' + "Оценка построена по зарегистрированным сделкам Росреестра — подходящих " + "объявлений поблизости почти нет. Объявления ниже приведены справочно, " + "для наглядности рынка.
" + ) + + +def _reliability_note_html(estimate: AggregatedEstimate, n_shown: int) -> str: + """#pdf-honesty: surfaces `AggregatedEstimate.relaxations`/`reliability` + (estimator.py #oblast-F cascade + #oblast-E deals-priority) — the web report + already shows this (LowConfidenceBanner); the PDF stayed silent, a + client-visible discrepancy between the two. Empty string (no-op) when + reliability=='ok' and relaxations is empty — the common, unrelaxed case, + byte-identical to the report before these fields existed.""" + if estimate.reliability == "ok" and not estimate.relaxations: + return "" + if estimate.relaxations: + detail = "Подбор аналогов расширен: " + ", ".join( + _html.escape(r) for r in estimate.relaxations + ) + else: + # relaxations пуст, но reliability всё же не 'ok' (напр. тонкая выборка, + # которую каскад ослаблений не смог расширить, см. estimator.py + # #oblast-F) — n_shown, не сырой n_analogs (та же #pdf-honesty логика, + # что и в счётчике выше страницы). + detail = f"Оценка построена по небольшой выборке ({n_shown} шт.)" + return f""" +Диапазон цен в объявлениях
diff --git a/tradein-mvp/backend/tests/test_estimator_headline_sufficiency.py b/tradein-mvp/backend/tests/test_estimator_headline_sufficiency.py index 3ae13ef7..a4033a0f 100644 --- a/tradein-mvp/backend/tests/test_estimator_headline_sufficiency.py +++ b/tradein-mvp/backend/tests/test_estimator_headline_sufficiency.py @@ -1,22 +1,55 @@ -"""#oblast-E — headline sufficiency gate (money-path audit, 2026-08-02). +"""#oblast-E — headline sufficiency gate (money-path audit, 2026-08-02, priority +RESTORED 2026-08-10) + #oblast-F — never-block relaxation cascade (product +decision, 2026-08-10). -Live-prod repro that motivated this gate: Серов 2к/45м², n=3 scraped listings → -headline 42 391 ₽/м² (−36% vs the city ДКП corridor, 54 126 ₽/м²); a neighbouring -street in the same town swung ±66% on 1-2 different random listings. Каменск- -Уральский returned a LITERAL 0 ₽ for a room/area combo with no local ДКП match -either, with no honest refusal surfaced. Первоуральск (0 listings) already fell -back to the (pre-existing) ДКП deals-headline fallback correctly — this gate -routes the THIN (1..HEADLINE_LISTINGS_MIN_N-1 listings) case into that SAME, -already-tested path instead of trusting a 1-4-lot median as the headline. +History: + 1. #oblast-E (2026-08-02) SUPPRESSED a thin (1..HEADLINE_LISTINGS_MIN_N-1) + listings sample to a literal zero, forcing the anchor/#oblast-D-deals- + fallback/insufficient_data chain to take over — motivated by a live + Серов repro (n=3 → 42 391 ₽/м², −36% vs the town's ДКП corridor of + 54 126 ₽/м²). + 2. #oblast-F (2026-08-10, first pass) reversed that suppression WHOLESALE — + a thin sample always kept its own median, even when a much more reliable + deals corridor was available. That accidentally REOPENED the exact Серов + bug #oblast-E existed to close. + 3. #oblast-E priority RESTORED (2026-08-10, same day, product correction): + "никогда не блокировать вывод" ≠ "предпочитать шумную медиану по 3 + объявлениям надёжному коридору по 54 сделкам". Final 3-way rule, in + `_price_from_inputs`'s gate: + - n_analogs >= HEADLINE_LISTINGS_MIN_N → listings median (unaffected). + - 0 < n_analogs < HEADLINE_LISTINGS_MIN_N AND a usable ДКП corridor + exists (count >= DEALS_HEADLINE_FALLBACK_MIN_N, median_ppm2 > 0) → + listings aggregate suppressed to zero, headline ceded to the + #oblast-D deals-headline-fallback chain (original #oblast-E + behaviour, restored). `PricingResult.deals_headline_due_to_thin_ + listings=True` — estimate_quality() adds relaxation label "оценка по + сделкам — мало объявлений рядом" and caps reliability at 'low'. + Listings display cards are NOT hidden (unlike original #oblast-E) — + `listings_clean` stays intact and estimate_quality() still surfaces + them as context even though they no longer drive n_analogs/median. + - 0 < n_analogs < HEADLINE_LISTINGS_MIN_N AND no usable ДКП corridor → + #oblast-F: keep the real thin median (never refuse outright). + Real refusal ("недостаточно данных") now happens ONLY at genuine n=0 + (no listings AND no usable anchor/deals) — the never-block requirement + with an honest, priority-ordered source selection. + +`estimate_quality()` tries to grow a thin sample FIRST via the #oblast-F +relaxation cascade (room-adjacency / freshness / novostroyki / radius, see +estimator.py module docstring) BEFORE `_price_from_inputs` (tested here in +Layer 1) ever runs the 3-way gate above — `listings` here is whatever that +cascade could find. Two layers: 1. `_price_from_inputs` unit tests (no DB, no estimate_quality overhead) — - boundary behaviour of the gate itself. + boundary behaviour of the gate itself: the 3-way rule, low-reliability + wording, listings_clean/listings_headline_thin_n/deals_headline_due_to_ + thin_listings bookkeeping. 2. `estimate_quality` integration tests — proves the money-path invariants - that matter to a caller: literal 0 never leaks as a "confident" price, - display `analogs` cards never outnumber what `n_analogs` claims, and the - explanation text describes what actually happened (not a stock "аналогов - не найдено" when some WERE found, just too few). + that matter to a caller: thin+usable-deals routes to the deals corridor + (Серов repro), thin+no-deals keeps its own median, display `analogs` + cards are shown either way, and the #oblast-F room-adjacency relaxation + (studio↔1-комн) actually grows a thin sample and is reported via + `AggregatedEstimate.relaxations` / `reliability`. """ from __future__ import annotations @@ -103,15 +136,14 @@ def test_threshold_is_five_not_lower() -> None: assert HEADLINE_LISTINGS_MIN_N == 5 -def test_four_listings_below_threshold_suppressed_no_fallback() -> None: - """n=4 (< 5), no ДКП signal → headline suppressed to the honest zero state, - NOT the naive median of 4 listings.""" +def test_four_listings_below_threshold_kept_not_suppressed() -> None: + """#oblast-F: n=4 (< 5) → the REAL 4-listing median is kept (product decision + 2026-08-10 — never zero out a thin-but-real sample), just flagged low.""" pr = _call(listings=_lots([200_000.0, 210_000.0, 220_000.0, 230_000.0])) - assert pr.median_ppm2 == 0.0 - assert pr.median_price == 0 - assert pr.n_analogs == 0 - assert pr.range_low == 0 - assert pr.range_high == 0 + assert pr.median_ppm2 == 215_000.0 + assert pr.n_analogs == 4 + assert pr.median_price == round(215_000.0 * 45.0) + assert pr.confidence == "low" def test_five_listings_at_threshold_not_suppressed() -> None: @@ -122,18 +154,21 @@ def test_five_listings_at_threshold_not_suppressed() -> None: assert pr.median_price == round(210_000.0 * 45.0) -def test_one_listing_below_threshold_suppressed() -> None: - """n=1 — the sharpest form of the Серов bug (a single random lot deciding - the whole headline) — must be suppressed exactly like n=4.""" +def test_one_listing_below_threshold_kept_not_suppressed() -> None: + """#oblast-F: n=1 — the sharpest thin case — still keeps its own (single-lot) + median rather than being zeroed; confidence stays 'low'.""" pr = _call(listings=_lots([200_000.0])) - assert pr.median_ppm2 == 0.0 - assert pr.n_analogs == 0 + assert pr.median_ppm2 == 200_000.0 + assert pr.n_analogs == 1 + assert pr.confidence == "low" def test_thin_sample_with_sufficient_deals_uses_deals_headline() -> None: - """n=3 listings (thin) + a usable ДКП corridor → headline comes from the - deal corridor median, NOT the 3-listing median (live Серов repro: 3 - listings gave 42 391 vs the honest ДКП-based ~54 126).""" + """#oblast-E priority RESTORED (2026-08-10 product correction): a thin + (n=3) listings sample must NOT outrank a usable ДКП deals corridor — this + is the exact live Серов repro #oblast-E exists for (3 noisy listings gave + 42 391 ₽/м², the honest 54-deal corridor gives 65 957 ₽/м²). Headline + comes from the deal corridor median, NOT the 3-listing median.""" dkp_raw = { "count": 54, "low_ppm2": 44_000, @@ -151,12 +186,23 @@ def