feat(tradein): per-cadastral-quarter price index in estimator (#764)
Apply quarter_price_index (FDW) via gap-correction to avoid double-counting location already encoded in the analog median: adjusted_ppm2 = base_median_ppm2 × (target_index / avg_analog_index) median_price/range_low/range_high scale by the same factor. Guards (no double-count + no regression): - feature flag estimate_quarter_index_enabled (default on) - pure-radius path only: skip if same-building anchor fired (anchor_tier) or any Avito IMV anchor influenced median/range (imv_anchor_present) - Guard-2: skip when >0.6 of analogs share the target quarter - sparse fallback: no row / n_deals<10 / FDW error → no-op - bimodal guard: price_index>2.0 & n_deals<50 → no-op (heterogeneous quarters) Analog quarter from listings.building_cadastral_number (added to analog SELECT); target quarter from dadata.house_cadnum. Batched single FDW lookup for analog indexes. Disclosure appended to confidence_explanation. Refs #764
This commit is contained in:
parent
99c751f834
commit
a749151679
3 changed files with 1168 additions and 4 deletions
|
|
@ -90,6 +90,24 @@ class Settings(BaseSettings):
|
|||
asking_to_sold_haircut: float = 0.05 # дефолтная asking→sold скидка (banded по ppm²)
|
||||
estimate_fsd_k: float = 1.65 # множитель FSD → полуширина диапазона
|
||||
|
||||
# ── #764: per-cadastral-quarter price index correction ───────────────────
|
||||
# Gap-correction: квартальный индекс применяется ТОЛЬКО в pure-radius пути
|
||||
# (когда same-building anchor и IMV-blend не сработали). Корректирует РАЗРЫВ
|
||||
# между квартальным уровнем целевого объекта и усреднённым квартальным уровнем
|
||||
# аналогов — не дублирует location, уже заложенный в медиану аналогов.
|
||||
# Формула: adjusted_ppm2 = base_ppm2 × target_index / avg_analog_index.
|
||||
# За флагом: False → точно текущее поведение (backward-compatible).
|
||||
estimate_quarter_index_enabled: bool = True
|
||||
# Минимальное число сделок в квартале (sparse fallback: меньше — no-op).
|
||||
estimate_quarter_index_min_n_deals: int = 10
|
||||
# Guard-2 (no double-count): если доля аналогов ИЗ ТОГО ЖЕ квартала > порога —
|
||||
# аналоги уже несут локацию квартала → skip (location in median).
|
||||
estimate_quarter_match_skip_ratio: float = 0.6
|
||||
# Bimodal/nominal guard (backtest 2026-05-31): структурно неоднородные кварталы
|
||||
# дают индекс > 2.0 при малой выборке → no-op чтобы избежать регрессию.
|
||||
estimate_quarter_index_max_for_small_n: float = 2.0
|
||||
estimate_quarter_index_small_n_threshold: int = 50
|
||||
|
||||
# ── Estimate enrichment time-budgets (#654) ──────────────────────────────
|
||||
# POST /estimate делает несколько ПОСЛЕДОВАТЕЛЬНЫХ блокирующих сетевых
|
||||
# вызовов (geocode → Overpass → Yandex valuation → IMV → Cian). Yandex
|
||||
|
|
|
|||
|
|
@ -788,6 +788,133 @@ def _apply_imv_blend(
|
|||
return new_median, new_range_high, new_ppm2, blended, anchor_total
|
||||
|
||||
|
||||
# ── #764: per-cadastral-quarter price index correction ───────────────────────
|
||||
|
||||
|
||||
def _quarter_from_cadastre(cad_num: str | None) -> str | None:
|
||||
"""Извлечь кадастровый номер квартала из кадастрового номера дома/квартиры.
|
||||
|
||||
Формат: AA:BB:CCCCCC или AA:BB:CCCCCCC (6 или 7 цифр в третьей части).
|
||||
Возвращаем первые три двоеточие-разделённых компонента.
|
||||
Пример: "66:41:0204016:350" → "66:41:0204016".
|
||||
При отсутствии или некорректном формате → None.
|
||||
"""
|
||||
if not cad_num:
|
||||
return None
|
||||
parts = cad_num.split(":")
|
||||
if len(parts) < 3:
|
||||
return None
|
||||
quarter = ":".join(parts[:3])
|
||||
# Проверяем что третья часть — числовая (квартал, не мусор)
|
||||
if not parts[2].isdigit():
|
||||
return None
|
||||
return quarter
|
||||
|
||||
|
||||
def _lookup_quarter_index(
|
||||
db: Session,
|
||||
*,
|
||||
quarter_cad_number: str,
|
||||
min_n_deals: int,
|
||||
) -> tuple[float, int] | None:
|
||||
"""Поиск price_index для кадастрового квартала в FDW-таблице quarter_price_index.
|
||||
|
||||
Возвращает (price_index, n_deals) или None при отсутствии строки / n_deals < min_n_deals
|
||||
/ любой FDW-ошибке (graceful — backward-compatible).
|
||||
Использует CAST(:q AS varchar) — psycopg v3 convention.
|
||||
"""
|
||||
try:
|
||||
row = (
|
||||
db.execute(
|
||||
text(
|
||||
"""
|
||||
SELECT price_index, n_deals
|
||||
FROM quarter_price_index
|
||||
WHERE quarter_cad_number = CAST(:q AS varchar)
|
||||
AND n_deals >= CAST(:min_n AS bigint)
|
||||
LIMIT 1
|
||||
"""
|
||||
),
|
||||
{"q": quarter_cad_number, "min_n": min_n_deals},
|
||||
)
|
||||
.mappings()
|
||||
.first()
|
||||
)
|
||||
except Exception as exc:
|
||||
logger.warning("quarter_price_index FDW lookup failed (graceful, no-op): %s", exc)
|
||||
return None
|
||||
if row is None:
|
||||
return None
|
||||
return float(row["price_index"]), int(row["n_deals"])
|
||||
|
||||
|
||||
def _lookup_quarter_indexes(
|
||||
db: Session,
|
||||
*,
|
||||
quarter_cad_numbers: list[str],
|
||||
min_n_deals: int,
|
||||
) -> dict[str, float]:
|
||||
"""Батч-поиск price_index для списка кадастровых кварталов (одним SQL-запросом).
|
||||
|
||||
Возвращает {quarter_cad_number: price_index} только для кварталов, у которых
|
||||
n_deals >= min_n_deals. Кварталы без записи или с n_deals < порога — не попадают
|
||||
в словарь. При любой FDW-ошибке → {} (graceful, avg_analog_index остаётся 1.0).
|
||||
"""
|
||||
if not quarter_cad_numbers:
|
||||
return {}
|
||||
distinct = list(dict.fromkeys(quarter_cad_numbers)) # сохраняем порядок, убираем дубли
|
||||
try:
|
||||
rows = (
|
||||
db.execute(
|
||||
text(
|
||||
"""
|
||||
SELECT quarter_cad_number, price_index
|
||||
FROM quarter_price_index
|
||||
WHERE quarter_cad_number = ANY(CAST(:quarters AS varchar[]))
|
||||
AND n_deals >= CAST(:min_n AS bigint)
|
||||
"""
|
||||
),
|
||||
{"quarters": distinct, "min_n": min_n_deals},
|
||||
)
|
||||
.mappings()
|
||||
.all()
|
||||
)
|
||||
except Exception as exc:
|
||||
logger.warning("quarter_price_index batch FDW lookup failed (graceful, no-op): %s", exc)
|
||||
return {}
|
||||
return {str(row["quarter_cad_number"]): float(row["price_index"]) for row in rows}
|
||||
|
||||
|
||||
def _apply_quarter_index(
|
||||
*,
|
||||
base_median_ppm2: float,
|
||||
base_median_price: int,
|
||||
base_range_low: int,
|
||||
base_range_high: int,
|
||||
target_index: float,
|
||||
avg_analog_index: float,
|
||||
) -> tuple[float, int, int, int, float]:
|
||||
"""Чистая (testable без БД) gap-correction квартального индекса (#764).
|
||||
|
||||
Корректирует ТОЛЬКО разрыв между квартальным уровнем целевого объекта и
|
||||
усреднённым квартальным уровнем аналогов:
|
||||
factor = target_index / avg_analog_index
|
||||
adjusted_ppm2 = base_median_ppm2 × factor
|
||||
|
||||
Все ценовые выходы масштабируются одним и тем же factor → median/range
|
||||
остаются геометрически консистентными.
|
||||
|
||||
Returns (adjusted_ppm2, adjusted_median_price, adjusted_range_low,
|
||||
adjusted_range_high, factor).
|
||||
"""
|
||||
factor = target_index / avg_analog_index
|
||||
adjusted_ppm2 = base_median_ppm2 * factor
|
||||
adjusted_median_price = round(base_median_price * factor)
|
||||
adjusted_range_low = round(base_range_low * factor)
|
||||
adjusted_range_high = round(base_range_high * factor)
|
||||
return adjusted_ppm2, adjusted_median_price, adjusted_range_low, adjusted_range_high, factor
|
||||
|
||||
|
||||
def _fetch_dkp_corridor(
|
||||
db: Session,
|
||||
*,
|
||||
|
|
@ -1837,6 +1964,10 @@ async def estimate_quality(
|
|||
# blend'ом и расширяем верх диапазона. Всё за флагом + null-guard (no-op без IMV).
|
||||
# ВАЖНО (v2): IMV-blend выполняется ТОЛЬКО когда same-building anchor НЕ сработал
|
||||
# (anchor_tier is None) — не накладываем blend поверх уже-построенного якоря дома.
|
||||
# #764: imv_anchor_present — любой IMV-anchor повлиял на estimate (median OR range).
|
||||
# Guard-1b использует этот флаг чтобы пропустить квартальный индекс при любом
|
||||
# IMV-влиянии, не только при blended (range_high расширяется даже без blend).
|
||||
imv_anchor_present: bool = False
|
||||
avito_imv_summary: AvitoImvSummary | None = None
|
||||
if (
|
||||
anchor_tier is None
|
||||
|
|
@ -1883,6 +2014,7 @@ async def estimate_quality(
|
|||
)
|
||||
|
||||
if anchor_total is not None:
|
||||
imv_anchor_present = True
|
||||
new_median, new_range_high, new_ppm2, blended, anchor_used = _apply_imv_blend(
|
||||
median_price=median_price,
|
||||
range_high=range_high,
|
||||
|
|
@ -1942,6 +2074,146 @@ async def estimate_quality(
|
|||
),
|
||||
)
|
||||
|
||||
# ── #764: per-cadastral-quarter price index gap-correction ──────────────────
|
||||
# Применяется ТОЛЬКО в pure-radius пути (Guard-1): когда same-building anchor
|
||||
# не сработал И IMV-blend не поднял медиану. Оба механизма уже учитывают
|
||||
# location пространственно — наложение индекса сверху даёт double-count.
|
||||
# Формула: adjusted_ppm2 = base_ppm2 × (target_index / avg_analog_index),
|
||||
# где avg_analog_index = взвешенная по ppm² медиана аналогов, чьи кварталы
|
||||
# известны. Если аналоги без кадастрового номера — avg_analog_index=1.0 (no-op).
|
||||
if (
|
||||
settings.estimate_quarter_index_enabled
|
||||
and anchor_tier is None # Guard-1a: same-building anchor не сработал
|
||||
and not imv_anchor_present # Guard-1b: IMV-anchor не повлиял (median или range)
|
||||
and median_price > 0
|
||||
and payload.area_m2
|
||||
):
|
||||
# Резолвим квартал target'а: Primary — DaData house_cadnum.
|
||||
target_quarter: str | None = _quarter_from_cadastre(
|
||||
dadata.house_cadnum if dadata is not None else None
|
||||
)
|
||||
# Fallback: building_cadastral_number из самих аналогов (если все в 1 доме
|
||||
# — Tier S path; тогда кадастровый номер квартала тот же). Не применяем
|
||||
# PostGIS point-in-quarter: нет готовой geometry-таблицы кварталов в tradein DB.
|
||||
if target_quarter is None:
|
||||
for lot in listings_clean:
|
||||
cq = _quarter_from_cadastre(lot.get("building_cadastral_number"))
|
||||
if cq is not None:
|
||||
target_quarter = cq
|
||||
break
|
||||
|
||||
if target_quarter is not None:
|
||||
qindex_result = _lookup_quarter_index(
|
||||
db,
|
||||
quarter_cad_number=target_quarter,
|
||||
min_n_deals=settings.estimate_quarter_index_min_n_deals,
|
||||
)
|
||||
if qindex_result is not None:
|
||||
target_qi, target_n_deals = qindex_result
|
||||
|
||||
# Bimodal/nominal guard (Guard-4): структурно неоднородный квартал
|
||||
# при малой выборке → no-op (regr. Радищева 66:41:0401017 et al.)
|
||||
if (
|
||||
target_qi > settings.estimate_quarter_index_max_for_small_n
|
||||
and target_n_deals < settings.estimate_quarter_index_small_n_threshold
|
||||
):
|
||||
logger.info(
|
||||
"quarter_index: bimodal guard triggered "
|
||||
"(index=%.3f n=%d < %d) for %s — no-op",
|
||||
target_qi,
|
||||
target_n_deals,
|
||||
settings.estimate_quarter_index_small_n_threshold,
|
||||
target_quarter,
|
||||
)
|
||||
else:
|
||||
# Вычисляем квартал каждого аналога ОДИН РАЗ — переиспользуем
|
||||
# для Guard-2 (same-quarter ratio) и avg_analog_index weighting.
|
||||
# lot_quarters_for_guard2: все лоты с известным кварталом (как раньше).
|
||||
# analog_quarters: только лоты с известным кварталом И ценой (для весов).
|
||||
lot_quarters_for_guard2: list[str] = []
|
||||
analog_quarters: list[tuple[str, float]] = []
|
||||
for lot in listings_clean:
|
||||
lq = _quarter_from_cadastre(lot.get("building_cadastral_number"))
|
||||
if lq is None:
|
||||
continue
|
||||
lot_quarters_for_guard2.append(lq)
|
||||
lp = lot.get("price_per_m2")
|
||||
if lp:
|
||||
analog_quarters.append((lq, float(lp)))
|
||||
|
||||
# Guard-2: доля аналогов ИЗ ТОГО ЖЕ квартала > skip_ratio →
|
||||
# location уже в медиане — пропускаем.
|
||||
same_quarter_count = sum(
|
||||
1 for lq in lot_quarters_for_guard2 if lq == target_quarter
|
||||
)
|
||||
same_quarter_ratio = (
|
||||
same_quarter_count / len(listings_clean) if listings_clean else 0.0
|
||||
)
|
||||
if same_quarter_ratio > settings.estimate_quarter_match_skip_ratio:
|
||||
logger.info(
|
||||
"quarter_index: Guard-2 skip (same-quarter ratio=%.2f > %.2f)"
|
||||
" for %s",
|
||||
same_quarter_ratio,
|
||||
settings.estimate_quarter_match_skip_ratio,
|
||||
target_quarter,
|
||||
)
|
||||
else:
|
||||
# Вычисляем avg_analog_index — ppm²-взвешенное среднее по
|
||||
# аналогам, чьи кварталы известны И присутствуют в индексе.
|
||||
# Один батч-запрос вместо N последовательных FDW roundtrips.
|
||||
# Аналоги без кадастрового номера — игнорируем (не штрафуем).
|
||||
distinct_analog_quarters = list(
|
||||
dict.fromkeys(lq for lq, _lp in analog_quarters)
|
||||
)
|
||||
analog_index_map = _lookup_quarter_indexes(
|
||||
db,
|
||||
quarter_cad_numbers=distinct_analog_quarters,
|
||||
min_n_deals=settings.estimate_quarter_index_min_n_deals,
|
||||
)
|
||||
weighted_sum = 0.0
|
||||
weight_total = 0.0
|
||||
for lq, lp in analog_quarters:
|
||||
lot_qi = analog_index_map.get(lq)
|
||||
if lot_qi is None:
|
||||
continue
|
||||
weighted_sum += lp * lot_qi
|
||||
weight_total += lp
|
||||
|
||||
avg_analog_index = weighted_sum / weight_total if weight_total > 0 else 1.0
|
||||
|
||||
(
|
||||
median_ppm2,
|
||||
median_price,
|
||||
range_low,
|
||||
range_high,
|
||||
qi_factor,
|
||||
) = _apply_quarter_index(
|
||||
base_median_ppm2=median_ppm2,
|
||||
base_median_price=median_price,
|
||||
base_range_low=range_low,
|
||||
base_range_high=range_high,
|
||||
target_index=target_qi,
|
||||
avg_analog_index=avg_analog_index,
|
||||
)
|
||||
analogs_with_qi = sum(
|
||||
1 for lq, _lp in analog_quarters if lq in analog_index_map
|
||||
)
|
||||
logger.info(
|
||||
"quarter_index: applied target=%s target_qi=%.3f"
|
||||
" avg_analog_qi=%.3f factor=%.3f"
|
||||
" (same_quarter_ratio=%.2f analogs_with_qi=%d)",
|
||||
target_quarter,
|
||||
target_qi,
|
||||
avg_analog_index,
|
||||
qi_factor,
|
||||
same_quarter_ratio,
|
||||
analogs_with_qi,
|
||||
)
|
||||
explanation = (explanation or "") + (
|
||||
f" Учтена локация квартала" f" (индекс цен квартала ×{qi_factor:.2f})."
|
||||
)
|
||||
sources_used_pre = sorted(set(sources_used_pre) | {"quarter_index"})
|
||||
|
||||
# 4c (cont.). expected_sold_* выводим ЗДЕСЬ — ПОСЛЕ #651 IMV-blend / SB-anchor,
|
||||
# которые могли поднять median_price/median_ppm2 и расширить range_high. Применяем
|
||||
# ratio к POST-якорным значениям → asking (median_price_rub) и sold
|
||||
|
|
@ -2626,7 +2898,8 @@ _ANALOG_SELECT_COLS = """
|
|||
rooms, area_m2, floor, total_floors,
|
||||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at
|
||||
scraped_at,
|
||||
building_cadastral_number
|
||||
"""
|
||||
|
||||
_COMMON_WHERE = """
|
||||
|
|
@ -2737,7 +3010,8 @@ def _fetch_analogs(
|
|||
rooms, area_m2, floor, total_floors,
|
||||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at, distance_m, relevance_score
|
||||
scraped_at, distance_m, relevance_score,
|
||||
building_cadastral_number
|
||||
FROM base
|
||||
WHERE rn_addr <= :max_per_addr
|
||||
ORDER BY scraped_at DESC
|
||||
|
|
@ -2787,7 +3061,8 @@ def _fetch_analogs(
|
|||
rooms, area_m2, floor, total_floors,
|
||||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at, distance_m, relevance_score
|
||||
scraped_at, distance_m, relevance_score,
|
||||
building_cadastral_number
|
||||
FROM base
|
||||
WHERE rn_addr <= :max_per_addr
|
||||
ORDER BY scraped_at DESC
|
||||
|
|
@ -2874,7 +3149,8 @@ def _fetch_analogs(
|
|||
rooms, area_m2, floor, total_floors,
|
||||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at, distance_m, relevance_score
|
||||
scraped_at, distance_m, relevance_score,
|
||||
building_cadastral_number
|
||||
FROM base
|
||||
WHERE rn_addr <= :max_per_addr
|
||||
ORDER BY relevance_score
|
||||
|
|
@ -2929,6 +3205,7 @@ def _fetch_analogs(
|
|||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at,
|
||||
building_cadastral_number,
|
||||
ST_Distance(geom::geography, ST_MakePoint(:lon, :lat)::geography)
|
||||
AS distance_m,
|
||||
(
|
||||
|
|
@ -3004,6 +3281,7 @@ def _fetch_analogs(
|
|||
price_rub, price_per_m2,
|
||||
listing_date, days_on_market, photo_urls,
|
||||
scraped_at,
|
||||
building_cadastral_number,
|
||||
distance_m,
|
||||
relevance_score
|
||||
FROM base
|
||||
|
|
|
|||
868
tradein-mvp/backend/tests/test_estimator_quarter_index.py
Normal file
868
tradein-mvp/backend/tests/test_estimator_quarter_index.py
Normal file
|
|
@ -0,0 +1,868 @@
|
|||
"""Unit tests for #764 — per-cadastral-quarter price index gap-correction.
|
||||
|
||||
Tests покрывают:
|
||||
- _quarter_from_cadastre: парсинг кадастрового номера в квартал
|
||||
- _apply_quarter_index: чистая математика корректировки (без БД)
|
||||
- _lookup_quarter_index: DB-хелпер с мокнутой Session
|
||||
- Guard-1: anchor_tier не None → no-op (same-building anchor)
|
||||
- Guard-1b: IMV-blended → no-op
|
||||
- Guard-2: >0.6 аналогов в целевом квартале → no-op
|
||||
- Sparse fallback: нет строки / n_deals < min_n → no-op
|
||||
- Bimodal guard: price_index>2.0 AND n_deals<50 → no-op
|
||||
- Flag off → точное старое поведение
|
||||
|
||||
Паттерн: os.environ.setdefault перед импортом (как test_estimator_pure_units.py).
|
||||
Чистые хелперы — без БД. DB-хелпер — мокнутая Session.
|
||||
estimate_quality-level тесты — через anyio.run + полный stub-пач всех I/O.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import os
|
||||
from typing import Any
|
||||
from unittest.mock import AsyncMock, MagicMock, patch
|
||||
|
||||
os.environ.setdefault("DATABASE_URL", "postgresql+psycopg://test:test@localhost:5432/test")
|
||||
|
||||
import anyio
|
||||
import pytest
|
||||
|
||||
from app.services.estimator import (
|
||||
_apply_quarter_index,
|
||||
_lookup_quarter_index,
|
||||
_lookup_quarter_indexes,
|
||||
_quarter_from_cadastre,
|
||||
)
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# _quarter_from_cadastre
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_standard() -> None:
|
||||
"""Нормальный кадастровый номер дома → квартал (первые три части)."""
|
||||
assert _quarter_from_cadastre("66:41:0204016:350") == "66:41:0204016"
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_7digit_block() -> None:
|
||||
"""7-значный блок квартала → корректно."""
|
||||
assert _quarter_from_cadastre("66:41:0401017:100") == "66:41:0401017"
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_none_input() -> None:
|
||||
assert _quarter_from_cadastre(None) is None
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_empty_string() -> None:
|
||||
assert _quarter_from_cadastre("") is None
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_too_few_parts() -> None:
|
||||
assert _quarter_from_cadastre("66:41") is None
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_non_numeric_third_part() -> None:
|
||||
"""Третья часть не числовая → None (не кадастровый квартал)."""
|
||||
assert _quarter_from_cadastre("66:41:BADDATA:100") is None
|
||||
|
||||
|
||||
def test_quarter_from_cadastre_no_fourth_part() -> None:
|
||||
"""Три части без объекта — само по себе квартал: возвращаем как есть."""
|
||||
result = _quarter_from_cadastre("66:41:0204016")
|
||||
assert result == "66:41:0204016"
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# _apply_quarter_index — чистая математика
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_apply_quarter_index_basic_math() -> None:
|
||||
"""Базовый кейс: target_index=1.2, avg_analog=1.0 → factor=1.2."""
|
||||
ppm2, median, low, high, factor = _apply_quarter_index(
|
||||
base_median_ppm2=100_000.0,
|
||||
base_median_price=5_000_000,
|
||||
base_range_low=4_000_000,
|
||||
base_range_high=6_000_000,
|
||||
target_index=1.2,
|
||||
avg_analog_index=1.0,
|
||||
)
|
||||
assert abs(factor - 1.2) < 1e-9
|
||||
assert abs(ppm2 - 120_000.0) < 1.0
|
||||
assert median == round(5_000_000 * 1.2)
|
||||
assert low == round(4_000_000 * 1.2)
|
||||
assert high == round(6_000_000 * 1.2)
|
||||
|
||||
|
||||
def test_apply_quarter_index_gap_correction() -> None:
|
||||
"""Gap-correction: target=1.3, avg_analog=1.1 → factor=1.3/1.1 ≈ 1.182."""
|
||||
_, median, _, _, factor = _apply_quarter_index(
|
||||
base_median_ppm2=200_000.0,
|
||||
base_median_price=10_000_000,
|
||||
base_range_low=8_000_000,
|
||||
base_range_high=12_000_000,
|
||||
target_index=1.3,
|
||||
avg_analog_index=1.1,
|
||||
)
|
||||
expected_factor = 1.3 / 1.1
|
||||
assert abs(factor - expected_factor) < 1e-9
|
||||
assert median == round(10_000_000 * expected_factor)
|
||||
|
||||
|
||||
def test_apply_quarter_index_same_quarter_noop() -> None:
|
||||
"""target_index == avg_analog_index → factor=1.0, медиана не меняется."""
|
||||
_, median, low, high, factor = _apply_quarter_index(
|
||||
base_median_ppm2=150_000.0,
|
||||
base_median_price=7_500_000,
|
||||
base_range_low=6_000_000,
|
||||
base_range_high=9_000_000,
|
||||
target_index=1.05,
|
||||
avg_analog_index=1.05,
|
||||
)
|
||||
assert abs(factor - 1.0) < 1e-9
|
||||
assert median == 7_500_000
|
||||
assert low == 6_000_000
|
||||
assert high == 9_000_000
|
||||
|
||||
|
||||
def test_apply_quarter_index_downcorrection() -> None:
|
||||
"""target_index < avg_analog_index → factor < 1.0 (коррекция вниз тоже работает)."""
|
||||
_, median, _, _, factor = _apply_quarter_index(
|
||||
base_median_ppm2=100_000.0,
|
||||
base_median_price=5_000_000,
|
||||
base_range_low=4_000_000,
|
||||
base_range_high=6_000_000,
|
||||
target_index=0.9,
|
||||
avg_analog_index=1.0,
|
||||
)
|
||||
assert factor < 1.0
|
||||
assert median == round(5_000_000 * 0.9)
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# _lookup_quarter_index — мокнутая Session
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_lookup_quarter_index_returns_row() -> None:
|
||||
"""Нормальная строка → возвращает (price_index, n_deals)."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.first.return_value = {
|
||||
"price_index": 1.25,
|
||||
"n_deals": 42,
|
||||
}
|
||||
result = _lookup_quarter_index(mock_db, quarter_cad_number="66:41:0204016", min_n_deals=10)
|
||||
assert result is not None
|
||||
qi, n = result
|
||||
assert abs(qi - 1.25) < 1e-9
|
||||
assert n == 42
|
||||
|
||||
|
||||
def test_lookup_quarter_index_none_when_no_row() -> None:
|
||||
"""Нет строки → None."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.first.return_value = None
|
||||
result = _lookup_quarter_index(mock_db, quarter_cad_number="66:41:9999999", min_n_deals=10)
|
||||
assert result is None
|
||||
|
||||
|
||||
def test_lookup_quarter_index_fdw_exception_graceful() -> None:
|
||||
"""FDW exception → None (graceful, no re-raise)."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.side_effect = RuntimeError("FDW connection refused")
|
||||
result = _lookup_quarter_index(mock_db, quarter_cad_number="66:41:0204016", min_n_deals=10)
|
||||
assert result is None
|
||||
|
||||
|
||||
def test_lookup_quarter_index_no_cast_colon_colon_in_sql() -> None:
|
||||
"""SQL текст хелпера не должен содержать :x::type (psycopg v3 rule)."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.first.return_value = None
|
||||
_lookup_quarter_index(mock_db, quarter_cad_number="66:41:0204016", min_n_deals=10)
|
||||
args, _ = mock_db.execute.call_args
|
||||
sql_text = str(args[0])
|
||||
# psycopg v3: CAST(:x AS type), never :x::type
|
||||
import re
|
||||
|
||||
assert not re.search(r":[a-z_]+::[a-z]", sql_text), f"::type cast found in SQL: {sql_text}"
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# _lookup_quarter_indexes (plural) — батч-хелпер
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_returns_dict() -> None:
|
||||
"""Нормальный результат → словарь {quarter: price_index}."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.all.return_value = [
|
||||
{"quarter_cad_number": "66:41:0204016", "price_index": 1.25},
|
||||
{"quarter_cad_number": "66:41:9998888", "price_index": 0.95},
|
||||
]
|
||||
result = _lookup_quarter_indexes(
|
||||
mock_db,
|
||||
quarter_cad_numbers=["66:41:0204016", "66:41:9998888"],
|
||||
min_n_deals=10,
|
||||
)
|
||||
assert result == {"66:41:0204016": 1.25, "66:41:9998888": 0.95}
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_empty_input_returns_empty() -> None:
|
||||
"""Пустой список кварталов → {} без обращения к БД."""
|
||||
mock_db = MagicMock()
|
||||
result = _lookup_quarter_indexes(mock_db, quarter_cad_numbers=[], min_n_deals=10)
|
||||
assert result == {}
|
||||
mock_db.execute.assert_not_called()
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_fdw_exception_returns_empty() -> None:
|
||||
"""FDW exception → {} (graceful, no re-raise)."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.side_effect = RuntimeError("FDW connection refused")
|
||||
result = _lookup_quarter_indexes(
|
||||
mock_db,
|
||||
quarter_cad_numbers=["66:41:0204016"],
|
||||
min_n_deals=10,
|
||||
)
|
||||
assert result == {}
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_deduplicates_input() -> None:
|
||||
"""Дублирующиеся кварталы в списке — передаются в БД без дублей."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.all.return_value = [
|
||||
{"quarter_cad_number": "66:41:0204016", "price_index": 1.1},
|
||||
]
|
||||
_lookup_quarter_indexes(
|
||||
mock_db,
|
||||
quarter_cad_numbers=["66:41:0204016", "66:41:0204016", "66:41:0204016"],
|
||||
min_n_deals=5,
|
||||
)
|
||||
passed_params = mock_db.execute.call_args[0][1]
|
||||
assert passed_params["quarters"] == ["66:41:0204016"]
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_no_cast_colon_colon_in_sql() -> None:
|
||||
"""Батч-хелпер: SQL не содержит :x::type (psycopg v3 rule)."""
|
||||
import re
|
||||
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.all.return_value = []
|
||||
_lookup_quarter_indexes(
|
||||
mock_db,
|
||||
quarter_cad_numbers=["66:41:0204016"],
|
||||
min_n_deals=10,
|
||||
)
|
||||
args, _ = mock_db.execute.call_args
|
||||
sql_text = str(args[0])
|
||||
assert not re.search(r":[a-z_]+::[a-z]", sql_text), f"::type cast found in SQL: {sql_text}"
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_uses_any_cast_array_idiom() -> None:
|
||||
"""SQL батч-хелпера содержит ANY(CAST(:quarters AS varchar[])) — pgpsycopg3 idiom."""
|
||||
mock_db = MagicMock()
|
||||
mock_db.execute.return_value.mappings.return_value.all.return_value = []
|
||||
_lookup_quarter_indexes(
|
||||
mock_db,
|
||||
quarter_cad_numbers=["66:41:0204016"],
|
||||
min_n_deals=10,
|
||||
)
|
||||
args, _ = mock_db.execute.call_args
|
||||
sql_text = str(args[0])
|
||||
assert "ANY(CAST(:quarters AS varchar[]))" in sql_text
|
||||
|
||||
|
||||
def test_lookup_quarter_indexes_multi_quarter_factor_matches_single() -> None:
|
||||
"""Батч возвращает те же значения, что N одиночных вызовов — математика идентична.
|
||||
|
||||
Аналоги: 2 лота из квартала A (index=1.2, ppm2=100k),
|
||||
1 лот из квартала B (index=0.8, ppm2=200k).
|
||||
Ожидаемый avg_analog_index = (100k*1.2 + 100k*1.2 + 200k*0.8) / (100k+100k+200k)
|
||||
= (120k + 120k + 160k) / 400k = 400k/400k = 1.0.
|
||||
"""
|
||||
index_map = {"66:41:AAAAAAA": 1.2, "66:41:BBBBBBB": 0.8}
|
||||
analog_lots = [
|
||||
("66:41:AAAAAAA", 100_000.0),
|
||||
("66:41:AAAAAAA", 100_000.0),
|
||||
("66:41:BBBBBBB", 200_000.0),
|
||||
]
|
||||
weighted_sum = sum(ppm2 * index_map[q] for q, ppm2 in analog_lots)
|
||||
weight_total = sum(ppm2 for _, ppm2 in analog_lots)
|
||||
avg_analog_index = weighted_sum / weight_total
|
||||
assert abs(avg_analog_index - 1.0) < 1e-9, f"Expected 1.0, got {avg_analog_index}"
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Helpers for estimate_quality integration tests (full I/O stub)
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
_BASE_PPM2 = 150_000.0
|
||||
_AREA = 40.0
|
||||
|
||||
|
||||
def _make_listing_qi(
|
||||
*,
|
||||
price_per_m2: float = _BASE_PPM2,
|
||||
area_m2: float = _AREA,
|
||||
building_cadastral_number: str | None = None,
|
||||
) -> dict[str, Any]:
|
||||
from datetime import UTC, datetime
|
||||
|
||||
price_rub = price_per_m2 * area_m2
|
||||
return {
|
||||
"source": "cian",
|
||||
"source_url": "https://cian.ru/offer/1",
|
||||
"address": "ЕКБ, ул. Тестовая, 5",
|
||||
"lat": 56.838,
|
||||
"lon": 60.595,
|
||||
"rooms": 1,
|
||||
"area_m2": area_m2,
|
||||
"floor": 4,
|
||||
"total_floors": 16,
|
||||
"price_rub": price_rub,
|
||||
"price_per_m2": price_per_m2,
|
||||
"listing_date": datetime(2026, 5, 1),
|
||||
"days_on_market": 10,
|
||||
"photo_urls": [],
|
||||
"scraped_at": datetime(2026, 5, 20, tzinfo=UTC),
|
||||
"distance_m": 100.0,
|
||||
"relevance_score": 0.1,
|
||||
"building_cadastral_number": building_cadastral_number,
|
||||
}
|
||||
|
||||
|
||||
def _make_fake_geo_qi():
|
||||
from app.services.geocoder import GeocodeResult
|
||||
|
||||
return GeocodeResult(
|
||||
lat=56.838,
|
||||
lon=60.595,
|
||||
full_address="Свердловская обл., Екатеринбург, ул. Тестовая, 5",
|
||||
provider="nominatim",
|
||||
)
|
||||
|
||||
|
||||
def _make_payload_qi(rooms: int = 1, area_m2: float = _AREA):
|
||||
from app.schemas.trade_in import TradeInEstimateInput
|
||||
|
||||
return TradeInEstimateInput(
|
||||
address="ЕКБ, ул. Тестовая, 5",
|
||||
area_m2=area_m2,
|
||||
rooms=rooms,
|
||||
floor=4,
|
||||
total_floors=16,
|
||||
)
|
||||
|
||||
|
||||
def _make_fake_dadata(house_cadnum: str | None):
|
||||
"""Minimal DadataAddressResult stub с нужным house_cadnum."""
|
||||
from app.services.dadata import DadataAddressResult
|
||||
|
||||
return DadataAddressResult(
|
||||
canonical_address="Свердловская обл., Екатеринбург, ул. Тестовая, 5",
|
||||
house_cadnum=house_cadnum,
|
||||
house_fias_id=None,
|
||||
lat=56.838,
|
||||
lon=60.595,
|
||||
qc_geo=1,
|
||||
qc_house=1,
|
||||
kladr_id=None,
|
||||
okato=None,
|
||||
oktmo=None,
|
||||
metro=[],
|
||||
raw={},
|
||||
)
|
||||
|
||||
|
||||
def _run_estimate_qi(
|
||||
analogs: list[dict[str, Any]],
|
||||
dadata_cadnum: str | None,
|
||||
qi_lookup_result: tuple[float, int] | None,
|
||||
*,
|
||||
flag_enabled: bool = True,
|
||||
anchor_tier_override: str | None = None,
|
||||
):
|
||||
"""Запускает estimate_quality с полным stub-пачем I/O; возвращает AggregatedEstimate."""
|
||||
from app.services.estimator import estimate_quality
|
||||
|
||||
db = MagicMock()
|
||||
payload = _make_payload_qi()
|
||||
|
||||
dadata_obj = _make_fake_dadata(dadata_cadnum) if dadata_cadnum is not None else None
|
||||
|
||||
# Батч-хелпер возвращает словарь: для каждого переданного квартала — тот же индекс,
|
||||
# что qi_lookup_result[0], если qi_lookup_result не None; иначе пустой dict.
|
||||
def _fake_lookup_indexes(db_arg, *, quarter_cad_numbers, min_n_deals):
|
||||
if qi_lookup_result is None:
|
||||
return {}
|
||||
return {q: qi_lookup_result[0] for q in quarter_cad_numbers}
|
||||
|
||||
async def _run():
|
||||
with (
|
||||
patch(
|
||||
"app.services.estimator.geocode",
|
||||
new=AsyncMock(return_value=_make_fake_geo_qi()),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.dadata_clean_address",
|
||||
new=AsyncMock(return_value=dadata_obj),
|
||||
),
|
||||
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_deals", return_value=[]),
|
||||
patch("app.services.estimator._fetch_dkp_corridor", return_value=None),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_imv_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_yandex_valuation_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.estimate_via_cian_valuation",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_asking_sold_ratio",
|
||||
return_value=(None, None),
|
||||
),
|
||||
patch("app.services.estimator._fetch_house_imv_anchor", return_value=None),
|
||||
# Stub singular target-quarter lookup
|
||||
patch(
|
||||
"app.services.estimator._lookup_quarter_index",
|
||||
return_value=qi_lookup_result,
|
||||
),
|
||||
# Stub batched analog-quarter lookup
|
||||
patch(
|
||||
"app.services.estimator._lookup_quarter_indexes",
|
||||
side_effect=_fake_lookup_indexes,
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.settings.estimate_quarter_index_enabled",
|
||||
flag_enabled,
|
||||
),
|
||||
):
|
||||
return await estimate_quality(payload, db)
|
||||
|
||||
return anyio.run(_run)
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Integration: gap-correction applied in pure-radius path
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
_OTHER_QUARTER = "66:41:9998888"
|
||||
_TARGET_QUARTER = "66:41:0204016"
|
||||
|
||||
_ANALOGS_OTHER_QUARTER = [
|
||||
_make_listing_qi(
|
||||
price_per_m2=_BASE_PPM2,
|
||||
building_cadastral_number=f"{_OTHER_QUARTER}:100",
|
||||
)
|
||||
for _ in range(3)
|
||||
]
|
||||
|
||||
|
||||
def test_quarter_index_correction_applied() -> None:
|
||||
"""Gap-correction срабатывает в pure-radius пути.
|
||||
|
||||
target_index=1.2, avg_analog_index=1.0 (аналоги без известного квартала →
|
||||
avg=1.0) → factor=1.2 → median должен вырасти на ×1.2.
|
||||
"""
|
||||
base_median = round(_BASE_PPM2 * _AREA) # 6_000_000
|
||||
|
||||
# Аналоги из ДРУГОГО квартала (building_cadastral_number = OTHER_QUARTER:100)
|
||||
# _lookup_quarter_index для аналогов вернёт тот же (1.2, 30) что и для target —
|
||||
# avg_analog_index = 1.2, factor = 1.2/1.2 = 1.0 (no change!).
|
||||
# Чтобы увидеть ненулевую коррекцию, делаем аналоги БЕЗ кадастрового номера
|
||||
# → avg_analog_index = 1.0 → factor = 1.2.
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs_no_cadnum,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=(1.2, 30),
|
||||
flag_enabled=True,
|
||||
)
|
||||
expected_median = round(base_median * 1.2)
|
||||
assert est.median_price_rub == expected_median
|
||||
# Disclosure должна содержать упоминание квартала
|
||||
assert est.confidence_explanation is not None
|
||||
assert "квартал" in est.confidence_explanation.lower()
|
||||
|
||||
|
||||
def test_quarter_index_flag_off_exact_old_behavior() -> None:
|
||||
"""При estimate_quarter_index_enabled=False — точное старое поведение."""
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs_no_cadnum,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=(1.2, 30),
|
||||
flag_enabled=False,
|
||||
)
|
||||
base_median = round(_BASE_PPM2 * _AREA)
|
||||
assert est.median_price_rub == base_median
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Guard-2: >0.6 аналогов в целевом квартале → no-op
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_guard2_skip_when_majority_analogs_in_target_quarter() -> None:
|
||||
"""Guard-2: >60% аналогов из целевого квартала → индекс не применяется."""
|
||||
# 4 аналога в target квартале, 1 в другом → ratio = 4/5 = 0.8 > 0.6 → skip
|
||||
target_cadnum = f"{_TARGET_QUARTER}:100"
|
||||
other_cadnum = f"{_OTHER_QUARTER}:100"
|
||||
analogs = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=target_cadnum),
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=target_cadnum),
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=target_cadnum),
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=target_cadnum),
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=other_cadnum),
|
||||
]
|
||||
base_median = round(_BASE_PPM2 * _AREA)
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=(1.5, 30), # high index — but guard-2 should skip
|
||||
flag_enabled=True,
|
||||
)
|
||||
# Медиана НЕ должна изменяться
|
||||
assert est.median_price_rub == base_median
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Sparse fallback: нет строки → no-op
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_sparse_fallback_no_row_noop() -> None:
|
||||
"""_lookup_quarter_index вернул None → no-op, медиана не меняется."""
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
base_median = round(_BASE_PPM2 * _AREA)
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs_no_cadnum,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=None, # sparse
|
||||
flag_enabled=True,
|
||||
)
|
||||
assert est.median_price_rub == base_median
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Bimodal guard: price_index>2.0 AND n_deals<50 → no-op
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_bimodal_guard_skips_high_index_small_n() -> None:
|
||||
"""Bimodal guard: price_index=3.5, n_deals=20 → no-op."""
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
base_median = round(_BASE_PPM2 * _AREA)
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs_no_cadnum,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=(3.5, 20), # index>2.0 AND n<50 → bimodal guard
|
||||
flag_enabled=True,
|
||||
)
|
||||
assert est.median_price_rub == base_median
|
||||
|
||||
|
||||
def test_bimodal_guard_allows_high_index_large_n() -> None:
|
||||
"""Bimodal guard НЕ срабатывает при price_index>2.0 если n_deals>=50."""
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
base_median = round(_BASE_PPM2 * _AREA)
|
||||
est = _run_estimate_qi(
|
||||
analogs=analogs_no_cadnum,
|
||||
dadata_cadnum=f"{_TARGET_QUARTER}:350",
|
||||
qi_lookup_result=(2.5, 60), # index>2.0 но n=60>=50 → применяется
|
||||
flag_enabled=True,
|
||||
)
|
||||
# factor = 2.5 → медиана должна измениться
|
||||
assert est.median_price_rub != base_median
|
||||
assert est.median_price_rub == round(base_median * 2.5)
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Guard-1a: anchor_tier не None → correction не применяется
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_guard1_anchor_tier_prevents_correction() -> None:
|
||||
"""Guard-1a: когда same-building anchor сработал (anchor_tier='A'),
|
||||
квартальный индекс не применяется (double-count guard).
|
||||
|
||||
Стабим _compute_same_building_anchor, чтобы вернул непустой anchor dict
|
||||
и _fetch_anchor_comps вернул comps → anchor_tier будет 'A'.
|
||||
"""
|
||||
from app.services.estimator import estimate_quality
|
||||
|
||||
db = MagicMock()
|
||||
payload = _make_payload_qi()
|
||||
dadata_obj = _make_fake_dadata(f"{_TARGET_QUARTER}:350")
|
||||
|
||||
fake_anchor = {
|
||||
"anchor_ppm2": _BASE_PPM2 * 1.5,
|
||||
"anchor_sold_ppm2": _BASE_PPM2 * 1.4,
|
||||
"fsd": 0.05,
|
||||
"confidence": "high",
|
||||
"n": 3,
|
||||
"cv": 0.05,
|
||||
"comp_min_ppm2": _BASE_PPM2 * 1.3,
|
||||
"comp_max_ppm2": _BASE_PPM2 * 1.7,
|
||||
"used_uplift": False,
|
||||
"haircut": 0.05,
|
||||
}
|
||||
fake_comps = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2 * 1.5, building_cadastral_number=None)
|
||||
for _ in range(3)
|
||||
]
|
||||
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
|
||||
async def _run():
|
||||
with (
|
||||
patch(
|
||||
"app.services.estimator.geocode",
|
||||
new=AsyncMock(return_value=_make_fake_geo_qi()),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.dadata_clean_address",
|
||||
new=AsyncMock(return_value=dadata_obj),
|
||||
),
|
||||
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_no_cadnum), False, "W"),
|
||||
),
|
||||
patch("app.services.estimator._fetch_deals", return_value=[]),
|
||||
patch("app.services.estimator._fetch_dkp_corridor", return_value=None),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_imv_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_yandex_valuation_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.estimate_via_cian_valuation",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_asking_sold_ratio",
|
||||
return_value=(None, None),
|
||||
),
|
||||
patch("app.services.estimator._fetch_house_imv_anchor", return_value=None),
|
||||
# same-building anchor FIRES → anchor_tier = 'A'
|
||||
patch(
|
||||
"app.services.estimator._fetch_anchor_comps",
|
||||
return_value=(fake_comps, "A"),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._compute_same_building_anchor",
|
||||
return_value=fake_anchor,
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._lookup_quarter_index",
|
||||
return_value=(1.5, 30), # would apply if not guarded
|
||||
),
|
||||
):
|
||||
return await estimate_quality(payload, db)
|
||||
|
||||
est = anyio.run(_run)
|
||||
|
||||
# Медиана от anchor = anchor_ppm2 * repair_coef(=1.0 нет ремонта) * area
|
||||
anchor_median = round(_BASE_PPM2 * 1.5 * _AREA)
|
||||
assert est.median_price_rub == anchor_median
|
||||
# explanation не содержит упоминания квартала (guard-1 сработал)
|
||||
assert (
|
||||
est.confidence_explanation is None
|
||||
or "квартал" not in (est.confidence_explanation or "").lower()
|
||||
)
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Guard-1b: IMV-blended → correction не применяется
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_guard1b_imv_blend_prevents_correction() -> None:
|
||||
"""Guard-1b: IMV-blend повышал медиану → квартальный индекс не применяется."""
|
||||
from app.services.estimator import estimate_quality
|
||||
|
||||
db = MagicMock()
|
||||
payload = _make_payload_qi()
|
||||
dadata_obj = _make_fake_dadata(f"{_TARGET_QUARTER}:350")
|
||||
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
|
||||
# IMV anchor сильно выше медианы → blend сработает
|
||||
imv_anchor = {
|
||||
"recommended_price": 30_000_000, # ≫ base_median 6М × 1.15
|
||||
"lower_price": 25_000_000,
|
||||
"higher_price": 35_000_000,
|
||||
"market_count": 500,
|
||||
"rooms": 1,
|
||||
"area_m2": _AREA,
|
||||
}
|
||||
|
||||
async def _run():
|
||||
with (
|
||||
patch(
|
||||
"app.services.estimator.geocode",
|
||||
new=AsyncMock(return_value=_make_fake_geo_qi()),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.dadata_clean_address",
|
||||
new=AsyncMock(return_value=dadata_obj),
|
||||
),
|
||||
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_no_cadnum), False, "W"),
|
||||
),
|
||||
patch("app.services.estimator._fetch_deals", return_value=[]),
|
||||
patch("app.services.estimator._fetch_dkp_corridor", return_value=None),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_imv_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_yandex_valuation_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.estimate_via_cian_valuation",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_asking_sold_ratio",
|
||||
return_value=(None, None),
|
||||
),
|
||||
# IMV anchor fires (returns non-None) → blend will trigger
|
||||
patch(
|
||||
"app.services.estimator._fetch_house_imv_anchor",
|
||||
return_value=imv_anchor,
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._lookup_quarter_index",
|
||||
return_value=(1.5, 30), # would apply if not guarded
|
||||
),
|
||||
):
|
||||
return await estimate_quality(payload, db)
|
||||
|
||||
est = anyio.run(_run)
|
||||
|
||||
# IMV blend: base 6М, anchor 30М, w=0.5 → 18М
|
||||
blended_median = round(6_000_000 * 0.5 + 30_000_000 * 0.5)
|
||||
assert est.median_price_rub == blended_median
|
||||
# explanation не должна содержать квартального дисклоужера
|
||||
assert "квартал" not in (est.confidence_explanation or "").lower()
|
||||
|
||||
|
||||
def test_guard1b_imv_anchor_below_blend_threshold_prevents_correction() -> None:
|
||||
"""Guard-1b: IMV anchor присутствует но ниже blend-порога (blended=False).
|
||||
|
||||
До фикса #764: imv_blended=False → квартальный индекс применялся поверх
|
||||
IMV-расширенного range_high (double-influence). После фикса: imv_anchor_present=True
|
||||
→ quarter index не применяется вне зависимости от blended.
|
||||
"""
|
||||
from app.services.estimator import estimate_quality
|
||||
|
||||
db = MagicMock()
|
||||
payload = _make_payload_qi()
|
||||
dadata_obj = _make_fake_dadata(f"{_TARGET_QUARTER}:350")
|
||||
|
||||
analogs_no_cadnum = [
|
||||
_make_listing_qi(price_per_m2=_BASE_PPM2, building_cadastral_number=None) for _ in range(3)
|
||||
]
|
||||
|
||||
# IMV anchor НИЖЕ blend-порога: base_median = 6_000_000, threshold=1.15 → порог 6.9М.
|
||||
# anchor=6_500_000 < 6.9М → blended=False, но range_high IMV-расширен.
|
||||
imv_anchor_below_threshold = {
|
||||
"recommended_price": 6_500_000,
|
||||
"lower_price": 5_800_000,
|
||||
"higher_price": 7_200_000,
|
||||
"market_count": 100,
|
||||
"rooms": 1,
|
||||
"area_m2": _AREA,
|
||||
}
|
||||
|
||||
async def _run():
|
||||
with (
|
||||
patch(
|
||||
"app.services.estimator.geocode",
|
||||
new=AsyncMock(return_value=_make_fake_geo_qi()),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.dadata_clean_address",
|
||||
new=AsyncMock(return_value=dadata_obj),
|
||||
),
|
||||
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_no_cadnum), False, "W"),
|
||||
),
|
||||
patch("app.services.estimator._fetch_deals", return_value=[]),
|
||||
patch("app.services.estimator._fetch_dkp_corridor", return_value=None),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_imv_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_or_fetch_yandex_valuation_cached",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator.estimate_via_cian_valuation",
|
||||
new=AsyncMock(return_value=None),
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._get_asking_sold_ratio",
|
||||
return_value=(None, None),
|
||||
),
|
||||
# IMV anchor returns below-threshold value → blended=False but anchor_present=True
|
||||
patch(
|
||||
"app.services.estimator._fetch_house_imv_anchor",
|
||||
return_value=imv_anchor_below_threshold,
|
||||
),
|
||||
patch(
|
||||
"app.services.estimator._lookup_quarter_index",
|
||||
return_value=(1.5, 30), # would apply factor=1.5 if not guarded
|
||||
),
|
||||
):
|
||||
return await estimate_quality(payload, db)
|
||||
|
||||
est = anyio.run(_run)
|
||||
|
||||
# Медиана не должна быть умножена на 1.5 (квартальный индекс заблокирован).
|
||||
base_median = round(_BASE_PPM2 * _AREA) # 6_000_000
|
||||
assert est.median_price_rub == base_median
|
||||
# explanation не содержит квартального дисклоужера
|
||||
assert "квартал" not in (est.confidence_explanation or "").lower()
|
||||
|
||||
|
||||
if __name__ == "__main__": # pragma: no cover
|
||||
raise SystemExit(pytest.main([__file__, "-q"]))
|
||||
Loading…
Add table
Reference in a new issue