deep-review #1964: v_objective_lots_latest has NO premise/district filter inside,
so a consumer's outer WHERE cannot push below DISTINCT ON → the view materializes
the WHOLE table (Parallel Seq Scan + external Sort 1.76M rows, ~55MB spill) on
every query. For REQUEST-PATH consumers inside analyze_parcel this is a ~19x latency
regression vs the pre-#1964 raw-table plan.
DISPROVEN remedy (NOT applied): a full index on the physflat-key does NOT help —
DISTINCT ON selects ol.* (51 cols, width≈945) so index-only-unique is impossible;
the planner ignores the index (seq-scan+sort still cheaper) and even forced it is
~3.9 s. A 142MB index for zero request-path benefit + slower bulk-INSERT during
objective-scrape is wrong. Honors original #1964 decision "no new index".
Prod EXPLAIN (Академический / 3km radius, 2026-06-28):
consumer via view inline (this commit)
concepts median 5854 ms 1640 ms (bitmap district + sort)
parcels district 5854 ms 1640 ms
parcels geo-median 6443 ms 122 ms (NestedLoop geo->complex bitmap)
parcels obj_pricing 5721 ms 441 ms (project bitmap per nearby ЖК)
FIX: keep v_objective_lots_latest ONLY for batch/background/cached consumers
(supply_layers L1, competitors._SOLD_COUNT_SQL, special_indices [/forecast bg task
30-180s], admin, landing). Revert the 4 request-path consumers inside analyze_parcel
to inline DISTINCT ON (physflat-key, latest snapshot) with the filter pushed INTO the
CTE so the district/spatial/project index applies:
- concepts._OBJECTIVE_MEDIAN_SQL
- parcels.py district price block
- parcels.py geo-radius median (complex_id-scoped)
- parcels.py obj_pricing CTE (project_name-scoped; aggregates over deduped set)
Migration 175 header CORRECTED: accurately states the partial mig-173 index does NOT
serve the view (qual can't push below DISTINCT ON), the full index is disproven/not
added, and which consumers use the view vs inline. No DDL change (still view-only).
Tests: +guards (concepts/obj_pricing dedup inline, not view; obj_pricing physflat
DISTINCT ON; perf-pushdown scope preserved). 965 passed.
Финальная часть эпика #1953: пользователь выбирает типовые дома
(тип × этажность × число секций) вместо авто max-FAR раскладки, формируя
building_program из Stage 3a.
Бэкенд:
- GET /api/v1/concepts/house-types — read-only каталог HOUSE_TYPES
(section_type, label_ru, footprint w×d + sqm, default_floors, housing_class)
как single source of truth; фронт ничего не хардкодит.
- Схема HouseTypeCatalog / HouseTypeCatalogItem в schemas/concept.py.
- Тесты эндпоинта: полнота каталога + совпадение ключей с available_section_types.
Кодген: api-types.ts перегенерён (dump OpenAPI → openapi-typescript →
project-local prettier 3.9.0); 2-й прогон без диффа.
Фронтенд:
- useHouseTypes() (TanStack useQuery, staleTime Infinity) в concept-api.ts;
building_program в ConceptInput, placed_count/requested_count в ConceptVariant.
- HouseProgramPicker: toggle «Авто (max-FAR)» (default, program omit → greedy)
vs «Выбрать дома» (список каталожных типов, count 1-50 / floors 1-40, дефолт
из каталога; габариты/этажность/класс как подсказка). Смонтирован в
Section7Concept и на странице /concept.
- Partial-fit заметка в ConceptVariantsResult: при placed<requested честное
«Разместилось N из M секций — участок вмещает меньше» (нейтрально, не ошибка).
The old test_no_program_reproduces_greedy_output_unchanged only compared
building_program=None (default) vs explicit None — both through the new
_Placer code — so it proved the two None branches agree but did NOT pin
the greedy geometry; it would still pass if the _Placer extraction had
drifted the output. test_placement.py only checks invariants, never
concrete counts/TEAP, so there was no anti-regression guard that the
greedy path is byte-identical after the refactor.
Replace it with two tests:
- test_greedy_output_matches_golden_pin: hard-coded literals per strategy
on the fixed _BIG_PARCEL — (features, built_area_sqm, total_floor_area_sqm,
apartments_count) — frozen from the current (== pre-refactor) output, so
any future deterministic drift in greedy placement FAILS.
- test_explicit_none_program_equals_default_greedy: keeps the None-branch
equivalence check (default vs explicit None go one greedy path).
Extend the concept generator so ConceptInput can carry an optional
building_program (list of typed houses from a catalog). When present,
placement lays out EXACTLY that program — for each item, place `count`
sections of the catalog footprint at the item's floors — instead of the
greedy max-FAR coverage-cap sweep. When absent, the existing greedy
behavior is unchanged (byte-for-byte backward-compatible).
- catalog.py: hardcoded HOUSE_TYPES (panel_econom, monolith_comfort,
tower_business, lowrise_comfort, townhouse) — sane-default catalog,
promote to DB later; get_house_type / available_section_types lookups.
- schema: additive BuildingProgramItem {section_type, floors, count} and
ConceptInput.building_program (default None -> greedy). ConceptVariant
gains optional placed_count / requested_count (partial-fit signal).
- placement: shared _Placer (collision/STRtree/setback machine extracted
from greedy sweep, reused — no duplication); place_program +
place_program_variant; branch in place_all_strategies on
building_program. Mixed-floor TEAP via exact per-floor-group aggregation
(GFA = sum(area_i * floors_i), no rounding drift).
- partial fit: when the parcel can't fit all sections, place as many as
fit and report placed_count < requested_count (no hard-422); zero-fit
still raises ParcelGeometryError (-> 422).
- API: validate program section_type keys against the catalog (unknown ->
422) before placement.
- tests: catalog integrity, greedy backward-compat, exact 2-item program +
TEAP reflection, over-packed partial placement, API program path.
- regenerate frontend api-types.ts (OpenAPI codegen gate stays green).
Code-review follow-up: /recompute hardcoded price_source="objective_district_median"
for any body-supplied market_price_per_sqm, mislabeling the honesty-flag once
Stage 2b forwards a price whose genuine source differs (objective_geo_radius /
district_reference / class_norm from financial_estimate).
- schemas/concept.py: add optional price_source: str | None to MassingProgram.
- api/v1/concepts.py: on FAST path use payload.price_source if provided, else the
default label (now a module-level constant _DEFAULT_PRERESOLVED_SOURCE). DB-fallback
and class-norm paths keep their own resolved source unchanged.
- tests: assert body-provided price_source echoes through to financial.price_source
(not overwritten), and the default label applies when the front omits it.
Stage 2a of epic #1953: backend service + endpoint for live economic recompute
driven by the interactive 3D massing (Stage 2b debounced sliders).
- teap.py: add pure synthesize_teap_from_program(total_footprint_sqm, floors,
site_area_sqm, housing_class, sections) — builds a TEAP from the SCALAR
aggregate footprint × floors, mirroring synthesize_teap_from_buildability and
reusing the same shared norm constants (_OFFICE_SHARE_OF_GFA / _EFFICIENCY_BY_CLASS
/ _AVG_APARTMENT_SQM / _PARKING_PER_APARTMENT) — single source of truth.
- schemas/concept.py: add MassingProgram (program contract, optional pre-resolved
market_price_per_sqm + parcel_centroid_wkt) and MassingRecomputeOutput (teap + financial).
- api/v1/concepts.py: add POST /api/v1/concepts/recompute — synthesize TEAP → run
the existing pure compute_financial. FAST path uses body market_price_per_sqm
(no DB); else _lookup_market_price by centroid via run_in_threadpool; else class norm.
- tests: synthesize_teap_from_program (gfa math, parity with compute_teap, class
efficiency, sections no-op) + endpoint (200, coherent output, price passthrough
skips DB, DB fallback, class-norm default, floors validation).
_SUPPLY_BATCH_SQL джойнил domrf_kn_flats по ОДНОЙ глобальной дате
(f.snapshot_date = MAX(snapshot_date) по всей таблице). Но domrf_kn_flats —
ПО-ОБЪЕКТНЫЙ time-series: каждый ЖК скрейпится в свой день. На единственной
глобал-max дате присутствует обычно 1 объект → у остальных 0 квартир →
supply_units_in_radius=0 для всех строк 4.2 Планировки → frontend показывал
«Срок продажи 0 мес» и «% продано —». Регрессия от #1944 (objects-first
дедуп snapshot'ов объектов, который сам по себе корректен).
Фикс: flats_latest CTE (DISTINCT ON (obj_id) ... ORDER BY obj_id,
snapshot_date DESC, id DESC) берёт для КАЖДОГО obj_id его собственный
последний снимок и джойнится к nearby. objects-first MATERIALIZED дедуп
(#1944) сохранён → fan-out по снимкам не возвращается. Глобальный
db.scalar(MAX(snapshot_date)) + :latest_snap bind удалены.
Прод (66:41:0205010:287, r=1км, 9 объектов): supply 0 (global-max) → 2675
(per-object, 4 объекта имеют flats на разных датах 2026-05-17/05-05; ни один
не на глобал-max 2026-06-22). Данные flats частично сломаны (#1945, отдельно),
но фикс корректно двигает supply с 0 к реальным per-object числам.
Тесты: новый guard test_supply_joins_flats_per_object_latest_snapshot;
обновлены mock-фабрики (db.scalar больше не вызывается).
Корень «−1.00 везде» (эпик #1953): compute_demand_supply_forecast брал
district-wide unit_velocity (847.5/мес, ВСЕ классы/комнаты) как спрос и
весь district-сток (~63k доступных) как предложение для КАЖДОЙ ячейки
what_to_build → один и тот же ratio во всех ячейках → все deficit_index
прижаты к −1.0. Плюс objective_lots — append-per-snapshot (~2.9× инфляция
строк), что симметрично раздувало обе базы → даже сегментация без дедупа
осталась бы вырожденной.
Фикс (blast radius — ТОЛЬКО forecast/deficit calc; platform-wide dedup = #1964):
- market_metrics.compute_market_metrics: +obj_class/+room_bucket (+cache key).
_STOCK_SQL и _SALES_WINDOW_SQL дедуплят до ПОСЛЕДНЕГО снапшота на физлот
(DISTINCT ON project_name,corpus_name,section,floor,lot_number ORDER BY …
snapshot_date DESC,id DESC), затем агрегируют. Class-фильтр (LOWER=LOWER,
class lowercase) + room-bucket (Source-B room_area-вокабуляр, зеркало
sales_series.room_area_bucket_of → what_to_build фильтрует без перевода).
ROLLUP/GROUPING сохранён; confidence считается на дедуплицированных counts.
- demand_supply_forecast: base_pace и open-сток теперь ПОСЕГМЕНТНЫЕ
(market_metrics(obj_class,room_bucket)). При заданном сегменте L2/L3
(hidden/future) ИСКЛЮЧЕНЫ из баланса — они класс/формат-агностичны, иначе
двоились бы по всем ячейкам. +_market_room_bucket VOCAB-мост (валидирующий
pass-through Source-B меток; неизвестное → None = без фильтра, не тихий 0-rows).
- what_to_build/_DEFAULT_CLASSES и recommendation Economy-маппинг: «эконом»→
«стандарт» (в objective_lots эконома НЕТ, стандарт=483k → раньше ячейка
матчила 0 строк и молча выпадала).
- report_assembler honesty-guard: если ВСЯ сетка прижата к ±1.0
(degenerate-fallback) — не эмитим «строить»/«избегать», показываем
«недостаточно гранулярных данных для посегментного вывода».
- data/sql/173_objective_lots_physflat_idx.sql: partial index под DISTINCT ON
(Index Only Scan + Unique, без Sort на 1.75M строк; idempotent, BEGIN/COMMIT).
Prod-verify (parcel 66:41:0205010:287, Железнодорожный, h=24): ячейки
ДИФФЕРЕНЦИРУЮТ (12 measured, 7 distinct) вместо all −1.0; MOI комфорт/студия
38.5 vs стандарт/студия 244.3 (точное совпадение с ожидаемым).
Тесты: регрессия «ячейки различаются (не all −1.0)» + vocab-translation +
honesty-guard + посегментное предложение. ruff clean; no :name::type.
Replace dev-jargon with plain RU across the §22 forecast report (6.2/6.3/6.5
+ deficit/затоварка legend). Source of truth = backend reason strings + the
frontend RU maps; text/render-only, no scoring/forecast math touched.
Backend (source of truth):
- scenarios.py _COLLAPSE_REASON_LOW_BETA: «β rate-sensitivity не прошёл gate …»
→ «чувствительность к ставке не оценена на коротком ряде ЕКБ → один базовый
сценарий вместо трёх».
- confidence_engine.py _coverage_factor: drop «domrf↔objective» jargon, say it
affects будущее предложение/конкуренцию. New _history_factor: «глубина истории
N мес» + на что влияет + связь с 6.2 (короткий ряд → один сценарий).
Frontend (both Section-6 families — live analysis page + ptica cockpit):
- Deficit legend: −1 затоварка / 0 баланс / +1 острый дефицит + actionable
трактовка; MOI tied to «сколько месяцев район распродаёт предложение».
- 6.2 heading «Почему один сценарий, а не три» over the collapse reason.
- 6.3 render confidence.rationale + weakest-link rule («скорее завысим
недоверие, чем недооценим риск»); FACTOR_RU gains confounded_window/
advisory_cap; factor notes shown.
- 6.5 «Вес»→«Оценка»; overall verdict vs 0.5; «Риск избытка предложения»→
«Запас по предложению»; §-refs moved from reason into tooltip; 6 special-
index 1-line «что это + куда лучше» descriptions; 0.00-score reasons shown.
Tests: confidence_engine (history/coverage notes), stripSectionRefs vitest.
Refs #1963
Миграция 172 убрала english 'Comfort' из v_bucket_success_score (NULL-class
→ 'не указан'), но _bucket_success_ranking в analytics_queries.py матчил
COALESCE(:cls, 'Comfort') — после 172 ни одна строка не равна 'Comfort' →
recommend_mix success-boost (#25, /analytics/recommend квартирография) тихо
возвращал [] (silent degradation). LIVE на проде с момента ручного применения
172 (graceful empty, без краша).
Fix:
- call site (recommend_mix): передаём target_class_db (уже переведённый через
_class_to_db_vocab english→русский), а не сырой english target_class.
Чинит и латентный pre-existing баг: Business/Elite никогда не матчили
русский источник.
- SQL default 'Comfort' → 'Комфорт' (массовый класс ЕКБ, 723 объекта).
NULL-class ('не указан') в default-путь намеренно не попадают
(документировано в docstring).
Prod-verified: OLD 'Comfort' default → 0 строк; NEW 'Комфорт' default → 5.
Tests:
- _bucket_success_ranking: реальный запрос (не замокан) с русским default —
SQL содержит 'Комфорт', не 'Comfort'; ranking непустой.
- recommend_mix: english 'Business' переводится в 'Бизнес' перед ranking
(раньше тест патчил _bucket_success_ranking→[] и баг не ловил;
helper теперь умеет patch_success_ranking=False).
- #1960: позитивный тест выбора quarter_rosreestr basis (deals≥5, все
higher fallback'и None) → median_price_basis='quarter_rosreestr'.
FIX A (#1955) «Что хорошо продаётся»: убран фантомный класс 'Comfort'.
- Миграция 172: v_bucket_success_score COALESCE(obj_class,'Comfort')
→ COALESCE(obj_class,'не указан'). Английский литерал заполнял 397 NULL
и сливался отдельным классом от русского 'Комфорт' → визуальные дубли
бакетов в UI. Источник уже канонически-русский (проверено на проде),
synonym-mapping не нужен.
- parcels.py: obj_class протаскивается в success-ranking query + dict.
- TS SuccessRankingBucket.obj_class добавлен.
FIX B (#1960) «Медиана рынка» = 64k (квартальная росреестровская n=1 ДКП):
- district.median_price_per_m2 больше не COALESCE(median_12m, ekb_ref) в SQL.
Basis-приоритет (newbuild-first): Objective по имени района →
geo_radius (Objective в 3км) → ekb_districts reference →
квартальная росреестровская медиана ТОЛЬКО при deals_count≥5.
Для 66:41:0205010:287: 64k → 132690 (geo_radius, newbuild-consistent).
- median_price_basis добавлен в payload + TS type (nullable median).
- Frontend null-guards для нового nullable median.
Tests: +4 (geo_radius basis, objective-приоритет, deals-guard, obj_class
passthrough); обновлены district-моки в 9 analyze-тестах под новую
SQL-сигнатуру.
Code-review nits (без изменения поведения):
- quarter_dump_lookup.py: комментарий про дедуп ошибочно утверждал, что зоны без
рег-номера не схлопываются. На деле reg_numb_border — TEXT NOT NULL (0 NULL на
проде), а DISTINCT ON в PG считает NULL равными. Переформулировано на точное.
- frontend/types/nspd.ts: убран устаревший 'cad_zouit' из списка возможных
group_key (бэкенд теперь эмитит только protected/engineering/okn/natural/other).
#1954 — площадь «—»: COALESCE(land_record_area, specified_area, declared_area)
в EGRN-блоке analyze (land_record_area NULL у 12809/42233 участков, но
specified_area заполнена). area_m2 для 66:41:0205010:287 теперь 106378 (был NULL).
#1954 — «Обновлено» сломано: cost_registration_date — мёртвая колонка (0/42233);
репойнт на updated_at (42233/42233 заполнено). Ключ ответа last_egrn_update_date
не меняется (additive value fix).
#1958 — confidence-фактор «Прогноз спрос/предложение» дублировался ×4
(по фактору на горизонт). Сворачиваем в один weakest-link'ом (MIN ранга
по горизонтам) в _component_confidences до confidence-движка (#990).
#1957 — ЗОУИТ backend:
- _get_cad_zouit_overlaps: DISTINCT ON (reg_numb_border) — дедуп дубль-строк
(одна физ. зона 2× с разным category_name). group_key cad_zouit→protected.
- _get_zouit_overlaps (dump path): subcategory→RU-тип карта (26→СЗЗ и др.,
коды сверены кросс-джойном dump↔cad_zouit на проде); type_zone из карты,
reg из props.options.reg_numb_border. Раньше отдавал blank-строки.
- унификация group_key (protected/engineering/okn/natural/other) + top-level
reg_numb_border в обоих путях.
UP038-модернизация isinstance в report_assembler (pre-commit ruff 0.7.4).
Frontend note (#1957): nspd_zouit_overlaps теперь всегда group_key из набора
{protected,engineering,okn,natural,other} — сырой 'cad_zouit' больше не отдаётся;
оба пути несут type_zone + reg_numb_border.
Tests: +4 _component_confidences collapse, +6 ЗОУИТ (subcategory map, dump
typing, DISTINCT ON), schema-test обновлён на protected. 451 passed targeted.
The data-freshness monitor classified by run RECENCY only, so the domrf_kn
FLATS loader running status=done but extracting 0 flats for ~5 weeks went
undetected — and the kn source watched objects_count (healthy ~1548), not
flats_count (the broken =0 metric).
Add an opt-in zero-output check: an otherwise-fresh run-ledger source (recent
success, would-be fresh by age) that produced 0 work-rows in the 7d window is
downgraded to status="failed" (so scrape_freshness_check alerts), with an
additive "reason". Guards: alert_on_zero_output flag, run-ledger only
(timestamp_col is None), status=="ok" (age-stale/failed already covered), and
upd_7d==0 (SUM of the source's own work_col over done-runs).
Registry: new kn_flats source (kn_scrape_runs, work_col=flats_count, critical,
flag on) — watches the column that was broken; existing kn (objects_count)
unchanged. Flag also enabled on objective (rows_lots, critical). nspd/nspd_geo/
cadastre left unflagged (legitimate-0 / data-table).
JSON additive only (new nullable "reason" key; endpoint is dict[str,Any], no
frontend consumer / no codegen needed). 4 new tests (downgrade, no-false-
positive, age-precedence, registry). code-reviewer APPROVE.
Would have caught #1945 within ~8-14d instead of 5 weeks.
domrf_kn_objects is a snapshot dimension (UNIQUE (obj_id, snapshot_date), ~8
snapshots/obj_id). _SUPPLY_BATCH_SQL joined flats to ALL object-snapshot rows
(no o.snapshot_date filter), counting each flat ~8.5x → supply_units_in_radius
inflated ~8.5x, sold_pct_of_supply deflated ~8.5x, is_oversold under-fired
(all user-facing, best_layouts.py:571-611; sold_pct=deals/supply is a raw
ratio so no canceling).
Fix: dedup objects to one row per obj_id (latest-snapshot coords) via
DISTINCT ON in an objects-first MATERIALIZED CTE, then join domrf_kn_flats via
idx_kn_flats_obj. units now = one count per flat (prod cross-check at radius
1.5km: units == count(*) == count(DISTINCT f.id) == 9612 for 65 objects;
correction factor 8.56x at 1.5km, 9.13x at 1.0km). This also aligns the supply
denominator with the deals numerator (_COMPETITORS_IN_RADIUS_SQL already uses
DISTINCT ON latest snapshot).
Perf bonus: objects-first avoids the parallel seq scan of the ~376k-row flats
snapshot. radius 1.5km / snapshot 2026-05-17: 240ms/~28k buffers/6712 disk
reads -> 49ms/1554 buffers/0 disk reads (~5x).
Tests: add SQL-text fan-out guard (DISTINCT ON + MATERIALIZED, no bare
flats->objects join); update stale EXPLAIN mirror in test_phantom_columns.
USER-FACING: best-layouts supply/sold_pct/is_oversold/sell-out-months shift
~8.5x toward correct (frontend BestLayoutsBlock only; ТЗ recommendation + PDF
unchanged — they derive from sum_deals, not supply). Deep-reviewed (APPROVE).
The objective_corpus_room_month index migration (#1942) was committed to
backend/data/sql/171_*, but the main deploy runner applies migrations from
repo-root data/sql/*.sql (deploy.yml:280) and the path trigger is data/sql/**.
So the file never ran — prod still seq-scans (verified post-deploy: index
absent, query 265ms). Move it to data/sql/171_* (alongside 170) so it deploys.
No SQL change; the index itself was dry-run-verified on prod (BEGIN/ROLLBACK).
_get_ekb_median() (velocity.py:706) runs on EVERY POST /parcels/{cad}/analyze
(the hottest endpoint) and seq-scanned the whole objective_corpus_room_month
(~95MB, ~12159 buffers, 144ms) — its predicates (report_month >= now-6mo AND
deals_total_count > 0) had no usable index (the 5 existing report_month indexes
aren't partial on deals_total_count; a bare range matches 27% of rows, so the
planner correctly chose Seq Scan).
Add partial b-tree (report_month) WHERE deals_total_count > 0 (~280kB, 8.9%
selectivity). Prod EXPLAIN (BEGIN/ROLLBACK): 144ms→38ms (~3.8x), buffers
12281→3136 (-74%); planner uses it naturally (Index/Bitmap scan). Independently
dry-run-verified: Index Only Scan, 2747 buffers.
Write cost negligible (objective_corpus_room_month written only by weekly ETL,
not request-path). Idempotent (IF NOT EXISTS); plain CREATE INDEX (not
CONCURRENTLY, can't run in the migration's BEGIN/COMMIT) — sub-second build,
SHARE lock blocks only the weekly ETL writer, not analyze readers.
Found via pg_stat_user_tables seq-scan audit + database-expert EXPLAIN analysis.
App-level logs (logging.getLogger("app.*")) had no StreamHandler, so their
INFO was lost — `docker logs gendesign-backend-1` showed only uvicorn loggers.
Useful lines (e.g. "OSRM road-distance applied", RBAC, analyze) were invisible
when debugging on the VPS (only Sentry/GlitchTip received them as breadcrumbs).
main.py now attaches one StreamHandler to the root "app" logger at
APP_LOG_LEVEL (default INFO), idempotent (marker guard), propagate=True so the
Sentry LoggingIntegration on root still gets breadcrumbs/events and there's no
stdout double (root has no stdout handler). API process only — celery worker
(celery_app.py) untouched, since its loaders log per-sync and would be noisy.
Flood-safe on the API: the analyze hot path logs INFO per-request (parcels.py
has 3 logger.info), not per-row.
Adds tests/test_app_logging.py (handler present, level INFO, propagate intact).
Refs #1926
Walk-relevant POIs (school/shop/park/kindergarten/pharmacy/stops) now route
via a FOOT OSRM graph (osrm-walk service), car-relevant POIs (mall/hospital +
unknown) keep the DRIVING graph. Validation showed driving overstated
pedestrian-proximity distance — median 1.6-2.9x straight-line (#39).
- config: osrm_walk_local_url + osrm_walk_categories (frozenset, 9 walk cats)
- osrm_client_local: base_url override on get_road_distances_m (default unchanged)
- _apply_osrm_road_distances: split POIs by category, per-group OSRM call with
independent graceful fallback (one server down -> its group keeps straight-line),
in-place write-back by original index; never raises; flag-OFF byte-identical
- docker-compose: osrm-walk service (foot graph, internal, mem_limit 1.5g)
- build_osrm.sh: CAR=0 gate for foot-only refresh (doesn't touch live car graph)
- tests: per-category split, per-group fallback, asymmetric intra-group write-back
Still flag-gated (use_osrm_distances OFF) — enabling is a product decision.
Refs #39
kn (DOM.РФ ЖК) runs weekly (Mon cron) but thresholds were 2/5 → false "stale" every Wed-Sun. Recalibrate to 8/14 (weekly, mirrors objective) + regression test. Surfaced by the freshness monitor's first prod run. Refs #73
smr/sale_inflation_annual in _build_monthly_cashflow, gated by module constants =0.0 → byte-identical zero-diff (deep-review verified IEEE-754 identity + all 57 prior tests unchanged). Land not escalated; marketing/VAT/tax retimed on escalated revenue (total preserved). Capability wired+tested, OFF until calibrated indices + invariant reframe. No schema/codegen. Refs #1881
