def format_address(self, country, components,
minimal_only=True, tag_components=True, replace_aliases=True,
template_replacements=False):
template = self.config.get(country.upper())
if not template:
return None
template_text = template['address_template']
if replace_aliases:
self.replace_aliases(components)
if minimal_only and not self.minimal_components(components):
return None
if template_replacements:
self.apply_replacements(template, components)
if tag_components:
template_text = self.tag_template_separators(template_text)
components = {k: u' '.join([u'{}/{}'.format(t.replace(' ', ''), k.replace(' ', '_'))
for t, c in tokenize(v)])
for k, v in components.iteritems()}
text = self.render_template(template_text, components, tagged=tag_components)
text = self.post_replacements(template, text)
return text
def format_address(self,
country,
components,
minimal_only=True,
tag_components=True,
replace_aliases=True,
template_replacements=False):
template = self.config.get(country.upper())
if not template:
return None
template_text = template['address_template']
if replace_aliases:
self.replace_aliases(components)
if minimal_only and not self.minimal_components(components):
return None
if template_replacements:
self.apply_replacements(template, components)
if tag_components:
template_text = self.tag_template_separators(template_text)
components = {
k: u' '.join([
u'{}/{}'.format(t.replace(' ', ''), k.replace(' ', '_'))
for t, c in tokenize(v)
])
for k, v in components.iteritems()
}
text = self.render_template(template_text,
components,
tagged=tag_components)
text = self.post_replacements(template, text)
return text
def disambiguate_language(text, languages):
valid_languages = OrderedDict(languages)
script_langs = {}
read_len = 0
while read_len < len(text):
script, script_len, is_ascii = get_string_script(text[read_len:])
if script != LATIN_SCRIPT:
script_langs[script] = set([
l for l, d in languages
if l in script_languages.get(script, [])
])
read_len += script_len
num_defaults = sum(
(1 for lang, default in valid_languages.iteritems() if default))
tokens = [
(c, t.rstrip('.'))
for t, c in tokenize(safe_decode(text).replace(u'-', u' ').lower())
]
current_lang = None
possible_lang = None
seen_languages = set()
for c, t, data in street_types_gazetteer.filter(tokens):
if c is token_types.PHRASE:
valid = []
data = [d.split('|') for d in data]
potentials = [l for l, c, s in data if l in valid_languages]
for lang, canonical, stopword in data:
canonical = int(canonical)
stopword = int(stopword)
if lang not in valid_languages or (stopword
and len(potentials) > 1):
continue
is_default = valid_languages[lang]
lang_valid = is_default or not seen_languages or lang in seen_languages
if lang_valid and ((canonical and not stopword) or
(is_default and len(potentials) == 1)):
valid.append(lang)
elif is_default and num_defaults > 1 and current_lang is not None and current_lang != lang:
return AMBIGUOUS_LANGUAGE
elif stopword and canonical and not is_default and lang in seen_languages:
valid.append(lang)
elif not seen_languages and len(potentials) == 1 and len(
t[0][1]) > 1:
possible_lang = lang if possible_lang is None or possible_lang == lang else None
if seen_languages and valid and not any((l in seen_languages for l in valid)) and \
(not any((valid_languages.get(l) for l in valid)) or any((valid_languages.get(l) for l in seen_languages))):
return AMBIGUOUS_LANGUAGE
if len(valid) == 1:
current_lang = valid[0]
else:
valid_default = [l for l in valid if valid_languages.get(l)]
if len(valid_default
) == 1 and current_lang is not None and valid_default[
0] != current_lang:
return AMBIGUOUS_LANGUAGE
elif len(valid_default) == 1:
current_lang = valid_default[0]
if any((current_lang not in langs
for script, langs in script_langs.iteritems())):
return AMBIGUOUS_LANGUAGE
seen_languages.update(valid)
if current_lang is not None:
return current_lang
elif possible_lang is not None:
if not any((possible_lang not in langs
for script, langs in script_langs.iteritems())):
return possible_lang
else:
return AMBIGUOUS_LANGUAGE
return UNKNOWN_LANGUAGE
