Find Dutch blackface parades

sintenpietengilde_intochten.py

import re
import requests
import pandas as pd
from xml.dom.minidom import parseString as xml_parse


MAP_URL = 'http://sintenpietengilde.nl/Kaart.html'
KML_OUT = 'sintenpietengilde_kaart.kml'
XLS_OUT = 'sintenpietengilde_kaart.xlsx'

## helper functions

def el_to_list(tree, tag):
    return [i.firstChild.nodeValue for i in tree.getElementsByTagName(tag)]

def get_date(s):
    try:
        return re.compile(r'(\d+ (?:November|December))').findall(s)[0]
    except:
        return s

def get_org(s):
    try:
        return re.compile(r'\((.*)(?:<br/>|\))').findall(x)[0]
    except:
        return s

## get data

src = requests.get(MAP_URL).text
kml_extractor = re.compile(r'<kml .*</kml>')
kml = kml_extractor.findall(src)[0]
t = xml_parse(kml)

## create spreadsheet

d = pd.DataFrame({'name': el_to_list(t, 'name'),
                  'description': el_to_list(t, 'description'),
                  'coordinates': el_to_list(t, 'coordinates')}).set_index('name')

d['date'] = d['description'].map(get_date)
d['org'] = d['description'].map(get_org)

## output

with open(KML_OUT, 'w') as f:
    f.write(t.toprettyxml())

d.drop('description', 1).to_excel(XLS_OUT, 'intochten')