def __str__(self) -> str:
if (self.begin_name_token is None or self.end_name_token is None):
return "?"
mt = MetaToken(self.begin_name_token, self.end_name_token)
if (self.type_value is None):
return "{0}: {1}".format(self.rank, str(mt))
else:
return "{0}: {1} ({2})".format(self.rank, str(mt), self.type_value)
def try_attach_nord_west(t: 'Token') -> 'MetaToken':
if (not (isinstance(t, TextToken))):
return None
tok = MiscLocationHelper.__m_nords.try_parse(t, TerminParseAttr.NO)
if (tok is None):
return None
res = MetaToken._new509(t, t, t.morph)
t1 = None
if ((t.next0_ is not None and t.next0_.is_hiphen
and not t.is_whitespace_after) and not t.is_whitespace_after):
t1 = t.next0_.next0_
elif (t.morph.class0_.is_adjective
and (t.whitespaces_after_count < 2)):
t1 = t.next0_
if (t1 is not None):
tok = MiscLocationHelper.__m_nords.try_parse(
t1, TerminParseAttr.NO)
if ((tok) is not None):
res.end_token = tok.end_token
res.morph = tok.morph
return res
def __deserialize_token(stream : Stream, kit : 'AnalysisKit', vers : int) -> 'Token':
from pullenti.ner.MetaToken import MetaToken
from pullenti.ner.ReferentToken import ReferentToken
typ = SerializerHelper.deserialize_short(stream)
if (typ == (0)):
return None
t = None
if (typ == (1)):
t = (TextToken(None, kit))
elif (typ == (2)):
t = (NumberToken(None, None, None, NumberSpellingType.DIGIT, kit))
elif (typ == (3)):
t = (ReferentToken(None, None, None, kit))
else:
t = (MetaToken(None, None, kit))
t._deserialize(stream, kit, vers)
if (isinstance(t, MetaToken)):
tt = SerializerHelper.deserialize_tokens(stream, kit, vers)
if (tt is not None):
t._m_begin_token = tt
while tt is not None:
t._m_end_token = tt
tt = tt.next0_
return t
def tryAttachNordWest(t : 'Token') -> 'MetaToken':
""" Выделение существительных и прилагательных типа "северо-западное", "южное"
Args:
t(Token):
"""
if (not ((isinstance(t, TextToken)))):
return None
tok = MiscLocationHelper.__m_nords.tryParse(t, TerminParseAttr.NO)
if (tok is None):
return None
res = MetaToken._new561(t, t, t.morph)
t1 = None
if ((t.next0_ is not None and t.next0_.is_hiphen and not t.is_whitespace_after) and not t.is_whitespace_after):
t1 = t.next0_.next0_
elif (t.morph.class0_.is_adjective and (t.whitespaces_after_count < 2)):
t1 = t.next0_
if (t1 is not None):
tok = MiscLocationHelper.__m_nords.tryParse(t1, TerminParseAttr.NO)
if ((tok) is not None):
res.end_token = tok.end_token
res.morph = tok.morph
return res
def tryParse(t: 'Token',
add_units: 'TerminCollection',
can_be_set: bool = True,
can_units_absent: bool = False) -> 'MeasureToken':
""" Выделение вместе с наименованием
Args:
t(Token):
"""
if (not ((isinstance(t, TextToken)))):
return None
if (t.is_table_control_char):
return None
t0 = t
whd = None
minmax = 0
wrapminmax1516 = RefOutArgWrapper(minmax)
tt = NumbersWithUnitToken._isMinOrMax(t0, wrapminmax1516)
minmax = wrapminmax1516.value
if (tt is not None):
t = tt.next0_
npt = NounPhraseHelper.tryParse(
t,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt is None):
whd = NumbersWithUnitToken._tryParseWHL(t)
if (whd is not None):
npt = NounPhraseToken(t0, whd.end_token)
elif (t0.isValue("КПД", None)):
npt = NounPhraseToken(t0, t0)
elif ((isinstance(t0, TextToken)) and t0.length_char > 3
and t0.getMorphClassInDictionary().is_undefined):
npt = NounPhraseToken(t0, t0)
else:
return None
elif (NumberHelper.tryParseRealNumber(t, True) is not None):
return None
else:
dtok = DateItemToken.tryAttach(t, None)
if (dtok is not None):
return None
t1 = npt.end_token
t = npt.end_token
name_ = MetaToken._new561(npt.begin_token, npt.end_token, npt.morph)
units = None
units2 = None
internals_ = list()
not0_ = False
tt = t1.next0_
first_pass3037 = True
while True:
if first_pass3037: first_pass3037 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
break
if (tt.is_table_control_char):
break
wrapminmax1510 = RefOutArgWrapper(minmax)
tt2 = NumbersWithUnitToken._isMinOrMax(tt, wrapminmax1510)
minmax = wrapminmax1510.value
if (tt2 is not None):
tt = tt2
t = tt
t1 = t
continue
if ((tt.isValue("БЫТЬ", None) or tt.isValue("ДОЛЖЕН", None)
or tt.isValue("ДОЛЖНЫЙ", None)) or tt.isValue("МОЖЕТ", None)
or
((tt.isValue("СОСТАВЛЯТЬ", None)
and not tt.getMorphClassInDictionary().is_adjective))):
t = tt
t1 = t
if (tt.previous.isValue("НЕ", None)):
not0_ = True
continue
www = NumbersWithUnitToken._tryParseWHL(tt)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
continue
if (len(internals_) > 0 and tt.is_comma_and):
continue
if (tt.isValue("ПРИ", None) or len(internals_) > 0):
mt1 = MeasureToken.tryParse(tt.next0_, add_units, False, False)
if (mt1 is not None and mt1.reliable):
internals_.append(mt1)
tt = mt1.end_token
t = tt
t1 = t
continue
if ((isinstance(tt, NumberToken))
and (tt).typ == NumberSpellingType.WORDS):
npt3 = NounPhraseHelper.tryParse(
tt, NounPhraseParseAttr.PARSENUMERICASADJECTIVE, 0)
if (npt3 is not None):
tt = npt3.end_token
t1 = tt
if (len(internals_) == 0):
name_.end_token = t1
continue
mt0 = NumbersWithUnitToken.tryParse(tt, add_units, False, False)
if (mt0 is not None):
break
if (((tt.is_comma or tt.isChar('('))) and tt.next0_ is not None):
www = NumbersWithUnitToken._tryParseWHL(tt.next0_)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
if (tt.next0_ is not None and tt.next0_.is_comma):
tt = tt.next0_
t1 = tt
if (tt.next0_ is not None and tt.next0_.isChar(')')):
tt = tt.next0_
t1 = tt
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
t = uu[len(uu) - 1].end_token
t1 = t
units = uu
if (tt.isChar('(') and t1.next0_ is not None
and t1.next0_.isChar(')')):
tt = t1.next0_
t = tt
t1 = t
continue
elif (t1.next0_ is not None and t1.next0_.isChar('(')):
uu = UnitToken.tryParseList(t1.next0_.next0_,
add_units, False)
if (uu is not None and uu[len(uu) - 1].end_token.next0_
is not None and
uu[len(uu) - 1].end_token.next0_.isChar(')')):
units2 = uu
tt = uu[len(uu) - 1].end_token.next0_
t = tt
t1 = t
continue
if (uu is not None and len(uu) > 0 and not uu[0].is_doubt):
break
if (BracketHelper.canBeStartOfSequence(tt, False, False)):
br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
t = tt
t1 = t
continue
if (tt.isValue("НЕ", None) and tt.next0_ is not None):
mc = tt.next0_.getMorphClassInDictionary()
if (mc.is_adverb or mc.is_misc):
break
continue
if (tt.isValue("ЯМЗ", None)):
pass
npt2 = NounPhraseHelper.tryParse(
tt,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt2 is None):
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
to = NumbersWithUnitToken.M_TERMINS.tryParse(
tt, TerminParseAttr.NO)
if (to is not None):
if ((isinstance(to.end_token.next0_, TextToken))
and to.end_token.next0_.is_letters):
pass
else:
break
t1 = tt
continue
mc = tt.getMorphClassInDictionary()
if (((isinstance(tt, TextToken)) and tt.chars.is_letter
and tt.length_char > 1)
and (((tt.chars.is_all_upper or mc.is_adverb
or mc.is_undefined) or mc.is_adjective))):
uu = UnitToken.tryParseList(tt, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = tt
continue
if (tt.is_comma):
continue
if (tt.isChar('.')):
if (not MiscHelper.canBeStartOfSentence(tt.next0_)):
