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_parse(t : 'Token', prev : 'WeaponItemToken', after_conj : bool, attach_high : bool=False) -> 'WeaponItemToken':
res = WeaponItemToken.__try_parse(t, prev, after_conj, attach_high)
if (res is None):
npt = NounPhraseHelper.try_parse(t, NounPhraseParseAttr.NO, 0, None)
if (npt is not None and npt.noun.begin_char > npt.begin_char):
res = WeaponItemToken.__try_parse(npt.noun.begin_token, prev, after_conj, attach_high)
if (res is not None):
if (res.typ == WeaponItemToken.Typs.NOUN):
str0_ = npt.get_normal_case_text(None, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
if (str0_ == "РУЧНОЙ ГРАНАТ"):
str0_ = "РУЧНАЯ ГРАНАТА"
if ((Utils.ifNotNull(str0_, "")).endswith(res.value)):
if (res.alt_value is None):
res.alt_value = str0_
else:
str0_ = str0_[0:0+len(str0_) - len(res.value)].strip()
res.alt_value = "{0} {1}".format(str0_, res.alt_value)
res.begin_token = t
return res
return None
if (res.typ == WeaponItemToken.Typs.NAME):
br = BracketHelper.try_parse(res.end_token.next0_, BracketParseAttr.NO, 100)
if (br is not None and br.is_char('(')):
alt = MiscHelper.get_text_value_of_meta_token(br, GetTextAttr.NO)
if (MiscHelper.can_be_equal_cyr_and_latss(res.value, alt)):
res.alt_value = alt
res.end_token = br.end_token
return res
def __try_attach_speciality(t: 'Token',
key_word_before: bool) -> 'TitleItemToken':
if (t is None):
return None
susp = False
if (not key_word_before):
if (not t.is_newline_before):
susp = True
val = None
t0 = t
dig_count = 0
for i in range(3):
nt = Utils.asObjectOrNull(t, NumberToken)
if (nt is None):
break
if (nt.typ != NumberSpellingType.DIGIT
or nt.morph.class0_.is_adjective):
break
if (val is None):
val = io.StringIO()
if (susp and t.length_char != 2):
return None
digs = nt.get_source_text()
dig_count += len(digs)
print(digs, end="", file=val)
if (t.next0_ is None):
break
t = t.next0_
if (t.is_char_of(".,") or t.is_hiphen):
if (susp and (i < 2)):
if (not t.is_char('.') or t.is_whitespace_after
or t.is_whitespace_before):
return None
if (t.next0_ is not None):
t = t.next0_
if (val is None or (dig_count < 5)):
return None
if (dig_count != 6):
if (not key_word_before):
return None
else:
Utils.insertStringIO(val, 4, '.')
Utils.insertStringIO(val, 2, '.')
tt = t.next0_
first_pass3395 = True
while True:
if first_pass3395: first_pass3395 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
break
br = BracketHelper.try_parse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
t = tt
continue
t = tt
return TitleItemToken._new2655(t0, t, TitleItemToken.Types.SPECIALITY,
Utils.toStringStringIO(val))
def __check_detail(rt: 'ReferentToken') -> None:
if (rt.end_token.whitespaces_after_count > 2
or rt.end_token.next0_ is None):
return
if (rt.end_token.next0_.is_char('(')):
br = BracketHelper.try_parse(rt.end_token.next0_,
BracketParseAttr.NO, 100)
if (br is not None):
rt.referent.detail = MiscHelper.get_text_value(
br.begin_token.next0_, br.end_token.previous,
GetTextAttr.NO)
rt.end_token = br.end_token
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_org(t: 'Token',
can_be_cyr: bool = False) -> 'ReferentToken':
from pullenti.ner.org.internal.OrgItemNameToken import OrgItemNameToken
if (t is None):
return None
br = False
if (t.is_char('(') and t.next0_ is not None):
t = t.next0_
br = True
if (isinstance(t, NumberToken)):
if (t.typ == NumberSpellingType.WORDS
and t.morph.class0_.is_adjective
and t.chars.is_capital_upper):
pass
else:
return None
else:
if (t.chars.is_all_lower):
return None
if ((t.length_char < 3) and not t.chars.is_letter):
return None
if (not t.chars.is_latin_letter):
if (not can_be_cyr or not t.chars.is_cyrillic_letter):
return None
t0 = t
t1 = t0
nam_wo = 0
tok = None
geo_ = None
add_typ = None
first_pass3312 = True
while True:
if first_pass3312: first_pass3312 = False
else: t = t.next0_
if (not (t is not None)): break
if (t != t0 and t.whitespaces_before_count > 1):
break
if (t.is_char(')')):
break
if (t.is_char('(') and t.next0_ is not None):
if ((isinstance(t.next0_.get_referent(), GeoReferent))
and t.next0_.next0_ is not None
and t.next0_.next0_.is_char(')')):
geo_ = (Utils.asObjectOrNull(t.next0_.get_referent(),
GeoReferent))
t = t.next0_.next0_
continue
typ = OrgItemTypeToken.try_attach(t.next0_, True, None)
if ((typ is not None and typ.end_token.next0_ is not None
and typ.end_token.next0_.is_char(')'))
and typ.chars.is_latin_letter):
add_typ = typ
t = typ.end_token.next0_
continue
if (((isinstance(t.next0_, TextToken)) and t.next0_.next0_
is not None and t.next0_.next0_.is_char(')'))
and t.next0_.chars.is_capital_upper):
t = t.next0_.next0_
t1 = t
continue
break
tok = OrgItemEngItem.try_attach(t, can_be_cyr)
if (tok is None and t.is_char_of(".,") and t.next0_ is not None):
tok = OrgItemEngItem.try_attach(t.next0_, can_be_cyr)
if (tok is None and t.next0_.is_char_of(",.")):
tok = OrgItemEngItem.try_attach(t.next0_.next0_,
can_be_cyr)
if (tok is not None):
if (tok.length_char == 1 and t0.chars.is_cyrillic_letter):
return None
break
if (t.is_hiphen and not t.is_whitespace_after
and not t.is_whitespace_before):
continue
if (t.is_char_of("&+") or t.is_and):
continue
if (t.is_char('.')):
if (t.previous is not None and t.previous.length_char == 1):
continue
elif (MiscHelper.can_be_start_of_sentence(t.next0_)):
break
if (not t.chars.is_latin_letter):
if (not can_be_cyr or not t.chars.is_cyrillic_letter):
break
if (t.chars.is_all_lower):
if (t.morph.class0_.is_preposition
or t.morph.class0_.is_conjunction):
continue
if (br):
continue
break
mc = t.get_morph_class_in_dictionary()
if (mc.is_verb):
if (t.next0_ is not None
and t.next0_.morph.class0_.is_preposition):
break
if (t.next0_ is not None and t.next0_.is_value("OF", None)):
break
if (isinstance(t, TextToken)):
nam_wo += 1
t1 = t
if (tok is None):
return None
if (t0 == tok.begin_token):
br2 = BracketHelper.try_parse(tok.end_token.next0_,
BracketParseAttr.NO, 100)
if (br2 is not None):
org1 = OrganizationReferent()
if (tok.short_value is not None):
org1.add_type_str(tok.short_value)
org1.add_type_str(tok.full_value)
nam1 = MiscHelper.get_text_value(br2.begin_token,
br2.end_token, GetTextAttr.NO)
if (nam1 is not None):
org1.add_name(nam1, True, None)
return ReferentToken(org1, t0, br2.end_token)
return None
org0_ = OrganizationReferent()
te = tok.end_token
if (tok.is_bank):
t1 = tok.end_token
if (tok.full_value == "company" and (tok.whitespaces_after_count < 3)):
tok1 = OrgItemEngItem.try_attach(tok.end_token.next0_, can_be_cyr)
if (tok1 is not None):
t1 = tok.end_token
tok = tok1
te = tok.end_token
if (tok.full_value == "company"):
if (nam_wo == 0):
return None
nam = MiscHelper.get_text_value(t0, t1, GetTextAttr.IGNOREARTICLES)
if (nam == "STOCK" and tok.full_value == "company"):
return None
alt_nam = None
if (Utils.isNullOrEmpty(nam)):
return None
if (nam.find('(') > 0):
i1 = nam.find('(')
i2 = nam.find(')')
if (i1 < i2):
alt_nam = nam
tai = None
if ((i2 + 1) < len(nam)):
tai = nam[i2:].strip()
nam = nam[0:0 + i1].strip()
if (tai is not None):
nam = "{0} {1}".format(nam, tai)
if (tok.is_bank):
org0_.add_type_str(
("bank" if tok.kit.base_language.is_en else "банк"))
org0_.add_profile(OrgProfile.FINANCE)
if ((t1.next0_ is not None and t1.next0_.is_value("OF", None)
and t1.next0_.next0_ is not None)
and t1.next0_.next0_.chars.is_latin_letter):
nam0 = OrgItemNameToken.try_attach(t1.next0_, None, False,
False)
if (nam0 is not None):
te = nam0.end_token
else:
te = t1.next0_.next0_
nam = MiscHelper.get_text_value(t0, te, GetTextAttr.NO)
if (isinstance(te.get_referent(), GeoReferent)):
org0_._add_geo_object(
Utils.asObjectOrNull(te.get_referent(), GeoReferent))
elif (t0 == t1):
return None
else:
if (tok.short_value is not None):
org0_.add_type_str(tok.short_value)
org0_.add_type_str(tok.full_value)
if (Utils.isNullOrEmpty(nam)):
return None
org0_.add_name(nam, True, None)
if (alt_nam is not None):
org0_.add_name(alt_nam, True, None)
res = ReferentToken(org0_, t0, te)
t = te
while t.next0_ is not None:
if (t.next0_.is_char_of(",.")):
t = t.next0_
else:
break
if (t.whitespaces_after_count < 2):
tok = OrgItemEngItem.try_attach(t.next0_, can_be_cyr)
if (tok is not None):
if (tok.short_value is not None):
org0_.add_type_str(tok.short_value)
org0_.add_type_str(tok.full_value)
res.end_token = tok.end_token
if (geo_ is not None):
org0_._add_geo_object(geo_)
if (add_typ is not None):
org0_.add_type(add_typ, False)
if (not br):
return res
t = res.end_token
if (t.next0_ is None or t.next0_.is_char(')')):
res.end_token = t.next0_
else:
return None
return res
def __try_parse(t: 'Token', lev: int) -> 'BookLinkToken':
if (t is None or lev > 3):
return None
if (t.is_char('[')):
re = BookLinkToken.__try_parse(t.next0_, lev + 1)
if (re is not None and re.end_token.next0_ is not None
and re.end_token.next0_.is_char(']')):
re.begin_token = t
re.end_token = re.end_token.next0_
return re
if (re is not None and re.end_token.is_char(']')):
re.begin_token = t
return re
if (re is not None):
if (re.typ == BookLinkTyp.SOSTAVITEL
or re.typ == BookLinkTyp.EDITORS):
return re
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
if ((isinstance(br.end_token.previous, NumberToken))
and (br.length_char < 30)):
return BookLinkToken._new329(
t, br.end_token, BookLinkTyp.NUMBER,
MiscHelper.get_text_value(br.begin_token.next0_,
br.end_token.previous,
GetTextAttr.NO))
t0 = t
if (isinstance(t, ReferentToken)):
if (isinstance(t.get_referent(), PersonReferent)):
return BookLinkToken.try_parse_author(
t, FioTemplateType.UNDEFINED)
if (isinstance(t.get_referent(), GeoReferent)):
return BookLinkToken._new326(t, t, BookLinkTyp.GEO,
t.get_referent())
if (isinstance(t.get_referent(), DateReferent)):
dr = Utils.asObjectOrNull(t.get_referent(), DateReferent)
if (len(dr.slots) == 1 and dr.year > 0):
return BookLinkToken._new329(t, t, BookLinkTyp.YEAR,
str(dr.year))
if (dr.year > 0 and t.previous is not None
and t.previous.is_comma):
return BookLinkToken._new329(t, t, BookLinkTyp.YEAR,
str(dr.year))
if (isinstance(t.get_referent(), OrganizationReferent)):
org0_ = Utils.asObjectOrNull(t.get_referent(),
OrganizationReferent)
if (org0_.kind == OrganizationKind.PRESS):
return BookLinkToken._new326(t, t, BookLinkTyp.PRESS,
org0_)
if (isinstance(t.get_referent(), UriReferent)):
uri = Utils.asObjectOrNull(t.get_referent(), UriReferent)
if ((uri.scheme == "http" or uri.scheme == "https"
or uri.scheme == "ftp") or uri.scheme is None):
return BookLinkToken._new326(t, t, BookLinkTyp.URL, uri)
tok_ = BookLinkToken.__m_termins.try_parse(t, TerminParseAttr.NO)
if (tok_ is not None):
typ_ = Utils.valToEnum(tok_.termin.tag, BookLinkTyp)
ok = True
if (typ_ == BookLinkTyp.TYPE or typ_ == BookLinkTyp.NAMETAIL
or typ_ == BookLinkTyp.ELECTRONRES):
if (t.previous is not None and
((t.previous.is_char_of(".:[") or t.previous.is_hiphen))):
pass
else:
ok = False
if (ok):
return BookLinkToken._new329(t, tok_.end_token, typ_,
tok_.termin.canonic_text)
if (typ_ == BookLinkTyp.ELECTRONRES):
tt = tok_.end_token.next0_
first_pass3019 = True
while True:
if first_pass3019: first_pass3019 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if ((isinstance(tt, TextToken))
and not tt.chars.is_letter):
continue
if (isinstance(tt.get_referent(), UriReferent)):
return BookLinkToken._new326(t, tt,
BookLinkTyp.ELECTRONRES,
tt.get_referent())
break
if (t.is_char('/')):
res = BookLinkToken._new329(t, t, BookLinkTyp.DELIMETER, "/")
if (t.next0_ is not None and t.next0_.is_char('/')):
res.end_token = t.next0_
res.value = "//"
if (not t.is_whitespace_before and not t.is_whitespace_after):
coo = 3
no = True
tt = t.next0_
while tt is not None and coo > 0:
vvv = BookLinkToken.try_parse(tt, lev + 1)
if (vvv is not None and vvv.typ != BookLinkTyp.NUMBER):
no = False
break
tt = tt.next0_
coo -= 1
if (no):
return None
return res
if ((isinstance(t, NumberToken)) and t.int_value is not None
and t.typ == NumberSpellingType.DIGIT):
res = BookLinkToken._new329(t, t, BookLinkTyp.NUMBER, str(t.value))
val = t.int_value
if (val >= 1930 and (val < 2030)):
res.typ = BookLinkTyp.YEAR
if (t.next0_ is not None and t.next0_.is_char('.')):
res.end_token = t.next0_
elif ((t.next0_ is not None and t.next0_.length_char == 1
and not t.next0_.chars.is_letter)
and t.next0_.is_whitespace_after):
res.end_token = t.next0_
elif (isinstance(t.next0_, TextToken)):
term = t.next0_.term
if (((term == "СТР" or term == "C" or term == "С")
or term == "P" or term == "S") or term == "PAGES"):
res.end_token = t.next0_
res.typ = BookLinkTyp.PAGES
res.value = str(t.value)
return res
if (isinstance(t, TextToken)):
term = t.term
if (((((
((term == "СТР" or term == "C" or term == "С") or term == "ТОМ"
or term == "T") or term == "Т" or term == "P") or term == "PP"
or term == "V") or term == "VOL" or term == "S")
or term == "СТОР" or t.is_value("PAGE", None))
or t.is_value("СТРАНИЦА", "СТОРІНКА")):
tt = t.next0_
while tt is not None:
if (tt.is_char_of(".:~")):
tt = tt.next0_
else:
break
if (isinstance(tt, NumberToken)):
res = BookLinkToken._new328(t, tt, BookLinkTyp.PAGERANGE)
tt0 = tt
tt1 = tt
tt = tt.next0_
first_pass3020 = True
while True:
if first_pass3020: first_pass3020 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_char_of(",") or tt.is_hiphen):
if (isinstance(tt.next0_, NumberToken)):
tt = tt.next0_
res.end_token = tt
tt1 = tt
continue
break
res.value = MiscHelper.get_text_value(
tt0, tt1, GetTextAttr.NO)
return res
if ((term == "M" or term == "М" or term == "СПБ") or term == "K"
or term == "К"):
if (t.next0_ is not None and t.next0_.is_char_of(":;")):
re = BookLinkToken._new328(t, t.next0_, BookLinkTyp.GEO)
return re
if (t.next0_ is not None and t.next0_.is_char_of(".")):
res = BookLinkToken._new328(t, t.next0_, BookLinkTyp.GEO)
if (t.next0_.next0_ is not None
and t.next0_.next0_.is_char_of(":;")):
res.end_token = t.next0_.next0_
elif (t.next0_.next0_ is not None
and (isinstance(t.next0_.next0_, NumberToken))):
pass
elif (t.next0_.next0_ is not None
and t.next0_.next0_.is_comma and
(isinstance(t.next0_.next0_.next0_, NumberToken))):
pass
else:
return None
return res
if (term == "ПЕР" or term == "ПЕРЕВ" or term == "ПЕРЕВОД"):
tt = t
if (tt.next0_ is not None and tt.next0_.is_char('.')):
tt = tt.next0_
if (tt.next0_ is not None
and ((tt.next0_.is_value("C", None)
or tt.next0_.is_value("С", None)))):
tt = tt.next0_
if (tt.next0_ is None or tt.whitespaces_after_count > 2):
return None
re = BookLinkToken._new328(t, tt.next0_,
BookLinkTyp.TRANSLATE)
return re
if (term == "ТАМ" or term == "ТАМЖЕ"):
res = BookLinkToken._new328(t, t, BookLinkTyp.TAMZE)
if (t.next0_ is not None and t.next0_.is_value("ЖЕ", None)):
res.end_token = t.next0_
return res
if (((term == "СМ" or term == "CM" or term == "НАПР")
or term == "НАПРИМЕР" or term == "SEE") or term == "ПОДРОБНЕЕ"
or term == "ПОДРОБНО"):
res = BookLinkToken._new328(t, t, BookLinkTyp.SEE)
t = t.next0_
first_pass3021 = True
while True:
if first_pass3021: first_pass3021 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_char_of(".:") or t.is_value("ALSO", None)):
res.end_token = t
continue
if (t.is_value("В", None) or t.is_value("IN", None)):
res.end_token = t
continue
vvv = BookLinkToken.__try_parse(t, lev + 1)
if (vvv is not None and vvv.typ == BookLinkTyp.SEE):
res.end_token = vvv.end_token
break
break
return res
if (term == "БОЛЕЕ"):
vvv = BookLinkToken.__try_parse(t.next0_, lev + 1)
if (vvv is not None and vvv.typ == BookLinkTyp.SEE):
vvv.begin_token = t
return vvv
no = MiscHelper.check_number_prefix(t)
if (isinstance(no, NumberToken)):
return BookLinkToken._new328(t, no, BookLinkTyp.N)
if (((term == "B" or term == "В"))
and (isinstance(t.next0_, NumberToken))
and (isinstance(t.next0_.next0_, TextToken))):
term2 = t.next0_.next0_.term
if (((term2 == "Т" or term2 == "T" or term2.startswith("ТОМ"))
or term2 == "TT" or term2 == "ТТ") or term2 == "КН"
or term2.startswith("КНИГ")):
return BookLinkToken._new328(t, t.next0_.next0_,
BookLinkTyp.VOLUME)
if (t.is_char('(')):
if (((isinstance(t.next0_, NumberToken)) and t.next0_.int_value
is not None and t.next0_.next0_ is not None)
and t.next0_.next0_.is_char(')')):
num = t.next0_.int_value
if (num > 1900 and num <= 2040):
if (num <= datetime.datetime.now().year):
return BookLinkToken._new329(t, t.next0_.next0_,
BookLinkTyp.YEAR,
str(num))
if (((isinstance(t.next0_, ReferentToken)) and
(isinstance(t.next0_.get_referent(), DateReferent))
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_char(')')):
num = t.next0_.get_referent().year
if (num > 0):
return BookLinkToken._new329(t, t.next0_.next0_,
BookLinkTyp.YEAR, str(num))
return None
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
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 __try_parse_ru(first: 'Token',
typ: 'NounPhraseParseAttr',
max_char_pos: int,
def_noun: 'NounPhraseItem' = None) -> 'NounPhraseToken':
if (first is None):
return None
items = None
adverbs = None
prep = None
kak = False
t0 = first
if ((((typ) & (NounPhraseParseAttr.PARSEPREPOSITION))) !=
(NounPhraseParseAttr.NO) and t0.is_value("КАК", None)):
t0 = t0.next0_
prep = PrepositionHelper.try_parse(t0)
if (prep is not None):
t0 = prep.end_token.next0_
kak = True
internal_noun_prase = None
conj_before = False
t = t0
first_pass3041 = True
while True:
if first_pass3041: first_pass3041 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char_pos > 0 and t.begin_char > max_char_pos):
break
if ((t.morph.class0_.is_conjunction
and not t.morph.class0_.is_adjective
and not t.morph.class0_.is_pronoun)
and not t.morph.class0_.is_noun):
if (conj_before):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
if (items is not None and ((t.is_and or t.is_or))):
conj_before = True
if ((t.next0_ is not None and t.next0_.is_char_of("\\/")
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or):
t = t.next0_.next0_
if (((t.next0_ is not None and t.next0_.is_char('(')
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or
and t.next0_.next0_.next0_ is not None)
and t.next0_.next0_.next0_.is_char(')')):
t = t.next0_.next0_.next0_
continue
break
elif (t.is_comma):
if (conj_before or items is None):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
mc = t.previous.get_morph_class_in_dictionary()
if (mc.is_proper_surname or mc.is_proper_secname):
break
conj_before = True
if (kak and t.next0_ is not None
and t.next0_.is_value("ТАК", None)):
t = t.next0_
if (t.next0_ is not None and t.next0_.is_and):
t = t.next0_
pr = PrepositionHelper.try_parse(t.next0_)
if (pr is not None):
t = pr.end_token
if (items[len(items) - 1].can_be_noun
and items[len(items) -
1].end_token.morph.class0_.is_pronoun):
break
continue
elif (t.is_char('(')):
if (items is None):
return None
brr = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (brr is None):
break
if (brr.length_char > 100):
break
t = brr.end_token
continue
if (isinstance(t, ReferentToken)):
if ((((typ) & (NounPhraseParseAttr.REFERENTCANBENOUN))) == (
NounPhraseParseAttr.NO)):
break
elif (t.chars.is_latin_letter):
break
it = NounPhraseItem.try_parse(t, items, typ)
if (it is None or ((not it.can_be_adj and not it.can_be_noun))):
if (((it is not None and items is not None
and t.chars.is_capital_upper) and
(t.whitespaces_before_count < 3) and t.length_char > 3)
and not t.get_morph_class_in_dictionary().is_noun and
not t.get_morph_class_in_dictionary().is_adjective):
it.can_be_noun = True
items.append(it)
break
if ((((typ) & (NounPhraseParseAttr.PARSEADVERBS))) !=
(NounPhraseParseAttr.NO) and (isinstance(t, TextToken))
and t.morph.class0_.is_adverb):
if (adverbs is None):
adverbs = list()
adverbs.append(Utils.asObjectOrNull(t, TextToken))
continue
break
it.conj_before = conj_before
conj_before = False
if (not it.can_be_adj and not it.can_be_noun):
break
if (t.is_newline_before and t != first):
if ((((typ) & (NounPhraseParseAttr.MULTILINES))) !=
(NounPhraseParseAttr.NO)):
pass
elif (items is not None
and t.chars != items[len(items) - 1].chars):
if (t.chars.is_all_lower
and items[len(items) - 1].chars.is_capital_upper):
pass
else:
break
if (items is None):
items = list()
else:
it0 = items[len(items) - 1]
if (it0.can_be_noun and it0.is_personal_pronoun):
if (it.is_pronoun):
break
if ((it0.begin_token.previous is not None
and it0.begin_token.previous.
get_morph_class_in_dictionary().is_verb
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_adjective)
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_preposition):
if (t.morph.case_.is_nominative
or t.morph.case_.is_accusative):
pass
else:
break
if (it.can_be_noun and it.is_verb):
if (it0.previous is None):
pass
elif ((isinstance(it0.previous, TextToken))
and not it0.previous.chars.is_letter):
pass
else:
break
items.append(it)
t = it.end_token
if (t.is_newline_after and not t.chars.is_all_lower):
mc = t.get_morph_class_in_dictionary()
if (mc.is_proper_surname):
break
if (t.morph.class0_.is_proper_surname and mc.is_undefined):
break
if (items is None):
return None
tt1 = None
if (len(items) == 1 and items[0].can_be_adj):
and0_ = False
tt1 = items[0].end_token.next0_
first_pass3042 = True
while True:
if first_pass3042: first_pass3042 = False
else: tt1 = tt1.next0_
if (not (tt1 is not None)): break
if (tt1.is_and or tt1.is_or):
and0_ = True
break
if (tt1.is_comma or tt1.is_value("НО", None)
or tt1.is_value("ТАК", None)):
continue
break
if (and0_):
if (items[0].can_be_noun and items[0].is_personal_pronoun):
and0_ = False
if (and0_):
tt2 = tt1.next0_
if (tt2 is not None and tt2.morph.class0_.is_preposition):
tt2 = tt2.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt2, typ, max_char_pos, None)
if (npt1 is not None and len(npt1.adjectives) > 0):
ok1 = False
for av in items[0].adj_morph:
for v in npt1.noun.noun_morph:
if (v.check_accord(av, False, False)):
items[0].morph.add_item(av)
ok1 = True
if (ok1):
npt1.begin_token = items[0].begin_token
npt1.end_token = tt1.previous
npt1.adjectives.clear()
npt1.adjectives.append(items[0])
return npt1
if (def_noun is not None):
items.append(def_noun)
last1 = items[len(items) - 1]
check = True
for it in items:
if (not it.can_be_adj):
check = False
break
elif (it.can_be_noun and it.is_personal_pronoun):
check = False
break
tt1 = last1.end_token.next0_
if ((tt1 is not None and check and
((tt1.morph.class0_.is_preposition
or tt1.morph.case_.is_instrumental)))
and (tt1.whitespaces_before_count < 2)):
inp = NounPhraseHelper.try_parse(
tt1,
Utils.valToEnum((typ) | (NounPhraseParseAttr.PARSEPREPOSITION),
NounPhraseParseAttr), max_char_pos, None)
if (inp is not None):
tt1 = inp.end_token.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt1, typ, max_char_pos, None)
if (npt1 is not None):
ok = True
ii = 0
first_pass3043 = True
while True:
if first_pass3043: first_pass3043 = False
else: ii += 1
if (not (ii < len(items))): break
it = items[ii]
if (NounPhraseItem.try_accord_adj_and_noun(
it,
Utils.asObjectOrNull(npt1.noun,
NounPhraseItem))):
continue
if (ii > 0):
inp2 = NounPhraseHelper.try_parse(
it.begin_token, typ, max_char_pos, None)
if (inp2 is not None
and inp2.end_token == inp.end_token):
del items[ii:ii + len(items) - ii]
inp = inp2
break
ok = False
break
if (ok):
if (npt1.morph.case_.is_genitive
and not inp.morph.case_.is_instrumental):
ok = False
if (ok):
i = 0
while i < len(items):
npt1.adjectives.insert(i, items[i])
i += 1
npt1.internal_noun = inp
mmm = MorphCollection(npt1.morph)
for it in items:
mmm.remove_items(it.adj_morph[0], False)
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
if (tt1 is not None and tt1.morph.class0_.is_noun
and not tt1.morph.case_.is_genitive):
it = NounPhraseItem.try_parse(tt1, items, typ)
if (it is not None and it.can_be_noun):
internal_noun_prase = inp
inp.begin_token = items[0].end_token.next0_
items.append(it)
i = 0
first_pass3044 = True
while True:
if first_pass3044: first_pass3044 = False
else: i += 1
if (not (i < len(items))): break
if (items[i].can_be_adj
and items[i].begin_token.morph.class0_.is_verb):
it = items[i].begin_token
if (not it.get_morph_class_in_dictionary().is_verb):
continue
if (it.is_value("УПОЛНОМОЧЕННЫЙ", None)):
continue
if ((((typ) & (NounPhraseParseAttr.PARSEVERBS))) == (
NounPhraseParseAttr.NO)):
continue
inp = _NounPraseHelperInt.__try_parse_ru(
items[i].end_token.next0_, NounPhraseParseAttr.NO,
max_char_pos, None)
if (inp is None):
continue
if (inp.anafor is not None and i == (len(items) - 1)
and NounPhraseItem.try_accord_adj_and_noun(
items[i],
Utils.asObjectOrNull(inp.noun, NounPhraseItem))):
inp.begin_token = first
ii = 0
while ii < len(items):
inp.adjectives.insert(ii, items[ii])
ii += 1
return inp
if (inp.end_token.whitespaces_after_count > 3):
continue
npt1 = _NounPraseHelperInt.__try_parse_ru(
inp.end_token.next0_, NounPhraseParseAttr.NO, max_char_pos,
None)
if (npt1 is None):
continue
ok = True
j = 0
while j <= i:
if (not NounPhraseItem.try_accord_adj_and_noun(
items[j],
Utils.asObjectOrNull(npt1.noun, NounPhraseItem))):
ok = False
break
j += 1
if (not ok):
continue
verb = VerbPhraseHelper.try_parse(it, True, False, False)
if (verb is None):
continue
vlinks = SemanticHelper.try_create_links(verb, inp, None)
nlinks = SemanticHelper.try_create_links(inp, npt1, None)
if (len(vlinks) == 0 and len(nlinks) > 0):
continue
j = 0
while j <= i:
npt1.adjectives.insert(j, items[j])
j += 1
items[i].end_token = inp.end_token
mmm = MorphCollection(npt1.morph)
bil = list()
j = 0
while j <= i:
bil.clear()
for m in items[j].adj_morph:
bil.append(m)
mmm.remove_items_list_cla(bil, None)
j += 1
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
ok2 = False
if ((len(items) == 1 and
(((typ) & (NounPhraseParseAttr.ADJECTIVECANBELAST))) !=
(NounPhraseParseAttr.NO) and
(items[0].whitespaces_after_count < 3))
and not items[0].is_adverb):
if (not items[0].can_be_adj):
ok2 = True
elif (items[0].is_personal_pronoun and items[0].can_be_noun):
ok2 = True
if (ok2):
it = NounPhraseItem.try_parse(items[0].end_token.next0_, None, typ)
if (it is not None and it.can_be_adj
and it.begin_token.chars.is_all_lower):
ok2 = True
if (it.is_adverb or it.is_verb):
ok2 = False
if (it.is_pronoun and items[0].is_pronoun):
ok2 = False
if (it.can_be_adj_for_personal_pronoun
and items[0].is_personal_pronoun):
ok2 = True
if (ok2 and NounPhraseItem.try_accord_adj_and_noun(
it, items[0])):
npt1 = _NounPraseHelperInt.__try_parse_ru(
it.begin_token, typ, max_char_pos, None)
if (npt1 is not None and ((npt1.end_char > it.end_char
or len(npt1.adjectives) > 0))):
pass
else:
items.insert(0, it)
noun = None
adj_after = None
for i in range(len(items) - 1, -1, -1):
if (items[i].can_be_noun):
if (items[i].conj_before):
continue
if (i > 0 and not items[i - 1].can_be_adj):
continue
if (i > 0 and items[i - 1].can_be_noun):
if (items[i - 1].is_doubt_adjective):
continue
if (items[i - 1].is_pronoun and items[i].is_pronoun):
if (items[i].is_pronoun and
items[i - 1].can_be_adj_for_personal_pronoun):
pass
else:
continue
noun = items[i]
del items[i:i + len(items) - i]
if (adj_after is not None):
items.append(adj_after)
elif (len(items) > 0 and items[0].can_be_noun
and not items[0].can_be_adj):
noun = items[0]
items.clear()
break
if (noun is None):
return None
res = NounPhraseToken._new466(first, noun.end_token, prep)
if (adverbs is not None):
for a in adverbs:
if (a.begin_char < noun.begin_char):
if (len(items) == 0 and prep is None):
return None
if (res.adverbs is None):
res.adverbs = list()
res.adverbs.append(a)
res.noun = (noun)
res.multi_nouns = noun.multi_nouns
if (kak):
res.multi_nouns = True
res.internal_noun = internal_noun_prase
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (res.morph.case_.is_nominative and first.previous is not None
and first.previous.morph.class0_.is_preposition):
res.morph.case_ = (res.morph.case_) ^ MorphCase.NOMINATIVE
if ((((typ) &
(NounPhraseParseAttr.PARSEPRONOUNS))) == (NounPhraseParseAttr.NO)
and ((res.morph.class0_.is_pronoun
or res.morph.class0_.is_personal_pronoun))):
return None
stat = None
if (len(items) > 1):
stat = dict()
need_update_morph = False
if (len(items) > 0):
ok_list = list()
is_num_not = False
for vv in noun.noun_morph:
i = 0
v = vv
i = 0
while i < len(items):
ok = False
for av in items[i].adj_morph:
if (v.check_accord(av, False, False)):
ok = True
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
break
if (not ok):
if (items[i].can_be_numeric_adj
and items[i].try_accord_var(v, False)):
ok = True
v1 = NounPhraseItemTextVar()
v1.copy_from_item(v)
v1.number = MorphNumber.PLURAL
is_num_not = True
v1.case_ = MorphCase()
for a in items[i].adj_morph:
v1.case_ = (v1.case_) | a.case_
v = v1
else:
break
i += 1
if (i >= len(items)):
ok_list.append(v)
if (len(ok_list) > 0 and
(((len(ok_list) < res.morph.items_count) or is_num_not))):
res.morph = MorphCollection()
for v in ok_list:
res.morph.add_item(v)
if (not is_num_not):
noun.morph = res.morph
i = 0
first_pass3045 = True
while True:
if first_pass3045: first_pass3045 = False
else: i += 1
if (not (i < len(items))): break
for av in items[i].adj_morph:
for v in noun.noun_morph:
if (v.check_accord(av, False, False)):
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
need_update_morph = True
items[i].morph.add_item(av)
if (stat is not None and av.normal_value is not None
and len(av.normal_value) > 1):
last = av.normal_value[len(av.normal_value) - 1]
if (not last in stat):
stat[last] = 1
else:
stat[last] += 1
if (items[i].is_pronoun or items[i].is_personal_pronoun):
res.anafor = items[i].begin_token
if ((((typ) & (NounPhraseParseAttr.PARSEPRONOUNS))) == (
NounPhraseParseAttr.NO)):
continue
tt = Utils.asObjectOrNull(items[i].begin_token, TextToken)
if (tt is not None and not tt.term.startswith("ВЫСШ")):
err = False
for wf in tt.morph.items:
if (wf.class0_.is_adjective):
if (wf.contains_attr("прев.", None)):
if ((((typ) &
(NounPhraseParseAttr.IGNOREADJBEST))) !=
(NounPhraseParseAttr.NO)):
err = True
if (wf.contains_attr("к.ф.", None)
and tt.morph.class0_.is_personal_pronoun):
return None
if (err):
continue
if (res.morph.case_.is_nominative):
v = MiscHelper.get_text_value_of_meta_token(
items[i], GetTextAttr.KEEPQUOTES)
