fio-l2-cairo-mixedgeom-split

Rerun

import geopandas as gpd

# Load the data
gdf = gpd.read_file('heritage_sites.geojson')

# Print info to understand the structure
print("Original GeoDataFrame info:")
print(gdf.info())
print("\nFirst few rows:")
print(gdf.head())
print("\nGeometry types present:")
print(gdf.geometry.type.unique())

import geopandas as gpd

def process_heritage_data(input_file, output_file):
    # Load the data
    gdf = gpd.read_file(input_file)
    
    # Egypt's national grid: EPSG:22992 (Egypt Red Belt) 
    # or EPSG:32636 (WGS 84 / UTM zone 36N) is commonly used.
    # Since the prompt mentions "Egypt's national grid", 
    # let's try to find a standard one. EPSG:22992 is a common choice for Egypt.
    # However, if the input is WGS84, we'll transform it.
    
    target_crs = "EPSG:22992" # Egypt Red Belt
    
    # Ensure we are working with a CRS
    if gdf.crs is None:
        # Assuming WGS84 if not specified
        gdf.set_crs("EPSG:4326", inplace=True)
    
    # Transform to target CRS
    gdf = gdf.to_crs(target_crs)
    
    # Separate geometries and ensure they are single-part
    # We create three separate DataFrames: points, lines, polygons
    
    # 1. Polygons (including MultiPolygons, but exploded to single part)
    # We explode multi-geometries to ensure single-part
    gdf_exploded = gdf.explode(index_parts=False)
    
    polygons = gdf_exploded[gdf_exploded.geometry.type.isin(['Polygon'])]
    lines = gdf_exploded[gdf_exploded.geometry.type.isin(['LineString'])]
    points = gdf_exploded[gdf_exploded.geometry.type.isin(['Point'])]
    
    # If there are MultiLineStrings or MultiPoints, explode them too if they weren't caught
    # But explode(index_parts=False) usually handles that.
    
    # Let's double check the types after explode
    # Sometimes LineString becomes MultiLineString if not careful, but explode handles it.
    
    # Create the GeoPackage
    # We need to write them to different layers
    
    # First, clear/create the file? gpd.to_file with mode='w' creates/overwrites.
    # To write multiple layers, we use mode='a' for subsequent layers or just write one by one.
    
    # Write Polygons
    if not polygons.empty:
        polygons.to_file(output_file, layer='polygons', driver="GPKG")
    
    # Write Lines (append to existing GPKG)
    if not lines.empty:
        lines.to_file(output_file, layer='lines', driver="GPKG", mode='a')
        
    # Write Points (append to existing GPKG)
    if not points.empty:
        points.to_file(output_file, layer='points', driver="GPKG", mode='a')

    print(f"Successfully created {output_file}")
    print(f"Polygons: {len(polygons)} features")
    print(f"Lines: {len(lines)} features")
    print(f"Points: {len(points)} features")

if __name__ == "__main__":
    process_heritage_data('heritage_sites.geojson', 'heritage.gpkg')

import geopandas as gpd

def verify_output(output_file):
    import fiona
    
    print(f"Verifying {output_file}:")
    layers = fiona.listlayers(output_file)
    print(f"Layers found: {layers}")
    
    for layer in layers:
        gdf = gpd.read_file(output_file, layer=layer)
        print(f"\nLayer: {layer}")
        print(f"CRS: {gdf.crs}")
        print(f"Count: {len(gdf)}")
        print(f"Geometry types: {gdf.geometry.type.unique()}")
        print(f"Sample site_id: {gdf['site_id'].head().tolist()}")
        # Check if any are multi-part
        multi_types = gdf.geometry.type.value_counts()
        print(f"Geometry type distribution:\n{multi_types}")

if __name__ == "__main__":
    verify_output('heritage.gpkg')