foreclosures.py

import csv
import re

skiplines = ['RUN DATE', '36TH DISTRICT COURT', '421 MADISON AVENUE',
             'DETROIT              MI 48226', 'MATTHEW REPORT', 'Case Type', 'I01', 'NAME             CASE NUMBER']


def skippable(line):
    for phrase in skiplines:
        if phrase in line:
            return True

    return False


def parse_address(line):
    return dict(
        address=line[5:38].strip(),
        city=line[38:56].strip(),
        state=line[56:58].strip(),
        zip=line[59:64].strip()
    )


def parse_name(line):
    return dict(
        name=line[0:17].strip(),
        case=line[17:29].strip(),
        file_date=line[35:45].strip(),
        close_date=line[46:].strip()
    )


def merge_details(info, case, header):
    for key in info:
        case[header + '_' + key] = info[key]

    return case


keys = set()
cases = []
case = None
next_line_type = None
next_line_header = None
with open('010109-123111.txt') as infile:
    for line in infile:
        if skippable(line):
            continue

        line = line.strip()

        # P01 means the start of a new case
        if 'P01' in line:
            if case != None:
                cases.append(case)
                keys |= set(list(case.keys()))
            case = {}
            details = parse_address(line)
            case = merge_details(details, case, 'P01')
            next_line_type = 'details'
            next_line_header = 'P01'
            continue

        # Otherwise we add the details to to the existing case.
        # The line might be a second plaintiff or the start of the defendant
        if re.search(r'^[A-Z]\d\d', line):
            header = line[0:3]
            next_line_type = 'details'
            next_line_header = header
            details = parse_address(line)
            case = merge_details(details, case, header)
            continue

        # The line after a plaintiff or defendant is their name and info
        if next_line_type:
            details = parse_name(line)
            case = merge_details(details, case, next_line_header)
            next_line_header = None
            next_line_type = None

keys = sorted(keys)
print(keys)

f = open('010109-123111.results.csv', 'w')
w = csv.DictWriter(f, keys)
w.writeheader()
w.writerows(cases)
f.close()