def replace_tokens(string):
for word in get_words(string):
for token in REPLACEMENT_TOKENS:
if token == word:
string = string.replace(word, REPLACEMENT_TOKENS[token])
break
return string
def remove_house_token(string):
words = get_words(string, True)
for index, word in enumerate(words):
if index not in xrange(len(words) - 1):
break
if word in HOUSE_TOKENS:
for next_words in words[index + 1:]:
if is_separator(next_words):
continue
if next_words.isdigit():
words[index] = u''
break
return u''.join(words)
def _parse(self, token_types, get_tokens, set_token, is_parse_default_token=lambda: False, known_tokens=[]):
for token_type in token_types:
words = get_words(self.address, True)
token_indexes_for_type = []
for index, word in filter(lambda (index, word): word == token_type, enumerate(words)):
if index not in xrange(len(words) - 1):
break
(token, token_indexes) = get_tokens(words, index + 1, known_tokens)
if token:
set_token(token, token_type)
token_indexes.append(index)
token_indexes_for_type.extend(token_indexes)
for index in sorted(token_indexes_for_type, reverse=True):
words.pop(index)
self.address = u''.join(words)
def spaces_round_second_hyphen_in_word(string):
for word in get_words(string):
base_word = word
index = -1
hyphen_count = 0
while True:
index = word.find('-', index + 1)
if index == -1 or index + 1 >= len(word):
break
hyphen_count += 1
if hyphen_count > 1:
word = insert_space(word, index + 1)
word = insert_space(word, index)
string = string.replace(base_word, word)
break
return string
def separate_house_and_housing(string):
words = get_words(string, True)
index = -1
while True:
index += 1
if index + 1 > len(words):
break
word = words[index]
for token in HOUSE_NUMBER_SEPARATORS:
subwords = word.split(token)
if len(subwords) == 2 and\
len(filter(is_digit_or_one_alpha, subwords)) == 2 and\
len(filter(lambda subword: subword.isdigit(), subwords)) > 0:
words[index] = subwords[0]
words.insert(index + 1, subwords[1])
words.insert(index + 1, u', корп. ' if subwords[1].isdigit() else u' ')
return u''.join(words)
def replace_house_number_tokens(string):
words = get_words(string, True)
index = -1
while True:
index += 1
if index not in xrange(len(words) - 1):
break
word = words[index]
for token in filter(lambda token: word == token, HOUSE_NUMBER_TOKENS):
for next_word in words[index + 1:]:
if is_separator(next_word):
continue
if is_digit_or_one_alpha(next_word):
words[index] = u'корп.'
for previous_index, previous_word in enumerate(words[index-1::-1]):
if previous_word == u',':
break
if is_separator(previous_word):
continue
words[index - previous_index] = u',%s' % words[index - previous_index]
break
return u''.join(words)
def rearranged_words(string):
words = get_words(string, True)
for index, word in enumerate(words):
for token in REARRANGED_WORDS:
if token == word:
previous_word_index = index
while True:
previous_word_index -= 1
if 0 > previous_word_index:
break
previous_word = words[previous_word_index]
if previous_word == ',':
break
if is_separator(previous_word):
continue
if previous_word[0].isupper():
words[index], words[previous_word_index] = words[previous_word_index], words[index]
break
string = u''.join(words)
for token in REARRANGED_WORDS_REPLACEMENT_TOKENS:
while token in string:
string = string.replace(token, REARRANGED_WORDS_REPLACEMENT_TOKENS[token])
return string
def remove_house_token_from_names(string):
words = get_words(string, True)
for index, word in enumerate(words):
if word[-len(HOUSE_TOKEN):] == HOUSE_TOKEN:
words[index] = word[:-len(HOUSE_TOKEN)]
return u''.join(words)