test_thin_sample_with_sufficient_deals_uses_deals_headline() -> None: ) assert pr.n_analogs == 0, "honest: 0 scraped-listing analogs back this headline" assert pr.confidence == "low" + assert pr.deals_headline_due_to_thin_listings is True + assert pr.listings_clean, "listings_clean must stay intact — display cards still show them" + # #4: explanation must not falsely claim "рядом нет объявлений" (some WERE + # found, just ceded priority to the more reliable deals corridor) and must + # NOT also carry the separate "Оценка построена по N аналогам" thin-kept + # wording (that phrasing is reserved for the no-usable-corridor branch). + assert pr.explanation is not None + assert "рядом нет актуальных объявлений" not in pr.explanation.lower() + assert "сделкам росреестра" in pr.explanation.lower() + assert "оценка построена по 3" not in pr.explanation.lower() -def test_thin_sample_with_insufficient_deals_stays_zero() -> None: +def test_thin_sample_with_thin_deals_also_uses_real_listings_median() -> None: """n=3 listings (thin) + a ДКП corridor that is ITSELF too thin - (< DEALS_HEADLINE_FALLBACK_MIN_N) → neither source is trusted; honest zero, - not a fabricated number from either side.""" + (< DEALS_HEADLINE_FALLBACK_MIN_N) → the corridor is NOT usable, so + #oblast-F's never-block rule applies: the real listings median is kept + rather than refusing (neither source alone would justify a hard zero).""" dkp_raw = { "count": 1, "low_ppm2": 40_000, @@ -165,49 +211,51 @@ def test_thin_sample_with_insufficient_deals_stays_zero() -> None: "period_months": 12, } pr = _call(listings=_lots([42_391.0, 26_818.0, 75_058.0]), dkp_raw=dkp_raw) - assert pr.median_ppm2 == 0.0 - assert pr.median_price == 0 - assert pr.n_analogs == 0 + assert pr.median_ppm2 == 42_391.0 + assert pr.n_analogs == 3 + assert pr.deals_headline_due_to_thin_listings is False -def test_thin_sample_explanation_is_honest_about_count() -> None: - """The explanation for a thin-but-nonzero sample must say HOW MANY listings - were found (not the generic 'ничего не найдено' text used for a genuine - zero-listing case) — #4 in the task: explanation must match reality.""" +def test_thin_sample_explanation_is_honest_about_low_accuracy() -> None: + """#4 (task spec): the explanation for a thin-but-real sample must read as + "small sample, lower accuracy" — NOT the old refusal-flavoured "минимум для + оценки по рынку" copy, and NOT the generic zero-analogs text.""" pr = _call(listings=_lots([200_000.0, 210_000.0])) # n=2 assert pr.explanation is not None assert "2" in pr.explanation - assert "недостаточно" in pr.explanation.lower() - # Must NOT reuse the "nothing found at all" copy — 2 listings WERE found. + assert "выборка мала" in pr.explanation.lower() + assert "точность снижена" in pr.explanation.lower() + assert "минимум для оценки по рынку" not in pr.explanation.lower() assert "не найдено аналогов" not in pr.explanation.lower() -def test_thin_sample_deals_fallback_explanation_does_not_claim_zero_listings() -> None: - """#4: once the ДКП fallback fires for a thin (not zero) sample, the - explanation must not falsely claim 'рядом нет объявлений' — some WERE - found, just not enough to trust.""" +def test_zero_listings_with_sufficient_deals_still_uses_deals_headline() -> None: + """Control: the #oblast-D deals-headline-fallback path is UNCHANGED for + GENUINELY zero listings (n=0) — #oblast-F only affects the 1..N-1 thin + case, not the true-zero case, which still needs a fallback source.""" dkp_raw = { - "count": 20, - "low_ppm2": 40_000, - "median_ppm2": 60_000, - "high_ppm2": 80_000, + "count": 54, + "low_ppm2": 44_000, + "median_ppm2": 65_957, + "high_ppm2": 89_000, "period_months": 12, } - pr = _call(listings=_lots([200_000.0, 210_000.0, 220_000.0]), dkp_raw=dkp_raw) + pr = _call(listings=[], dkp_raw=dkp_raw) + assert pr.median_ppm2 == 65_957.0 + assert pr.n_analogs == 0 + assert pr.confidence == "low" assert pr.explanation is not None - assert "рядом нет актуальных объявлений" not in pr.explanation.lower() + assert "рядом нет актуальных объявлений" in pr.explanation.lower() assert "сделкам росреестра" in pr.explanation.lower() -def test_thin_sample_listings_clean_preserved_for_anchor_ghost_guard() -> None: - """Regression guard: the gate must suppress the AGGREGATE (median/n_analogs) - without clearing `listings_clean` itself — the same-building anchor's own - ghost-anchor guard (#1871) reads `listings_clean` truthiness to tell - "genuinely zero nearby listings" from "some nearby, just too few to trust - as headline", and conflating the two was caught regressing - test_estimator_split_corridor_1871.py during this change.""" +def test_thin_sample_listings_clean_preserved_and_thin_n_still_tracked() -> None: + """listings_clean stays intact (unchanged invariant — same-building anchor's + ghost-anchor guard #1871 depends on it) AND, post-#oblast-F, n_analogs is + the REAL count (not zeroed) while listings_headline_thin_n still marks the + sample as thin for the low-reliability note upstream.""" pr = _call(listings=_lots([200_000.0, 210_000.0, 220_000.0])) - assert pr.n_analogs == 0 + assert pr.n_analogs == 3 assert len(pr.listings_clean) == 3 assert pr.listings_headline_thin_n == 3 @@ -219,6 +267,42 @@ def test_sufficient_sample_listings_headline_thin_n_is_zero() -> None: assert pr.listings_headline_thin_n == 0 +def test_repair_coefficient_now_applies_to_thin_sample() -> None: + """#oblast-F: pre-#oblast-F, the repair-state coefficient was skipped for a + thin sample because the headline was already zeroed (applying it would be a + no-op). Now that the real median is kept, the coefficient must apply.""" + pr_no_repair = _call(listings=_lots([200_000.0, 210_000.0])) # n=2, thin + pr = _price_from_inputs( + listings=_lots([200_000.0, 210_000.0]), + area_m2=45.0, + rooms=2, + repair_state="excellent", + floor=5, + total_floors=9, + target_year=None, + analog_tier="W", + fallback_used=False, + area_widened=False, + anchor_comps=[], + anchor_tier_fetched=None, + dkp_raw=None, + imv_anchor=None, + imv_eval=None, + yandex_val_present=False, + cian_val_present=False, + ratio_resolver=lambda _appm2: (None, None), + quarter_index_lookup=lambda q: None, + quarter_indexes_lookup=lambda qs: {}, + target_house_cadnum=None, + dadata_coarse=False, + geo=_geo(), + dadata_qc_geo=None, + ) + assert ( + pr.median_price != pr_no_repair.median_price + ), "repair coefficient must be applied even for a thin (#oblast-E-flagged) sample" + + # ───────────────────────────────────────────────────────────────────────────── # Layer 2 — `estimate_quality` integration tests (full stub-patched I/O path) # ───────────────────────────────────────────────────────────────────────────── @@ -261,24 +345,31 @@ def _serov_payload() -> Any: def _run_estimate( *, - analogs: list[dict[str, Any]], + analogs: list[dict[str, Any]] | None = None, dkp_raw: dict[str, Any] | None, + fetch_analogs_side_effect: Any = None, + payload: Any = None, + geo: GeocodeResult | None = None, ) -> Any: from app.services.estimator import estimate_quality db = MagicMock() - payload = _serov_payload() + payload = payload or _serov_payload() + geo = geo or _geo() + + fetch_analogs_kwargs: dict[str, Any] = ( + {"side_effect": fetch_analogs_side_effect} + if fetch_analogs_side_effect is not None + else {"return_value": (list(analogs or []), False, "W")} + ) async def _run() -> Any: with ( - patch("app.services.estimator.geocode", new=AsyncMock(return_value=_geo())), + patch("app.services.estimator.geocode", new=AsyncMock(return_value=geo)), patch("app.services.estimator.dadata_clean_address", new=AsyncMock(return_value=None)), patch("app.services.estimator.match_house_readonly", return_value=None), patch("app.services.estimator.get_house_metadata", new=AsyncMock(return_value=None)), - patch( - "app.services.estimator._fetch_analogs", - return_value=(list(analogs), False, "W"), - ), + patch("app.services.estimator._fetch_analogs", **fetch_analogs_kwargs), patch("app.services.estimator._fetch_anchor_comps", return_value=([], None)), patch("app.services.estimator._fetch_deals", return_value=[]), patch( @@ -301,40 +392,47 @@ def _run_estimate( return anyio.run(_run) -def test_e2e_thin_no_deals_never_leaks_literal_zero_as_confident_price() -> None: - """Каменск-Уральский-style repro: thin listings, no usable ДКП corridor — - median_price_rub must be 0 AND insufficient_data must be True TOGETHER - (the AggregatedEstimate.insufficient_data computed_field invariant that - stops a literal 0 ₽ reaching the user as a confident number).""" +def test_e2e_thin_sample_no_relaxation_help_keeps_real_median() -> None: + """#oblast-F: 2 thin listings, no ДКП, and the mocked `_fetch_analogs` always + returns the SAME 2 listings regardless of relaxation params (none of them + help) — median_price_rub must be the REAL non-zero 2-listing median, + insufficient_data False, n_analogs=2, confidence='low', reliability + 'very_low' (n<3), relaxations empty (nothing actually helped).""" analogs = [ _make_listing(price_per_m2=200_000.0, address="ул. Ленина, 5"), _make_listing(price_per_m2=210_000.0, address="ул. Ленина, 7"), ] est = _run_estimate(analogs=analogs, dkp_raw=None) - assert est.median_price_rub == 0 - assert est.insufficient_data is True - assert est.n_analogs == 0 + assert est.median_price_rub == round(205_000.0 * 45.0) + assert est.insufficient_data is False + assert est.n_analogs == 2 assert est.confidence == "low" + assert est.relaxations == [] + assert est.reliability == "very_low" -def test_e2e_thin_sample_display_cards_never_outnumber_n_analogs() -> None: - """The 2 thin listings must NOT be surfaced as `analogs` display cards while - n_analogs reports 0 — that would be the same dishonesty (confident-looking - UI) this whole gate exists to remove.""" +def test_e2e_thin_sample_display_cards_match_n_analogs() -> None: + """#oblast-F: display `analogs` cards are NO LONGER suppressed for a thin + sample — they must match n_analogs exactly (both = 2), never hidden.""" analogs = [ _make_listing(price_per_m2=200_000.0, address="ул. Ленина, 5"), _make_listing(price_per_m2=210_000.0, address="ул. Ленина, 7"), ] est = _run_estimate(analogs=analogs, dkp_raw=None) - assert est.n_analogs == 0 - assert est.analogs == [] + assert est.n_analogs == 2 + assert len(est.analogs) == 2 def test_e2e_serov_repro_thin_sample_routes_to_deals_headline() -> None: - """Live Серов repro (n=3 scraped listings, wide ДКП corridor available): - headline must come from the deal corridor, not the noisy 3-listing median, - and the estimate must be honestly non-'insufficient' (a real number, low - confidence, deals-sourced).""" + """Live Серов repro (n=3 scraped listings, wide ДКП corridor available) — + #oblast-E priority RESTORED: headline must come from the deal corridor, + not the noisy 3-listing median. Also proves the #4 task-spec requirements + layered on top of the restored priority: the estimate is honestly non- + 'insufficient' (a real number, low confidence), reliability is capped at + 'low' (not 'very_low' — a 54-deal corridor is real signal), the + relaxation label names the source switch, AND the 3 thin listings are + still shown as display cards (not discarded) even though they no longer + drive n_analogs/median.""" analogs = [ _make_listing(price_per_m2=42_391.0, address="ул. Льва Толстого, 8А"), _make_listing(price_per_m2=26_818.0, address="ул. Кирова, 4"), @@ -354,12 +452,15 @@ def test_e2e_serov_repro_thin_sample_routes_to_deals_headline() -> None: assert est.confidence == "low" assert est.confidence_explanation is not None assert "сделкам росреестра" in est.confidence_explanation.lower() + assert est.reliability == "low", "a 54-deal corridor is real signal, not 'very_low'" + assert "оценка по сделкам — мало объявлений рядом" in est.relaxations + assert len(est.analogs) == 3, "thin listings must still surface as display cards" def test_e2e_sufficient_five_analogs_unaffected_control() -> None: """Control (mirrors the Екатеринбург prod check in the PR): a sample that clears the threshold is priced exactly as before — headline is the real - listings median, all 5 analogs counted.""" + listings median, all 5 analogs counted, no relaxations needed.""" analogs = [ _make_listing(price_per_m2=195_000.0, address="ул. Ленина, 5"), _make_listing(price_per_m2=205_000.0, address="ул. Ленина, 7"), @@ -371,3 +472,78 @@ def test_e2e_sufficient_five_analogs_unaffected_control() -> None: assert est.median_price_per_m2 == 210_000 assert est.n_analogs == 5 assert est.insufficient_data is False + assert est.relaxations == [] + assert est.reliability == "low" # n=5 falls in the 3..7 bucket + + +def test_e2e_rooms_relaxation_includes_studios_when_thin() -> None: + """#oblast-F step (a) — the exact scenario from the task spec: rooms=1 thin + sample (studio-adjacent building, live prod repro Академика Парина 46/5) → + cascade retries with rooms IN (0,1) and finds a trustworthy sample there. + Asserts: studios pulled in, `relaxations` names it, real non-zero median, + reliability downgraded to 'low' (thin base sample).""" + from app.schemas.trade_in import TradeInEstimateInput + + exact_rooms1 = [ + _make_listing(price_per_m2=150_000.0, address="ул. Парина, 1", area_m2=23.0), + _make_listing(price_per_m2=155_000.0, address="ул. Парина, 2", area_m2=23.0), + ] + studio_pool = [ + *exact_rooms1, + _make_listing(price_per_m2=140_000.0, address="ул. Парина, 3", area_m2=20.0), + _make_listing(price_per_m2=145_000.0, address="ул. Парина, 4", area_m2=21.0), + _make_listing(price_per_m2=148_000.0, address="ул. Парина, 5", area_m2=22.0), + ] + + def _fetch_analogs_stub(*_args: Any, **kwargs: Any) -> tuple[list[dict[str, Any]], bool, str]: + if kwargs.get("rooms_min") == 0 and kwargs.get("rooms_max") == 1: + return list(studio_pool), False, "W" + return list(exact_rooms1), False, "W" + + geo = GeocodeResult( + lat=56.838, + lon=60.595, + full_address="Свердловская обл., Екатеринбург, ул. Парина, 46/5", + provider="nominatim", + ) + payload = TradeInEstimateInput( + address="ЕКБ, ул. Парина, 46/5", + area_m2=23.1, + rooms=1, + ) + + est = _run_estimate( + dkp_raw=None, + fetch_analogs_side_effect=_fetch_analogs_stub, + payload=payload, + geo=geo, + ) + + assert "учтены студии" in est.relaxations + assert est.median_price_rub > 0 + assert est.reliability == "low" + assert est.n_analogs == 5 + + +def test_e2e_radius_relaxation_respects_explicit_user_radius() -> None: + """#oblast-F step (d) contract: when the user explicitly picked radius_m + (#2044), the cascade must NOT auto-expand past it — mirrors the existing + radius-fallback contract above (no auto-expansion beyond user's choice).""" + from app.schemas.trade_in import TradeInEstimateInput + + thin = [ + _make_listing(price_per_m2=200_000.0, address="ул. Ленина, 5"), + _make_listing(price_per_m2=210_000.0, address="ул. Ленина, 7"), + ] + payload = TradeInEstimateInput( + address="Серов, ул. Ленина, 5", + area_m2=45.0, + rooms=2, + floor=5, + total_floors=9, + city_hint="Серов", + radius_m=1500, + ) + est = _run_estimate(analogs=thin, dkp_raw=None, payload=payload) + assert not any("радиус расширен" in r for r in est.relaxations) + assert est.search_radius_m == 1500 diff --git a/tradein-mvp/backend/tests/test_pdf_security.py b/tradein-mvp/backend/tests/test_pdf_security.py index da5722ec..7d6d21ed 100644 --- a/tradein-mvp/backend/tests/test_pdf_security.py +++ b/tradein-mvp/backend/tests/test_pdf_security.py @@ -423,6 +423,86 @@ def test_build_listings_page_none_year_built_no_crash() -> None: assert "РЫНОК КВАРТИР" in html +# ── #pdf-honesty (#oblast-E deals-priority regression fix, 2026-08-10) ──────── +# n_analogs==0 (headline ceded to the ДКП deals corridor, estimator.py +# `deals_headline_due_to_thin_listings`) with a non-empty `analogs` display list +# (thin listings kept as reference cards) used to print "0 шт." above a +# non-empty examples table — a client-visible contradiction that leaked into +# the PDF handed to clients. See _build_listings_page / _deals_sourced_thin_ +# listings_note_html / _reliability_note_html. + + +def test_listings_page_zero_analogs_shown_cards_no_false_zero_count() -> None: + """The exact bug: n_analogs=0 + 3 shown analogs must NOT print '0 шт.' — + falls back to the actually-shown population (3) and adds an honest + deals-sourced footnote.""" + analogs = [ + _analog(address="ул. Льва Толстого, 8А"), + _analog(address="ул. Кирова, 4"), + _analog(address="ул. Льва Толстого, 34"), + ] + est = _estimate(n_analogs=0, analogs=analogs, sources_used=["avito"]) + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "0 шт." not in html + assert "3 шт." in html + assert "Оценка построена по зарегистрированным сделкам Росреестра" in html + assert "почти нет" in html + + +def test_listings_page_zero_analogs_empty_cards_stays_honest_zero() -> None: + """Control: genuinely zero listings (no cards to show either) — '0 шт.' is + honest here, and the deals-sourced footnote (which explains a MISMATCH) + must NOT appear since there is nothing to reconcile.""" + est = _estimate(n_analogs=0, analogs=[]) + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "0 шт." in html + assert "Оценка построена по зарегистрированным сделкам Росреестра" not in html + + +def test_listings_page_healthy_sample_keeps_full_n_analogs_not_capped_len() -> None: + """Control/regression guard for the max() choice: a healthy sample where + n_analogs (15) EXCEEDS the capped display list (10, AggregatedEstimate's + own top-10 cap) must keep printing the full honest count (15 шт.), NOT + silently understate it to len(analogs) (10 шт.).""" + analogs = [_analog(address=f"ул. Тест, {i}") for i in range(10)] + est = _estimate(n_analogs=15, analogs=analogs, sources_used=["avito"]) + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "15 шт." in html + assert "10 шт." not in html + + +def test_listings_page_relaxations_warning_shown_with_labels() -> None: + """relaxations non-empty → warning block present, names the labels, and + reliability != 'ok' — mirrors what the web LowConfidenceBanner already + shows (see AggregatedEstimate docstring).""" + est = _estimate(relaxations=["учтены студии", "радиус расширен до 3000 м"], reliability="low") + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "Точность оценки снижена." in html + assert "учтены студии" in html + assert "радиус расширен до 3000 м" in html + + +def test_listings_page_reliability_downgraded_no_relaxations_fallback_text() -> None: + """reliability != 'ok' but relaxations is empty (cascade couldn't grow a + thin sample, estimator.py #oblast-F) → warning block still shown, with a + fallback sentence (not an empty label list).""" + est = _estimate(n_analogs=2, reliability="very_low", relaxations=[]) + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "Точность оценки снижена." in html + assert "небольшой выборке" in html + + +def test_listings_page_no_warning_block_when_ok_and_no_relaxations() -> None: + """Control: the common/unrelaxed case (reliability='ok' default, no + relaxations) — no warning block at all, byte-identical to the report + before these fields existed.""" + est = _estimate() + assert est.reliability == "ok" + assert est.relaxations == [] + html = mod._build_listings_page(est, _SNAPSHOT, _GENERIC) + assert "Точность оценки снижена." not in html + + def test_build_deals_page_none_year_built_no_crash() -> None: snap = dict(_SNAPSHOT) snap["year_built"] = None diff --git a/tradein-mvp/backend/tests/test_street_deals_endpoint.py b/tradein-mvp/backend/tests/test_street_deals_endpoint.py index c1ea1249..003883e1 100644 --- a/tradein-mvp/backend/tests/test_street_deals_endpoint.py +++ b/tradein-mvp/backend/tests/test_street_deals_endpoint.py @@ -107,6 +107,64 @@ def test_extract_street_name_parametrized(address: str | None, expected: str | N assert extract_street_name(address) == expected +@pytest.mark.parametrize( + "address,expected", + [ + # Live-prod repro (2026-08-10): DaData format — abbreviations WITHOUT a + # trailing dot ("ул Академика Парина", not "ул. ..."), plus a leading + # postal index + admin parts ("620105, Свердловская обл, г + # Екатеринбург, Академический р-н, ..."). Old `_STREET_KW_RE` required + # the dot → keyword never matched on ANY DaData address → street-deals + # / sales-vs-listings endpoints silently returned empty for every + # DaData-geocoded request, not just this one. + ( + "620105, Свердловская обл, г Екатеринбург, Академический р-н, " + "ул Академика Парина, д 46/5", + "Академика Парина", + ), + # Same address, WITH the dot — must give the identical result (dot + # optional, not dot-forbidden). + ( + "620105, Свердловская обл, г Екатеринбург, Академический р-н, " + "ул. Академика Парина, д 46/5", + "Академика Парина", + ), + # Same address, full word "улица" — the alternation-order/backtracking + # concern: "ул\\.?" must NOT eat the "ул" prefix of "улица" and leave + # "ица ..." behind. + ( + "620105, Свердловская обл, г Екатеринбург, Академический р-н, " + "улица Академика Парина, д 46/5", + "Академика Парина", + ), + # Without the leading postal index — same admin prefix otherwise. + ( + "Свердловская обл, г Екатеринбург, Академический р-н, ул Академика Парина, д 46/5", + "Академика Парина", + ), + # Bare street+house, no admin prefix at all. + ("ул Академика Парина, д 46/5", "Академика Парина"), + # Other dot-optional abbreviations from _STREET_KW_RE (пр/пер/ш/наб/пл/мкр). + ("г Екатеринбург, пр Ленина, 5", "Ленина"), + ("г Екатеринбург, пер Красный, 4", "Красный"), + ("г Екатеринбург, наб Реки Исеть, 1", "Реки Исеть"), + # "ул. X" / "ул X" / "улица X" must all agree (no dot-optional regression). + ("Екатеринбург, ул. Малышева, 1", "Малышева"), + ("Екатеринбург, ул Малышева, 1", "Малышева"), + ("Екатеринбург, улица Малышева, 1", "Малышева"), + ], +) +def test_extract_street_name_dadata_no_dot_abbreviations( + address: str | None, expected: str | None +) -> None: + """#pdf-honesty/street-deals live-prod fix (2026-08-10): DaData addresses + use dot-less abbreviations ("ул", "пр", "пер", "ш", "наб", "пл", "мкр") + — _STREET_KW_RE must match them exactly like the dotted forms.""" + from app.services.estimator import extract_street_name + + assert extract_street_name(address) == expected + + # ── Helpers ─────────────────────────────────────────────────────────────────── diff --git a/tradein-mvp/frontend/src/app/v2/page.tsx b/tradein-mvp/frontend/src/app/v2/page.tsx index 49cc2846..f0164c23 100644 --- a/tradein-mvp/frontend/src/app/v2/page.tsx +++ b/tradein-mvp/frontend/src/app/v2/page.tsx @@ -20,6 +20,7 @@ import TopNav from "@/components/trade-in/v2/TopNav"; import HeroBar from "@/components/trade-in/v2/HeroBar"; import ParamsPanel from "@/components/trade-in/v2/ParamsPanel"; import ResultPanel from "@/components/trade-in/v2/ResultPanel"; +import { LowConfidenceBanner } from "@/components/trade-in/v2/LowConfidenceBanner"; import { ObjectSummary } from "@/components/trade-in/v2/ObjectSummary"; import { LeadForm } from "@/components/trade-in/v2/LeadForm"; import { Footer } from "@/components/trade-in/v2/Footer"; @@ -584,6 +585,18 @@ export default function TradeInV2Page() { // — no hydration drift, same reason the PDF control is gated behind `mounted`. const hasEstimate = mounted && estimate != null && !insufficient; + // fix (never-block estimate) — reliability/relaxations are optional on the + // wire (old/cached estimates predate the backend fields), default to the + // "nothing to disclose" values so a stale response never fabricates a + // warning. LowConfidenceBanner mounts above the result whenever the sample + // was thin (reliability !== "ok") or the backend had to relax the search to + // produce a price at all (relaxations.length > 0) — never on insufficient + // (no price at all — that stays InsufficientPanel, no banner to layer over). + const reliability = estimate?.reliability ?? "ok"; + const relaxations = estimate?.relaxations ?? []; + const showLowConfidenceBanner = + !insufficient && (reliability !== "ok" || relaxations.length > 0); + // ── Mapped presentation data (memoised so nav/drawer toggles don't recompute // geometry). ────────────────────────────────────────────────────────── const report = useMemo( @@ -772,12 +785,31 @@ export default function TradeInV2Page() { /> ); } else if (estimate && !insufficient && resultPanelData) { + // Banner is a sibling ABOVE ResultPanel, not a change to ResultPanel + // itself — the wrapper only replaces the direct grid child; ResultPanel's + // own markup/props are untouched from before this fix. middleContent = ( -