def tryParse(t: 'Token',
loc_onto: 'IntOntologyCollection') -> 'NamedItemToken':
if (t is None):
return None
if (isinstance(t, ReferentToken)):
r = t.getReferent()
if ((r.type_name == "PERSON" or r.type_name == "PERSONPROPERTY" or
(isinstance(r, GeoReferent)))
or r.type_name == "ORGANIZATION"):
return NamedItemToken._new1635(t, t, r, t.morph)
return None
typ = NamedItemToken.__m_types.tryParse(t, TerminParseAttr.NO)
nam = NamedItemToken.__m_names.tryParse(t, TerminParseAttr.NO)
if (typ is not None):
if (not ((isinstance(t, TextToken)))):
return None
res = NamedItemToken._new1636(typ.begin_token, typ.end_token,
typ.morph, typ.chars)
res.kind = (Utils.valToEnum(typ.termin.tag, NamedEntityKind))
res.type_value = typ.termin.canonic_text
if ((nam is not None and nam.end_token == typ.end_token
and not t.chars.is_all_lower) and (Utils.valToEnum(
nam.termin.tag, NamedEntityKind)) == res.kind):
res.name_value = nam.termin.canonic_text
res.is_wellknown = True
return res
if (nam is not None):
if (nam.begin_token.chars.is_all_lower):
return None
res = NamedItemToken._new1636(nam.begin_token, nam.end_token,
nam.morph, nam.chars)
res.kind = (Utils.valToEnum(nam.termin.tag, NamedEntityKind))
res.name_value = nam.termin.canonic_text
ok = True
if (not t.is_whitespace_before and t.previous is not None):
ok = False
elif (not t.is_whitespace_after and t.next0_ is not None):
if (t.next0_.isCharOf(",.;!?")
and t.next0_.is_whitespace_after):
pass
else:
ok = False
if (ok):
res.is_wellknown = True
res.type_value = (Utils.asObjectOrNull(nam.termin.tag2, str))
return res
adj = MiscLocationHelper.tryAttachNordWest(t)
if (adj is not None):
if (adj.morph.class0_.is_noun):
if (adj.end_token.isValue("ВОСТОК", None)):
if (adj.begin_token == adj.end_token):
return None
re = NamedItemToken._new1638(t, adj.end_token, adj.morph)
re.kind = NamedEntityKind.LOCATION
re.name_value = MiscHelper.getTextValue(
t, adj.end_token,
GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
re.is_wellknown = True
return re
return None
if (adj.whitespaces_after_count > 2):
return None
if ((isinstance(adj.end_token.next0_, ReferentToken)) and
(isinstance(adj.end_token.next0_.getReferent(), GeoReferent))):
re = NamedItemToken._new1638(t, adj.end_token.next0_,
adj.end_token.next0_.morph)
re.kind = NamedEntityKind.LOCATION
re.name_value = MiscHelper.getTextValue(
t, adj.end_token.next0_,
GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
re.is_wellknown = True
re.ref = adj.end_token.next0_.getReferent()
return re
res = NamedItemToken.tryParse(adj.end_token.next0_, loc_onto)
if (res is not None and res.kind == NamedEntityKind.LOCATION):
s = adj.getNormalCaseText(MorphClass.ADJECTIVE, True,
res.morph.gender, False)
if (s is not None):
if (res.name_value is None):
res.name_value = s.upper()
else:
res.name_value = "{0} {1}".format(
s.upper(), res.name_value)
res.type_value = (None)
res.begin_token = t
res.chars = t.chars
res.is_wellknown = True
return res
if (t.chars.is_capital_upper
and not MiscHelper.canBeStartOfSentence(t)):
npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0)
if (npt is not None and len(npt.adjectives) > 0):
test = NamedItemToken.tryParse(npt.noun.begin_token, loc_onto)
if (test is not None and test.end_token == npt.end_token
and test.type_value is not None):
test.begin_token = t
tmp = io.StringIO()
for a in npt.adjectives:
s = a.getNormalCaseText(MorphClass.ADJECTIVE, True,
test.morph.gender, False)
if (tmp.tell() > 0):
print(' ', end="", file=tmp)
print(s, end="", file=tmp)
test.name_value = Utils.toStringStringIO(tmp)
test.chars = t.chars
if (test.kind == NamedEntityKind.LOCATION):
test.is_wellknown = True
return test
if ((BracketHelper.isBracket(t, True) and t.next0_ is not None
and t.next0_.chars.is_letter)
and not t.next0_.chars.is_all_lower):
br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100)
if (br is not None):
res = NamedItemToken(t, br.end_token)
res.is_in_bracket = True
res.name_value = MiscHelper.getTextValue(
t, br.end_token, GetTextAttr.NO)
nam = NamedItemToken.__m_names.tryParse(
t.next0_, TerminParseAttr.NO)
if (nam is not None
and nam.end_token == br.end_token.previous):
res.kind = (Utils.valToEnum(nam.termin.tag,
NamedEntityKind))
res.is_wellknown = True
res.name_value = nam.termin.canonic_text
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter
and not t.chars.is_all_lower) and t.length_char > 2):
res = NamedItemToken._new1638(t, t, t.morph)
str0_ = (t).term
if (str0_.endswith("О") or str0_.endswith("И")
or str0_.endswith("Ы")):
res.name_value = str0_
else:
res.name_value = t.getNormalCaseText(None, False,
MorphGender.UNDEFINED,
False)
res.chars = t.chars
if (((not t.is_whitespace_after and t.next0_ is not None
and t.next0_.is_hiphen) and
(isinstance(t.next0_.next0_, TextToken))
and not t.next0_.next0_.is_whitespace_after)
and t.chars.is_cyrillic_letter
== t.next0_.next0_.chars.is_cyrillic_letter):
res.end_token = t.next0_.next0_
t = res.end_token
res.name_value = "{0}-{1}".format(
res.name_value,
t.getNormalCaseText(None, False, MorphGender.UNDEFINED,
False))
return res
return None
def process(self, kit: 'AnalysisKit') -> None:
ad = kit.getAnalyzerData(self)
models = TerminCollection()
objs_by_model = dict()
obj_by_names = TerminCollection()
t = kit.first_token
first_pass3158 = True
while True:
if first_pass3158: first_pass3158 = False
else: t = t.next0_
if (not (t is not None)): break
its = WeaponItemToken.tryParseList(t, 10)
if (its is None):
continue
rts = self.__tryAttach(its, False)
if (rts is not None):
for rt in rts:
rt.referent = ad.registerReferent(rt.referent)
kit.embedToken(rt)
t = (rt)
for s in rt.referent.slots:
if (s.type_name == WeaponReferent.ATTR_MODEL):
mod = str(s.value)
for k in range(2):
if (not str.isdigit(mod[0])):
li = []
wrapli2638 = RefOutArgWrapper(None)
inoutres2639 = Utils.tryGetValue(
objs_by_model, mod, wrapli2638)
li = wrapli2638.value
if (not inoutres2639):
li = list()
objs_by_model[mod] = li
if (not rt.referent in li):
li.append(rt.referent)
models.addStr(mod, li, None, False)
if (k > 0):
break
brand = rt.referent.getStringValue(
WeaponReferent.ATTR_BRAND)
if (brand is None):
break
mod = "{0} {1}".format(brand, mod)
elif (s.type_name == WeaponReferent.ATTR_NAME):
obj_by_names.add(
Termin._new117(str(s.value), rt.referent))
if (len(objs_by_model) == 0 and len(obj_by_names.termins) == 0):
return
t = kit.first_token
first_pass3159 = True
while True:
if first_pass3159: first_pass3159 = False
else: t = t.next0_
if (not (t is not None)): break
br = BracketHelper.tryParse(t, BracketParseAttr.NO, 10)
if (br is not None):
toks = obj_by_names.tryParse(t.next0_, TerminParseAttr.NO)
if (toks is not None
and toks.end_token.next0_ == br.end_token):
rt0 = ReferentToken(
Utils.asObjectOrNull(toks.termin.tag, Referent),
br.begin_token, br.end_token)
kit.embedToken(rt0)
t = (rt0)
continue
if (not ((isinstance(t, TextToken)))):
continue
if (not t.chars.is_letter):
continue
tok = models.tryParse(t, TerminParseAttr.NO)
if (tok is None):
if (not t.chars.is_all_lower):
tok = obj_by_names.tryParse(t, TerminParseAttr.NO)
if (tok is None):
continue
if (not tok.is_whitespace_after):
if (tok.end_token.next0_ is None
or not tok.end_token.next0_.isCharOf(",.)")):
if (not BracketHelper.isBracket(tok.end_token.next0_,
False)):
continue
tr = None
li = Utils.asObjectOrNull(tok.termin.tag, list)
if (li is not None and len(li) == 1):
tr = li[0]
else:
tr = (Utils.asObjectOrNull(tok.termin.tag, Referent))
if (tr is not None):
tit = WeaponItemToken.tryParse(tok.begin_token.previous, None,
False, True)
if (tit is not None and tit.typ == WeaponItemToken.Typs.BRAND):
tr.addSlot(WeaponReferent.ATTR_BRAND, tit.value, False, 0)
tok.begin_token = tit.begin_token
rt0 = ReferentToken(tr, tok.begin_token, tok.end_token)
kit.embedToken(rt0)
t = (rt0)
continue
def createNickname(pr : 'PersonReferent', t : 'Token') -> 'Token':
""" Выделить кличку
Args:
pr(PersonReferent):
t(Token): начальный токен
Returns:
Token: если не null, то последний токен клички, а в pr запишет саму кличку
"""
has_keyw = False
is_br = False
first_pass3097 = True
while True:
if first_pass3097: first_pass3097 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_hiphen or t.is_comma or t.isCharOf(".:;")):
continue
if (t.morph.class0_.is_preposition):
continue
if (t.isChar('(')):
is_br = True
continue
if ((t.isValue("ПРОЗВИЩЕ", "ПРІЗВИСЬКО") or t.isValue("КЛИЧКА", None) or t.isValue("ПСЕВДОНИМ", "ПСЕВДОНІМ")) or t.isValue("ПСЕВДО", None) or t.isValue("ПОЗЫВНОЙ", "ПОЗИВНИЙ")):
has_keyw = True
continue
break
if (not has_keyw or t is None):
return None
if (BracketHelper.isBracket(t, True)):
br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100)
if (br is not None):
ni = MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = br.end_token
tt = t.next0_
first_pass3098 = True
while True:
if first_pass3098: first_pass3098 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_comma_and):
continue
if (not BracketHelper.isBracket(tt, True)):
break
br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100)
if (br is None):
break
ni = MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
tt = br.end_token
t = tt
if (is_br and t.next0_ is not None and t.next0_.isChar(')')):
t = t.next0_
return t
else:
pli = PersonItemToken.tryAttachList(t, None, PersonItemToken.ParseAttr.NO, 10)
if (pli is not None and ((len(pli) == 1 or len(pli) == 2))):
ni = MiscHelper.getTextValue(pli[0].begin_token, pli[len(pli) - 1].end_token, GetTextAttr.NO)
if (ni is not None):
pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = pli[len(pli) - 1].end_token
if (is_br and t.next0_ is not None and t.next0_.isChar(')')):
t = t.next0_
return t
return None
def __try_parse(t : 'Token', prev : 'WeaponItemToken', after_conj : bool, attach_high : bool=False) -> 'WeaponItemToken':
if (t is None):
return None
if (BracketHelper.is_bracket(t, True)):
wit = WeaponItemToken.__try_parse(t.next0_, prev, after_conj, attach_high)
if (wit is not None):
if (wit.end_token.next0_ is None):
wit.begin_token = t
return wit
if (BracketHelper.is_bracket(wit.end_token.next0_, True)):
wit.begin_token = t
wit.end_token = wit.end_token.next0_
return wit
tok = WeaponItemToken.M_ONTOLOGY.try_parse(t, TerminParseAttr.NO)
if (tok is not None):
res = WeaponItemToken(t, tok.end_token)
res.typ = (Utils.valToEnum(tok.termin.tag, WeaponItemToken.Typs))
if (res.typ == WeaponItemToken.Typs.NOUN):
res.value = tok.termin.canonic_text
if (tok.termin.tag2 is not None):
res.is_doubt = True
tt = res.end_token.next0_
first_pass3426 = True
while True:
if first_pass3426: first_pass3426 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.whitespaces_before_count > 2):
break
wit = WeaponItemToken.__try_parse(tt, None, False, False)
if (wit is not None):
if (wit.typ == WeaponItemToken.Typs.BRAND):
res.__inner_tokens.append(wit)
tt = wit.end_token
res.end_token = tt
continue
break
if (not (isinstance(tt, TextToken))):
break
mc = tt.get_morph_class_in_dictionary()
if (mc == MorphClass.ADJECTIVE):
if (res.alt_value is None):
res.alt_value = res.value
if (res.alt_value.endswith(res.value)):
res.alt_value = res.alt_value[0:0+len(res.alt_value) - len(res.value)]
res.alt_value = "{0}{1} {2}".format(res.alt_value, tt.term, res.value)
res.end_token = tt
continue
break
return res
if (res.typ == WeaponItemToken.Typs.BRAND or res.typ == WeaponItemToken.Typs.NAME):
res.value = tok.termin.canonic_text
return res
if (res.typ == WeaponItemToken.Typs.MODEL):
res.value = tok.termin.canonic_text
if (isinstance(tok.termin.tag2, list)):
li = Utils.asObjectOrNull(tok.termin.tag2, list)
for to in li:
wit = WeaponItemToken._new2758(t, tok.end_token, Utils.valToEnum(to.tag, WeaponItemToken.Typs), to.canonic_text, tok.begin_token == tok.end_token)
res.__inner_tokens.append(wit)
if (to.additional_vars is not None and len(to.additional_vars) > 0):
wit.alt_value = to.additional_vars[0].canonic_text
res.__correct_model()
return res
nnn = MiscHelper.check_number_prefix(t)
if (nnn is not None):
tit = TransItemToken._attach_number(nnn, True)
if (tit is not None):
res = WeaponItemToken._new2759(t, tit.end_token, WeaponItemToken.Typs.NUMBER)
res.value = tit.value
res.alt_value = tit.alt_value
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter and t.chars.is_all_upper) and (t.length_char < 4)):
if ((t.next0_ is not None and ((t.next0_.is_hiphen or t.next0_.is_char('.'))) and (t.next0_.whitespaces_after_count < 2)) and (isinstance(t.next0_.next0_, NumberToken))):
res = WeaponItemToken._new2760(t, t.next0_, WeaponItemToken.Typs.MODEL, True)
res.value = t.term
res.__correct_model()
return res
if ((isinstance(t.next0_, NumberToken)) and not t.is_whitespace_after):
res = WeaponItemToken._new2760(t, t, WeaponItemToken.Typs.MODEL, True)
res.value = t.term
res.__correct_model()
return res
if (t.term == "СП" and (t.whitespaces_after_count < 3) and (isinstance(t.next0_, TextToken))):
pp = WeaponItemToken.__try_parse(t.next0_, None, False, False)
if (pp is not None and ((pp.typ == WeaponItemToken.Typs.MODEL or pp.typ == WeaponItemToken.Typs.BRAND))):
res = WeaponItemToken._new2759(t, t, WeaponItemToken.Typs.NOUN)
res.value = "ПИСТОЛЕТ"
res.alt_value = "СЛУЖЕБНЫЙ ПИСТОЛЕТ"
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter and not t.chars.is_all_lower) and t.length_char > 2):
ok = False
if (prev is not None and ((prev.typ == WeaponItemToken.Typs.NOUN or prev.typ == WeaponItemToken.Typs.MODEL or prev.typ == WeaponItemToken.Typs.BRAND))):
ok = True
elif (prev is None and t.previous is not None and t.previous.is_comma_and):
ok = True
if (ok):
res = WeaponItemToken._new2760(t, t, WeaponItemToken.Typs.NAME, True)
res.value = t.term
if ((t.next0_ is not None and t.next0_.is_hiphen and (isinstance(t.next0_.next0_, TextToken))) and t.next0_.next0_.chars == t.chars):
res.value = "{0}-{1}".format(res.value, t.next0_.next0_.term)
res.end_token = t.next0_.next0_
if (prev is not None and prev.typ == WeaponItemToken.Typs.NOUN):
res.typ = WeaponItemToken.Typs.BRAND
if (res.end_token.next0_ is not None and res.end_token.next0_.is_hiphen and (isinstance(res.end_token.next0_.next0_, NumberToken))):
res.typ = WeaponItemToken.Typs.MODEL
res.__correct_model()
elif (not res.end_token.is_whitespace_after and (isinstance(res.end_token.next0_, NumberToken))):
res.typ = WeaponItemToken.Typs.MODEL
res.__correct_model()
return res
if (t.is_value("МАРКА", None)):
res = WeaponItemToken.__try_parse(t.next0_, prev, after_conj, False)
if (res is not None and res.typ == WeaponItemToken.Typs.BRAND):
res.begin_token = t
return res
if (BracketHelper.can_be_start_of_sequence(t.next0_, True, False)):
br = BracketHelper.try_parse(t.next0_, BracketParseAttr.NO, 100)
if (br is not None):
return WeaponItemToken._new2764(t, br.end_token, WeaponItemToken.Typs.BRAND, MiscHelper.get_text_value(br.begin_token, br.end_token, GetTextAttr.NO))
if (((isinstance(t, TextToken)) and (isinstance(t.next0_, TextToken)) and t.next0_.length_char > 1) and not t.next0_.chars.is_all_lower):
return WeaponItemToken._new2764(t, t.next0_, WeaponItemToken.Typs.BRAND, t.term)
if (t.is_value("КАЛИБР", "КАЛІБР")):
tt1 = t.next0_
if (tt1 is not None and ((tt1.is_hiphen or tt1.is_char(':')))):
tt1 = tt1.next0_
num = NumbersWithUnitToken.try_parse(tt1, None, False, False, False, False)
if (num is not None and num.single_val is not None):
return WeaponItemToken._new2764(t, num.end_token, WeaponItemToken.Typs.CALIBER, NumberHelper.double_to_string(num.single_val))
if (isinstance(t, NumberToken)):
num = NumbersWithUnitToken.try_parse(t, None, False, False, False, False)
if (num is not None and num.single_val is not None):
if (len(num.units) == 1 and num.units[0].unit is not None and num.units[0].unit.name_cyr == "мм"):
return WeaponItemToken._new2764(t, num.end_token, WeaponItemToken.Typs.CALIBER, NumberHelper.double_to_string(num.single_val))
if (num.end_token.next0_ is not None and num.end_token.next0_.is_value("КАЛИБР", "КАЛІБР")):
return WeaponItemToken._new2764(t, num.end_token.next0_, WeaponItemToken.Typs.CALIBER, NumberHelper.double_to_string(num.single_val))
if (t.is_value("ПРОИЗВОДСТВО", "ВИРОБНИЦТВО")):
tt1 = t.next0_
if (tt1 is not None and ((tt1.is_hiphen or tt1.is_char(':')))):
tt1 = tt1.next0_
if (isinstance(tt1, ReferentToken)):
if ((isinstance(tt1.get_referent(), OrganizationReferent)) or (isinstance(tt1.get_referent(), GeoReferent))):
return WeaponItemToken._new2769(t, tt1, WeaponItemToken.Typs.DEVELOPER, tt1.get_referent())
return None
def tryParse(
self,
t0: 'Token',
pars: 'TerminParseAttr' = TerminParseAttr.NO) -> 'TerminToken':
""" Попробовать привязать термин
Args:
t0(Token):
fullWordsOnly:
"""
from pullenti.ner.core.MiscHelper import MiscHelper
from pullenti.ner.core.BracketHelper import BracketHelper
if (t0 is None):
return None
term = None
if (isinstance(t0, TextToken)):
term = (t0).term
if (self.acronym_smart is not None
and (((pars) &
(TerminParseAttr.FULLWORDSONLY))) == (TerminParseAttr.NO)
and term is not None):
if (self.acronym_smart == term):
if (t0.next0_ is not None and t0.next0_.isChar('.')
and not t0.is_whitespace_after):
return TerminToken._new606(t0, t0.next0_, self)
else:
return TerminToken._new606(t0, t0, self)
t1 = Utils.asObjectOrNull(t0, TextToken)
tt = Utils.asObjectOrNull(t0, TextToken)
i = 0
while i < len(self.acronym):
if (tt is None):
break
term1 = tt.term
if (len(term1) != 1 or tt.is_whitespace_after):
break
if (i > 0 and tt.is_whitespace_before):
break
if (term1[0] != self.acronym[i]):
break
if (tt.next0_ is None or not tt.next0_.isChar('.')):
break
t1 = (Utils.asObjectOrNull(tt.next0_, TextToken))
tt = (Utils.asObjectOrNull(tt.next0_.next0_, TextToken))
i += 1
if (i >= len(self.acronym)):
return TerminToken._new606(t0, t1, self)
if (self.acronym is not None and term is not None
and self.acronym == term):
if (t0.chars.is_all_upper or self.acronym_can_be_lower
or ((not t0.chars.is_all_lower and len(term) >= 3))):
return TerminToken._new606(t0, t0, self)
if (self.acronym is not None and t0.chars.is_last_lower
and t0.length_char > 3):
if (t0.isValue(self.acronym, None)):
return TerminToken._new606(t0, t0, self)
cou = 0
i = 0
while i < len(self.terms):
if (self.terms[i].is_hiphen):
cou -= 1
else:
cou += 1
i += 1
if (len(self.terms) > 0
and ((not self.ignore_terms_order or cou == 1))):
t1 = t0
tt = t0
e0_ = None
eup = None
ok = True
mc = None
dont_change_mc = False
i = 0
first_pass2812 = True
while True:
if first_pass2812: first_pass2812 = False
else: i += 1
if (not (i < len(self.terms))): break
if (self.terms[i].is_hiphen):
continue
if (tt is not None and tt.is_hiphen and i > 0):
tt = tt.next0_
if (i > 0 and tt is not None):
if ((((pars) & (TerminParseAttr.IGNOREBRACKETS))) !=
(TerminParseAttr.NO) and not tt.chars.is_letter
and BracketHelper.isBracket(tt, False)):
tt = tt.next0_
if (((((pars) & (TerminParseAttr.CANBEGEOOBJECT))) !=
(TerminParseAttr.NO) and i > 0 and
(isinstance(tt, ReferentToken)))
and tt.getReferent().type_name == "GEO"):
tt = tt.next0_
if ((isinstance(tt, ReferentToken)) and e0_ is None):
eup = tt
e0_ = (tt).end_token
tt = (tt).begin_token
if (tt is None):
ok = False
break
if (not self.terms[i].checkByToken(tt)):
if (tt.next0_ is not None and tt.isChar('.')
and self.terms[i].checkByToken(tt.next0_)):
tt = tt.next0_
elif (((i > 0 and tt.next0_ is not None and
(isinstance(tt, TextToken))) and
((tt.morph.class0_.is_preposition
or MiscHelper.isEngArticle(tt)))
and self.terms[i].checkByToken(tt.next0_))
and not self.terms[i - 1].is_pattern_any):
tt = tt.next0_
else:
ok = False
if (((i + 2) < len(self.terms))
and self.terms[i + 1].is_hiphen
and self.terms[i + 2].checkByPrefToken(
self.terms[i],
Utils.asObjectOrNull(tt, TextToken))):
i += 2
ok = True
elif (((not tt.is_whitespace_after
and tt.next0_ is not None and
(isinstance(tt, TextToken))) and
(tt).length_char == 1
and tt.next0_.isCharOf("\"'`’“”"))
and not tt.next0_.is_whitespace_after
and (isinstance(tt.next0_.next0_, TextToken))):
if (self.terms[i].checkByStrPrefToken(
(tt).term,
Utils.asObjectOrNull(
tt.next0_.next0_, TextToken))):
ok = True
tt = tt.next0_.next0_
if (not ok):
if (i > 0 and
(((pars) &
(TerminParseAttr.IGNORESTOPWORDS))) !=
(TerminParseAttr.NO)):
if (isinstance(tt, TextToken)):
if (not tt.chars.is_letter):
tt = tt.next0_
i -= 1
continue
mc1 = tt.getMorphClassInDictionary()
if (mc1.is_conjunction
or mc1.is_preposition):
tt = tt.next0_
i -= 1
continue
if (isinstance(tt, NumberToken)):
tt = tt.next0_
i -= 1
continue
break
if (tt.morph.items_count > 0 and not dont_change_mc):
mc = MorphCollection(tt.morph)
if (((mc.class0_.is_noun or mc.class0_.is_verb))
and not mc.class0_.is_adjective):
if (((i + 1) < len(self.terms))
and self.terms[i + 1].is_hiphen):
pass
else:
dont_change_mc = True
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
dont_change_mc = True
if (tt == e0_):
tt = eup
eup = (None)
e0_ = (None)
if (e0_ is None):
t1 = tt
tt = tt.next0_
if (ok and i >= len(self.terms)):
if (t1.next0_ is not None and t1.next0_.isChar('.')
and self.abridges is not None):
for a in self.abridges:
if (a.tryAttach(t0) is not None):
t1 = t1.next0_
break
if (t0 != t1 and t0.morph.class0_.is_adjective):
npt = NounPhraseHelper.tryParse(t0, NounPhraseParseAttr.NO,
0)
if (npt is not None and npt.end_char <= t1.end_char):
mc = npt.morph
return TerminToken._new611(t0, t1, mc)
if (len(self.terms) > 1 and self.ignore_terms_order):
terms_ = list(self.terms)
t1 = t0
tt = t0
while len(terms_) > 0:
if (tt != t0 and tt is not None and tt.is_hiphen):
tt = tt.next0_
if (tt is None):
break
j = 0
while j < len(terms_):
if (terms_[j].checkByToken(tt)):
break
j += 1
if (j >= len(terms_)):
if (tt != t0 and (((pars) &
(TerminParseAttr.IGNORESTOPWORDS))) !=
(TerminParseAttr.NO)):
if (isinstance(tt, TextToken)):
if (not tt.chars.is_letter):
tt = tt.next0_
continue
mc1 = tt.getMorphClassInDictionary()
if (mc1.is_conjunction or mc1.is_preposition):
tt = tt.next0_
continue
if (isinstance(tt, NumberToken)):
tt = tt.next0_
continue
break
del terms_[j]
t1 = tt
tt = tt.next0_
for i in range(len(terms_) - 1, -1, -1):
if (terms_[i].is_hiphen):
del terms_[i]
if (len(terms_) == 0):
return TerminToken(t0, t1)
if (self.abridges is not None and
(((pars) &
(TerminParseAttr.FULLWORDSONLY))) == (TerminParseAttr.NO)):
res = None
for a in self.abridges:
r = a.tryAttach(t0)
if (r is None):
continue
if (r.abridge_without_point and len(self.terms) > 0):
if (not ((isinstance(t0, TextToken)))):
continue
if (a.parts[0].value != (t0).term):
continue
if (res is None or (res.length_char < r.length_char)):
res = r
if (res is not None):
return res
return None
def __try_parse_ru(first: 'Token',
typ: 'NounPhraseParseAttr',
max_char_pos: int,
def_noun: 'NounPhraseItem' = None) -> 'NounPhraseToken':
if (first is None):
return None
items = None
adverbs = None
prep = None
kak = False
t0 = first
if ((((typ) & (NounPhraseParseAttr.PARSEPREPOSITION))) !=
(NounPhraseParseAttr.NO) and t0.is_value("КАК", None)):
t0 = t0.next0_
prep = PrepositionHelper.try_parse(t0)
if (prep is not None):
t0 = prep.end_token.next0_
kak = True
internal_noun_prase = None
conj_before = False
t = t0
first_pass3041 = True
while True:
if first_pass3041: first_pass3041 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char_pos > 0 and t.begin_char > max_char_pos):
break
if ((t.morph.class0_.is_conjunction
and not t.morph.class0_.is_adjective
and not t.morph.class0_.is_pronoun)
and not t.morph.class0_.is_noun):
if (conj_before):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
if (items is not None and ((t.is_and or t.is_or))):
conj_before = True
if ((t.next0_ is not None and t.next0_.is_char_of("\\/")
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or):
t = t.next0_.next0_
if (((t.next0_ is not None and t.next0_.is_char('(')
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or
and t.next0_.next0_.next0_ is not None)
and t.next0_.next0_.next0_.is_char(')')):
t = t.next0_.next0_.next0_
continue
break
elif (t.is_comma):
if (conj_before or items is None):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
mc = t.previous.get_morph_class_in_dictionary()
if (mc.is_proper_surname or mc.is_proper_secname):
break
conj_before = True
if (kak and t.next0_ is not None
and t.next0_.is_value("ТАК", None)):
t = t.next0_
if (t.next0_ is not None and t.next0_.is_and):
t = t.next0_
pr = PrepositionHelper.try_parse(t.next0_)
if (pr is not None):
t = pr.end_token
if (items[len(items) - 1].can_be_noun
and items[len(items) -
1].end_token.morph.class0_.is_pronoun):
break
continue
elif (t.is_char('(')):
if (items is None):
return None
brr = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (brr is None):
break
if (brr.length_char > 100):
break
t = brr.end_token
continue
if (isinstance(t, ReferentToken)):
if ((((typ) & (NounPhraseParseAttr.REFERENTCANBENOUN))) == (
NounPhraseParseAttr.NO)):
break
elif (t.chars.is_latin_letter):
break
it = NounPhraseItem.try_parse(t, items, typ)
if (it is None or ((not it.can_be_adj and not it.can_be_noun))):
if (((it is not None and items is not None
and t.chars.is_capital_upper) and
(t.whitespaces_before_count < 3) and t.length_char > 3)
and not t.get_morph_class_in_dictionary().is_noun and
not t.get_morph_class_in_dictionary().is_adjective):
it.can_be_noun = True
items.append(it)
break
if ((((typ) & (NounPhraseParseAttr.PARSEADVERBS))) !=
(NounPhraseParseAttr.NO) and (isinstance(t, TextToken))
and t.morph.class0_.is_adverb):
if (adverbs is None):
adverbs = list()
adverbs.append(Utils.asObjectOrNull(t, TextToken))
continue
break
it.conj_before = conj_before
conj_before = False
if (not it.can_be_adj and not it.can_be_noun):
break
if (t.is_newline_before and t != first):
if ((((typ) & (NounPhraseParseAttr.MULTILINES))) !=
(NounPhraseParseAttr.NO)):
pass
elif (items is not None
and t.chars != items[len(items) - 1].chars):
if (t.chars.is_all_lower
and items[len(items) - 1].chars.is_capital_upper):
pass
else:
break
if (items is None):
items = list()
else:
it0 = items[len(items) - 1]
if (it0.can_be_noun and it0.is_personal_pronoun):
if (it.is_pronoun):
break
if ((it0.begin_token.previous is not None
and it0.begin_token.previous.
