Skip to content

Instantly share code, notes, and snippets.

@amatmv

amatmv/parse_comer_cnmc.py

Last active October 9, 2017 12:15
Show Gist options
  • Save amatmv/07c16238284256009e7f0588e04832a3 to your computer and use it in GitHub Desktop.
Save amatmv/07c16238284256009e7f0588e04832a3 to your computer and use it in GitHub Desktop.
Download ZIP
Update XML with REE comers info with the info of the CNMC
Raw
parse_comer_cnmc.py
# -*- coding: utf-8 -*-
import click
import csv
import re
import os
import subprocess
from xml.etree.ElementTree import Element, SubElement
from xml.etree import ElementTree as ET
from xml.dom import minidom
import sys
from fuzzywuzzy import process, utils
import unicodedata
import difflib
import pprint
######################################### INFORMATION ############################
# Script that takes the CNMC list of comers in csv format and tries to combine the
# information that gives with the XML with the REE information.
# The CNMC provides more information and the main goal is to assign the CNMC
# identification code to the companies that already are in the XMl
### FORMAT OF THE CNMC CSV:
# 0 - CODIGO CNMC;
# 1 - RAZON SOCIAL;
# 2 - DOMICILIO SOCIAL;
# 3 - C.P;
# 4 - MUNICIPIO;
# 5 - PROVINCIA;
# 6 - TLF;
# 7 - ÁMBITO ACTUACION;
# 8 - C.I.F.;
# 9 - FECHA INICIO;
# 10 - FECHA BAJA;
# 11 - PG. WEB
MATCHING_MIN_RATIO = 92
AMBIT_CODES = [
'N', # 'Nacional'
'P', # 'Peninsular'
'AL', # 'Alicante'
'AN', # 'Andorra'
'AR', # 'Aragón'
'B', # 'Barcelona'
'CAT', # 'Cataluña'
'GE', # 'Gerona'
'MU', # 'Murcia'
'RI', # 'La Rioja'
'T', # 'Tarragona'
'V', # 'Comunidad Valenciana'
'MM', # 'Mallorca-Menorca'
'IF', # 'Ibiza-Formentera'
'GC', # 'Gran Canaria'
'TF', # 'Tenerife'
'FL', # 'Fuerteventura-Lanzarote'
'PA', # 'La Palma'
'LG', # 'La Gomera'
'HI', # 'El Hierro'
'CE', # 'Ceuta'
'ML', # 'Melilla'
]
PROVINCIES = {
'Albacete': '02',
'Alicante': '03',
'Almeria': '04',
'Álava': '01',
'Asturias': '33',
'Principado de Asturias': '33',
'Badajoz': '06',
'Islas Baleares': '07',
'Barcelona': '08',
'Vizcaya': '48',
'Bizkaia': '48',
'Burgos': '09',
'Cantabria': '39',
'Castellón': '12',
'Ceuta': '51',
'Ciudad Real': '13',
'La Coruña': '15',
'Cuenca': '16',
'Cáceres': '10',
'Cádiz': '11',
'Córdoba': '14',
'Desconocida': '00',
'Extranjero': '60',
'Guipúzcoa': '20',
'Gerona': '17',
'Granada': '18',
'Guadalajara': '19',
'Huelva': '21',
'Huesca': '22',
'Jaén': '23',
'Jaen': '23',
'León': '24',
'Lérida': '25',
'Lugo': '27',
'Madrid': '28',
'Melilla': '52',
'Murcia': '30',
'Málaga': '29',
'Navarra': '31',
'Ourense': '32',
'Palencia': '34',
'Las Palmas': '35',
'Pontevedra': '36',
'La Rioja': '26',
'Salamanca': '37',
'Tenerife': '38',
'Santa Cruz de Tenerife': '38',
'Segovia': '40',
'Sevilla': '41',
'Soria': '42',
'Tarragona': '43',
'Teruel': '44',
'Toledo': '45',
'Valencia': '46',
'Valladolid': '47',
'Zamora': '49',
'Zaragoza': '50',
'Sevila': '05'
}
@click.command()
@click.option(
'-i',
'--csv-file',
default='201709_Listado Comercializadores_CNMC.csv',
help=u'Nom del fitxer de la CNMC, en format .csv a parsejar per a actualitzar la informació de les comercialitzadores. Default: "201509_Listado Comercializadores_CNMC.csv"'
)
@click.option(
'-f',
'--xml-file',
default='res_participant_comer_data.xml',
help=u'Nom del fitxer de la CNMC, en format .csv a parsejar. Default: "201509_Listado Comercializadores_CNMC.csv"'
)
@click.option(
'-o',
'--output-xml',
default='res_participant_comer_data.xml',
