ScraperWiki for Denver Streets and Sidewalks data https://scraperwiki.com/scrapers/denver_streets_and_sidewalks/

denver_street_closures.py

# TODO -
#   Geocode addresses
#   Parse start/end date
#   

import scraperwiki
import urllib2
import re
from bs4 import BeautifulSoup

def get_dates(date_string):
    re_date = "Start Date:\s+(\d+\/\d+\/\d+)\s+-\s+End Date:\s+(\d+\/\d+\/\d+)"
    regex = re.compile(re_date)
    r = regex.search(date_string)
    return r.groups()


url = "https://www.denvergov.org/Portals/707/documents/mydenverdrive/1-22-25-2013.pdf"
xml = scraperwiki.pdftoxml(urllib2.urlopen(url).read())
parsed = BeautifulSoup(xml).text.split("\n")
filtered_list = parsed[parsed.index('Location: '):]
closures = []

i = 0
current_closure = -1

while i < len(filtered_list):
    text = filtered_list[i]
    if text == "Location: ":
        closures.append({})
        current_closure = len(closures) - 1
        i += 1
        closures[current_closure]['location'] = filtered_list[i]
    elif text == "Type: ":
        i += 1
        closures[current_closure]['type'] = filtered_list[i]
    elif text == "Date: ":
        i += 1
        dates = get_dates(filtered_list[i])
        closures[current_closure]['start_date'] = dates[0]
        closures[current_closure]['end_date'] = dates[1]
    elif text == "Time: ":
        i += 1
        closures[current_closure]['time'] = filtered_list[i]
    elif text == "Purpose: ":
        i += 1
        closures[current_closure]['purpose'] = filtered_list[i]
    elif text == "Contractor: ":
        closures[current_closure]['contractor'] = filtered_list[i]
    
    i+= 1


for closure in closures:
    scraperwiki.sqlite.save(unique_keys=['location'], data=closure)