Generar Nomenclator de referencias catastrales en Redis

cat2redis

from argparse import ArgumentParser
import os
from lxml import etree
import redis

EXEC_DIR = os.path.dirname(__file__)

root = "/home/josetomas/Descargas/catastro_alicante_2014"

gmdns = "{http://www.isotc211.org/2005/gmd}"
gcons = "{http://www.isotc211.org/2005/gco}"
gmlns = "{http://www.opengis.net/gml/3.2}"
cpns = "{urn:x-inspire:specification:gmlas:CadastralParcels:3.0}"


def savemun(rediscache, munid, epsgid, pubtype, pubtitle, pubdate):
    key = "m:{0}:{1}".format(munid, pubtype)
    t = rediscache.pipeline()
    t.hset(key, "epsgid", epsgid)
    t.hset(key, "pubtitle", pubtitle)
    t.hset(key, "pubdate", pubdate)
    t.execute()


def saveparcel(rediscache, ref, x, y, created, areaval, pubtype, munid):
    key = "{0}:{1}".format(ref, pubtype)
    t = rediscache.pipeline()
    t.hset(key, "x", x)
    t.hset(key, "y", y)
    t.hset(key, "created", created)
    #t.hset(key, "areaval", areaval)
    t.hset(key, "munid", munid)
    t.execute()
    return


def cat2redis(sourcemd, source, rediscache):
    context = etree.iterparse(sourcemd, events=("end",),
        tag= gmdns + "MD_DataIdentification")
    for event, elem in context:
        publication = elem.find(
            gmdns + "citation").find(
                gmdns + "CI_Citation")
        pubtitle = publication.find(
            gmdns + "title").find(
                gcons + "CharacterString").text
        pubid = publication.find(
            gmdns + "identifier").find(
                gmdns + "RS_Identifier").find(
                    gmdns + "authority").find(
                        gmdns + "CI_Citation").find(
                            gmdns + "title").find(
                                gcons + "CharacterString").text.rsplit(".", 2)
        pubtype = pubid[1]
        munid = pubid[2]
        pubdate = publication.find(
            gmdns + "identifier").find(
                gmdns + "RS_Identifier").find(
                    gmdns + "authority").find(
                        gmdns + "CI_Citation").find(
                            gmdns + "date").find(
                                gmdns + "CI_Date").find(
                                    gmdns + "date").find(
                                        gcons + "Date").text
        epsgid = publication.find(
            gmdns + "identifier").find(
                gmdns + "RS_Identifier").find(
                    gmdns + "code").find(
                        gcons + "CharacterString").text
        elem.clear()
        savemun(rediscache, munid, epsgid, pubtype, pubtitle, pubdate)

    context = etree.iterparse(source, events=("end",),
        tag= gmlns + "featureMember")
    count = 0
    for event, elem in context:
        cat = elem[0]
        if cat.tag == cpns + "CadastralParcel":
            ref = cat.find(cpns + "nationalCadastralReference").text
            refpoint = cat.find(cpns + "referencePoint")
            gmlpoint = refpoint.find(gmlns + "Point")
            coord = gmlpoint.find(gmlns + "pos").text.split()
            x = coord[0]
            y = coord[1]
            created = cat.find(cpns + "beginLifespanVersion").text
            areaval = cat.find(cpns + "areaValue").text
        elif cat.tag == cpns + "CadastralZoning":
            ref = cat.find(cpns + "nationalCadastalZoningReference").text
            refpoint = cat.find(cpns + "referencePoint")
            gmlpoint = refpoint.find(gmlns + "Point")
            coord = gmlpoint.find(gmlns + "pos").text.split()
            x = coord[0]
            y = coord[1]
            created = cat.find(cpns + "beginLifespanVersion").text
            areaval = "-1"
        elem.clear()
        saveparcel(rediscache, ref, x, y, created, areaval, pubtype, munid)
        count += 1
    print(("Total records written: {0}".format(count)))
    return


def main():
    parser = ArgumentParser()
    parser.add_argument("-d", "--directory", default=EXEC_DIR,
        help="Directorio en que se almacena la cartografia catastral")
    args = parser.parse_args()
    rediscache = redis.StrictRedis()
    for dirname, subdirs, files in os.walk(args.directory):
        sourcemd = None
        source = None
        if len(files) == 2:
            for f in files:
                if f.endswith(".xml"):
                    sourcemd = os.path.join(dirname, f)
                elif f.endswith(".gml"):
                    source = os.path.join(dirname, f)
        if sourcemd is not None and source is not None:
            print(("Processing file {0}".format(source)))
            cat2redis(sourcemd, source, rediscache)
    rediscache.bgsave()

"""Esta utilidad lee las referencias catastrales de ficheros GML en un
directorio y las almacena como como pares referencia/centroide en Redis.

El directorio raiz debe contener 2 subdirectorios por cada municipio:
    - uno para la cartografia de urbana;
    - otro para la cartografia de rustica.

Esta estructura de subdirectorios se puede generar desde el directorio
raiz en que se encuentren los ficheros comprimidos de catastro, ejecutando
la siguiente sentencia en bash:
    $ for f in *.zip;do unzip "$f" "-d ${f%.zip}";done

Cada subdirectorio debe contener:
    - el fichero .xml de metadatos
    - el fichero .gml de cartografia

Ejemplo:
    $ python cat2redis -d /home/user/directorio_raiz_de_ficheros_gml
"""
if __name__ == "__main__":
    main()