From f345d41535971d9d9cbc744f03ff4f6ab38181f5 Mon Sep 17 00:00:00 2001
From: lmiranda <leobmiranda@gmail.com>
Date: Sun, 18 Jan 2026 23:44:08 -0500
Subject: [PATCH] fix: Seed multi-year housing data for rent trend charts

The seed script now inserts housing data for years 2019-2024 to
support rent trend line visualizations.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
---
 scripts/data/seed_amenity_data.py | 80 ++++++++++++++++++++++++++++---
 1 file changed, 73 insertions(+), 7 deletions(-)

diff --git a/scripts/data/seed_amenity_data.py b/scripts/data/seed_amenity_data.py
index 5e26578..cdeb3ea 100644
--- a/scripts/data/seed_amenity_data.py
+++ b/scripts/data/seed_amenity_data.py
@@ -171,9 +171,11 @@ def seed_census_housing() -> int:
 
 
 def seed_housing_mart() -> int:
-    """Seed housing mart with rental and affordability data."""
+    """Seed housing mart with rental and affordability data for multiple years."""
     engine = create_engine(DATABASE_URL)
+    total = 0
 
+    # First update existing NULL records
     with engine.begin() as conn:
         result = conn.execute(
             text(
@@ -186,10 +188,6 @@ def seed_housing_mart() -> int:
         )
         rows = [dict(row._mapping) for row in result]
 
-        if not rows:
-            print("No NULL housing mart values found")
-            return 0
-
         for row in rows:
             avg_rent = random.randint(1800, 3200)
             income = random.randint(55000, 180000)
@@ -230,9 +228,77 @@ def seed_housing_mart() -> int:
                     "renter": round(random.uniform(25, 70), 1),
                 },
             )
+            total += 1
 
-    print(f"Seeded housing mart data for {len(rows)} records")
-    return len(rows)
+    # Then insert multi-year data for trend charts
+    years = [2019, 2020, 2022, 2023, 2024]
+    with engine.begin() as conn:
+        result = conn.execute(
+            text(
+                "SELECT neighbourhood_id, name, geometry FROM public.dim_neighbourhood"
+            )
+        )
+        neighbourhoods = [dict(row._mapping) for row in result]
+
+        for n in neighbourhoods:
+            for year in years:
+                # Check if exists
+                exists = conn.execute(
+                    text(
+                        """
+                    SELECT 1 FROM public_marts.mart_neighbourhood_housing
+                    WHERE neighbourhood_id = :nid AND year = :year
+                """
+                    ),
+                    {"nid": n["neighbourhood_id"], "year": year},
+                ).fetchone()
+
+                if exists:
+                    continue
+
+                base_rent = random.randint(1800, 2800)
+                year_factor = (year - 2019) * random.randint(50, 150)
+                avg_rent = base_rent + year_factor
+                income = random.randint(55000, 180000)
+                rent_to_income = round((avg_rent * 12 / income) * 100, 2)
+
+                conn.execute(
+                    text(
+                        """
+                    INSERT INTO public_marts.mart_neighbourhood_housing
+                    (neighbourhood_id, neighbourhood_name, geometry, year,
+                     avg_rent_bachelor, avg_rent_1bed, avg_rent_2bed, avg_rent_3bed,
+                     vacancy_rate, rent_to_income_pct, affordability_index, is_affordable,
+                     median_household_income, pct_owner_occupied, pct_renter_occupied)
+                    VALUES
+                    (:nid, :name, :geom, :year,
+                     :bachelor, :onebed, :twobed, :threebed,
+                     :vacancy, :rent_income, :afford_idx, :is_afford,
+                     :income, :owner, :renter)
+                """
+                    ),
+                    {
+                        "nid": n["neighbourhood_id"],
+                        "name": n["name"],
+                        "geom": n["geometry"],
+                        "year": year,
+                        "bachelor": avg_rent - 500,
+                        "onebed": avg_rent - 300,
+                        "twobed": avg_rent,
+                        "threebed": avg_rent + 400,
+                        "vacancy": round(random.uniform(0.5, 4.5), 1),
+                        "rent_income": rent_to_income,
+                        "afford_idx": round(rent_to_income / 30 * 100, 1),
+                        "is_afford": rent_to_income <= 30,
+                        "income": income,
+                        "owner": round(random.uniform(30, 75), 1),
+                        "renter": round(random.uniform(25, 70), 1),
+                    },
+                )
+                total += 1
+
+    print(f"Seeded housing mart data for {total} records")
+    return total
 
 
 def seed_overview_mart() -> int: