school-business-closing-detector.py

# Michael A. Schade, www.mschade.me

URL = 'http://www.ksdk.com/weather/severe_weather/cancellations_closings/default.aspx'

closings = {
    'School':   [],
    'Business': [],
}

def get_closings():
    """Downloads and parses the KSDK school and business cancellation page,
    returning only the data pertaining to closings that we need."""
    from urllib2 import urlopen
    data = urlopen(URL).read()
    
    from lxml import html
    data    = html.document_fromstring(data)
    try:
        data = data.get_element_by_id('schooclosings_dg')
    except KeyError:
        return None
    data    = data.findall('./tr/td/table/tr/td')[1::2]
    return zip(*[iter(data)]*2)

def add_closings():
    """Maintains the list of closed schools and businesses and notifies only
    for new cancellations."""
    added       = False
    count       = {'School': 0, 'Business': 0}
    closings    = get_closings()
    if closings is None:
        print "No closings at this time."
    else:
        for entity, place_type in closings:
            entity      = entity.text_content()
            place_type  = place_type.text_content()
            if place_type == '': place_type = 'School'
            if entity not in closings[place_type]:
                added = True
                closings[place_type].append(entity)
                count[place_type] += 1
                print 'Adding new %s: %s' % (place_type, entity)
        if added:
            import time
            now = time.strftime('%I:%M %p', time.localtime())
            print '%s) %d schools (%d new) and %d businesses (%d new) canceled as of now.' % (
                now,
                len(closings['School']),
                count['School'],
                len(closings['Business']),
                count['Business']
            )
            print '=' * 78

import time
while True:
    add_closings()
    time.sleep(120)