help=u'Nom del fitxer de sortida on anirà, en format XML, la informació de les comercialitzadores. Default: "res_participant_comer_data"'
)
def parser(**kwargs):
ree_info = {}
cnmc_info = {}
empreses = {}
foreign_companies = []
uncertain_matchings = {}
adresses = {}
def str_prep(str):
str_tr = ''.join((c for c in unicodedata.normalize('NFD', str)
if unicodedata.category(c) != 'Mn'))
return str_tr.lower().replace('.', '').replace(',', '')
def prettify(elem):
rough_string = ET.tostring(elem)
reparsed = minidom.parseString(rough_string)
return reparsed.toprettyxml(indent=" ")
def best_match(cnmc_register, empreses):
def get_key_by_ree_name(empresa):
for record, value in empreses_unicode.items():
if value == empresa:
return record
cnmc_name = cnmc_register[1]
empreses_unicode = {record: str_prep(unicode(empresa)) for (record, empresa) in empreses.items()}
name_utf = str_prep(unicode(cnmc_name))
uncertain_matchings_list = []
best = difflib.get_close_matches(name_utf, empreses_unicode.values())
if best:
if difflib.SequenceMatcher(None, name_utf, best[0]).ratio() > 0.5:
return get_key_by_ree_name(best[0])
uncertain_matchings_list.append(best[0])
new_best = process.extract(name_utf, empreses_unicode.values(), limit=5)
if new_best[0][1] >= MATCHING_MIN_RATIO:
return get_key_by_ree_name(new_best[0][0])
if difflib.SequenceMatcher(None, name_utf, new_best[0][0]).ratio()*new_best[0][1] >= 60 and difflib.SequenceMatcher(None, ''.join(name_utf.split()[0]), ''.join(new_best[0][0].split()[0])).ratio() > 0.8:
return get_key_by_ree_name(new_best[0][0])
uncertain_matchings_list.append(new_best[0][0])
if uncertain_matchings_list:
# print "participant of:", new_best[0][0], " =", get_key_by_ree_name(new_best[0][0])
uncertain_matchings.update({get_key_by_ree_name(new_best[0][0]): [uncertain_matchings_list, cnmc_register]})
return None
def xml_dict(root_element):
dic = {}
for record in root_element.iterfind('record'):
field_dict = {}
for field in record.findall('field'):
field_dict[field.attrib['name']] = field.text
if field.attrib['name'] == 'name':
empreses[record.attrib['id']] = field.text
dic[record.attrib['id']] = field_dict
return dic
def update_info(ree_matching, cnmc_company_register):
ree_info[ree_matching].update({
'ref2': cnmc_company_register[0],
'vat': cnmc_company_register[8],
'website': cnmc_company_register[11],
'ambit': cnmc_company_register[7],
'state': cnmc_company_register[5]
})
def ask(string):
r = raw_input(string)
return not r or r == 'y' or r == 'Y'
def update_foreign_companies():
for l in foreign_companies:
best = best_match(l, empreses)
if best is not None:
update_info(best, l)
# <record model="res.partner.address" id="res_partner_address_comer_0762">
# <field name="partner_id" ref="res_partner_comer_0762"/>
# <field name="name">SOM ENERGÍA, S.C.C.L.</field>
# <field name="state_id" search="[('code', '=', '17')]"/>
# <field name="id_municipi" search="[('ine', '=', '17079')]"/>
# <field name="phone">900103605</field>
# <!-- tipovia, nombre via, número, escalera, planta, puerta, cp -->
# <field name="tv" search="[('codi', '=', 'CL')]"/>
# <field name="nv">PIC DE PEGUERA</field>
# <field name="pnp">15</field>
# <field name="es">A</field>
# <field name="pt">1</field>
# <field name="pu">16</field>
# <field name="zip">17003</field>
# </record>
def create_address_record(parent, cnmc_record, info):
ree_code = cnmc_record[12:]
print info
address_record = SubElement(parent, 'record', {
'model': 'res.partner.address',
'id': 'res_partner_address_comer_{}'.format(ree_code)
})
# Participant
participant = {
'name': 'participant_id',
'ref': 'participant_{}'.format(ree_code)
}
child = SubElement(address_record, 'field', participant)
# Name of the comer
child = SubElement(address_record, 'field', {'name':'name'})
child.text = cnmc_record[9]
# State
print info.keys()
child = SubElement(address_record, 'field', {
'state_id': '[(\'code\', \'=\', \'{}\')]'.format(PROVINCIES[info['state']])}
)
# <field search="[('codi','=','13')]" model='res.comunitat_autonoma' name='comunitat_autonoma'/>
def make_xml_from_dict():
top = Element('openerp')
print "KEYS:", len(ree_info.keys())
for record in ree_info.keys():
top.insert(0,ET.Comment(record))
child = SubElement(top, 'record', {
'model': 'giscedata.participant',
'id': record
})
for field in ree_info[record].keys():
if field not in ('ambit', 'state'):
second_child = SubElement(child, 'field', {'name': field})
second_child.text = ree_info[record][field]
second_child = SubElement(child, 'field', {
'name': 'category_id',
'eval': '[(6, 0, [res_partner_cat_comer])]'
}
)
if 'ambit' in ree_info[record].keys():
for ambit in ree_info[record]['ambit'].split(';'):
if ambit in AMBIT_CODES:
second_child = SubElement(child, 'field', {
'name': 'country_id',
'ref': 'base.es'
})
break
for record in ree_info.keys():
create_address_record(top, record, ree_info[record])
with open('sortida.xml', 'w+') as xml_output:
xml_output.write(prettify(top))
xml_output.close()
##############################################################
######################### PARSER #############################
##############################################################
with open(kwargs['xml_file'], 'r') as comer_data:
xml = ET.parse(comer_data)
root_element = xml.getroot()
ree_info = xml_dict(root_element)
with open(kwargs['csv_file'], 'r') as f:
cnmc_comers = csv.reader(f, delimiter=';')
good = wrong = total = foreign_counter = 0
for l in cnmc_comers:
if l[5] not in ('Extranjero', 'PROVINCIA'):
best = best_match(l, empreses)
if best is None:
wrong += 1
else:
update_info(best, l)
good += 1
else:
foreign_companies.append(l)
foreign_counter += 1
total += 1
print "=================================================================="
print "Total of registers processed:", total
print "Correctly matched and combined:", good
print "Registers that could't be matched:", wrong
print "=================================================================="
print "The REE file has some foreign companies."
if ask("Do you want to update the foreign companies of the xml?(Y/n)\n> "):
update_foreign_companies()
print "LIST OF UNCERTAIN MATCHINGS WITH -RELATIVELY- ELEVATED MATCHING RATIO:"
for key in uncertain_matchings.keys():
print "For the company \'", key, "\' there are:"
for value in uncertain_matchings[key][0]:
print " - ", value
if ask("Do you want to resolve manually the uncertain matchings?(Y/n)\n> "):
for company in uncertain_matchings.keys():
print "For the company ->", company, ", the most similar matching are:\n", uncertain_matchings[company][0][0].decode(encoding='utf-8')
r = raw_input("Do you want to combine the information? If so, which of them? (0-" + str(len(uncertain_matchings[company][0])-1) + "/N)")
if r >= 0 and r <= len(uncertain_matchings[company][0])-1:
update_info(company, uncertain_matchings[company][r])
del(uncertain_matchings[company])
for uncertain_company in uncertain_matchings.keys():''
update_info(uncertain_company, uncertain_matchings[uncertain_company][1])
make_xml_from_dict()
if __name__ == '__main__':
parser()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment