def __calc_newline_between_coef(fr1: 'FragToken', fr2: 'FragToken') -> int:
if (fr1.newlines_after_count > 1):
return 1
tt = fr1.begin_token
while tt is not None and tt.end_char <= fr1.end_char:
if (BracketHelper.can_be_start_of_sequence(tt, False, False)):
br = BracketHelper.try_parse(tt,
BracketParseAttr.CANBEMANYLINES,
100)
if (br is not None and br.end_char >= fr2.begin_char):
return -1
tt = tt.next0_
t = fr1.end_token
if (t.is_char_of(":;.")):
return 1
if ((isinstance(t, TextToken))
and ((t.morph.class0_.is_preposition
or t.morph.class0_.is_conjunction))):
return -1
t1 = fr2.begin_token
if (isinstance(t1, TextToken)):
if (t1.chars.is_all_lower):
return -1
if (BracketHelper.can_be_start_of_sequence(t1, False, False)):
if (t.chars.is_all_lower):
return -1
elif (isinstance(t1, NumberToken)):
if (t.chars.is_all_lower):
return -1
if (t.chars.is_all_lower):
if (fr2.end_token.is_char(';')):
return -1
return 0
def try_attach_city(li: typing.List['CityItemToken'],
ad: 'AnalyzerDataWithOntology',
always: bool = False) -> 'ReferentToken':
if (li is None):
return None
oi = None
if (len(li) > 2 and li[0].typ == CityItemToken.ItemType.MISC
and li[1].typ == CityItemToken.ItemType.NOUN):
li[1].doubtful = False
del li[0]
res = None
if (res is None and len(li) > 1):
res = CityAttachHelper.__try4(li)
if (res is not None and res.end_char <= li[1].end_char):
res = (None)
if (res is None):
wrapoi1125 = RefOutArgWrapper(None)
res = CityAttachHelper.__try1(li, wrapoi1125, ad)
oi = wrapoi1125.value
if (res is None):
wrapoi1126 = RefOutArgWrapper(None)
res = CityAttachHelper.__try_noun_name(li, wrapoi1126, False)
oi = wrapoi1126.value
if (res is None):
wrapoi1127 = RefOutArgWrapper(None)
res = CityAttachHelper.__try_name_exist(li, wrapoi1127, False)
oi = wrapoi1127.value
if (res is None):
res = CityAttachHelper.__try4(li)
if (res is None and always):
wrapoi1128 = RefOutArgWrapper(None)
res = CityAttachHelper.__try_noun_name(li, wrapoi1128, True)
oi = wrapoi1128.value
if (res is None and always):
if (AddressItemToken.try_attach_org(li[0].begin_token)
is not None):
pass
else:
wrapoi1129 = RefOutArgWrapper(None)
res = CityAttachHelper.__try_name_exist(li, wrapoi1129, True)
oi = wrapoi1129.value
if (res is None):
return None
if (res is not None and res.morph is not None):
pass
if (res.begin_token.previous is not None):
if (res.begin_token.previous.is_value("ТЕРРИТОРИЯ", None)):
res.begin_token = res.begin_token.previous
res.morph = res.begin_token.morph
if ((BracketHelper.can_be_start_of_sequence(
res.begin_token.previous, False, False)
and BracketHelper.can_be_end_of_sequence(
res.end_token.next0_, False, None, False)
and res.begin_token.previous.previous is not None)
and res.begin_token.previous.previous.is_value(
"ТЕРРИТОРИЯ", None)):
res.begin_token = res.begin_token.previous.previous
res.morph = res.begin_token.morph
res.end_token = res.end_token.next0_
return res
def process(self, kit : 'AnalysisKit') -> None:
# Основная функция выделения объектов
ad = Utils.asObjectOrNull(kit.get_analyzer_data(self), AnalyzerDataWithOntology)
for k in range(2):
detect_new_denoms = False
dt = datetime.datetime.now()
t = kit.first_token
first_pass3147 = True
while True:
if first_pass3147: first_pass3147 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_whitespace_before):
pass
elif (t.previous is not None and ((t.previous.is_char_of(",") or BracketHelper.can_be_start_of_sequence(t.previous, False, False)))):
pass
else:
continue
rt0 = self.__try_attach_spec(t)
if (rt0 is not None):
rt0.referent = ad.register_referent(rt0.referent)
kit.embed_token(rt0)
t = (rt0)
continue
if (not t.chars.is_letter):
continue
if (not self.__can_be_start_of_denom(t)):
continue
ot = None
ot = ad.local_ontology.try_attach(t, None, False)
if (ot is not None and (isinstance(ot[0].item.referent, DenominationReferent))):
if (self.__check_attach(ot[0].begin_token, ot[0].end_token)):
cl = Utils.asObjectOrNull(ot[0].item.referent.clone(), DenominationReferent)
cl.occurrence.clear()
rt = ReferentToken(cl, ot[0].begin_token, ot[0].end_token)
kit.embed_token(rt)
t = (rt)
continue
if (k > 0):
continue
if (t is not None and t.kit.ontology is not None):
ot = t.kit.ontology.attach_token(DenominationReferent.OBJ_TYPENAME, t)
if ((ot) is not None):
if (self.__check_attach(ot[0].begin_token, ot[0].end_token)):
dr = DenominationReferent()
dr.merge_slots(ot[0].item.referent, True)
rt = ReferentToken(ad.register_referent(dr), ot[0].begin_token, ot[0].end_token)
kit.embed_token(rt)
t = (rt)
continue
rt0 = self.try_attach(t, False)
if (rt0 is not None):
rt0.referent = ad.register_referent(rt0.referent)
kit.embed_token(rt0)
detect_new_denoms = True
t = (rt0)
if (len(ad.local_ontology.items) > 1000):
break
if (not detect_new_denoms):
break
def _add_name(self, begin: 'Token', end: 'Token') -> 'Termin':
if (BracketHelper.can_be_start_of_sequence(begin, True, False)):
br = BracketHelper.try_parse(begin, BracketParseAttr.NO, 100)
if (br is not None and br.end_token == end):
begin = begin.next0_
end = end.previous
val = MiscHelper.get_text_value(
begin, end,
Utils.valToEnum(
(GetTextAttr.KEEPREGISTER) | (GetTextAttr.KEEPQUOTES),
GetTextAttr))
if (val is None):
return None
if (val.endswith(".") and not val.endswith("..")):
val = val[0:0 + len(val) - 1].strip()
self.add_slot(TitlePageReferent.ATTR_NAME, val, False, 0)
return Termin(val.upper())
def __try_attach_pure_terr(li: typing.List['TerrItemToken'],
ad: 'AnalyzerData') -> 'ReferentToken':
aid = None
t = li[0].end_token.next0_
if (t is None):
return None
tt = t
if (BracketHelper.can_be_start_of_sequence(tt, True, False)):
tt = tt.next0_
if (len(li) > 1):
tmp = list(li)
del tmp[0]
rt0 = TerrAttachHelper.try_attach_territory(
tmp, ad, False, None, None)
if (rt0 is None and len(tmp) == 2):
if (((tmp[0].termin_item is None
and tmp[1].termin_item is not None))
or ((tmp[0].termin_item is not None
and tmp[1].termin_item is None))):
if (aid is None):
rt0 = TerrAttachHelper.try_attach_territory(
tmp, ad, True, None, None)
if (rt0 is not None):
if (rt0.referent.is_state):
return None
rt0.begin_token = li[0].begin_token
rt0.morph = li[0].morph
return rt0
if (aid is None):
aid = AddressItemToken.try_attach_org(tt)
if (aid is not None):
rt = aid.create_geo_org_terr()
if (rt is None):
return None
rt.begin_token = li[0].begin_token
t1 = rt.end_token
if (tt != t and BracketHelper.can_be_end_of_sequence(
t1.next0_, False, None, False)):
t1 = t1.next0_
rt.end_token = t1
return rt
return None
def try_parse(t : 'Token', add_units : 'TerminCollection', can_be_set : bool=True, can_units_absent : bool=False, is_resctriction : bool=False, is_subval : bool=False) -> 'MeasureToken':
if (not (isinstance(t, TextToken))):
return None
if (t.is_table_control_char):
return None
t0 = t
whd = None
minmax = 0
wrapminmax1625 = RefOutArgWrapper(minmax)
tt = NumbersWithUnitToken._is_min_or_max(t0, wrapminmax1625)
minmax = wrapminmax1625.value
if (tt is not None):
t = tt.next0_
npt = NounPhraseHelper.try_parse(t, Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) | (NounPhraseParseAttr.IGNOREBRACKETS), NounPhraseParseAttr), 0, None)
if (npt is None):
whd = NumbersWithUnitToken._try_parsewhl(t)
if (whd is not None):
npt = NounPhraseToken(t0, whd.end_token)
elif (t0.is_value("КПД", None)):
npt = NounPhraseToken(t0, t0)
elif ((isinstance(t0, TextToken)) and t0.length_char > 3 and t0.get_morph_class_in_dictionary().is_undefined):
npt = NounPhraseToken(t0, t0)
elif (t0.is_value("T", None) and t0.chars.is_all_lower):
npt = NounPhraseToken(t0, t0)
t = t0
if (t.next0_ is not None and t.next0_.is_char('=')):
npt.end_token = t.next0_
elif ((isinstance(t0, TextToken)) and t0.chars.is_letter and is_subval):
if (NumbersWithUnitToken.try_parse(t, add_units, False, False, False, False) is not None):
return None
npt = NounPhraseToken(t0, t0)
t = t0.next0_
while t is not None:
if (t.whitespaces_before_count > 2):
break
elif (not (isinstance(t, TextToken))):
break
elif (not t.chars.is_letter):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
npt.end_token = t
else:
break
elif (NumbersWithUnitToken.try_parse(t, add_units, False, False, False, False) is not None):
break
else:
npt.end_token = t
t = t.next0_
else:
return None
elif (NumberHelper.try_parse_real_number(t, True, False) is not None):
return None
else:
dtok = DateItemToken.try_attach(t, None, False)
if (dtok is not None):
return None
t1 = npt.end_token
t = npt.end_token
name_ = MetaToken._new509(npt.begin_token, npt.end_token, npt.morph)
units = None
units2 = None
internals_ = list()
not0_ = False
tt = t1.next0_
first_pass3305 = True
while True:
if first_pass3305: first_pass3305 = 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
wrapminmax1617 = RefOutArgWrapper(minmax)
tt2 = NumbersWithUnitToken._is_min_or_max(tt, wrapminmax1617)
minmax = wrapminmax1617.value
if (tt2 is not None):
tt = tt2
t = tt
t1 = t
continue
if ((tt.is_value("БЫТЬ", None) or tt.is_value("ДОЛЖЕН", None) or tt.is_value("ДОЛЖНЫЙ", None)) or tt.is_value("МОЖЕТ", None) or ((tt.is_value("СОСТАВЛЯТЬ", None) and not tt.get_morph_class_in_dictionary().is_adjective))):
t = tt
t1 = t
if (tt.previous.is_value("НЕ", None)):
not0_ = True
continue
www = NumbersWithUnitToken._try_parsewhl(tt)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
continue
if (tt.is_value("ПРИ", None)):
mt1 = MeasureToken.try_parse(tt.next0_, add_units, False, False, True, False)
if (mt1 is not None):
internals_.append(mt1)
tt = mt1.end_token
t = tt
t1 = t
continue
n1 = NumbersWithUnitToken.try_parse(tt.next0_, add_units, False, False, False, False)
if (n1 is not None and len(n1.units) > 0):
mt1 = MeasureToken._new1612(n1.begin_token, n1.end_token, n1)
internals_.append(mt1)
tt = mt1.end_token
t = tt
t1 = t
continue
if (tt.is_value("ПО", None) and tt.next0_ is not None and tt.next0_.is_value("U", None)):
tt = tt.next0_
t = tt
t1 = t
continue
if (len(internals_) > 0):
if (tt.is_char(':')):
break
mt1 = MeasureToken.try_parse(tt.next0_, add_units, False, False, True, 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.try_parse(tt, NounPhraseParseAttr.PARSENUMERICASADJECTIVE, 0, None)
if (npt3 is not None):
tt = npt3.end_token
t1 = tt
if (len(internals_) == 0):
name_.end_token = t1
continue
if (((tt.is_hiphen and not tt.is_whitespace_before and not tt.is_whitespace_after) and (isinstance(tt.next0_, NumberToken)) and (isinstance(tt.previous, TextToken))) and tt.previous.chars.is_all_upper):
t = tt.next0_
tt = t
t1 = tt
if (len(internals_) == 0):
name_.end_token = t1
continue
if (((isinstance(tt, NumberToken)) and not tt.is_whitespace_before and (isinstance(tt.previous, TextToken))) and tt.previous.chars.is_all_upper):
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = t1
continue
if ((((isinstance(tt, NumberToken)) and not tt.is_whitespace_after and tt.next0_.is_hiphen) and not tt.next0_.is_whitespace_after and (isinstance(tt.next0_.next0_, TextToken))) and tt.next0_.next0_.length_char > 2):
tt = tt.next0_.next0_
t = tt
t1 = t
npt1 = NounPhraseHelper.try_parse(tt, NounPhraseParseAttr.NO, 0, None)
if (npt1 is not None and npt1.end_char > tt.end_char):
tt = npt1.end_token
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = t1
continue
if ((isinstance(tt, NumberToken)) and tt.previous is not None):
if (tt.previous.is_value("USB", None)):
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = t1
ttt = tt.next0_
while ttt is not None:
if (ttt.is_whitespace_before):
break
if (ttt.is_char_of(",:")):
break
tt = ttt
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = t1
ttt = ttt.next0_
continue
mt0 = NumbersWithUnitToken.try_parse(tt, add_units, False, False, False, False)
if (mt0 is not None):
npt1 = NounPhraseHelper.try_parse(tt, Utils.valToEnum((NounPhraseParseAttr.PARSENUMERICASADJECTIVE) | (NounPhraseParseAttr.PARSEPREPOSITION), NounPhraseParseAttr), 0, None)
if (npt1 is not None and npt1.end_char > mt0.end_char):
tt = npt1.end_token
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = t1
continue
break
if (((tt.is_comma or tt.is_char('('))) and tt.next0_ is not None):
www = NumbersWithUnitToken._try_parsewhl(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_.is_char(')')):
tt = tt.next0_
t1 = tt
continue
uu = UnitToken.try_parse_list(tt.next0_, add_units, False)
if (uu is not None):
t = uu[len(uu) - 1].end_token
t1 = t
units = uu
if (tt.is_char('(') and t1.next0_ is not None and t1.next0_.is_char(')')):
tt = t1.next0_
t = tt
t1 = t
continue
elif (t1.next0_ is not None and t1.next0_.is_char('(')):
uu = UnitToken.try_parse_list(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_.is_char(')')):
units2 = uu
tt = uu[len(uu) - 1].end_token.next0_
t = tt
t1 = t
continue
www = NumbersWithUnitToken._try_parsewhl(t1.next0_)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
continue
if (uu is not None and len(uu) > 0 and not uu[0].is_doubt):
break
if (t1.next0_ is not None):
if (t1.next0_.is_table_control_char or t1.is_newline_after):
break
units = (None)
if (BracketHelper.can_be_start_of_sequence(tt, False, False) and not (isinstance(tt.next0_, NumberToken))):
br = BracketHelper.try_parse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
t = tt
t1 = t
continue
if (tt.is_value("НЕ", None) and tt.next0_ is not None):
mc = tt.next0_.get_morph_class_in_dictionary()
if (mc.is_adverb or mc.is_misc):
break
continue
if (tt.is_value("ЯМЗ", None)):
pass
npt2 = NounPhraseHelper.try_parse(tt, Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) | (NounPhraseParseAttr.IGNOREBRACKETS) | (NounPhraseParseAttr.PARSEPRONOUNS), NounPhraseParseAttr), 0, None)
if (npt2 is None):
if (tt.morph.class0_.is_preposition or tt.morph.class0_.is_conjunction):
to = NumbersWithUnitToken.M_TERMINS.try_parse(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.get_morph_class_in_dictionary()
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.try_parse_list(tt, add_units, False)
if (uu is not None):
if (uu[0].length_char > 1 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.is_char('.')):
if (not MiscHelper.can_be_start_of_sentence(tt.next0_)):
continue
uu = UnitToken.try_parse_list(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.is_value("ПРЕДЕЛ", None) or t.is_value("ГРАНИЦА", None) or t.is_value("ДИАПАЗОН", None)):
pass
elif (t.chars.is_letter):
name_.end_token = t1
t11 = t1
t1 = t1.next0_
first_pass3306 = True
while True:
if first_pass3306: first_pass3306 = False
else: t1 = t1.next0_
if (not (t1 is not None)): break
if (t1.is_table_control_char):
pass
elif (t1.is_char_of(":,_")):
if (is_resctriction):
return None
www = NumbersWithUnitToken._try_parsewhl(t1.next0_)
if (www is not None):
whd = www
t = www.end_token
t1 = t
continue
uu = UnitToken.try_parse_list(t1.next0_, add_units, False)
if (uu is not None):
if (uu[0].length_char > 1 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
continue
if (t1.is_char(':')):
li = list()
ttt = t1.next0_
first_pass3307 = True
while True:
if first_pass3307: first_pass3307 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.is_hiphen or ttt.is_table_control_char):
continue
if ((isinstance(ttt, TextToken)) and not ttt.chars.is_letter):
continue
mt1 = MeasureToken.try_parse(ttt, add_units, True, True, False, True)
if (mt1 is None):
break
li.append(mt1)
ttt = mt1.end_token
if (ttt.next0_ is not None and ttt.next0_.is_char(';')):
ttt = ttt.next0_
if (ttt.is_char(';')):
pass
elif (ttt.is_newline_after and mt1.is_newline_before):
pass
else:
break
if (len(li) > 1):
res0 = MeasureToken._new1618(t0, li[len(li) - 1].end_token, li, True)
if (internals_ is not None and len(internals_) > 0):
res0.internal_ex = internals_[0]
nam = MiscHelper.get_text_value_of_meta_token(name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
li[0].begin_token = t0
for v in li:
v.name = "{0} ({1})".format(nam, Utils.ifNotNull(v.name, "")).strip()
if (v.nums is not None and len(v.nums.units) == 0 and units is not None):
v.nums.units = units
return res0
elif (t1.is_hiphen and t1.is_whitespace_after and t1.is_whitespace_before):
pass
elif (t1.is_hiphen and t1.next0_ is not None and t1.next0_.is_char('(')):
pass
else:
break
if (t1 is None):
return None
mts = NumbersWithUnitToken.try_parse_multi(t1, add_units, False, not0_, True, is_resctriction)
if (mts is None):
if (units is not None and len(units) > 0):
if (t1 is None or t1.previous.is_char(':')):
mts = list()
if (t1 is None):
t1 = t11
while t1 is not None and t1.next0_ is not None:
pass
t1 = t1.next0_
else:
t1 = t1.previous
mts.append(NumbersWithUnitToken._new1619(t0, t1, math.nan))
if (mts is None):
return None
mt = mts[0]
if (mt.begin_token == mt.end_token and not (isinstance(mt.begin_token, NumberToken))):
return None
if (not is_subval and name_.begin_token.morph.class0_.is_preposition):
name_.begin_token = name_.begin_token.next0_
if (mt.whl is not None):
whd = mt.whl
for kk in range(10):
if (whd is not None and whd.end_token == name_.end_token):
name_.end_token = whd.begin_token.previous
continue
if (units is not None):
if (units[len(units) - 1].end_token == name_.end_token):
name_.end_token = units[0].begin_token.previous
continue
break
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._new1620(t0, mts[len(mts) - 1].end_token, name_.morph, True)
res1.name = MiscHelper.get_text_value_of_meta_token(name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
k = 0
while k < len(mts):
ttt = MeasureToken._new1612(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.is_char('±')):
nn = NumbersWithUnitToken._try_parse(tt1, add_units, True, False, False)
if (nn is not None and nn.plus_minus_percent):
res1.end_token = nn.end_token
res1.nums = nn
if (len(nn.units) > 0 and units is None and len(mt.units) == 0):
for m in mts:
m.units = nn.units
return res1
if (not mt.is_whitespace_before):
if (mt.begin_token.previous is None):
return None
if (mt.begin_token.previous.is_char_of(":),") or mt.begin_token.previous.is_table_control_char or mt.begin_token.previous.is_value("IP", None)):
pass
elif (mt.begin_token.is_hiphen and len(mt.units) > 0 and not mt.units[0].is_doubt):
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.try_parse_list(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._new1622(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.get_text_value_of_meta_token(name_, (GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE if not is_subval else GetTextAttr.NO))
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.__parse_internals(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.try_parse_multi(t1, add_units, False, False, 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.can_be_equals(mts[0].units, mts1[0].units)):
res.is_set = True
res.nums = (None)
res.internals.append(MeasureToken._new1612(mt.begin_token, mt.end_token, mt))
res.internals.append(MeasureToken._new1612(mts1[0].begin_token, mts1[0].end_token, mts1[0]))
res.end_token = mts1[0].end_token
return res
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 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 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 _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 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 __try_parse(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.is_char('(')):
is_bracket_regime = True
blt = BookLinkToken.try_parse(t, 0)
if (blt is None):
blt = BookLinkToken.try_parse_author(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.get_referent()
if (isinstance(r, PersonReferent)):
pass
elif (is_in_lit and r is not None
and r.type_name == "DECREE"):
pass
else:
return None
first_pass3025 = True
while True:
if first_pass3025: first_pass3025 = 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.can_be_start_of_sequence(t, True, False)):
break
if (not t.chars.is_letter):
continue
bbb = BookLinkToken.try_parse(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.get_referent(), BookLinkRefReferent)):
book_prev = tt.get_referent().book
break
tt = tt.previous
max0_ -= 1
blt1 = BookLinkToken.try_parse_author(t, FioTemplateType.UNDEFINED)
if (blt1 is not None and blt1.typ == BookLinkTyp.PERSON):
regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS
else:
ok = False
tt = t
first_pass3026 = True
while True:
if first_pass3026: first_pass3026 = 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.get_referent() is not None
and tt.get_referent().type_name == "DECREE"):
ok = True
decree = tt
break
bbb = BookLinkToken.try_parse(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.try_parse_author(
tt, FioTemplateType.UNDEFINED) is not None):
ok = True
break
bbb = BookLinkToken.try_parse(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.check_link_before(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_pass3027 = True
while True:
if first_pass3027: first_pass3027 = 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.is_char_of(".;") or t.is_comma_and):
continue
if (t.is_char('/')):
break
if ((t.is_char('(') and t.next0_ is not None
and t.next0_.is_value("EDS", None))
and t.next0_.next0_ is not None
and t.next0_.next0_.is_char(')')):
t = t.next0_.next0_.next0_
break
blt = BookLinkToken.try_parse_author(t, prev_pers_templ)
if (blt is None and t.previous is not None
and t.previous.is_and):
blt = BookLinkToken.try_parse_author(
t.previous, FioTemplateType.UNDEFINED)
if (blt is None):
if ((isinstance(t.get_referent(), OrganizationReferent))
and blt00 is not None):
bbb2 = BookLinkToken.try_parse(t.next0_, 0)
if (bbb2 is not None):
if (bbb2.typ == BookLinkTyp.YEAR):
res.add_slot(BookLinkReferent.ATTR_AUTHOR,
t.get_referent(), False, 0)
res.year = int(bbb2.value)
coef += 0.5
t = bbb2.end_token.next0_
break
if (blt.typ == BookLinkTyp.PERSON):
tt2 = blt.end_token.next0_
bbb2 = BookLinkToken.try_parse(tt2, 0)
if (bbb2 is not None):
if (bbb2.typ == BookLinkTyp.YEAR):
res.year = int(bbb2.value)
coef += 0.5
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.is_char('.')))):
tt11 = blt.end_token.next0_
nex = BookLinkToken.try_parse(tt11, 0)
if (nex is not None
and nex.typ == BookLinkTyp.ANDOTHERS):
pass
else:
if (tt11 is None):
break
if (tt11.is_char('/') and tt11.next0_ is not None
and tt11.next0_.is_char('/')):
break
if (tt11.is_char(':')):
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.is_char_of(",.;")
and tt11.next0_ is not None):
tt11 = tt11.next0_
nex = BookLinkToken.try_parse(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.try_parse_author(
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.__add_author(res, blt)
coef += 1
t = blt.end_token
if (isinstance(t.get_referent(), 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 += 0.5
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.find_slot(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.get_slot_value(BookLinkReferent.ATTR_AUTHOR) is None):
return None
tn0 = t
tn1 = None
uri = None
next_num = None
nn = 0
wrapnn376 = RefOutArgWrapper(0)
inoutres377 = Utils.tryParseInt(Utils.ifNotNull(num, ""), wrapnn376)
nn = wrapnn376.value
if (inoutres377):
next_num = str((nn + 1))
br = (BracketHelper.try_parse(
t,
Utils.valToEnum(
(BracketParseAttr.CANCONTAINSVERBS) |
(BracketParseAttr.CANBEMANYLINES), BracketParseAttr), 100) if
BracketHelper.can_be_start_of_sequence(t, True, False) else None)
if (br is not None):
t = t.next0_
pages = None
first_pass3028 = True
while True:
if first_pass3028: first_pass3028 = 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.try_attach(t)
if (tit is not None):
if ((tit.typ == TitleItemToken.Types.TYP and tn0 == t
and br is None)
and BracketHelper.can_be_start_of_sequence(
tit.end_token.next0_, True, False)):
br = BracketHelper.try_parse(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.can_be_start_of_sentence(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.try_parse(
t.previous,
Utils.valToEnum(
((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.PARSEADVERBS) |
(NounPhraseParseAttr.PARSENUMERICASADJECTIVE))
| (NounPhraseParseAttr.MULTILINES),
NounPhraseParseAttr), 0, None)
if (nnn is not None and nnn.end_char >= t.end_char):
pass
else:
break
if (t.is_char_of(".;") and t.whitespaces_after_count > 0):
tit = TitleItemToken.try_attach(t.next0_)
if ((tit) is not None):
if (tit.typ == TitleItemToken.Types.TYP):
break
stop = True
words = 0
notwords = 0
tt = t.next0_
first_pass3029 = True
while True:
if first_pass3029: first_pass3029 = False
else: tt = tt.next0_
if (not (tt is not None)): break
blt0 = BookLinkToken.try_parse(tt, 0)
if (blt0 is None):
if (tt.is_newline_before):
break
if ((isinstance(tt, TextToken)) and not tt.
get_morph_class_in_dictionary().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.try_parse(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.is_char_of(".;")
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.is_char_of(":.")):
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._new372(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.try_parse(t, 0)
if (blt0 is not None):
if (isinstance(blt0.ref, UriReferent)):
uri_re.add_slot(
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._new372(decree.get_referent())
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.is_char_of(",.")
or tt.is_hiphen)):
tt = tt.next0_
blt0 = BookLinkToken.try_parse(tt, 0)
if (blt0 is not None and blt0.typ == BookLinkTyp.PAGERANGE):
rts0 = list()
bref0 = BookLinkRefReferent._new372(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.is_char_of(":,~")):
tn0 = tn0.next0_
else:
break
while tn1 is not None and tn1.begin_char > tn0.begin_char:
if (tn1.is_char_of(".;,:(~") or tn1.is_hiphen
or tn1.is_value("РЕД", None)):
pass
else:
break
tn1 = tn1.previous
nam = MiscHelper.get_text_value(
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.find_slot(BookLinkReferent.ATTR_AUTHOR, None, True) is None
and not BookLinkToken.check_link_before(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_pass3030 = True
while True:
if first_pass3030: first_pass3030 = 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.get_text_value(
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_pass3031 = True
while True:
if first_pass3031: first_pass3031 = 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.can_be_start_of_sequence(t, False, False)):
br = BracketHelper.try_parse(t,
BracketParseAttr.CANBEMANYLINES,
100)
if (br is not None and br.length_char > 300):
br = (None)
blt = BookLinkToken.try_parse(t, 0)
if (t.is_newline_before and not t.is_char('/')
and not t.previous.is_char('/')):
if (blt is not None and blt.typ == BookLinkTyp.NUMBER):
break
if (t.previous.is_char_of(":")):
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.try_parse_author(
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.try_parse(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.try_parse(
t.previous,
Utils.valToEnum(
((NounPhraseParseAttr.MULTILINES) |
(NounPhraseParseAttr.PARSEADVERBS) |
(NounPhraseParseAttr.PARSEPREPOSITION)) |
(NounPhraseParseAttr.PARSEVERBS) |
(NounPhraseParseAttr.PARSEPRONOUNS),
NounPhraseParseAttr), 0, None)
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.is_char_of(".,") or t.is_hiphen):
continue
if (t.is_value("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.try_parse_author(t, prev_pers_templ)
if (blt2 is not None and blt2.typ == BookLinkTyp.PERSON):
prev_pers_templ = blt2.person_template
if (not edits):
BookLinkAnalyzer.__add_author(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 += 0.5
if (isinstance(blt.ref, UriReferent)):
res.add_slot(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 += 0.5
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_.is_char(')')):
coef += (2)
if (res.name is not None
and res.find_slot(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.is_char('.')):
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.check_link_before(t0, num)):
coef += (2)
elif (BookLinkToken.check_link_after(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.try_attach_list(
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.is_value(ln, None)):
del rr.referent.slots[i]
if (not exi):
rr.referent.add_slot(PersonReferent.ATTR_LASTNAME,
pits0[0].value, False, 0)
rts = list()
bref = BookLinkRefReferent._new372(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 __try_parse_thesis(t: 'Token') -> 'ReferentToken':
if (t is None):
return None
t0 = t
tt = t
mc = tt.get_morph_class_in_dictionary()
preamb = None
if (mc.is_conjunction):
return None
if (t.is_value("LET", None)):
return None
if (mc.is_preposition or mc.is_misc or mc.is_adverb):
if (not MiscHelper.is_eng_article(tt)):
tt = tt.next0_
first_pass3131 = True
while True:
if first_pass3131: first_pass3131 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_comma):
break
if (tt.is_char('(')):
br = BracketHelper.try_parse(tt, BracketParseAttr.NO,
100)
if (br is not None):
tt = br.end_token
continue
if (MiscHelper.can_be_start_of_sentence(tt)):
break
npt0 = NounPhraseHelper.try_parse(
tt,
Utils.valToEnum(
(NounPhraseParseAttr.PARSENUMERICASADJECTIVE) |
(NounPhraseParseAttr.REFERENTCANBENOUN),
NounPhraseParseAttr), 0, None)
if (npt0 is not None):
tt = npt0.end_token
continue
if (tt.get_morph_class_in_dictionary().is_verb):
break
if (tt is None or not tt.is_comma or tt.next0_ is None):
return None
preamb = MetaToken(t0, tt.previous)
tt = tt.next0_
t1 = tt
mc = tt.get_morph_class_in_dictionary()
npt = NounPhraseHelper.try_parse(
tt,
Utils.valToEnum((NounPhraseParseAttr.PARSENUMERICASADJECTIVE) |
(NounPhraseParseAttr.REFERENTCANBENOUN) |
(NounPhraseParseAttr.PARSEADVERBS),
NounPhraseParseAttr), 0, None)
if (npt is None and (isinstance(tt, TextToken))):
if (tt.chars.is_all_upper):
npt = NounPhraseToken(tt, tt)
elif (not tt.chars.is_all_lower):
if (mc.is_proper or preamb is not None):
npt = NounPhraseToken(tt, tt)
if (npt is None):
return None
if (mc.is_personal_pronoun):
return None
t2 = npt.end_token.next0_
if (t2 is None or MiscHelper.can_be_start_of_sentence(t2)
or not (isinstance(t2, TextToken))):
return None
if (not t2.get_morph_class_in_dictionary().is_verb):
return None
t3 = t2
tt = t2.next0_
while tt is not None:
if (not tt.get_morph_class_in_dictionary().is_verb):
break
tt = tt.next0_
first_pass3132 = True
while True:
if first_pass3132: first_pass3132 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.next0_ is None):
t3 = tt
break
if (tt.is_char_of(".;!?")):
if (MiscHelper.can_be_start_of_sentence(tt.next0_)):
t3 = tt
break
if (not (isinstance(tt, TextToken))):
continue
if (BracketHelper.can_be_start_of_sequence(tt, False, False)):
br = BracketHelper.try_parse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
continue
tt = t3
if (t3.is_char_of(";.!?")):
tt = tt.previous
txt = MiscHelper.get_text_value(
t2, tt,
Utils.valToEnum(
(GetTextAttr.KEEPREGISTER) | (GetTextAttr.KEEPQUOTES),
GetTextAttr))
if (txt is None or (len(txt) < 15)):
return None
if (t0 != t1):
tt = t1.previous
if (tt.is_comma):
tt = tt.previous
txt0 = MiscHelper.get_text_value(
t0, tt,
Utils.valToEnum(
(GetTextAttr.KEEPREGISTER) | (GetTextAttr.KEEPQUOTES),
GetTextAttr))
if (txt0 is not None and len(txt0) > 10):
if (t0.chars.is_capital_upper):
txt0 = ((str.lower(txt0[0])) + txt0[1:])
txt = "{0}, {1}".format(txt, txt0)
tt = t1
if (MiscHelper.is_eng_article(tt)):
tt = tt.next0_
nam = MiscHelper.get_text_value(tt, t2.previous,
GetTextAttr.KEEPQUOTES)
if (nam.startswith("SO-CALLED")):
nam = nam[9:].strip()
dr = DefinitionReferent()
dr.kind = DefinitionKind.ASSERTATION
dr.add_slot(DefinitionReferent.ATTR_TERMIN, nam, False, 0)
dr.add_slot(DefinitionReferent.ATTR_VALUE, txt, False, 0)
return ReferentToken(dr, t0, t3)
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 __get_name_without_brackets(begin: 'Token',
end: 'Token',
normalize_first_noun_group: bool = False,
normal_first_group_single: bool = False,
ignore_geo_referent: bool = False) -> str:
res = None
if (BracketHelper.can_be_start_of_sequence(begin, False, False)
and BracketHelper.can_be_end_of_sequence(
end, False, begin, False)):
begin = begin.next0_
end = end.previous
if (normalize_first_noun_group
and not begin.morph.class0_.is_preposition):
npt = NounPhraseHelper.try_parse(
begin, NounPhraseParseAttr.REFERENTCANBENOUN, 0, None)
if (npt is not None):