get_morph_class_in_dictionary().is_verb
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_adjective)
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_preposition):
if (t.morph.case_.is_nominative
or t.morph.case_.is_accusative):
pass
else:
break
if (it.can_be_noun and it.is_verb):
if (it0.previous is None):
pass
elif ((isinstance(it0.previous, TextToken))
and not it0.previous.chars.is_letter):
pass
else:
break
items.append(it)
t = it.end_token
if (t.is_newline_after and not t.chars.is_all_lower):
mc = t.get_morph_class_in_dictionary()
if (mc.is_proper_surname):
break
if (t.morph.class0_.is_proper_surname and mc.is_undefined):
break
if (items is None):
return None
tt1 = None
if (len(items) == 1 and items[0].can_be_adj):
and0_ = False
tt1 = items[0].end_token.next0_
first_pass3042 = True
while True:
if first_pass3042: first_pass3042 = False
else: tt1 = tt1.next0_
if (not (tt1 is not None)): break
if (tt1.is_and or tt1.is_or):
and0_ = True
break
if (tt1.is_comma or tt1.is_value("НО", None)
or tt1.is_value("ТАК", None)):
continue
break
if (and0_):
if (items[0].can_be_noun and items[0].is_personal_pronoun):
and0_ = False
if (and0_):
tt2 = tt1.next0_
if (tt2 is not None and tt2.morph.class0_.is_preposition):
tt2 = tt2.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt2, typ, max_char_pos, None)
if (npt1 is not None and len(npt1.adjectives) > 0):
ok1 = False
for av in items[0].adj_morph:
for v in npt1.noun.noun_morph:
if (v.check_accord(av, False, False)):
items[0].morph.add_item(av)
ok1 = True
if (ok1):
npt1.begin_token = items[0].begin_token
npt1.end_token = tt1.previous
npt1.adjectives.clear()
npt1.adjectives.append(items[0])
return npt1
if (def_noun is not None):
items.append(def_noun)
last1 = items[len(items) - 1]
check = True
for it in items:
if (not it.can_be_adj):
check = False
break
elif (it.can_be_noun and it.is_personal_pronoun):
check = False
break
tt1 = last1.end_token.next0_
if ((tt1 is not None and check and
((tt1.morph.class0_.is_preposition
or tt1.morph.case_.is_instrumental)))
and (tt1.whitespaces_before_count < 2)):
inp = NounPhraseHelper.try_parse(
tt1,
Utils.valToEnum((typ) | (NounPhraseParseAttr.PARSEPREPOSITION),
NounPhraseParseAttr), max_char_pos, None)
if (inp is not None):
tt1 = inp.end_token.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt1, typ, max_char_pos, None)
if (npt1 is not None):
ok = True
ii = 0
first_pass3043 = True
while True:
if first_pass3043: first_pass3043 = False
else: ii += 1
if (not (ii < len(items))): break
it = items[ii]
if (NounPhraseItem.try_accord_adj_and_noun(
it,
Utils.asObjectOrNull(npt1.noun,
NounPhraseItem))):
continue
if (ii > 0):
inp2 = NounPhraseHelper.try_parse(
it.begin_token, typ, max_char_pos, None)
if (inp2 is not None
and inp2.end_token == inp.end_token):
del items[ii:ii + len(items) - ii]
inp = inp2
break
ok = False
break
if (ok):
if (npt1.morph.case_.is_genitive
and not inp.morph.case_.is_instrumental):
ok = False
if (ok):
i = 0
while i < len(items):
npt1.adjectives.insert(i, items[i])
i += 1
npt1.internal_noun = inp
mmm = MorphCollection(npt1.morph)
for it in items:
mmm.remove_items(it.adj_morph[0], False)
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
if (tt1 is not None and tt1.morph.class0_.is_noun
and not tt1.morph.case_.is_genitive):
it = NounPhraseItem.try_parse(tt1, items, typ)
if (it is not None and it.can_be_noun):
internal_noun_prase = inp
inp.begin_token = items[0].end_token.next0_
items.append(it)
i = 0
first_pass3044 = True
while True:
if first_pass3044: first_pass3044 = False
else: i += 1
if (not (i < len(items))): break
if (items[i].can_be_adj
and items[i].begin_token.morph.class0_.is_verb):
it = items[i].begin_token
if (not it.get_morph_class_in_dictionary().is_verb):
continue
if (it.is_value("УПОЛНОМОЧЕННЫЙ", None)):
continue
if ((((typ) & (NounPhraseParseAttr.PARSEVERBS))) == (
NounPhraseParseAttr.NO)):
continue
inp = _NounPraseHelperInt.__try_parse_ru(
items[i].end_token.next0_, NounPhraseParseAttr.NO,
max_char_pos, None)
if (inp is None):
continue
if (inp.anafor is not None and i == (len(items) - 1)
and NounPhraseItem.try_accord_adj_and_noun(
items[i],
Utils.asObjectOrNull(inp.noun, NounPhraseItem))):
inp.begin_token = first
ii = 0
while ii < len(items):
inp.adjectives.insert(ii, items[ii])
ii += 1
return inp
if (inp.end_token.whitespaces_after_count > 3):
continue
npt1 = _NounPraseHelperInt.__try_parse_ru(
inp.end_token.next0_, NounPhraseParseAttr.NO, max_char_pos,
None)
if (npt1 is None):
continue
ok = True
j = 0
while j <= i:
if (not NounPhraseItem.try_accord_adj_and_noun(
items[j],
Utils.asObjectOrNull(npt1.noun, NounPhraseItem))):
ok = False
break
j += 1
if (not ok):
continue
verb = VerbPhraseHelper.try_parse(it, True, False, False)
if (verb is None):
continue
vlinks = SemanticHelper.try_create_links(verb, inp, None)
nlinks = SemanticHelper.try_create_links(inp, npt1, None)
if (len(vlinks) == 0 and len(nlinks) > 0):
continue
j = 0
while j <= i:
npt1.adjectives.insert(j, items[j])
j += 1
items[i].end_token = inp.end_token
mmm = MorphCollection(npt1.morph)
bil = list()
j = 0
while j <= i:
bil.clear()
for m in items[j].adj_morph:
bil.append(m)
mmm.remove_items_list_cla(bil, None)
j += 1
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
ok2 = False
if ((len(items) == 1 and
(((typ) & (NounPhraseParseAttr.ADJECTIVECANBELAST))) !=
(NounPhraseParseAttr.NO) and
(items[0].whitespaces_after_count < 3))
and not items[0].is_adverb):
if (not items[0].can_be_adj):
ok2 = True
elif (items[0].is_personal_pronoun and items[0].can_be_noun):
ok2 = True
if (ok2):
it = NounPhraseItem.try_parse(items[0].end_token.next0_, None, typ)
if (it is not None and it.can_be_adj
and it.begin_token.chars.is_all_lower):
ok2 = True
if (it.is_adverb or it.is_verb):
ok2 = False
if (it.is_pronoun and items[0].is_pronoun):
ok2 = False
if (it.can_be_adj_for_personal_pronoun
and items[0].is_personal_pronoun):
ok2 = True
if (ok2 and NounPhraseItem.try_accord_adj_and_noun(
it, items[0])):
npt1 = _NounPraseHelperInt.__try_parse_ru(
it.begin_token, typ, max_char_pos, None)
if (npt1 is not None and ((npt1.end_char > it.end_char
or len(npt1.adjectives) > 0))):
pass
else:
items.insert(0, it)
noun = None
adj_after = None
for i in range(len(items) - 1, -1, -1):
if (items[i].can_be_noun):
if (items[i].conj_before):
continue
if (i > 0 and not items[i - 1].can_be_adj):
continue
if (i > 0 and items[i - 1].can_be_noun):
if (items[i - 1].is_doubt_adjective):
continue
if (items[i - 1].is_pronoun and items[i].is_pronoun):
if (items[i].is_pronoun and
items[i - 1].can_be_adj_for_personal_pronoun):
pass
else:
continue
noun = items[i]
del items[i:i + len(items) - i]
if (adj_after is not None):
items.append(adj_after)
elif (len(items) > 0 and items[0].can_be_noun
and not items[0].can_be_adj):
noun = items[0]
items.clear()
break
if (noun is None):
return None
res = NounPhraseToken._new466(first, noun.end_token, prep)
if (adverbs is not None):
for a in adverbs:
if (a.begin_char < noun.begin_char):
if (len(items) == 0 and prep is None):
return None
if (res.adverbs is None):
res.adverbs = list()
res.adverbs.append(a)
res.noun = (noun)
res.multi_nouns = noun.multi_nouns
if (kak):
res.multi_nouns = True
res.internal_noun = internal_noun_prase
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (res.morph.case_.is_nominative and first.previous is not None
and first.previous.morph.class0_.is_preposition):
res.morph.case_ = (res.morph.case_) ^ MorphCase.NOMINATIVE
if ((((typ) &
(NounPhraseParseAttr.PARSEPRONOUNS))) == (NounPhraseParseAttr.NO)
and ((res.morph.class0_.is_pronoun
or res.morph.class0_.is_personal_pronoun))):
return None
stat = None
if (len(items) > 1):
stat = dict()
need_update_morph = False
if (len(items) > 0):
ok_list = list()
is_num_not = False
for vv in noun.noun_morph:
i = 0
v = vv
i = 0
while i < len(items):
ok = False
for av in items[i].adj_morph:
if (v.check_accord(av, False, False)):
ok = True
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
break
if (not ok):
if (items[i].can_be_numeric_adj
and items[i].try_accord_var(v, False)):
ok = True
v1 = NounPhraseItemTextVar()
v1.copy_from_item(v)
v1.number = MorphNumber.PLURAL
is_num_not = True
v1.case_ = MorphCase()
for a in items[i].adj_morph:
v1.case_ = (v1.case_) | a.case_
v = v1
else:
break
i += 1
if (i >= len(items)):
ok_list.append(v)
if (len(ok_list) > 0 and
(((len(ok_list) < res.morph.items_count) or is_num_not))):
res.morph = MorphCollection()
for v in ok_list:
res.morph.add_item(v)
if (not is_num_not):
noun.morph = res.morph
i = 0
first_pass3045 = True
while True:
if first_pass3045: first_pass3045 = False
else: i += 1
if (not (i < len(items))): break
for av in items[i].adj_morph:
for v in noun.noun_morph:
if (v.check_accord(av, False, False)):
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
need_update_morph = True
items[i].morph.add_item(av)
if (stat is not None and av.normal_value is not None
and len(av.normal_value) > 1):
last = av.normal_value[len(av.normal_value) - 1]
if (not last in stat):
stat[last] = 1
else:
stat[last] += 1
if (items[i].is_pronoun or items[i].is_personal_pronoun):
res.anafor = items[i].begin_token
if ((((typ) & (NounPhraseParseAttr.PARSEPRONOUNS))) == (
NounPhraseParseAttr.NO)):
continue
tt = Utils.asObjectOrNull(items[i].begin_token, TextToken)
if (tt is not None and not tt.term.startswith("ВЫСШ")):
err = False
for wf in tt.morph.items:
if (wf.class0_.is_adjective):
if (wf.contains_attr("прев.", None)):
if ((((typ) &
(NounPhraseParseAttr.IGNOREADJBEST))) !=
(NounPhraseParseAttr.NO)):
err = True
if (wf.contains_attr("к.ф.", None)
and tt.morph.class0_.is_personal_pronoun):
return None
if (err):
continue
if (res.morph.case_.is_nominative):
v = MiscHelper.get_text_value_of_meta_token(
items[i], GetTextAttr.KEEPQUOTES)
if (not Utils.isNullOrEmpty(v)):
if (items[i].get_normal_case_text(
None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED,
False) != v):
wf = NounPhraseItemTextVar(items[i].morph, None)
wf.normal_value = v
wf.class0_ = MorphClass.ADJECTIVE
wf.case_ = res.morph.case_
if (res.morph.case_.is_prepositional
or res.morph.gender == MorphGender.NEUTER
or res.morph.gender == MorphGender.FEMINIE):
items[i].morph.add_item(wf)
else:
items[i].morph.insert_item(0, wf)
res.adjectives.append(items[i])
if (items[i].end_char > res.end_char):
res.end_token = items[i].end_token
i = 0
first_pass3046 = True
while True:
if first_pass3046: first_pass3046 = False
else: i += 1
if (not (i < (len(res.adjectives) - 1))): break
if (res.adjectives[i].whitespaces_after_count > 5):
if (res.adjectives[i].chars != res.adjectives[i + 1].chars):
if (not res.adjectives[i + 1].chars.is_all_lower):
return None
if (res.adjectives[i].chars.is_all_upper
and res.adjectives[i + 1].chars.is_capital_upper):
return None
if (res.adjectives[i].chars.is_capital_upper
and res.adjectives[i + 1].chars.is_all_upper):
return None
if (res.adjectives[i].whitespaces_after_count > 10):
if (res.adjectives[i].newlines_after_count == 1):
if (res.adjectives[i].chars.is_capital_upper and i == 0
and res.adjectives[i + 1].chars.is_all_lower):
continue
if (res.adjectives[i].chars == res.adjectives[
i + 1].chars):
continue
return None
if (need_update_morph):
noun.morph = MorphCollection()
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (len(res.adjectives) > 0):
if (noun.begin_token.previous is not None):
if (noun.begin_token.previous.is_comma_and):
if (res.adjectives[0].begin_char > noun.begin_char):
pass
else:
return None
zap = 0
and0_ = 0
cou = 0
last_and = False
i = 0
while i < (len(res.adjectives) - 1):
te = res.adjectives[i].end_token.next0_
if (te is None):
return None
if (te.is_char('(')):
pass
elif (te.is_comma):
zap += 1
last_and = False
elif (te.is_and or te.is_or):
and0_ += 1
last_and = True
if (not res.adjectives[i].begin_token.morph.class0_.is_pronoun
):
cou += 1
i += 1
if ((zap + and0_) > 0):
if (and0_ > 1):
return None
elif (and0_ == 1 and not last_and):
return None
if ((zap + and0_) != cou):
if (and0_ == 1):
pass
else:
return None
last = Utils.asObjectOrNull(
res.adjectives[len(res.adjectives) - 1], NounPhraseItem)
if (last.is_pronoun and not last_and):
return None
if (stat is not None):
for adj in items:
if (adj.morph.items_count > 1):
w1 = Utils.asObjectOrNull(adj.morph.get_indexer_item(0),
NounPhraseItemTextVar)
w2 = Utils.asObjectOrNull(adj.morph.get_indexer_item(1),
NounPhraseItemTextVar)
if ((len(w1.normal_value) < 2)
or (len(w2.normal_value) < 2)):
break
l1 = w1.normal_value[len(w1.normal_value) - 1]
l2 = w2.normal_value[len(w2.normal_value) - 1]
i1 = 0
i2 = 0
wrapi1468 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l1, wrapi1468)
i1 = wrapi1468.value
wrapi2467 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l2, wrapi2467)
i2 = wrapi2467.value
if (i1 < i2):
adj.morph.remove_item(1)
adj.morph.insert_item(0, w2)
if (res.begin_token.get_morph_class_in_dictionary().is_verb
and len(items) > 0):
if (not res.begin_token.chars.is_all_lower
or res.begin_token.previous is None):
pass
elif (res.begin_token.previous.morph.class0_.is_preposition):
pass
else:
comma = False
tt = res.begin_token.previous
first_pass3047 = True
while True:
if first_pass3047: first_pass3047 = False
else: tt = tt.previous
if (not (tt is not None and tt.end_char <= res.end_char)):
break
if (tt.morph.class0_.is_adverb):
continue
if (tt.is_char_of(".;")):
break
if (tt.is_comma):
comma = True
continue
if (tt.is_value("НЕ", None)):
continue
if (((tt.morph.class0_.is_noun
or tt.morph.class0_.is_proper)) and comma):
for it in res.begin_token.morph.items:
if (it.class0_.is_verb
and (isinstance(it, MorphWordForm))):
if (tt.morph.check_accord(it, False, False)):
if (res.morph.case_.is_instrumental):
return None
break
if (res.begin_token == res.end_token):
mc = res.begin_token.get_morph_class_in_dictionary()
if (mc.is_adverb):
if (res.begin_token.previous is not None and
res.begin_token.previous.morph.class0_.is_preposition):
pass
elif (mc.is_noun and not mc.is_preposition
and not mc.is_conjunction):
pass
elif (res.begin_token.is_value("ВЕСЬ", None)):
pass
else:
return None
if (def_noun is not None and def_noun.end_token == res.end_token
and len(res.adjectives) > 0):
res.end_token = res.adjectives[len(res.adjectives) - 1].end_token
return res
def tryParse(t: 'Token',
add_units: 'TerminCollection',
can_be_set: bool = True,
can_units_absent: bool = False) -> 'MeasureToken':
""" Выделение вместе с наименованием
Args:
t(Token):
"""
if (not ((isinstance(t, TextToken)))):
return None
if (t.is_table_control_char):
return None
t0 = t
whd = None
minmax = 0
wrapminmax1516 = RefOutArgWrapper(minmax)
tt = NumbersWithUnitToken._isMinOrMax(t0, wrapminmax1516)
minmax = wrapminmax1516.value
if (tt is not None):
t = tt.next0_
npt = NounPhraseHelper.tryParse(
t,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt is None):
whd = NumbersWithUnitToken._tryParseWHL(t)
if (whd is not None):
npt = NounPhraseToken(t0, whd.end_token)
elif (t0.isValue("КПД", None)):
npt = NounPhraseToken(t0, t0)
elif ((isinstance(t0, TextToken)) and t0.length_char > 3
and t0.getMorphClassInDictionary().is_undefined):
npt = NounPhraseToken(t0, t0)
else:
return None
elif (NumberHelper.tryParseRealNumber(t, True) is not None):
return None
else:
dtok = DateItemToken.tryAttach(t, None)
if (dtok is not None):
return None
t1 = npt.end_token
t = npt.end_token
name_ = MetaToken._new561(npt.begin_token, npt.end_token, npt.morph)
units = None
units2 = None
internals_ = list()
not0_ = False
tt = t1.next0_
first_pass3037 = True
while True:
if first_pass3037: first_pass3037 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
break
if (tt.is_table_control_char):
break
wrapminmax1510 = RefOutArgWrapper(minmax)
tt2 = NumbersWithUnitToken._isMinOrMax(tt, wrapminmax1510)
minmax = wrapminmax1510.value
if (tt2 is not None):
tt = tt2
t = tt
t1 = t
continue
if ((tt.isValue("БЫТЬ", None) or tt.isValue("ДОЛЖЕН", None)
or tt.isValue("ДОЛЖНЫЙ", None)) or tt.isValue("МОЖЕТ", None)
or
((tt.isValue("СОСТАВЛЯТЬ", None)
and not tt.getMorphClassInDictionary().is_adjective))):
t = tt
t1 = t
if (tt.previous.isValue("НЕ", None)):
not0_ = True
continue
www = NumbersWithUnitToken._tryParseWHL(tt)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
continue
if (len(internals_) > 0 and tt.is_comma_and):
continue
if (tt.isValue("ПРИ", None) or len(internals_) > 0):
mt1 = MeasureToken.tryParse(tt.next0_, add_units, False, False)
if (mt1 is not None and mt1.reliable):
internals_.append(mt1)
tt = mt1.end_token
t = tt
t1 = t
continue
if ((isinstance(tt, NumberToken))
and (tt).typ == NumberSpellingType.WORDS):
npt3 = NounPhraseHelper.tryParse(
tt, NounPhraseParseAttr.PARSENUMERICASADJECTIVE, 0)
if (npt3 is not None):
tt = npt3.end_token
t1 = tt
if (len(internals_) == 0):
name_.end_token = t1
continue
mt0 = NumbersWithUnitToken.tryParse(tt, add_units, False, False)
if (mt0 is not None):
break
if (((tt.is_comma or tt.isChar('('))) and tt.next0_ is not None):
www = NumbersWithUnitToken._tryParseWHL(tt.next0_)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
if (tt.next0_ is not None and tt.next0_.is_comma):
tt = tt.next0_
t1 = tt
if (tt.next0_ is not None and tt.next0_.isChar(')')):
tt = tt.next0_
t1 = tt
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
t = uu[len(uu) - 1].end_token
t1 = t
units = uu
if (tt.isChar('(') and t1.next0_ is not None
and t1.next0_.isChar(')')):
tt = t1.next0_
t = tt
t1 = t
continue
elif (t1.next0_ is not None and t1.next0_.isChar('(')):
uu = UnitToken.tryParseList(t1.next0_.next0_,
add_units, False)
if (uu is not None and uu[len(uu) - 1].end_token.next0_
is not None and
uu[len(uu) - 1].end_token.next0_.isChar(')')):
units2 = uu
tt = uu[len(uu) - 1].end_token.next0_
t = tt
t1 = t
continue
if (uu is not None and len(uu) > 0 and not uu[0].is_doubt):
break
if (BracketHelper.canBeStartOfSequence(tt, False, False)):
br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
t = tt
t1 = t
continue
if (tt.isValue("НЕ", None) and tt.next0_ is not None):
mc = tt.next0_.getMorphClassInDictionary()
if (mc.is_adverb or mc.is_misc):
break
continue
if (tt.isValue("ЯМЗ", None)):
pass
npt2 = NounPhraseHelper.tryParse(
tt,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt2 is None):
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
to = NumbersWithUnitToken.M_TERMINS.tryParse(
tt, TerminParseAttr.NO)
if (to is not None):
if ((isinstance(to.end_token.next0_, TextToken))
and to.end_token.next0_.is_letters):
pass
else:
break
t1 = tt
continue
mc = tt.getMorphClassInDictionary()
if (((isinstance(tt, TextToken)) and tt.chars.is_letter
and tt.length_char > 1)
and (((tt.chars.is_all_upper or mc.is_adverb
or mc.is_undefined) or mc.is_adjective))):
uu = UnitToken.tryParseList(tt, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = tt
continue
if (tt.is_comma):
continue
if (tt.isChar('.')):
if (not MiscHelper.canBeStartOfSentence(tt.next0_)):
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
break
tt = npt2.end_token
t = tt
t1 = t
if (len(internals_) > 0):
pass
elif (t.isValue("ПРЕДЕЛ", None) or t.isValue("ГРАНИЦА", None)
or t.isValue("ДИАПАЗОН", None)):
pass
elif (t.chars.is_letter):
name_.end_token = t1
t1 = t1.next0_
first_pass3038 = True
while True:
if first_pass3038: first_pass3038 = False
else: t1 = t1.next0_
if (not (t1 is not None)): break
if (t1.is_table_control_char):
pass
elif (t1.isCharOf(":,_")):
www = NumbersWithUnitToken._tryParseWHL(t1.next0_)
if (www is not None):
whd = www
t = www.end_token
t1 = t
continue
elif (t1.is_hiphen and t1.is_whitespace_after
and t1.is_whitespace_before):
pass
else:
break
if (t1 is None):
return None
mts = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, not0_)
if (mts is None):
return None
mt = mts[0]
if (name_.begin_token.morph.class0_.is_preposition):
name_.begin_token = name_.begin_token.next0_
if (len(mts) > 1 and len(internals_) == 0):
if (len(mt.units) == 0):
if (units is not None):
for m in mts:
m.units = units
res1 = MeasureToken._new1511(t0, mts[len(mts) - 1].end_token,
name_.morph, True)
res1.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
k = 0
while k < len(mts):
ttt = MeasureToken._new1506(mts[k].begin_token,
mts[k].end_token, mts[k])
if (whd is not None):
nams = Utils.asObjectOrNull(whd.tag, list)
if (k < len(nams)):
ttt.name = nams[k]
res1.internals.append(ttt)
k += 1
tt1 = res1.end_token.next0_
if (tt1 is not None and tt1.isChar('±')):
nn = NumbersWithUnitToken._tryParse(tt1, add_units, True,
False)
if (nn is not None and nn.plus_minus_percent):
res1.end_token = nn.end_token
res1.nums = nn
return res1
if (not mt.is_whitespace_before):
if (mt.begin_token.previous is None):
return None
if (mt.begin_token.previous.isCharOf(":),")
or mt.begin_token.previous.is_table_control_char):
pass
else:
return None
if (len(mt.units) == 0 and units is not None):
mt.units = units
if (mt.div_num is not None and len(units) > 1
and len(mt.div_num.units) == 0):
i = 1
while i < len(units):
if (units[i].pow0_ == -1):
j = i
while j < len(units):
mt.div_num.units.append(units[j])
units[j].pow0_ = (-units[j].pow0_)
j += 1
del mt.units[i:i + len(units) - i]
break
i += 1
if ((minmax < 0) and mt.single_val is not None):
mt.from_val = mt.single_val
mt.from_include = True
mt.single_val = (None)
if (minmax > 0 and mt.single_val is not None):
mt.to_val = mt.single_val
mt.to_include = True
mt.single_val = (None)
if (len(mt.units) == 0):
units = UnitToken.tryParseList(mt.end_token.next0_, add_units,
True)
if (units is None):
if (can_units_absent):
pass
else:
return None
else:
mt.units = units
res = MeasureToken._new1513(t0, mt.end_token, name_.morph, internals_)
if (((not t0.is_whitespace_before and t0.previous is not None
and t0 == name_.begin_token) and t0.previous.is_hiphen
and not t0.previous.is_whitespace_before)
and (isinstance(t0.previous.previous, TextToken))):
name_.begin_token = res.begin_token = name_.begin_token.previous.previous
res.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
res.nums = mt
for u in res.nums.units:
if (u.keyword is not None):
if (u.keyword.begin_char >= res.begin_char):
res.reliable = True
res.__parseInternals(add_units)
if (len(res.internals) > 0 or not can_be_set):
return res
t1 = res.end_token.next0_
if (t1 is not None and t1.is_comma_and):
t1 = t1.next0_
mts1 = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, False)
if ((mts1 is not None and len(mts1) == 1 and
(t1.whitespaces_before_count < 3)) and len(mts1[0].units) > 0
and not UnitToken.canBeEquals(mts[0].units, mts1[0].units)):
res.is_set = True
res.nums = (None)
res.internals.append(
MeasureToken._new1506(mt.begin_token, mt.end_token, mt))
res.internals.append(
MeasureToken._new1506(mts1[0].begin_token, mts1[0].end_token,
mts1[0]))
res.end_token = mts1[0].end_token
return res
def __calc_rank_and_value(self, min_newlines_count: int) -> bool:
self.rank = 0
if (self.begin_token.chars.is_all_lower):
self.rank -= 30
words = 0
up_words = 0
notwords = 0
line_number = 0
tstart = self.begin_token
tend = self.end_token
t = self.begin_token
first_pass3396 = True
while True:
if first_pass3396: first_pass3396 = False
else: t = t.next0_
if (not (t != self.end_token.next0_ and t is not None
and t.end_char <= self.end_token.end_char)):
break
if (t.is_newline_before):
pass
tit = TitleItemToken.try_attach(t)
if (tit is not None):
if (tit.typ == TitleItemToken.Types.THEME
or tit.typ == TitleItemToken.Types.TYPANDTHEME):
if (t != self.begin_token):
if (line_number > 0):
return False
notwords = 0
up_words = notwords
words = up_words
tstart = tit.end_token.next0_
t = tit.end_token
if (t.next0_ is None):
return False
if (t.next0_.chars.is_letter
and t.next0_.chars.is_all_lower):
self.rank += 20
else:
self.rank += 100
tstart = t.next0_
if (tit.typ == TitleItemToken.Types.TYPANDTHEME):
self.type_value = tit.value
continue
if (tit.typ == TitleItemToken.Types.TYP):
if (t == self.begin_token):
if (tit.end_token.is_newline_after):
self.type_value = tit.value
self.rank += 5
tstart = tit.end_token.next0_
t = tit.end_token
words += 1
if (tit.begin_token != tit.end_token):
words += 1
if (tit.chars.is_all_upper):
up_words += 1
continue
if (tit.typ == TitleItemToken.Types.DUST
or tit.typ == TitleItemToken.Types.SPECIALITY):
if (t == self.begin_token):
return False
self.rank -= 20
if (tit.typ == TitleItemToken.Types.SPECIALITY):
self.speciality = tit.value
t = tit.end_token
continue
if (tit.typ == TitleItemToken.Types.CONSULTANT
or tit.typ == TitleItemToken.Types.BOSS
or tit.typ == TitleItemToken.Types.EDITOR):
t = tit.end_token
if (t.next0_ is not None and
((t.next0_.is_char_of(":") or t.next0_.is_hiphen
or t.whitespaces_after_count > 4))):
self.rank -= 10
else:
self.rank -= 2
continue
return False
blt = BookLinkToken.try_parse(t, 0)
if (blt is not None):
if (blt.typ == BookLinkTyp.MISC or blt.typ == BookLinkTyp.N
or blt.typ == BookLinkTyp.PAGES):
self.rank -= 10
elif (blt.typ == BookLinkTyp.N
or blt.typ == BookLinkTyp.PAGERANGE):
self.rank -= 20
if (t == self.begin_token and BookLinkToken.try_parse_author(
t, FioTemplateType.UNDEFINED) is not None):
self.rank -= 20
if (t.is_newline_before and t != self.begin_token):
line_number += 1
if (line_number > 4):
return False
if (t.chars.is_all_lower):
self.rank += 10
elif (t.previous.is_char('.')):
self.rank -= 10
elif (t.previous.is_char_of(",-")):
self.rank += 10
else:
npt = NounPhraseHelper.try_parse(t.previous,
NounPhraseParseAttr.NO, 0,
None)
if (npt is not None and npt.end_char >= t.end_char):
self.rank += 10
if (t != self.begin_token
and t.newlines_before_count > min_newlines_count):
self.rank -= (t.newlines_before_count - min_newlines_count)
bst = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (bst is not None and bst.is_quote_type
and bst.end_token.end_char <= self.end_token.end_char):
if (words == 0):
tstart = bst.begin_token
self.rank += 10
if (bst.end_token == self.end_token):
tend = self.end_token
self.rank += 10
rli = t.get_referents()
if (rli is not None):
for r in rli:
if (isinstance(r, OrganizationReferent)):
if (t.is_newline_before):
self.rank -= 10
else:
self.rank -= 4
continue
if ((isinstance(r, GeoReferent))
or (isinstance(r, PersonReferent))):
if (t.is_newline_before):
self.rank -= 5
if (t.is_newline_after or t.next0_ is None):
self.rank -= 20
elif (t.next0_.is_hiphen
or (isinstance(t.next0_, NumberToken))
or (isinstance(t.next0_.get_referent(),
DateReferent))):
self.rank -= 20
elif (t != self.begin_token):
self.rank -= 20
continue
if ((isinstance(r, GeoReferent))
or (isinstance(r, DenominationReferent))):
continue
if ((isinstance(r, UriReferent))
or (isinstance(r, PhoneReferent))):
return False
if (t.is_newline_before):
self.rank -= 4
else:
self.rank -= 2
if (t == self.begin_token and (isinstance(
self.end_token.get_referent(), PersonReferent))):
self.rank -= 10
words += 1
if (t.chars.is_all_upper):
up_words += 1
if (t == self.begin_token):
if (t.is_newline_after):
self.rank -= 10
elif (t.next0_ is not None and t.next0_.is_char('.')
and t.next0_.is_newline_after):
self.rank -= 10
continue
if (isinstance(t, NumberToken)):
if (t.typ == NumberSpellingType.WORDS):
words += 1
if (t.chars.is_all_upper):
up_words += 1
else:
notwords += 1
continue
pat = PersonAttrToken.try_attach(
t, None, PersonAttrToken.PersonAttrAttachAttrs.NO)
if (pat is not None):
if (t.is_newline_before):
if (not pat.morph.case_.is_undefined
and not pat.morph.case_.is_nominative):
pass
elif (pat.chars.is_all_upper):
pass
else:
self.rank -= 20
elif (t.chars.is_all_lower):
self.rank -= 1
while t is not None:
words += 1
if (t.chars.is_all_upper):
up_words += 1
if (t == pat.end_token):
break
t = t.next0_
continue
oitt = OrgItemTypeToken.try_attach(t, True, None)
if (oitt is not None):
if (oitt.morph.number != MorphNumber.PLURAL
and not oitt.is_doubt_root_word):
if (not oitt.morph.case_.is_undefined
and not oitt.morph.case_.is_nominative):
words += 1
if (t.chars.is_all_upper):
up_words += 1
else:
self.rank -= 4
if (t == self.begin_token):
self.rank -= 5
else:
words += 1
if (t.chars.is_all_upper):
up_words += 1
t = oitt.end_token
continue
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is not None):
if (tt.is_char('©')):
self.rank -= 10
if (tt.is_char('_')):
self.rank -= 1
if (tt.chars.is_letter):
if (tt.length_char > 2):
words += 1
if (t.chars.is_all_upper):
up_words += 1
elif (not tt.is_char(',')):
notwords += 1
if (tt.is_pure_verb):
self.rank -= 30
words -= 1
break
if (tt == self.end_token):
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
self.rank -= 10
elif (tt.is_char('.')):
self.rank += 5
elif (tt.is_char_of("._")):
self.rank -= 5
self.rank += words
self.rank -= notwords
if ((words < 1) and (self.rank < 50)):
return False
if (tstart is None or tend is None):
return False
if (tstart.end_char > tend.end_char):
return False
tit1 = TitleItemToken.try_attach(self.end_token.next0_)
if (tit1 is not None
and ((tit1.typ == TitleItemToken.Types.TYP
or tit1.typ == TitleItemToken.Types.SPECIALITY))):
if (tit1.end_token.is_newline_after):
self.rank += 15
else:
self.rank += 10
if (tit1.typ == TitleItemToken.Types.SPECIALITY):
self.speciality = tit1.value
if (up_words > 4 and up_words > (math.floor((0.8 * (words))))):
if (tstart.previous is not None and
(isinstance(tstart.previous.get_referent(), PersonReferent))):
self.rank += (5 + up_words)
self.begin_name_token = tstart
self.end_name_token = tend
return True
def try_attach_requisites(t : 'Token', cur : 'InstrumentParticipantReferent', other : 'InstrumentParticipantReferent', cant_be_empty : bool=False) -> 'ReferentToken':
if (t is None or cur is None):
return None
if (t.is_table_control_char):
return None
err = 0
spec_chars = 0
rt = None
t0 = t
is_in_tab_cell = False
cou = 0
tt = t.next0_
while tt is not None and (cou < 300):
if (tt.is_table_control_char):
is_in_tab_cell = True
break
tt = tt.next0_; cou += 1
first_pass3286 = True
while True:
if first_pass3286: first_pass3286 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.begin_char == 8923):
pass
if (t.is_table_control_char):
if (t != t0):
if (rt is not None):
rt.end_token = t.previous
elif (not cant_be_empty):
rt = ReferentToken(cur, t0, t.previous)
break
else:
continue
if ((t.is_char_of(":.") or t.is_value("М", None) or t.is_value("M", None)) or t.is_value("П", None)):
if (rt is not None):
rt.end_token = t
continue
pp = ParticipantToken.try_attach_to_exist(t, cur, other)
if (pp is not None):
if (pp.referent != cur):
break
if (rt is None):
rt = ReferentToken(cur, t, t)
rt.end_token = pp.end_token
err = 0
continue
if (t.is_newline_before):
iii = InstrToken.parse(t, 0, None)
if (iii is not None):
if (iii.typ == ILTypes.APPENDIX):
break
if (t.whitespaces_before_count > 25 and not is_in_tab_cell):
if (t != t0):
if (t.previous is not None and t.previous.is_char_of(",;")):
pass
elif (t.newlines_before_count > 1):
break
if ((isinstance(t.get_referent(), PersonReferent)) or (isinstance(t.get_referent(), OrganizationReferent))):
if (not cur._contains_ref(t.get_referent())):
break
if ((t.is_char_of(";:,.") or t.is_hiphen or t.morph.class0_.is_preposition) or t.morph.class0_.is_conjunction):
continue
if (t.is_char_of("_/\\")):
spec_chars += 1
if (spec_chars > 10 and rt is None):
rt = ReferentToken(cur, t0, t)
if (rt is not None):
rt.end_token = t
continue
if (t.is_newline_before and (isinstance(t, NumberToken))):
break
if (t.is_value("ОФИС", None)):
if (BracketHelper.can_be_start_of_sequence(t.next0_, True, False)):
br = BracketHelper.try_parse(t.next0_, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
continue
if ((isinstance(t.next0_, TextToken)) and not t.next0_.chars.is_all_lower):
t = t.next0_
continue
r = t.get_referent()
if ((((isinstance(r, PersonReferent)) or (isinstance(r, AddressReferent)) or (isinstance(r, UriReferent))) or (isinstance(r, OrganizationReferent)) or (isinstance(r, PhoneReferent))) or (isinstance(r, PersonIdentityReferent)) or (isinstance(r, BankDataReferent))):
if (other is not None and other.find_slot(None, r, True) is not None):
if (not (isinstance(r, UriReferent))):
break
if (rt is None):
rt = ReferentToken(cur, t, t)
if (cur.find_slot(InstrumentParticipantReferent.ATTR_DELEGATE, r, True) is not None):
pass
else:
cur.add_slot(InstrumentParticipantReferent.ATTR_REF, r, False, 0)
rt.end_token = t
err = 0
else:
if ((isinstance(t, TextToken)) and t.length_char > 1):
err += 1
if (is_in_tab_cell and rt is not None):
if (err > 300):
break
elif (err > 4):
break
return rt
def _process(begin : 'Token', max_char_pos : int, kit : 'AnalysisKit', end_token : 'Token') -> 'TitlePageReferent':
end_token.value = begin
res = TitlePageReferent()
term = None
lines = Line.parse(begin, 30, 1500, max_char_pos)
if (len(lines) < 1):
return None
cou = len(lines)
min_newlines_count = 10
lines_count_stat = dict()
i = 0
while i < len(lines):
if (TitleNameToken.can_be_start_of_text_or_content(lines[i].begin_token, lines[i].end_token)):
cou = i
break
j = lines[i].newlines_before_count
if (i > 0 and j > 0):
if (not j in lines_count_stat):
lines_count_stat[j] = 1
else:
lines_count_stat[j] += 1
i += 1
max0_ = 0
for kp in lines_count_stat.items():
if (kp[1] > max0_):
max0_ = kp[1]
min_newlines_count = kp[0]
end_char = (lines[cou - 1].end_char if cou > 0 else 0)
if (max_char_pos > 0 and end_char > max_char_pos):
end_char = max_char_pos
names = list()
i = 0
while i < cou:
if (i == 6):
pass
j = i
while (j < cou) and (j < (i + 5)):
if (i == 6 and j == 8):
pass
if (j > i):
if (lines[j - 1].is_pure_en and lines[j].is_pure_ru):
break
if (lines[j - 1].is_pure_ru and lines[j].is_pure_en):
break
if (lines[j].newlines_before_count >= (min_newlines_count * 2)):
break
ttt = TitleNameToken.try_parse(lines[i].begin_token, lines[j].end_token, min_newlines_count)
if (ttt is not None):
if (lines[i].is_pure_en):
ttt.morph.language = MorphLang.EN
elif (lines[i].is_pure_ru):
ttt.morph.language = MorphLang.RU
names.append(ttt)
j += 1
i += 1
TitleNameToken.sort(names)
name_rt = None
if (len(names) > 0):
i0 = 0
if (names[i0].morph.language.is_en):
ii = 1
while ii < len(names):
if (names[ii].morph.language.is_ru and names[ii].rank > 0):
i0 = ii
break
ii += 1
term = res._add_name(names[i0].begin_name_token, names[i0].end_name_token)
if (names[i0].type_value is not None):
res._add_type(names[i0].type_value)
if (names[i0].speciality is not None):
res.speciality = names[i0].speciality
rt = ReferentToken(res, names[i0].begin_token, names[i0].end_token)
if (kit is not None):
kit.embed_token(rt)
else:
res.add_occurence(TextAnnotation(rt.begin_token, rt.end_token))
end_token.value = rt.end_token
name_rt = rt
if (begin.begin_char == rt.begin_char):
begin = (rt)
if (term is not None and kit is not None):
t = kit.first_token
first_pass3397 = True
while True:
if first_pass3397: first_pass3397 = False
else: t = t.next0_
if (not (t is not None)): break
tok = term.try_parse(t, TerminParseAttr.NO)
if (tok is None):
continue
t0 = t
t1 = tok.end_token
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
if (BracketHelper.can_be_start_of_sequence(t0.previous, False, False) and BracketHelper.can_be_end_of_sequence(t1.next0_, False, None, False)):
t0 = t0.previous
t1 = t1.next0_
rt = ReferentToken(res, t0, t1)
kit.embed_token(rt)
t = (rt)
pr = PersonRelations()
pers_typ = TitleItemToken.Types.UNDEFINED
pers_types = pr.rel_types
t = begin
first_pass3398 = True
while True:
if first_pass3398: first_pass3398 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char_pos > 0 and t.begin_char > max_char_pos):
break
if (t == name_rt):
continue
tpt = TitleItemToken.try_attach(t)
if (tpt is not None):
pers_typ = TitleItemToken.Types.UNDEFINED
if (tpt.typ == TitleItemToken.Types.TYP):
if (len(res.types) == 0):
res._add_type(tpt.value)
elif (len(res.types) == 1):
ty = res.types[0].upper()
if (ty == "РЕФЕРАТ"):
res._add_type(tpt.value)
elif (ty == "АВТОРЕФЕРАТ"):
if (tpt.value == "КАНДИДАТСКАЯ ДИССЕРТАЦИЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат кандидатской диссертации", True, 0)
elif (tpt.value == "ДОКТОРСКАЯ ДИССЕРТАЦИЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат докторской диссертации", True, 0)
elif (tpt.value == "МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат магистерской диссертации", True, 0)
elif (tpt.value == "КАНДИДАТСЬКА ДИСЕРТАЦІЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат кандидатської дисертації", True, 0)
elif (tpt.value == "ДОКТОРСЬКА ДИСЕРТАЦІЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат докторської дисертації", True, 0)
elif (tpt.value == "МАГІСТЕРСЬКА ДИСЕРТАЦІЯ"):
res.add_slot(TitlePageReferent.ATTR_TYPE, "автореферат магістерської дисертації", True, 0)
else:
res._add_type(tpt.value)
elif (tpt.value == "РЕФЕРАТ" or tpt.value == "АВТОРЕФЕРАТ"):
if (not tpt.value in ty):
res._add_type(tpt.value)
elif (tpt.typ == TitleItemToken.Types.SPECIALITY):
if (res.speciality is None):
res.speciality = tpt.value
elif (tpt.typ in pers_types):
pers_typ = tpt.typ
t = tpt.end_token
if (t.end_char > end_token.value.end_char):
end_token.value = t
if (t.next0_ is not None and t.next0_.is_char_of(":-")):
t = t.next0_
continue
if (t.end_char > end_char):
break
rli = t.get_referents()
if (rli is None):
continue
if (not t.is_newline_before and (isinstance(t.previous, TextToken))):
s = t.previous.term
if (s == "ИМЕНИ" or s == "ИМ"):
continue
if (s == "." and t.previous.previous is not None and t.previous.previous.is_value("ИМ", None)):
continue
for r in rli:
if (isinstance(r, PersonReferent)):
if (r != rli[0]):
continue
p = Utils.asObjectOrNull(r, PersonReferent)
if (pers_typ != TitleItemToken.Types.UNDEFINED):
if (t.previous is not None and t.previous.is_char('.')):
pers_typ = TitleItemToken.Types.UNDEFINED
typ = pr.calc_typ_from_attrs(p)
if (typ != TitleItemToken.Types.UNDEFINED):
pr.add(p, typ, 1)
pers_typ = typ
elif (pers_typ != TitleItemToken.Types.UNDEFINED):
pr.add(p, pers_typ, 1)
elif (t.previous is not None and t.previous.is_char('©')):
pers_typ = TitleItemToken.Types.WORKER
pr.add(p, pers_typ, 1)
else:
tt = t.next0_
first_pass3399 = True
while True:
if first_pass3399: first_pass3399 = False
else: tt = tt.next0_
if (not (tt is not None)): break
rr = tt.get_referent()
if (rr == res):
pers_typ = TitleItemToken.Types.WORKER
break
if (isinstance(rr, PersonReferent)):
if (pr.calc_typ_from_attrs(Utils.asObjectOrNull(r, PersonReferent)) != TitleItemToken.Types.UNDEFINED):
break
else:
continue
if (rr is not None):
break
tpt = TitleItemToken.try_attach(tt)
if (tpt is not None):
if (tpt.typ != TitleItemToken.Types.TYP and tpt.typ != TitleItemToken.Types.TYPANDTHEME):
break
tt = tpt.end_token
if (tt.end_char > end_token.value.end_char):
end_token.value = tt
continue
if (pers_typ == TitleItemToken.Types.UNDEFINED):
tt = t.previous
while tt is not None:
rr = tt.get_referent()
if (rr == res):
pers_typ = TitleItemToken.Types.WORKER
break
if (rr is not None):
break
if ((tt.is_value("СТУДЕНТ", None) or tt.is_value("СТУДЕНТКА", None) or tt.is_value("СЛУШАТЕЛЬ", None)) or tt.is_value("ДИПЛОМНИК", None) or tt.is_value("ИСПОЛНИТЕЛЬ", None)):
pers_typ = TitleItemToken.Types.WORKER
break
tpt = TitleItemToken.try_attach(tt)
if (tpt is not None and tpt.typ != TitleItemToken.Types.TYP):
break
tt = tt.previous
if (pers_typ != TitleItemToken.Types.UNDEFINED):
pr.add(p, pers_typ, 1)
else:
pr.add(p, pers_typ, 0.5)
if (t.end_char > end_token.value.end_char):
end_token.value = t
continue
if (r == rli[0]):
pers_typ = TitleItemToken.Types.UNDEFINED
if (isinstance(r, DateReferent)):
if (res.date is None):
res.date = Utils.asObjectOrNull(r, DateReferent)
if (t.end_char > end_token.value.end_char):
end_token.value = t
elif (isinstance(r, GeoReferent)):
if (res.city is None and r.is_city):
res.city = Utils.asObjectOrNull(r, GeoReferent)
if (t.end_char > end_token.value.end_char):
end_token.value = t
if (isinstance(r, OrganizationReferent)):
org0_ = Utils.asObjectOrNull(r, OrganizationReferent)
if ("курс" in org0_.types and org0_.number is not None):
i = 0
wrapi2673 = RefOutArgWrapper(0)
inoutres2674 = Utils.tryParseInt(org0_.number, wrapi2673)
i = wrapi2673.value
if (inoutres2674):
if (i > 0 and (i < 8)):
res.student_year = i
while org0_.higher is not None:
if (org0_.kind != OrganizationKind.DEPARTMENT):
break
org0_ = org0_.higher
if (org0_.kind != OrganizationKind.DEPARTMENT):
if (res.org0_ is None):
res.org0_ = org0_
elif (OrganizationReferent.can_be_higher(res.org0_, org0_)):
res.org0_ = org0_
if (t.end_char > end_token.value.end_char):
end_token.value = t
if ((isinstance(r, UriReferent)) or (isinstance(r, GeoReferent))):
if (t.end_char > end_token.value.end_char):
end_token.value = t
for ty in pers_types:
for p in pr.get_persons(ty):
if (pr.get_attr_name_for_type(ty) is not None):
res.add_slot(pr.get_attr_name_for_type(ty), p, False, 0)
if (res.get_slot_value(TitlePageReferent.ATTR_AUTHOR) is None):
for p in pr.get_persons(TitleItemToken.Types.UNDEFINED):
res.add_slot(TitlePageReferent.ATTR_AUTHOR, p, False, 0)
break
if (res.city is None and res.org0_ is not None):
s = res.org0_.find_slot(OrganizationReferent.ATTR_GEO, None, True)
if (s is not None and (isinstance(s.value, GeoReferent))):
if (s.value.is_city):
res.city = Utils.asObjectOrNull(s.value, GeoReferent)
if (res.date is None):
t = begin
first_pass3400 = True
while True:
if first_pass3400: first_pass3400 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= end_char)): break
city = Utils.asObjectOrNull(t.get_referent(), GeoReferent)
if (city is None):
continue
if (isinstance(t.next0_, TextToken)):
if (t.next0_.is_char_of(":,") or t.next0_.is_hiphen):
t = t.next0_
rt = t.kit.process_referent(DateAnalyzer.ANALYZER_NAME, t.next0_)
if (rt is not None):
rt.save_to_local_ontology()
res.date = Utils.asObjectOrNull(rt.referent, DateReferent)
if (kit is not None):
kit.embed_token(rt)
break
if (len(res.slots) == 0):
return None
else:
return res
def create_nickname(pr : 'PersonReferent', t : 'Token') -> 'Token':
has_keyw = False
is_br = False
first_pass3367 = True
while True:
if first_pass3367: first_pass3367 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_hiphen or t.is_comma or t.is_char_of(".:;")):
continue
if (t.morph.class0_.is_preposition):
continue
if (t.is_char('(')):
is_br = True
continue
if ((t.is_value("ПРОЗВИЩЕ", "ПРІЗВИСЬКО") or t.is_value("КЛИЧКА", None) or t.is_value("ПСЕВДОНИМ", "ПСЕВДОНІМ")) or t.is_value("ПСЕВДО", None) or t.is_value("ПОЗЫВНОЙ", "ПОЗИВНИЙ")):
has_keyw = True
continue
break
if (not has_keyw or t is None):
return None
if (BracketHelper.is_bracket(t, True)):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
ni = MiscHelper.get_text_value(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = br.end_token
tt = t.next0_
first_pass3368 = True
while True:
if first_pass3368: first_pass3368 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_comma_and):
continue
if (not BracketHelper.is_bracket(tt, True)):
break
br = BracketHelper.try_parse(tt, BracketParseAttr.NO, 100)
if (br is None):
break
ni = MiscHelper.get_text_value(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
tt = br.end_token
t = tt
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
return t
else:
ret = None
first_pass3369 = True
while True:
if first_pass3369: first_pass3369 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_comma_and):
continue
if (ret is not None and t.chars.is_all_lower):
break
if (t.whitespaces_before_count > 2):
break
pli = PersonItemToken.try_attach_list(t, None, PersonItemToken.ParseAttr.NO, 10)
if (pli is not None and ((len(pli) == 1 or len(pli) == 2))):
ni = MiscHelper.get_text_value(pli[0].begin_token, pli[len(pli) - 1].end_token, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = pli[len(pli) - 1].end_token
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
ret = t
continue
if ((isinstance(t, ReferentToken)) and not t.chars.is_all_lower and t.begin_token == t.end_token):
val = MiscHelper.get_text_value_of_meta_token(Utils.asObjectOrNull(t, ReferentToken), GetTextAttr.NO)
pr.add_slot(PersonReferent.ATTR_NICKNAME, val, False, 0)
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
ret = t
continue
break
return ret
return None
def __tryParse(t: 'Token',
is_in_lit: bool,
max_char: int = 0) -> typing.List['ReferentToken']:
if (t is None):
return None
is_bracket_regime = False
if (t.previous is not None and t.previous.isChar('(')):
is_bracket_regime = True
blt = BookLinkToken.tryParse(t, 0)
if (blt is None):
blt = BookLinkToken.tryParseAuthor(t, FioTemplateType.UNDEFINED)
if (blt is None and not is_bracket_regime):
return None
t0 = t
coef = 0
is_electr_res = False
decree = None
regtyp = BookLinkAnalyzer.RegionTyp.UNDEFINED
num = None
spec_see = None
book_prev = None
if (is_bracket_regime):
regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS
elif (blt.typ == BookLinkTyp.PERSON):
if (not is_in_lit):
return None
regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS
elif (blt.typ == BookLinkTyp.NUMBER):
num = blt.value
t = blt.end_token.next0_
if (t is None or t.is_newline_before):
return None
if (not t.is_whitespace_before):
if (isinstance(t, NumberToken)):
n = (t).value
if ((((n == "3" or n == "0")) and not t.is_whitespace_after
and (isinstance(t.next0_, TextToken)))
and t.next0_.chars.is_all_lower):
pass
else:
return None
elif (not ((isinstance(t, TextToken)))
or t.chars.is_all_lower):
r = t.getReferent()
if (isinstance(r, PersonReferent)):
pass
elif (is_in_lit and r is not None
and r.type_name == "DECREE"):
pass
else:
return None
first_pass2757 = True
while True:
if first_pass2757: first_pass2757 = False
else: t = t.next0_
if (not (t is not None)): break
if (isinstance(t, NumberToken)):
break
if (not ((isinstance(t, TextToken)))):
break
if (BracketHelper.canBeStartOfSequence(t, True, False)):
break
if (not t.chars.is_letter):
continue
bbb = BookLinkToken.tryParse(t, 0)
if (bbb is not None):
if (bbb.typ == BookLinkTyp.TAMZE):
spec_see = bbb
t = bbb.end_token.next0_
break
if (bbb.typ == BookLinkTyp.SEE):
t = bbb.end_token
continue
break
if (spec_see is not None and spec_see.typ == BookLinkTyp.TAMZE):
coef += 1
max0_ = 1000
tt = t0
while tt is not None and max0_ > 0:
if (isinstance(tt.getReferent(), BookLinkRefReferent)):
book_prev = (tt.getReferent()).book
break
tt = tt.previous
max0_ -= 1
blt1 = BookLinkToken.tryParseAuthor(t, FioTemplateType.UNDEFINED)
if (blt1 is not None and blt1.typ == BookLinkTyp.PERSON):
regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS
else:
ok = False
tt = t
first_pass2758 = True
while True:
if first_pass2758: first_pass2758 = False
else: tt = (None if tt is None else tt.next0_)
if (not (tt is not None)): break
if (tt.is_newline_before):
break
if (is_in_lit and tt.getReferent() is not None
and tt.getReferent().type_name == "DECREE"):
ok = True
decree = tt
break
bbb = BookLinkToken.tryParse(tt, 0)
if (bbb is None):
continue
if (bbb.typ == BookLinkTyp.ELECTRONRES):
is_electr_res = True
ok = True
break
if (bbb.typ == BookLinkTyp.DELIMETER):
tt = bbb.end_token.next0_
if (BookLinkToken.tryParseAuthor(
tt, FioTemplateType.UNDEFINED) is not None):
ok = True
break
bbb = BookLinkToken.tryParse(tt, 0)
if (bbb is not None):
if (bbb.typ == BookLinkTyp.EDITORS
or bbb.typ == BookLinkTyp.TRANSLATE
or bbb.typ == BookLinkTyp.SOSTAVITEL):
ok = True
break
if (not ok and not is_in_lit):
if (BookLinkToken.checkLinkBefore(t0, num)):
pass
else:
return None
regtyp = BookLinkAnalyzer.RegionTyp.NAME
else:
return None
res = BookLinkReferent()
corr_authors = list()
t00 = t
blt00 = None
start_of_name = None
prev_pers_templ = FioTemplateType.UNDEFINED
if (regtyp == BookLinkAnalyzer.RegionTyp.AUTHORS):
first_pass2759 = True
while True:
if first_pass2759: first_pass2759 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char > 0 and t.begin_char >= max_char):
break
if (t.isCharOf(".;") or t.is_comma_and):
continue
if (t.isChar('/')):
break
if ((t.isChar('(') and t.next0_ is not None
and t.next0_.isValue("EDS", None))
and t.next0_.next0_ is not None
and t.next0_.next0_.isChar(')')):
t = t.next0_.next0_.next0_
break
blt = BookLinkToken.tryParseAuthor(t, prev_pers_templ)
if (blt is None and t.previous is not None
and t.previous.is_and):
blt = BookLinkToken.tryParseAuthor(
t.previous, FioTemplateType.UNDEFINED)
if (blt is None):
if ((isinstance(t.getReferent(), OrganizationReferent))
and blt00 is not None):
bbb2 = BookLinkToken.tryParse(t.next0_, 0)
if (bbb2 is not None):
if (bbb2.typ == BookLinkTyp.YEAR):
res.addSlot(BookLinkReferent.ATTR_AUTHOR,
t.getReferent(), False, 0)
res.year = int(bbb2.value)
coef += .5
t = bbb2.end_token.next0_
break
if (blt.typ == BookLinkTyp.PERSON):
tt2 = blt.end_token.next0_
bbb2 = BookLinkToken.tryParse(tt2, 0)
if (bbb2 is not None):
if (bbb2.typ == BookLinkTyp.YEAR):
res.year = int(bbb2.value)
coef += .5
blt.end_token = bbb2.end_token
blt00 = (None)
if (blt00 is not None
and ((blt00.end_token.next0_ == blt.begin_token
or blt.begin_token.previous.isChar('.')))):
tt11 = blt.end_token.next0_
nex = BookLinkToken.tryParse(tt11, 0)
if (nex is not None
and nex.typ == BookLinkTyp.ANDOTHERS):
pass
else:
if (tt11 is None):
break
if (tt11.isChar('/') and tt11.next0_ is not None
and tt11.next0_.isChar('/')):
break
if (tt11.isChar(':')):
break
if ((str(blt).find('.') < 0)
and str(blt00).find('.') > 0):
break
if ((isinstance(tt11, TextToken))
and tt11.chars.is_all_lower):
break
if (tt11.isCharOf(",.;")
and tt11.next0_ is not None):
tt11 = tt11.next0_
nex = BookLinkToken.tryParse(tt11, 0)
if (nex is not None
and nex.typ != BookLinkTyp.PERSON
and nex.typ != BookLinkTyp.ANDOTHERS):
break
elif (
(blt00 is not None
and blt00.person_template != FioTemplateType.UNDEFINED
and blt.person_template != blt00.person_template)
and blt.person_template
== FioTemplateType.NAMESURNAME):
if (blt.end_token.next0_ is None
or not blt.end_token.next0_.is_comma_and):
break
if (BookLinkToken.tryParseAuthor(
blt.end_token.next0_.next0_,
FioTemplateType.UNDEFINED) is not None):
pass
else:
break
if (blt00 is None and blt.person_template
== FioTemplateType.NAMESURNAME):
tt = blt.end_token.next0_
if (tt is not None and tt.is_hiphen):
tt = tt.next0_
if (isinstance(tt, NumberToken)):
break
BookLinkAnalyzer.__addAuthor(res, blt)
coef += 1
t = blt.end_token
if (isinstance(t.getReferent(), PersonReferent)):
corr_authors.append(
Utils.asObjectOrNull(t, ReferentToken))
blt00 = blt
prev_pers_templ = blt.person_template
start_of_name = blt.start_of_name
if ((start_of_name) is not None):
t = t.next0_
break
continue
if (blt.typ == BookLinkTyp.ANDOTHERS):
coef += .5
t = blt.end_token.next0_
res.authors_and_other = True
break
break
if (t is None):
return None
if ((t.is_newline_before and t != t0 and num is None) and res.findSlot(
BookLinkReferent.ATTR_AUTHOR, None, True) is None):
return None
if (start_of_name is None):
if (t.chars.is_all_lower):
coef -= (1)
if (t.chars.is_latin_letter and not is_electr_res and num is None):
if (res.getSlotValue(BookLinkReferent.ATTR_AUTHOR) is None):
return None
tn0 = t
tn1 = None
uri = None
next_num = None
wrapnn393 = RefOutArgWrapper(0)
inoutres394 = Utils.tryParseInt(Utils.ifNotNull(num, ""), wrapnn393)
nn = wrapnn393.value
if (inoutres394):
next_num = str((nn + 1))
br = (BracketHelper.tryParse(
t,
Utils.valToEnum(
(BracketParseAttr.CANCONTAINSVERBS) |
(BracketParseAttr.CANBEMANYLINES), BracketParseAttr), 100)
if BracketHelper.canBeStartOfSequence(t, True, False) else None)
if (br is not None):
t = t.next0_
pages = None
first_pass2760 = True
while True:
if first_pass2760: first_pass2760 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char > 0 and t.begin_char >= max_char):
break
if (br is not None and br.end_token == t):
tn1 = t
break
tit = TitleItemToken.tryAttach(t)
if (tit is not None):
if ((tit.typ == TitleItemToken.Types.TYP and tn0 == t
and br is None) and BracketHelper.canBeStartOfSequence(
tit.end_token.next0_, True, False)):
br = BracketHelper.tryParse(tit.end_token.next0_,
BracketParseAttr.NO, 100)
if (br is not None):
coef += (1)
if (num is not None):
coef += 1
tn0 = br.begin_token
tn1 = br.end_token
res.typ = tit.value.lower()
t = br.end_token.next0_
break
if (t.is_newline_before and t != tn0):
if (br is not None and (t.end_char < br.end_char)):
pass
elif (not MiscHelper.canBeStartOfSentence(t)):
pass
else:
if (t.newlines_before_count > 1):
break
if ((isinstance(t, NumberToken)) and num is not None
and (t).int_value is not None):
if (num == str(((t).int_value - 1))):
break
elif (num is not None):
pass
else:
nnn = NounPhraseHelper.tryParse(
t.previous,
Utils.valToEnum(
((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.PARSEADVERBS) |
(NounPhraseParseAttr.PARSENUMERICASADJECTIVE))
| (NounPhraseParseAttr.MULTILINES),
NounPhraseParseAttr), 0)
if (nnn is not None and nnn.end_char >= t.end_char):
pass
else:
break
if (t.isCharOf(".;") and t.whitespaces_after_count > 0):
tit = TitleItemToken.tryAttach(t.next0_)
if ((tit) is not None):
if (tit.typ == TitleItemToken.Types.TYP):
break
stop = True
words = 0
notwords = 0
tt = t.next0_
first_pass2761 = True
while True:
if first_pass2761: first_pass2761 = False
else: tt = tt.next0_
if (not (tt is not None)): break
blt0 = BookLinkToken.tryParse(tt, 0)
if (blt0 is None):
if (tt.is_newline_before):
break
if ((isinstance(tt, TextToken)) and
not tt.getMorphClassInDictionary().is_undefined
):
words += 1
else:
notwords += 1
if (words > 6 and words > (notwords * 4)):
stop = False
break
continue
if ((blt0.typ == BookLinkTyp.DELIMETER
or blt0.typ == BookLinkTyp.TRANSLATE
or blt0.typ == BookLinkTyp.TYPE)
or blt0.typ == BookLinkTyp.GEO
or blt0.typ == BookLinkTyp.PRESS):
stop = False
break
if (br is not None
and br.end_token.previous.end_char > t.end_char):
stop = False
if (stop):
break
if (t == decree):
t = t.next0_
break
blt = BookLinkToken.tryParse(t, 0)
if (blt is None):
tn1 = t
continue
if (blt.typ == BookLinkTyp.DELIMETER):
break
if (((blt.typ == BookLinkTyp.MISC or blt.typ
== BookLinkTyp.TRANSLATE or blt.typ == BookLinkTyp.NAMETAIL)
or blt.typ == BookLinkTyp.TYPE
or blt.typ == BookLinkTyp.VOLUME)
or blt.typ == BookLinkTyp.PAGERANGE
or blt.typ == BookLinkTyp.PAGES):
coef += 1
break
if (blt.typ == BookLinkTyp.GEO or blt.typ == BookLinkTyp.PRESS):
if (t.previous.is_hiphen or t.previous.isCharOf(".;")
or blt.add_coef > 0):
break
if (blt.typ == BookLinkTyp.YEAR):
if (t.previous is not None and t.previous.is_comma):
break
if (blt.typ == BookLinkTyp.ELECTRONRES):
is_electr_res = True
break
if (blt.typ == BookLinkTyp.URL):
if (t == tn0 or t.previous.isCharOf(":.")):
is_electr_res = True
break
tn1 = t
if (tn1 is None and start_of_name is None):
if (is_electr_res):
uri_re = BookLinkReferent()
rt0 = ReferentToken(uri_re, t00, t)
rts0 = list()
bref0 = BookLinkRefReferent._new389(uri_re)
if (num is not None):
bref0.number = num
rt01 = ReferentToken(bref0, t0, rt0.end_token)
ok = False
while t is not None:
if (t.is_newline_before):
break
blt0 = BookLinkToken.tryParse(t, 0)
if (blt0 is not None):
if (isinstance(blt0.ref, UriReferent)):
uri_re.addSlot(
BookLinkReferent.ATTR_URL,
Utils.asObjectOrNull(blt0.ref, UriReferent),
False, 0)
ok = True
t = blt0.end_token
rt0.end_token = rt01.end_token = t
t = t.next0_
if (ok):
rts0.append(rt01)
rts0.append(rt0)
return rts0
if (decree is not None and num is not None):
rts0 = list()
bref0 = BookLinkRefReferent._new389(decree.getReferent())
if (num is not None):
bref0.number = num
rt01 = ReferentToken(bref0, t0, decree)
t = decree.next0_
while t is not None:
if (t.is_newline_before):
break
if (isinstance(t, TextToken)):
if ((t).is_pure_verb):
return None
rt01.end_token = t
t = t.next0_
rts0.append(rt01)
return rts0
if (book_prev is not None):
tt = t
while tt is not None and ((tt.isCharOf(",.") or tt.is_hiphen)):
tt = tt.next0_
blt0 = BookLinkToken.tryParse(tt, 0)
if (blt0 is not None and blt0.typ == BookLinkTyp.PAGERANGE):
rts0 = list()
bref0 = BookLinkRefReferent._new389(book_prev)
if (num is not None):
bref0.number = num
bref0.pages = blt0.value
rt00 = ReferentToken(bref0, t0, blt0.end_token)
rts0.append(rt00)
return rts0
return None
if (br is not None
and ((tn1 == br.end_token or tn1 == br.end_token.previous))):
tn0 = tn0.next0_
tn1 = tn1.previous
if (start_of_name is None):
while tn0 is not None:
if (tn0.isCharOf(":,~")):
tn0 = tn0.next0_
else:
break
while tn1 is not None and tn1.begin_char > tn0.begin_char:
if (tn1.isCharOf(".;,:(~") or tn1.is_hiphen
or tn1.isValue("РЕД", None)):
pass
else:
break
tn1 = tn1.previous
nam = MiscHelper.getTextValue(
tn0, tn1,
Utils.valToEnum(
(GetTextAttr.KEEPQUOTES) | (GetTextAttr.KEEPREGISTER),
GetTextAttr))
if (start_of_name is not None):
if (nam is None or (len(nam) < 3)):
nam = start_of_name
else:
nam = "{0}{1}{2}".format(
start_of_name, (" " if tn0.is_whitespace_before else ""),
nam)
if (nam is None):
return None
res.name = nam
if (num is None and not is_in_lit):
if (len(nam) < 20):
return None
coef -= (2)
if (len(nam) > 500):
coef -= (math.floor(len(nam) / 500))
if (is_bracket_regime):
coef -= 1
if (len(nam) > 200):
if (num is None):
return None
if (res.findSlot(BookLinkReferent.ATTR_AUTHOR, None, True) is None
and not BookLinkToken.checkLinkBefore(t0, num)):
return None
en = 0
ru = 0
ua = 0
cha = 0
nocha = 0
chalen = 0
lt0 = tn0
lt1 = tn1
if (tn1 is None):
if (t is None):
return None
lt0 = t0
lt1 = t
tn1 = t.previous
tt = lt0
while tt is not None and tt.end_char <= lt1.end_char:
if ((isinstance(tt, TextToken)) and tt.chars.is_letter):
if (tt.chars.is_latin_letter):
en += 1
elif (tt.morph.language.is_ua):
ua += 1
elif (tt.morph.language.is_ru):
ru += 1
if (tt.length_char > 2):
cha += 1
chalen += tt.length_char
elif (not ((isinstance(tt, ReferentToken)))):
nocha += 1
tt = tt.next0_
if (ru > (ua + en)):
res.lang = "RU"
elif (ua > (ru + en)):
res.lang = "UA"
elif (en > (ru + ua)):
res.lang = "EN"
if (nocha > 3 and nocha > cha and start_of_name is None):
if (nocha > (math.floor(chalen / 3))):
coef -= (2)
if (res.lang == "EN"):
tt = tn0.next0_
first_pass2762 = True
while True:
if first_pass2762: first_pass2762 = False
else: tt = tt.next0_
if (not (tt is not None and (tt.end_char < tn1.end_char))):
break
if (tt.is_comma and tt.next0_ is not None
and ((not tt.next0_.chars.is_all_lower or
(isinstance(tt.next0_, ReferentToken))))):
if (tt.next0_.next0_ is not None
and tt.next0_.next0_.is_comma_and):
if (isinstance(tt.next0_, ReferentToken)):
pass
else:
continue
nam = MiscHelper.getTextValue(
tn0, tt.previous,
Utils.valToEnum((GetTextAttr.KEEPQUOTES) |
(GetTextAttr.KEEPREGISTER),
GetTextAttr))
if (nam is not None and len(nam) > 15):
res.name = nam
break
rt = ReferentToken(res, t00, tn1)
authors = True
edits = False
br = (None)
first_pass2763 = True
while True:
if first_pass2763: first_pass2763 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char > 0 and t.begin_char >= max_char):
break
if (BracketHelper.canBeStartOfSequence(t, False, False)):
br = BracketHelper.tryParse(t, BracketParseAttr.CANBEMANYLINES,
100)
if (br is not None and br.length_char > 300):
br = (None)
blt = BookLinkToken.tryParse(t, 0)
if (t.is_newline_before and not t.isChar('/')
and not t.previous.isChar('/')):
if (blt is not None and blt.typ == BookLinkTyp.NUMBER):
break
if (t.previous.isCharOf(":")):
pass
elif (blt is not None and ((
((blt.typ == BookLinkTyp.DELIMETER or blt.typ
== BookLinkTyp.PAGERANGE or blt.typ == BookLinkTyp.PAGES)
or blt.typ == BookLinkTyp.GEO or blt.typ
== BookLinkTyp.PRESS) or blt.typ == BookLinkTyp.N))):
pass
elif (num is not None and BookLinkToken.tryParseAuthor(
t, FioTemplateType.UNDEFINED) is not None):
pass
elif (num is not None and blt is not None
and blt.typ != BookLinkTyp.NUMBER):
pass
elif (br is not None and (t.end_char < br.end_char)
and t.begin_char > br.begin_char):
pass
else:
ok = False
mmm = 50
tt = t.next0_
while tt is not None and mmm > 0:
if (tt.is_newline_before):
blt2 = BookLinkToken.tryParse(tt, 0)
if (blt2 is not None
and blt2.typ == BookLinkTyp.NUMBER
and blt2.value == next_num):
ok = True
break
if (blt2 is not None):
if (blt2.typ == BookLinkTyp.PAGES
or blt2.typ == BookLinkTyp.GEO
or blt2.typ == BookLinkTyp.PRESS):
ok = True
break
tt = tt.next0_
mmm -= 1
if (not ok):
npt = NounPhraseHelper.tryParse(
t.previous,
Utils.valToEnum(
((NounPhraseParseAttr.MULTILINES) |
(NounPhraseParseAttr.PARSEADVERBS) |
(NounPhraseParseAttr.PARSEPREPOSITION)) |
(NounPhraseParseAttr.PARSEVERBS) |
(NounPhraseParseAttr.PARSEPRONOUNS),
NounPhraseParseAttr), 0)
if (npt is not None and npt.end_char >= t.end_char):
ok = True
if (not ok):
break
rt.end_token = t
if (blt is not None):
rt.end_token = blt.end_token
if (t.isCharOf(".,") or t.is_hiphen):
continue
if (t.isValue("С", None)):
pass
if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and blt is not None
and blt.typ == BookLinkTyp.EDITORS):
edits = True
t = blt.end_token
coef += 1
continue
if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and blt is not None
and blt.typ == BookLinkTyp.SOSTAVITEL):
edits = False
t = blt.end_token
coef += 1
continue
if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and authors):
blt2 = BookLinkToken.tryParseAuthor(t, prev_pers_templ)
if (blt2 is not None and blt2.typ == BookLinkTyp.PERSON):
prev_pers_templ = blt2.person_template
if (not edits):
BookLinkAnalyzer.__addAuthor(res, blt2)
coef += 1
t = blt2.end_token
continue
if (blt2 is not None and blt2.typ == BookLinkTyp.ANDOTHERS):
if (not edits):
res.authors_and_other = True
coef += 1
t = blt2.end_token
continue
authors = False
if (blt is None):
continue
if (blt.typ == BookLinkTyp.ELECTRONRES
or blt.typ == BookLinkTyp.URL):
is_electr_res = True
if (blt.typ == BookLinkTyp.ELECTRONRES):
coef += 1.5
else:
coef += .5
if (isinstance(blt.ref, UriReferent)):
res.addSlot(BookLinkReferent.ATTR_URL,
Utils.asObjectOrNull(blt.ref, UriReferent),
False, 0)
elif (blt.typ == BookLinkTyp.YEAR):
if (res.year == 0):
res.year = int(blt.value)
coef += .5
elif (blt.typ == BookLinkTyp.DELIMETER):
coef += 1
if (blt.length_char == 2):
regtyp = BookLinkAnalyzer.RegionTyp.SECOND
else:
regtyp = BookLinkAnalyzer.RegionTyp.FIRST
elif (
(((blt.typ == BookLinkTyp.MISC or blt.typ == BookLinkTyp.TYPE
or blt.typ == BookLinkTyp.PAGES) or blt.typ
== BookLinkTyp.NAMETAIL or blt.typ == BookLinkTyp.TRANSLATE)
or blt.typ == BookLinkTyp.PRESS
or blt.typ == BookLinkTyp.VOLUME)
or blt.typ == BookLinkTyp.N):
coef += 1
elif (blt.typ == BookLinkTyp.PAGERANGE):
pages = blt
coef += 1
if (is_bracket_regime and blt.end_token.next0_ is not None
and blt.end_token.next0_.isChar(')')):
coef += (2)
if (res.name is not None
and res.findSlot(BookLinkReferent.ATTR_AUTHOR,
None, True) is not None):
coef = (10)
elif (blt.typ == BookLinkTyp.GEO
and ((regtyp == BookLinkAnalyzer.RegionTyp.SECOND
or regtyp == BookLinkAnalyzer.RegionTyp.FIRST))):
coef += 1
elif (blt.typ == BookLinkTyp.GEO and t.previous is not None
and t.previous.isChar('.')):
coef += 1
elif (blt.typ == BookLinkTyp.ANDOTHERS):
coef += 1
if (authors):
res.authors_and_other = True
coef += blt.add_coef
t = blt.end_token
if ((coef < 2.5) and num is not None):
if (BookLinkToken.checkLinkBefore(t0, num)):
coef += (2)
elif (BookLinkToken.checkLinkAfter(rt.end_token, num)):
coef += (1)
if (rt.length_char > 500):
return None
if (is_in_lit):
coef += 1
if (coef < 2.5):
if (is_electr_res and uri is not None):
pass
elif (coef >= 2 and is_in_lit):
pass
else:
return None
for rr in corr_authors:
pits0 = PersonItemToken.tryAttachList(
rr.begin_token, None,
PersonItemToken.ParseAttr.CANINITIALBEDIGIT, 10)
if (pits0 is None or (len(pits0) < 2)):
continue
if (pits0[0].typ == PersonItemToken.ItemType.VALUE):
exi = False
for i in range(len(rr.referent.slots) - 1, -1, -1):
s = rr.referent.slots[i]
if (s.type_name == PersonReferent.ATTR_LASTNAME):
ln = Utils.asObjectOrNull(s.value, str)
if (ln is None):
continue
if (ln == pits0[0].value):
exi = True
continue
if (ln.find('-') > 0):
ln = ln[0:0 + ln.find('-')]
if (pits0[0].begin_token.isValue(ln, None)):
del rr.referent.slots[i]
if (not exi):
rr.referent.addSlot(PersonReferent.ATTR_LASTNAME,
pits0[0].value, False, 0)
rts = list()
bref = BookLinkRefReferent._new389(res)
if (num is not None):
bref.number = num
rt1 = ReferentToken(bref, t0, rt.end_token)
if (pages is not None):
if (pages.value is not None):
bref.pages = pages.value
rt.end_token = pages.begin_token.previous
rts.append(rt1)
rts.append(rt)
return rts
def process(self, kit: 'AnalysisKit') -> None:
ad = kit.getAnalyzerData(self)
is_lit_block = 0
refs_by_num = dict()
rts = []
t = kit.first_token
first_pass2754 = True
while True:
if first_pass2754: first_pass2754 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.isChar('(')):
br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100)
if (br is not None and br.length_char > 70
and (br.length_char < 400)):
if (br.is_newline_after
or ((br.end_token.next0_ is not None
and br.end_token.next0_.isCharOf(".;")))):
rts = BookLinkAnalyzer.__tryParse(
t.next0_, False, br.end_char)
if (rts is not None and len(rts) >= 1):
if (len(rts) > 1):
rts[1].referent = ad.registerReferent(
rts[1].referent)
kit.embedToken(rts[1])
(rts[0].referent).book = Utils.asObjectOrNull(
rts[1].referent, BookLinkReferent)
if (rts[0].begin_char == rts[1].begin_char):
rts[0].begin_token = rts[1]
if (rts[0].end_char == rts[1].end_char):
rts[0].end_token = rts[1]
rts[0].begin_token = t
rts[0].end_token = br.end_token
(rts[0].referent).typ = BookLinkRefType.INLINE
rts[0].referent = ad.registerReferent(
rts[0].referent)
kit.embedToken(rts[0])
t = (rts[0])
continue
if (not t.is_newline_before):
continue
if (is_lit_block <= 0):
tt = BookLinkToken.parseStartOfLitBlock(t)
if (tt is not None):
is_lit_block = 5
t = tt
continue
rts = BookLinkAnalyzer.__tryParse(t, is_lit_block > 0, 0)
if (rts is None or (len(rts) < 1)):
is_lit_block -= 1
if ((is_lit_block) < 0):
is_lit_block = 0
continue
is_lit_block += 1
if ((is_lit_block) > 5):
is_lit_block = 5
if (len(rts) > 1):
rts[1].referent = ad.registerReferent(rts[1].referent)
kit.embedToken(rts[1])
(rts[0].referent).book = Utils.asObjectOrNull(
rts[1].referent, BookLinkReferent)
if (rts[0].begin_char == rts[1].begin_char):
rts[0].begin_token = rts[1]
if (rts[0].end_char == rts[1].end_char):
rts[0].end_token = rts[1]
re = Utils.asObjectOrNull(rts[0].referent, BookLinkRefReferent)
re = (Utils.asObjectOrNull(ad.registerReferent(re),
BookLinkRefReferent))
rts[0].referent = (re)
kit.embedToken(rts[0])
t = (rts[0])
if (re.number is not None):
li = []
wrapli385 = RefOutArgWrapper(None)
inoutres386 = Utils.tryGetValue(refs_by_num, re.number,
wrapli385)
li = wrapli385.value
if (not inoutres386):
li = list()
refs_by_num[re.number] = li
li.append(re)
t = kit.first_token
first_pass2755 = True
while True:
if first_pass2755: first_pass2755 = False
else: t = t.next0_
if (not (t is not None)): break
if (not ((isinstance(t, TextToken)))):
continue
rt = BookLinkAnalyzer.__tryParseShortInline(t)
if (rt is None):
continue
re = Utils.asObjectOrNull(rt.referent, BookLinkRefReferent)
li = []
wrapli387 = RefOutArgWrapper(None)
inoutres388 = Utils.tryGetValue(refs_by_num,
Utils.ifNotNull(re.number, ""),
wrapli387)
li = wrapli387.value
if (not inoutres388):
continue
i = 0
while i < len(li):
if (t.begin_char < li[i].occurrence[0].begin_char):
break
i += 1
if (i >= len(li)):
continue
re.book = li[i].book
if (re.pages is None):
re.pages = li[i].pages
re.typ = BookLinkRefType.INLINE
re = (Utils.asObjectOrNull(ad.registerReferent(re),
BookLinkRefReferent))
rt.referent = (re)
kit.embedToken(rt)
t = (rt)
def try_attach(self, t : 'Token', for_ontology : bool=False) -> 'ReferentToken':
if (t is None):
return None
rt0 = self.__try_attach_spec(t)
if (rt0 is not None):
return rt0
if (t.chars.is_all_lower):
if (not t.is_whitespace_after and (isinstance(t.next0_, NumberToken))):
if (t.previous is None or t.is_whitespace_before or t.previous.is_char_of(",:")):
pass
else:
return None
else:
return None
tmp = io.StringIO()
t1 = t
hiph = False
ok = True
nums = 0
chars = 0
w = t1.next0_
first_pass3148 = True
while True:
if first_pass3148: first_pass3148 = False
else: w = w.next0_
if (not (w is not None)): break
if (w.is_whitespace_before and not for_ontology):
break
if (w.is_char_of("/\\_") or w.is_hiphen):
hiph = True
print('-', end="", file=tmp)
continue
hiph = False
nt = Utils.asObjectOrNull(w, NumberToken)
if (nt is not None):
if (nt.typ != NumberSpellingType.DIGIT):
break
t1 = (nt)
print(nt.get_source_text(), end="", file=tmp)
nums += 1
continue
tt = Utils.asObjectOrNull(w, TextToken)
if (tt is None):
break
if (tt.length_char > 3):
ok = False
break
if (not str.isalpha(tt.term[0])):
if (tt.is_char_of(",:") or BracketHelper.can_be_end_of_sequence(tt, False, None, False)):
break
if (not tt.is_char_of("+*&^#@!")):
ok = False
break
chars += 1
t1 = (tt)
print(tt.get_source_text(), end="", file=tmp)
if (not for_ontology):
if ((tmp.tell() < 1) or not ok or hiph):
return None
if (tmp.tell() > 12):
return None
last = Utils.getCharAtStringIO(tmp, tmp.tell() - 1)
if (last == '!'):
return None
if ((nums + chars) == 0):
return None
if (not self.__check_attach(t, t1)):
return None
new_dr = DenominationReferent()
new_dr._add_value(t, t1)
return ReferentToken(new_dr, t, t1)
def __try_attach(t: 'Token', prev: typing.List['DateItemToken'],
detail_regime: bool) -> 'DateItemToken':
from pullenti.ner.measure.internal.MeasureToken import MeasureToken
if (t is None):
return None
nt = Utils.asObjectOrNull(t, NumberToken)
begin = t
end = t
is_in_brack = False
if ((BracketHelper.can_be_start_of_sequence(t, False, False)
and t.next0_ is not None and (isinstance(t.next0_, NumberToken)))
and BracketHelper.can_be_end_of_sequence(
t.next0_.next0_, False, None, False)):
nt = (Utils.asObjectOrNull(t.next0_, NumberToken))
end = t.next0_.next0_
is_in_brack = True
if ((t.is_newline_before and BracketHelper.is_bracket(t, False) and
(isinstance(t.next0_, NumberToken)))
and BracketHelper.is_bracket(t.next0_.next0_, False)):
nt = (Utils.asObjectOrNull(t.next0_, NumberToken))
end = t.next0_.next0_
is_in_brack = True
if (nt is not None):
if (nt.int_value is None):
return None
if (nt.typ == NumberSpellingType.WORDS):
if (nt.morph.class0_.is_noun
and not nt.morph.class0_.is_adjective):
if (t.next0_ is not None
and ((t.next0_.is_value("КВАРТАЛ", None)
or t.next0_.is_value("ПОЛУГОДИЕ", None)
or t.next0_.is_value("ПІВРІЧЧЯ", None)))):
pass
else:
return None
if (NumberHelper.try_parse_age(nt) is not None):
return None
tt = None
res = DateItemToken._new628(begin, end,
DateItemToken.DateItemType.NUMBER,
nt.int_value, nt.morph)
if ((res.int_value == 20 and (isinstance(nt.next0_, NumberToken))
and nt.next0_.int_value is not None)
and nt.next0_.length_char == 2 and prev is not None):
num = 2000 + nt.next0_.int_value
if ((num < 2030) and len(prev) > 0 and prev[len(prev) - 1].typ
== DateItemToken.DateItemType.MONTH):
ok = False
if (nt.whitespaces_after_count == 1):
ok = True
elif (nt.is_newline_after and nt.is_newline_after):
ok = True
if (ok):
nt = (Utils.asObjectOrNull(nt.next0_, NumberToken))
res.end_token = nt
res.int_value = num
if (res.int_value == 20 or res.int_value == 201):
tt = t.next0_
if (tt is not None and tt.is_char('_')):
while tt is not None:
if (not tt.is_char('_')):
break
tt = tt.next0_
tt = DateItemToken.__test_year_rus_word(tt, False)
if (tt is not None):
res.int_value = 0
res.end_token = tt
res.typ = DateItemToken.DateItemType.YEAR
return res
if (res.int_value <= 12 and t.next0_ is not None
and (t.whitespaces_after_count < 3)):
tt = t.next0_
if (tt.is_value("ЧАС", None)):
if (((isinstance(t.previous, TextToken))
and not t.previous.chars.is_letter
and not t.is_whitespace_before)
and (isinstance(t.previous.previous, NumberToken))
and not t.previous.is_whitespace_before):
pass
else:
res.typ = DateItemToken.DateItemType.HOUR
res.end_token = tt
tt = tt.next0_
if (tt is not None and tt.is_char('.')):
res.end_token = tt
tt = tt.next0_
first_pass3072 = True
while True:
if first_pass3072: first_pass3072 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_value("УТРО", "РАНОК")):
res.end_token = tt
res.typ = DateItemToken.DateItemType.HOUR
return res
if (tt.is_value("ВЕЧЕР", "ВЕЧІР")):
res.end_token = tt
res.int_value += 12
res.typ = DateItemToken.DateItemType.HOUR
return res
if (tt.is_value("ДЕНЬ", None)):
res.end_token = tt
if (res.int_value < 10):
res.int_value += 12
res.typ = DateItemToken.DateItemType.HOUR
return res
if (tt.is_value("НОЧЬ", "НІЧ")):
res.end_token = tt
if (res.int_value == 12):
res.int_value = 0
elif (res.int_value > 9):
res.int_value += 12
res.typ = DateItemToken.DateItemType.HOUR
return res
if (tt.is_comma or tt.morph.class0_.is_adverb):
continue
break
if (res.typ == DateItemToken.DateItemType.HOUR):
return res
can_be_year_ = True
if (prev is not None and len(prev) > 0 and prev[len(prev) - 1].typ
== DateItemToken.DateItemType.MONTH):
pass
elif ((prev is not None and len(prev) >= 4 and
prev[len(prev) - 1].typ == DateItemToken.DateItemType.DELIM)
and prev[len(prev) - 2].can_by_month):
pass
elif (nt.next0_ is not None
and ((nt.next0_.is_value("ГОД", None)
or nt.next0_.is_value("РІК", None)))):
if (res.int_value < 1000):
can_be_year_ = False
tt = DateItemToken.__test_year_rus_word(nt.next0_, False)
if (tt is not None and DateItemToken.__is_new_age(tt.next0_)):
res.typ = DateItemToken.DateItemType.YEAR
res.end_token = tt
elif (can_be_year_):
if (res.can_be_year
or res.typ == DateItemToken.DateItemType.NUMBER):
tt = DateItemToken.__test_year_rus_word(
nt.next0_, res.is_newline_before)
if ((tt) is not None):
if ((tt.is_value("Г", None)
and not tt.is_whitespace_before
and t.previous is not None)
and ((t.previous.is_value("КОРПУС", None)
or t.previous.is_value("КОРП", None)))):
pass
elif (
(((nt.next0_.is_value("Г", None) and
(t.whitespaces_before_count < 3) and t.previous
is not None) and t.previous.is_value("Я", None)
and t.previous.previous is not None)
and t.previous.previous.is_char_of("\\/")
and t.previous.previous.previous is not None)
and t.previous.previous.previous.is_value(
"А", None)):
return None
elif (nt.next0_.length_char == 1
and not res.can_be_year
and ((prev is None or
((len(prev) > 0 and prev[len(prev) - 1].typ
!= DateItemToken.DateItemType.DELIM))))):
pass
else:
res.end_token = tt
res.typ = DateItemToken.DateItemType.YEAR
res.lang = tt.morph.language
elif (tt is not None and (nt.whitespaces_after_count < 2)
and (nt.end_char - nt.begin_char) == 1):
res.end_token = tt
res.typ = DateItemToken.DateItemType.YEAR
res.lang = tt.morph.language
if (nt.previous is not None):
if (nt.previous.is_value("В", "У")
or nt.previous.is_value("К", None)
or nt.previous.is_value("ДО", None)):
tt = DateItemToken.__test_year_rus_word(nt.next0_, False)
if ((tt) is not None):
ok = False
if ((res.int_value < 100)
and (isinstance(tt, TextToken)) and
((tt.term == "ГОДА" or tt.term == "РОКИ"))):
pass
else:
ok = True
if (nt.previous.is_value("ДО", None)
and nt.next0_.is_value("Г", None)):
cou = 0
ttt = nt.previous.previous
while ttt is not None and (cou < 10):
mt = MeasureToken.try_parse(
ttt, None, False, False, False, False)
if (mt is not None
and mt.end_char > nt.end_char):
ok = False
break
ttt = ttt.previous
cou += 1
if (ok):
res.end_token = tt
res.typ = DateItemToken.DateItemType.YEAR
res.lang = tt.morph.language
res.begin_token = nt.previous
elif (((nt.previous.is_value("IN", None)
or nt.previous.is_value("SINCE", None)))
and res.can_be_year):
uu = (NumbersWithUnitToken.try_parse(
nt, None, False, False, False, False)
if nt.previous.is_value("IN", None) else None)
if (uu is not None and len(uu.units) > 0):
pass
else:
res.typ = DateItemToken.DateItemType.YEAR
res.begin_token = nt.previous
elif (nt.previous.is_value("NEL", None)
or nt.previous.is_value("DEL", None)):
if (res.can_be_year):
res.typ = DateItemToken.DateItemType.YEAR
res.lang = MorphLang.IT
res.begin_token = nt.previous
elif (nt.previous.is_value("IL", None) and res.can_be_day):
res.lang = MorphLang.IT
res.begin_token = nt.previous
t1 = res.end_token.next0_
if (t1 is not None):
if (t1.is_value("ЧАС", "ГОДИНА") or t1.is_value("HOUR", None)):
if ((((prev is not None and len(prev) == 2
and prev[0].can_be_hour)
and prev[1].typ == DateItemToken.DateItemType.DELIM
and not prev[1].is_whitespace_after)
and not prev[1].is_whitespace_after
and res.int_value >= 0) and (res.int_value < 59)):
prev[0].typ = DateItemToken.DateItemType.HOUR
res.typ = DateItemToken.DateItemType.MINUTE
res.end_token = t1
elif (res.int_value < 24):
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
res.typ = DateItemToken.DateItemType.HOUR
res.end_token = t1
elif ((res.int_value < 60)
and ((t1.is_value("МИНУТА", "ХВИЛИНА") or t1.is_value(
"МИН", None) or t.is_value("MINUTE", None)))):
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
res.typ = DateItemToken.DateItemType.MINUTE
res.end_token = t1
elif (
(res.int_value < 60) and
((t1.is_value("СЕКУНДА", None) or t1.is_value("СЕК", None)
or t1.is_value("SECOND", None)))):
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
res.typ = DateItemToken.DateItemType.SECOND
res.end_token = t1
elif ((res.int_value < 30)
and ((t1.is_value("ВЕК", "ВІК")
or t1.is_value("СТОЛЕТИЕ", "СТОЛІТТЯ")))):
res.typ = DateItemToken.DateItemType.CENTURY
res.end_token = t1
elif (res.int_value <= 4 and t1.is_value("КВАРТАЛ", None)):
res.typ = DateItemToken.DateItemType.QUARTAL
res.end_token = t1
elif (res.int_value <= 2
and ((t1.is_value("ПОЛУГОДИЕ", None)
or t1.is_value("ПІВРІЧЧЯ", None)))):
res.typ = DateItemToken.DateItemType.HALFYEAR
res.end_token = t1
return res
t0 = Utils.asObjectOrNull(t, TextToken)
if (t0 is None):
return None
txt = t0.get_source_text()
if ((txt[0] == 'I' or txt[0] == 'X' or txt[0] == 'Х')
or txt[0] == 'V'):
lat = NumberHelper.try_parse_roman(t)
if (lat is not None and lat.end_token.next0_ is not None
and lat.int_value is not None):
val = lat.int_value
tt = lat.end_token.next0_
if (tt.is_value("КВАРТАЛ", None) and val > 0 and val <= 4):
return DateItemToken._new629(
t, tt, DateItemToken.DateItemType.QUARTAL, val)
if (tt.is_value("ПОЛУГОДИЕ", "ПІВРІЧЧЯ") and val > 0
and val <= 2):
return DateItemToken._new629(
t, lat.end_token.next0_,
DateItemToken.DateItemType.HALFYEAR, val)
if (tt.is_value("ВЕК", "ВІК")
or tt.is_value("СТОЛЕТИЕ", "СТОЛІТТЯ")):
return DateItemToken._new629(
t, lat.end_token.next0_,
DateItemToken.DateItemType.CENTURY, val)
if (tt.is_value("В", None) and tt.next0_ is not None
and tt.next0_.is_char('.')):
if (prev is not None and len(prev) > 0
and prev[len(prev) - 1].typ
== DateItemToken.DateItemType.POINTER):
return DateItemToken._new629(
t, tt.next0_, DateItemToken.DateItemType.CENTURY,
val)
if (DateItemToken.__is_new_age(tt.next0_.next0_)):
return DateItemToken._new629(
t, tt.next0_, DateItemToken.DateItemType.CENTURY,
val)
if (tt.is_hiphen):
lat2 = NumberHelper.try_parse_roman(tt.next0_)
if (lat2 is not None and lat2.int_value is not None
and lat2.end_token.next0_ is not None):
if (lat2.end_token.next0_.is_value("ВЕК", "ВІК")
or lat2.end_token.next0_.is_value(
"СТОЛЕТИЕ", "СТОЛІТТЯ")):
ddd = DateItemToken.try_attach(
tt.next0_, None, False)
return DateItemToken._new634(
t, lat.end_token,
DateItemToken.DateItemType.CENTURY, val,
((ddd.new_age if ddd is not None else 0)))
if (t is not None and t.is_value("НАПРИКІНЦІ", None)):
return DateItemToken._new635(t, t,
DateItemToken.DateItemType.POINTER,
"конец")
if (t is not None and t.is_value("ДОНЕДАВНА", None)):
return DateItemToken._new635(t, t,
DateItemToken.DateItemType.POINTER,
"сегодня")
if (prev is None):
if (t is not None):
if (t.is_value("ОКОЛО", "БІЛЯ")
or t.is_value("ПРИМЕРНО", "ПРИБЛИЗНО")
or t.is_value("ABOUT", None)):
return DateItemToken._new635(
t, t, DateItemToken.DateItemType.POINTER, "около")
if (t.is_value("ОК", None) or t.is_value("OK", None)):
if (t.next0_ is not None and t.next0_.is_char('.')):
return DateItemToken._new635(
t, t.next0_, DateItemToken.DateItemType.POINTER,
"около")
return DateItemToken._new635(
t, t, DateItemToken.DateItemType.POINTER, "около")
tok = DateItemToken.M_SEASONS.try_parse(t, TerminParseAttr.NO)
if ((tok is not None and
(Utils.valToEnum(tok.termin.tag, DatePointerType))
== DatePointerType.SUMMER and t.morph.language.is_ru)
and (isinstance(t, TextToken))):
str0_ = t.term
if (str0_ != "ЛЕТОМ" and str0_ != "ЛЕТА" and str0_ != "ЛЕТО"):
tok = (None)
if (tok is not None):
return DateItemToken._new629(
t, tok.end_token, DateItemToken.DateItemType.POINTER,
Utils.valToEnum(tok.termin.tag, DatePointerType))
npt = NounPhraseHelper.try_parse(t, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
tok = DateItemToken.M_SEASONS.try_parse(npt.end_token,
TerminParseAttr.NO)
if ((tok is not None and
(Utils.valToEnum(tok.termin.tag, DatePointerType))
== DatePointerType.SUMMER and t.morph.language.is_ru)
and (isinstance(t, TextToken))):
str0_ = t.term
if (str0_ != "ЛЕТОМ" and str0_ != "ЛЕТА" and str0_ != "ЛЕТО"):
tok = (None)
if (tok is not None):
return DateItemToken._new629(
t, tok.end_token, DateItemToken.DateItemType.POINTER,
Utils.valToEnum(tok.termin.tag, DatePointerType))
typ_ = DateItemToken.DateItemType.NUMBER
if (npt.noun.is_value("КВАРТАЛ", None)):
typ_ = DateItemToken.DateItemType.QUARTAL
elif (npt.end_token.is_value("ПОЛУГОДИЕ", None)
or npt.end_token.is_value("ПІВРІЧЧЯ", None)):
typ_ = DateItemToken.DateItemType.HALFYEAR
elif (npt.end_token.is_value("НАЧАЛО", None)
or npt.end_token.is_value("ПОЧАТОК", None)):
return DateItemToken._new635(
t, npt.end_token, DateItemToken.DateItemType.POINTER,
"начало")
elif (npt.end_token.is_value("СЕРЕДИНА", None)):
return DateItemToken._new635(
t, npt.end_token, DateItemToken.DateItemType.POINTER,
"середина")
elif (npt.end_token.is_value("КОНЕЦ", None)
or npt.end_token.is_value("КІНЕЦЬ", None)
or npt.end_token.is_value("НАПРИКІНЕЦЬ", None)):
return DateItemToken._new635(
t, npt.end_token, DateItemToken.DateItemType.POINTER,
"конец")
elif (npt.end_token.is_value("ВРЕМЯ", None)
and len(npt.adjectives) > 0
and npt.end_token.previous.is_value("НАСТОЯЩЕЕ", None)):
return DateItemToken._new635(
t, npt.end_token, DateItemToken.DateItemType.POINTER,
"сегодня")
elif (npt.end_token.is_value("ЧАС", None)
and len(npt.adjectives) > 0
and npt.end_token.previous.is_value("ДАНИЙ", None)):
return DateItemToken._new635(
t, npt.end_token, DateItemToken.DateItemType.POINTER,
"сегодня")
if (typ_ != DateItemToken.DateItemType.NUMBER or detail_regime):
delta = 0
if (len(npt.adjectives) > 0):
if (npt.adjectives[0].is_value("ПОСЛЕДНИЙ", "ОСТАННІЙ")):
return DateItemToken._new629(
t0, npt.end_token, typ_,
(4 if typ_ == DateItemToken.DateItemType.QUARTAL
else 2))
if (npt.adjectives[0].is_value("ПРЕДЫДУЩИЙ", "ПОПЕРЕДНІЙ")
or npt.adjectives[0].is_value("ПРОШЛЫЙ", None)):
delta = -1
elif (npt.adjectives[0].is_value("СЛЕДУЮЩИЙ", None)
or npt.adjectives[0].is_value("ПОСЛЕДУЮЩИЙ", None)
or npt.adjectives[0].is_value("НАСТУПНИЙ", None)):
delta = 1
else:
return None
cou = 0
tt = t.previous
first_pass3073 = True
while True:
if first_pass3073: first_pass3073 = False
else: tt = tt.previous
if (not (tt is not None)): break
if (cou > 200):
break
dr = Utils.asObjectOrNull(tt.get_referent(),
DateRangeReferent)
if (dr is None):
continue
if (typ_ == DateItemToken.DateItemType.QUARTAL):
ii = dr.quarter_number
if (ii < 1):
continue
ii += delta
if ((ii < 1) or ii > 4):
continue
return DateItemToken._new629(t0, npt.end_token, typ_,
ii)
if (typ_ == DateItemToken.DateItemType.HALFYEAR):
ii = dr.halfyear_number
if (ii < 1):
continue
ii += delta
if ((ii < 1) or ii > 2):
continue
return DateItemToken._new629(t0, npt.end_token, typ_,
ii)
term = t0.term
if (not str.isalnum(term[0])):
if (t0.is_char_of(".\\/:") or t0.is_hiphen):
return DateItemToken._new635(t0, t0,
DateItemToken.DateItemType.DELIM,
term)
elif (t0.is_char(',')):
return DateItemToken._new635(t0, t0,
DateItemToken.DateItemType.DELIM,
term)
else:
return None
if (term == "O" or term == "О"):
if ((isinstance(t.next0_, NumberToken))
and not t.is_whitespace_after
and len(t.next0_.value) == 1):
return DateItemToken._new629(t, t.next0_,
DateItemToken.DateItemType.NUMBER,
t.next0_.int_value)
if (str.isalpha(term[0])):
inf = DateItemToken.M_MONTHES.try_parse(t, TerminParseAttr.NO)
if (inf is not None and inf.termin.tag is None):
inf = DateItemToken.M_MONTHES.try_parse(
inf.end_token.next0_, TerminParseAttr.NO)
if (inf is not None and (isinstance(inf.termin.tag, int))):
return DateItemToken._new653(inf.begin_token, inf.end_token,
DateItemToken.DateItemType.MONTH,
inf.termin.tag, inf.termin.lang)
return None
def attach_first(self, p : 'InstrumentParticipantReferent', min_char : int, max_char : int) -> 'ReferentToken':
t = None
tt0 = self.begin_token
refs = list()
t = tt0.previous
first_pass3287 = True
while True:
if first_pass3287: first_pass3287 = False
else: t = t.previous
if (not (t is not None and t.begin_char >= min_char)): break
if (t.is_newline_after):
if (t.newlines_after_count > 1):
break
if (isinstance(t.next0_, NumberToken)):
break
tt = ParticipantToken.__try_attach_contract_ground(t, p, False)
if (tt is not None):
continue
r = t.get_referent()
if (((((isinstance(r, OrganizationReferent)) or (isinstance(r, PhoneReferent)) or (isinstance(r, PersonReferent))) or (isinstance(r, PersonPropertyReferent)) or (isinstance(r, AddressReferent))) or (isinstance(r, UriReferent)) or (isinstance(r, PersonIdentityReferent))) or (isinstance(r, BankDataReferent))):
if (not r in refs):
refs.insert(0, r)
tt0 = t
if (len(refs) > 0):
for r in refs:
if (r != refs[0] and (isinstance(refs[0], OrganizationReferent)) and (((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent))))):
p.add_slot(InstrumentParticipantReferent.ATTR_DELEGATE, r, False, 0)
else:
p.add_slot(InstrumentParticipantReferent.ATTR_REF, r, False, 0)
rt = ReferentToken(p, tt0, self.end_token)
t = self.end_token.next0_
if (BracketHelper.is_bracket(t, False)):
t = t.next0_
if (t is not None and t.is_char(',')):
t = t.next0_
first_pass3288 = True
while True:
if first_pass3288: first_pass3288 = False
else: t = t.next0_
if (not (t is not None and ((max_char == 0 or t.begin_char <= max_char)))): break
if (t.is_value("СТОРОНА", None)):
break
r = t.get_referent()
if (((((isinstance(r, OrganizationReferent)) or (isinstance(r, PhoneReferent)) or (isinstance(r, PersonReferent))) or (isinstance(r, PersonPropertyReferent)) or (isinstance(r, AddressReferent))) or (isinstance(r, UriReferent)) or (isinstance(r, PersonIdentityReferent))) or (isinstance(r, BankDataReferent))):
if ((((isinstance(r, PersonPropertyReferent)) and t.next0_ is not None and t.next0_.is_comma) and (isinstance(t.next0_.next0_, ReferentToken)) and (isinstance(t.next0_.next0_.get_referent(), PersonReferent))) and not t.next0_.is_newline_after):
pe = Utils.asObjectOrNull(t.next0_.next0_.get_referent(), PersonReferent)
pe.add_slot(PersonReferent.ATTR_ATTR, r, False, 0)
r = (pe)
t = t.next0_.next0_
is_delegate = False
if (t.previous.is_value("ЛИЦО", None) or t.previous.is_value("ИМЯ", None)):
is_delegate = True
if (t.previous.is_value("КОТОРЫЙ", None) and t.previous.previous is not None and ((t.previous.previous.is_value("ИМЯ", None) or t.previous.previous.is_value("ЛИЦО", None)))):
is_delegate = True
p.add_slot((InstrumentParticipantReferent.ATTR_DELEGATE if (((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent)))) and is_delegate else InstrumentParticipantReferent.ATTR_REF), r, False, 0)
rt.end_token = t
continue
tt = ParticipantToken.__try_attach_contract_ground(t, p, False)
if (tt is not None):
rt.end_token = tt
t = rt.end_token
if (rt.begin_char == tt.begin_char):
rt.begin_token = tt
continue
if (t.is_value("В", None) and t.next0_ is not None and t.next0_.is_value("ЛИЦО", None)):
t = t.next0_
continue
if (t.is_value("ОТ", None) and t.next0_ is not None and t.next0_.is_value("ИМЯ", None)):
t = t.next0_
continue
if (t.is_value("ПО", None) and t.next0_ is not None and t.next0_.is_value("ПОРУЧЕНИЕ", None)):
t = t.next0_
continue
if (t.is_newline_before):
break
if (t.get_morph_class_in_dictionary() == MorphClass.VERB):
if ((not t.is_value("УДОСТОВЕРЯТЬ", None) and not t.is_value("ПРОЖИВАТЬ", None) and not t.is_value("ЗАРЕГИСТРИРОВАТЬ", None)) and not t.is_value("ДЕЙСТВОВАТЬ", None)):
break
if (t.is_and and t.previous is not None and t.previous.is_comma):
break
if (t.is_and and t.next0_.get_referent() is not None):
if (isinstance(t.next0_.get_referent(), OrganizationReferent)):
break
pe = Utils.asObjectOrNull(t.next0_.get_referent(), PersonReferent)
if (pe is not None):
has_ip = False
for s in pe.slots:
if (s.type_name == PersonReferent.ATTR_ATTR):
if (str(s.value).startswith("индивидуальный предприниматель")):
has_ip = True
break
if (has_ip):
break
t = rt.begin_token
while t is not None and t.end_char <= rt.end_char:
tt = ParticipantToken.__try_attach_contract_ground(t, p, True)
if (tt is not None):
if (tt.end_char > rt.end_char):
rt.end_token = tt
t = tt
t = t.next0_
return rt
def tryAttachOrg(t : 'Token', can_be_cyr : bool=False) -> 'ReferentToken':
from pullenti.ner.org.internal.OrgItemNameToken import OrgItemNameToken
if (t is None):
return None
br = False
if (t.isChar('(') and t.next0_ is not None):
t = t.next0_
br = True
if (isinstance(t, NumberToken)):
if ((t).typ == NumberSpellingType.WORDS and t.morph.class0_.is_adjective and t.chars.is_capital_upper):
pass
else:
return None
else:
if (t.chars.is_all_lower):
return None
if ((t.length_char < 3) and not t.chars.is_letter):
return None
if (not t.chars.is_latin_letter):
if (not can_be_cyr or not t.chars.is_cyrillic_letter):
return None
t0 = t
t1 = t0
nam_wo = 0
tok = None
geo_ = None
add_typ = None
first_pass3043 = True
while True:
if first_pass3043: first_pass3043 = False
else: t = t.next0_
if (not (t is not None)): break
if (t != t0 and t.whitespaces_before_count > 1):
break
if (t.isChar(')')):
break
if (t.isChar('(') and t.next0_ is not None):
if ((isinstance(t.next0_.getReferent(), GeoReferent)) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')):
geo_ = (Utils.asObjectOrNull(t.next0_.getReferent(), GeoReferent))
t = t.next0_.next0_
continue
typ = OrgItemTypeToken.tryAttach(t.next0_, True, None)
if ((typ is not None and typ.end_token.next0_ is not None and typ.end_token.next0_.isChar(')')) and typ.chars.is_latin_letter):
add_typ = typ
t = typ.end_token.next0_
continue
if (((isinstance(t.next0_, TextToken)) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')) and t.next0_.chars.is_capital_upper):
t = t.next0_.next0_
t1 = t
continue
break
tok = OrgItemEngItem.tryAttach(t, can_be_cyr)
if (tok is None and t.isCharOf(".,") and t.next0_ is not None):
tok = OrgItemEngItem.tryAttach(t.next0_, can_be_cyr)
if (tok is None and t.next0_.isCharOf(",.")):
tok = OrgItemEngItem.tryAttach(t.next0_.next0_, can_be_cyr)
if (tok is not None):
if (tok.length_char == 1 and t0.chars.is_cyrillic_letter):
return None
break
if (t.is_hiphen and not t.is_whitespace_after and not t.is_whitespace_before):
continue
if (t.isCharOf("&+") or t.is_and):
continue
if (t.isChar('.')):
if (t.previous is not None and t.previous.length_char == 1):
continue
elif (MiscHelper.canBeStartOfSentence(t.next0_)):
break
if (not t.chars.is_latin_letter):
if (not can_be_cyr or not t.chars.is_cyrillic_letter):
break
if (t.chars.is_all_lower):
if (t.morph.class0_.is_preposition or t.morph.class0_.is_conjunction):
continue
if (br):
continue
break
mc = t.getMorphClassInDictionary()
if (mc.is_verb):
if (t.next0_ is not None and t.next0_.morph.class0_.is_preposition):
break
if (t.next0_ is not None and t.next0_.isValue("OF", None)):
break
if (isinstance(t, TextToken)):
nam_wo += 1
t1 = t
if (tok is None):
return None
if (t0 == tok.begin_token):
br2 = BracketHelper.tryParse(tok.end_token.next0_, BracketParseAttr.NO, 100)
if (br2 is not None):
org1 = OrganizationReferent()
if (tok.short_value is not None):
org1.addTypeStr(tok.short_value)
org1.addTypeStr(tok.full_value)
nam1 = MiscHelper.getTextValue(br2.begin_token, br2.end_token, GetTextAttr.NO)
if (nam1 is not None):
org1.addName(nam1, True, None)
return ReferentToken(org1, t0, br2.end_token)
return None
org0_ = OrganizationReferent()
te = tok.end_token
if (tok.is_bank):
t1 = tok.end_token
if (tok.full_value == "company" and (tok.whitespaces_after_count < 3)):
tok1 = OrgItemEngItem.tryAttach(tok.end_token.next0_, can_be_cyr)
if (tok1 is not None):
t1 = tok.end_token
tok = tok1
te = tok.end_token
if (tok.full_value == "company"):
if (nam_wo == 0):
return None
nam = MiscHelper.getTextValue(t0, t1, GetTextAttr.IGNOREARTICLES)
if (nam == "STOCK" and tok.full_value == "company"):
return None
alt_nam = None
if (Utils.isNullOrEmpty(nam)):
return None
if (nam.find('(') > 0):
i1 = nam.find('(')
i2 = nam.find(')')
if (i1 < i2):
alt_nam = nam
tai = None
if ((i2 + 1) < len(nam)):
tai = nam[i2:].strip()
nam = nam[0:0+i1].strip()
if (tai is not None):
nam = "{0} {1}".format(nam, tai)
if (tok.is_bank):
org0_.addTypeStr(("bank" if tok.kit.base_language.is_en else "банк"))
org0_.addProfile(OrgProfile.FINANCE)
if ((t1.next0_ is not None and t1.next0_.isValue("OF", None) and t1.next0_.next0_ is not None) and t1.next0_.next0_.chars.is_latin_letter):
nam0 = OrgItemNameToken.tryAttach(t1.next0_, None, False, False)
if (nam0 is not None):
te = nam0.end_token
else:
te = t1.next0_.next0_
nam = MiscHelper.getTextValue(t0, te, GetTextAttr.NO)
if (isinstance(te.getReferent(), GeoReferent)):
org0_._addGeoObject(Utils.asObjectOrNull(te.getReferent(), GeoReferent))
elif (t0 == t1):
return None
else:
if (tok.short_value is not None):
org0_.addTypeStr(tok.short_value)
org0_.addTypeStr(tok.full_value)
if (Utils.isNullOrEmpty(nam)):
return None
org0_.addName(nam, True, None)
if (alt_nam is not None):
org0_.addName(alt_nam, True, None)
res = ReferentToken(org0_, t0, te)
t = te
while t.next0_ is not None:
if (t.next0_.isCharOf(",.")):
t = t.next0_
else:
break
if (t.whitespaces_after_count < 2):
tok = OrgItemEngItem.tryAttach(t.next0_, can_be_cyr)
if (tok is not None):
if (tok.short_value is not None):
org0_.addTypeStr(tok.short_value)
org0_.addTypeStr(tok.full_value)
res.end_token = tok.end_token
if (geo_ is not None):
org0_._addGeoObject(geo_)
if (add_typ is not None):
org0_.addType(add_typ, False)
if (not br):
return res
t = res.end_token
if (t.next0_ is None or t.next0_.isChar(')')):
res.end_token = t.next0_
else:
return None
return res
def __try_attach_contract_ground(t : 'Token', ip : 'InstrumentParticipantReferent', can_be_passport : bool=False) -> 'Token':
ok = False
first_pass3289 = True
while True:
if first_pass3289: first_pass3289 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_char(',') or t.morph.class0_.is_preposition):
continue
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
continue
if (t.is_value("ОСНОВАНИЕ", None) or t.is_value("ДЕЙСТВОВАТЬ", None) or t.is_value("ДЕЙСТВУЮЩИЙ", None)):
ok = True
if (t.next0_ is not None and t.next0_.is_char('(')):
br = BracketHelper.try_parse(t.next0_, BracketParseAttr.NO, 100)
if (br is not None and (br.length_char < 10)):
t = br.end_token
continue
dr = Utils.asObjectOrNull(t.get_referent(), DecreeReferent)
if (dr is not None):
ip.ground = dr
return t
pir = Utils.asObjectOrNull(t.get_referent(), PersonIdentityReferent)
if (pir is not None and can_be_passport):
if (pir.typ is not None and not "паспорт" in pir.typ):
ip.ground = pir
return t
if (t.is_value("УСТАВ", None)):
ip.ground = t.get_normal_case_text(MorphClass.NOUN, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
return t
if (t.is_value("ДОВЕРЕННОСТЬ", None)):
dts = DecreeToken.try_attach_list(t.next0_, None, 10, False)
if (dts is None):
has_spec = False
ttt = t.next0_
first_pass3290 = True
while True:
if first_pass3290: first_pass3290 = False
else: ttt = ttt.next0_
if (not (ttt is not None and ((ttt.end_char - t.end_char) < 200))): break
if (ttt.is_comma):
continue
if (ttt.is_value("УДОСТОВЕРИТЬ", None) or ttt.is_value("УДОСТОВЕРЯТЬ", None)):
has_spec = True
continue
dt = DecreeToken.try_attach(ttt, None, False)
if (dt is not None):
if (dt.typ == DecreeToken.ItemType.DATE or dt.typ == DecreeToken.ItemType.NUMBER):
dts = DecreeToken.try_attach_list(ttt, None, 10, False)
break
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
if (npt.end_token.is_value("НОТАРИУС", None)):
ttt = npt.end_token
has_spec = True
continue
if (ttt.get_referent() is not None):
if (has_spec):
continue
break
if (dts is not None and len(dts) > 0):
t0 = t
dr = DecreeReferent()
dr.typ = "ДОВЕРЕННОСТЬ"
for d in dts:
if (d.typ == DecreeToken.ItemType.DATE):
dr._add_date(d)
t = d.end_token
elif (d.typ == DecreeToken.ItemType.NUMBER):
dr._add_number(d)
t = d.end_token
else:
break
ad = t.kit.get_analyzer_data_by_analyzer_name(InstrumentAnalyzer.ANALYZER_NAME)
ip.ground = ad.register_referent(dr)
rt = ReferentToken(Utils.asObjectOrNull(ip.ground, Referent), t0, t)
t.kit.embed_token(rt)
return rt
ip.ground = "ДОВЕРЕННОСТЬ"
return t
break
return None
def try_attach_territory(
li: typing.List['TerrItemToken'],
ad: 'AnalyzerData',
attach_always: bool = False,
cits: typing.List['CityItemToken'] = None,
exists: typing.List['GeoReferent'] = None) -> 'ReferentToken':
if (li is None or len(li) == 0):
return None
ex_obj = None
new_name = None
adj_list = list()
noun = None
add_noun = None
rt = TerrAttachHelper.__try_attach_moscowao(li, ad)
if (rt is not None):
return rt
if (li[0].termin_item is not None
and li[0].termin_item.canonic_text == "ТЕРРИТОРИЯ"):
res2 = TerrAttachHelper.__try_attach_pure_terr(li, ad)
return res2
if (len(li) == 2):
if (li[0].rzd is not None and li[1].rzd_dir is not None):
rzd = GeoReferent()
rzd._add_name(li[1].rzd_dir)
rzd._add_typ_ter(li[0].kit.base_language)
rzd.add_slot(GeoReferent.ATTR_REF, li[0].rzd.referent, False,
0)
rzd.add_ext_referent(li[0].rzd)
return ReferentToken(rzd, li[0].begin_token, li[1].end_token)
if (li[1].rzd is not None and li[0].rzd_dir is not None):
rzd = GeoReferent()
rzd._add_name(li[0].rzd_dir)
rzd._add_typ_ter(li[0].kit.base_language)
rzd.add_slot(GeoReferent.ATTR_REF, li[1].rzd.referent, False,
0)
rzd.add_ext_referent(li[1].rzd)
return ReferentToken(rzd, li[0].begin_token, li[1].end_token)
can_be_city_before = False
adj_terr_before = False
if (cits is not None):
if (cits[0].typ == CityItemToken.ItemType.CITY):
can_be_city_before = True
elif (cits[0].typ == CityItemToken.ItemType.NOUN
and len(cits) > 1):
can_be_city_before = True
k = 0
k = 0
while k < len(li):
if (li[k].onto_item is not None):
if (ex_obj is not None or new_name is not None):
break
if (noun is not None):
if (k == 1):
if (noun.termin_item.canonic_text == "РАЙОН"
or noun.termin_item.canonic_text == "ОБЛАСТЬ"
or noun.termin_item.canonic_text == "СОЮЗ"):
if (isinstance(li[k].onto_item.referent,
GeoReferent)):
if (li[k].onto_item.referent.is_state):
break
ok = False
tt = li[k].end_token.next0_
if (tt is None):
ok = True
elif (tt.is_char_of(",.")):
ok = True
if (not ok):
ok = MiscLocationHelper.check_geo_object_before(
li[0].begin_token)
if (not ok):
adr = AddressItemToken.try_parse(
tt, None, False, False, None)
if (adr is not None):
if (adr.typ ==
AddressItemToken.ItemType.STREET):
ok = True
if (not ok):
break
if (li[k].onto_item is not None):
if (noun.begin_token.is_value("МО", None)
or noun.begin_token.is_value("ЛО", None)):
return None
ex_obj = li[k]
elif (li[k].termin_item is not None):
if (noun is not None):
break
if (li[k].termin_item.is_always_prefix and k > 0):
break
if (k > 0 and li[k].is_doubt):
if (li[k].begin_token == li[k].end_token
and li[k].begin_token.is_value("ЗАО", None)):
break
if (li[k].termin_item.is_adjective
or li[k].is_geo_in_dictionary):
adj_list.append(li[k])
else:
if (ex_obj is not None):
geo_ = Utils.asObjectOrNull(ex_obj.onto_item.referent,
GeoReferent)
if (geo_ is None):
break
if (ex_obj.is_adjective and
((li[k].termin_item.canonic_text == "СОЮЗ" or
li[k].termin_item.canonic_text == "ФЕДЕРАЦИЯ"))):
str0_ = str(ex_obj.onto_item)
if (not li[k].termin_item.canonic_text in str0_):
return None
if (li[k].termin_item.canonic_text == "РАЙОН"
or li[k].termin_item.canonic_text == "ОКРУГ"
or li[k].termin_item.canonic_text == "КРАЙ"):
tmp = io.StringIO()
for s in geo_.slots:
if (s.type_name == GeoReferent.ATTR_TYPE):
print("{0};".format(s.value),
end="",
file=tmp,
flush=True)
if (not li[k].termin_item.canonic_text
in Utils.toStringStringIO(tmp).upper()):
if (k != 1 or new_name is not None):
break
new_name = li[0]
new_name.is_adjective = True
new_name.onto_item = (None)
ex_obj = (None)
noun = li[k]
if (k == 0):
tt = TerrItemToken.try_parse(
li[k].begin_token.previous, None, True, False,
None)
if (tt is not None and tt.morph.class0_.is_adjective):
adj_terr_before = True
else:
if (ex_obj is not None):
break
if (new_name is not None):
break
new_name = li[k]
k += 1
name = None
alt_name = None
full_name = None
morph_ = None
if (ex_obj is not None):
if (ex_obj.is_adjective and not ex_obj.morph.language.is_en
and noun is None):
if (attach_always and ex_obj.end_token.next0_ is not None):
npt = NounPhraseHelper.try_parse(ex_obj.begin_token,
NounPhraseParseAttr.NO, 0,
None)
if (ex_obj.end_token.next0_.is_comma_and):
pass
elif (npt is None):
pass
else:
str0_ = StreetItemToken.try_parse(
ex_obj.end_token.next0_, None, False, None, False)
if (str0_ is not None):
if (str0_.typ == StreetItemType.NOUN
and str0_.end_token == npt.end_token):
return None
else:
cit = CityItemToken.try_parse(ex_obj.end_token.next0_,
None, False, None)
if (cit is not None
and ((cit.typ == CityItemToken.ItemType.NOUN
or cit.typ == CityItemToken.ItemType.CITY))):
npt = NounPhraseHelper.try_parse(
ex_obj.begin_token, NounPhraseParseAttr.NO, 0,
None)
if (npt is not None
and npt.end_token == cit.end_token):
pass
else:
return None
elif (ex_obj.begin_token.is_value("ПОДНЕБЕСНЫЙ", None)):
pass
else:
return None
if (noun is None and ex_obj.can_be_city):
cit0 = CityItemToken.try_parse_back(
ex_obj.begin_token.previous)
if (cit0 is not None
and cit0.typ != CityItemToken.ItemType.PROPERNAME):
return None
if (ex_obj.is_doubt and noun is None):
ok2 = False
if (TerrAttachHelper.__can_be_geo_after(
ex_obj.end_token.next0_)):
ok2 = True
elif (not ex_obj.can_be_surname and not ex_obj.can_be_city):
if ((ex_obj.end_token.next0_ is not None
and ex_obj.end_token.next0_.is_char(')')
and ex_obj.begin_token.previous is not None)
and ex_obj.begin_token.previous.is_char('(')):
ok2 = True
elif (ex_obj.chars.is_latin_letter
and ex_obj.begin_token.previous is not None):
if (ex_obj.begin_token.previous.is_value("IN", None)):
ok2 = True
elif (ex_obj.begin_token.previous.is_value(
"THE", None) and
ex_obj.begin_token.previous.previous is not None
and
ex_obj.begin_token.previous.previous.is_value(
"IN", None)):
ok2 = True
if (not ok2):
cit0 = CityItemToken.try_parse_back(
ex_obj.begin_token.previous)
if (cit0 is not None
and cit0.typ != CityItemToken.ItemType.PROPERNAME):
pass
elif (MiscLocationHelper.check_geo_object_before(
ex_obj.begin_token.previous)):
pass
else:
return None
name = ex_obj.onto_item.canonic_text
morph_ = ex_obj.morph
elif (new_name is not None):
if (noun is None):
return None
j = 1
while j < k:
if (li[j].is_newline_before and not li[0].is_newline_before):
if (BracketHelper.can_be_start_of_sequence(
li[j].begin_token, False, False)):
pass
else:
return None
j += 1
morph_ = noun.morph
if (new_name.is_adjective):
if (noun.termin_item.acronym == "АО"):
if (noun.begin_token != noun.end_token):
return None
if (new_name.morph.gender != MorphGender.FEMINIE):
return None
geo_before = None
tt0 = li[0].begin_token.previous
if (tt0 is not None and tt0.is_comma_and):
tt0 = tt0.previous
if (not li[0].is_newline_before and tt0 is not None):
geo_before = (Utils.asObjectOrNull(tt0.get_referent(),
GeoReferent))
if (Utils.indexOfList(li, noun, 0) < Utils.indexOfList(
li, new_name, 0)):
if (noun.termin_item.is_state):
return None
if (new_name.can_be_surname and geo_before is None):
if (((noun.morph.case_)
& new_name.morph.case_).is_undefined):
return None
if (MiscHelper.is_exists_in_dictionary(
new_name.begin_token, new_name.end_token,
(MorphClass.ADJECTIVE) | MorphClass.PRONOUN
| MorphClass.VERB)):
if (noun.begin_token != new_name.begin_token):
if (geo_before is None):
if (len(li) == 2 and
TerrAttachHelper.__can_be_geo_after(
li[1].end_token.next0_)):
pass
elif (len(li) == 3
and li[2].termin_item is not None
and TerrAttachHelper.__can_be_geo_after(
li[2].end_token.next0_)):
pass
elif (new_name.is_geo_in_dictionary):
pass
elif (new_name.end_token.is_newline_after):
pass
else:
return None
npt = NounPhraseHelper.try_parse(
new_name.end_token, NounPhraseParseAttr.PARSEPRONOUNS,
0, None)
if (npt is not None
and npt.end_token != new_name.end_token):
if (len(li) >= 3 and li[2].termin_item is not None
and npt.end_token == li[2].end_token):
add_noun = li[2]
else:
return None
rtp = new_name.kit.process_referent(
"PERSON", new_name.begin_token)
if (rtp is not None):
return None
name = ProperNameHelper.get_name_ex(
new_name.begin_token, new_name.end_token,
MorphClass.ADJECTIVE, MorphCase.UNDEFINED,
noun.termin_item.gender, False, False)
else:
ok = False
if (((k + 1) < len(li)) and li[k].termin_item is None
and li[k + 1].termin_item is not None):
ok = True
elif ((k < len(li)) and li[k].onto_item is not None):
ok = True
elif (k == len(li) and not new_name.is_adj_in_dictionary):
ok = True
elif (MiscLocationHelper.check_geo_object_before(
li[0].begin_token) or can_be_city_before):
ok = True
elif (MiscLocationHelper.check_geo_object_after(
li[k - 1].end_token, False)):
ok = True
elif (len(li) == 3 and k == 2):
cit = CityItemToken.try_parse(li[2].begin_token, None,
False, None)
if (cit is not None):
if (cit.typ == CityItemToken.ItemType.CITY
or cit.typ == CityItemToken.ItemType.NOUN):
ok = True
elif (len(li) == 2):
ok = TerrAttachHelper.__can_be_geo_after(
li[len(li) - 1].end_token.next0_)
if (not ok and not li[0].is_newline_before
and not li[0].chars.is_all_lower):
rt00 = li[0].kit.process_referent(
"PERSONPROPERTY", li[0].begin_token.previous)
if (rt00 is not None):
ok = True
if (noun.termin_item is not None
and noun.termin_item.is_strong
and new_name.is_adjective):
ok = True
if (noun.is_doubt and len(adj_list) == 0
and geo_before is None):
return None
name = ProperNameHelper.get_name_ex(
new_name.begin_token, new_name.end_token,
MorphClass.ADJECTIVE, MorphCase.UNDEFINED,
noun.termin_item.gender, False, False)
if (not ok and not attach_always):
if (MiscHelper.is_exists_in_dictionary(
new_name.begin_token, new_name.end_token,
(MorphClass.ADJECTIVE) | MorphClass.PRONOUN
| MorphClass.VERB)):
if (exists is not None):
for e0_ in exists:
if (e0_.find_slot(GeoReferent.ATTR_NAME,
name, True) is not None):
ok = True
break
if (not ok):
return None
full_name = "{0} {1}".format(
ProperNameHelper.get_name_ex(li[0].begin_token,
noun.begin_token.previous,
MorphClass.ADJECTIVE,
MorphCase.UNDEFINED,
noun.termin_item.gender,
False, False),
noun.termin_item.canonic_text)
else:
if (not attach_always or
((noun.termin_item is not None
and noun.termin_item.canonic_text == "ФЕДЕРАЦИЯ"))):
is_latin = noun.chars.is_latin_letter and new_name.chars.is_latin_letter
if (Utils.indexOfList(li, noun, 0) > Utils.indexOfList(
li, new_name, 0)):
if (not is_latin):
return None
if (not new_name.is_district_name
and not BracketHelper.can_be_start_of_sequence(
new_name.begin_token, False, False)):
if (len(adj_list) == 0
and MiscHelper.is_exists_in_dictionary(
new_name.begin_token, new_name.end_token,
(MorphClass.NOUN) | MorphClass.PRONOUN)):
if (len(li) == 2 and noun.is_city_region
and (noun.whitespaces_after_count < 2)):
pass
else:
return None
if (not is_latin):
if ((noun.termin_item.is_region
and not attach_always and
((not adj_terr_before or new_name.is_doubt)))
and not noun.is_city_region and
not noun.termin_item.is_specific_prefix):
if (not MiscLocationHelper.
check_geo_object_before(
noun.begin_token)):
if (not noun.is_doubt and noun.begin_token
!= noun.end_token):
pass
elif ((noun.termin_item.is_always_prefix
and len(li) == 2 and li[0] == noun)
and li[1] == new_name):
pass
else:
return None
if (noun.is_doubt and len(adj_list) == 0):
if (noun.termin_item.acronym == "МО"
or noun.termin_item.acronym == "ЛО"):
if (k == (len(li) - 1)
and li[k].termin_item is not None):
add_noun = li[k]
k += 1
elif (len(li) == 2 and noun == li[0]
and str(new_name).endswith("совет")):
pass
else:
return None
else:
return None
pers = new_name.kit.process_referent(
"PERSON", new_name.begin_token)
if (pers is not None):
return None
name = MiscHelper.get_text_value(new_name.begin_token,
new_name.end_token,
GetTextAttr.NO)
if (new_name.begin_token != new_name.end_token):
ttt = new_name.begin_token.next0_
while ttt is not None and ttt.end_char <= new_name.end_char:
if (ttt.chars.is_letter):
ty = TerrItemToken.try_parse(
ttt, None, False, False, None)
if ((ty is not None and ty.termin_item is not None
and noun is not None)
and ((noun.termin_item.canonic_text
in ty.termin_item.canonic_text
or ty.termin_item.canonic_text
in noun.termin_item.canonic_text))):
name = MiscHelper.get_text_value(
new_name.begin_token, ttt.previous,
GetTextAttr.NO)
break
ttt = ttt.next0_
if (len(adj_list) > 0):
npt = NounPhraseHelper.try_parse(adj_list[0].begin_token,
NounPhraseParseAttr.NO, 0,
None)
if (npt is not None and npt.end_token == noun.end_token):
alt_name = "{0} {1}".format(
npt.get_normal_case_text(None,
MorphNumber.UNDEFINED,
MorphGender.UNDEFINED,
False), name)
else:
if ((len(li) == 1 and noun is not None
and noun.end_token.next0_ is not None) and (isinstance(
noun.end_token.next0_.get_referent(), GeoReferent))):
g = Utils.asObjectOrNull(noun.end_token.next0_.get_referent(),
GeoReferent)
if (noun.termin_item is not None):
tyy = noun.termin_item.canonic_text.lower()
ooo = False
if (g.find_slot(GeoReferent.ATTR_TYPE, tyy, True)
is not None):
ooo = True
elif (tyy.endswith("район") and g.find_slot(
GeoReferent.ATTR_TYPE, "район", True) is not None):
ooo = True
if (ooo):
return ReferentToken._new734(g, noun.begin_token,
noun.end_token.next0_,
noun.begin_token.morph)
if ((len(li) == 1 and noun == li[0]
and li[0].termin_item is not None)
and TerrItemToken.try_parse(li[0].end_token.next0_, None,
True, False, None) is None and
TerrItemToken.try_parse(li[0].begin_token.previous, None,
True, False, None) is None):
if (li[0].morph.number == MorphNumber.PLURAL):
return None
cou = 0
str0_ = li[0].termin_item.canonic_text.lower()
tt = li[0].begin_token.previous
first_pass3158 = True
while True:
if first_pass3158: first_pass3158 = False
else: tt = tt.previous
if (not (tt is not None)): break
if (tt.is_newline_after):
cou += 10
else:
cou += 1
if (cou > 500):
break
g = Utils.asObjectOrNull(tt.get_referent(), GeoReferent)
if (g is None):
continue
ok = True
cou = 0
tt = li[0].end_token.next0_
first_pass3159 = True
while True:
if first_pass3159: first_pass3159 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
cou += 10
else:
cou += 1
if (cou > 500):
break
tee = TerrItemToken.try_parse(tt, None, True, False,
None)
if (tee is None):
continue
ok = False
break
if (ok):
ii = 0
while g is not None and (ii < 3):
if (g.find_slot(GeoReferent.ATTR_TYPE, str0_, True)
is not None):
return ReferentToken._new734(
g, li[0].begin_token, li[0].end_token,
noun.begin_token.morph)
g = g.higher
ii += 1
break
return None
ter = None
if (ex_obj is not None and (isinstance(ex_obj.tag, GeoReferent))):
ter = (Utils.asObjectOrNull(ex_obj.tag, GeoReferent))
else:
ter = GeoReferent()
if (ex_obj is not None):
geo_ = Utils.asObjectOrNull(ex_obj.onto_item.referent,
GeoReferent)
if (geo_ is not None and not geo_.is_city):
ter._merge_slots2(geo_, li[0].kit.base_language)
else:
ter._add_name(name)
if (noun is None and ex_obj.can_be_city):
ter._add_typ_city(li[0].kit.base_language)
else:
pass
elif (new_name is not None):
ter._add_name(name)
if (alt_name is not None):
ter._add_name(alt_name)
if (noun is not None):
if (noun.termin_item.canonic_text == "АО"):
ter._add_typ(
("АВТОНОМНИЙ ОКРУГ" if li[0].kit.base_language.is_ua
else "АВТОНОМНЫЙ ОКРУГ"))
elif (noun.termin_item.canonic_text == "МУНИЦИПАЛЬНОЕ СОБРАНИЕ"
or noun.termin_item.canonic_text
== "МУНІЦИПАЛЬНЕ ЗБОРИ"):
ter._add_typ(("МУНІЦИПАЛЬНЕ УТВОРЕННЯ"
if li[0].kit.base_language.is_ua else
"МУНИЦИПАЛЬНОЕ ОБРАЗОВАНИЕ"))
elif (noun.termin_item.acronym == "МО"
and add_noun is not None):
ter._add_typ(add_noun.termin_item.canonic_text)
else:
if (noun.termin_item.canonic_text == "СОЮЗ"
and ex_obj is not None
and ex_obj.end_char > noun.end_char):
return ReferentToken._new734(ter, ex_obj.begin_token,
ex_obj.end_token,
ex_obj.morph)
ter._add_typ(noun.termin_item.canonic_text)
if (noun.termin_item.is_region and ter.is_state):
ter._add_typ_reg(li[0].kit.base_language)
if (ter.is_state and ter.is_region):
for a in adj_list:
if (a.termin_item.is_region):
ter._add_typ_reg(li[0].kit.base_language)
break
if (ter.is_state):
if (full_name is not None):
ter._add_name(full_name)
res = ReferentToken(ter, li[0].begin_token, li[k - 1].end_token)
if (noun is not None and noun.morph.class0_.is_noun):
res.morph = noun.morph
else:
res.morph = MorphCollection()
ii = 0
while ii < k:
for v in li[ii].morph.items:
bi = MorphBaseInfo()
bi.copy_from(v)
if (noun is not None):
if (bi.class0_.is_adjective):
bi.class0_ = MorphClass.NOUN
res.morph.add_item(bi)
ii += 1
if (li[0].termin_item is not None
and li[0].termin_item.is_specific_prefix):
res.begin_token = li[0].end_token.next0_
if (add_noun is not None and add_noun.end_char > res.end_char):
res.end_token = add_noun.end_token
if ((isinstance(res.begin_token.previous, TextToken))
and (res.whitespaces_before_count < 2)):
tt = Utils.asObjectOrNull(res.begin_token.previous, TextToken)
if (tt.term == "АР"):
for ty in ter.typs:
if ("республика" in ty or "республіка" in ty):
res.begin_token = tt
break
return res
def try_attach(t : 'Token', p1 : 'InstrumentParticipantReferent'=None, p2 : 'InstrumentParticipantReferent'=None, is_contract : bool=False) -> 'ParticipantToken':
if (t is None):
return None
tt = t
br = False
if (p1 is None and p2 is None and is_contract):
r1 = t.get_referent()
if ((r1 is not None and t.next0_ is not None and t.next0_.is_comma_and) and (isinstance(t.next0_.next0_, ReferentToken))):
r2 = t.next0_.next0_.get_referent()
if (r1.type_name == r2.type_name):
ttt = t.next0_.next0_.next0_
refs = list()
refs.append(r1)
refs.append(r2)
first_pass3282 = True
while True:
if first_pass3282: first_pass3282 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if ((ttt.is_comma_and and ttt.next0_ is not None and ttt.next0_.get_referent() is not None) and ttt.next0_.get_referent().type_name == r1.type_name):
ttt = ttt.next0_
if (not ttt.get_referent() in refs):
refs.append(ttt.get_referent())
continue
break
first_pass3283 = True
while True:
if first_pass3283: first_pass3283 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.is_comma or ttt.morph.class0_.is_preposition):
continue
if ((ttt.is_value("ИМЕНОВАТЬ", None) or ttt.is_value("ДАЛЬНЕЙШИЙ", None) or ttt.is_value("ДАЛЕЕ", None)) or ttt.is_value("ТЕКСТ", None)):
continue
if (ttt.is_value("ДОГОВАРИВАТЬСЯ", None)):
continue
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None and npt.noun.is_value("СТОРОНА", None) and npt.morph.number != MorphNumber.SINGULAR):
re = ParticipantToken._new1573(t, npt.end_token, ParticipantToken.Kinds.NAMEDASPARTS)
re.parts = refs
return re
break
if ((isinstance(r1, OrganizationReferent)) or (isinstance(r1, PersonReferent))):
has_br = False
has_named = False
if (isinstance(r1, PersonReferent)):
if (t.previous is not None and t.previous.is_value("ЛИЦО", None)):
return None
elif (t.previous is not None and ((t.previous.is_value("ВЫДАВАТЬ", None) or t.previous.is_value("ВЫДАТЬ", None)))):
return None
ttt = t.begin_token
while ttt is not None and (ttt.end_char < t.end_char):
if (ttt.is_char('(')):
has_br = True
elif ((ttt.is_value("ИМЕНОВАТЬ", None) or ttt.is_value("ДАЛЬНЕЙШИЙ", None) or ttt.is_value("ДАЛЕЕ", None)) or ttt.is_value("ТЕКСТ", None)):
has_named = True
elif ((ttt.is_comma or ttt.morph.class0_.is_preposition or ttt.is_hiphen) or ttt.is_char(':')):
pass
elif (isinstance(ttt, ReferentToken)):
pass
elif (has_br or has_named):
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.REFERENTCANBENOUN, 0, None)
if (npt is None):
break
if (has_br):
if (npt.end_token.next0_ is None or not npt.end_token.next0_.is_char(')')):
break
if (not has_named):
if (ParticipantToken.M_ONTOLOGY.try_parse(ttt, TerminParseAttr.NO) is None):
break
re = ParticipantToken._new1573(t, t, ParticipantToken.Kinds.NAMEDAS)
re.typ = npt.get_normal_case_text(None, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
re.parts = list()
re.parts.append(r1)
return re
ttt = ttt.next0_
has_br = False
has_named = False
end_side = None
brr = None
add_refs = None
ttt = t.next0_
first_pass3284 = True
while True:
if first_pass3284: first_pass3284 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if ((isinstance(ttt, NumberToken)) and (isinstance(ttt.next0_, TextToken)) and ttt.next0_.term == "СТОРОНЫ"):
ttt = ttt.next0_
end_side = ttt
if (ttt.next0_ is not None and ttt.next0_.is_comma):
ttt = ttt.next0_
if (ttt.next0_ is not None and ttt.next0_.is_and):
break
if (brr is not None and ttt.begin_char > brr.end_char):
brr = (None)
if (BracketHelper.can_be_start_of_sequence(ttt, False, False)):
brr = BracketHelper.try_parse(ttt, BracketParseAttr.NO, 100)
if (brr is not None and (brr.length_char < 7) and ttt.is_char('(')):
ttt = brr.end_token
brr = (None)
continue
elif ((ttt.is_value("ИМЕНОВАТЬ", None) or ttt.is_value("ДАЛЬНЕЙШИЙ", None) or ttt.is_value("ДАЛЕЕ", None)) or ttt.is_value("ТЕКСТ", None)):
has_named = True
elif ((ttt.is_comma or ttt.morph.class0_.is_preposition or ttt.is_hiphen) or ttt.is_char(':')):
pass
elif (brr is not None or has_named):
if (BracketHelper.can_be_start_of_sequence(ttt, True, False)):
ttt = ttt.next0_
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.REFERENTCANBENOUN, 0, None)
typ22 = None
if (npt is not None):
ttt = npt.end_token
if (npt.end_token.is_value("ДОГОВОР", None)):
continue
else:
ttok = None
if (isinstance(ttt, MetaToken)):
ttok = ParticipantToken.M_ONTOLOGY.try_parse(ttt.begin_token, TerminParseAttr.NO)
if (ttok is not None):
typ22 = ttok.termin.canonic_text
elif (has_named and ttt.morph.class0_.is_adjective):
typ22 = ttt.get_normal_case_text(MorphClass.ADJECTIVE, MorphNumber.UNDEFINED, MorphGender.UNDEFINED, False)
elif (brr is not None):
continue
else:
break
if (BracketHelper.can_be_end_of_sequence(ttt.next0_, True, None, False)):
ttt = ttt.next0_
if (brr is not None):
if (ttt.next0_ is None):
ttt = brr.end_token
continue
ttt = ttt.next0_
if (not has_named and typ22 is None):
if (ParticipantToken.M_ONTOLOGY.try_parse(npt.begin_token, TerminParseAttr.NO) is None):
break
re = ParticipantToken._new1573(t, ttt, ParticipantToken.Kinds.NAMEDAS)
re.typ = (Utils.ifNotNull(typ22, npt.get_normal_case_text(None, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)))
re.parts = list()
re.parts.append(r1)
return re
elif ((ttt.is_value("ЗАРЕГИСТРИРОВАННЫЙ", None) or ttt.is_value("КАЧЕСТВО", None) or ttt.is_value("ПРОЖИВАЮЩИЙ", None)) or ttt.is_value("ЗАРЕГ", None)):
pass
elif (ttt.get_referent() == r1):
pass
elif ((isinstance(ttt.get_referent(), PersonIdentityReferent)) or (isinstance(ttt.get_referent(), AddressReferent))):
if (add_refs is None):
add_refs = list()
add_refs.append(ttt.get_referent())
else:
prr = ttt.kit.process_referent("PERSONPROPERTY", ttt)
if (prr is not None):
ttt = prr.end_token
continue
if (isinstance(ttt.get_referent(), GeoReferent)):
continue
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
if ((npt.noun.is_value("МЕСТО", None) or npt.noun.is_value("ЖИТЕЛЬСТВО", None) or npt.noun.is_value("ПРЕДПРИНИМАТЕЛЬ", None)) or npt.noun.is_value("ПОЛ", None) or npt.noun.is_value("РОЖДЕНИЕ", None)):
ttt = npt.end_token
continue
if (ttt.is_newline_before):
break
if (ttt.length_char < 3):
continue
mc = ttt.get_morph_class_in_dictionary()
if (mc.is_adverb or mc.is_adjective):
continue
if (ttt.chars.is_all_upper):
continue
break
if (end_side is not None or ((add_refs is not None and t.previous is not None and t.previous.is_and))):
re = ParticipantToken._new1573(t, Utils.ifNotNull(end_side, t), ParticipantToken.Kinds.NAMEDAS)
re.typ = (None)
re.parts = list()
re.parts.append(r1)
if (add_refs is not None):
re.parts.extend(add_refs)
return re
too = ParticipantToken.M_ONTOLOGY.try_parse(t, TerminParseAttr.NO)
if (too is not None):
if ((isinstance(t.previous, TextToken)) and t.previous.is_value("ЛИЦО", None)):
too = (None)
if (too is not None and too.termin.tag is not None and too.termin.canonic_text != "СТОРОНА"):
tt1 = too.end_token.next0_
if (tt1 is not None):
if (tt1.is_hiphen or tt1.is_char(':')):
tt1 = tt1.next0_
if (isinstance(tt1, ReferentToken)):
r1 = tt1.get_referent()
if ((isinstance(r1, PersonReferent)) or (isinstance(r1, OrganizationReferent))):
re = ParticipantToken._new1573(t, tt1, ParticipantToken.Kinds.NAMEDAS)
re.typ = too.termin.canonic_text
re.parts = list()
re.parts.append(r1)
return re
add_typ1 = (None if p1 is None else p1.typ)
add_typ2 = (None if p2 is None else p2.typ)
if (BracketHelper.can_be_start_of_sequence(tt, False, False) and tt.next0_ is not None):
br = True
tt = tt.next0_
term1 = None
term2 = None
if (add_typ1 is not None and add_typ1.find(' ') > 0 and not add_typ1.startswith("СТОРОНА")):
term1 = Termin(add_typ1)
if (add_typ2 is not None and add_typ2.find(' ') > 0 and not add_typ2.startswith("СТОРОНА")):
term2 = Termin(add_typ2)
named = False
typ_ = None
t1 = None
t0 = tt
first_pass3285 = True
while True:
if first_pass3285: first_pass3285 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.morph.class0_.is_preposition and typ_ is not None):
continue
if (tt.is_char_of("(:)") or tt.is_hiphen):
continue
if (tt.is_table_control_char):
break
if (tt.is_newline_before and tt != t0):
if (isinstance(tt, NumberToken)):
break
if ((isinstance(tt, TextToken)) and (isinstance(tt.previous, TextToken))):
if (tt.previous.is_value(tt.term, None)):
break
if (BracketHelper.is_bracket(tt, False)):
continue
tok = (ParticipantToken.M_ONTOLOGY.try_parse(tt, TerminParseAttr.NO) if ParticipantToken.M_ONTOLOGY is not None else None)
if (tok is not None and (isinstance(tt.previous, TextToken))):
if (tt.previous.is_value("ЛИЦО", None)):
return None
if (tok is None):
if (add_typ1 is not None and ((MiscHelper.is_not_more_than_one_error(add_typ1, tt) or (((isinstance(tt, MetaToken)) and tt.begin_token.is_value(add_typ1, None)))))):
if (typ_ is not None):
if (not ParticipantToken.__is_types_equal(add_typ1, typ_)):
break
typ_ = add_typ1
t1 = tt
continue
if (add_typ2 is not None and ((MiscHelper.is_not_more_than_one_error(add_typ2, tt) or (((isinstance(tt, MetaToken)) and tt.begin_token.is_value(add_typ2, None)))))):
if (typ_ is not None):
if (not ParticipantToken.__is_types_equal(add_typ2, typ_)):
break
typ_ = add_typ2
t1 = tt
continue
if (tt.chars.is_letter):
if (term1 is not None):
tok1 = term1.try_parse(tt, TerminParseAttr.NO)
if (tok1 is not None):
if (typ_ is not None):
if (not ParticipantToken.__is_types_equal(add_typ1, typ_)):
break
typ_ = add_typ1
tt = tok1.end_token
t1 = tt
continue
if (term2 is not None):
tok2 = term2.try_parse(tt, TerminParseAttr.NO)
if (tok2 is not None):
if (typ_ is not None):
if (not ParticipantToken.__is_types_equal(add_typ2, typ_)):
break
typ_ = add_typ2
tt = tok2.end_token
t1 = tt
continue
if (named and tt.get_morph_class_in_dictionary().is_noun):
if (not tt.chars.is_all_lower or BracketHelper.is_bracket(tt.previous, True)):
if (DecreeToken.is_keyword(tt, False) is None):
val = tt.get_normal_case_text(MorphClass.NOUN, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
if (typ_ is not None):
if (not ParticipantToken.__is_types_equal(typ_, val)):
break
typ_ = val
t1 = tt
continue
if (named and typ_ is None and is_contract):
if ((isinstance(tt, TextToken)) and tt.chars.is_cyrillic_letter and tt.chars.is_capital_upper):
dc = tt.get_morph_class_in_dictionary()
if (dc.is_undefined or dc.is_noun):
dt = DecreeToken.try_attach(tt, None, False)
ok = True
if (dt is not None):
ok = False
elif (tt.is_value("СТОРОНА", None)):
ok = False
if (ok):
typ_ = tt.lemma
t1 = tt
continue
if (dc.is_adjective):
npt = NounPhraseHelper.try_parse(tt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None and len(npt.adjectives) > 0 and npt.noun.get_morph_class_in_dictionary().is_noun):
typ_ = npt.get_normal_case_text(None, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
t1 = npt.end_token
continue
if (tt == t):
break
if ((isinstance(tt, NumberToken)) or tt.is_char('.')):
break
if (tt.length_char < 4):
if (typ_ is not None):
continue
break
if (tok.termin.tag is None):
named = True
else:
if (typ_ is not None):
break
if (tok.termin.canonic_text == "СТОРОНА"):
tt1 = tt.next0_
if (tt1 is not None and tt1.is_hiphen):
tt1 = tt1.next0_
if (not (isinstance(tt1, NumberToken))):
break
if (tt1.is_newline_before):
break
typ_ = "{0} {1}".format(tok.termin.canonic_text, tt1.value)
t1 = tt1
else:
typ_ = tok.termin.canonic_text
t1 = tok.end_token
break
tt = tok.end_token
if (typ_ is None):
return None
if (not named and t1 != t and not typ_.startswith("СТОРОНА")):
if (not ParticipantToken.__is_types_equal(typ_, add_typ1) and not ParticipantToken.__is_types_equal(typ_, add_typ2)):
return None
if (BracketHelper.can_be_end_of_sequence(t1.next0_, False, None, False)):
t1 = t1.next0_
if (not t.is_whitespace_before and BracketHelper.can_be_start_of_sequence(t.previous, False, False)):
t = t.previous
elif (BracketHelper.can_be_start_of_sequence(t, False, False) and BracketHelper.can_be_end_of_sequence(t1.next0_, True, t, True)):
t1 = t1.next0_
if (br and t1.next0_ is not None and BracketHelper.can_be_end_of_sequence(t1.next0_, False, None, False)):
t1 = t1.next0_
res = ParticipantToken._new1578(t, t1, (ParticipantToken.Kinds.NAMEDAS if named else ParticipantToken.Kinds.PURE), typ_)
if (t.is_char(':')):
res.begin_token = t.next0_
return res
def tryAttach(t : 'Token') -> 'TitleItemToken':
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is not None):
t1 = tt
if (tt.term == "ТЕМА"):
tit = TitleItemToken.tryAttach(tt.next0_)
if (tit is not None and tit.typ == TitleItemToken.Types.TYP):
t1 = tit.end_token
if (t1.next0_ is not None and t1.next0_.isChar(':')):
t1 = t1.next0_
return TitleItemToken._new2501(t, t1, TitleItemToken.Types.TYPANDTHEME, tit.value)
if (tt.next0_ is not None and tt.next0_.isChar(':')):
t1 = tt.next0_
return TitleItemToken(tt, t1, TitleItemToken.Types.THEME)