if (not Utils.isNullOrEmpty(v)):
if (items[i].get_normal_case_text(
None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED,
False) != v):
wf = NounPhraseItemTextVar(items[i].morph, None)
wf.normal_value = v
wf.class0_ = MorphClass.ADJECTIVE
wf.case_ = res.morph.case_
if (res.morph.case_.is_prepositional
or res.morph.gender == MorphGender.NEUTER
or res.morph.gender == MorphGender.FEMINIE):
items[i].morph.add_item(wf)
else:
items[i].morph.insert_item(0, wf)
res.adjectives.append(items[i])
if (items[i].end_char > res.end_char):
res.end_token = items[i].end_token
i = 0
first_pass3046 = True
while True:
if first_pass3046: first_pass3046 = False
else: i += 1
if (not (i < (len(res.adjectives) - 1))): break
if (res.adjectives[i].whitespaces_after_count > 5):
if (res.adjectives[i].chars != res.adjectives[i + 1].chars):
if (not res.adjectives[i + 1].chars.is_all_lower):
return None
if (res.adjectives[i].chars.is_all_upper
and res.adjectives[i + 1].chars.is_capital_upper):
return None
if (res.adjectives[i].chars.is_capital_upper
and res.adjectives[i + 1].chars.is_all_upper):
return None
if (res.adjectives[i].whitespaces_after_count > 10):
if (res.adjectives[i].newlines_after_count == 1):
if (res.adjectives[i].chars.is_capital_upper and i == 0
and res.adjectives[i + 1].chars.is_all_lower):
continue
if (res.adjectives[i].chars == res.adjectives[
i + 1].chars):
continue
return None
if (need_update_morph):
noun.morph = MorphCollection()
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (len(res.adjectives) > 0):
if (noun.begin_token.previous is not None):
if (noun.begin_token.previous.is_comma_and):
if (res.adjectives[0].begin_char > noun.begin_char):
pass
else:
return None
zap = 0
and0_ = 0
cou = 0
last_and = False
i = 0
while i < (len(res.adjectives) - 1):
te = res.adjectives[i].end_token.next0_
if (te is None):
return None
if (te.is_char('(')):
pass
elif (te.is_comma):
zap += 1
last_and = False
elif (te.is_and or te.is_or):
and0_ += 1
last_and = True
if (not res.adjectives[i].begin_token.morph.class0_.is_pronoun
):
cou += 1
i += 1
if ((zap + and0_) > 0):
if (and0_ > 1):
return None
elif (and0_ == 1 and not last_and):
return None
if ((zap + and0_) != cou):
if (and0_ == 1):
pass
else:
return None
last = Utils.asObjectOrNull(
res.adjectives[len(res.adjectives) - 1], NounPhraseItem)
if (last.is_pronoun and not last_and):
return None
if (stat is not None):
for adj in items:
if (adj.morph.items_count > 1):
w1 = Utils.asObjectOrNull(adj.morph.get_indexer_item(0),
NounPhraseItemTextVar)
w2 = Utils.asObjectOrNull(adj.morph.get_indexer_item(1),
NounPhraseItemTextVar)
if ((len(w1.normal_value) < 2)
or (len(w2.normal_value) < 2)):
break
l1 = w1.normal_value[len(w1.normal_value) - 1]
l2 = w2.normal_value[len(w2.normal_value) - 1]
i1 = 0
i2 = 0
wrapi1468 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l1, wrapi1468)
i1 = wrapi1468.value
wrapi2467 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l2, wrapi2467)
i2 = wrapi2467.value
if (i1 < i2):
adj.morph.remove_item(1)
adj.morph.insert_item(0, w2)
if (res.begin_token.get_morph_class_in_dictionary().is_verb
and len(items) > 0):
if (not res.begin_token.chars.is_all_lower
or res.begin_token.previous is None):
pass
elif (res.begin_token.previous.morph.class0_.is_preposition):
pass
else:
comma = False
tt = res.begin_token.previous
first_pass3047 = True
while True:
if first_pass3047: first_pass3047 = False
else: tt = tt.previous
if (not (tt is not None and tt.end_char <= res.end_char)):
break
if (tt.morph.class0_.is_adverb):
continue
if (tt.is_char_of(".;")):
break
if (tt.is_comma):
comma = True
continue
if (tt.is_value("НЕ", None)):
continue
if (((tt.morph.class0_.is_noun
or tt.morph.class0_.is_proper)) and comma):
for it in res.begin_token.morph.items:
if (it.class0_.is_verb
and (isinstance(it, MorphWordForm))):
if (tt.morph.check_accord(it, False, False)):
if (res.morph.case_.is_instrumental):
return None
break
if (res.begin_token == res.end_token):
mc = res.begin_token.get_morph_class_in_dictionary()
if (mc.is_adverb):
if (res.begin_token.previous is not None and
res.begin_token.previous.morph.class0_.is_preposition):
pass
elif (mc.is_noun and not mc.is_preposition
and not mc.is_conjunction):
pass
elif (res.begin_token.is_value("ВЕСЬ", None)):
pass
else:
return None
if (def_noun is not None and def_noun.end_token == res.end_token
and len(res.adjectives) > 0):
res.end_token = res.adjectives[len(res.adjectives) - 1].end_token
return res
def __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 create_nickname(pr : 'PersonReferent', t : 'Token') -> 'Token':
has_keyw = False
is_br = False
first_pass3367 = True
while True:
if first_pass3367: first_pass3367 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_hiphen or t.is_comma or t.is_char_of(".:;")):
continue
if (t.morph.class0_.is_preposition):
continue
if (t.is_char('(')):
is_br = True
continue
if ((t.is_value("ПРОЗВИЩЕ", "ПРІЗВИСЬКО") or t.is_value("КЛИЧКА", None) or t.is_value("ПСЕВДОНИМ", "ПСЕВДОНІМ")) or t.is_value("ПСЕВДО", None) or t.is_value("ПОЗЫВНОЙ", "ПОЗИВНИЙ")):
has_keyw = True
continue
break
if (not has_keyw or t is None):
return None
if (BracketHelper.is_bracket(t, True)):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
ni = MiscHelper.get_text_value(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = br.end_token
tt = t.next0_
first_pass3368 = True
while True:
if first_pass3368: first_pass3368 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_comma_and):
continue
if (not BracketHelper.is_bracket(tt, True)):
break
br = BracketHelper.try_parse(tt, BracketParseAttr.NO, 100)
if (br is None):
break
ni = MiscHelper.get_text_value(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
tt = br.end_token
t = tt
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
return t
else:
ret = None
first_pass3369 = True
while True:
if first_pass3369: first_pass3369 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_comma_and):
continue
if (ret is not None and t.chars.is_all_lower):
break
if (t.whitespaces_before_count > 2):
break
pli = PersonItemToken.try_attach_list(t, None, PersonItemToken.ParseAttr.NO, 10)
if (pli is not None and ((len(pli) == 1 or len(pli) == 2))):
ni = MiscHelper.get_text_value(pli[0].begin_token, pli[len(pli) - 1].end_token, GetTextAttr.NO)
if (ni is not None):
pr.add_slot(PersonReferent.ATTR_NICKNAME, ni, False, 0)
t = pli[len(pli) - 1].end_token
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
ret = t
continue
if ((isinstance(t, ReferentToken)) and not t.chars.is_all_lower and t.begin_token == t.end_token):
val = MiscHelper.get_text_value_of_meta_token(Utils.asObjectOrNull(t, ReferentToken), GetTextAttr.NO)
pr.add_slot(PersonReferent.ATTR_NICKNAME, val, False, 0)
if (is_br and t.next0_ is not None and t.next0_.is_char(')')):
t = t.next0_
ret = t
continue
break
return ret
return None
def process(self, kit: 'AnalysisKit') -> None:
ad = kit.get_analyzer_data(self)
models = TerminCollection()
objs_by_model = dict()
obj_by_names = TerminCollection()
t = kit.first_token
first_pass3428 = True
while True:
if first_pass3428: first_pass3428 = False
else: t = t.next0_
if (not (t is not None)): break
its = WeaponItemToken.try_parse_list(t, 10)
if (its is None):
continue
rts = self.__try_attach(its, False)
if (rts is not None):
for rt in rts:
rt.referent = ad.register_referent(rt.referent)
kit.embed_token(rt)
t = (rt)
for s in rt.referent.slots:
if (s.type_name == WeaponReferent.ATTR_MODEL):
mod = str(s.value)
for k in range(2):
if (not str.isdigit(mod[0])):
li = []
wrapli2804 = RefOutArgWrapper(None)
inoutres2805 = Utils.tryGetValue(
objs_by_model, mod, wrapli2804)
li = wrapli2804.value
if (not inoutres2805):
li = list()
objs_by_model[mod] = li
if (not rt.referent in li):
li.append(rt.referent)
models.add_string(mod, li, None, False)
if (k > 0):
break
brand = rt.referent.get_string_value(
WeaponReferent.ATTR_BRAND)
if (brand is None):
break
mod = "{0} {1}".format(brand, mod)
elif (s.type_name == WeaponReferent.ATTR_NAME):
obj_by_names.add(
Termin._new100(str(s.value), rt.referent))
if (len(objs_by_model) == 0 and len(obj_by_names.termins) == 0):
return
t = kit.first_token
first_pass3429 = True
while True:
if first_pass3429: first_pass3429 = False
else: t = t.next0_
if (not (t is not None)): break
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 10)
if (br is not None):
toks = obj_by_names.try_parse(t.next0_, TerminParseAttr.NO)
if (toks is not None
and toks.end_token.next0_ == br.end_token):
rt0 = ReferentToken(
Utils.asObjectOrNull(toks.termin.tag, Referent),
br.begin_token, br.end_token)
kit.embed_token(rt0)
t = (rt0)
continue
if (not (isinstance(t, TextToken))):
continue
if (not t.chars.is_letter):
continue
tok = models.try_parse(t, TerminParseAttr.NO)
if (tok is None):
if (not t.chars.is_all_lower):
tok = obj_by_names.try_parse(t, TerminParseAttr.NO)
if (tok is None):
continue
if (not tok.is_whitespace_after):
if (tok.end_token.next0_ is None
or not tok.end_token.next0_.is_char_of(",.)")):
if (not BracketHelper.is_bracket(tok.end_token.next0_,
False)):
continue
tr = None
li = Utils.asObjectOrNull(tok.termin.tag, list)
if (li is not None and len(li) == 1):
tr = li[0]
else:
tr = (Utils.asObjectOrNull(tok.termin.tag, Referent))
if (tr is not None):
tit = WeaponItemToken.try_parse(tok.begin_token.previous, None,
False, True)
if (tit is not None and tit.typ == WeaponItemToken.Typs.BRAND):
tr.add_slot(WeaponReferent.ATTR_BRAND, tit.value, False, 0)
tok.begin_token = tit.begin_token
rt0 = ReferentToken(tr, tok.begin_token, tok.end_token)
kit.embed_token(rt0)
t = (rt0)
continue
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 __correct_person(res : 'InstrToken') -> 'InstrToken':
spec_chars = 0
if (not res.is_pure_person):
