_parser.py

import json, urllib2, re, csv
import xml.etree.ElementTree as ET

patterns = {
	'YEAR': re.compile('(19|20)\d{2}'),
	'ZIP': re.compile('(\d{5})'),
	'MILES': re.compile('\d+,?(\d|X|k)+\smiles')
}

occurrences = {
	'CASH': re.compile('cash'),
	'OBO': re.compile('obo|obo.'),
}

def find_groups(obs,str):
	for name, pattern in patterns.iteritems():
		match = pattern.search(str)
		if match:
			obs[name] = match.group(0)
		else:
			obs[name] = ''

def find_occurrences(obs, str):
	for name, pattern in occurrences.iteritems():
		if pattern.search(str):
			obs[name] = True
		else:
			obs[name] = False

tree = ET.parse('motorcycles.xml')
root = tree.getroot()
channel = root.find('channel')
items = channel.findall('item')
print 'found ' + str(len(items)) + ' items'
with open('motorcycles.csv','wb') as csvfile:
	writer = csv.writer(csvfile, quoting=csv.QUOTE_MINIMAL)
	observations = []
	for item in items:
		obs = {}
		obs['price'] = item.find('title').text.split(':')[0].replace('$','').replace(',','')
		obs['pub_date'] = item.find('pubDate').text
		description = item.find('description').text
		if isinstance(description, basestring):
			description = description.encode('ascii','ignore').lower()
			obs['description'] = description
			find_occurrences(obs, description)
			find_groups(obs, description)
			observations.append(obs)
	writer.writerow(observations[0].keys())
	for obs in observations:
		writer.writerow(obs.values())

motorcycles.csv