def test_strip_spaces(self):
res = strip_spaces(
remove_stopwords(
expand_abbreviations(
separate_postcode(normalise(single_line(
self.address)))[0])))
self.assertEqual(res, '3BRISLEEAVENUENORTHSHIELDS')
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:
print(">> BEST MATCH")
print(best_match[0])
elif len(best_match) == 0:
print(">> NO MATCH")
else:
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')
def test_separate_postcode(self):
res = separate_postcode(normalise(single_line(self.address)))
self.assertEqual(res[0], '3 THE BRISLEE AVE NORTH SHIELDS')
self.assertEqual(res[1], 'NE30 2SQ')
def test_normalise(self):
res = normalise(single_line(self.address))
self.assertEqual(res, '3 THE BRISLEE AVE NORTH SHIELDS NE30 2SQ')
def test_single_line(self):
res = single_line(self.address)
self.assertEqual(res,
' 3 The Brislee Ave, North Shields ne30 2sq')