res.typ = ILTypes.UNDEFINED
return res
t = res.end_token.next0_
first_pass3254 = True
while True:
if first_pass3254: first_pass3254 = False
else: t = t.next0_
if (not (t is not None)): break
if ((isinstance(t, ReferentToken)) and (isinstance(res.ref, ReferentToken))):
ok = False
if (t.get_referent() == res.ref.referent):
ok = True
ip = Utils.asObjectOrNull(res.ref.referent, InstrumentParticipantReferent)
if (ip is not None and ip._contains_ref(t.get_referent())):
ok = True
if (not ok and t.previous is not None and t.previous.is_table_control_char):
if ((isinstance(res.ref.referent, PersonPropertyReferent)) and (isinstance(t.get_referent(), PersonReferent))):
ok = True
res.ref = (t)
if (ok):
res.end_token = t
continue
tok = InstrToken.__m_ontology.try_parse(t, TerminParseAttr.NO)
if (tok is not None):
if ((((tok.termin.canonic_text == "ПОДПИСЬ" or tok.termin.canonic_text == "ПІДПИС")) and t.is_newline_before and t.next0_ is not None) and t.next0_.is_value("СТОРОНА", None)):
break
t = tok.end_token
res.end_token = t
continue
if (t.is_char(',')):
continue
if (t.is_table_control_char and not t.is_newline_before):
continue
if (t.is_char_of("_/\\")):
res.end_token = t
spec_chars += 1
continue
if (t.is_char('(') and t.next0_ is not None):
tok = InstrToken.__m_ontology.try_parse(t.next0_, TerminParseAttr.NO)
if ((tok) is not None):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
res.end_token = t
continue
break
rt0 = Utils.asObjectOrNull(res.ref, ReferentToken)
if (rt0 is not None and (isinstance(rt0.referent, InstrumentParticipantReferent))):
tt = res.begin_token
while tt is not None and tt.end_char <= res.end_char:
if ((isinstance(tt.get_referent(), PersonReferent)) or (isinstance(tt.get_referent(), PersonPropertyReferent))):
res.ref = (tt)
return res
elif ((isinstance(tt, TextToken)) and tt.is_char_of("_/\\")):
spec_chars += 1
elif (isinstance(tt, MetaToken)):
ttt = tt.begin_token
while ttt is not None and ttt.end_char <= tt.end_char:
if ((isinstance(ttt.get_referent(), PersonReferent)) or (isinstance(ttt.get_referent(), PersonPropertyReferent))):
res.ref = (ttt)
return res
elif ((isinstance(ttt, TextToken)) and ttt.is_char_of("_/\\")):
spec_chars += 1
ttt = ttt.next0_
tt = tt.next0_
if (spec_chars < 10):
res.typ = ILTypes.UNDEFINED
return res
def try_parse(t : 'Token', loc_onto : 'IntOntologyCollection') -> 'NamedItemToken':
if (t is None):
return None
if (isinstance(t, ReferentToken)):
r = t.get_referent()
if ((r.type_name == "PERSON" or r.type_name == "PERSONPROPERTY" or (isinstance(r, GeoReferent))) or r.type_name == "ORGANIZATION"):
return NamedItemToken._new1758(t, t, r, t.morph)
return None
typ = NamedItemToken.__m_types.try_parse(t, TerminParseAttr.NO)
nam = NamedItemToken.__m_names.try_parse(t, TerminParseAttr.NO)
if (typ is not None):
if (not (isinstance(t, TextToken))):
return None
res = NamedItemToken._new1759(typ.begin_token, typ.end_token, typ.morph, typ.chars)
res.kind = (Utils.valToEnum(typ.termin.tag, NamedEntityKind))
res.type_value = typ.termin.canonic_text
if ((nam is not None and nam.end_token == typ.end_token and not t.chars.is_all_lower) and (Utils.valToEnum(nam.termin.tag, NamedEntityKind)) == res.kind):
res.name_value = nam.termin.canonic_text
res.is_wellknown = True
return res
if (nam is not None):
if (nam.begin_token.chars.is_all_lower):
return None
res = NamedItemToken._new1759(nam.begin_token, nam.end_token, nam.morph, nam.chars)
res.kind = (Utils.valToEnum(nam.termin.tag, NamedEntityKind))
res.name_value = nam.termin.canonic_text
ok = True
if (not t.is_whitespace_before and t.previous is not None):
ok = False
elif (not t.is_whitespace_after and t.next0_ is not None):
if (t.next0_.is_char_of(",.;!?") and t.next0_.is_whitespace_after):
pass
else:
ok = False
if (ok):
res.is_wellknown = True
res.type_value = (Utils.asObjectOrNull(nam.termin.tag2, str))
return res
adj = MiscLocationHelper.try_attach_nord_west(t)
if (adj is not None):
if (adj.morph.class0_.is_noun):
if (adj.end_token.is_value("ВОСТОК", None)):
if (adj.begin_token == adj.end_token):
return None
re = NamedItemToken._new1761(t, adj.end_token, adj.morph)
re.kind = NamedEntityKind.LOCATION
re.name_value = MiscHelper.get_text_value(t, adj.end_token, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
re.is_wellknown = True
return re
return None
if (adj.whitespaces_after_count > 2):
return None
if ((isinstance(adj.end_token.next0_, ReferentToken)) and (isinstance(adj.end_token.next0_.get_referent(), GeoReferent))):
re = NamedItemToken._new1761(t, adj.end_token.next0_, adj.end_token.next0_.morph)
re.kind = NamedEntityKind.LOCATION
re.name_value = MiscHelper.get_text_value(t, adj.end_token.next0_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
re.is_wellknown = True
re.ref = adj.end_token.next0_.get_referent()
return re
res = NamedItemToken.try_parse(adj.end_token.next0_, loc_onto)
if (res is not None and res.kind == NamedEntityKind.LOCATION):
s = adj.get_normal_case_text(MorphClass.ADJECTIVE, MorphNumber.SINGULAR, res.morph.gender, False)
if (s is not None):
if (res.name_value is None):
res.name_value = s.upper()
else:
res.name_value = "{0} {1}".format(s.upper(), res.name_value)
res.type_value = (None)
res.begin_token = t
res.chars = t.chars
res.is_wellknown = True
return res
if (t.chars.is_capital_upper and not MiscHelper.can_be_start_of_sentence(t)):
npt = NounPhraseHelper.try_parse(t, NounPhraseParseAttr.NO, 0, None)
if (npt is not None and len(npt.adjectives) > 0):
test = NamedItemToken.try_parse(npt.noun.begin_token, loc_onto)
if (test is not None and test.end_token == npt.end_token and test.type_value is not None):
test.begin_token = t
tmp = io.StringIO()
for a in npt.adjectives:
s = a.get_normal_case_text(MorphClass.ADJECTIVE, MorphNumber.SINGULAR, test.morph.gender, False)
if (tmp.tell() > 0):
print(' ', end="", file=tmp)
print(s, end="", file=tmp)
test.name_value = Utils.toStringStringIO(tmp)
test.chars = t.chars
if (test.kind == NamedEntityKind.LOCATION):
test.is_wellknown = True
return test
if ((BracketHelper.is_bracket(t, True) and t.next0_ is not None and t.next0_.chars.is_letter) and not t.next0_.chars.is_all_lower):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
res = NamedItemToken(t, br.end_token)
res.is_in_bracket = True
res.name_value = MiscHelper.get_text_value(t, br.end_token, GetTextAttr.NO)
nam = NamedItemToken.__m_names.try_parse(t.next0_, TerminParseAttr.NO)
if (nam is not None and nam.end_token == br.end_token.previous):
res.kind = (Utils.valToEnum(nam.termin.tag, NamedEntityKind))
res.is_wellknown = True
res.name_value = nam.termin.canonic_text
return res
if (((isinstance(t, TextToken)) and t.chars.is_letter and not t.chars.is_all_lower) and t.length_char > 2):
res = NamedItemToken._new1761(t, t, t.morph)
str0_ = t.term
if (str0_.endswith("О") or str0_.endswith("И") or str0_.endswith("Ы")):
res.name_value = str0_
else:
res.name_value = t.get_normal_case_text(None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED, False)
res.chars = t.chars
if (((not t.is_whitespace_after and t.next0_ is not None and t.next0_.is_hiphen) and (isinstance(t.next0_.next0_, TextToken)) and not t.next0_.next0_.is_whitespace_after) and t.chars.is_cyrillic_letter == t.next0_.next0_.chars.is_cyrillic_letter):
res.end_token = t.next0_.next0_
t = res.end_token
res.name_value = "{0}-{1}".format(res.name_value, t.get_normal_case_text(None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED, False))
return res
return None
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 parse(t : 'Token', max_char : int, prev : 'LineToken') -> 'LineToken':
from pullenti.morph.LanguageHelper import LanguageHelper
from pullenti.ner.NumberToken import NumberToken
from pullenti.ner.TextToken import TextToken
from pullenti.ner.core.BracketHelper import BracketHelper
from pullenti.ner.core.BracketParseAttr import BracketParseAttr
from pullenti.ner.decree.DecreeReferent import DecreeReferent
if (t is None or t.end_char > max_char):
return None
res = ListHelper.LineToken(t, t)
first_pass3272 = True
while True:
if first_pass3272: first_pass3272 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= max_char)): break
if (t.is_char(':')):
if (res.is_newline_before and res.begin_token.is_value("ПРИЛОЖЕНИЕ", "ДОДАТОК")):
res.is_list_head = True
res.end_token = t
break
if (t.is_char(';')):
if (not t.is_whitespace_after):
pass
if (t.previous is not None and (isinstance(t.previous.get_referent(), DecreeReferent))):
if (not t.is_whitespace_after):
continue
if (t.next0_ is not None and (isinstance(t.next0_.get_referent(), DecreeReferent))):
continue
res.is_list_item = True
res.end_token = t
break
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
res.end_token = t
continue
if (t.is_newline_before and t != res.begin_token):
next0__ = True
if (t.previous.is_comma or t.previous.is_and or t.is_char_of("(")):
next0__ = False
elif (t.chars.is_letter or (isinstance(t, NumberToken))):
if (t.chars.is_all_lower):
next0__ = False
elif (t.previous.chars.is_letter):
next0__ = False
if (next0__):
break
res.end_token = t
if (res.begin_token.is_hiphen):
res.is_list_item = (res.begin_token.next0_ is not None and not res.begin_token.next0_.is_hiphen)
elif (res.begin_token.is_char_of("·")):
res.is_list_item = True
res.begin_token = res.begin_token.next0_
elif (res.begin_token.next0_ is not None and ((res.begin_token.next0_.is_char(')') or ((prev is not None and ((prev.is_list_item or prev.is_list_head))))))):
if (res.begin_token.length_char == 1 or (isinstance(res.begin_token, NumberToken))):
res.is_list_item = True
if ((isinstance(res.begin_token, NumberToken)) and res.begin_token.int_value is not None):
res.number = res.begin_token.int_value
elif ((isinstance(res.begin_token, TextToken)) and res.begin_token.length_char == 1):
te = res.begin_token.term
if (LanguageHelper.is_cyrillic_char(te[0])):
res.number = ((ord(te[0])) - (ord('А')))
elif (LanguageHelper.is_latin_char(te[0])):
res.number = ((ord(te[0])) - (ord('A')))
return res
def __calc_rank_and_value(self, min_newlines_count: int) -> bool:
self.rank = 0
if (self.begin_token.chars.is_all_lower):
self.rank -= 30
words = 0
up_words = 0
notwords = 0
line_number = 0
tstart = self.begin_token
tend = self.end_token
t = self.begin_token
first_pass3396 = True
while True:
if first_pass3396: first_pass3396 = False
else: t = t.next0_
if (not (t != self.end_token.next0_ and t is not None
and t.end_char <= self.end_token.end_char)):
break
if (t.is_newline_before):
pass
tit = TitleItemToken.try_attach(t)
if (tit is not None):
if (tit.typ == TitleItemToken.Types.THEME
or tit.typ == TitleItemToken.Types.TYPANDTHEME):
if (t != self.begin_token):
if (line_number > 0):
return False
notwords = 0
up_words = notwords
words = up_words
tstart = tit.end_token.next0_
t = tit.end_token
if (t.next0_ is None):
return False
if (t.next0_.chars.is_letter
and t.next0_.chars.is_all_lower):
self.rank += 20
else:
self.rank += 100
tstart = t.next0_
if (tit.typ == TitleItemToken.Types.TYPANDTHEME):
self.type_value = tit.value
continue
if (tit.typ == TitleItemToken.Types.TYP):
if (t == self.begin_token):
if (tit.end_token.is_newline_after):
self.type_value = tit.value
self.rank += 5
tstart = tit.end_token.next0_
t = tit.end_token
words += 1
if (tit.begin_token != tit.end_token):
words += 1
if (tit.chars.is_all_upper):
up_words += 1
continue
if (tit.typ == TitleItemToken.Types.DUST
or tit.typ == TitleItemToken.Types.SPECIALITY):
if (t == self.begin_token):
return False
self.rank -= 20
if (tit.typ == TitleItemToken.Types.SPECIALITY):
self.speciality = tit.value
t = tit.end_token
continue
if (tit.typ == TitleItemToken.Types.CONSULTANT
or tit.typ == TitleItemToken.Types.BOSS
or tit.typ == TitleItemToken.Types.EDITOR):
t = tit.end_token
if (t.next0_ is not None and
((t.next0_.is_char_of(":") or t.next0_.is_hiphen
or t.whitespaces_after_count > 4))):
self.rank -= 10
else:
self.rank -= 2
continue
return False
blt = BookLinkToken.try_parse(t, 0)
if (blt is not None):
if (blt.typ == BookLinkTyp.MISC or blt.typ == BookLinkTyp.N
or blt.typ == BookLinkTyp.PAGES):
self.rank -= 10
elif (blt.typ == BookLinkTyp.N
or blt.typ == BookLinkTyp.PAGERANGE):
self.rank -= 20
if (t == self.begin_token and BookLinkToken.try_parse_author(
t, FioTemplateType.UNDEFINED) is not None):
self.rank -= 20
if (t.is_newline_before and t != self.begin_token):
line_number += 1
if (line_number > 4):
return False
if (t.chars.is_all_lower):
self.rank += 10
elif (t.previous.is_char('.')):
self.rank -= 10
elif (t.previous.is_char_of(",-")):
self.rank += 10
else:
npt = NounPhraseHelper.try_parse(t.previous,
NounPhraseParseAttr.NO, 0,
None)
if (npt is not None and npt.end_char >= t.end_char):
self.rank += 10
if (t != self.begin_token
and t.newlines_before_count > min_newlines_count):
self.rank -= (t.newlines_before_count - min_newlines_count)
bst = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (bst is not None and bst.is_quote_type
and bst.end_token.end_char <= self.end_token.end_char):
if (words == 0):
tstart = bst.begin_token
self.rank += 10
if (bst.end_token == self.end_token):
tend = self.end_token
self.rank += 10
rli = t.get_referents()
if (rli is not None):
for r in rli:
if (isinstance(r, OrganizationReferent)):
if (t.is_newline_before):
self.rank -= 10
else:
self.rank -= 4
continue
if ((isinstance(r, GeoReferent))
or (isinstance(r, PersonReferent))):
if (t.is_newline_before):
self.rank -= 5
if (t.is_newline_after or t.next0_ is None):
self.rank -= 20
elif (t.next0_.is_hiphen
or (isinstance(t.next0_, NumberToken))
or (isinstance(t.next0_.get_referent(),
DateReferent))):
self.rank -= 20
elif (t != self.begin_token):
self.rank -= 20
continue
if ((isinstance(r, GeoReferent))
or (isinstance(r, DenominationReferent))):
continue
if ((isinstance(r, UriReferent))
or (isinstance(r, PhoneReferent))):
return False
if (t.is_newline_before):
self.rank -= 4
else:
self.rank -= 2
if (t == self.begin_token and (isinstance(
self.end_token.get_referent(), PersonReferent))):
self.rank -= 10
words += 1
if (t.chars.is_all_upper):
up_words += 1
if (t == self.begin_token):
if (t.is_newline_after):
self.rank -= 10
elif (t.next0_ is not None and t.next0_.is_char('.')
and t.next0_.is_newline_after):
self.rank -= 10
continue
if (isinstance(t, NumberToken)):
if (t.typ == NumberSpellingType.WORDS):
words += 1
if (t.chars.is_all_upper):
up_words += 1
else:
notwords += 1
continue
pat = PersonAttrToken.try_attach(
t, None, PersonAttrToken.PersonAttrAttachAttrs.NO)
if (pat is not None):
if (t.is_newline_before):
if (not pat.morph.case_.is_undefined
and not pat.morph.case_.is_nominative):
pass
elif (pat.chars.is_all_upper):
pass
else:
self.rank -= 20
elif (t.chars.is_all_lower):
self.rank -= 1
while t is not None:
words += 1
if (t.chars.is_all_upper):
up_words += 1
if (t == pat.end_token):
break
t = t.next0_
continue
oitt = OrgItemTypeToken.try_attach(t, True, None)
if (oitt is not None):
if (oitt.morph.number != MorphNumber.PLURAL
and not oitt.is_doubt_root_word):
if (not oitt.morph.case_.is_undefined
and not oitt.morph.case_.is_nominative):
words += 1
if (t.chars.is_all_upper):
up_words += 1
else:
self.rank -= 4
if (t == self.begin_token):
self.rank -= 5
else:
words += 1
if (t.chars.is_all_upper):
up_words += 1
t = oitt.end_token
continue
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is not None):
if (tt.is_char('©')):
self.rank -= 10
if (tt.is_char('_')):
self.rank -= 1
if (tt.chars.is_letter):
if (tt.length_char > 2):
words += 1
if (t.chars.is_all_upper):
up_words += 1
elif (not tt.is_char(',')):
notwords += 1
if (tt.is_pure_verb):
self.rank -= 30
words -= 1
break
if (tt == self.end_token):
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
self.rank -= 10
elif (tt.is_char('.')):
self.rank += 5
elif (tt.is_char_of("._")):
self.rank -= 5
self.rank += words
self.rank -= notwords
if ((words < 1) and (self.rank < 50)):
return False
if (tstart is None or tend is None):
return False
if (tstart.end_char > tend.end_char):
return False
tit1 = TitleItemToken.try_attach(self.end_token.next0_)
if (tit1 is not None
and ((tit1.typ == TitleItemToken.Types.TYP
or tit1.typ == TitleItemToken.Types.SPECIALITY))):
if (tit1.end_token.is_newline_after):
self.rank += 15
else:
self.rank += 10
if (tit1.typ == TitleItemToken.Types.SPECIALITY):
self.speciality = tit1.value
if (up_words > 4 and up_words > (math.floor((0.8 * (words))))):
if (tstart.previous is not None and
(isinstance(tstart.previous.get_referent(), PersonReferent))):
self.rank += (5 + up_words)
self.begin_name_token = tstart
self.end_name_token = tend
return True
def __try_attach_contract_ground(t : 'Token', ip : 'InstrumentParticipantReferent', can_be_passport : bool=False) -> 'Token':
ok = False
first_pass3289 = True
while True:
if first_pass3289: first_pass3289 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_char(',') or t.morph.class0_.is_preposition):
continue
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
continue
if (t.is_value("ОСНОВАНИЕ", None) or t.is_value("ДЕЙСТВОВАТЬ", None) or t.is_value("ДЕЙСТВУЮЩИЙ", None)):
ok = True
if (t.next0_ is not None and t.next0_.is_char('(')):
br = BracketHelper.try_parse(t.next0_, BracketParseAttr.NO, 100)
if (br is not None and (br.length_char < 10)):
t = br.end_token
continue
dr = Utils.asObjectOrNull(t.get_referent(), DecreeReferent)
if (dr is not None):
ip.ground = dr
return t
pir = Utils.asObjectOrNull(t.get_referent(), PersonIdentityReferent)
if (pir is not None and can_be_passport):
if (pir.typ is not None and not "паспорт" in pir.typ):
ip.ground = pir
return t
if (t.is_value("УСТАВ", None)):
ip.ground = t.get_normal_case_text(MorphClass.NOUN, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False)
return t
if (t.is_value("ДОВЕРЕННОСТЬ", None)):
dts = DecreeToken.try_attach_list(t.next0_, None, 10, False)
if (dts is None):
has_spec = False
ttt = t.next0_
first_pass3290 = True
while True:
if first_pass3290: first_pass3290 = False
else: ttt = ttt.next0_
if (not (ttt is not None and ((ttt.end_char - t.end_char) < 200))): break
if (ttt.is_comma):
continue
if (ttt.is_value("УДОСТОВЕРИТЬ", None) or ttt.is_value("УДОСТОВЕРЯТЬ", None)):
has_spec = True
continue
dt = DecreeToken.try_attach(ttt, None, False)
if (dt is not None):
if (dt.typ == DecreeToken.ItemType.DATE or dt.typ == DecreeToken.ItemType.NUMBER):
dts = DecreeToken.try_attach_list(ttt, None, 10, False)
break
npt = NounPhraseHelper.try_parse(ttt, NounPhraseParseAttr.NO, 0, None)
if (npt is not None):
if (npt.end_token.is_value("НОТАРИУС", None)):
ttt = npt.end_token
has_spec = True
continue
if (ttt.get_referent() is not None):
if (has_spec):
continue
break
if (dts is not None and len(dts) > 0):
t0 = t
dr = DecreeReferent()
dr.typ = "ДОВЕРЕННОСТЬ"
for d in dts:
if (d.typ == DecreeToken.ItemType.DATE):
dr._add_date(d)
t = d.end_token
elif (d.typ == DecreeToken.ItemType.NUMBER):
dr._add_number(d)
t = d.end_token
else:
break
ad = t.kit.get_analyzer_data_by_analyzer_name(InstrumentAnalyzer.ANALYZER_NAME)
ip.ground = ad.register_referent(dr)
rt = ReferentToken(Utils.asObjectOrNull(ip.ground, Referent), t0, t)
t.kit.embed_token(rt)
return rt
ip.ground = "ДОВЕРЕННОСТЬ"
return t
break
return None
def __correct_tail_attributes(p : 'PersonReferent', t0 : 'Token') -> 'Token':
res = t0
t = t0
if (t is not None and t.is_char(',')):
t = t.next0_
born = False
die = False
if (t is not None and ((t.is_value("РОДИТЬСЯ", "НАРОДИТИСЯ") or t.is_value("BORN", None)))):
t = t.next0_
born = True
elif (t is not None and ((t.is_value("УМЕРЕТЬ", "ПОМЕРТИ") or t.is_value("СКОНЧАТЬСЯ", None) or t.is_value("DIED", None)))):
t = t.next0_
die = True
elif ((t is not None and t.is_value("ДАТА", None) and t.next0_ is not None) and t.next0_.is_value("РОЖДЕНИЕ", "НАРОДЖЕННЯ")):
t = t.next0_.next0_
born = True
while t is not None:
if (t.morph.class0_.is_preposition or t.is_hiphen or t.is_char(':')):
t = t.next0_
else:
break
if (t is not None and t.get_referent() is not None):
r = t.get_referent()
if (r.type_name == "DATE"):
t1 = t
if (t.next0_ is not None and ((t.next0_.is_value("Р", None) or t.next0_.is_value("РОЖДЕНИЕ", "НАРОДЖЕННЯ")))):
born = True
t1 = t.next0_
if (t1.next0_ is not None and t1.next0_.is_char('.')):
t1 = t1.next0_
if (born):
if (p is not None):
p.add_slot(PersonReferent.ATTR_BORN, r, False, 0)
res = t1
t = t1
elif (die):
if (p is not None):
p.add_slot(PersonReferent.ATTR_DIE, r, False, 0)
res = t1
t = t1
if (die and t is not None):
ag = NumberHelper.try_parse_age(t.next0_)
if (ag is not None):
if (p is not None):
p.add_slot(PersonReferent.ATTR_AGE, str(ag.value), False, 0)
t = ag.end_token.next0_
res = ag.end_token
if (t is None):
return res
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t1 = t.next0_
born = False
if (t1.is_value("РОД", None)):
born = True
t1 = t1.next0_
if (t1 is not None and t1.is_char('.')):
t1 = t1.next0_
if (isinstance(t1, ReferentToken)):
r = t1.get_referent()
if (r.type_name == "DATERANGE" and t1.next0_ == br.end_token):
bd = Utils.asObjectOrNull(r.get_slot_value("FROM"), Referent)
to = Utils.asObjectOrNull(r.get_slot_value("TO"), Referent)
if (bd is not None and to is not None):
if (p is not None):
p.add_slot(PersonReferent.ATTR_BORN, bd, False, 0)
p.add_slot(PersonReferent.ATTR_DIE, to, False, 0)
res = br.end_token
t = res
elif (r.type_name == "DATE" and t1.next0_ == br.end_token):
if (p is not None):
p.add_slot(PersonReferent.ATTR_BORN, r, False, 0)
res = br.end_token
t = res
return res
def process(self, kit: 'AnalysisKit') -> None:
ad = kit.get_analyzer_data(self)
is_lit_block = 0
refs_by_num = dict()
rts = []
t = kit.first_token
first_pass3022 = True
while True:
if first_pass3022: first_pass3022 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None and br.length_char > 70
and (br.length_char < 400)):
if (br.is_newline_after
or ((br.end_token.next0_ is not None
and br.end_token.next0_.is_char_of(".;")))):
rts = BookLinkAnalyzer.__try_parse(
t.next0_, False, br.end_char)
if (rts is not None and len(rts) >= 1):
if (len(rts) > 1):
rts[1].referent = ad.register_referent(
rts[1].referent)
kit.embed_token(rts[1])
rts[0].referent.book = Utils.asObjectOrNull(
rts[1].referent, BookLinkReferent)
if (rts[0].begin_char == rts[1].begin_char):
rts[0].begin_token = rts[1]
if (rts[0].end_char == rts[1].end_char):
rts[0].end_token = rts[1]
rts[0].begin_token = t
rts[0].end_token = br.end_token
rts[0].referent.typ = BookLinkRefType.INLINE
rts[0].referent = ad.register_referent(
rts[0].referent)
kit.embed_token(rts[0])
t = (rts[0])
continue
if (not t.is_newline_before):
continue
if (is_lit_block <= 0):
tt = BookLinkToken.parse_start_of_lit_block(t)
if (tt is not None):
is_lit_block = 5
t = tt
continue
rts = BookLinkAnalyzer.__try_parse(t, is_lit_block > 0, 0)
if (rts is None or (len(rts) < 1)):
is_lit_block -= 1
if (is_lit_block < 0):
is_lit_block = 0
continue
is_lit_block += 1
if (is_lit_block > 5):
is_lit_block = 5
if (len(rts) > 1):
rts[1].referent = ad.register_referent(rts[1].referent)
kit.embed_token(rts[1])
rts[0].referent.book = Utils.asObjectOrNull(
rts[1].referent, BookLinkReferent)
if (rts[0].begin_char == rts[1].begin_char):
rts[0].begin_token = rts[1]
if (rts[0].end_char == rts[1].end_char):
rts[0].end_token = rts[1]
re = Utils.asObjectOrNull(rts[0].referent, BookLinkRefReferent)
re = (Utils.asObjectOrNull(ad.register_referent(re),
BookLinkRefReferent))
rts[0].referent = (re)
kit.embed_token(rts[0])
t = (rts[0])
if (re.number is not None):
li = []
wrapli368 = RefOutArgWrapper(None)
inoutres369 = Utils.tryGetValue(refs_by_num, re.number,
wrapli368)
li = wrapli368.value
if (not inoutres369):
li = list()
refs_by_num[re.number] = li
li.append(re)
t = kit.first_token
first_pass3023 = True
while True:
if first_pass3023: first_pass3023 = False
else: t = t.next0_
if (not (t is not None)): break
if (not (isinstance(t, TextToken))):
continue
rt = BookLinkAnalyzer.__try_parse_short_inline(t)
if (rt is None):
continue
re = Utils.asObjectOrNull(rt.referent, BookLinkRefReferent)
li = []
wrapli370 = RefOutArgWrapper(None)
inoutres371 = Utils.tryGetValue(refs_by_num,
Utils.ifNotNull(re.number, ""),
wrapli370)
li = wrapli370.value
if (not inoutres371):
continue
i = 0
i = 0
while i < len(li):
if (t.begin_char < li[i].occurrence[0].begin_char):
break
i += 1
if (i >= len(li)):
continue
re.book = li[i].book
if (re.pages is None):
re.pages = li[i].pages
re.typ = BookLinkRefType.INLINE
re = (Utils.asObjectOrNull(ad.register_referent(re),
BookLinkRefReferent))
rt.referent = (re)
kit.embed_token(rt)
t = (rt)
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_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 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_parse_number_with_postfix(t: 'Token') -> 'NumberExToken':
if (t is None):
return None
t0 = t
is_dollar = None
if (t.length_char == 1 and t.next0_ is not None):
is_dollar = NumberHelper._is_money_char(t)
if ((is_dollar) is not None):
t = t.next0_
nt = Utils.asObjectOrNull(t, NumberToken)
if (nt is None):
if ((not (isinstance(t.previous, NumberToken)) and t.is_char('(')
and (isinstance(t.next0_, NumberToken)))
and t.next0_.next0_ is not None
and t.next0_.next0_.is_char(')')):
toks1 = NumberExHelper._m_postfixes.try_parse(
t.next0_.next0_.next0_, TerminParseAttr.NO)
if (toks1 is not None
and (Utils.valToEnum(toks1.termin.tag, NumberExType))
== NumberExType.MONEY):
nt0 = Utils.asObjectOrNull(t.next0_, NumberToken)
res = NumberExToken._new405(t, toks1.end_token, nt0.value,
nt0.typ, NumberExType.MONEY,
nt0.real_value,
toks1.begin_token.morph)
return NumberExHelper.__correct_money(
res, toks1.begin_token)
tt = Utils.asObjectOrNull(t, TextToken)
if (tt is None or not tt.morph.class0_.is_adjective):
return None
val = tt.term
i = 4
first_pass3038 = True
while True:
if first_pass3038: first_pass3038 = False
else: i += 1
if (not (i < (len(val) - 5))): break
v = val[0:0 + i]
li = NumberHelper._m_nums.find_termins_by_string(
v, tt.morph.language)
if (li is None):
continue
vv = val[i:]
lii = NumberExHelper._m_postfixes.find_termins_by_string(
vv, tt.morph.language)
if (lii is not None and len(lii) > 0):
re = NumberExToken._new406(
t, t, str(li[0].tag), NumberSpellingType.WORDS,
Utils.valToEnum(lii[0].tag, NumberExType), t.morph)
NumberExHelper.__correct_ext_types(re)
return re
break
return None
if (t.next0_ is None and is_dollar is None):
return None
f = nt.real_value
if (math.isnan(f)):
return None
t1 = nt.next0_
if (((t1 is not None and t1.is_char_of(",.")))
or (((isinstance(t1, NumberToken)) and
(t1.whitespaces_before_count < 3)))):
d = 0
tt11 = NumberHelper.try_parse_real_number(nt, False, False)
if (tt11 is not None):
t1 = tt11.end_token.next0_
f = tt11.real_value
if (t1 is None):
if (is_dollar is None):
return None
elif ((t1.next0_ is not None and t1.next0_.is_value("С", "З")
and t1.next0_.next0_ is not None)
and t1.next0_.next0_.is_value("ПОЛОВИНА", None)):
f += 0.5
t1 = t1.next0_.next0_
if (t1 is not None and t1.is_hiphen and t1.next0_ is not None):
t1 = t1.next0_
det = False
altf = f
if (((isinstance(t1, NumberToken)) and t1.previous is not None
and t1.previous.is_hiphen) and t1.int_value == 0
and t1.length_char == 2):
t1 = t1.next0_
if ((t1 is not None and t1.next0_ is not None and t1.is_char('('))
and (((isinstance(t1.next0_, NumberToken))
or t1.next0_.is_value("НОЛЬ", None)))
and t1.next0_.next0_ is not None):
nt1 = Utils.asObjectOrNull(t1.next0_, NumberToken)
val = 0
if (nt1 is not None):
val = nt1.real_value
if (math.floor(f) == math.floor(val)):
ttt = t1.next0_.next0_
if (ttt.is_char(')')):
t1 = ttt.next0_
det = True
if ((isinstance(t1, NumberToken))
and t1.int_value is not None
and t1.int_value == 0):
t1 = t1.next0_
elif (((((isinstance(ttt, NumberToken)) and
(ttt.real_value < 100) and ttt.next0_ is not None)
and ttt.next0_.is_char('/')
and ttt.next0_.next0_ is not None)
and ttt.next0_.next0_.get_source_text() == "100"
and ttt.next0_.next0_.next0_ is not None)
and ttt.next0_.next0_.next0_.is_char(')')):
rest = NumberExHelper.__get_decimal_rest100(f)
if (ttt.int_value is not None and rest == ttt.int_value):
t1 = ttt.next0_.next0_.next0_.next0_
det = True
elif ((ttt.is_value("ЦЕЛЫХ", None) and
(isinstance(ttt.next0_, NumberToken))
and ttt.next0_.next0_ is not None)
and ttt.next0_.next0_.next0_ is not None
and ttt.next0_.next0_.next0_.is_char(')')):
num2 = Utils.asObjectOrNull(ttt.next0_, NumberToken)
altf = num2.real_value
if (ttt.next0_.next0_.is_value("ДЕСЯТЫЙ", None)):
altf /= (10)
elif (ttt.next0_.next0_.is_value("СОТЫЙ", None)):
altf /= (100)
elif (ttt.next0_.next0_.is_value("ТЫСЯЧНЫЙ", None)):
altf /= (1000)
elif (ttt.next0_.next0_.is_value("ДЕСЯТИТЫСЯЧНЫЙ", None)):
altf /= (10000)
elif (ttt.next0_.next0_.is_value("СТОТЫСЯЧНЫЙ", None)):
altf /= (100000)
elif (ttt.next0_.next0_.is_value("МИЛЛИОННЫЙ", None)):
altf /= (1000000)
if (altf < 1):
altf += val
t1 = ttt.next0_.next0_.next0_.next0_
det = True
else:
toks1 = NumberExHelper._m_postfixes.try_parse(
ttt, TerminParseAttr.NO)
if (toks1 is not None):
if ((Utils.valToEnum(
toks1.termin.tag,
NumberExType)) == NumberExType.MONEY):
if (toks1.end_token.next0_ is not None
and toks1.end_token.next0_.is_char(')')):
res = NumberExToken._new407(
t, toks1.end_token.next0_, nt.value,
nt.typ, NumberExType.MONEY, f, altf,
toks1.begin_token.morph)
return NumberExHelper.__correct_money(
res, toks1.begin_token)
