def __correctModel(self) -> None:
tt = self.end_token.next0_
if (tt is None or tt.whitespaces_before_count > 2):
return
if (tt.isValue(":\\/.", None) or tt.is_hiphen):
tt = tt.next0_
if (isinstance(tt, NumberToken)):
tmp = io.StringIO()
print((tt).value, end="", file=tmp)
is_lat = LanguageHelper.isLatinChar(self.value[0])
self.end_token = tt
tt = tt.next0_
first_pass3157 = True
while True:
if first_pass3157: first_pass3157 = 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.isLatinChar(ch) and not is_lat):
ch2 = LanguageHelper.getCyrForLat(ch)
if (ch2 != (chr(0))):
ch = ch2
elif (LanguageHelper.isCyrillicChar(ch) and is_lat):
ch2 = LanguageHelper.getLatForCyr(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.createCyrLatAlternative(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_.isCharOf("\\/")))):
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 find(self, key: str) -> 'Termin':
if (Utils.isNullOrEmpty(key)):
return None
li = []
if (LanguageHelper.isLatinChar(key[0])):
li = self.__FindInTree(key, MorphLang.EN)
else:
li = self.__FindInTree(key, MorphLang.RU)
if (li is None):
li = self.__FindInTree(key, MorphLang.UA)
return (li[0] if li is not None and len(li) > 0 else None)
def _mergeSlots2(self, obj : 'Referent', lang : 'MorphLang') -> None:
merge_statistic = True
for s in obj.slots:
if (s.type_name == GeoReferent.ATTR_NAME or s.type_name == GeoReferent.ATTR_TYPE):
nam = s.value
if (LanguageHelper.isLatinChar(nam[0])):
if (not lang.is_en):
continue
elif (lang.is_en):
continue
if (LanguageHelper.endsWith(nam, " ССР")):
continue
self.addSlot(s.type_name, s.value, False, (s.count if merge_statistic else 0))
if (self.findSlot(GeoReferent.ATTR_NAME, None, True) is None and obj.findSlot(GeoReferent.ATTR_NAME, None, True) is not None):
for s in obj.slots:
if (s.type_name == GeoReferent.ATTR_NAME):
self.addSlot(s.type_name, s.value, False, (s.count if merge_statistic else 0))
if (self.findSlot(GeoReferent.ATTR_TYPE, None, True) is None and obj.findSlot(GeoReferent.ATTR_TYPE, None, True) is not None):
for s in obj.slots:
if (s.type_name == GeoReferent.ATTR_TYPE):
self.addSlot(s.type_name, s.value, False, (s.count if merge_statistic else 0))
if (self.is_territory):
if (((self.alpha2 is not None or self.findSlot(GeoReferent.ATTR_TYPE, "государство", True) is not None or self.findSlot(GeoReferent.ATTR_TYPE, "держава", True) is not None) or self.findSlot(GeoReferent.ATTR_TYPE, "империя", True) is not None or self.findSlot(GeoReferent.ATTR_TYPE, "імперія", True) is not None) or self.findSlot(GeoReferent.ATTR_TYPE, "state", True) is not None):
s = self.findSlot(GeoReferent.ATTR_TYPE, "территория", True)
if (s is not None):
self.slots.remove(s)
if (self.is_state):
for s in self.slots:
if (s.type_name == GeoReferent.ATTR_TYPE and ((str(s.value) == "регион" or str(s.value) == "регіон" or str(s.value) == "region"))):
self.slots.remove(s)
break
if (self.is_city):
s = Utils.ifNotNull(self.findSlot(GeoReferent.ATTR_TYPE, "город", True), Utils.ifNotNull(self.findSlot(GeoReferent.ATTR_TYPE, "місто", True), self.findSlot(GeoReferent.ATTR_TYPE, "city", True)))
if (s is not None):
for ss in self.slots:
if (ss.type_name == GeoReferent.ATTR_TYPE and ss != s and GeoReferent.__isCity(ss.value)):
self.slots.remove(s)
break
has = False
i = 0
while i < len(self.slots):
if (self.slots[i].type_name == GeoReferent.ATTR_HIGHER):
if (not has):
has = True
else:
del self.slots[i]
i -= 1
i += 1
self._mergeExtReferents(obj)
def __toFullString(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.convertFirstCharUpperAndOtherLower(id0_)
sss = self.getStringValue("NAMETYPE")
if (sss == "china"):
last_name_first = True
n = self.getStringValue(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.__findForSurname(PersonReferent.ATTR_FIRSTNAME, n, False)
if (s is not None):
print("{0}".format(s), end="", file=res, flush=True)
if (PersonReferent.__isInitial(s)):
print('.', end="", file=res)
else:
print(' ', end="", file=res)
s = self.__findForSurname(PersonReferent.ATTR_MIDDLENAME, n, False)
if (s is not None):
print("{0}".format(s), end="", file=res, flush=True)
if (PersonReferent.__isInitial(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.isCyrillicChar(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.isLatinChar(ss[0])):
nl = ss
break
if (nl is not None):
nal = self.__findForSurname(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.convertFirstCharUpperAndOtherLower(Utils.toStringStringIO(res))
else:
n = self.getStringValue(PersonReferent.ATTR_FIRSTNAME)
if ((n) is not None):
s = self.__findForSurname(PersonReferent.ATTR_MIDDLENAME, n, False)
if (s is not None):
n = "{0} {1}".format(n, s)
n = MiscHelper.convertFirstCharUpperAndOtherLower(n)
nik = self.getStringValue(PersonReferent.ATTR_NICKNAME)
tit = self.__findShortestKingTitul(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 "?"
def parse(t: 'Token', max_char: int, prev: 'LineToken') -> 'LineToken':
from pullenti.ner.TextToken import TextToken
from pullenti.ner.NumberToken import NumberToken
from pullenti.morph.LanguageHelper import LanguageHelper
from pullenti.ner.core.BracketParseAttr import BracketParseAttr
from pullenti.ner.core.BracketHelper import BracketHelper
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_pass3004 = True
while True:
if first_pass3004: first_pass3004 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= max_char)): break
if (t.isChar(':')):
if (res.is_newline_before and res.begin_token.isValue(
"ПРИЛОЖЕНИЕ", "ДОДАТОК")):
res.is_list_head = True
res.end_token = t
break
if (t.isChar(';')):
if (not t.is_whitespace_after):
pass
if (t.previous is not None and (isinstance(
t.previous.getReferent(), DecreeReferent))):
if (not t.is_whitespace_after):
continue
if (t.next0_ is not None and (isinstance(
t.next0_.getReferent(), DecreeReferent))):
continue
res.is_list_item = True
res.end_token = t
break
if (t.isChar('(')):
br = BracketHelper.tryParse(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.isCharOf("(")):
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.isCharOf("·")):
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_.isChar(')') 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.isCyrillicChar(te[0])):
res.number = ((ord(te[0])) - (ord('А')))
elif (LanguageHelper.isLatinChar(te[0])):
res.number = ((ord(te[0])) - (ord('A')))
return res
