Parse Forclosure Data from softcode

parse_forclosure.py

# for parsing the output of https://www.tylertech.com/solutions-products/softcode-product-suite
# parses the forclosure pdf output and produces an csv file with the location, amount owed, lender and auction date
# first run pdf2html and then this script

import bs4
import pprint
import re
import csv

f=open("sheriff_foreclosuresales_ls.html")
lines=[]
for x in f:
   lines.append(x)

objects= {}

junk = [
    '<br/>',
    '<br>',
    '</br>',
    '\n',
    '<b>',
    '</b>',
    '&#160;'
]

def clean(value):
    for j in junk:
            value = value.replace(j,'')
    return value

for key in ('Location','Current Sale','Plaintiff') :
    n = 0 
    for x in xrange(1,len(lines)):
        if key not in lines[x]:
            continue
        n = n + 1
        value = clean(lines[x+1])
        
        if n not in objects:
            aobject={
                key : value
            }
            objects[n]=aobject
        else:
            nkey = key
            if key == 'Plaintiff' :
                if re.match('[\d+,\.]+$',value):
                    nkey = 'Owed'
                    value = float(value.replace(',',''))

                    # just grab the next line
                    objects[n]['Plaintiff']=clean(lines[x+2])
                    
            objects[n][nkey]=value

with open('results.csv', 'w') as csvfile:
    fieldnames = ['Location', 'Owed', 'Current Sale', 'Plaintiff']
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
    writer.writeheader()
    for x in objects:
        writer.writerow(objects[x])