gistfile1.txt

#!/usr/bin/env python

# on a unix-type system, it runs like this:
# ./parse_csv.py INPUTFILE > OUTPUTFILE
# where INPUTFILE is a .csv file like the dump we were given
# it creates OUTPUTFILE, which is a .csv file like the input file,
# but with the additional parsed-out fields (OWNER_ADDR,
# OWNER_APT_SUITE, SITE_ADDR, and SITE_APT_SUITE).

import sys
import csv
import re


fields = ["APN","RNO","OWNER1","OWNER2","OWNERADDR","OWNERCITY",
          "OWNERSTATE","OWNERZIP","SITEADDR","SITECITY",
          "SITESTATE","SITEZIP"]

new_fields = ["APN","RNO","OWNER1","OWNER2","OWNERADDR",
              "OWNER_ADDR","OWNER_APT_SUITE","OWNERCITY",
              "OWNERSTATE","OWNERZIP","SITEADDR","SITE_ADDR",
              "SITE_APT_SUITE","SITECITY","SITESTATE","SITEZIP"]


suite_re = re.compile("^(.*?)\s+(#|STE #|STE|UNIT|ROOM|MS#|RM|APT|BOX|FSC)([\s\d].*?)$")

reader = csv.reader(open(sys.argv[1]))

def pull_suite(str):
    m = re.match(suite_re,str)
    if m:
        return (m.group(1), m.group(2) + m.group(3))
    else:
        return (str, "")

print ",".join(["\"%s\"" % f for f in new_fields])

header_row = reader.next()

for row in reader:
    row_fields = dict(zip(fields, row))

    (own_addr, own_ste) = pull_suite(row_fields["OWNERADDR"])
    row_fields["OWNER_ADDR"] = own_addr
    row_fields["OWNER_APT_SUITE"] = own_ste

    (site_addr, site_ste) = pull_suite(row_fields["SITEADDR"])
    row_fields["SITE_ADDR"] = site_addr
    row_fields["SITE_APT_SUITE"] = site_ste

    values = [row_fields[f] for f in new_fields]
    print ",".join(["\"%s\"" % v for v in values])