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
break
tt = npt2.end_token
t = tt
t1 = t
if (len(internals_) > 0):
pass
elif (t.isValue("ПРЕДЕЛ", None) or t.isValue("ГРАНИЦА", None)
or t.isValue("ДИАПАЗОН", None)):
pass
elif (t.chars.is_letter):
name_.end_token = t1
t1 = t1.next0_
first_pass3038 = True
while True:
if first_pass3038: first_pass3038 = False
else: t1 = t1.next0_
if (not (t1 is not None)): break
if (t1.is_table_control_char):
pass
elif (t1.isCharOf(":,_")):
www = NumbersWithUnitToken._tryParseWHL(t1.next0_)
if (www is not None):
whd = www
t = www.end_token
t1 = t
continue
elif (t1.is_hiphen and t1.is_whitespace_after
and t1.is_whitespace_before):
pass
else:
break
if (t1 is None):
return None
mts = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, not0_)
if (mts is None):
return None
mt = mts[0]
if (name_.begin_token.morph.class0_.is_preposition):
name_.begin_token = name_.begin_token.next0_
if (len(mts) > 1 and len(internals_) == 0):
if (len(mt.units) == 0):
if (units is not None):
for m in mts:
m.units = units
res1 = MeasureToken._new1511(t0, mts[len(mts) - 1].end_token,
name_.morph, True)
res1.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
k = 0
while k < len(mts):
ttt = MeasureToken._new1506(mts[k].begin_token,
mts[k].end_token, mts[k])
if (whd is not None):
nams = Utils.asObjectOrNull(whd.tag, list)
if (k < len(nams)):
ttt.name = nams[k]
res1.internals.append(ttt)
k += 1
tt1 = res1.end_token.next0_
if (tt1 is not None and tt1.isChar('±')):
nn = NumbersWithUnitToken._tryParse(tt1, add_units, True,
False)
if (nn is not None and nn.plus_minus_percent):
res1.end_token = nn.end_token
res1.nums = nn
return res1
if (not mt.is_whitespace_before):
if (mt.begin_token.previous is None):
return None
if (mt.begin_token.previous.isCharOf(":),")
or mt.begin_token.previous.is_table_control_char):
pass
else:
return None
if (len(mt.units) == 0 and units is not None):
mt.units = units
if (mt.div_num is not None and len(units) > 1
and len(mt.div_num.units) == 0):
i = 1
while i < len(units):
if (units[i].pow0_ == -1):
j = i
while j < len(units):
mt.div_num.units.append(units[j])
units[j].pow0_ = (-units[j].pow0_)
j += 1
del mt.units[i:i + len(units) - i]
break
i += 1
if ((minmax < 0) and mt.single_val is not None):
mt.from_val = mt.single_val
mt.from_include = True
mt.single_val = (None)
if (minmax > 0 and mt.single_val is not None):
mt.to_val = mt.single_val
mt.to_include = True
mt.single_val = (None)
if (len(mt.units) == 0):
units = UnitToken.tryParseList(mt.end_token.next0_, add_units,
True)
if (units is None):
if (can_units_absent):
pass
else:
return None
else:
mt.units = units
res = MeasureToken._new1513(t0, mt.end_token, name_.morph, internals_)
if (((not t0.is_whitespace_before and t0.previous is not None
and t0 == name_.begin_token) and t0.previous.is_hiphen
and not t0.previous.is_whitespace_before)
and (isinstance(t0.previous.previous, TextToken))):
name_.begin_token = res.begin_token = name_.begin_token.previous.previous
res.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
res.nums = mt
for u in res.nums.units:
if (u.keyword is not None):
if (u.keyword.begin_char >= res.begin_char):
res.reliable = True
res.__parseInternals(add_units)
if (len(res.internals) > 0 or not can_be_set):
return res
t1 = res.end_token.next0_
if (t1 is not None and t1.is_comma_and):
t1 = t1.next0_
mts1 = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, False)
if ((mts1 is not None and len(mts1) == 1 and
(t1.whitespaces_before_count < 3)) and len(mts1[0].units) > 0
and not UnitToken.canBeEquals(mts[0].units, mts1[0].units)):
res.is_set = True
res.nums = (None)
res.internals.append(
MeasureToken._new1506(mt.begin_token, mt.end_token, mt))
res.internals.append(
MeasureToken._new1506(mts1[0].begin_token, mts1[0].end_token,
mts1[0]))
res.end_token = mts1[0].end_token
return res
def 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 _tryParseWHL(t: 'Token') -> 'MetaToken':
""" Это распознавание написаний ГхШхВ
Args:
t(Token):
"""
if (not ((isinstance(t, TextToken)))):
return None
if (t.isCharOf(":-")):
re0 = NumbersWithUnitToken._tryParseWHL(t.next0_)
if (re0 is not None):
return re0
if (t.isCharOf("(")):
re0 = NumbersWithUnitToken._tryParseWHL(t.next0_)
if (re0 is not None):
if (re0.end_token.next0_ is not None
and re0.end_token.next0_.isChar(')')):
re0.begin_token = t
re0.end_token = re0.end_token.next0_
return re0
txt = (t).term
nams = None
if (len(txt) == 5 and txt[1] == 'Х' and txt[3] == 'Х'):
nams = list()
for i in range(3):
ch = txt[i * 2]
if (ch == 'Г'):
nams.append("ГЛУБИНА")
elif (ch == 'В'):
nams.append("ВЫСОТА")
elif (ch == 'Ш'):
nams.append("ШИРИНА")
elif (ch == 'Д'):
nams.append("ДЛИНА")
else:
return None
return MetaToken._new809(t, t, nams)
t0 = t
t1 = t
while t is not None:
if (not ((isinstance(t, TextToken)))
or ((t.whitespaces_before_count > 1 and t != t0))):
break
term = (t).term
nam = None
if ((t.isValue("ДЛИНА", None) or t.isValue("ДЛИННА", None)
or term == "Д") or term == "ДЛ" or term == "ДЛИН"):
nam = "ДЛИНА"
elif ((t.isValue("ШИРИНА", None) or t.isValue("ШИРОТА", None)
or term == "Ш") or term == "ШИР" or term == "ШИРИН"):
nam = "ШИРИНА"
elif ((t.isValue("ГЛУБИНА", None) or term == "Г" or term == "ГЛ")
or term == "ГЛУБ"):
nam = "ГЛУБИНА"
elif (t.isValue("ВЫСОТА", None) or term == "В" or term == "ВЫС"):
nam = "ВЫСОТА"
else:
break
if (nams is None):
nams = list()
nams.append(nam)
t1 = t
if (t.next0_ is not None and t.next0_.isChar('.')):
t = t.next0_
t1 = t
if (t.next0_ is None):
break
if (MeasureHelper.isMultChar(t.next0_) or t.next0_.is_comma
or t.next0_.isCharOf("\\/")):
t = t.next0_
t = t.next0_
if (nams is None or (len(nams) < 2)):
return None
return MetaToken._new809(t0, t1, nams)
def __tryParseThesis(t: 'Token') -> 'ReferentToken':
if (t is None):
return None
t0 = t
tt = t
mc = tt.getMorphClassInDictionary()
preamb = None
if (mc.is_conjunction):
return None
if (t.isValue("LET", None)):
return None
if (mc.is_preposition or mc.is_misc or mc.is_adverb):
if (not MiscHelper.isEngArticle(tt)):
tt = tt.next0_
first_pass2871 = True
while True:
if first_pass2871: first_pass2871 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_comma):
break
if (tt.isChar('(')):
br = BracketHelper.tryParse(tt, BracketParseAttr.NO,
100)
if (br is not None):
tt = br.end_token
continue
if (MiscHelper.canBeStartOfSentence(tt)):
break
npt0 = NounPhraseHelper.tryParse(
tt,
Utils.valToEnum(
(NounPhraseParseAttr.PARSENUMERICASADJECTIVE) |
(NounPhraseParseAttr.REFERENTCANBENOUN),
NounPhraseParseAttr), 0)
if (npt0 is not None):
tt = npt0.end_token
continue
if (tt.getMorphClassInDictionary().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.getMorphClassInDictionary()
npt = NounPhraseHelper.tryParse(
tt,
Utils.valToEnum((NounPhraseParseAttr.PARSENUMERICASADJECTIVE) |
(NounPhraseParseAttr.REFERENTCANBENOUN) |
(NounPhraseParseAttr.PARSEADVERBS),
NounPhraseParseAttr), 0)
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.canBeStartOfSentence(t2)
or not ((isinstance(t2, TextToken)))):
return None
if (not t2.getMorphClassInDictionary().is_verb):
return None
t3 = t2
tt = t2.next0_
while tt is not None:
if (not tt.getMorphClassInDictionary().is_verb):
break
tt = tt.next0_
first_pass2872 = True
while True:
if first_pass2872: first_pass2872 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.next0_ is None):
t3 = tt
break
if (tt.isCharOf(".;!?")):
if (MiscHelper.canBeStartOfSentence(tt.next0_)):
t3 = tt
break
if (not ((isinstance(tt, TextToken)))):
continue
if (BracketHelper.canBeStartOfSequence(tt, False, False)):
br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
continue
tt = t3
if (t3.isCharOf(";.!?")):
tt = tt.previous
txt = MiscHelper.getTextValue(
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.getTextValue(
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.isEngArticle(tt)):
tt = tt.next0_
nam = MiscHelper.getTextValue(tt, t2.previous, GetTextAttr.KEEPQUOTES)
if (nam.startswith("SO-CALLED")):
nam = nam[9:].strip()
dr = DefinitionReferent()
dr.kind = DefinitionKind.ASSERTATION
dr.addSlot(DefinitionReferent.ATTR_TERMIN, nam, False, 0)
dr.addSlot(DefinitionReferent.ATTR_VALUE, txt, False, 0)
return ReferentToken(dr, t0, t3)
def parse_near_items(t : 'Token', t1 : 'Token', lev : int, prev : typing.List['SentItem']) -> typing.List['SentItem']:
if (lev > 100):
return None
if (t is None or t.begin_char > t1.end_char):
return None
res = list()
if (isinstance(t, ReferentToken)):
res.append(SentItem(Utils.asObjectOrNull(t, MetaToken)))
return res
delim = DelimToken.try_parse(t)
if (delim is not None):
res.append(SentItem(delim))
return res
conj = ConjunctionHelper.try_parse(t)
if (conj is not None):
res.append(SentItem(conj))
return res
prep_ = PrepositionHelper.try_parse(t)
t111 = (t if prep_ is None else prep_.end_token.next0_)
if ((isinstance(t111, NumberToken)) and ((t111.morph.class0_.is_adjective and not t111.morph.class0_.is_noun))):
t111 = (None)
num = (None if t111 is None else NumbersWithUnitToken.try_parse(t111, None, False, False, False, False))
if (num is not None):
if (len(num.units) == 0):
npt1 = NounPhraseHelper.try_parse(num.end_token.next0_, SentItem.__m_npt_attrs, 0, None)
if (npt1 is None and num.end_token.next0_ is not None and num.end_token.next0_.is_value("РАЗ", None)):
npt1 = NounPhraseToken(num.end_token.next0_, num.end_token.next0_)
npt1.noun = MetaToken(num.end_token.next0_, num.end_token.next0_)
if (npt1 is not None and prep_ is not None):
if (npt1.noun.end_token.is_value("РАЗ", None)):
npt1.morph.remove_items(prep_.next_case, False)
elif (((npt1.morph.case_) & prep_.next_case).is_undefined):
npt1 = (None)
else:
npt1.morph.remove_items(prep_.next_case, False)
if ((npt1 is not None and npt1.end_token.is_value("ОНИ", None) and npt1.preposition is not None) and npt1.preposition.normal == "ИЗ"):
npt1.morph = MorphCollection(num.end_token.morph)
npt1.preposition = (None)
nn = str(num)
si1 = SentItem(npt1)
if (nn == "1" and (isinstance(num.end_token, NumberToken)) and num.end_token.end_token.is_value("ОДИН", None)):
a = SemAttribute._new2946(SemAttributeType.ONEOF, num.end_token.end_token.get_normal_case_text(None, MorphNumber.SINGULAR, MorphGender.UNDEFINED, False))
aex = SemAttributeEx._new2945(num, a)
si1.attrs = list()
si1.attrs.append(aex)
else:
si1.quant = SemQuantity(nn, num.begin_token, num.end_token)
if (prep_ is not None):
si1.prep = prep_.normal
res.append(si1)
return res
if (npt1 is not None):
si1 = SentItem._new2948(npt1, SemQuantity(str(num), num.begin_token, num.end_token))
if (prep_ is not None):
si1.prep = prep_.normal
if (npt1.end_token.is_value("РАЗ", None)):
si1.typ = SentItemType.FORMULA
if (((npt1.morph.number) & (MorphNumber.PLURAL)) == (MorphNumber.UNDEFINED) and si1.quant.spelling != "1"):
ok = False
if (si1.quant.spelling.endswith("1")):
ok = True
elif (si1.typ == SentItemType.FORMULA):
ok = True
elif (si1.quant.spelling.endswith("2") and npt1.morph.case_.is_genitive):
ok = True
elif (si1.quant.spelling.endswith("3") and npt1.morph.case_.is_genitive):
ok = True
elif (si1.quant.spelling.endswith("4") and npt1.morph.case_.is_genitive):
ok = True
if (ok):
npt1.morph = MorphCollection()
npt1.morph.number = MorphNumber.PLURAL
res.append(si1)
return res
num.begin_token = t
num.morph = MorphCollection(num.end_token.morph)
si = SentItem(num)
if (prep_ is not None):
si.prep = prep_.normal
res.append(si)
if (si.prep == "НА"):
aa = AdverbToken.try_parse(si.end_token.next0_)
if (aa is not None and ((aa.typ == SemAttributeType.LESS or aa.typ == SemAttributeType.GREAT))):
si.add_attr(aa)
si.end_token = aa.end_token
return res
mc = t.get_morph_class_in_dictionary()
adv = AdverbToken.try_parse(t)
npt = NounPhraseHelper.try_parse(t, SentItem.__m_npt_attrs, 0, None)
if (npt is not None and (isinstance(npt.end_token, TextToken)) and npt.end_token.term == "БЫЛИ"):
npt = (None)
if (npt is not None and adv is not None):
if (adv.end_char > npt.end_char):
npt = (None)
elif (adv.end_char == npt.end_char):
res.append(SentItem(npt))
res.append(SentItem(adv))
return res
if (npt is not None and len(npt.adjectives) == 0):
if (npt.end_token.is_value("КОТОРЫЙ", None) and t.previous is not None and t.previous.is_comma_and):
res1 = SentItem.__parse_subsent(npt, t1, lev + 1, prev)
if (res1 is not None):
return res1
if (npt.end_token.is_value("СКОЛЬКО", None)):
tt1 = npt.end_token.next0_
if (tt1 is not None and tt1.is_value("ВСЕГО", None)):
tt1 = tt1.next0_
npt1 = NounPhraseHelper.try_parse(tt1, NounPhraseParseAttr.NO, 0, None)
if (npt1 is not None and not npt1.morph.case_.is_undefined and prep_ is not None):
if (((prep_.next_case) & npt1.morph.case_).is_undefined):
npt1 = (None)
else:
npt1.morph.remove_items(prep_.next_case, False)
if (npt1 is not None):
npt1.begin_token = npt.begin_token
npt1.preposition = npt.preposition
npt1.adjectives.append(MetaToken(npt.end_token, npt.end_token))
npt = npt1
if (npt.end_token.morph.class0_.is_adjective):
if (VerbPhraseHelper.try_parse(t, True, False, False) is not None):
npt = (None)
vrb = None
if (npt is not None and len(npt.adjectives) > 0):
vrb = VerbPhraseHelper.try_parse(t, True, False, False)
if (vrb is not None and vrb.first_verb.is_participle):
npt = (None)
elif (adv is None or npt is not None):
vrb = VerbPhraseHelper.try_parse(t, True, False, False)
if (npt is not None):
res.append(SentItem(npt))
if (vrb is not None and not vrb.first_verb.is_participle and not vrb.first_verb.is_dee_participle):
vars0_ = list()
for wf in vrb.first_verb.morph.items:
if (wf.class0_.is_verb and (isinstance(wf, MorphWordForm)) and wf.is_in_dictionary):
vars0_.append(Utils.asObjectOrNull(wf, MorphWordForm))
if (len(vars0_) < 2):
res.append(SentItem(vrb))
else:
vrb.first_verb.verb_morph = vars0_[0]
res.append(SentItem(vrb))
i = 1
while i < len(vars0_):
vrb = VerbPhraseHelper.try_parse(t, False, False, False)
if (vrb is None):
break
vrb.first_verb.verb_morph = vars0_[i]
res.append(SentItem(vrb))
i += 1
if (vars0_[0].misc.mood == MorphMood.IMPERATIVE and vars0_[1].misc.mood != MorphMood.IMPERATIVE):
rr = res[0]
res[0] = res[1]
res[1] = rr
return res
if (vrb is not None):
res1 = SentItem.__parse_participles(vrb, t1, lev + 1)
if (res1 is not None):
res.extend(res1)
if (len(res) > 0):
return res
if (adv is not None):
if (adv.typ == SemAttributeType.OTHER):
npt1 = NounPhraseHelper.try_parse(adv.end_token.next0_, SentItem.__m_npt_attrs, 0, None)
if (npt1 is not None and npt1.end_token.is_value("ОНИ", None) and npt1.preposition is not None):
si1 = SentItem(npt1)
a = SemAttribute._new2946(SemAttributeType.OTHER, adv.end_token.get_normal_case_text(None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED, False))
aex = SemAttributeEx._new2945(num, a)
si1.attrs = list()
si1.attrs.append(aex)
if (prep_ is not None):
si1.prep = prep_.normal
res.append(si1)
return res
for i in range(len(prev) - 1, -1, -1):
if (prev[i].attrs is not None):
for a in prev[i].attrs:
if (a.attr.typ == SemAttributeType.ONEOF):
si1 = SentItem(prev[i].source)
aa = SemAttribute._new2946(SemAttributeType.OTHER, adv.end_token.get_normal_case_text(None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED, False))
aex = SemAttributeEx._new2945(adv, aa)
si1.attrs = list()
si1.attrs.append(aex)
if (prep_ is not None):
si1.prep = prep_.normal
si1.begin_token = adv.begin_token
si1.end_token = adv.end_token
res.append(si1)
return res
res.append(SentItem(adv))
return res
if (mc.is_adjective):
npt = NounPhraseToken._new2953(t, t, MorphCollection(t.morph))
npt.noun = MetaToken(t, t)
res.append(SentItem(npt))
return res
return None
def checkNds(t : 'Token', nds : float=18, nds_mustbe_money : bool=False) -> 'MetaToken':
""" Проверка корректности НДС для суммы
Args:
t(Token): Указывает на значение, для которой должно далее следовать НДС
nds(float):
"""
if (t is None or nds <= 0):
return None
m = Utils.asObjectOrNull(t.getReferent(), MoneyReferent)
if (m is None):
return None
has_nds = False
has_nds_perc = False
has_all = False
incl = False
m1 = None
ndst0 = None
ndst1 = None
tt = t.next0_
first_pass2835 = True
while True:
if first_pass2835: first_pass2835 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.isValue("НДС", None)):
has_nds = True
ndst1 = tt
ndst0 = ndst1
continue
if (isinstance(tt, ReferentToken)):
m1 = (Utils.asObjectOrNull(tt.getReferent(), MoneyReferent))
break
if (isinstance(tt, NumberToken)):
ne = NumberHelper.tryParseNumberWithPostfix(tt)
if (ne is not None and ne.ex_typ == NumberExType.PERCENT):
if (math.fabs(ne.real_value - nds) > .0001):
ok = False
if (has_nds):
ok = True
if (ok):
return MetaToken._new809(tt, ne.end_token, "Размер НДС должен быть {0}%, а не {1}%".format(nds, ne.real_value))
ndst1 = ne.end_token
tt = ndst1
has_nds_perc = True
continue
if (tt.isValue("ВСЕГО", None)):
has_all = True
continue
if (tt.isValue("ТОМ", None) or tt.isValue("ЧИСЛО", None) or tt.isValue("ВКЛЮЧАЯ", None)):
incl = True
continue
if ((tt.isValue("КРОМЕ", None) or tt.isValue("ТОГО", None) or tt.isValue("РАЗМЕР", None)) or tt.isValue("СУММА", None) or tt.isValue("СТАВКА", None)):
continue
if (((tt.isValue("Т", None) and tt.next0_ is not None and tt.next0_.isChar('.')) and tt.next0_.next0_ is not None and tt.next0_.next0_.isValue("Ч", None)) and tt.next0_.next0_.next0_ is not None and tt.next0_.next0_.next0_.isChar('.')):
incl = True
tt = tt.next0_.next0_.next0_
continue
if (not tt.chars.is_letter or tt.morph.class0_.is_preposition):
continue
break
if (not has_nds):
return None
if (m1 is None):
if (nds_mustbe_money):
return MetaToken._new809(ndst0, ndst1, "Размер НДС должен быть в денежном выражении")
return None
if (has_all):
return None
must_be = m.real_value
must_be = (must_be * ((nds / (100))))
if (incl):
must_be /= (((1) + ((nds / (100)))))
dd = must_be * (100)
dd -= (math.floor(dd))
dd /= (100)
must_be -= dd
if (dd >= .005):
must_be += .01
real = m1.real_value
delta = must_be - real
if (delta < 0):
delta = (- delta)
if (delta > .011):
if ((delta < 1) and m1.rest == 0 and m.rest == 0):
pass
else:
mr = MoneyReferent._new811(m1.currency, must_be)
return MetaToken._new809(t, tt, "Размер НДС должен быть {0}, а не {1}".format(DecreeHelper.__outMoney(mr), DecreeHelper.__outMoney(m1)))
if (incl):
return None
m2 = None
has_all = False
tt = tt.next0_
first_pass2836 = True
while True:
if first_pass2836: first_pass2836 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (isinstance(tt, ReferentToken)):
m2 = (Utils.asObjectOrNull(tt.getReferent(), MoneyReferent))
break
if (not tt.chars.is_letter or tt.morph.class0_.is_preposition):
continue
if (tt.isValue("ВСЕГО", None)):
has_all = True
continue
if (tt.isValue("НДС", None) or tt.isValue("ВМЕСТЕ", None)):
continue
break
if (m2 is not None and has_all):
must_be = (m.real_value + m1.real_value)
delta = (must_be - m2.real_value)
if (delta < 0):
delta = (- delta)
if (delta > .01):
mr = MoneyReferent._new811(m1.currency, must_be)
err = "Всего с НДС должно быть {0}, а не {1}".format(DecreeHelper.__outMoney(mr), DecreeHelper.__outMoney(m2))
return MetaToken._new809(t, tt, err)
return None
def _try_parsewhl(t : 'Token') -> 'MetaToken':
if (not (isinstance(t, TextToken))):
return None
if (t.is_char_of(":-")):
re0 = NumbersWithUnitToken._try_parsewhl(t.next0_)
if (re0 is not None):
return re0
if (t.is_char_of("(")):
re0 = NumbersWithUnitToken._try_parsewhl(t.next0_)
if (re0 is not None):
if (re0.end_token.next0_ is not None and re0.end_token.next0_.is_char(')')):
re0.begin_token = t
re0.end_token = re0.end_token.next0_
return re0
txt = t.term
nams = None
if (len(txt) == 5 and ((txt[1] == 'Х' or txt[1] == 'X')) and ((txt[3] == 'Х' or txt[3] == 'X'))):
nams = list()
for i in range(3):
ch = txt[i * 2]
if (ch == 'Г'):
nams.append("ГЛУБИНА")
elif (ch == 'В' or ch == 'H' or ch == 'Н'):
nams.append("ВЫСОТА")
elif (ch == 'Ш' or ch == 'B' or ch == 'W'):
nams.append("ШИРИНА")
elif (ch == 'Д' or ch == 'L'):
nams.append("ДЛИНА")
elif (ch == 'D'):
nams.append("ДИАМЕТР")
else:
return None
return MetaToken._new836(t, t, nams)
t0 = t
t1 = t
while t is not None:
if (not (isinstance(t, TextToken)) or ((t.whitespaces_before_count > 1 and t != t0))):
break
term = t.term
if (term.endswith("X") or term.endswith("Х")):
term = term[0:0+len(term) - 1]
nam = None
if (((t.is_value("ДЛИНА", None) or t.is_value("ДЛИННА", None) or term == "Д") or term == "ДЛ" or term == "ДЛИН") or term == "L"):
nam = "ДЛИНА"
elif (((t.is_value("ШИРИНА", None) or t.is_value("ШИРОТА", None) or term == "Ш") or term == "ШИР" or term == "ШИРИН") or term == "W" or term == "B"):
nam = "ШИРИНА"
elif ((t.is_value("ГЛУБИНА", None) or term == "Г" or term == "ГЛ") or term == "ГЛУБ"):
nam = "ГЛУБИНА"
elif ((t.is_value("ВЫСОТА", None) or term == "В" or term == "ВЫС") or term == "H" or term == "Н"):
nam = "ВЫСОТА"
elif (t.is_value("ДИАМЕТР", None) or term == "D" or term == "ДИАМ"):
nam = "ДИАМЕТР"
else:
break
if (nams is None):
nams = list()
nams.append(nam)
t1 = t
if (t.next0_ is not None and t.next0_.is_char('.')):
t = t.next0_
t1 = t
if (t.next0_ is None):
break
if (MeasureHelper.is_mult_char(t.next0_) or t.next0_.is_comma or t.next0_.is_char_of("\\/")):
t = t.next0_
t = t.next0_
if (nams is None or (len(nams) < 2)):
return None
return MetaToken._new836(t0, t1, nams)