Python script to check GeoServer layergroups integrity

check_layergroups.py

#!/usr/bin/env python
import os
import glob 
import lxml.etree as ET
import argparse

parser = argparse.ArgumentParser()
parser.add_argument("path", help="path to GeoServer data directory")
args = parser.parse_args()

# "/usr/local/apache-tomcat-8.0.33/webapps/geoserver_2.13/data/workspaces"
_path = args.path

_lyr_dic = {}
_lg_ids = []
_lyr_ids = []

def process_layergroup(path):
    file = open(lg, "r") 
    s = file.read() 
    root = ET.fromstring(s)
    layergroup_id = root.xpath('/layerGroup/id/text()')[0]
    _lg_ids.append(layergroup_id)
    layergroup_childs = []
    layer_childs = []

    for item in root.xpath("/layerGroup/publishables/published[@type='layerGroup']/id/text()"):
        layergroup_childs.append(item)

    for item in root.xpath("/layerGroup/publishables/published[@type='layer']/id/text()"):
        layer_childs.append(item)   

    _lyr_dic[lg] = (layergroup_id, layergroup_childs, layer_childs)

def process_layer(path):
    file = open(lyr, "r") 
    s = file.read() 
    root = ET.fromstring(s)
    layer_id = root.xpath('/layer/id/text()')[0]
    _lyr_ids.append(layer_id)

def proces_dictionary_item(key, value):
    lg_id = value[0]
    lg_childs = value[1]
    l_childs = value[2]

    for child in lg_childs:
        if child not in _lg_ids:
            raise AssertionError('Child layergroup {} of layergroup {} not found in data directory'.format(child, key))

    for child in l_childs:
        if child not in _lyr_ids:
            raise AssertionError('Child layer {} of layergroup {} not found in data directory'.format(child, key))

if __name__ == '__main__':
    for root, dirs, filelist in os.walk(_path):
        if 'layergroups' in dirs: # insert logic to find the folder you want 
            lg = os.path.join(root, "layergroups")
            # one layergroups directory can have multiple layergroup.xml files
            glob_pattern = os.path.join(lg, "*.xml")
            for lg in glob.glob(glob_pattern):
                process_layergroup(lg)

        if 'layer.xml' in filelist:
            # layer directory can have only one layer.xml file
            lyr = os.path.join(root, "layer.xml")
            process_layer(lyr)

    # python 3
    # for key, value in d.items():
    # python 2
    for key, value in _lyr_dic.iteritems():    
        proces_dictionary_item(key,value)

    print("All layergroup children are accounted for.")