From 3780b3cf2fea803e544d8cb86d8f8fa5fc97edc8 Mon Sep 17 00:00:00 2001
From: James Pattinson <james.pattinson@sasalliance.org>
Date: Thu, 4 Dec 2025 18:29:09 +0000
Subject: [PATCH] External DB init changes

---
 backend/Dockerfile      |  7 ++++---
 backend/entrypoint.sh   | 26 +++++++++++++++++++++++++-
 backend/seed_data.py    | 28 ++++++++++++++++------------
 docker-compose.prod.yml |  2 ++
 4 files changed, 47 insertions(+), 16 deletions(-)

diff --git a/backend/Dockerfile b/backend/Dockerfile
index 34e2bec..28fd1e9 100644
--- a/backend/Dockerfile
+++ b/backend/Dockerfile
@@ -19,11 +19,12 @@ RUN pip install --no-cache-dir -r requirements.txt
 # Copy application code
 COPY . .
 
-# Make entrypoint executable
-RUN chmod +x /app/entrypoint.sh
+# Make scripts executable
+RUN chmod +x /app/entrypoint.sh && \
+    chmod +x /app/seed_data.py
 
 # Expose port
 EXPOSE 8000
 
 # Use entrypoint script
-ENTRYPOINT ["/app/entrypoint.sh"]
\ No newline at end of file
+ENTRYPOINT ["/bin/bash", "/app/entrypoint.sh"]
\ No newline at end of file
diff --git a/backend/entrypoint.sh b/backend/entrypoint.sh
index 7f0c132..3c86422 100644
--- a/backend/entrypoint.sh
+++ b/backend/entrypoint.sh
@@ -2,7 +2,7 @@
 # Docker entrypoint script for PPR API
 # Handles database migrations and data seeding automatically
 
-set -e
+# Note: We don't use 'set -e' here because we need to handle specific exit codes from Python scripts
 
 echo "========================================="
 echo "PPR API Container Starting"
@@ -139,6 +139,30 @@ elif [ $DB_STATE -eq 0 ]; then
     else
         echo "✓ Database is up to date"
     fi
+    
+    # Check if reference data needs to be loaded
+    python3 << EOF
+from sqlalchemy import create_engine, text
+from app.core.config import settings
+
+engine = create_engine(settings.database_url)
+with engine.connect() as conn:
+    airport_count = conn.execute(text("SELECT COUNT(*) FROM airports")).fetchone()[0]
+    aircraft_count = conn.execute(text("SELECT COUNT(*) FROM aircraft")).fetchone()[0]
+    
+    if airport_count == 0 or aircraft_count == 0:
+        print("Reference data missing - will load")
+        exit(10)
+    else:
+        print(f"Reference data exists (airports: {airport_count}, aircraft: {aircraft_count})")
+        exit(0)
+EOF
+    
+    if [ $? -eq 10 ]; then
+        echo "Loading reference data..."
+        python3 /app/seed_data.py
+        echo "✓ Reference data loaded"
+    fi
 else
     echo "✗ Database check failed"
     exit 1
diff --git a/backend/seed_data.py b/backend/seed_data.py
index 655f3df..a478aba 100644
--- a/backend/seed_data.py
+++ b/backend/seed_data.py
@@ -29,13 +29,15 @@ def load_airports(db, csv_path):
     batch = []
     
     with open(csv_path, 'r', encoding='utf-8') as f:
-        reader = csv.DictReader(f)
+        reader = csv.reader(f)  # CSV has no headers
         for row in reader:
+            if len(row) < 4:
+                continue  # Skip invalid rows
             airport = Airport(
-                icao=row['icao'],
-                iata=row.get('iata') if row.get('iata') else None,
-                name=row['name'],
-                country=row['country']
+                icao=row[0].strip('"'),
+                iata=row[1].strip('"') if row[1].strip('"') else None,
+                name=row[2].strip('"'),
+                country=row[3].strip('"')
             )
             batch.append(airport)
             loaded += 1
@@ -73,15 +75,17 @@ def load_aircraft(db, csv_path):
     batch = []
     
     with open(csv_path, 'r', encoding='utf-8') as f:
-        reader = csv.DictReader(f)
+        reader = csv.reader(f)  # CSV has no headers
         for row in reader:
+            if len(row) < 6:
+                continue  # Skip invalid rows
             aircraft = Aircraft(
-                icao24=row.get('icao24') if row.get('icao24') else None,
-                registration=row.get('registration') if row.get('registration') else None,
-                manufacturer_icao=row.get('manufacturericao') if row.get('manufacturericao') else None,
-                type_code=row.get('typecode') if row.get('typecode') else None,
-                manufacturer_name=row.get('manufacturername') if row.get('manufacturername') else None,
-                model=row.get('model') if row.get('model') else None
+                icao24=row[0].strip('"') if row[0].strip('"') else None,
+                registration=row[1].strip('"') if row[1].strip('"') else None,
+                manufacturer_icao=row[2].strip('"') if row[2].strip('"') else None,
+                type_code=row[3].strip('"') if row[3].strip('"') else None,
+                manufacturer_name=row[4].strip('"') if row[4].strip('"') else None,
+                model=row[5].strip('"') if row[5].strip('"') else None
             )
             batch.append(aircraft)
             loaded += 1
diff --git a/docker-compose.prod.yml b/docker-compose.prod.yml
index 8d0cdf5..ebcb26a 100644
--- a/docker-compose.prod.yml
+++ b/docker-compose.prod.yml
@@ -36,6 +36,8 @@ services:
       - ./db-init:/db-init:ro  # Mount CSV data for seeding
     networks:
       - app_network
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
     deploy:
       resources:
         limits: