feat(velocity): per-room-bucket breakdown in VelocityResult (#163)
* fix(velocity): :window_interval::interval cast syntax — same bug as PR #152 SQLAlchemy text() parser confuses ':window_interval' (named param) followed by '::interval' (cast operator) — exactly как было в PR #152 с :weights::jsonb. Result: psycopg видит literal ':window_interval' в SQL → syntax error → exception caught by velocity catch → return None → UI shows null. ## Fix ':window_interval::interval' → 'CAST(:window_interval AS interval)' (2 мест: sales_rows query + _get_ekb_median percentile). Pre-push code-reviewer должен был catch — добавим в feedback что для каждого text() SQL grep ':[a-z]*::' before push. Refs: PR #158 deploy verify, e2e velocity = null root cause * feat(velocity): add per-room-bucket breakdown to VelocityResult Add third SQL query (bucket_rows) aggregating deals_total_count by room_bucket ('студия','1','2','3','4+') across mapped competitors. New fields: - VelocityResult.by_room_bucket: aggregate {units, sqm, complexes_count} - sample_competitors[].by_room_bucket: {bucket: units} Bucket query wrapped в SAVEPOINT — failure degrades gracefully к empty dict без abort outer tx. CAST(:window_interval AS interval) pattern per PR #160 lesson. Test coverage: 3 new tests (aggregation, empty fallback, sample entries). Closes part of #161 --------- Co-authored-by: lekss361 <claudestars@proton.me>
This commit is contained in:
lekss361
GitHub
parent
30d78c5030
commit
d7fbaa0528
2 changed files with 177 additions and 4 deletions
|
|
@ -46,6 +46,7 @@ class VelocityResult:
|
|||
period_start: str # YYYY-MM
|
||||
period_end: str # YYYY-MM
|
||||
sample_competitors: list[dict[str, Any]] # top-5 для UI
|
||||
by_room_bucket: dict[str, dict[str, Any]] # агрегат по комнатности
|
||||
|
||||
def as_dict(self) -> dict[str, Any]:
|
||||
return {
|
||||
|
|
@ -57,6 +58,7 @@ class VelocityResult:
|
|||
"months_observed": self.months_observed,
|
||||
"period": {"start": self.period_start, "end": self.period_end},
|
||||
"sample_competitors": self.sample_competitors,
|
||||
"by_room_bucket": self.by_room_bucket,
|
||||
}
|
||||
|
||||
|
||||
|
|
@ -207,6 +209,75 @@ def compute_velocity(
|
|||
if not sales_rows:
|
||||
return None
|
||||
|
||||
# ── Step 2b: разбивка по комнатности (room_bucket) ───────────────────────
|
||||
# Тот же маппинг domrf_obj_id → project_name. Агрегируем по room_bucket
|
||||
# для отображения структуры спроса в UI.
|
||||
try:
|
||||
with db.begin_nested():
|
||||
bucket_rows = (
|
||||
db.execute(
|
||||
text(
|
||||
"""
|
||||
WITH mapped AS (
|
||||
SELECT cm.domrf_obj_id AS obj_id,
|
||||
cm.objective_complex_name
|
||||
FROM objective_complex_mapping cm
|
||||
WHERE cm.domrf_obj_id = ANY(:obj_ids)
|
||||
)
|
||||
SELECT
|
||||
m.obj_id,
|
||||
crm.room_bucket,
|
||||
SUM(crm.deals_total_count) AS units_sold,
|
||||
SUM(COALESCE(crm.deals_total_vol_m2,
|
||||
crm.deals_total_count * 45.0)) AS sqm_sold
|
||||
FROM objective_corpus_room_month crm
|
||||
JOIN mapped m ON m.objective_complex_name = crm.project_name
|
||||
WHERE crm.report_month >= (CURRENT_DATE - CAST(:window_interval AS interval))
|
||||
AND crm.deals_total_count > 0
|
||||
GROUP BY m.obj_id, crm.room_bucket
|
||||
"""
|
||||
),
|
||||
{
|
||||
"obj_ids": obj_ids,
|
||||
"window_interval": f"{months_window} months",
|
||||
},
|
||||
)
|
||||
.mappings()
|
||||
.all()
|
||||
)
|
||||
except Exception:
|
||||
logger.warning("velocity: bucket breakdown query failed, continuing without it")
|
||||
bucket_rows = []
