Fix shapefile loading issue - add startup script with fallback downloads

Dockerfile

@@ -16,6 +16,9 @@ RUN apt-get update && apt-get install -y \
     libspatialindex-dev \
     libffi-dev \
     git \
+    git-lfs \
+    curl \
+    wget \
     && rm -rf /var/lib/apt/lists/*
 
 # Set environment variables for GDAL
@@ -49,4 +52,4 @@ USER user
 
 # Command to run the application
 # Note: Hugging Face Spaces expects the app to run on port 7860
-CMD ["python", "-c", "import sys; sys.path.append('.'); from app import app; app.run(host='0.0.0.0', port=7860, debug=False)"]
+CMD ["python", "startup.py"]

plot_generator.py

@@ -20,7 +20,7 @@ class IndiaMapPlotter:
         
         Parameters:
         plots_dir (str): Directory to save plots
-        shapefile_path (str): Path to the India districts shapefile
+        shapefile_path (str): Path to India boundary shapefile
         """
         self.plots_dir = Path(plots_dir)
         self.plots_dir.mkdir(exist_ok=True)

requirements.txt

@@ -11,4 +11,5 @@ jinja2==3.1.2
 python-dateutil==2.8.2
 plotly==6.3.0
 kaleido
-geopandas
+geopandas
+shapely

startup.py

@@ -0,0 +1,118 @@
+#!/usr/bin/env python3
+"""
+Startup script to ensure all required files are available before running the main app
+"""
+import os
+import sys
+import subprocess
+import urllib.request
+import zipfile
+import geopandas as gpd
+from pathlib import Path
+
+def ensure_shapefile_exists():
+    """Ensure the India State Boundary shapefile exists"""
+    shapefile_path = "shapefiles/India_State_Boundary.shp"
+    
+    if os.path.exists(shapefile_path):
+        print(f"✓ Shapefile found at {shapefile_path}")
+        return True
+    
+    print(f"✗ Shapefile not found at {shapefile_path}")
+    
+    # Try to pull from git lfs if available
+    try:
+        if os.path.exists(".git"):
+            print("Attempting to pull LFS files...")
+            result = subprocess.run(["git", "lfs", "pull"], capture_output=True, text=True)
+            if result.returncode == 0 and os.path.exists(shapefile_path):
+                print("✓ Successfully pulled LFS files")
+                return True
+            else:
+                print("✗ Git LFS pull failed or file still missing")
+    except Exception as e:
+        print(f"✗ Git LFS not available: {e}")
+    
+    # Download alternative shapefile
+    print("Downloading alternative India boundary data...")
+    try:
+        # Create shapefiles directory
+        os.makedirs("shapefiles", exist_ok=True)
+        
+        # Download from Natural Earth (reliable source)
+        url = "https://www.naturalearthdata.com/http//www.naturalearthdata.com/download/50m/cultural/ne_50m_admin_0_countries.zip"
+        zip_path = "temp_countries.zip"
+        
+        print("Downloading Natural Earth country boundaries...")
+        urllib.request.urlretrieve(url, zip_path)
+        
+        # Extract
+        with zipfile.ZipFile(zip_path, 'r') as zip_ref:
+            zip_ref.extractall("temp_extract")
+        
+        # Load and filter for India
+        world_data = gpd.read_file("temp_extract/ne_50m_admin_0_countries.shp")
+        india_data = world_data[world_data['NAME'] == 'India']
+        
+        if len(india_data) == 0:
+            # Try alternative name matching
+            india_data = world_data[world_data['NAME_EN'] == 'India']
+        
+        if len(india_data) > 0:
+            # Save as our expected shapefile
+            india_data.to_file(shapefile_path)
+            print(f"✓ Successfully created {shapefile_path}")
+        else:
+            raise Exception("Could not find India in the world data")
+        
+        # Cleanup
+        os.remove(zip_path)
+        import shutil
+        shutil.rmtree("temp_extract", ignore_errors=True)
+        
+        return True
+        
+    except Exception as e:
+        print(f"✗ Failed to download alternative shapefile: {e}")
+        
+        # Create a simple fallback
+        print("Creating fallback boundary...")
+        try:
+            from shapely.geometry import Polygon
+            
+            # Simple India bounding box
+            india_bounds = [68.0, 6.0, 97.5, 37.0]  # [min_lon, min_lat, max_lon, max_lat]
+            
+            polygon = Polygon([
+                (india_bounds[0], india_bounds[1]),  # SW
+                (india_bounds[2], india_bounds[1]),  # SE  
+                (india_bounds[2], india_bounds[3]),  # NE
+                (india_bounds[0], india_bounds[3]),  # NW
+                (india_bounds[0], india_bounds[1])   # Close
+            ])
+            
+            gdf = gpd.GeoDataFrame([1], geometry=[polygon], crs="EPSG:4326")
+            gdf['NAME'] = 'India'
+            gdf.to_file(shapefile_path)
+            
+            print(f"✓ Created fallback boundary at {shapefile_path}")
+            return True
+            
+        except Exception as e2:
+            print(f"✗ Failed to create fallback: {e2}")
+            return False
+
+if __name__ == "__main__":
+    print("🚀 Starting CAMS Pollution Dashboard...")
+    print("Checking required files...")
+    
+    if ensure_shapefile_exists():
+        print("✓ All required files are ready")
+        print("Starting Flask application...")
+        
+        # Import and run the main app
+        from app import app
+        app.run(host='0.0.0.0', port=7860, debug=False)
+    else:
+        print("✗ Failed to ensure required files exist")
+        sys.exit(1)