if (tt.term == "ПО" or tt.term == "НА"):
if (tt.next0_ is not None and tt.next0_.isValue("ТЕМА", None)):
t1 = tt.next0_
if (t1.next0_ is not None and t1.next0_.isChar(':')):
t1 = t1.next0_
return TitleItemToken(tt, t1, TitleItemToken.Types.THEME)
if (tt.term == "ПЕРЕВОД" or tt.term == "ПЕР"):
tt2 = tt.next0_
if (tt2 is not None and tt2.isChar('.')):
tt2 = tt2.next0_
if (isinstance(tt2, TextToken)):
if ((tt2).term == "C" or (tt2).term == "С"):
tt2 = tt2.next0_
if (isinstance(tt2, TextToken)):
return TitleItemToken(t, tt2, TitleItemToken.Types.TRANSLATE)
if (tt.term == "СЕКЦИЯ" or tt.term == "SECTION" or tt.term == "СЕКЦІЯ"):
t1 = tt.next0_
if (t1 is not None and t1.isChar(':')):
t1 = t1.next0_
br = BracketHelper.tryParse(t1, BracketParseAttr.NO, 100)
if (br is not None):
t1 = br.end_token
elif (t1 != tt.next0_):
while t1 is not None:
if (t1.is_newline_after):
break
t1 = t1.next0_
if (t1 is None):
return None
if (t1 != tt.next0_):
return TitleItemToken(tt, t1, TitleItemToken.Types.DUST)
t1 = (None)
if (tt.isValue("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")):
t1 = tt.next0_
elif (tt.morph.class0_.is_preposition and tt.next0_ is not None and tt.next0_.isValue("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")):
t1 = tt.next0_.next0_
elif (tt.isChar('/') and tt.is_newline_before):
t1 = tt.next0_
if (t1 is not None):
if (t1.isCharOf(":") or t1.is_hiphen):
t1 = t1.next0_
spec = TitleItemToken.__tryAttachSpeciality(t1, True)
if (spec is not None):
spec.begin_token = t
return spec
sss = TitleItemToken.__tryAttachSpeciality(t, False)
if (sss is not None):
return sss
if (isinstance(t, ReferentToken)):
return None
npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0)
if (npt is not None):
s = npt.getNormalCaseText(None, False, MorphGender.UNDEFINED, False)
tok = TitleItemToken.M_TERMINS.tryParse(npt.end_token, TerminParseAttr.NO)
if (tok is not None):
ty = Utils.valToEnum(tok.termin.tag, TitleItemToken.Types)
if (ty == TitleItemToken.Types.TYP):
tit = TitleItemToken.tryAttach(tok.end_token.next0_)
if (tit is not None and tit.typ == TitleItemToken.Types.THEME):
return TitleItemToken._new2501(npt.begin_token, tit.end_token, TitleItemToken.Types.TYPANDTHEME, s)
if (s == "РАБОТА" or s == "РОБОТА" or s == "ПРОЕКТ"):
return None
t1 = tok.end_token
if (s == "ДИССЕРТАЦИЯ" or s == "ДИСЕРТАЦІЯ"):
err = 0
ttt = t1.next0_
first_pass3125 = True
while True:
if first_pass3125: first_pass3125 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.morph.class0_.is_preposition):
continue
if (ttt.isValue("СОИСКАНИЕ", "")):
continue
npt1 = NounPhraseHelper.tryParse(ttt, NounPhraseParseAttr.NO, 0)
if (npt1 is not None and npt1.noun.isValue("СТЕПЕНЬ", "СТУПІНЬ")):
ttt = npt1.end_token
t1 = ttt
continue
rt = t1.kit.processReferent("PERSON", ttt)
if (rt is not None and (isinstance(rt.referent, PersonPropertyReferent))):
ppr = Utils.asObjectOrNull(rt.referent, PersonPropertyReferent)
if (ppr.name == "доктор наук"):
t1 = rt.end_token
s = "ДОКТОРСКАЯ ДИССЕРТАЦИЯ"
break
elif (ppr.name == "кандидат наук"):
t1 = rt.end_token
s = "КАНДИДАТСКАЯ ДИССЕРТАЦИЯ"
break
elif (ppr.name == "магистр"):
t1 = rt.end_token
s = "МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ"
break
if (ttt.isValue("ДОКТОР", None) or ttt.isValue("КАНДИДАТ", None) or ttt.isValue("МАГИСТР", "МАГІСТР")):
t1 = ttt
npt1 = NounPhraseHelper.tryParse(ttt.next0_, NounPhraseParseAttr.NO, 0)
if (npt1 is not None and npt1.end_token.isValue("НАУК", None)):
t1 = npt1.end_token
s = ("МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ" if ttt.isValue("МАГИСТР", "МАГІСТР") else ("ДОКТОРСКАЯ ДИССЕРТАЦИЯ" if ttt.isValue("ДОКТОР", None) else "КАНДИДАТСКАЯ ДИССЕРТАЦИЯ"))
break
err += 1
if ((err) > 3):
break
if (t1.next0_ is not None and t1.next0_.isChar('.')):
t1 = t1.next0_
if (s.endswith("ОТЧЕТ") and t1.next0_ is not None and t1.next0_.isValue("О", None)):
npt1 = NounPhraseHelper.tryParse(t1.next0_, NounPhraseParseAttr.PARSEPREPOSITION, 0)
if (npt1 is not None and npt1.morph.case_.is_prepositional):
t1 = npt1.end_token
return TitleItemToken._new2501(npt.begin_token, t1, ty, s)
tok1 = TitleItemToken.M_TERMINS.tryParse(t, TerminParseAttr.NO)
if (tok1 is not None):
t1 = tok1.end_token
re = TitleItemToken(tok1.begin_token, t1, Utils.valToEnum(tok1.termin.tag, TitleItemToken.Types))
return re
if (BracketHelper.canBeStartOfSequence(t, False, False)):
tok1 = TitleItemToken.M_TERMINS.tryParse(t.next0_, TerminParseAttr.NO)
if (tok1 is not None and BracketHelper.canBeEndOfSequence(tok1.end_token.next0_, False, None, False)):
t1 = tok1.end_token.next0_
return TitleItemToken(tok1.begin_token, t1, Utils.valToEnum(tok1.termin.tag, TitleItemToken.Types))
return None
def __TryParse(t: 'Token',
prev: 'WeaponItemToken',
after_conj: bool,
attach_high: bool = False) -> 'WeaponItemToken':
if (t is None):
return None
if (BracketHelper.isBracket(t, True)):
wit = WeaponItemToken.__TryParse(t.next0_, prev, after_conj,
attach_high)
if (wit is not None):
if (wit.end_token.next0_ is None):
wit.begin_token = t
return wit
if (BracketHelper.isBracket(wit.end_token.next0_, True)):
wit.begin_token = t
wit.end_token = wit.end_token.next0_
return wit
tok = WeaponItemToken.M_ONTOLOGY.tryParse(t, TerminParseAttr.NO)
if (tok is not None):
res = WeaponItemToken(t, tok.end_token)
res.typ = (Utils.valToEnum(tok.termin.tag, WeaponItemToken.Typs))
if (res.typ == WeaponItemToken.Typs.NOUN):
res.value = tok.termin.canonic_text
if (tok.termin.tag2 is not None):
res.is_doubt = True
tt = res.end_token.next0_
first_pass3156 = True
while True:
if first_pass3156: first_pass3156 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.whitespaces_before_count > 2):
break
wit = WeaponItemToken.__TryParse(tt, None, False, False)
if (wit is not None):
if (wit.typ == WeaponItemToken.Typs.BRAND):
res.__inner_tokens.append(wit)
tt = wit.end_token
res.end_token = tt
continue
break
if (not ((isinstance(tt, TextToken)))):
break
mc = tt.getMorphClassInDictionary()
if (mc == MorphClass.ADJECTIVE):
if (res.alt_value is None):
res.alt_value = res.value
if (res.alt_value.endswith(res.value)):
res.alt_value = res.alt_value[0:0 +
len(res.alt_value) -
len(res.value)]
res.alt_value = "{0}{1} {2}".format(
res.alt_value, (tt).term, res.value)
res.end_token = tt
continue
break
return res
if (res.typ == WeaponItemToken.Typs.BRAND
or res.typ == WeaponItemToken.Typs.NAME):
res.value = tok.termin.canonic_text
return res
if (res.typ == WeaponItemToken.Typs.MODEL):
res.value = tok.termin.canonic_text
if (isinstance(tok.termin.tag2, list)):
li = Utils.asObjectOrNull(tok.termin.tag2, list)
for to in li:
wit = WeaponItemToken._new2600(
t, tok.end_token,
Utils.valToEnum(to.tag, WeaponItemToken.Typs),
to.canonic_text, tok.begin_token == tok.end_token)
res.__inner_tokens.append(wit)
if (to.additional_vars is not None
and len(to.additional_vars) > 0):
wit.alt_value = to.additional_vars[0].canonic_text
res.__correctModel()
return res
nnn = MiscHelper.checkNumberPrefix(t)
if (nnn is not None):
tit = TransItemToken._attachNumber(nnn, True)
if (tit is not None):
res = WeaponItemToken._new2601(t, tit.end_token,
WeaponItemToken.Typs.NUMBER)
res.value = tit.value
res.alt_value = tit.alt_value
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter
and t.chars.is_all_upper) and (t.length_char < 4)):
if ((t.next0_ is not None and
((t.next0_.is_hiphen or t.next0_.isChar('.'))) and
(t.next0_.whitespaces_after_count < 2))
and (isinstance(t.next0_.next0_, NumberToken))):
res = WeaponItemToken._new2602(t, t.next0_,
WeaponItemToken.Typs.MODEL,
True)
res.value = (t).term
res.__correctModel()
return res
if ((isinstance(t.next0_, NumberToken))
and not t.is_whitespace_after):
res = WeaponItemToken._new2602(t, t,
WeaponItemToken.Typs.MODEL,
True)
res.value = (t).term
res.__correctModel()
return res
if ((t).term == "СП" and (t.whitespaces_after_count < 3)
and (isinstance(t.next0_, TextToken))):
pp = WeaponItemToken.__TryParse(t.next0_, None, False, False)
if (pp is not None
and ((pp.typ == WeaponItemToken.Typs.MODEL
or pp.typ == WeaponItemToken.Typs.BRAND))):
res = WeaponItemToken._new2601(t, t,
WeaponItemToken.Typs.NOUN)
res.value = "ПИСТОЛЕТ"
res.alt_value = "СЛУЖЕБНЫЙ ПИСТОЛЕТ"
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter
and not t.chars.is_all_lower) and t.length_char > 2):
ok = False
if (prev is not None
and ((prev.typ == WeaponItemToken.Typs.NOUN
or prev.typ == WeaponItemToken.Typs.MODEL
or prev.typ == WeaponItemToken.Typs.BRAND))):
ok = True
elif (prev is None and t.previous is not None
and t.previous.is_comma_and):
ok = True
if (ok):
res = WeaponItemToken._new2602(t, t, WeaponItemToken.Typs.NAME,
True)
res.value = (t).term
if ((t.next0_ is not None and t.next0_.is_hiphen and
(isinstance(t.next0_.next0_, TextToken)))
and t.next0_.next0_.chars == t.chars):
res.value = "{0}-{1}".format(res.value,
(t.next0_.next0_).term)
res.end_token = t.next0_.next0_
if (prev is not None
and prev.typ == WeaponItemToken.Typs.NOUN):
res.typ = WeaponItemToken.Typs.BRAND
if (res.end_token.next0_ is not None
and res.end_token.next0_.is_hiphen and
(isinstance(res.end_token.next0_.next0_, NumberToken))):
res.typ = WeaponItemToken.Typs.MODEL
res.__correctModel()
elif (not res.end_token.is_whitespace_after
and (isinstance(res.end_token.next0_, NumberToken))):
res.typ = WeaponItemToken.Typs.MODEL
res.__correctModel()
return res
return None
def try_parse_list(t : 'Token', max_count : int=10) -> typing.List['WeaponItemToken']:
tr = WeaponItemToken.try_parse(t, None, False, False)
if (tr is None):
return None
if (tr.typ == WeaponItemToken.Typs.CLASS or tr.typ == WeaponItemToken.Typs.DATE):
return None
tr0 = tr
res = list()
if (len(tr.__inner_tokens) > 0):
res.extend(tr.__inner_tokens)
if (res[0].begin_char > tr.begin_char):
res[0].begin_token = tr.begin_token
res.append(tr)
t = tr.end_token.next0_
if (tr.typ == WeaponItemToken.Typs.NOUN):
while t is not None:
if (t.is_char(':') or t.is_hiphen):
pass
else:
break
t = t.next0_
and_conj = False
first_pass3425 = True
while True:
if first_pass3425: first_pass3425 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_count > 0 and len(res) >= max_count):
break
if (t.is_char(':')):
continue
if (tr0.typ == WeaponItemToken.Typs.NOUN):
if (t.is_hiphen and t.next0_ is not None):
t = t.next0_
tr = WeaponItemToken.try_parse(t, tr0, False, False)
if (tr is None):
if (BracketHelper.can_be_end_of_sequence(t, True, None, False) and t.next0_ is not None):
if (tr0.typ == WeaponItemToken.Typs.MODEL or tr0.typ == WeaponItemToken.Typs.BRAND):
tt1 = t.next0_
if (tt1 is not None and tt1.is_comma):
tt1 = tt1.next0_
tr = WeaponItemToken.try_parse(tt1, tr0, False, False)
if (tr is None and (isinstance(t, ReferentToken))):
rt = Utils.asObjectOrNull(t, ReferentToken)
if (rt.begin_token == rt.end_token and (isinstance(rt.begin_token, TextToken))):
tr = WeaponItemToken.try_parse(rt.begin_token, tr0, False, False)
if (tr is not None and tr.begin_token == tr.end_token):
tr.begin_token = tr.end_token = t
if (tr is None and t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token.next0_
if (tt is not None and tt.is_comma):
tt = tt.next0_
tr = WeaponItemToken.try_parse(tt, tr0, False, False)
if (tr is not None and tr.typ == WeaponItemToken.Typs.NUMBER):
pass
else:
tr = (None)
if (tr is None and t.is_hiphen):
if (tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL):
tr = WeaponItemToken.try_parse(t.next0_, tr0, False, False)
if (tr is None and t.is_comma):
if ((tr0.typ == WeaponItemToken.Typs.NAME or tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL) or tr0.typ == WeaponItemToken.Typs.CLASS or tr0.typ == WeaponItemToken.Typs.DATE):
tr = WeaponItemToken.try_parse(t.next0_, tr0, True, False)
if (tr is not None):
if (tr.typ == WeaponItemToken.Typs.NUMBER):
pass
else:
tr = (None)
if (tr is None):
break
if (t.is_newline_before):
if (tr.typ != WeaponItemToken.Typs.NUMBER):
break
if (len(tr.__inner_tokens) > 0):
res.extend(tr.__inner_tokens)
res.append(tr)
tr0 = tr
t = tr.end_token
if (and_conj):
break
i = 0
while i < (len(res) - 1):
if (res[i].typ == WeaponItemToken.Typs.MODEL and res[i + 1].typ == WeaponItemToken.Typs.MODEL):
res[i].end_token = res[i + 1].end_token
res[i].value = "{0}{1}{2}".format(res[i].value, ('-' if res[i].end_token.next0_ is not None and res[i].end_token.next0_.is_hiphen else ' '), res[i + 1].value)
del res[i + 1]
i -= 1
i += 1
return res
def getNameEx(begin: 'Token',
end: 'Token',
cla: 'MorphClass',
mc: 'MorphCase',
gender: 'MorphGender' = MorphGender.UNDEFINED,
ignore_brackets_and_hiphens: bool = False,
ignore_geo_referent: bool = False) -> str:
if (end is None or begin is None):
return None
if (begin.end_char > end.begin_char and begin != end):
return None
res = io.StringIO()
prefix = None
t = begin
first_pass2809 = True
while True:
if first_pass2809: first_pass2809 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= end.end_char)): break
if (res.tell() > 1000):
break
if (t.is_table_control_char):
continue
if (ignore_brackets_and_hiphens):
if (BracketHelper.isBracket(t, False)):
if (t == end):
break
if (t.isCharOf("(<[")):
br = BracketHelper.tryParse(t, BracketParseAttr.NO,
100)
if (br is not None and br.end_char <= end.end_char):
tmp = ProperNameHelper.getNameEx(
br.begin_token.next0_, br.end_token.previous,
MorphClass.UNDEFINED, MorphCase.UNDEFINED,
MorphGender.UNDEFINED,
ignore_brackets_and_hiphens, False)
if (tmp is not None):
if ((br.end_char == end.end_char
and br.begin_token.next0_
== br.end_token.previous and
not br.begin_token.next0_.chars.is_letter)
and not ((isinstance(
br.begin_token.next0_,
ReferentToken)))):
pass
else:
print(" {0}{1}{2}".format(
t.getSourceText(), tmp,
br.end_token.getSourceText()),
end="",
file=res,
flush=True)
t = br.end_token
continue
if (t.is_hiphen):
if (t == end):
break
elif (t.is_whitespace_before or t.is_whitespace_after):
continue
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is not None):
if (not ignore_brackets_and_hiphens):
if ((tt.next0_ is not None and tt.next0_.is_hiphen and
(isinstance(tt.next0_.next0_, TextToken)))
and tt != end and tt.next0_ != end):
if (prefix is None):
prefix = tt.term
else:
prefix = "{0}-{1}".format(prefix, tt.term)
t = tt.next0_
if (t == end):
break
else:
continue
s = None
if (cla.value != (0) or not mc.is_undefined
or gender != MorphGender.UNDEFINED):
for wff in tt.morph.items:
wf = Utils.asObjectOrNull(wff, MorphWordForm)
if (wf is None):
continue
if (cla.value != (0)):
if ((((wf.class0_.value) & (cla.value))) == 0):
continue
if (not mc.is_undefined):
if (((wf.case_) & mc).is_undefined):
continue
if (gender != MorphGender.UNDEFINED):
if ((((wf.gender) &
(gender))) == (MorphGender.UNDEFINED)):
continue
if (s is None or wf.normal_case == tt.term):
s = wf.normal_case
if (s is None and gender != MorphGender.UNDEFINED):
for wff in tt.morph.items:
wf = Utils.asObjectOrNull(wff, MorphWordForm)
if (wf is None):
continue
if (cla.value != (0)):
if ((((wf.class0_.value) & (cla.value))) == 0):
continue
if (not mc.is_undefined):
if (((wf.case_) & mc).is_undefined):
continue
if (s is None or wf.normal_case == tt.term):
s = wf.normal_case
if (s is None):
s = tt.term
if (tt.chars.is_last_lower and tt.length_char > 2):
s = tt.getSourceText()
for i in range(len(s) - 1, -1, -1):
if (str.isupper(s[i])):
s = s[0:0 + i + 1]
break
if (prefix is not None):
delim = "-"
if (ignore_brackets_and_hiphens):
delim = " "
s = "{0}{1}{2}".format(prefix, delim, s)
prefix = (None)
if (res.tell() > 0 and len(s) > 0):
if (str.isalnum(s[0])):
ch0 = Utils.getCharAtStringIO(res, res.tell() - 1)
if (ch0 == '-'):
pass
else:
print(' ', end="", file=res)
elif (not ignore_brackets_and_hiphens
and BracketHelper.canBeStartOfSequence(
tt, False, False)):
print(' ', end="", file=res)
print(s, end="", file=res)
elif (isinstance(t, NumberToken)):
if (res.tell() > 0):
if (not t.is_whitespace_before and Utils.getCharAtStringIO(
res,
res.tell() - 1) == '-'):
pass
else:
print(' ', end="", file=res)
nt = Utils.asObjectOrNull(t, NumberToken)
if ((t.morph.class0_.is_adjective
and nt.typ == NumberSpellingType.WORDS
and nt.begin_token == nt.end_token)
and (isinstance(nt.begin_token, TextToken))):
print((nt.begin_token).term, end="", file=res)
else:
print(nt.value, end="", file=res)
elif (isinstance(t, MetaToken)):
if ((ignore_geo_referent and t != begin
and t.getReferent() is not None)
and t.getReferent().type_name == "GEO"):
continue
s = ProperNameHelper.getNameEx(
(t).begin_token, (t).end_token, cla, mc, gender,
ignore_brackets_and_hiphens, ignore_geo_referent)
if (not Utils.isNullOrEmpty(s)):
if (res.tell() > 0):
if (not t.is_whitespace_before
and Utils.getCharAtStringIO(
res,
res.tell() - 1) == '-'):
pass
else:
print(' ', end="", file=res)
print(s, end="", file=res)
if (t == end):
break
if (res.tell() == 0):
return None
return Utils.toStringStringIO(res)
def try_attach(t: 'Token') -> 'ParenthesisToken':
if (t is None):
return None
tok = ParenthesisToken.__m_termins.try_parse(t, TerminParseAttr.NO)
if (tok is not None):
res = ParenthesisToken(t, tok.end_token)
return res
if (not (isinstance(t, TextToken))):
return None
mc = t.get_morph_class_in_dictionary()
ok = False
t1 = None
if (mc.is_adverb):
ok = True
elif (mc.is_adjective):
if (t.morph.contains_attr("сравн.", None)
and t.morph.contains_attr("кач.прил.", None)):
ok = True
if (ok and t.next0_ is not None):
if (t.next0_.is_char(',')):
return ParenthesisToken(t, t)
t1 = t.next0_
if (t1.get_morph_class_in_dictionary() == MorphClass.VERB):
if (t1.morph.contains_attr("н.вр.", None)
and t1.morph.contains_attr("нес.в.", None)
and t1.morph.contains_attr("дейст.з.", None)):
return ParenthesisToken(t, t1)
t1 = (None)
if ((t.is_value("В", None) and t.next0_ is not None
and t.next0_.is_value("СООТВЕТСТВИЕ", None))
and t.next0_.next0_ is not None
and t.next0_.next0_.morph.class0_.is_preposition):
t1 = t.next0_.next0_.next0_
elif (t.is_value("СОГЛАСНО", None)):
t1 = t.next0_
elif (t.is_value("В", None) and t.next0_ is not None):
if (t.next0_.is_value("СИЛА", None)):
t1 = t.next0_.next0_
elif (t.next0_.morph.class0_.is_adjective
or t.next0_.morph.class0_.is_pronoun):
npt = NounPhraseHelper.try_parse(t.next0_,
NounPhraseParseAttr.NO, 0,
None)
if (npt is not None):
if (npt.noun.is_value("ВИД", None)
or npt.noun.is_value("СЛУЧАЙ", None)
or npt.noun.is_value("СФЕРА", None)):
return ParenthesisToken(t, npt.end_token)
if (t1 is not None):
if (t1.next0_ is not None):
npt1 = NounPhraseHelper.try_parse(t1, NounPhraseParseAttr.NO,
0, None)
if (npt1 is not None):
if (npt1.noun.is_value("НОРМА", None)
or npt1.noun.is_value("ПОЛОЖЕНИЕ", None)
or npt1.noun.is_value("УКАЗАНИЕ", None)):
t1 = npt1.end_token.next0_
r = t1.get_referent()
if (r is not None):
res = ParenthesisToken._new1115(t, t1, r)
if (t1.next0_ is not None and t1.next0_.is_comma):
sila = False
ttt = t1.next0_.next0_
first_pass3133 = True
while True:
if first_pass3133: first_pass3133 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.is_value("СИЛА", None)
or ttt.is_value("ДЕЙСТВИЕ", None)):
sila = True
continue
if (ttt.is_comma):
if (sila):
res.end_token = ttt.previous
break
if (BracketHelper.can_be_start_of_sequence(
ttt, False, False)):
break
return res
npt = NounPhraseHelper.try_parse(t1, NounPhraseParseAttr.NO, 0,
None)
if (npt is not None):
return ParenthesisToken(t, npt.end_token)
tt = t
if (tt.is_value("НЕ", None) and t is not None):
tt = tt.next0_
if (tt.morph.class0_.is_preposition and tt is not None):
tt = tt.next0_
npt1 = NounPhraseHelper.try_parse(tt, NounPhraseParseAttr.NO, 0,
None)
if (npt1 is not None):
tt = npt1.end_token
if (tt.next0_ is not None and tt.next0_.is_comma):
return ParenthesisToken(t, tt.next0_)
if (npt1.noun.is_value("ОЧЕРЕДЬ", None)):
return ParenthesisToken(t, tt)
if (t.is_value("ВЕДЬ", None)):
return ParenthesisToken(t, t)
return None
def __getNameWithoutBrackets(begin: 'Token',
end: 'Token',
normalize_first_noun_group: bool = False,
normal_first_group_single: bool = False,
ignore_geo_referent: bool = False) -> str:
""" Получить строковое значение между токенами, при этом исключая кавычки и скобки
Args:
begin(Token): начальный токен
end(Token): конечный токен
normalize_first_noun_group(bool): нормализовывать ли первую именную группу (именит. падеж)
normal_first_group_single(bool): приводить ли к единственному числу первую именную группу
ignore_geo_referent(bool): игнорировать внутри географические сущности
"""
res = None
if (BracketHelper.canBeStartOfSequence(begin, False, False) and
BracketHelper.canBeEndOfSequence(end, False, begin, False)):
begin = begin.next0_
end = end.previous
if (normalize_first_noun_group
and not begin.morph.class0_.is_preposition):
npt = NounPhraseHelper.tryParse(
begin, NounPhraseParseAttr.REFERENTCANBENOUN, 0)
if (npt is not None):
if (npt.noun.getMorphClassInDictionary().is_undefined
and len(npt.adjectives) == 0):
npt = (None)
if (npt is not None and npt.end_token.end_char > end.end_char):
npt = (None)
if (npt is not None):
res = npt.getNormalCaseText(None, normal_first_group_single,
MorphGender.UNDEFINED, False)
te = npt.end_token.next0_
if (((te is not None and te.next0_ is not None and te.is_comma)
and (isinstance(te.next0_, TextToken))
and te.next0_.end_char <= end.end_char)
and te.next0_.morph.class0_.is_verb
and te.next0_.morph.class0_.is_adjective):
for it in te.next0_.morph.items:
if (it.gender == npt.morph.gender
or (((it.gender) & (npt.morph.gender))) !=
(MorphGender.UNDEFINED)):
if (not (
(it.case_) & npt.morph.case_).is_undefined):
if (it.number == npt.morph.number or
(((it.number) & (npt.morph.number))) !=
(MorphNumber.UNDEFINED)):
var = (te.next0_).term
if (isinstance(it, MorphWordForm)):
var = (it).normal_case
bi = MorphBaseInfo._new549(
MorphClass.ADJECTIVE, npt.morph.gender,
npt.morph.number, npt.morph.language)
var = Morphology.getWordform(var, bi)
if (var is not None):
res = "{0}, {1}".format(res, var)
te = te.next0_.next0_
break
if (te is not None and te.end_char <= end.end_char):
s = ProperNameHelper.getNameEx(te, end,
MorphClass.UNDEFINED,
MorphCase.UNDEFINED,
MorphGender.UNDEFINED, True,
ignore_geo_referent)
if (not Utils.isNullOrEmpty(s)):
if (not str.isalnum(s[0])):
res = "{0}{1}".format(res, s)
else:
res = "{0} {1}".format(res, s)
elif ((isinstance(begin, TextToken))
and begin.chars.is_cyrillic_letter):
mm = begin.getMorphClassInDictionary()
if (not mm.is_undefined):
res = begin.getNormalCaseText(mm, False,
MorphGender.UNDEFINED, False)
if (begin.end_char < end.end_char):
res = "{0} {1}".format(
res,
ProperNameHelper.getNameEx(begin.next0_, end,
MorphClass.UNDEFINED,
MorphCase.UNDEFINED,
MorphGender.UNDEFINED,
True, False))
if (res is None):
res = ProperNameHelper.getNameEx(begin, end, MorphClass.UNDEFINED,
MorphCase.UNDEFINED,
MorphGender.UNDEFINED, True,
ignore_geo_referent)
if (not Utils.isNullOrEmpty(res)):
k = 0
i = len(res) - 1
while i >= 0:
if (res[i] == '*' or Utils.isWhitespace(res[i])):
pass
else:
break
i -= 1
k += 1
if (k > 0):
if (k == len(res)):
return None
res = res[0:0 + len(res) - k]
return res
def tryParseNumberWithPostfix(t: 'Token') -> 'NumberExToken':
""" Выделение стандартных мер, типа: 10 кв.м. """
if (t is None):
return None
t0 = t
is_dollar = None
if (t.length_char == 1 and t.next0_ is not None):
is_dollar = NumberHelper._isMoneyChar(t)
if ((is_dollar) is not None):
t = t.next0_
nt = Utils.asObjectOrNull(t, NumberToken)
if (nt is None):
if ((not ((isinstance(t.previous, NumberToken))) and t.isChar('(')
and (isinstance(t.next0_, NumberToken)))
and t.next0_.next0_ is not None
and t.next0_.next0_.isChar(')')):
toks1 = NumberExHelper._m_postfixes.tryParse(
t.next0_.next0_.next0_, TerminParseAttr.NO)
if (toks1 is not None
and (Utils.valToEnum(toks1.termin.tag, NumberExType))
== NumberExType.MONEY):
nt0 = Utils.asObjectOrNull(t.next0_, NumberToken)
res = NumberExToken._new471(t, toks1.end_token, nt0.value,
nt0.typ, NumberExType.MONEY,
nt0.real_value,
toks1.begin_token.morph)
return NumberExHelper.__correctMoney(
res, toks1.begin_token)
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is None or not tt.morph.class0_.is_adjective):
return None
val = tt.term
i = 4
first_pass2785 = True
while True:
if first_pass2785: first_pass2785 = False
else: i += 1
if (not (i < (len(val) - 5))): break
v = val[0:0 + i]
li = NumberHelper._m_nums.tryAttachStr(v, tt.morph.language)
if (li is None):
continue
vv = val[i:]
lii = NumberExHelper._m_postfixes.tryAttachStr(
vv, tt.morph.language)
if (lii is not None and len(lii) > 0):
re = NumberExToken._new472(
t, t, str((li[0].tag)), NumberSpellingType.WORDS,
Utils.valToEnum(lii[0].tag, NumberExType), t.morph)
NumberExHelper.__correctExtTypes(re)
return re
break
return None
if (t.next0_ is None and is_dollar is None):
return None
f = nt.real_value
t1 = nt.next0_
if (((t1 is not None and t1.isCharOf(",.")))
or (((isinstance(t1, NumberToken)) and
(t1.whitespaces_before_count < 3)))):
tt11 = NumberHelper.tryParseRealNumber(nt, False)
if (tt11 is not None):
t1 = tt11.end_token.next0_
f = tt11.real_value
if (t1 is None):
if (is_dollar is None):
return None
elif ((t1.next0_ is not None and t1.next0_.isValue("С", "З")
and t1.next0_.next0_ is not None)
and t1.next0_.next0_.isValue("ПОЛОВИНА", None)):
f += .5
t1 = t1.next0_.next0_
if (t1 is not None and t1.is_hiphen and t1.next0_ is not None):
t1 = t1.next0_
det = False
altf = f
if (((isinstance(t1, NumberToken)) and t1.previous is not None
and t1.previous.is_hiphen) and (t1).int_value == 0
and t1.length_char == 2):
t1 = t1.next0_
if ((t1 is not None and t1.next0_ is not None and t1.isChar('('))
and (((isinstance(t1.next0_, NumberToken))
or t1.next0_.isValue("НОЛЬ", None)))
and t1.next0_.next0_ is not None):
nt1 = Utils.asObjectOrNull(t1.next0_, NumberToken)
val = 0
if (nt1 is not None):
val = nt1.real_value
if (math.floor(f) == math.floor(val)):
ttt = t1.next0_.next0_
if (ttt.isChar(')')):
t1 = ttt.next0_
det = True
if ((isinstance(t1, NumberToken))
and (t1).int_value is not None
and (t1).int_value == 0):
t1 = t1.next0_
elif (
((((isinstance(ttt, NumberToken)) and
((ttt).real_value < 100) and ttt.next0_ is not None) and
ttt.next0_.isChar('/') and ttt.next0_.next0_ is not None)
and ttt.next0_.next0_.getSourceText() == "100"
and ttt.next0_.next0_.next0_ is not None)
and ttt.next0_.next0_.next0_.isChar(')')):
rest = NumberExHelper.__getDecimalRest100(f)
if ((ttt).int_value is not None
and rest == (ttt).int_value):
t1 = ttt.next0_.next0_.next0_.next0_
det = True
elif ((ttt.isValue("ЦЕЛЫХ", None) and
(isinstance(ttt.next0_, NumberToken))
and ttt.next0_.next0_ is not None)
and ttt.next0_.next0_.next0_ is not None
and ttt.next0_.next0_.next0_.isChar(')')):
num2 = Utils.asObjectOrNull(ttt.next0_, NumberToken)
altf = num2.real_value
if (ttt.next0_.next0_.isValue("ДЕСЯТЫЙ", None)):
altf /= (10)
elif (ttt.next0_.next0_.isValue("СОТЫЙ", None)):
altf /= (100)
elif (ttt.next0_.next0_.isValue("ТЫСЯЧНЫЙ", None)):
altf /= (1000)
elif (ttt.next0_.next0_.isValue("ДЕСЯТИТЫСЯЧНЫЙ", None)):
altf /= (10000)
elif (ttt.next0_.next0_.isValue("СТОТЫСЯЧНЫЙ", None)):
altf /= (100000)
elif (ttt.next0_.next0_.isValue("МИЛЛИОННЫЙ", None)):
altf /= (1000000)
if (altf < 1):
altf += val
t1 = ttt.next0_.next0_.next0_.next0_
det = True
else:
toks1 = NumberExHelper._m_postfixes.tryParse(
ttt, TerminParseAttr.NO)
if (toks1 is not None):
if ((Utils.valToEnum(
toks1.termin.tag,
NumberExType)) == NumberExType.MONEY):
if (toks1.end_token.next0_ is not None
and toks1.end_token.next0_.isChar(')')):
res = NumberExToken._new473(
t, toks1.end_token.next0_, nt.value,
nt.typ, NumberExType.MONEY, f, altf,
toks1.begin_token.morph)
return NumberExHelper.__correctMoney(
res, toks1.begin_token)
res2 = NumberExHelper.tryParseNumberWithPostfix(t1.next0_)
if (res2 is not None and res2.end_token.next0_ is not None
and res2.end_token.next0_.isChar(')')):
if (res2.int_value is not None):
res2.begin_token = t
res2.end_token = res2.end_token.next0_
res2.alt_real_value = res2.real_value
res2.real_value = f
NumberExHelper.__correctExtTypes(res2)
if (res2.whitespaces_after_count < 2):
toks2 = NumberExHelper._m_postfixes.tryParse(
res2.end_token.next0_, TerminParseAttr.NO)
if (toks2 is not None):
if ((Utils.valToEnum(
toks2.termin.tag, NumberExType)
) == NumberExType.MONEY):
res2.end_token = toks2.end_token
return res2
elif (nt1 is not None and nt1.typ == NumberSpellingType.WORDS
and nt.typ == NumberSpellingType.DIGIT):
altf = nt1.real_value
ttt = t1.next0_.next0_
if (ttt.isChar(')')):
t1 = ttt.next0_
det = True
if (not det):
altf = f
if ((t1 is not None and t1.isChar('(') and t1.next0_ is not None)
and t1.next0_.isValue("СУММА", None)):
br = BracketHelper.tryParse(t1, BracketParseAttr.NO, 100)
if (br is not None):
t1 = br.end_token.next0_
if (is_dollar is not None):
te = None
if (t1 is not None):
te = t1.previous
else:
t1 = t0
while t1 is not None:
if (t1.next0_ is None):
te = t1
t1 = t1.next0_
if (te is None):
return None
if (te.is_hiphen and te.next0_ is not None):
if (te.next0_.isValue("МИЛЛИОННЫЙ", None)):
f *= (1000000)
altf *= (1000000)
te = te.next0_
elif (te.next0_.isValue("МИЛЛИАРДНЫЙ", None)):
f *= (1000000000)
altf *= (1000000000)
te = te.next0_
if (not te.is_whitespace_after
and (isinstance(te.next0_, TextToken))):
if (te.next0_.isValue("M", None)):
f *= (1000000)
altf *= (1000000)
te = te.next0_
elif (te.next0_.isValue("BN", None)):
f *= (1000000000)
altf *= (1000000000)
te = te.next0_
return NumberExToken._new474(t0, te, "", nt.typ,
NumberExType.MONEY, f, altf,
is_dollar)
if (t1 is None or ((t1.is_newline_before and not det))):
return None
toks = NumberExHelper._m_postfixes.tryParse(t1, TerminParseAttr.NO)
if ((toks is None and det and (isinstance(t1, NumberToken)))
and (t1).value == "0"):
toks = NumberExHelper._m_postfixes.tryParse(
t1.next0_, TerminParseAttr.NO)
if (toks is not None):
t1 = toks.end_token
if (not t1.isChar('.') and t1.next0_ is not None
and t1.next0_.isChar('.')):
if ((isinstance(t1, TextToken)) and t1.isValue(
toks.termin.terms[0].canonical_text, None)):
pass
elif (not t1.chars.is_letter):
pass
else:
t1 = t1.next0_
if (toks.termin.canonic_text == "LTL"):
return None
if (toks.begin_token == t1):
if (t1.morph.class0_.is_preposition
or t1.morph.class0_.is_conjunction):
if (t1.is_whitespace_before and t1.is_whitespace_after):
return None
ty = Utils.valToEnum(toks.termin.tag, NumberExType)
res = NumberExToken._new473(t, t1, nt.value, nt.typ, ty, f, altf,
toks.begin_token.morph)
if (ty != NumberExType.MONEY):
NumberExHelper.__correctExtTypes(res)
return res
return NumberExHelper.__correctMoney(res, toks.begin_token)
pfx = NumberExHelper.__attachSpecPostfix(t1)
if (pfx is not None):
pfx.begin_token = t
pfx.value = nt.value
pfx.typ = nt.typ
pfx.real_value = f
pfx.alt_real_value = altf
return pfx
if (t1.next0_ is not None and ((t1.morph.class0_.is_preposition
or t1.morph.class0_.is_conjunction))):
if (t1.isValue("НА", None)):
pass
else:
nn = NumberExHelper.tryParseNumberWithPostfix(t1.next0_)
if (nn is not None):
return NumberExToken._new476(t, t, nt.value, nt.typ,
nn.ex_typ, f, altf,
nn.ex_typ2, nn.ex_typ_param)
if (not t1.is_whitespace_after and (isinstance(t1.next0_, NumberToken))
and (isinstance(t1, TextToken))):
term = (t1).term
ty = NumberExType.UNDEFINED
if (term == "СМХ" or term == "CMX"):
ty = NumberExType.SANTIMETER
elif (term == "MX" or term == "МХ"):
ty = NumberExType.METER
elif (term == "MMX" or term == "ММХ"):
ty = NumberExType.MILLIMETER
if (ty != NumberExType.UNDEFINED):
return NumberExToken._new477(t, t1, nt.value, nt.typ, ty, f,
altf, True)
return None
def try_attach(t: 'Token',
prev: typing.List['DateItemToken'],
detail_regime: bool = False) -> 'DateItemToken':
if (t is None):
return None
t0 = t
if (t0.is_char('_')):
t = t.next0_
while t is not None:
if (t.is_newline_before):
return None
if (not t.is_char('_')):
break
t = t.next0_
elif (BracketHelper.can_be_start_of_sequence(t0, True, False)):
ok = False
t = t.next0_
while t is not None:
if (BracketHelper.can_be_end_of_sequence(t, True, t0, False)):
ok = True
break
elif (not t.is_char('_')):
break
t = t.next0_
if (not ok):
t = t0
else:
t = t.next0_
while t is not None:
if (not t.is_char('_')):
break
t = t.next0_
elif ((isinstance(t0, TextToken)) and t0.is_value("THE", None)):
res0 = DateItemToken.__try_attach(t.next0_, prev, detail_regime)
if (res0 is not None):
res0.begin_token = t
return res0
res = DateItemToken.__try_attach(t, prev, detail_regime)
if (res is None):
return None
res.begin_token = t0
if (not res.is_whitespace_after and res.end_token.next0_ is not None
and res.end_token.next0_.is_char('_')):
t = res.end_token.next0_
while t is not None:
if (not t.is_char('_')):
break
else:
res.end_token = t
t = t.next0_
if (res.typ == DateItemToken.DateItemType.YEAR
or res.typ == DateItemToken.DateItemType.CENTURY
or res.typ == DateItemToken.DateItemType.NUMBER):
tok = None
ii = 0
t = res.end_token.next0_
if (t is not None and t.is_value("ДО", None)):
tok = DateItemToken.M_NEW_AGE.try_parse(
t.next0_, TerminParseAttr.NO)
ii = -1
elif (t is not None and t.is_value("ОТ", "ВІД")):
tok = DateItemToken.M_NEW_AGE.try_parse(
t.next0_, TerminParseAttr.NO)
ii = 1
else:
tok = DateItemToken.M_NEW_AGE.try_parse(t, TerminParseAttr.NO)
ii = 1
if (tok is not None):
res.new_age = (-1 if ii < 0 else 1)
res.end_token = tok.end_token
if (res.typ == DateItemToken.DateItemType.NUMBER):
res.typ = DateItemToken.DateItemType.YEAR
return res
def __correctTailAttributes(p : 'PersonReferent', t0 : 'Token') -> 'Token':
res = t0
t = t0
if (t is not None and t.isChar(',')):
t = t.next0_
born = False
die = False
if (t is not None and ((t.isValue("РОДИТЬСЯ", "НАРОДИТИСЯ") or t.isValue("BORN", None)))):
t = t.next0_
born = True
elif (t is not None and ((t.isValue("УМЕРЕТЬ", "ПОМЕРТИ") or t.isValue("СКОНЧАТЬСЯ", None) or t.isValue("DIED", None)))):
t = t.next0_
die = True
elif ((t is not None and t.isValue("ДАТА", None) and t.next0_ is not None) and t.next0_.isValue("РОЖДЕНИЕ", "НАРОДЖЕННЯ")):
t = t.next0_.next0_
born = True
while t is not None:
if (t.morph.class0_.is_preposition or t.is_hiphen or t.isChar(':')):
t = t.next0_
else:
break
if (t is not None and t.getReferent() is not None):
r = t.getReferent()
if (r.type_name == "DATE"):
t1 = t
if (t.next0_ is not None and ((t.next0_.isValue("Р", None) or t.next0_.isValue("РОЖДЕНИЕ", "НАРОДЖЕННЯ")))):
born = True
t1 = t.next0_
if (t1.next0_ is not None and t1.next0_.isChar('.')):
t1 = t1.next0_
if (born):
if (p is not None):
p.addSlot(PersonReferent.ATTR_BORN, r, False, 0)
res = t1
t = t1
elif (die):
if (p is not None):
p.addSlot(PersonReferent.ATTR_DIE, r, False, 0)
res = t1
t = t1
if (die and t is not None):
ag = NumberHelper.tryParseAge(t.next0_)
if (ag is not None):
if (p is not None):
p.addSlot(PersonReferent.ATTR_AGE, str(ag.value), False, 0)
t = ag.end_token.next0_
res = ag.end_token
if (t is None):
return res
if (t.isChar('(')):
br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100)
if (br is not None):
t1 = t.next0_
born = False
if (t1.isValue("РОД", None)):
born = True
t1 = t1.next0_
if (t1 is not None and t1.isChar('.')):
t1 = t1.next0_
if (isinstance(t1, ReferentToken)):
r = t1.getReferent()
if (r.type_name == "DATERANGE" and t1.next0_ == br.end_token):
bd = Utils.asObjectOrNull(r.getSlotValue("FROM"), Referent)
to = Utils.asObjectOrNull(r.getSlotValue("TO"), Referent)
if (bd is not None and to is not None):
if (p is not None):
p.addSlot(PersonReferent.ATTR_BORN, bd, False, 0)
p.addSlot(PersonReferent.ATTR_DIE, to, False, 0)
res = br.end_token
t = res
elif (r.type_name == "DATE" and t1.next0_ == br.end_token):
if (p is not None):
p.addSlot(PersonReferent.ATTR_BORN, r, False, 0)
res = br.end_token
t = res
return res
def __try_parse(t: 'Token', lev: int) -> 'BookLinkToken':
if (t is None or lev > 3):
return None
if (t.is_char('[')):
re = BookLinkToken.__try_parse(t.next0_, lev + 1)
if (re is not None and re.end_token.next0_ is not None
and re.end_token.next0_.is_char(']')):
re.begin_token = t
re.end_token = re.end_token.next0_
return re
if (re is not None and re.end_token.is_char(']')):
re.begin_token = t
return re
if (re is not None):
if (re.typ == BookLinkTyp.SOSTAVITEL
or re.typ == BookLinkTyp.EDITORS):
return re
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
if ((isinstance(br.end_token.previous, NumberToken))
and (br.length_char < 30)):
return BookLinkToken._new329(
t, br.end_token, BookLinkTyp.NUMBER,
MiscHelper.get_text_value(br.begin_token.next0_,
br.end_token.previous,
GetTextAttr.NO))
t0 = t
if (isinstance(t, ReferentToken)):
if (isinstance(t.get_referent(), PersonReferent)):
return BookLinkToken.try_parse_author(
t, FioTemplateType.UNDEFINED)
if (isinstance(t.get_referent(), GeoReferent)):
return BookLinkToken._new326(t, t, BookLinkTyp.GEO,
t.get_referent())
if (isinstance(t.get_referent(), DateReferent)):
dr = Utils.asObjectOrNull(t.get_referent(), DateReferent)
if (len(dr.slots) == 1 and dr.year > 0):
return BookLinkToken._new329(t, t, BookLinkTyp.YEAR,
str(dr.year))
if (dr.year > 0 and t.previous is not None
and t.previous.is_comma):
return BookLinkToken._new329(t, t, BookLinkTyp.YEAR,
str(dr.year))
if (isinstance(t.get_referent(), OrganizationReferent)):
org0_ = Utils.asObjectOrNull(t.get_referent(),
OrganizationReferent)
if (org0_.kind == OrganizationKind.PRESS):
return BookLinkToken._new326(t, t, BookLinkTyp.PRESS,
org0_)
if (isinstance(t.get_referent(), UriReferent)):
uri = Utils.asObjectOrNull(t.get_referent(), UriReferent)
if ((uri.scheme == "http" or uri.scheme == "https"
or uri.scheme == "ftp") or uri.scheme is None):
return BookLinkToken._new326(t, t, BookLinkTyp.URL, uri)
tok_ = BookLinkToken.__m_termins.try_parse(t, TerminParseAttr.NO)
if (tok_ is not None):
typ_ = Utils.valToEnum(tok_.termin.tag, BookLinkTyp)
ok = True
if (typ_ == BookLinkTyp.TYPE or typ_ == BookLinkTyp.NAMETAIL
or typ_ == BookLinkTyp.ELECTRONRES):
if (t.previous is not None and
((t.previous.is_char_of(".:[") or t.previous.is_hiphen))):
pass
else:
ok = False
if (ok):
return BookLinkToken._new329(t, tok_.end_token, typ_,
tok_.termin.canonic_text)
if (typ_ == BookLinkTyp.ELECTRONRES):
tt = tok_.end_token.next0_
first_pass3019 = True
while True:
if first_pass3019: first_pass3019 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if ((isinstance(tt, TextToken))
and not tt.chars.is_letter):
continue
if (isinstance(tt.get_referent(), UriReferent)):
return BookLinkToken._new326(t, tt,
BookLinkTyp.ELECTRONRES,
tt.get_referent())
break
if (t.is_char('/')):
res = BookLinkToken._new329(t, t, BookLinkTyp.DELIMETER, "/")
if (t.next0_ is not None and t.next0_.is_char('/')):
res.end_token = t.next0_
res.value = "//"
if (not t.is_whitespace_before and not t.is_whitespace_after):
coo = 3
no = True
tt = t.next0_
while tt is not None and coo > 0:
vvv = BookLinkToken.try_parse(tt, lev + 1)
if (vvv is not None and vvv.typ != BookLinkTyp.NUMBER):
no = False
break
tt = tt.next0_
coo -= 1
if (no):
return None
return res
if ((isinstance(t, NumberToken)) and t.int_value is not None
and t.typ == NumberSpellingType.DIGIT):
res = BookLinkToken._new329(t, t, BookLinkTyp.NUMBER, str(t.value))
val = t.int_value
if (val >= 1930 and (val < 2030)):
res.typ = BookLinkTyp.YEAR
if (t.next0_ is not None and t.next0_.is_char('.')):
res.end_token = t.next0_
elif ((t.next0_ is not None and t.next0_.length_char == 1
and not t.next0_.chars.is_letter)
and t.next0_.is_whitespace_after):
res.end_token = t.next0_
elif (isinstance(t.next0_, TextToken)):
term = t.next0_.term
if (((term == "СТР" or term == "C" or term == "С")
or term == "P" or term == "S") or term == "PAGES"):
res.end_token = t.next0_
res.typ = BookLinkTyp.PAGES
res.value = str(t.value)
return res
if (isinstance(t, TextToken)):
term = t.term
if (((((
((term == "СТР" or term == "C" or term == "С") or term == "ТОМ"
or term == "T") or term == "Т" or term == "P") or term == "PP"
or term == "V") or term == "VOL" or term == "S")
or term == "СТОР" or t.is_value("PAGE", None))
or t.is_value("СТРАНИЦА", "СТОРІНКА")):
tt = t.next0_
while tt is not None:
if (tt.is_char_of(".:~")):
tt = tt.next0_
else:
break
if (isinstance(tt, NumberToken)):
res = BookLinkToken._new328(t, tt, BookLinkTyp.PAGERANGE)
tt0 = tt
tt1 = tt
tt = tt.next0_
first_pass3020 = True
while True:
if first_pass3020: first_pass3020 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_char_of(",") or tt.is_hiphen):
if (isinstance(tt.next0_, NumberToken)):
tt = tt.next0_
res.end_token = tt
tt1 = tt
continue
break
res.value = MiscHelper.get_text_value(
tt0, tt1, GetTextAttr.NO)
return res
if ((term == "M" or term == "М" or term == "СПБ") or term == "K"
or term == "К"):
if (t.next0_ is not None and t.next0_.is_char_of(":;")):
re = BookLinkToken._new328(t, t.next0_, BookLinkTyp.GEO)
return re
if (t.next0_ is not None and t.next0_.is_char_of(".")):
res = BookLinkToken._new328(t, t.next0_, BookLinkTyp.GEO)
if (t.next0_.next0_ is not None
and t.next0_.next0_.is_char_of(":;")):
res.end_token = t.next0_.next0_
elif (t.next0_.next0_ is not None
and (isinstance(t.next0_.next0_, NumberToken))):
pass
elif (t.next0_.next0_ is not None
and t.next0_.next0_.is_comma and
(isinstance(t.next0_.next0_.next0_, NumberToken))):
pass
else:
return None
return res
if (term == "ПЕР" or term == "ПЕРЕВ" or term == "ПЕРЕВОД"):
tt = t
if (tt.next0_ is not None and tt.next0_.is_char('.')):
tt = tt.next0_
if (tt.next0_ is not None
and ((tt.next0_.is_value("C", None)
or tt.next0_.is_value("С", None)))):
tt = tt.next0_
if (tt.next0_ is None or tt.whitespaces_after_count > 2):
return None
re = BookLinkToken._new328(t, tt.next0_,
BookLinkTyp.TRANSLATE)
return re
if (term == "ТАМ" or term == "ТАМЖЕ"):
res = BookLinkToken._new328(t, t, BookLinkTyp.TAMZE)
if (t.next0_ is not None and t.next0_.is_value("ЖЕ", None)):
res.end_token = t.next0_
return res
if (((term == "СМ" or term == "CM" or term == "НАПР")
or term == "НАПРИМЕР" or term == "SEE") or term == "ПОДРОБНЕЕ"
or term == "ПОДРОБНО"):
res = BookLinkToken._new328(t, t, BookLinkTyp.SEE)
t = t.next0_
first_pass3021 = True
while True:
if first_pass3021: first_pass3021 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_char_of(".:") or t.is_value("ALSO", None)):
res.end_token = t
continue
if (t.is_value("В", None) or t.is_value("IN", None)):
res.end_token = t
continue
vvv = BookLinkToken.__try_parse(t, lev + 1)
if (vvv is not None and vvv.typ == BookLinkTyp.SEE):
res.end_token = vvv.end_token
break
break
return res
if (term == "БОЛЕЕ"):
vvv = BookLinkToken.__try_parse(t.next0_, lev + 1)
if (vvv is not None and vvv.typ == BookLinkTyp.SEE):
vvv.begin_token = t
return vvv
no = MiscHelper.check_number_prefix(t)
if (isinstance(no, NumberToken)):
return BookLinkToken._new328(t, no, BookLinkTyp.N)
if (((term == "B" or term == "В"))
and (isinstance(t.next0_, NumberToken))
and (isinstance(t.next0_.next0_, TextToken))):
term2 = t.next0_.next0_.term
if (((term2 == "Т" or term2 == "T" or term2.startswith("ТОМ"))
or term2 == "TT" or term2 == "ТТ") or term2 == "КН"
or term2.startswith("КНИГ")):
return BookLinkToken._new328(t, t.next0_.next0_,
BookLinkTyp.VOLUME)
if (t.is_char('(')):
if (((isinstance(t.next0_, NumberToken)) and t.next0_.int_value
is not None and t.next0_.next0_ is not None)
and t.next0_.next0_.is_char(')')):
num = t.next0_.int_value
if (num > 1900 and num <= 2040):
if (num <= datetime.datetime.now().year):
return BookLinkToken._new329(t, t.next0_.next0_,
BookLinkTyp.YEAR,
str(num))
if (((isinstance(t.next0_, ReferentToken)) and
(isinstance(t.next0_.get_referent(), DateReferent))
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_char(')')):
num = t.next0_.get_referent().year
if (num > 0):
return BookLinkToken._new329(t, t.next0_.next0_,
BookLinkTyp.YEAR, str(num))
return None