def find(self, key : str) -> 'Termin':
if (Utils.isNullOrEmpty(key)):
return None
li = [ ]
if (LanguageHelper.is_latin_char(key[0])):
li = self.__find_in_tree(key, MorphLang.EN)
else:
li = self.__find_in_tree(key, MorphLang.RU)
if (li is None):
li = self.__find_in_tree(key, MorphLang.UA)
return (li[0] if li is not None and len(li) > 0 else None)
def __correct_model(self) -> None:
tt = self.end_token.next0_
if (tt is None or tt.whitespaces_before_count > 2):
return
if (tt.is_value(":\\/.", None) or tt.is_hiphen):
tt = tt.next0_
if (isinstance(tt, NumberToken)):
tmp = io.StringIO()
print(tt.value, end="", file=tmp)
is_lat = LanguageHelper.is_latin_char(self.value[0])
self.end_token = tt
tt = tt.next0_
first_pass3427 = True
while True:
if first_pass3427: first_pass3427 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if ((isinstance(tt, TextToken)) and tt.length_char == 1 and tt.chars.is_letter):
if (not tt.is_whitespace_before or ((tt.previous is not None and tt.previous.is_hiphen))):
ch = tt.term[0]
self.end_token = tt
ch2 = chr(0)
if (LanguageHelper.is_latin_char(ch) and not is_lat):
ch2 = LanguageHelper.get_cyr_for_lat(ch)
if (ch2 != (chr(0))):
ch = ch2
elif (LanguageHelper.is_cyrillic_char(ch) and is_lat):
ch2 = LanguageHelper.get_lat_for_cyr(ch)
if (ch2 != (chr(0))):
ch = ch2
print(ch, end="", file=tmp)
continue
break
self.value = "{0}-{1}".format(self.value, Utils.toStringStringIO(tmp))
self.alt_value = MiscHelper.create_cyr_lat_alternative(self.value)
if (not self.end_token.is_whitespace_after and self.end_token.next0_ is not None and ((self.end_token.next0_.is_hiphen or self.end_token.next0_.is_char_of("\\/")))):
if (not self.end_token.next0_.is_whitespace_after and (isinstance(self.end_token.next0_.next0_, NumberToken))):
self.end_token = self.end_token.next0_.next0_
self.value = "{0}-{1}".format(self.value, self.end_token.value)
if (self.alt_value is not None):
self.alt_value = "{0}-{1}".format(self.alt_value, self.end_token.value)
def parse(t : 'Token', max_char : int, prev : 'LineToken') -> 'LineToken':
from pullenti.morph.LanguageHelper import LanguageHelper
from pullenti.ner.NumberToken import NumberToken
from pullenti.ner.TextToken import TextToken
from pullenti.ner.core.BracketHelper import BracketHelper
from pullenti.ner.core.BracketParseAttr import BracketParseAttr
from pullenti.ner.decree.DecreeReferent import DecreeReferent
if (t is None or t.end_char > max_char):
return None
res = ListHelper.LineToken(t, t)
first_pass3272 = True
while True:
if first_pass3272: first_pass3272 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= max_char)): break
if (t.is_char(':')):
if (res.is_newline_before and res.begin_token.is_value("ПРИЛОЖЕНИЕ", "ДОДАТОК")):
res.is_list_head = True
res.end_token = t
break
if (t.is_char(';')):
if (not t.is_whitespace_after):
pass
if (t.previous is not None and (isinstance(t.previous.get_referent(), DecreeReferent))):
if (not t.is_whitespace_after):
continue
if (t.next0_ is not None and (isinstance(t.next0_.get_referent(), DecreeReferent))):
continue
res.is_list_item = True
res.end_token = t
break
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
res.end_token = t
continue
if (t.is_newline_before and t != res.begin_token):
next0__ = True
if (t.previous.is_comma or t.previous.is_and or t.is_char_of("(")):
next0__ = False
elif (t.chars.is_letter or (isinstance(t, NumberToken))):
if (t.chars.is_all_lower):
next0__ = False
elif (t.previous.chars.is_letter):
next0__ = False
if (next0__):
break
res.end_token = t
if (res.begin_token.is_hiphen):
res.is_list_item = (res.begin_token.next0_ is not None and not res.begin_token.next0_.is_hiphen)
elif (res.begin_token.is_char_of("·")):
res.is_list_item = True
res.begin_token = res.begin_token.next0_
elif (res.begin_token.next0_ is not None and ((res.begin_token.next0_.is_char(')') or ((prev is not None and ((prev.is_list_item or prev.is_list_head))))))):
if (res.begin_token.length_char == 1 or (isinstance(res.begin_token, NumberToken))):
res.is_list_item = True
if ((isinstance(res.begin_token, NumberToken)) and res.begin_token.int_value is not None):
res.number = res.begin_token.int_value
elif ((isinstance(res.begin_token, TextToken)) and res.begin_token.length_char == 1):
te = res.begin_token.term
if (LanguageHelper.is_cyrillic_char(te[0])):
res.number = ((ord(te[0])) - (ord('А')))
elif (LanguageHelper.is_latin_char(te[0])):
res.number = ((ord(te[0])) - (ord('A')))
return res
def __to_full_string(self, last_name_first: bool,
lang: 'MorphLang') -> str:
id0_ = None
for a in self.slots:
if (a.type_name == PersonReferent.ATTR_IDENTITY):
s = str(a.value)
if (id0_ is None or len(s) > len(id0_)):
id0_ = s
if (id0_ is not None):
return MiscHelper.convert_first_char_upper_and_other_lower(id0_)
sss = self.get_string_value("NAMETYPE")
if (sss == "china"):
last_name_first = True
n = self.get_string_value(PersonReferent.ATTR_LASTNAME)
if (n is not None):
res = io.StringIO()
if (last_name_first):
print("{0} ".format(n), end="", file=res, flush=True)
s = self.__find_for_surname(PersonReferent.ATTR_FIRSTNAME, n,
False)
if (s is not None):
print("{0}".format(s), end="", file=res, flush=True)
if (PersonReferent.__is_initial(s)):
print('.', end="", file=res)
else:
print(' ', end="", file=res)
s = self.__find_for_surname(PersonReferent.ATTR_MIDDLENAME, n,
False)
if (s is not None):
print("{0}".format(s), end="", file=res, flush=True)
if (PersonReferent.__is_initial(s)):
print('.', end="", file=res)
else:
print(' ', end="", file=res)
if (not last_name_first):
print(n, end="", file=res)
elif (Utils.getCharAtStringIO(res, res.tell() - 1) == ' '):
Utils.setLengthStringIO(res, res.tell() - 1)
if (LanguageHelper.is_cyrillic_char(n[0])):
nl = None
for sl in self.slots:
if (sl.type_name == PersonReferent.ATTR_LASTNAME):
ss = Utils.asObjectOrNull(sl.value, str)
if (len(ss) > 0
and LanguageHelper.is_latin_char(ss[0])):
nl = ss
break
if (nl is not None):
nal = self.__find_for_surname(
PersonReferent.ATTR_FIRSTNAME, nl, False)
if (nal is None):
print(" ({0})".format(nl),
end="",
file=res,
flush=True)
elif (PersonReferent.SHOW_LASTNAME_ON_FIRST_POSITION):
print(" ({0} {1})".format(nl, nal),
end="",
file=res,
flush=True)
else:
print(" ({0} {1})".format(nal, nl),
end="",
file=res,
flush=True)
return MiscHelper.convert_first_char_upper_and_other_lower(
Utils.toStringStringIO(res))
else:
n = self.get_string_value(PersonReferent.ATTR_FIRSTNAME)
if ((n) is not None):
s = self.__find_for_surname(PersonReferent.ATTR_MIDDLENAME, n,
False)
if (s is not None):
n = "{0} {1}".format(n, s)
n = MiscHelper.convert_first_char_upper_and_other_lower(n)
nik = self.get_string_value(PersonReferent.ATTR_NICKNAME)
tit = self.__find_shortest_king_titul(False)
if (tit is not None):
n = "{0} {1}".format(tit, n)
if (nik is not None):
n = "{0} {1}".format(n, nik)
return n
return "?"