if (npt.noun.get_morph_class_in_dictionary().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.get_normal_case_text(
None, (MorphNumber.SINGULAR if normal_first_group_single
else MorphNumber.UNDEFINED), 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._new492(
MorphClass.ADJECTIVE, npt.morph.gender,
npt.morph.number, npt.morph.language)
var = MorphologyService.get_wordform(
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.get_name_ex(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.get_morph_class_in_dictionary()
if (not mm.is_undefined):
res = begin.get_normal_case_text(mm, MorphNumber.UNDEFINED,
MorphGender.UNDEFINED,
False)
if (begin.end_char < end.end_char):
res = "{0} {1}".format(
res,
ProperNameHelper.get_name_ex(
begin.next0_, end, MorphClass.UNDEFINED,
MorphCase.UNDEFINED, MorphGender.UNDEFINED,
True, False))
if (res is None):
res = ProperNameHelper.get_name_ex(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 get_name_ex(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_pass3064 = True
while True:
if first_pass3064: first_pass3064 = 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.is_bracket(t, False)):
if (t == end):
break
if (t.is_char_of("(<[")):
br = BracketHelper.try_parse(t, BracketParseAttr.NO,
100)
if (br is not None and br.end_char <= end.end_char):
tmp = ProperNameHelper.get_name_ex(
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.get_source_text(), tmp,
br.end_token.get_source_text()),
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.get_source_text()
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.can_be_start_of_sequence(
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.get_referent() is not None)
and t.get_referent().type_name == "GEO"):
continue
s = ProperNameHelper.get_name_ex(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') -> 'TitleItemToken':
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is not None):
t1 = tt
if (tt.term == "ТЕМА"):
tit = TitleItemToken.try_attach(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_.is_char(':')):
t1 = t1.next0_
return TitleItemToken._new2655(
t, t1, TitleItemToken.Types.TYPANDTHEME, tit.value)
if (tt.next0_ is not None and tt.next0_.is_char(':')):
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_.is_value("ТЕМА", None)):
t1 = tt.next0_
if (t1.next0_ is not None and t1.next0_.is_char(':')):
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.is_char('.')):
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.is_char(':')):
t1 = t1.next0_
br = BracketHelper.try_parse(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.is_value("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")):
t1 = tt.next0_
elif (tt.morph.class0_.is_preposition and tt.next0_ is not None
and tt.next0_.is_value("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")):
t1 = tt.next0_.next0_
elif (tt.is_char('/') and tt.is_newline_before):
t1 = tt.next0_
if (t1 is not None):
if (t1.is_char_of(":") or t1.is_hiphen):
t1 = t1.next0_
spec = TitleItemToken.__try_attach_speciality(t1, True)
if (spec is not None):
spec.begin_token = t
return spec
sss = TitleItemToken.__try_attach_speciality(t, False)
if (sss is not None):
return sss
if (isinstance(t, ReferentToken)):
return None
npt = NounPhraseHelper.try_parse(t, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
s = npt.get_normal_case_text(None, MorphNumber.UNDEFINED,
MorphGender.UNDEFINED, False)
tok = TitleItemToken.M_TERMINS.try_parse(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.try_attach(tok.end_token.next0_)
if (tit is not None
and tit.typ == TitleItemToken.Types.THEME):
return TitleItemToken._new2655(
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_pass3394 = True
while True:
if first_pass3394: first_pass3394 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.morph.class0_.is_preposition):
continue
if (ttt.is_value("СОИСКАНИЕ", "")):
continue
npt1 = NounPhraseHelper.try_parse(
ttt, NounPhraseParseAttr.NO, 0, None)
if (npt1 is not None and npt1.noun.is_value(
"СТЕПЕНЬ", "СТУПІНЬ")):
ttt = npt1.end_token
t1 = ttt
continue
rt = t1.kit.process_referent("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.is_value("ДОКТОР", None)
or ttt.is_value("КАНДИДАТ", None)
or ttt.is_value("МАГИСТР", "МАГІСТР")):
t1 = ttt
npt1 = NounPhraseHelper.try_parse(
ttt.next0_, NounPhraseParseAttr.NO, 0,
None)
if (npt1 is not None
and npt1.end_token.is_value(
"НАУК", None)):
t1 = npt1.end_token
s = ("МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ"
if ttt.is_value("МАГИСТР", "МАГІСТР") else
("ДОКТОРСКАЯ ДИССЕРТАЦИЯ" if ttt.is_value(
"ДОКТОР", None) else
"КАНДИДАТСКАЯ ДИССЕРТАЦИЯ"))
break
err += 1
if (err > 3):
break
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
if (s.endswith("ОТЧЕТ") and t1.next0_ is not None
and t1.next0_.is_value("О", None)):
npt1 = NounPhraseHelper.try_parse(
t1.next0_, NounPhraseParseAttr.PARSEPREPOSITION, 0,
None)
if (npt1 is not None
and npt1.morph.case_.is_prepositional):
t1 = npt1.end_token
return TitleItemToken._new2655(npt.begin_token, t1, ty, s)
tok1 = TitleItemToken.M_TERMINS.try_parse(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.can_be_start_of_sequence(t, False, False)):
tok1 = TitleItemToken.M_TERMINS.try_parse(t.next0_,
TerminParseAttr.NO)
if (tok1 is not None and BracketHelper.can_be_end_of_sequence(
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 parse(t : 'Token', max_char : int=0, prev : 'InstrToken'=None) -> 'InstrToken':
from pullenti.ner.instrument.internal.InstrToken1 import InstrToken1
is_start_of_line = False
t00 = t
if (t is not None):
is_start_of_line = t00.is_newline_before
while t is not None:
if (t.is_table_control_char and not t.is_char(chr(0x1F))):
if (t.is_newline_after and not is_start_of_line):
is_start_of_line = True
t = t.next0_
else:
break
if (t is None):
return None
if (t.is_newline_before):
is_start_of_line = True
if (is_start_of_line):
if ((t.is_value("СОДЕРЖИМОЕ", "ВМІСТ") or t.is_value("СОДЕРЖАНИЕ", "ЗМІСТ") or t.is_value("ОГЛАВЛЕНИЕ", "ЗМІСТ")) or ((t.is_value("СПИСОК", None) and t.next0_ is not None and t.next0_.is_value("РАЗДЕЛ", None)))):
cont = InstrToken1.parse(t, True, None, 0, None, False, 0, False, False)
if (cont is not None and cont.typ == InstrToken1.Types.INDEX):
return InstrToken(t, cont.end_token)
t0 = t
t1 = None
has_word = False
first_pass3255 = True
while True:
if first_pass3255: first_pass3255 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_newline_before and t != t0):
break
if (max_char > 0 and t.begin_char > max_char):
break
if (is_start_of_line and t == t0):
if (t.is_value("ГЛАВА", None)):
next0__ = InstrToken.parse(t.next0_, 0, None)
if (next0__ is not None and next0__.typ == ILTypes.PERSON):
next0__.begin_token = t
return next0__
tt = None
if ((isinstance(t.get_referent(), PersonReferent)) or (isinstance(t.get_referent(), PersonPropertyReferent)) or (isinstance(t.get_referent(), InstrumentParticipantReferent))):
return InstrToken.__correct_person(InstrToken._new1511(t00, t, ILTypes.PERSON, t))
is_ref = False
if (isinstance(t.get_referent(), PersonPropertyReferent)):
tt = t.next0_
is_ref = True
elif (prev is not None and prev.typ == ILTypes.PERSON):
rt = t.kit.process_referent(PersonAnalyzer.ANALYZER_NAME, t)
if (rt is not None):
if (isinstance(rt.referent, PersonReferent)):
return InstrToken._new1512(t00, rt.end_token, ILTypes.PERSON)
tt = rt.end_token.next0_
cou = 0
t11 = (None if tt is None else tt.previous)
first_pass3256 = True
while True:
if first_pass3256: first_pass3256 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_table_control_char):
continue
re = tt.get_referent()
if (isinstance(re, PersonReferent)):
return InstrToken._new1511(t00, tt, ILTypes.PERSON, tt)
if (isinstance(re, GeoReferent)):
t11 = tt
continue
if (re is not None):
break
if (DecreeToken.is_keyword(tt, False) is not None):
break
if (tt.is_newline_before):
cou += 1
if (cou > 4):
break
if (tt is None and is_ref):
return InstrToken._new1511(t00, Utils.ifNotNull(t11, t), ILTypes.PERSON, t)
dt = DecreeToken.try_attach(t, None, False)
if (dt is not None):
if (dt.typ == DecreeToken.ItemType.TYP and not t.chars.is_all_lower):
if (t != t0):
break
has_verb_ = False
tt = dt.end_token
while tt is not None:
if (tt.is_newline_before):
break
elif ((isinstance(tt, TextToken)) and tt.is_pure_verb):
has_verb_ = True
break
tt = tt.next0_
if (not has_verb_):
res2 = InstrToken._new1515(t0, dt.end_token, ILTypes.TYP, Utils.ifNotNull(dt.full_value, dt.value))
if (res2.value == "ДОПОЛНИТЕЛЬНОЕ СОГЛАШЕНИЕ" or res2.value == "ДОДАТКОВА УГОДА"):
if (res2.begin_char > 500 and res2.newlines_before_count > 1):
res2.typ = ILTypes.APPENDIX
return res2
if (dt.typ == DecreeToken.ItemType.NUMBER):
if (t != t0):
break
return InstrToken._new1515(t0, dt.end_token, ILTypes.REGNUMBER, dt.value)
if (dt.typ == DecreeToken.ItemType.ORG):
if (t != t0):
break
return InstrToken._new1517(t0, dt.end_token, ILTypes.ORGANIZATION, dt.ref, dt.value)
if (dt.typ == DecreeToken.ItemType.TERR):
if (t != t0):
break
re = InstrToken._new1517(t0, dt.end_token, ILTypes.GEO, dt.ref, dt.value)
t1 = re.end_token.next0_
if (t1 is not None and t1.is_char(',')):
t1 = t1.next0_
if (t1 is not None and t1.is_value("КРЕМЛЬ", None)):
re.end_token = t1
elif ((t1 is not None and t1.is_value("ДОМ", "БУДИНОК") and t1.next0_ is not None) and t1.next0_.is_value("СОВЕТ", "РАД")):
re.end_token = t1.next0_
if (t1.next0_.next0_ is not None and (isinstance(t1.next0_.next0_.get_referent(), GeoReferent))):
re.end_token = t1.next0_.next0_
return re
if (dt.typ == DecreeToken.ItemType.OWNER):
if (t != t0):
break
if (dt.ref is not None and str(dt.ref.referent).startswith("агент")):
dt = (None)
if (dt is not None):
res1 = InstrToken._new1517(t0, dt.end_token, ILTypes.PERSON, dt.ref, dt.value)
return InstrToken.__correct_person(res1)
if (BracketHelper.can_be_start_of_sequence(t, False, False)):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t1 = br.end_token
t = t1
continue
if (t.next0_ is not None and BracketHelper.can_be_end_of_sequence(t.next0_, False, None, False)):
t1 = t.next0_
t = t1
continue
if (isinstance(t, TextToken)):
if (t.is_char('_')):
t1 = t
continue
r = t.get_referent()
if (isinstance(r, DateReferent)):
tt = t
if (tt.next0_ is not None and tt.next0_.is_char_of(",;")):
tt = tt.next0_
if (not t.is_newline_before and not tt.is_newline_after):
t1 = tt
continue
if (not has_word):
return InstrToken._new1511(t, tt, ILTypes.DATE, t)
if (t != t0):
break
has_word = True
if (isinstance(r, InstrumentParticipantReferent)):
tt = t.begin_token
first_pass3257 = True
while True:
if first_pass3257: first_pass3257 = False
else: tt = tt.next0_
if (not (tt is not None and (tt.end_char < t.end_char))): break
rr = tt.get_referent()
if (rr is None):
continue
if ((isinstance(rr, OrganizationReferent)) or (isinstance(rr, BankDataReferent)) or (isinstance(rr, UriReferent))):
r = (None)
break
if ((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent)) or (isinstance(r, InstrumentParticipantReferent))):
if (t != t0):
break
if (isinstance(r, InstrumentParticipantReferent)):
pass
res1 = InstrToken._new1511(t, t, ILTypes.PERSON, t)
return InstrToken.__correct_person(res1)
if (isinstance(r, OrganizationReferent)):
if (t != t0):
break
return InstrToken._new1511(t, t, ILTypes.ORGANIZATION, t)
if (isinstance(r, DecreePartReferent)):
dpr = Utils.asObjectOrNull(r, DecreePartReferent)
if (dpr.appendix is not None):
if (t.is_newline_before or is_start_of_line):
if (t.is_newline_after or t.whitespaces_before_count > 30):
return InstrToken._new1515(t, t, ILTypes.APPENDIX, "ПРИЛОЖЕНИЕ")
ok = True
tt = t.next0_
first_pass3258 = True
while True:
if first_pass3258: first_pass3258 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
break
npt = NounPhraseHelper.try_parse(tt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
tt = npt.end_token
continue
ok = False
break
if (ok):
return InstrToken._new1515(t, t, ILTypes.APPENDIX, "ПРИЛОЖЕНИЕ")
if ((isinstance(r, DecreeReferent)) and r.kind == DecreeKind.PUBLISHER and t == t0):
res1 = InstrToken._new1512(t, t, ILTypes.APPROVED)
tt = t.next0_
first_pass3259 = True
while True:
if first_pass3259: first_pass3259 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_char_of(",;")):
continue
if ((isinstance(tt.get_referent(), DecreeReferent)) and tt.get_referent().kind == DecreeKind.PUBLISHER):
res1.end_token = t
else:
break
return res1
if (t.is_value("ЗА", None) and t.next0_ is not None and t.is_newline_before):
rr = t.next0_.get_referent()
if ((isinstance(rr, PersonReferent)) or (isinstance(rr, PersonPropertyReferent)) or (isinstance(rr, InstrumentParticipantReferent))):
if (t != t0):
break
res1 = InstrToken._new1511(t, t.next0_, ILTypes.PERSON, t.next0_)
t = t.next0_.next0_
if ((isinstance(rr, InstrumentParticipantReferent)) and t is not None):
r = t.get_referent()
if ((r) is not None):
if ((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent))):
res1.end_token = t
res1.ref = (t)
return res1
ii = 0
while ii < len(InstrToken._m_directives):
if (t.is_value(InstrToken._m_directives[ii], None)):
if (t.next0_ is not None and t.next0_.is_value("СЛЕДУЮЩЕЕ", "НАСТУПНЕ")):
if (t != t0):
break
t11 = t.next0_
ok = False
if (t11.next0_ is not None and t11.next0_.is_char_of(":.") and t11.next0_.is_newline_after):
ok = True
t11 = t11.next0_
if (ok):
return InstrToken._new1515(t, t11, ILTypes.DIRECTIVE, InstrToken._m_directives_norm[ii])
if (t.is_newline_after or ((t.next0_ is not None and t.next0_.is_char(':') and t.next0_.is_newline_after))):
if (t != t0):
break
if (not t.is_newline_before):
if ((InstrToken._m_directives_norm[ii] != "ПРИКАЗ" and InstrToken._m_directives_norm[ii] != "ПОСТАНОВЛЕНИЕ" and InstrToken._m_directives_norm[ii] != "НАКАЗ") and InstrToken._m_directives_norm[ii] != "ПОСТАНОВУ"):
break
return InstrToken._new1515(t, (t if t.is_newline_after else t.next0_), ILTypes.DIRECTIVE, InstrToken._m_directives_norm[ii])
break
ii += 1
if (t.is_newline_before and t.chars.is_letter and t.length_char == 1):
for d in InstrToken._m_directives:
t11 = MiscHelper.try_attach_word_by_letters(d, t, True)
if (t11 is not None):
if (t11.next0_ is not None and t11.next0_.is_char(':')):
t11 = t11.next0_
return InstrToken._new1512(t, t11, ILTypes.DIRECTIVE)
tte = (t.begin_token if isinstance(t, MetaToken) else t)
term = (tte.term if isinstance(tte, TextToken) else None)
if (is_start_of_line and not tte.chars.is_all_lower and t == t0):
npt = NounPhraseHelper.try_parse(tte, NounPhraseParseAttr.NO, 0, None)
if (npt is not None and ((term == "ПРИЛОЖЕНИЯ" or term == "ДОДАТКИ"))):
# if (tte.Next != null && tte.Next.IsChar(':'))
npt = (None)
if (npt is not None and npt.morph.case_.is_nominative and (isinstance(npt.end_token, TextToken))):
term1 = npt.end_token.term
if (((term1 == "ПРИЛОЖЕНИЕ" or term1 == "ДОДАТОК" or term1 == "МНЕНИЕ") or term1 == "ДУМКА" or term1 == "АКТ") or term1 == "ФОРМА" or term == "ЗАЯВКА"):
tt1 = npt.end_token.next0_
dt1 = DecreeToken.try_attach(tt1, None, False)
if (dt1 is not None and dt1.typ == DecreeToken.ItemType.NUMBER):
tt1 = dt1.end_token.next0_
elif (isinstance(tt1, NumberToken)):
tt1 = tt1.next0_
elif ((isinstance(tt1, TextToken)) and tt1.length_char == 1 and tt1.chars.is_letter):
tt1 = tt1.next0_
ok = True
if (tt1 is None):
ok = False
elif (tt1.is_value("В", "У")):
ok = False
elif (tt1.is_value("К", None) and tt1.is_newline_before):
return InstrToken._new1515(t, t, ILTypes.APPENDIX, term1)
elif (not tt1.is_newline_before and InstrToken._check_entered(tt1) is not None):
ok = False
elif (tt1 == t.next0_ and ((tt1.is_char(':') or ((tt1.is_value("НА", None) and term1 != "ЗАЯВКА"))))):
ok = False
if (ok):
br = BracketHelper.try_parse(tt1, BracketParseAttr.NO, 100)
if (br is not None):
tt1 = br.end_token.next0_
if (br.end_token.next0_ is None or not br.end_token.is_newline_after or br.end_token.next0_.is_char_of(";,")):
ok = False
if (tt1 is not None and tt1.is_value("ПРИЛОЖЕНИЕ", "ДОДАТОК")):
ok = False
if (prev is not None and prev.typ == ILTypes.APPENDIX):
ok = False
if (ok):
cou = 0
ttt = tte.previous
while ttt is not None and (cou < 300):
if (ttt.is_table_control_char):
if (not ttt.is_char(chr(0x1F))):
if (ttt == tte.previous and ttt.is_char(chr(0x1E))):
pass
else:
ok = False
break
ttt = ttt.previous; cou += 1
if (ok):
it1 = InstrToken1.parse(t, True, None, 0, None, False, 0, False, False)
if (it1 is not None):
if (it1.has_verb):
ok = False
if (ok and t.previous is not None):
ttp = t.previous
first_pass3260 = True
while True:
if first_pass3260: first_pass3260 = False
else: ttp = ttp.previous
if (not (ttp is not None)): break
if (ttp.is_table_control_char and not ttp.is_char(chr(0x1F))):
continue
if (BracketHelper.is_bracket(ttp, False) and not BracketHelper.can_be_end_of_sequence(ttp, False, None, False)):
continue
if (ttp.is_char_of(";:")):
ok = False
break
if ((ok and t.previous is not None and (t.newlines_before_count < 3)) and not t.is_newline_after):
lines = 0
ttp = t.previous
first_pass3261 = True
while True:
if first_pass3261: first_pass3261 = False
else: ttp = ttp.previous
if (not (ttp is not None)): break
if (not ttp.is_newline_before):
continue
while ttp is not None and (ttp.end_char < t.begin_char):
if (isinstance(ttp, NumberToken)):
pass
elif ((isinstance(ttp, TextToken)) and ttp.length_char > 1):
if (ttp.is_value("ПРИЛОЖЕНИЕ", "ДОДАТОК")):
ok = False
break
else:
break
ttp = ttp.next0_
lines += 1
if (lines > 1):
break
if (ok and ((term1 != "ПРИЛОЖЕНИЕ" and term1 != "ДОДАТОК" and term1 != "МНЕНИЕ"))):
if (t.newlines_before_count < 3):
ok = False
if (ok):
return InstrToken._new1515(t, t, ILTypes.APPENDIX, term1)
app = False
if ((((term == "ОСОБОЕ" or term == "ОСОБЛИВЕ")) and t.next0_ is not None and t.next0_.is_value("МНЕНИЕ", "ДУМКА")) and t == t0 and is_start_of_line):
app = True
if ((((term == "ДОПОЛНИТЕЛЬНОЕ" or term == "ДОДАТКОВА")) and t.next0_ is not None and t.next0_.is_value("СОГЛАШЕНИЕ", "УГОДА")) and t == t0 and is_start_of_line):
app = True
if (app):
tt = t.next0_
while tt is not None:
if (tt.is_newline_before):
break
elif (tt.get_morph_class_in_dictionary() == MorphClass.VERB):
app = False
break
tt = tt.next0_
if (app):
return InstrToken._new1512(t, t.next0_, ILTypes.APPENDIX)
if (not t.chars.is_all_lower and t == t0):
tt = InstrToken._check_approved(t)
if (tt is not None):
if (tt.next0_ is not None and (isinstance(tt.next0_.get_referent(), DecreeReferent))):
return InstrToken._new1511(t, tt, ILTypes.APPROVED, tt.next0_.get_referent())
dt1 = DecreeToken.try_attach(tt.next0_, None, False)
if (dt1 is not None and dt1.typ == DecreeToken.ItemType.TYP):
return InstrToken._new1512(t, tt, ILTypes.APPROVED)
t1 = t
is_start_of_line = False
if (t1 is None):
return None
res = InstrToken._new1512(t00, t1, ILTypes.UNDEFINED)
res.no_words = True
t = t0
first_pass3262 = True
while True:
if first_pass3262: first_pass3262 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= t1.end_char)): break
if (not (isinstance(t, TextToken))):
if (isinstance(t, ReferentToken)):
res.no_words = False
continue
if (not t.chars.is_letter):
continue
res.no_words = False
if (t.is_pure_verb):
res.has_verb = True
if (t0.is_value("ВОПРОС", "ПИТАННЯ") and t0.next0_ is not None and t0.next0_.is_char_of(":.")):
res.typ = ILTypes.QUESTION
return 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 try_create_canonic_decree_ref_uri(t: 'Token') -> 'CanonicDecreeRefUri':
if (not (isinstance(t, ReferentToken))):
return None
dr = Utils.asObjectOrNull(t.get_referent(), DecreeReferent)
res = None
if (dr is not None):
if (dr.kind == DecreeKind.PUBLISHER):
return None
res = CanonicDecreeRefUri._new833(t.kit.sofa.text, dr,
t.begin_char, t.end_char)
if ((t.previous is not None and t.previous.is_char('(')
and t.next0_ is not None) and t.next0_.is_char(')')):
return res
if (t.misc_attrs != 0):
return res
rt = Utils.asObjectOrNull(t, ReferentToken)
if (rt.begin_token.is_char('(') and rt.end_token.is_char(')')):
res = CanonicDecreeRefUri._new833(
t.kit.sofa.text, dr, rt.begin_token.next0_.begin_char,
rt.end_token.previous.end_char)
return res
next_decree_items = None
if ((t.next0_ is not None and t.next0_.is_comma_and and
(isinstance(t.next0_.next0_, ReferentToken))) and
(isinstance(t.next0_.next0_.get_referent(), DecreeReferent))):
next_decree_items = DecreeToken.try_attach_list(
t.next0_.next0_.begin_token, None, 10, False)
if (next_decree_items is not None
and len(next_decree_items) > 1):
i = 0
while i < (len(next_decree_items) - 1):
if (next_decree_items[i].is_newline_after):
del next_decree_items[i + 1:i + 1 +
len(next_decree_items) - i -
1]
break
i += 1
was_typ = False
was_num = False
tt = t.begin_token
first_pass3090 = True
while True:
if first_pass3090: first_pass3090 = False
else: tt = tt.next0_
if (not (tt is not None and tt.end_char <= t.end_char)): break
if (tt.begin_char == t.begin_char and tt.is_char('(')
and tt.next0_ is not None):
res.begin_char = tt.next0_.begin_char
if (tt.is_char('(') and tt.next0_ is not None
and tt.next0_.is_value("ДАЛЕЕ", None)):
if (res.end_char >= tt.begin_char):
res.end_char = tt.previous.end_char
break
if (tt.end_char == t.end_char and tt.is_char(')')):
res.end_char = tt.previous.end_char
tt1 = tt.previous
while tt1 is not None and tt1.begin_char >= res.begin_char:
if (tt1.is_char('(') and tt1.previous is not None):
if (res.begin_char < tt1.previous.begin_char):
res.end_char = tt1.previous.end_char
tt1 = tt1.previous
li = DecreeToken.try_attach_list(tt, None, 10, False)
if (li is not None and len(li) > 0):
ii = 0
while ii < (len(li) - 1):
if (li[ii].typ == DecreeToken.ItemType.TYP and
li[ii + 1].typ == DecreeToken.ItemType.TERR):
res.type_with_geo = MiscHelper.get_text_value(
li[ii].begin_token, li[ii + 1].end_token,
GetTextAttr.FIRSTNOUNGROUPTONOMINATIVESINGLE)
ii += 1
if ((next_decree_items is not None
and len(next_decree_items) > 1 and
(len(next_decree_items) < len(li)))
and next_decree_items[0].typ !=
DecreeToken.ItemType.TYP):
d = len(li) - len(next_decree_items)
j = 0
j = 0
while j < len(next_decree_items):
if (next_decree_items[j].typ != li[d + j].typ):
break
j += 1
if (j >= len(next_decree_items)):
del li[0:0 + d]
res.begin_char = li[0].begin_char
elif (
(next_decree_items is not None
and len(next_decree_items) == 1 and
next_decree_items[0].typ == DecreeToken.ItemType.NAME)
and len(li) == 2
and li[1].typ == DecreeToken.ItemType.NAME):
res.begin_char = li[1].begin_char
res.end_char = li[1].end_char
break
elif ((next_decree_items is not None
and len(next_decree_items) == 1
and next_decree_items[0].typ
== DecreeToken.ItemType.NUMBER) and
li[len(li) - 1].typ == DecreeToken.ItemType.NUMBER):
res.begin_char = li[len(li) - 1].begin_char
res.end_char = li[len(li) - 1].end_char
i = 0
first_pass3091 = True
while True:
if first_pass3091: first_pass3091 = False
else: i += 1
if (not (i < len(li))): break
l_ = li[i]
if (l_.begin_char > t.end_char):
del li[i:i + len(li) - i]
break
if (l_.typ == DecreeToken.ItemType.NAME):
if (not was_num):
if (dr.kind == DecreeKind.CONTRACT):
continue
if (((i + 1) < len(li)) and (
(li[i + 1].typ == DecreeToken.ItemType.DATE
or li[i + 1].typ
== DecreeToken.ItemType.NUMBER))):
continue
ee = l_.begin_token.previous.end_char
if (ee > res.begin_char and (ee < res.end_char)):
res.end_char = ee
break
if (l_.typ == DecreeToken.ItemType.NUMBER):
was_num = True
if (i == 0):
if (l_.typ == DecreeToken.ItemType.TYP):
was_typ = True
elif (l_.typ == DecreeToken.ItemType.OWNER
or l_.typ == DecreeToken.ItemType.ORG):
if (((i + 1) < len(li)) and
((li[1].typ == DecreeToken.ItemType.DATE
or li[1].typ
== DecreeToken.ItemType.NUMBER))):
was_typ = True
if (was_typ):
tt0 = l_.begin_token.previous
if (tt0 is not None and tt0.is_char('.')):
tt0 = tt0.previous
if (tt0 is not None and
((tt0.is_value("УТВЕРЖДЕННЫЙ", None)
or tt0.is_value("УТВЕРДИТЬ", None)
or tt0.is_value("УТВ", None)))):
if (l_.begin_char > res.begin_char):
res.begin_char = l_.begin_char
if (res.end_char < res.begin_char):
res.end_char = t.end_char
res.is_adopted = True
if (len(li) > 0):
tt = li[len(li) - 1].end_token
if (tt.is_char(')')):
tt = tt.previous
continue
if (was_typ):
na = DecreeToken.try_attach_name(tt, dr.typ0, True, False)
if (na is not None and tt.begin_char > t.begin_char):
tt1 = na.end_token.next0_
if (tt1 is not None and tt1.is_char_of(",()")):
tt1 = tt1.next0_
if (tt1 is not None and (tt1.end_char < t.end_char)):
if (tt1.is_value("УТВЕРЖДЕННЫЙ", None)
or tt1.is_value("УТВЕРДИТЬ", None)
or tt1.is_value("УТВ", None)):
tt = tt1
continue
if (tt.previous is not None
and tt.previous.is_char(':')
and na.end_char <= res.end_char):
res.begin_char = tt.begin_char
break
if (tt.previous.end_char > res.begin_char):
res.end_char = tt.previous.end_char
break
return res
dpr = Utils.asObjectOrNull(t.get_referent(), DecreePartReferent)
if (dpr is None):
return None
if ((t.previous is not None and t.previous.is_hiphen and
(isinstance(t.previous.previous, ReferentToken))) and (isinstance(
t.previous.previous.get_referent(), DecreePartReferent))):
if (DecreePartReferent.create_range_referent(
Utils.asObjectOrNull(t.previous.previous.get_referent(),
DecreePartReferent), dpr)
is not None):
return None
t1 = t
has_diap = False
diap_ref = None
if ((t.next0_ is not None and t.next0_.is_hiphen and
(isinstance(t.next0_.next0_, ReferentToken))) and
(isinstance(t.next0_.next0_.get_referent(), DecreePartReferent))):
diap = DecreePartReferent.create_range_referent(
Utils.asObjectOrNull(dpr, DecreePartReferent),
Utils.asObjectOrNull(t.next0_.next0_.get_referent(),
DecreePartReferent))
if (diap is not None):
dpr = diap
has_diap = True
t1 = t.next0_.next0_
diap_ref = (Utils.asObjectOrNull(t1, ReferentToken))
res = CanonicDecreeRefUri._new835(t.kit.sofa.text, dpr, t.begin_char,
t1.end_char, has_diap)
if ((t.previous is not None and t.previous.is_char('(')
and t1.next0_ is not None) and t1.next0_.is_char(')')):
return res
tt = t.begin_token
while tt is not None and tt.end_char <= t.end_char:
if (isinstance(tt.get_referent(), DecreeReferent)):
if (tt.begin_char > t.begin_char):
res.end_char = tt.previous.end_char
if (tt.previous.morph.class0_.is_preposition
and tt.previous.previous is not None):
res.end_char = tt.previous.previous.end_char
elif (tt.end_char < t.end_char):
res.begin_char = tt.begin_char
break
tt = tt.next0_
has_same_before = DecreeHelper.__has_same_decree(t, dpr, True)
has_same_after = DecreeHelper.__has_same_decree(t, dpr, False)
ptmin = PartToken.ItemType.PREFIX
ptmin2 = PartToken.ItemType.PREFIX
max0_ = 0
max2 = 0
for s in dpr.slots:
pt = PartToken._get_type_by_attr_name(s.type_name)
if (pt == PartToken.ItemType.PREFIX):
continue
co = PartToken._get_rank(pt)
if (co < 1):
if (pt == PartToken.ItemType.PART
and dpr.find_slot(DecreePartReferent.ATTR_CLAUSE, None,
True) is not None):
co = PartToken._get_rank(PartToken.ItemType.PARAGRAPH)
else:
continue
if (co > max0_):
max2 = max0_
ptmin2 = ptmin
max0_ = co
ptmin = pt
elif (co > max2):
max2 = co
ptmin2 = pt
if (ptmin != PartToken.ItemType.PREFIX):
tt = t.begin_token
while tt is not None and tt.end_char <= res.end_char:
if (tt.begin_char >= res.begin_char):
pt = PartToken.try_attach(tt, None, False, False)
if (pt is not None and pt.typ == ptmin):
res.begin_char = pt.begin_char
res.end_char = pt.end_char
if (pt.typ == PartToken.ItemType.APPENDIX
and pt.end_token.is_value("К", None)
and pt.begin_token != pt.end_token):
res.end_char = pt.end_token.previous.end_char
if (pt.end_char == t.end_char):
if ((t.next0_ is not None and t.next0_.is_comma_and
and
(isinstance(t.next0_.next0_, ReferentToken)))
and
(isinstance(t.next0_.next0_.get_referent(),
DecreePartReferent))):
tt1 = t.next0_.next0_.begin_token
ok = True
if (tt1.chars.is_letter):
ok = False
if (ok):
for v in pt.values:
res.begin_char = v.begin_char
res.end_char = v.end_char
break
if (not has_diap):
return res
break
tt = tt.next0_
if (has_diap and diap_ref is not None):
tt = diap_ref.begin_token
while tt is not None and tt.end_char <= diap_ref.end_char:
if (tt.is_char(',')):
break
if (tt != diap_ref.begin_token
and tt.is_whitespace_before):
break
res.end_char = tt.end_char
tt = tt.next0_
return res
if (((has_same_before or has_same_after))
and ptmin != PartToken.ItemType.PREFIX):
tt = t.begin_token
first_pass3092 = True
while True:
if first_pass3092: first_pass3092 = False
else: tt = tt.next0_
if (not (tt is not None and tt.end_char <= res.end_char)):
break
if (tt.begin_char >= res.begin_char):
pt = (PartToken.try_attach(tt, None, False, False)
if not has_same_before else None)
if (pt is not None):
if (pt.typ == ptmin):
for v in pt.values:
res.begin_char = v.begin_char
res.end_char = v.end_char
return res
tt = pt.end_token
continue
if ((isinstance(tt, NumberToken))
and tt.begin_char == res.begin_char):
res.end_char = tt.end_char
while tt is not None and tt.next0_ is not None:
if (not tt.next0_.is_char('.')
or tt.is_whitespace_after
or tt.next0_.is_whitespace_after):
break
if (not (isinstance(tt.next0_.next0_,
NumberToken))):
break
tt = tt.next0_.next0_
res.end_char = tt.end_char
if (tt.next0_ is not None and tt.next0_.is_hiphen):
if (isinstance(tt.next0_.next0_, NumberToken)):
tt = tt.next0_.next0_
res.end_char = tt.end_char
while tt is not None and tt.next0_ is not None:
if (not tt.next0_.is_char('.')
or tt.is_whitespace_after
or tt.next0_.is_whitespace_after):
break
if (not (isinstance(
tt.next0_.next0_, NumberToken))):
break
tt = tt.next0_.next0_
res.end_char = tt.end_char
elif (tt.next0_.next0_ is not None and
(isinstance(tt.next0_.next0_.get_referent(),
DecreePartReferent))
and has_diap):
res.end_char = tt.next0_.next0_.begin_token.end_char
return res
if (BracketHelper.can_be_start_of_sequence(
tt, True, False)
and tt.begin_char == res.begin_char
and has_same_before):
br = BracketHelper.try_parse(tt, BracketParseAttr.NO,
100)
if (br is not None
and br.end_token.previous == tt.next0_):
res.end_char = br.end_char
return res
return res
if (not has_same_before and not has_same_after
and ptmin != PartToken.ItemType.PREFIX):
tt = t.begin_token
while tt is not None and tt.end_char <= res.end_char:
if (tt.begin_char >= res.begin_char):
pts = PartToken.try_attach_list(tt, False, 40)
if (pts is None or len(pts) == 0):
break
i = 0
while i < len(pts):
if (pts[i].typ == ptmin):
res.begin_char = pts[i].begin_char
res.end_char = pts[i].end_char
tt = pts[i].end_token
if (tt.next0_ is not None and tt.next0_.is_hiphen):
if (isinstance(tt.next0_.next0_, NumberToken)):
res.end_char = tt.next0_.next0_.end_char
elif (tt.next0_.next0_ is not None
and (isinstance(
tt.next0_.next0_.get_referent(),
DecreePartReferent)) and has_diap):
res.end_char = tt.next0_.next0_.begin_token.end_char
return res
i += 1
tt = tt.next0_
return res