python3 renameGeoserverLayers.py data_directory

renameGeoserverLayers.py

#!/usr/bin/python3

import os
import sys
import shutil
import urllib.request
import json
import xml.etree.ElementTree as etree
import subprocess

if len(sys.argv) < 2:
    print("Error: Necesitas especificar un directorio")
    sys.exit()


GEOSERVER_WORKSPACE = os.getenv("GEOSERVER_WORKSPACE")
GEOSERVER_REST = os.getenv("GEOSERVER_REST")
GEOSERVER_USER = os.getenv("GEOSERVER_USER")
GEOSERVER_PASSWORD = os.getenv("GEOSERVER_PASSWORD")


CKAN_API = "https://datos.gob.mx/ckan-admin/api/3/action/package_show?id="
DGM_API = "https://api.datos.gob.mx/v1/resources?id="

rootDirectory = sys.argv[1]

def main():
    # get processed layers
    if os.path.exists("processed_layers"):
        processed_layers = open("processed_layers","r").read().splitlines()
    else:
        processed_layers = []

    for directory in os.listdir(rootDirectory):
        if "." in directory or directory in processed_layers:
            print("Omitiendo " + directory)
            continue

        resourceId = directory.split("__")[0].replace("_","-")

        print("Procesando {}...".format(directory), end=" ", flush=True)

        resourcePath = rootDirectory + directory + "/featuretype.xml"

        tree = etree.parse(resourcePath)
        if validCoordinates( tree.getroot().find("latLonBoundingBox") ):
            with urllib.request.urlopen(DGM_API + resourceId) as dgm_response:
                dgm_result = json.loads(dgm_response.read().decode("utf-8"))

                if len(dgm_result["results"]) >= 1:
                    dgm_resource = dgm_result["results"][0]

                    if len(directory.split("__")) > 1:
                        dgm_resource["name"] = dgm_resource["name"] + " - " + directory.split("__")[1]

                    with urllib.request.urlopen(CKAN_API + dgm_resource["package-id"]) as ckan_response:
                        ckan_resource = json.loads(ckan_response.read().decode("utf-8"))["result"]

                        processFeatureType(tree, dgm_resource, ckan_resource)

                        # write new xml
                        tree.write(resourcePath, encoding="UTF-8")

                        print("Metadatos actualizados correctamente")

                else:
                    print("No se encontro el recurso", json.dumps(dgm_result))
                    deleteLayer(directory)
        else:
            print("Extensión del layer no valido")
            deleteLayer(directory)
            # shutil.rmtree(rootDirectory + directory)

    print("----- Done")

def deleteLayer(layerName):
    print("Deleting layer: " + layerName)
    request = "curl -v -u {user}:{password} \
    -X DELETE \
    {geoserver_rest}/layers/{workspace}:{layer}.xml".format(user=GEOSERVER_USER, password=GEOSERVER_PASSWORD, layer=layerName, workspace=GEOSERVER_WORKSPACE, geoserver_rest=GEOSERVER_REST)

    print(request)
    subprocess.call( request , shell=True)


def validCoordinates(bounds):
    maxx = float(bounds.find("maxx").text)
    minx = float(bounds.find("minx").text)
    miny = float(bounds.find("miny").text)
    maxy = float(bounds.find("maxy").text)

    return (-180.0000 < minx and minx < 180.0000) and (-180.0000 < maxx and maxx < 180.0000) and (-90.0000 < miny and miny < 90.0000) and (-90.0000 < maxy and maxy < 90.0000)


def processFeatureType(tree, dgm_resource, ckan_resource):
    root = tree.getroot()

    # Layer showed title
    if dgm_resource["name"] is not None:
        root.find("title").text = dgm_resource["name"]

    # abstract
    if "description" in dgm_resource:
        if root.find("abstract") is None:
            root.append( etree.Element("abstract") )

        root.find("abstract").text = dgm_resource["description"]

    # tags
    tags = []

    ## organization
    if "organization" in dgm_resource and dgm_resource["organization"] is not None:
        tags.append(dgm_resource["organization"])

    if "tags" in ckan_resource:
        for tag in ckan_resource["tags"]:
            if tag["state"] == "active":
                tags.append(tag["display_name"])

    for tag in tags:
        keywords = root.find("keywords")

        etree.SubElement(keywords, "string").text = tag

# init
main()