a45f3acac787cde1b9eb74a858020fbad09670b8ff5b081869b5eca73b82",
"space-tab-drop-before-light": "401dded3d7fdff3e823d46eac1b3cef2d6e0eb05f48b5ea289cdd8ecd941dee0",
"space-tab-drop-after-dark": "a851621040e14f911b1080488f2c75e3d9bc0d5fef74fb5d0081f80b023ff0a2",
"plan-snap-endpoint-light": "db967076cd9fd50c2a2a1337e4ecbf4ef67cb7f800401f6227497c75415951d5",
"plan-snap-line-gaps-dark": "ff1b6bbf2e1b4272bb2f0e96d92136aa6960ee24acef1b63b4478e54441a4b58",
"wall-junctions-plan-preview-light": "fa81fa6c6dba92578a4beee5ed29f407443a3bc61affa25c44fcfefab0c7070c",
"wall-junctions-plan-t-dark": "f8a517d316a99ca9ee5b4ff56f515be15d440adfc551bbdaaa5e9d9a411a79bc",
"day-cycle-dawn-dark": "be3b891520637857b7f89dc88284753a15601ebae4db7db750714b9535348d33",
"day-cycle-day-dark": "1b791b8a773a5910a71338351ab12fdc36c22742a358b647ff4f295431fa4cfd",
"day-cycle-dusk-dark": "14182ad89301cb260ba21ee5077134ed408f038ea878ed4539b84f987182f910",
"day-cycle-night-dark": "407632fc54e13e6362b7e583d6e2b7bdd31b4093823846e4e630de0c8f542120",
"geometry-plan-editor-dark": "40a0083511c1ff9d0b8f1ec419af63e2a28f1f9a0ec3768387a10c89efe73905",
"safe-resize-handles-clamp-light": "f47a5bb09fa4415416d01ca5f4164014cc1d8dc7be54ea888e5399277104e94c",
"safe-resize-handles-clamp-dark": "9cb3cb83d2aea4260fdbf1bc1d4281deecedeb9c9dbb0adbed921470b9aaf397",
"space-tab-drop-before-light": "19c699fd035030318ce578fdb6427d6243cd3f1d955ea0fbf958cad385e9c7db",
"space-tab-drop-after-dark": "1f802274c3015ad6422ee9b50a8d67a412407513da09de188be3db2f7a3873e4",
"plan-snap-endpoint-light": "2185d5ec6726e07097ba1a911984a11d5ca9b460a1cd1e7b2a719255b7d0ed0b",
"plan-snap-line-gaps-dark": "276eb10f32be938640bd68a1cf837494f11cce8d10703721831360f3a1100d23",
"wall-junctions-plan-preview-light": "d92aad64d7aa8641c6743b51a7d13d9166b15c49f3d8aee17b8e61ff263a57a6",
"wall-junctions-plan-t-dark": "45a98164ff93b01fbb7f3ac936ee8636c36dd791c09ac3b160b7c7ad803696d1",
"wall-junctions-view-dark": "7b859c4f25f8a5dd4fca64d5b2f7aa64b65845fabbe83bc2388c2f99d29f71b1",
"junction-patch-resilience-plan-dark": "57ea7522c438e52292f226f975bb112930976b6e072b72322180d52ec0277f4f",
"junction-patch-resilience-view-dark": "1f0df26ed72bcc5bef06a8366d7043ad40b4a2bb744fa5a64248c29eae968424",
"wall-union-isolation-view-light": "291c045758be2814be88d12ab46bdc3d12cb99a03dbbcc9a1539c2c9a5931edf",
"wall-union-isolation-view-dark": "616fc6cc7144c9596ae52421027914e222f30f26de45215fa7e7b6175ca78543",
"multiwall-junction-bevel-view-dark": "e226d023fbab0502eb747395d43a46bb56bcf2be037ba4f9b7ad74ba406d9671",
"orthogonal-strip-cell-5-view-dark": "9b8e4bd22cbf1a82ff48f052822967c61179290a3234ae48ef547642329b5a90",
