fio-l2-cairo-mixedgeom-split

Rerun

import geopandas as gpd

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

# Inspect the data
print("Columns:", gdf.columns)
print("CRS:", gdf.crs)
print("Geometry types present:")
print(gdf.geom_type.unique())
print("\nFirst few rows:")
print(gdf.head())

import geopandas as gpd

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

# Target CRS: Egypt's National Grid (EPSG:22992 is often used for WGS84/UTM, 
# but let's check if there's a more specific one or use UTM Zone 36N which covers Egypt.
# Actually, Egypt National Grid is often EPSG:22992 or similar.
# However, standard UTM 36N (EPSG:32636) is very common.
# Let's check the prompt again: "Egypt's national grid".
# Egypt National Grid (Egypt Red Belt) is EPSG:22992.
# Let's use EPSG:22992.
TARGET_CRS = 'EPSG:22992'

# Ensure all geometries are single-part
# explode() will turn MultiPolygons/MultiLineStrings into separate single-part rows
gdf_exploded = gdf.explode(index_parts=False)

# Project to target CRS
gdf_exploded = gdf_exploded.to_crs(TARGET_CRS)

# Separate into layers
points = gdf_exploded[gdf_exploded.geometry.type == 'Point']
lines = gdf_exploded[gdf_exploded.geometry.type == 'LineString']
polygons = gdf_exploded[gdf_exploded.geometry.type == 'Polygon']

# Write to GPKG
output_file = 'heritage.gpkg'
with gpd.GeoDataFrame as gdf_empty:
    # We need to use engine='pyogrio' or similar if available for multi-layer writing 
    # but standard geopandas to_file works fine for single layers.
    # We will write them one by one.
    pass

# Writing layers
points.to_file(output_file, layer='points', driver='GPKG')
lines.to_file(output_file, layer='lines', driver='GPKG')
polygons.to_file(output_file, layer='polygons', driver='GPKG')

print(f"Successfully created {output_file}")
print(f"Layers: points ({len(points)}), lines ({len(lines)}), polygons ({len(polygons)})")
print(f"CRS: {gdf_exploded.crs}")

import geopandas as gpd
import os

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

# Target CRS: Egypt's National Grid
# EPSG:22992 is a common one for Egypt, though UTM 36N (32636) is also widely used.
# Given the prompt, I will use EPSG:22992.
TARGET_CRS = 'EPSG:22992'

# Ensure all geometries are single-part
# explode() will turn MultiPolygons/MultiLineStrings into separate single-part rows
gdf_exploded = gdf.explode(index_parts=False)

# Project to target CRS
gdf_exploded = gdf_exploded.to_crs(TARGET_CRS)

# Separate into layers
points = gdf_exploded[gdf_exploded.geometry.type == 'Point']
lines = gdf_exploded[gdf_exploded.geometry.type == 'LineString']
polygons = gdf_exploded[gdf_exploded.geometry.type == 'Polygon']

# Write to GPKG
output_file = 'heritage.gpkg'

if os.path.exists(output_file):
    os.remove(output_file)

# Writing layers
points.to_file(output_file, layer='points', driver='GPKG')
lines.to_file(output_file, layer='lines', driver='GPKG')
polygons.to_file(output_file, layer='polygons', driver='GPKG')

print(f"Successfully created {output_file}")
print(f"Layers: points ({len(points)}), lines ({len(lines)}), polygons ({len(polygons)})")
print(f"CRS: {gdf_exploded.crs}")

import geopandas as gpd

# Load the created GPKG to verify
gdf_points = gpd.read_file('heritage.gpkg', layer='points')
gdf_lines = gpd.read_file('heritage.gpkg', layer='lines')
gdf_polygons = gpd.read_file('heritage.gpkg', layer='polygons')

print("Points check:")
print(gdf_points.head())
print("\nLines check:")
print(gdf_lines.head())
print("\nPolygons check:")
print(gdf_polygons.head())

# Verify site_id is present in all
print("\nSite IDs in points:", gdf_points['site_id'].unique()[:5])
print("Site IDs in lines:", gdf_lines['site_id'].unique()[:5])
print("Site IDs in polygons:", gdf_polygons['site_id'].unique()[:5])

# Verify single part
print("\nGeometry types in points:", gdf_points.geom_type.unique())
print("Geometry types in lines:", gdf_lines.geom_type.unique())
print("Geometry types in polygons:", gdf_polygons.geom_type.unique())