def test_strip_spaces(self):
res = strip_spaces(
remove_stopwords(
expand_abbreviations(
separate_postcode(normalise(single_line(
self.address)))[0])))
self.assertEqual(res, '3BRISLEEAVENUENORTHSHIELDS')
def test_jaccard_index(self):
a1 = strip_spaces(
remove_stopwords(
expand_abbreviations(
normalise('Flat 22, 8 St. Andrews Cross, PLYMOUTH'))))
a2 = strip_spaces(
remove_stopwords(
expand_abbreviations(
normalise('Flat 22, 8 St. Andrews Cross, PLYMOUTH'))))
self.assertEqual(jaccard_index(a1, a2), 1)
a3 = ''
self.assertEqual(jaccard_index(a1, a3), 0)
a4 = strip_spaces(
remove_stopwords(
expand_abbreviations(
normalise('Flat 22, 8 St. Andrews Cross'))))
self.assertGreater(jaccard_index(a1, a4), 0.7)
a5 = strip_spaces(
remove_stopwords(expand_abbreviations(normalise('Flat 22'))))
self.assertLess(jaccard_index(a1, a5), 0.3)
to_match = sys.argv[1]
res = separate_postcode(normalise(to_match))
postcode = res[1]
to_match = res[0]
if postcode:
addresses = db.addresses.find({'postcode': postcode})
else:
res = re.split('\W+', to_match)
address = ' '.join(res[:min(len(res), 4)])
addresses = db.addresses.find(
{'phonetic': {
'$regex': '^' + phonetic(address)
}})
to_match = strip_spaces(remove_stopwords(expand_abbreviations(to_match)))
best_jaccard = 0
best_match = list()
for address in addresses:
lines = paf_to_lines(address)
line = separate_postcode(normalise(single_line(lines)))
line = strip_spaces(remove_stopwords(expand_abbreviations(line[0])))
idx = jaccard_index(to_match, line)
if idx > best_jaccard:
best_jaccard = idx
best_match = list()
best_match.append(address)
elif idx == best_jaccard:
best_match.append(address)
if len(best_match) == 1:
def test_expand_abbreviations(self):
res = expand_abbreviations(
separate_postcode(normalise(single_line(self.address)))[0])
self.assertEqual(res, '3 THE BRISLEE AVENUE NORTH SHIELDS')