"orthogonal-strip-cell-1-view-dark": "e59e51ad3645144161459c438a07420730c4dcfd2c33bd2b18e8c3e57ea84515",
"wall-key-roundtrip-view-dark": "df46b4e209f98b97958748097ddc646ddc64d8285c2651762de71ec2456ac690",
"junction-patch-resilience-plan-dark": "90191f637f4d1345d82c63977513524fde353c4e3f87de1a723bf7d3b1f8e91f",
"junction-patch-resilience-view-dark": "afd83241d9a1e1f19ab59c3f1e6fa7af058c3e42a7fafa5ba271ff92ccea8f91",
"wall-union-isolation-view-light": "106f8ac97b82107fed36aa4456c0b9f7a39eeb44be86864b1d4f083497cdbd6e",
"wall-union-isolation-view-dark": "914451a813546189bdc9ac22dcf158bbc822b4403080a3c2fd4c6b02bb7780a4",
"multiwall-junction-bevel-view-dark": "6b02d1d0c5af8b386791cc5764b3bc407129f6576b6d9591c64fd388a8a6865d",
"orthogonal-strip-cell-5-view-dark": "66a288f7e7fc6228c3c63f10e1df8df0656ca0cd36681d998b126b03acb70b5a",
"orthogonal-strip-cell-1-view-dark": "c23e6d0410b8f5c2b7fb0707f6bea40408730af08f9e7a0b166aa8c0a3d6c85f",
"wall-key-roundtrip-view-dark": "d7b0749d0f6c0cfa74ad4d9f76d8a7b7962fadf4e2648260b5647548a988341c",
"isometric-wall-junctions-dark": "d1956cbbde9a6a02953ce80eb8a0f74ca06ea268a4c60bb09b5b3981c26d9731",
"opening-placement-door-thick-wall-dark": "1157139aac7325b40e3869b8c3b5444509b4bc24a415ddf6f8d5402624f3a04a",
"opening-placement-passage-thick-wall-dark": "6d4a9a953b70401ca11cd6ce96c2e0e0781bbfff71e758dc57958a4163e2712d",
"opening-placement-passage-thick-wall-light": "b004b1395652c8a8eef5f5002b471ecf8f2f0347f158a14a056ff2b7382e4ea3",
"opening-symbol-room-wall-light": "562fed297210ef542de608e384c2d344dddd193a5b43615930c9f76faaf53b42",
"opening-placement-door-thick-wall-dark": "c21c2af812b847e7fd9a3a5b9df832db95266f81eba3d7b48c97d2b32619e7fe",
"opening-placement-passage-thick-wall-dark": "90f9a837898169c2d9ff9ec3b44b2a3d247ea034fd84b1753519da7b0c89ebdb",
"opening-placement-passage-thick-wall-light": "dd17884027d85f5727450cd08f3fe6daaed290770dbd04fe5ce10c4c2b9d43db",
"opening-symbol-room-wall-light": "403f1c6e1c59122331fbe99019a3ebd80de4106dc6d33f5f5f33feb748229645",
"opening-symbol-diagonal-partition-dark": "620102c00e77613cb341c906829bfcc7e9a5430075aa576f563adafa8bdd8f08",
"opening-symbol-flip-pairs-light": "33364f51fb65e49bfdd6c41d6e0c53542c89fadc45a3c6e5adc590144c4676ea",
"isometric-opening-symbol-parity-dark": "302daf5d90d4af5ef90036cc10e957e6fc909ea56f9f2af938ab850d4f608e34",
"geometry-devices-editor-dark": "4b91bdc00082e33d2529ea3cf7ce7af36383a84838f4a2c909463df24483b47b",
"geometry-decor-editor-dark": "8b03bc87d2353cb6ef8fb4e2dec3f0d3be85de903ac9346171ea4aaa31170a66",
"tray-wide-selection-en": "d140cf6a51c634a38d2e3661c4b61835d63673174fd934b85f9da0db4fa20d1a",
"tray-wide-tool-ru": "806ff7c5ae3d2e708a4682833af7dc28a88001ef09e4ca564c91a5e6023b297d",
"tray-medium-group-en": "5a1adbaa64d2b73f294003e2752a79008f7ec313ab41dd0514b6b38918b8c00e",
"tray-medium-selection-ru": "53c7ea5099772fd14c2e27ad414c4d57d87c020866992b7a0b0dd62a22fbeb7e",
"tray-narrow-palette-en": "5bc029bb7a4b077a45abc0c18f54360f781caa3684befb63cfa714efcfaf7edd",
"tray-narrow-tool-ru": "32e27435ec9f0ce6db2e2800ffd71b6a5fcb1ab7dde17454a3ba7dda680b8a45",
"geometry-diagonal-45-opening-dark": "a0167f3028cd8ef8283c92ae7e803940cf0eb0db45f1b137108a4e009741a5df",
"openings-thick-wall-dark": "c45bae26ec0b4f719f72e5b92e3b8e03260d7bda4b3182d4c563b50f656e68a1",
"openings-filled-tunnel-dark": "b916d3ce48bba7487ed4b548bf659b610243ebcde07206c21fff65867792d0d1",
"decor-over-opaque-hover-light": "553d279db44aa0a4a6855496036b0f6ddda9570cef133e09a30f29f41b7d0c31",
"decor-over-glow-base-dark": "8e5b8630798e31936f8e7dd52d1cba73a3f80fd3d3217a5ad59d50b5dc12a26d",
"openings-hidden-view-dark": "d99f365376e6ece668060181ce8c3a5ebdb93ec5a8c64e08b0aef5af592c6874",
"lighting-glow-sun-dark": "8599aea80a0479442f891adb30aecd647ddf217e28e3f0da5c21b41e21d746ec",
"device-value-badge-positions-dark": "ae163e96cd976e006eacd09ebc75440d600e81c063e38da27d53000e4e51c68d",
"device-icon-state-table-light": "969879fc7baeb193639a03701eb8bacc9469f6160b8d41a318db8da83590e51d",
"device-icon-state-table-dark": "045145a87ccde3894334009c6b9edd690654fd048c55b770e09d5ed79f9a7351",
"device-text-shell-long-light": "a96d16ac0a02ed322bab949e34ff7ce59da5f5db42e02185fac5fde495591bb1",
"device-text-shell-long-dark": "f301f3aa1e1eb382018c4329db2909a8d2990f5f917bde83e3de46ff4dc7416c",
"opening-symbol-flip-pairs-light": "6d780d1d06909e6e5a7ebb35987b9681d6902d78408e974dfe2fe587521990ce",
"isometric-opening-symbol-parity-dark": "1e254ed43da00a16d3048e9a2d29db0a43c7790991b1006ecb7205d002bd1f63",
"geometry-devices-editor-dark": "a57c8528f2075252dbc0ea11af23a6c33a3b533b4f80b43ab5eebaa918e12816",
"geometry-decor-editor-dark": "2a0d4368e49b51fc4fcf5ee72679efb7b7e094fd8b6f47682b0696ec909c423e",
"tray-wide-selection-en": "056f3b54d60bba83ef313f9c12fe697ea6a0b3d48db0470735394ee0d3f72c39",
"tray-wide-tool-ru": "f8e050c96cb3e78f1ab8d0b8b276b3d154950e795c5b620148b9e395330cffe9",
"tray-medium-group-en": "4513abc13f1bc24e4d04f14445069c8ce9785dc3fddcdc99bcee41b5eadb8d53",
"tray-medium-selection-ru": "bd97bee66778a65ebb849331724a307d0b089d6cdf5f3b85c6c905c413d5290f",
"tray-narrow-palette-en": "10073be9fdf658275f77643505c3523bcddce6a1890deda5b1820958b90613d1",
"tray-narrow-tool-ru": "95fa0f072efb595e1f9fe3f3287e7ce346fdee640e90dd70e4f5e5b31049ef29",
"geometry-diagonal-45-opening-dark": "ba9462322da34b20fff1ee1f950c516855dd7ac56131e13863f576e2d0ba5177",
"openings-thick-wall-dark": "539cdb9cb1460cd42c2e51901ea50f496e8214d02633e110e2780c3f177154db",
"openings-filled-tunnel-dark": "bb2a4343a285e2fd867c6090108589549facf87b1a9471f3ae331d743af6653c",
"decor-over-opaque-hover-light": "eae79026256367a4cfaa5f99393ec9c66dbf23908c50ccc0f541c139d1baeda0",
"decor-over-glow-base-dark": "72a83f08f832c219d8e0e1b6f94d689c250342aa95d30e8c1f79cf9f9548ccab",
"openings-hidden-view-dark": "19ab4fd2ffce0fea0fe609b007a3e60f7abd84b40935cbf3c5fa9dc1fda21046",
"lighting-glow-sun-dark": "08e2964df6fd03abad4b9a81f1dbd0fe3812fb6b5f7908aeb027ad2953b2f88c",
"device-value-badge-positions-dark": "75ab0210d697f952b8a8977649e84308de3ce3b59c84ec93071179a1d6cb015d",
"device-icon-state-table-light": "85c785824ccafc2d5f50e834ad2d746319d64ecb292d1b9cf7fab8661d5d410d",
"device-icon-state-table-dark": "03a90523132c3dbe6c78d44d01fe89df93e2d0474f91ef84e7107e6e4bc7d337",
"device-text-shell-long-light": "97a113b5164f373b4af2922c8e71bc1f063d8c2761b72c3182697fcb1aac3cab",
"device-text-shell-long-dark": "c8fe32aced704c7e7562f0c8499183fad848a977063ab731401cc01e7f20ef38",
"lighting-sun-window-state-only-dark": "4da798c4de6d1303d17be246e9a2e63b12da3f53a42136cf745d1a92ad5a4f21",
"lighting-fill-light-axis-split-dark": "6b3a4dd637c6fe7e43a131c44c4be232bd81edab417ea9af3860d82c506ad45f",
"lighting-fill-temp-axis-split-dark": "6025c6efd3ea3565c13e147db550b017eeb86da0be5dfc82a8b5bd7e623d5852",
"lighting-fill-lqi-axis-split-dark": "1ec5c18572fd5be471088286c4aa7c67c3885cb907080fcde44c3583ff794275",
"lighting-temp-glow-dark": "274bcd5a19ebbcb63fb64a5d21f838122f1916d428defcaa3c0356d8ea9809a5",
"lighting-temp-glow-light": "decc386073a1e93d8d457ec09daf83e8afc180977f5880ce56cc4834a2c203f0",
"lighting-custom-glow-dark": "32607e6001c4c8cd541cabf82142f3cbf7f2b298900a02f30c6f44ececbd5907",
"lighting-opaque-glow-two-doorways-dark": "bd9f49432c69c8ed622810e96f4708974e2024c58700b0cfa3e5ebd2edd7bc7b",
"lighting-custom-glow-light": "973ae32bb3bacbfe633041e89352a9ef1a2222b89a4792feef45c26bfa541459",
"lighting-temp-glow-no-sources-dark": "e19b7d86d5dae54f9afc0a1ef04a3848ebec283a5f60e68efdcc7380dff3b0a7",
"lighting-temp-glow-room-override-dark": "bebfc27f460cb75e32ab14076a01e3b9729763d7cecd3f457c1b696ed3e57da3",
"lighting-manual-auto-spill-overlap-dark": "471e39397361832a171f0309302be7d8ba4310d628e337478e23da44dc455ef8",
"hover-over-glow-dark": "6581433024060f773e00211d60eac7b6daffe442031ff076f861918419c71930",
"hover-nested-room-dark": "f05ba332a127b56c00a6a94eac727b168ed34cc3f0355f8b783827f7bf55714c",
"lighting-fill-light-axis-split-dark": "e45d44496c83ad92f8d5bd058821e6d71c97c48a2b66c1933206535aea2655c1",
"lighting-fill-temp-axis-split-dark": "895e1e0beee38bf659b1313ed9951353a7792d2e8ae3a8620c4cf7a4f6a84a4d",
"lighting-fill-lqi-axis-split-dark": "4a029b17419aaf60feab23f7f0cb385d97353e3aec0e39ed67e9ac94f6c08029",
"lighting-temp-glow-dark": "3ce504d83fda2ba5d772c25f8cbb47c379b335addd8650f6933b58f22f5655a7",
"lighting-temp-glow-light": "e9f2db0a686a9e20bf71564c10bab826a2a473c73d572a6b67ef02a735d61e8e",
"lighting-custom-glow-dark": "94cc6135573aa9462de9e7f309dea62e236668b84970bae35d48f1aad338f8f4",
"lighting-opaque-glow-two-doorways-dark": "ee3c53a408f6b62aa6b770925fa6948934e0c1c6e73877e379695bc260b0ceac",
"lighting-custom-glow-light": "aa9477aaff4d137abfcb6d67a5cf62b91f41b3631383ee8edb7499802a5c9b87",
"lighting-temp-glow-no-sources-dark": "cd992af54d1c9dcc58bbd0b21b15b475be6f19eaabcf9ea866bfa82f205460c9",
"lighting-temp-glow-room-override-dark": "66a418829b51379d024c542f8b9a12bd42d72afa81a6bf66561d6498ac4cdb45",
"lighting-manual-auto-spill-overlap-dark": "0dc780335317d6dd577c8d5539f33223228074e1c95e936e4666f50a8db09739",
"hover-over-glow-dark": "49136f6073662482ebf49a28f8f22b698d2c20a9cb64b5667d111b9835717968",
"hover-nested-room-dark": "b546f923fd31d28d2881f074e3a67c5d9d3f9450169f9118502ecafc98c4467d",
"junction-t-90-equal15-dark": "fed083eaaaf1588222002c23f7edd741645c76f57c49a3db8cc5ec22af0b6e7d",
"junction-t-90-bar50-leg15-dark": "e981f0597a55e9e9e623dd49417f33a0ebac18d18e0e66675a9af1f2224bd152",
"junction-t-90-bar70-leg15-dark": "4428fbe75679fab6e57ef61c5d5097d498960e2539bb9e759e1bae9a160ef174",
"junction-x-90-equal15-dark": "8fe0e262e5873714f94b9a6a76ed920881bc6b267e69153039f52a7102ec55d5",
"junction-x-90-mixed-dark": "deabce4c2b5fa7866a4f56d25404c6a515d47437298bde114442bc80a6339560",
"junction-x-45-alternating-dark": "56a85069baafd6e7af921389b55b916b74cc5d76df908782013b94c2b5b87752",
"junction-star5-equal15-dark": "1b9714dffa3f37b9f1e8d848244c4585d38539b74fe582e8d974e2b9e8fdd01a",
"junction-y-60-equal50-dark": "f90d46cfe79a95bb3ec73253d0b7f77948164e90c49561d79b1e0db655cc0847",
"junction-x-45-alternating-dark": "49192516aabc14e84a2da4d1361c1be57d24e5c947b199cdf8b6c55fd7318504",
"junction-star5-equal15-dark": "2fd822f73f486173700405208168300d27d4e360b17cb278c7ee7bc1cda3769e",
"junction-y-60-equal50-dark": "7a61dbfc36437674c2c2bf5f450a64797e21a124f42c0e45bfb56e3236e0b53e",
"junction-acute30-mixed-dark": "9ea64ae5864295a1d93a6819efb6e210de2599030fe650913862d99f40b11a69",
"junction-acute15-equal50-dark": "3327caf8bc1ccddb317aa7c4c330e09c098122a4f6d2e4df64502cdc033cb2e4",
"junction-acute15-equal50-dark": "e7e657137e8dcf6294d400f3114497318c055405dbc0d867603514b56c42b907",
"junction-splay10-170-dark": "ae27454bb4213237313f1fe9e2591ada48facf20baca5517968f92aebbfb330e",
"junction-t-virtual-arm-dark": "d325dc112e2f56c31d4511fd586adaa92b6785fc676c2a60ed5aba69c164fd6e",
"junction-x-virtual-through-dark": "cce24ac9d7678959844803c6306e65ab48f723ec0375ee6d3f2dbf1c7def7be1",
"junction-column-node-dark": "fed083eaaaf1588222002c23f7edd741645c76f57c49a3db8cc5ec22af0b6e7d",
"junction-draft-end-node-dark": "6453bb8463ffdc9a42dbd7acb171c6249fe6d5beec0eae0c789e40f48e65a1f9",
"junction-owner-repro-dark": "23ae216c2637b8ee9ea8f852002693f74d64021a66d4494e2086a53db8ca6520",
"large-house-zoom-040-dark": "df1a5515089b5e470ae550ce6da31a301be5079d3f7bb9130baad99d3d7c641e",
"large-house-zoom-250-dark": "88c0d22686fd9ef314866d0275a5ae785781c3d671e257b2c5f790f816cd65ee",
"large-house-warm-remount-dark": "9a61d668c5fbc61bb370c043771cee91707b3d40378cf89f81199882820bf95c",
"device-dialog-desktop-en": "9739886af3104c2fe3d6d3138293e9fd1a7f480f151decc26f12bb64cab69543",
"junction-309-step-dark": "f5e0e649ee19197026b87687fd3512b18536883f071348589ec3a7f0ea2a216f",
"junction-309-spike-dark": "98e709aa2fa58a01fe280635c6a3799e3a5f4158b2ada12c31c38818f92c4f44",
"junction-309-hump-dark": "abae96fd96ed8515d7b1e33b84858bdf08ac28a888898549874baa8c98202e7a",
"sharp-apex-legacy-dark": "93a46c66f7263a0325d0c4d50fa0677f3c902647ea5880960b99dc8fa04a01a5",
"junction-owner-repro-dark": "f5407d420449ebf897f94fcfcdaaf4630e818c043d3076b090f7d1944df5296b",
"large-house-zoom-040-dark": "736af1c340646c4c1cbab3d2270f67a535995e1ea1e2abe49c91d2c396052097",
"large-house-zoom-250-dark": "3f7abdd66a7b33301275983dbf83862bd24f91c65a7d731c411ebd3ed01b381d",
"large-house-warm-remount-dark": "1c14dba8cdbc4628984091d395c1ac14faead1d43338e44aa3cb57976e519df6",
"device-dialog-desktop-en": "a6637997458ef1def3d30ffaccb6bbc9728dbc4ba2f02400a7334915a8101bbe",
"device-dialog-mobile-ru": "c90be98e65c37963fd4443e413848ca68a412c9ee568d648aba76bf9232ee59d",
"toggle-entity-dialog-desktop-en": "f4924466de5e134c2d1f2456ff0c05f8aa8106f6b06398e87a263e3c536f16ea",
"toggle-entity-dialog-mobile-ru": "7d3aecd318c6c0774dd3ce1c21c0fafb1ae2be23cebed6f548fbf8bd11ab2f62",
"device-help-popover-light-ru": "add3275d74ced358a184ec9f864c83735f1196e2bff7d75fb3e12160cb85bfaf",
"decor-color-popover-mobile-ru": "51725859afc6a1e2d1b2ae8be270ab584608901623ef7e2d283b6b195644fd16",
"decor-color-popover-desktop-en": "8ad229fd3d433f969dc6d09c7677e9949d4f62503196bffbf17030386f7c7511",
"general-color-popover-desktop-en": "91c3b7ebae5146c62d172d1ce10ae9e375ed955f23e9dc860976373a334d8be5",
"decor-color-popover-mobile-ru": "fcdd138e396654e3517927c27d2062663c61860e8b9903471fa44651a4cbdefd",
"decor-color-popover-desktop-en": "98d769f5393afb2bad3fb5143a562da5b8bb9a5c8049f4c707706bbcb4f99f8d",
"general-color-popover-desktop-en": "008865918113e05847a7ad8ae0fdc068b5326acf1c591d855806b333084978a3",
"device-ripple-color-popover-mobile-ru": "3ab12c339106004c76fe3c7741d242269e157c40ddfa29afd8f13b445b0a7023",
"space-room-color-popover-desktop-ru": "d66d068a1b403f88a5bd3c57f4a9e55c336767c533ddfe3ae251cc4f9da40a74",
"backup-full-preview-desktop-en": "c9fe8ca1cf63b9ed98e2d517a2c06639d044d460011d958573e70c6b22bd7a2f",
"backup-plan-only-export-desktop-en": "c48b3ccdd7d85d36f132c2d7b4c5dba94a1027e3e3f2ef495b4eb89a23f8ca33",
"space-room-color-popover-desktop-ru": "946c550232ff3819bb44058eff6c5deadfccf85527c23065137ee02fa457aed9",
"backup-full-preview-desktop-en": "6b68188269fdbaddaf4f2ce8fc0d10ea41e050f0c08cdbd9507548871a3786ef",
"backup-plan-only-export-desktop-en": "f3c01ae191af41fcc815d27586fe40f965ae3fddbf782553076703ce695457d6",
"backup-space-preview-mobile-ru": "82cd93cb495ad77b92d635b4a3715601b1ae0e5e2974edcbbd461fe872c75bad",
"optimize-preflight-dialog-dark-en": "6e0706b951db73c7d699173d9cd0cd3cd8857238e3d7e95bc38eeb3691f907b6",
"optimize-preflight-dialog-light-ru": "a96af4b4428297d2c672f0d0e8f376e81c27765ecc941781a2d3873a054e2ec3",
"optimize-orphan-references-dark-en": "cdf4eb8c79a9961a1e961a4eed4107d4d849ebc5b922ad88ce07befee6b65a9a",
"optimize-orphan-references-light-ru": "c07151ac896e43bef6150128cbef8a5798ee113083961ed8eaf97113a0b56e1e",
"optimize-preflight-dialog-dark-en": "0473d7f16dd830aaddbec5c33f2b423fde48e2053a1094dbab052e6a2cbad9d4",
"optimize-preflight-dialog-light-ru": "a4eb6ef9930652bd7e55fbb725fd1bcc5a0e5b71e8060dbf6f772dd62cec5d36",
"optimize-orphan-references-dark-en": "f40417298e126fa804ddafb91d62b728c86a56c5a28b4fad12b732070faa4ad6",
"optimize-orphan-references-light-ru": "2d0319f681cfbe472ad5dede05d145a995dedee577d8d816e2fe46a519e019f5",
"card-editor-invalid-default-floor-light-en": "dbf251ece014cfee252e0dcb2d7c00eda66aa4b5bd125e4c83e248d4319d24b3",
"card-editor-invalid-default-floor-dark-ru": "422b74bb391b87e383350247ef1bc1f10ca48c8f6fe418f75bde94e62e3b8bc0",
"junction-309-step-dark": "f5e0e649ee19197026b87687fd3512b18536883f071348589ec3a7f0ea2a216f",
"junction-309-spike-dark": "98e709aa2fa58a01fe280635c6a3799e3a5f4158b2ada12c31c38818f92c4f44",
"junction-309-hump-dark": "abae96fd96ed8515d7b1e33b84858bdf08ac28a888898549874baa8c98202e7a"
"card-editor-invalid-default-floor-dark-ru": "422b74bb391b87e383350247ef1bc1f10ca48c8f6fe418f75bde94e62e3b8bc0"
}
}
Binary file not shown.

Before

Width:  |  Height:  |  Size: 130 KiB

After

Width:  |  Height:  |  Size: 130 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 102 KiB

After

Width:  |  Height:  |  Size: 102 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 137 KiB

After

Width:  |  Height:  |  Size: 137 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 118 KiB

After

Width:  |  Height:  |  Size: 118 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 176 KiB

After

Width:  |  Height:  |  Size: 176 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 80 KiB

After

Width:  |  Height:  |  Size: 80 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 33 KiB

After

Width:  |  Height:  |  Size: 33 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 36 KiB

After

Width:  |  Height:  |  Size: 36 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 348 KiB

After

Width:  |  Height:  |  Size: 348 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 61 KiB

After

Width:  |  Height:  |  Size: 61 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 58 KiB

After

Width:  |  Height:  |  Size: 58 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 46 KiB

After

Width:  |  Height:  |  Size: 46 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 45 KiB

After

Width:  |  Height:  |  Size: 44 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 63 KiB

After

Width:  |  Height:  |  Size: 63 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 117 KiB

After

Width:  |  Height:  |  Size: 117 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 290 KiB

After

Width:  |  Height:  |  Size: 290 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 279 KiB

After

Width:  |  Height:  |  Size: 279 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 44 KiB

After

Width:  |  Height:  |  Size: 44 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 335 KiB

After

Width:  |  Height:  |  Size: 332 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 45 KiB

After

Width:  |  Height:  |  Size: 45 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 45 KiB

After

Width:  |  Height:  |  Size: 45 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 264 KiB

After

Width:  |  Height:  |  Size: 264 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 279 KiB

After

Width:  |  Height:  |  Size: 280 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 47 KiB

After

Width:  |  Height:  |  Size: 47 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 182 KiB

After

Width:  |  Height:  |  Size: 182 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 64 KiB

After

Width:  |  Height:  |  Size: 63 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 66 KiB

After

Width:  |  Height:  |  Size: 65 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 177 KiB

After

Width:  |  Height:  |  Size: 176 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 150 KiB

After

Width:  |  Height:  |  Size: 150 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 164 KiB

After

Width:  |  Height:  |  Size: 164 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 29 KiB

After

Width:  |  Height:  |  Size: 29 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 25 KiB

After

Width:  |  Height:  |  Size: 24 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 16 KiB

After

Width:  |  Height:  |  Size: 16 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 31 KiB

After

Width:  |  Height:  |  Size: 31 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 340 KiB

After

Width:  |  Height:  |  Size: 337 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 41 KiB

After

Width:  |  Height:  |  Size: 40 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 52 KiB

After

Width:  |  Height:  |  Size: 52 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 25 KiB

After

Width:  |  Height:  |  Size: 25 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 30 KiB

After

Width:  |  Height:  |  Size: 30 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 172 KiB

After

Width:  |  Height:  |  Size: 172 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 75 KiB

After

Width:  |  Height:  |  Size: 75 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 139 KiB

After

Width:  |  Height:  |  Size: 139 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 189 KiB

After

Width:  |  Height:  |  Size: 189 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 188 KiB

After

Width:  |  Height:  |  Size: 188 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 62 KiB

After

Width:  |  Height:  |  Size: 62 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 62 KiB

After

Width:  |  Height:  |  Size: 62 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 62 KiB

After

Width:  |  Height:  |  Size: 62 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 178 KiB

After

Width:  |  Height:  |  Size: 178 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 209 KiB

After

Width:  |  Height:  |  Size: 209 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 161 KiB

After

Width:  |  Height:  |  Size: 160 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 186 KiB

After

Width:  |  Height:  |  Size: 186 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 184 KiB

After

Width:  |  Height:  |  Size: 184 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 63 KiB

After

Width:  |  Height:  |  Size: 62 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 180 KiB

After

Width:  |  Height:  |  Size: 180 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 44 KiB

After

Width:  |  Height:  |  Size: 42 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 339 KiB

After

Width:  |  Height:  |  Size: 337 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 338 KiB

After

Width:  |  Height:  |  Size: 336 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 357 KiB

After

Width:  |  Height:  |  Size: 356 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 31 KiB

After

Width:  |  Height:  |  Size: 31 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 30 KiB

After

Width:  |  Height:  |  Size: 30 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 58 KiB

After

Width:  |  Height:  |  Size: 57 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 56 KiB

After

Width:  |  Height:  |  Size: 56 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 60 KiB

After

Width:  |  Height:  |  Size: 60 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 107 KiB

After

Width:  |  Height:  |  Size: 110 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 123 KiB

After

Width:  |  Height:  |  Size: 130 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 74 KiB

After

Width:  |  Height:  |  Size: 93 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 78 KiB

After

Width:  |  Height:  |  Size: 96 KiB

Some files were not shown because too many files have changed in this diff Show More