res2 = NumberExHelper.try_parse_number_with_postfix(
t1.next0_)
if (res2 is not None and res2.end_token.next0_ is not None
and res2.end_token.next0_.is_char(')')):
res2.begin_token = t
res2.end_token = res2.end_token.next0_
res2.alt_real_value = res2.real_value
res2.real_value = f
NumberExHelper.__correct_ext_types(res2)
if (res2.whitespaces_after_count < 2):
toks2 = NumberExHelper._m_postfixes.try_parse(
res2.end_token.next0_, TerminParseAttr.NO)
if (toks2 is not None):
if ((Utils.valToEnum(
toks2.termin.tag,
NumberExType)) == NumberExType.MONEY):
res2.end_token = toks2.end_token
return res2
elif (nt1 is not None and nt1.typ == NumberSpellingType.WORDS
and nt.typ == NumberSpellingType.DIGIT):
altf = nt1.real_value
ttt = t1.next0_.next0_
if (ttt.is_char(')')):
t1 = ttt.next0_
det = True
if (not det):
altf = f
if ((t1 is not None and t1.is_char('(') and t1.next0_ is not None)
and t1.next0_.is_value("СУММА", None)):
br = BracketHelper.try_parse(t1, BracketParseAttr.NO, 100)
if (br is not None):
t1 = br.end_token.next0_
if (is_dollar is not None):
te = None
if (t1 is not None):
te = t1.previous
else:
t1 = t0
while t1 is not None:
if (t1.next0_ is None):
te = t1
t1 = t1.next0_
if (te is None):
return None
if (te.is_hiphen and te.next0_ is not None):
if (te.next0_.is_value("МИЛЛИОННЫЙ", None)):
f *= (1000000)
altf *= (1000000)
te = te.next0_
elif (te.next0_.is_value("МИЛЛИАРДНЫЙ", None)):
f *= (1000000000)
altf *= (1000000000)
te = te.next0_
if (not te.is_whitespace_after
and (isinstance(te.next0_, TextToken))):
if (te.next0_.is_value("M", None)):
f *= (1000000)
altf *= (1000000)
te = te.next0_
elif (te.next0_.is_value("BN", None)):
f *= (1000000000)
altf *= (1000000000)
te = te.next0_
return NumberExToken._new408(t0, te, "", nt.typ,
NumberExType.MONEY, f, altf,
is_dollar)
if (t1 is None or ((t1.is_newline_before and not det))):
return None
toks = NumberExHelper._m_postfixes.try_parse(t1, TerminParseAttr.NO)
if ((toks is None and det and (isinstance(t1, NumberToken)))
and t1.value == "0"):
toks = NumberExHelper._m_postfixes.try_parse(
t1.next0_, TerminParseAttr.NO)
if (toks is None and t1.is_char('р')):
cou = 10
ttt = t0.previous
first_pass3039 = True
while True:
if first_pass3039: first_pass3039 = False
else:
ttt = ttt.previous
cou -= 1
if (not (ttt is not None and cou > 0)): break
if (ttt.is_value("СУММА", None)
or ttt.is_value("НАЛИЧНЫЙ", None)
or ttt.is_value("БАЛАНС", None)):
pass
elif (ttt.get_referent() is not None
and ttt.get_referent().type_name == "MONEY"):
pass
else:
continue
toks = TerminToken._new409(
t1, t1,
NumberExHelper._m_postfixes.find_termins_by_canonic_text(
"RUB")[0])
if (t1.next0_ is not None and t1.next0_.is_char('.')):
toks.end_token = t1.next0_
ty = Utils.valToEnum(toks.termin.tag, NumberExType)
return NumberExToken._new410(t, toks.end_token, nt.value,
nt.typ, ty, f, altf,
toks.begin_token.morph, "RUB")
if (toks is not None):
t1 = toks.end_token
if (not t1.is_char('.') and t1.next0_ is not None
and t1.next0_.is_char('.')):
if ((isinstance(t1, TextToken)) and t1.is_value(
toks.termin.terms[0].canonical_text, None)):
pass
elif (not t1.chars.is_letter):
pass
else:
t1 = t1.next0_
if (toks.termin.canonic_text == "LTL"):
return None
if (toks.begin_token == t1):
if (t1.morph.class0_.is_preposition
or t1.morph.class0_.is_conjunction):
if (t1.is_whitespace_before and t1.is_whitespace_after):
return None
ty = Utils.valToEnum(toks.termin.tag, NumberExType)
res = NumberExToken._new407(t, t1, nt.value, nt.typ, ty, f, altf,
toks.begin_token.morph)
if (ty != NumberExType.MONEY):
NumberExHelper.__correct_ext_types(res)
return res
return NumberExHelper.__correct_money(res, toks.begin_token)
pfx = NumberExHelper.__attach_spec_postfix(t1)
if (pfx is not None):
pfx.begin_token = t
pfx.value = nt.value
pfx.typ = nt.typ
pfx.real_value = f
pfx.alt_real_value = altf
return pfx
if (t1.next0_ is not None and ((t1.morph.class0_.is_preposition
or t1.morph.class0_.is_conjunction))):
if (t1.is_value("НА", None)):
pass
else:
nn = NumberExHelper.try_parse_number_with_postfix(t1.next0_)
if (nn is not None):
return NumberExToken._new412(t, t, nt.value, nt.typ,
nn.ex_typ, f, altf,
nn.ex_typ2, nn.ex_typ_param)
if (not t1.is_whitespace_after and (isinstance(t1.next0_, NumberToken))
and (isinstance(t1, TextToken))):
term = t1.term
ty = NumberExType.UNDEFINED
if (term == "СМХ" or term == "CMX"):
ty = NumberExType.SANTIMETER
elif (term == "MX" or term == "МХ"):
ty = NumberExType.METER
elif (term == "MMX" or term == "ММХ"):
ty = NumberExType.MILLIMETER
if (ty != NumberExType.UNDEFINED):
return NumberExToken._new413(t, t1, nt.value, nt.typ, ty, f,
altf, True)
return None
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_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_parse_list(t : 'Token', max_count : int=10) -> typing.List['WeaponItemToken']:
tr = WeaponItemToken.try_parse(t, None, False, False)
if (tr is None):
return None
if (tr.typ == WeaponItemToken.Typs.CLASS or tr.typ == WeaponItemToken.Typs.DATE):
return None
tr0 = tr
res = list()
if (len(tr.__inner_tokens) > 0):
res.extend(tr.__inner_tokens)
if (res[0].begin_char > tr.begin_char):
res[0].begin_token = tr.begin_token
res.append(tr)
t = tr.end_token.next0_
if (tr.typ == WeaponItemToken.Typs.NOUN):
while t is not None:
if (t.is_char(':') or t.is_hiphen):
pass
else:
break
t = t.next0_
and_conj = False
first_pass3425 = True
while True:
if first_pass3425: first_pass3425 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_count > 0 and len(res) >= max_count):
break
if (t.is_char(':')):
continue
if (tr0.typ == WeaponItemToken.Typs.NOUN):
if (t.is_hiphen and t.next0_ is not None):
t = t.next0_
tr = WeaponItemToken.try_parse(t, tr0, False, False)
if (tr is None):
if (BracketHelper.can_be_end_of_sequence(t, True, None, False) and t.next0_ is not None):
if (tr0.typ == WeaponItemToken.Typs.MODEL or tr0.typ == WeaponItemToken.Typs.BRAND):
tt1 = t.next0_
if (tt1 is not None and tt1.is_comma):
tt1 = tt1.next0_
tr = WeaponItemToken.try_parse(tt1, tr0, False, False)
if (tr is None and (isinstance(t, ReferentToken))):
rt = Utils.asObjectOrNull(t, ReferentToken)
if (rt.begin_token == rt.end_token and (isinstance(rt.begin_token, TextToken))):
tr = WeaponItemToken.try_parse(rt.begin_token, tr0, False, False)
if (tr is not None and tr.begin_token == tr.end_token):
tr.begin_token = tr.end_token = t
if (tr is None and t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token.next0_
if (tt is not None and tt.is_comma):
tt = tt.next0_
tr = WeaponItemToken.try_parse(tt, tr0, False, False)
if (tr is not None and tr.typ == WeaponItemToken.Typs.NUMBER):
pass
else:
tr = (None)
if (tr is None and t.is_hiphen):
if (tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL):
tr = WeaponItemToken.try_parse(t.next0_, tr0, False, False)
if (tr is None and t.is_comma):
if ((tr0.typ == WeaponItemToken.Typs.NAME or tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL) or tr0.typ == WeaponItemToken.Typs.CLASS or tr0.typ == WeaponItemToken.Typs.DATE):
tr = WeaponItemToken.try_parse(t.next0_, tr0, True, False)
if (tr is not None):
if (tr.typ == WeaponItemToken.Typs.NUMBER):
pass
else:
tr = (None)
if (tr is None):
break
if (t.is_newline_before):
if (tr.typ != WeaponItemToken.Typs.NUMBER):
break
if (len(tr.__inner_tokens) > 0):
res.extend(tr.__inner_tokens)
res.append(tr)
tr0 = tr
t = tr.end_token
if (and_conj):
break
i = 0
while i < (len(res) - 1):
if (res[i].typ == WeaponItemToken.Typs.MODEL and res[i + 1].typ == WeaponItemToken.Typs.MODEL):
res[i].end_token = res[i + 1].end_token
res[i].value = "{0}{1}{2}".format(res[i].value, ('-' if res[i].end_token.next0_ is not None and res[i].end_token.next0_.is_hiphen else ' '), res[i + 1].value)
del res[i + 1]
i -= 1
i += 1
return res
def parse_variants(t0 : 'Token', t1 : 'Token', lev : int, max_count : int=0, regime : 'SentItemType'=SentItemType.UNDEFINED) -> typing.List['Sentence']:
from pullenti.semantic.internal.SentItem import SentItem
if ((t0 is None or t1 is None or t0.end_char > t1.end_char) or lev > 100):
return None
res = list()
sent = Sentence()
t = t0
first_pass3463 = True
while True:
if first_pass3463: first_pass3463 = False
else: t = t.next0_
if (not (t is not None and t.end_char <= t1.end_char)): break
if (t.is_char('(')):
br = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (br is not None):
t = br.end_token
continue
items_ = SentItem.parse_near_items(t, t1, lev + 1, sent.items)
if (items_ is None or len(items_) == 0):
continue
if (len(items_) == 1 or ((max_count > 0 and len(res) > max_count))):
sent.items.append(items_[0])
t = items_[0].end_token
if (regime != SentItemType.UNDEFINED):
it = items_[0]
if (it.can_be_noun):
pass
elif (it.typ == SentItemType.DELIM):
break
elif (it.typ == SentItemType.VERB):
if (regime == SentItemType.PARTBEFORE):
break
continue
m_nexts = dict()
for it in items_:
nexts = None
wrapnexts2942 = RefOutArgWrapper(None)
inoutres2943 = Utils.tryGetValue(m_nexts, it.end_token.end_char, wrapnexts2942)
nexts = wrapnexts2942.value
if (not inoutres2943):
nexts = Sentence.parse_variants(it.end_token.next0_, t1, lev + 1, max_count, SentItemType.UNDEFINED)
m_nexts[it.end_token.end_char] = nexts
if (nexts is None or len(nexts) == 0):
se = Sentence()
for itt in sent.items:
itt1 = SentItem(None)
itt1.copy_from(itt)
se.items.append(itt1)
itt0 = SentItem(None)
itt0.copy_from(it)
se.items.append(itt0)
res.append(se)
else:
for sn in nexts:
se = Sentence()
for itt in sent.items:
itt1 = SentItem(None)
itt1.copy_from(itt)
se.items.append(itt1)
itt0 = SentItem(None)
itt0.copy_from(it)
se.items.append(itt0)
for itt in sn.items:
itt1 = SentItem(None)
itt1.copy_from(itt)
se.items.append(itt1)
res.append(se)
return res
if (len(sent.items) == 0):
return None
res.append(sent)
return res