moving geocoding to a background task

backend/app.py

@@ -21,10 +21,9 @@ from starlette.middleware.base import BaseHTTPMiddleware
 from .config import settings
 from .data_loader import (
     clear_cache,
-    geocode_postcodes_bulk,
-    geocode_single_postcode,
     haversine_distance,
     load_school_data,
+    geocode_single_postcode,
 )
 from .data_loader import get_data_info as get_db_info
 from .database import init_db
@@ -256,7 +255,7 @@ async def get_schools(
     ]
     schools_df = df_latest[available_cols].drop_duplicates(subset=["urn"])
 
-    # Location-based search
+    # Location-based search (uses pre-geocoded data from database)
     search_coords = None
     if postcode:
         coords = geocode_single_postcode(postcode)
@@ -264,24 +263,7 @@ async def get_schools(
             search_coords = coords
             schools_df = schools_df.copy()
 
-            # Geocode school postcodes on-demand if not already cached
-            if "postcode" in schools_df.columns:
-                unique_postcodes = schools_df["postcode"].dropna().unique().tolist()
-                geocoded = geocode_postcodes_bulk(unique_postcodes)
-
-                # Add lat/long from geocoded data
-                schools_df["latitude"] = schools_df["postcode"].apply(
-                    lambda pc: geocoded.get(str(pc).strip().upper(), (None, None))[0]
-                    if pd.notna(pc)
-                    else None
-                )
-                schools_df["longitude"] = schools_df["postcode"].apply(
-                    lambda pc: geocoded.get(str(pc).strip().upper(), (None, None))[1]
-                    if pd.notna(pc)
-                    else None
-                )
-
-            # Filter by distance
+            # Filter by distance using pre-geocoded lat/long from database
             def calc_distance(row):
                 if pd.isna(row.get("latitude")) or pd.isna(row.get("longitude")):
                     return float("inf")

backend/data_loader.py

@@ -1,6 +1,9 @@
 """
 Data loading module that queries from PostgreSQL database.
 Provides efficient queries with caching and lazy loading.
+
+Note: School geocoding is handled by a separate cron job (scripts/geocode_schools.py).
+Only user search postcodes are geocoded on-demand via geocode_single_postcode().
 """
 
 import pandas as pd
@@ -15,58 +18,10 @@ from .config import settings
 from .database import SessionLocal, get_db_session
 from .models import School, SchoolResult
 
-# Cache for postcode geocoding
+# Cache for user search postcode geocoding (not for school data)
 _postcode_cache: Dict[str, Tuple[float, float]] = {}
 
 
-def geocode_postcodes_bulk(postcodes: list) -> Dict[str, Tuple[float, float]]:
-    """
-    Geocode postcodes in bulk using postcodes.io API.
-    Returns dict of postcode -> (latitude, longitude).
-    """
-    results = {}
-    
-    # Check cache first
-    uncached = []
-    for pc in postcodes:
-        if pc and isinstance(pc, str):
-            pc_upper = pc.strip().upper()
-            if pc_upper in _postcode_cache:
-                results[pc_upper] = _postcode_cache[pc_upper]
-            elif len(pc_upper) >= 5:
-                uncached.append(pc_upper)
-    
-    if not uncached:
-        return results
-    
-    uncached = list(set(uncached))
-    
-    # postcodes.io allows max 100 postcodes per request
-    batch_size = 100
-    for i in range(0, len(uncached), batch_size):
-        batch = uncached[i:i + batch_size]
-        try:
-            response = requests.post(
-                'https://api.postcodes.io/postcodes',
-                json={'postcodes': batch},
-                timeout=30
-            )
-            if response.status_code == 200:
-                data = response.json()
-                for item in data.get('result', []):
-                    if item and item.get('result'):
-                        pc = item['query'].upper()
-                        lat = item['result'].get('latitude')
-                        lon = item['result'].get('longitude')
-                        if lat and lon:
-                            results[pc] = (lat, lon)
-                            _postcode_cache[pc] = (lat, lon)
-        except Exception as e:
-            print(f"  Warning: Geocoding batch failed: {e}")
-    
-    return results
-
-
 def geocode_single_postcode(postcode: str) -> Optional[Tuple[float, float]]:
     """Geocode a single postcode using postcodes.io API."""
     if not postcode: