cleanup_geoserver.py

import os
import argparse
import xml.etree.ElementTree as ET

def get_published_layers(workspace_dir):
    published_layers = set()
    
    for root, dirs, files in os.walk(workspace_dir):
        for file in files:
            if file.endswith('.xml'):
                xml_path = os.path.join(root, file)
                try:
                    tree = ET.parse(xml_path)
                    root_elem = tree.getroot()
                    for layer in root_elem.findall(".//layer/name"):
                        published_layers.add(layer.text)
                    for coverage in root_elem.findall(".//coverage/name"):
                        published_layers.add(coverage.text)
                except Exception as e:
                    print(f"Error parsing {xml_path}: {e}")
    
    return published_layers

def scan_data_directory(data_dir, published_layers, delete=False):
    obsolete_files = []
    total_size = 0
    raster_extensions = ('.tif', '.tiff', '.img', '.ecw', '.jp2', '.nc', '.hdf', '.geotiff', '.geotif')
    vector_extensions = ('.shp', '.geojson')
    
    for root, dirs, files in os.walk(data_dir):
        for file in files:
            full_path = os.path.join(root, file)
            
            if file.lower().endswith(raster_extensions):
                layer_name = os.path.splitext(file)[0]
                if layer_name not in published_layers:
                    obsolete_files.append(full_path)
                    total_size += os.path.getsize(full_path)
                    if delete:
                        try:
                            os.remove(full_path)
                            print(f"Deleted: {full_path}")
                        except Exception as e:
                            print(f"Error deleting {full_path}: {e}")
            
            elif file.lower().endswith(vector_extensions):
                file_name, file_ext = os.path.splitext(file)
                layer_name = f"{os.path.basename(root)}:{file_name}"
                if layer_name not in published_layers:
                    for ext in ['.shp', '.shx', '.dbf', '.prj', '.cpg']:
                        file_to_remove = os.path.join(root, file_name + ext)
                        if os.path.exists(file_to_remove):
                            obsolete_files.append(file_to_remove)
                            total_size += os.path.getsize(file_to_remove)
                            if delete:
                                try:
                                    os.remove(file_to_remove)
                                    print(f"Deleted: {file_to_remove}")
                                except Exception as e:
                                    print(f"Error deleting {file_to_remove}: {e}")
    
    return obsolete_files, total_size

def main():
    parser = argparse.ArgumentParser(description="Scan GeoServer data directory for obsolete files")
    parser.add_argument("data_dir", help="Path to the GeoServer data directory")
    parser.add_argument("workspace_dir", help="Path to the GeoServer workspace directory containing XML configuration files")
    parser.add_argument("--delete", action="store_true", help="Delete obsolete files")
    
    args = parser.parse_args()
    
    try:
        published_layers = get_published_layers(args.workspace_dir)
    except Exception as e:
        print(f"Error: {e}")
        return
    
    obsolete_files, total_size = scan_data_directory(args.data_dir, published_layers, args.delete)
    
    if args.delete:
        print("Obsolete files deleted:")
    else:
        print("Obsolete files found (dry run):")
    
    for file in obsolete_files:
        print(file)
    
    print(f"\nTotal obsolete files: {len(obsolete_files)}")
    print(f"Total space used by obsolete files: {total_size / (1024 * 1024):.2f} MB")

if __name__ == "__main__":
    main()

Run with

python script.py /path/to/data/directory /path/to/workspace/directory --delete