|
||||
|
||||
# Агрегируем по room_bucket поверх всех конкурентов.
|
||||
by_bucket_agg: dict[str, dict[str, Any]] = {}
|
||||
per_comp_buckets: dict[int, dict[str, int]] = {}
|
||||
|
||||
for row in bucket_rows:
|
||||
bucket = str(row["room_bucket"])
|
||||
oid = int(row["obj_id"])
|
||||
units = int(row["units_sold"] or 0)
|
||||
sqm = float(row["sqm_sold"] or 0.0)
|
||||
|
||||
if bucket not in by_bucket_agg:
|
||||
by_bucket_agg[bucket] = {"units": 0, "sqm": 0.0, "complexes": set()}
|
||||
by_bucket_agg[bucket]["units"] += units
|
||||
by_bucket_agg[bucket]["sqm"] += sqm
|
||||
by_bucket_agg[bucket]["complexes"].add(oid)
|
||||
|
||||
if oid not in per_comp_buckets:
|
||||
per_comp_buckets[oid] = {}
|
||||
per_comp_buckets[oid][bucket] = units
|
||||
|
||||
by_room_bucket: dict[str, dict[str, Any]] = {
|
||||
bucket: {
|
||||
"units": data["units"],
|
||||
"sqm": round(data["sqm"], 0),
|
||||
"complexes_count": len(data["complexes"]),
|
||||
}
|
||||
for bucket, data in by_bucket_agg.items()
|
||||
}
|
||||
|
||||
total_sqm = sum(float(r["total_sqm"] or 0.0) for r in sales_rows)
|
||||
months_observed = max((int(r["months_with_data"] or 0) for r in sales_rows), default=0)
|
||||
period_start_dates = [r["period_start"] for r in sales_rows if r["period_start"]]
|
||||
|
|
@ -254,6 +325,7 @@ def compute_velocity(
|
|||
"obj_id": oid,
|
||||
**competitor_meta[oid],
|
||||
"total_sqm_period": round(sales_by_id.get(oid, 0.0), 0),
|
||||
"by_room_bucket": per_comp_buckets.get(oid, {}),
|
||||
}
|
||||
for oid in obj_ids
|
||||
if oid in competitor_meta
|
||||
|
|
@ -272,6 +344,7 @@ def compute_velocity(
|
|||
period_start=period_start,
|
||||
period_end=period_end,
|
||||
sample_competitors=sample,
|
||||
by_room_bucket=by_room_bucket,
|
||||
)
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -4,6 +4,7 @@ Mock-based — не требуют живой БД.
|
|||
Источник данных — objective_corpus_room_month (мигрировано с domrf_kn_sale_graph).
|
||||
Mock shape совместим: sales query возвращает те же aliases (obj_id, total_sqm,
|
||||
months_with_data, period_start, period_end) через GROUP BY domrf_obj_id.
|
||||
Третий вызов execute — bucket_rows (obj_id, room_bucket, units_sold, sqm_sold).
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
|
@ -59,11 +60,30 @@ def _sales_row(
|
|||
return r
|
||||
|
||||
|
||||
def _make_db(comp_rows: list, sales_rows: list) -> MagicMock:
|
||||
"""Сконструировать mock Session с двумя последовательными вызовами execute."""
|
||||
def _bucket_row(obj_id: int, room_bucket: str, units_sold: int, sqm_sold: float) -> MagicMock:
|
||||
r = MagicMock()
|
||||
r.__getitem__ = lambda self, k: {
|
||||
"obj_id": obj_id,
|
||||
"room_bucket": room_bucket,
|
||||
"units_sold": units_sold,
|
||||
"sqm_sold": sqm_sold,
|
||||
}[k]
|
||||
return r
|
||||
|
||||
|
||||
def _make_db(
|
||||
comp_rows: list,
|
||||
sales_rows: list,
|
||||
bucket_rows: list | None = None,
|
||||
) -> MagicMock:
|
||||
"""Сконструировать mock Session с тремя последовательными вызовами execute.
|
||||
|
||||
Порядок: comp_rows → sales_rows → bucket_rows.
|
||||
bucket_rows=None → пустой список (bucket query gracefully degraded).
|
||||
"""
|
||||
db = MagicMock()
|
||||
execute_results = []
|
||||
for rows in [comp_rows, sales_rows]:
|
||||
for rows in [comp_rows, sales_rows, bucket_rows if bucket_rows is not None else []]:
|
||||
result = MagicMock()
|
||||
result.mappings.return_value.all.return_value = rows
|
||||
execute_results.append(result)
|
||||
|
|
@ -196,7 +216,7 @@ def test_score_zero_when_no_sales_sqm():
|
|||
|
||||
|
||||
def test_as_dict_structure():
|
||||
"""as_dict() содержит все ожидаемые ключи."""
|
||||
"""as_dict() содержит все ожидаемые ключи, включая by_room_bucket."""
|
||||
vr = VelocityResult(
|
||||
competitors_count=5,
|
||||
monthly_velocity_sqm=3000.0,
|
||||
|
|
@ -207,6 +227,7 @@ def test_as_dict_structure():
|
|||
period_start="2024-11",
|
||||
period_end="2025-02",
|
||||
sample_competitors=[],
|
||||
by_room_bucket={"1": {"units": 10, "sqm": 450.0, "complexes_count": 2}},
|
||||
)
|
||||
d = vr.as_dict()
|
||||
assert "competitors_count" in d
|
||||
|
|
@ -216,6 +237,8 @@ def test_as_dict_structure():
|
|||
assert d["period"]["start"] == "2024-11"
|
||||
assert d["period"]["end"] == "2025-02"
|
||||
assert d["velocity_score"] == pytest.approx(0.333, abs=1e-3)
|
||||
assert "by_room_bucket" in d
|
||||
assert d["by_room_bucket"]["1"]["units"] == 10
|
||||
|
||||
|
||||
def test_sample_competitors_top5():
|
||||
|
|
@ -235,3 +258,80 @@ def test_sample_competitors_top5():
|
|||
assert len(result.sample_competitors) <= 5
|
||||
sqms = [c["total_sqm_period"] for c in result.sample_competitors]
|
||||
assert sqms == sorted(sqms, reverse=True)
|
||||
|
||||
|
||||
def test_by_room_bucket_aggregation():
|
||||
"""by_room_bucket агрегирует units/sqm поверх всех конкурентов корректно."""
|
||||
comp_rows = [_comp_row(1), _comp_row(2)]
|
||||
sales_rows = [
|
||||
_sales_row(1, total_sqm=3000.0, months=3),
|
||||
_sales_row(2, total_sqm=2000.0, months=3),
|
||||
]
|
||||
bucket_rows = [
|
||||
_bucket_row(1, "студия", units_sold=38, sqm_sold=1520.0),
|
||||
_bucket_row(1, "1", units_sold=22, sqm_sold=990.0),
|
||||
_bucket_row(2, "студия", units_sold=18, sqm_sold=720.0),
|
||||
_bucket_row(2, "1", units_sold=13, sqm_sold=585.0),
|
||||
]
|
||||
db = _make_db(comp_rows=comp_rows, sales_rows=sales_rows, bucket_rows=bucket_rows)
|
||||
|
||||
with patch(
|
||||
"app.services.site_finder.velocity._get_ekb_median",
|
||||
return_value=_EKB_MEDIAN_FALLBACK_SQM_PER_MONTH,
|
||||
):
|
||||
result = compute_velocity(db, parcel_geom_wkt=_PARCEL_WKT)
|
||||
|
||||
assert result is not None
|
||||
assert "студия" in result.by_room_bucket
|
||||
assert "1" in result.by_room_bucket
|
||||
# студия: 38+18=56 units, complexes from obj 1 and 2
|
||||
assert result.by_room_bucket["студия"]["units"] == 56
|
||||
assert result.by_room_bucket["студия"]["complexes_count"] == 2
|
||||
# 1-к: 22+13=35 units
|
||||
assert result.by_room_bucket["1"]["units"] == 35
|
||||
# sqm rounded
|
||||
assert result.by_room_bucket["студия"]["sqm"] == pytest.approx(2240.0)
|
||||
|
||||
|
||||
def test_by_room_bucket_empty_when_no_bucket_data():
|
||||
"""Если bucket query вернул пустой список — by_room_bucket пустой dict."""
|
||||
comp_rows = [_comp_row(1)]
|
||||
sales_rows = [_sales_row(1, total_sqm=5000.0, months=5)]
|
||||
db = _make_db(comp_rows=comp_rows, sales_rows=sales_rows, bucket_rows=[])
|
||||
|
||||
with patch(
|
||||
"app.services.site_finder.velocity._get_ekb_median",
|
||||
return_value=_EKB_MEDIAN_FALLBACK_SQM_PER_MONTH,
|
||||
):
|
||||
result = compute_velocity(db, parcel_geom_wkt=_PARCEL_WKT)
|
||||
|
||||
assert result is not None
|
||||
assert result.by_room_bucket == {}
|
||||
|
||||
|
||||
def test_sample_competitors_include_by_room_bucket():
|
||||
"""sample_competitors каждого элемента содержит by_room_bucket."""
|
||||
comp_rows = [_comp_row(1), _comp_row(2)]
|
||||
sales_rows = [
|
||||
_sales_row(1, total_sqm=6000.0, months=4),
|
||||
_sales_row(2, total_sqm=4000.0, months=4),
|
||||
]
|
||||
bucket_rows = [
|
||||
_bucket_row(1, "2", units_sold=30, sqm_sold=1800.0),
|
||||
_bucket_row(2, "2", units_sold=20, sqm_sold=1200.0),
|
||||
]
|
||||
db = _make_db(comp_rows=comp_rows, sales_rows=sales_rows, bucket_rows=bucket_rows)
|
||||
|
||||
with patch(
|
||||
"app.services.site_finder.velocity._get_ekb_median",
|
||||
return_value=_EKB_MEDIAN_FALLBACK_SQM_PER_MONTH,
|
||||
):
|
||||
result = compute_velocity(db, parcel_geom_wkt=_PARCEL_WKT)
|
||||
|
||||
assert result is not None
|
||||
for comp in result.sample_competitors:
|
||||
assert "by_room_bucket" in comp
|
||||
# obj_id=1 had bucket data
|
||||
top = result.sample_competitors[0]
|
||||
assert top["obj_id"] == 1
|
||||
assert top["by_room_bucket"].get("2") == 30
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue