def __try_parse_ru(first: 'Token',
typ: 'NounPhraseParseAttr',
max_char_pos: int,
def_noun: 'NounPhraseItem' = None) -> 'NounPhraseToken':
if (first is None):
return None
items = None
adverbs = None
prep = None
kak = False
t0 = first
if ((((typ) & (NounPhraseParseAttr.PARSEPREPOSITION))) !=
(NounPhraseParseAttr.NO) and t0.is_value("КАК", None)):
t0 = t0.next0_
prep = PrepositionHelper.try_parse(t0)
if (prep is not None):
t0 = prep.end_token.next0_
kak = True
internal_noun_prase = None
conj_before = False
t = t0
first_pass3041 = True
while True:
if first_pass3041: first_pass3041 = False
else: t = t.next0_
if (not (t is not None)): break
if (max_char_pos > 0 and t.begin_char > max_char_pos):
break
if ((t.morph.class0_.is_conjunction
and not t.morph.class0_.is_adjective
and not t.morph.class0_.is_pronoun)
and not t.morph.class0_.is_noun):
if (conj_before):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
if (items is not None and ((t.is_and or t.is_or))):
conj_before = True
if ((t.next0_ is not None and t.next0_.is_char_of("\\/")
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or):
t = t.next0_.next0_
if (((t.next0_ is not None and t.next0_.is_char('(')
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or
and t.next0_.next0_.next0_ is not None)
and t.next0_.next0_.next0_.is_char(')')):
t = t.next0_.next0_.next0_
continue
break
elif (t.is_comma):
if (conj_before or items is None):
break
if ((((typ) & (NounPhraseParseAttr.CANNOTHASCOMMAAND))) !=
(NounPhraseParseAttr.NO)):
break
mc = t.previous.get_morph_class_in_dictionary()
if (mc.is_proper_surname or mc.is_proper_secname):
break
conj_before = True
if (kak and t.next0_ is not None
and t.next0_.is_value("ТАК", None)):
t = t.next0_
if (t.next0_ is not None and t.next0_.is_and):
t = t.next0_
pr = PrepositionHelper.try_parse(t.next0_)
if (pr is not None):
t = pr.end_token
if (items[len(items) - 1].can_be_noun
and items[len(items) -
1].end_token.morph.class0_.is_pronoun):
break
continue
elif (t.is_char('(')):
if (items is None):
return None
brr = BracketHelper.try_parse(t, BracketParseAttr.NO, 100)
if (brr is None):
break
if (brr.length_char > 100):
break
t = brr.end_token
continue
if (isinstance(t, ReferentToken)):
if ((((typ) & (NounPhraseParseAttr.REFERENTCANBENOUN))) == (
NounPhraseParseAttr.NO)):
break
elif (t.chars.is_latin_letter):
break
it = NounPhraseItem.try_parse(t, items, typ)
if (it is None or ((not it.can_be_adj and not it.can_be_noun))):
if (((it is not None and items is not None
and t.chars.is_capital_upper) and
(t.whitespaces_before_count < 3) and t.length_char > 3)
and not t.get_morph_class_in_dictionary().is_noun and
not t.get_morph_class_in_dictionary().is_adjective):
it.can_be_noun = True
items.append(it)
break
if ((((typ) & (NounPhraseParseAttr.PARSEADVERBS))) !=
(NounPhraseParseAttr.NO) and (isinstance(t, TextToken))
and t.morph.class0_.is_adverb):
if (adverbs is None):
adverbs = list()
adverbs.append(Utils.asObjectOrNull(t, TextToken))
continue
break
it.conj_before = conj_before
conj_before = False
if (not it.can_be_adj and not it.can_be_noun):
break
if (t.is_newline_before and t != first):
if ((((typ) & (NounPhraseParseAttr.MULTILINES))) !=
(NounPhraseParseAttr.NO)):
pass
elif (items is not None
and t.chars != items[len(items) - 1].chars):
if (t.chars.is_all_lower
and items[len(items) - 1].chars.is_capital_upper):
pass
else:
break
if (items is None):
items = list()
else:
it0 = items[len(items) - 1]
if (it0.can_be_noun and it0.is_personal_pronoun):
if (it.is_pronoun):
break
if ((it0.begin_token.previous is not None
and it0.begin_token.previous.
get_morph_class_in_dictionary().is_verb
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_adjective)
and not it0.begin_token.previous.
get_morph_class_in_dictionary().is_preposition):
if (t.morph.case_.is_nominative
or t.morph.case_.is_accusative):
pass
else:
break
if (it.can_be_noun and it.is_verb):
if (it0.previous is None):
pass
elif ((isinstance(it0.previous, TextToken))
and not it0.previous.chars.is_letter):
pass
else:
break
items.append(it)
t = it.end_token
if (t.is_newline_after and not t.chars.is_all_lower):
mc = t.get_morph_class_in_dictionary()
if (mc.is_proper_surname):
break
if (t.morph.class0_.is_proper_surname and mc.is_undefined):
break
if (items is None):
return None
tt1 = None
if (len(items) == 1 and items[0].can_be_adj):
and0_ = False
tt1 = items[0].end_token.next0_
first_pass3042 = True
while True:
if first_pass3042: first_pass3042 = False
else: tt1 = tt1.next0_
if (not (tt1 is not None)): break
if (tt1.is_and or tt1.is_or):
and0_ = True
break
if (tt1.is_comma or tt1.is_value("НО", None)
or tt1.is_value("ТАК", None)):
continue
break
if (and0_):
if (items[0].can_be_noun and items[0].is_personal_pronoun):
and0_ = False
if (and0_):
tt2 = tt1.next0_
if (tt2 is not None and tt2.morph.class0_.is_preposition):
tt2 = tt2.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt2, typ, max_char_pos, None)
if (npt1 is not None and len(npt1.adjectives) > 0):
ok1 = False
for av in items[0].adj_morph:
for v in npt1.noun.noun_morph:
if (v.check_accord(av, False, False)):
items[0].morph.add_item(av)
ok1 = True
if (ok1):
npt1.begin_token = items[0].begin_token
npt1.end_token = tt1.previous
npt1.adjectives.clear()
npt1.adjectives.append(items[0])
return npt1
if (def_noun is not None):
items.append(def_noun)
last1 = items[len(items) - 1]
check = True
for it in items:
if (not it.can_be_adj):
check = False
break
elif (it.can_be_noun and it.is_personal_pronoun):
check = False
break
tt1 = last1.end_token.next0_
if ((tt1 is not None and check and
((tt1.morph.class0_.is_preposition
or tt1.morph.case_.is_instrumental)))
and (tt1.whitespaces_before_count < 2)):
inp = NounPhraseHelper.try_parse(
tt1,
Utils.valToEnum((typ) | (NounPhraseParseAttr.PARSEPREPOSITION),
NounPhraseParseAttr), max_char_pos, None)
if (inp is not None):
tt1 = inp.end_token.next0_
npt1 = _NounPraseHelperInt.__try_parse_ru(
tt1, typ, max_char_pos, None)
if (npt1 is not None):
ok = True
ii = 0
first_pass3043 = True
while True:
if first_pass3043: first_pass3043 = False
else: ii += 1
if (not (ii < len(items))): break
it = items[ii]
if (NounPhraseItem.try_accord_adj_and_noun(
it,
Utils.asObjectOrNull(npt1.noun,
NounPhraseItem))):
continue
if (ii > 0):
inp2 = NounPhraseHelper.try_parse(
it.begin_token, typ, max_char_pos, None)
if (inp2 is not None
and inp2.end_token == inp.end_token):
del items[ii:ii + len(items) - ii]
inp = inp2
break
ok = False
break
if (ok):
if (npt1.morph.case_.is_genitive
and not inp.morph.case_.is_instrumental):
ok = False
if (ok):
i = 0
while i < len(items):
npt1.adjectives.insert(i, items[i])
i += 1
npt1.internal_noun = inp
mmm = MorphCollection(npt1.morph)
for it in items:
mmm.remove_items(it.adj_morph[0], False)
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
if (tt1 is not None and tt1.morph.class0_.is_noun
and not tt1.morph.case_.is_genitive):
it = NounPhraseItem.try_parse(tt1, items, typ)
if (it is not None and it.can_be_noun):
internal_noun_prase = inp
inp.begin_token = items[0].end_token.next0_
items.append(it)
i = 0
first_pass3044 = True
while True:
if first_pass3044: first_pass3044 = False
else: i += 1
if (not (i < len(items))): break
if (items[i].can_be_adj
and items[i].begin_token.morph.class0_.is_verb):
it = items[i].begin_token
if (not it.get_morph_class_in_dictionary().is_verb):
continue
if (it.is_value("УПОЛНОМОЧЕННЫЙ", None)):
continue
if ((((typ) & (NounPhraseParseAttr.PARSEVERBS))) == (
NounPhraseParseAttr.NO)):
continue
inp = _NounPraseHelperInt.__try_parse_ru(
items[i].end_token.next0_, NounPhraseParseAttr.NO,
max_char_pos, None)
if (inp is None):
continue
if (inp.anafor is not None and i == (len(items) - 1)
and NounPhraseItem.try_accord_adj_and_noun(
items[i],
Utils.asObjectOrNull(inp.noun, NounPhraseItem))):
inp.begin_token = first
ii = 0
while ii < len(items):
inp.adjectives.insert(ii, items[ii])
ii += 1
return inp
if (inp.end_token.whitespaces_after_count > 3):
continue
npt1 = _NounPraseHelperInt.__try_parse_ru(
inp.end_token.next0_, NounPhraseParseAttr.NO, max_char_pos,
None)
if (npt1 is None):
continue
ok = True
j = 0
while j <= i:
if (not NounPhraseItem.try_accord_adj_and_noun(
items[j],
Utils.asObjectOrNull(npt1.noun, NounPhraseItem))):
ok = False
break
j += 1
if (not ok):
continue
verb = VerbPhraseHelper.try_parse(it, True, False, False)
if (verb is None):
continue
vlinks = SemanticHelper.try_create_links(verb, inp, None)
nlinks = SemanticHelper.try_create_links(inp, npt1, None)
if (len(vlinks) == 0 and len(nlinks) > 0):
continue
j = 0
while j <= i:
npt1.adjectives.insert(j, items[j])
j += 1
items[i].end_token = inp.end_token
mmm = MorphCollection(npt1.morph)
bil = list()
j = 0
while j <= i:
bil.clear()
for m in items[j].adj_morph:
bil.append(m)
mmm.remove_items_list_cla(bil, None)
j += 1
if (mmm.gender != MorphGender.UNDEFINED
or mmm.number != MorphNumber.UNDEFINED
or not mmm.case_.is_undefined):
npt1.morph = mmm
if (adverbs is not None):
if (npt1.adverbs is None):
npt1.adverbs = adverbs
else:
npt1.adverbs[0:0] = adverbs
npt1.begin_token = first
return npt1
ok2 = False
if ((len(items) == 1 and
(((typ) & (NounPhraseParseAttr.ADJECTIVECANBELAST))) !=
(NounPhraseParseAttr.NO) and
(items[0].whitespaces_after_count < 3))
and not items[0].is_adverb):
if (not items[0].can_be_adj):
ok2 = True
elif (items[0].is_personal_pronoun and items[0].can_be_noun):
ok2 = True
if (ok2):
it = NounPhraseItem.try_parse(items[0].end_token.next0_, None, typ)
if (it is not None and it.can_be_adj
and it.begin_token.chars.is_all_lower):
ok2 = True
if (it.is_adverb or it.is_verb):
ok2 = False
if (it.is_pronoun and items[0].is_pronoun):
ok2 = False
if (it.can_be_adj_for_personal_pronoun
and items[0].is_personal_pronoun):
ok2 = True
if (ok2 and NounPhraseItem.try_accord_adj_and_noun(
it, items[0])):
npt1 = _NounPraseHelperInt.__try_parse_ru(
it.begin_token, typ, max_char_pos, None)
if (npt1 is not None and ((npt1.end_char > it.end_char
or len(npt1.adjectives) > 0))):
pass
else:
items.insert(0, it)
noun = None
adj_after = None
for i in range(len(items) - 1, -1, -1):
if (items[i].can_be_noun):
if (items[i].conj_before):
continue
if (i > 0 and not items[i - 1].can_be_adj):
continue
if (i > 0 and items[i - 1].can_be_noun):
if (items[i - 1].is_doubt_adjective):
continue
if (items[i - 1].is_pronoun and items[i].is_pronoun):
if (items[i].is_pronoun and
items[i - 1].can_be_adj_for_personal_pronoun):
pass
else:
continue
noun = items[i]
del items[i:i + len(items) - i]
if (adj_after is not None):
items.append(adj_after)
elif (len(items) > 0 and items[0].can_be_noun
and not items[0].can_be_adj):
noun = items[0]
items.clear()
break
if (noun is None):
return None
res = NounPhraseToken._new466(first, noun.end_token, prep)
if (adverbs is not None):
for a in adverbs:
if (a.begin_char < noun.begin_char):
if (len(items) == 0 and prep is None):
return None
if (res.adverbs is None):
res.adverbs = list()
res.adverbs.append(a)
res.noun = (noun)
res.multi_nouns = noun.multi_nouns
if (kak):
res.multi_nouns = True
res.internal_noun = internal_noun_prase
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (res.morph.case_.is_nominative and first.previous is not None
and first.previous.morph.class0_.is_preposition):
res.morph.case_ = (res.morph.case_) ^ MorphCase.NOMINATIVE
if ((((typ) &
(NounPhraseParseAttr.PARSEPRONOUNS))) == (NounPhraseParseAttr.NO)
and ((res.morph.class0_.is_pronoun
or res.morph.class0_.is_personal_pronoun))):
return None
stat = None
if (len(items) > 1):
stat = dict()
need_update_morph = False
if (len(items) > 0):
ok_list = list()
is_num_not = False
for vv in noun.noun_morph:
i = 0
v = vv
i = 0
while i < len(items):
ok = False
for av in items[i].adj_morph:
if (v.check_accord(av, False, False)):
ok = True
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
break
if (not ok):
if (items[i].can_be_numeric_adj
and items[i].try_accord_var(v, False)):
ok = True
v1 = NounPhraseItemTextVar()
v1.copy_from_item(v)
v1.number = MorphNumber.PLURAL
is_num_not = True
v1.case_ = MorphCase()
for a in items[i].adj_morph:
v1.case_ = (v1.case_) | a.case_
v = v1
else:
break
i += 1
if (i >= len(items)):
ok_list.append(v)
if (len(ok_list) > 0 and
(((len(ok_list) < res.morph.items_count) or is_num_not))):
res.morph = MorphCollection()
for v in ok_list:
res.morph.add_item(v)
if (not is_num_not):
noun.morph = res.morph
i = 0
first_pass3045 = True
while True:
if first_pass3045: first_pass3045 = False
else: i += 1
if (not (i < len(items))): break
for av in items[i].adj_morph:
for v in noun.noun_morph:
if (v.check_accord(av, False, False)):
if (not ((av.case_) & v.case_).is_undefined
and av.case_ != v.case_):
v.case_ = av.case_ = (av.case_) & v.case_
need_update_morph = True
items[i].morph.add_item(av)
if (stat is not None and av.normal_value is not None
and len(av.normal_value) > 1):
last = av.normal_value[len(av.normal_value) - 1]
if (not last in stat):
stat[last] = 1
else:
stat[last] += 1
if (items[i].is_pronoun or items[i].is_personal_pronoun):
res.anafor = items[i].begin_token
if ((((typ) & (NounPhraseParseAttr.PARSEPRONOUNS))) == (
NounPhraseParseAttr.NO)):
continue
tt = Utils.asObjectOrNull(items[i].begin_token, TextToken)
if (tt is not None and not tt.term.startswith("ВЫСШ")):
err = False
for wf in tt.morph.items:
if (wf.class0_.is_adjective):
if (wf.contains_attr("прев.", None)):
if ((((typ) &
(NounPhraseParseAttr.IGNOREADJBEST))) !=
(NounPhraseParseAttr.NO)):
err = True
if (wf.contains_attr("к.ф.", None)
and tt.morph.class0_.is_personal_pronoun):
return None
if (err):
continue
if (res.morph.case_.is_nominative):
v = MiscHelper.get_text_value_of_meta_token(
items[i], GetTextAttr.KEEPQUOTES)
if (not Utils.isNullOrEmpty(v)):
if (items[i].get_normal_case_text(
None, MorphNumber.UNDEFINED, MorphGender.UNDEFINED,
False) != v):
wf = NounPhraseItemTextVar(items[i].morph, None)
wf.normal_value = v
wf.class0_ = MorphClass.ADJECTIVE
wf.case_ = res.morph.case_
if (res.morph.case_.is_prepositional
or res.morph.gender == MorphGender.NEUTER
or res.morph.gender == MorphGender.FEMINIE):
items[i].morph.add_item(wf)
else:
items[i].morph.insert_item(0, wf)
res.adjectives.append(items[i])
if (items[i].end_char > res.end_char):
res.end_token = items[i].end_token
i = 0
first_pass3046 = True
while True:
if first_pass3046: first_pass3046 = False
else: i += 1
if (not (i < (len(res.adjectives) - 1))): break
if (res.adjectives[i].whitespaces_after_count > 5):
if (res.adjectives[i].chars != res.adjectives[i + 1].chars):
if (not res.adjectives[i + 1].chars.is_all_lower):
return None
if (res.adjectives[i].chars.is_all_upper
and res.adjectives[i + 1].chars.is_capital_upper):
return None
if (res.adjectives[i].chars.is_capital_upper
and res.adjectives[i + 1].chars.is_all_upper):
return None
if (res.adjectives[i].whitespaces_after_count > 10):
if (res.adjectives[i].newlines_after_count == 1):
if (res.adjectives[i].chars.is_capital_upper and i == 0
and res.adjectives[i + 1].chars.is_all_lower):
continue
if (res.adjectives[i].chars == res.adjectives[
i + 1].chars):
continue
return None
if (need_update_morph):
noun.morph = MorphCollection()
for v in noun.noun_morph:
noun.morph.add_item(v)
res.morph = noun.morph
if (len(res.adjectives) > 0):
if (noun.begin_token.previous is not None):
if (noun.begin_token.previous.is_comma_and):
if (res.adjectives[0].begin_char > noun.begin_char):
pass
else:
return None
zap = 0
and0_ = 0
cou = 0
last_and = False
i = 0
while i < (len(res.adjectives) - 1):
te = res.adjectives[i].end_token.next0_
if (te is None):
return None
if (te.is_char('(')):
pass
elif (te.is_comma):
zap += 1
last_and = False
elif (te.is_and or te.is_or):
and0_ += 1
last_and = True
if (not res.adjectives[i].begin_token.morph.class0_.is_pronoun
):
cou += 1
i += 1
if ((zap + and0_) > 0):
if (and0_ > 1):
return None
elif (and0_ == 1 and not last_and):
return None
if ((zap + and0_) != cou):
if (and0_ == 1):
pass
else:
return None
last = Utils.asObjectOrNull(
res.adjectives[len(res.adjectives) - 1], NounPhraseItem)
if (last.is_pronoun and not last_and):
return None
if (stat is not None):
for adj in items:
if (adj.morph.items_count > 1):
w1 = Utils.asObjectOrNull(adj.morph.get_indexer_item(0),
NounPhraseItemTextVar)
w2 = Utils.asObjectOrNull(adj.morph.get_indexer_item(1),
NounPhraseItemTextVar)
if ((len(w1.normal_value) < 2)
or (len(w2.normal_value) < 2)):
break
l1 = w1.normal_value[len(w1.normal_value) - 1]
l2 = w2.normal_value[len(w2.normal_value) - 1]
i1 = 0
i2 = 0
wrapi1468 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l1, wrapi1468)
i1 = wrapi1468.value
wrapi2467 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l2, wrapi2467)
i2 = wrapi2467.value
if (i1 < i2):
adj.morph.remove_item(1)
adj.morph.insert_item(0, w2)
if (res.begin_token.get_morph_class_in_dictionary().is_verb
and len(items) > 0):
if (not res.begin_token.chars.is_all_lower
or res.begin_token.previous is None):
pass
elif (res.begin_token.previous.morph.class0_.is_preposition):
pass
else:
comma = False
tt = res.begin_token.previous
first_pass3047 = True
while True:
if first_pass3047: first_pass3047 = False
else: tt = tt.previous
if (not (tt is not None and tt.end_char <= res.end_char)):
break
if (tt.morph.class0_.is_adverb):
continue
if (tt.is_char_of(".;")):
break
if (tt.is_comma):
comma = True
continue
if (tt.is_value("НЕ", None)):
continue
if (((tt.morph.class0_.is_noun
or tt.morph.class0_.is_proper)) and comma):
for it in res.begin_token.morph.items:
if (it.class0_.is_verb
and (isinstance(it, MorphWordForm))):
if (tt.morph.check_accord(it, False, False)):
if (res.morph.case_.is_instrumental):
return None
break
if (res.begin_token == res.end_token):
mc = res.begin_token.get_morph_class_in_dictionary()
if (mc.is_adverb):
if (res.begin_token.previous is not None and
res.begin_token.previous.morph.class0_.is_preposition):
pass
elif (mc.is_noun and not mc.is_preposition
and not mc.is_conjunction):
pass
elif (res.begin_token.is_value("ВЕСЬ", None)):
pass
else:
return None
if (def_noun is not None and def_noun.end_token == res.end_token
and len(res.adjectives) > 0):
res.end_token = res.adjectives[len(res.adjectives) - 1].end_token
return res
def tryParse(t: 'Token',
add_units: 'TerminCollection',
can_be_set: bool = True,
can_units_absent: bool = False) -> 'MeasureToken':
""" Выделение вместе с наименованием
Args:
t(Token):
"""
if (not ((isinstance(t, TextToken)))):
return None
if (t.is_table_control_char):
return None
t0 = t
whd = None
minmax = 0
wrapminmax1516 = RefOutArgWrapper(minmax)
tt = NumbersWithUnitToken._isMinOrMax(t0, wrapminmax1516)
minmax = wrapminmax1516.value
if (tt is not None):
t = tt.next0_
npt = NounPhraseHelper.tryParse(
t,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt is None):
whd = NumbersWithUnitToken._tryParseWHL(t)
if (whd is not None):
npt = NounPhraseToken(t0, whd.end_token)
elif (t0.isValue("КПД", None)):
npt = NounPhraseToken(t0, t0)
elif ((isinstance(t0, TextToken)) and t0.length_char > 3
and t0.getMorphClassInDictionary().is_undefined):
npt = NounPhraseToken(t0, t0)
else:
return None
elif (NumberHelper.tryParseRealNumber(t, True) is not None):
return None
else:
dtok = DateItemToken.tryAttach(t, None)
if (dtok is not None):
return None
t1 = npt.end_token
t = npt.end_token
name_ = MetaToken._new561(npt.begin_token, npt.end_token, npt.morph)
units = None
units2 = None
internals_ = list()
not0_ = False
tt = t1.next0_
first_pass3037 = True
while True:
if first_pass3037: first_pass3037 = False
else: tt = tt.next0_
if (not (tt is not None)): break
if (tt.is_newline_before):
break
if (tt.is_table_control_char):
break
wrapminmax1510 = RefOutArgWrapper(minmax)
tt2 = NumbersWithUnitToken._isMinOrMax(tt, wrapminmax1510)
minmax = wrapminmax1510.value
if (tt2 is not None):
tt = tt2
t = tt
t1 = t
continue
if ((tt.isValue("БЫТЬ", None) or tt.isValue("ДОЛЖЕН", None)
or tt.isValue("ДОЛЖНЫЙ", None)) or tt.isValue("МОЖЕТ", None)
or
((tt.isValue("СОСТАВЛЯТЬ", None)
and not tt.getMorphClassInDictionary().is_adjective))):
t = tt
t1 = t
if (tt.previous.isValue("НЕ", None)):
not0_ = True
continue
www = NumbersWithUnitToken._tryParseWHL(tt)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
continue
if (len(internals_) > 0 and tt.is_comma_and):
continue
if (tt.isValue("ПРИ", None) or len(internals_) > 0):
mt1 = MeasureToken.tryParse(tt.next0_, add_units, False, False)
if (mt1 is not None and mt1.reliable):
internals_.append(mt1)
tt = mt1.end_token
t = tt
t1 = t
continue
if ((isinstance(tt, NumberToken))
and (tt).typ == NumberSpellingType.WORDS):
npt3 = NounPhraseHelper.tryParse(
tt, NounPhraseParseAttr.PARSENUMERICASADJECTIVE, 0)
if (npt3 is not None):
tt = npt3.end_token
t1 = tt
if (len(internals_) == 0):
name_.end_token = t1
continue
mt0 = NumbersWithUnitToken.tryParse(tt, add_units, False, False)
if (mt0 is not None):
break
if (((tt.is_comma or tt.isChar('('))) and tt.next0_ is not None):
www = NumbersWithUnitToken._tryParseWHL(tt.next0_)
if (www is not None):
whd = www
tt = www.end_token
t = tt
t1 = t
if (tt.next0_ is not None and tt.next0_.is_comma):
tt = tt.next0_
t1 = tt
if (tt.next0_ is not None and tt.next0_.isChar(')')):
tt = tt.next0_
t1 = tt
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
t = uu[len(uu) - 1].end_token
t1 = t
units = uu
if (tt.isChar('(') and t1.next0_ is not None
and t1.next0_.isChar(')')):
tt = t1.next0_
t = tt
t1 = t
continue
elif (t1.next0_ is not None and t1.next0_.isChar('(')):
uu = UnitToken.tryParseList(t1.next0_.next0_,
add_units, False)
if (uu is not None and uu[len(uu) - 1].end_token.next0_
is not None and
uu[len(uu) - 1].end_token.next0_.isChar(')')):
units2 = uu
tt = uu[len(uu) - 1].end_token.next0_
t = tt
t1 = t
continue
if (uu is not None and len(uu) > 0 and not uu[0].is_doubt):
break
if (BracketHelper.canBeStartOfSequence(tt, False, False)):
br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100)
if (br is not None):
tt = br.end_token
t = tt
t1 = t
continue
if (tt.isValue("НЕ", None) and tt.next0_ is not None):
mc = tt.next0_.getMorphClassInDictionary()
if (mc.is_adverb or mc.is_misc):
break
continue
if (tt.isValue("ЯМЗ", None)):
pass
npt2 = NounPhraseHelper.tryParse(
tt,
Utils.valToEnum((NounPhraseParseAttr.PARSEPREPOSITION) |
(NounPhraseParseAttr.IGNOREBRACKETS),
NounPhraseParseAttr), 0)
if (npt2 is None):
if (tt.morph.class0_.is_preposition
or tt.morph.class0_.is_conjunction):
to = NumbersWithUnitToken.M_TERMINS.tryParse(
tt, TerminParseAttr.NO)
if (to is not None):
if ((isinstance(to.end_token.next0_, TextToken))
and to.end_token.next0_.is_letters):
pass
else:
break
t1 = tt
continue
mc = tt.getMorphClassInDictionary()
if (((isinstance(tt, TextToken)) and tt.chars.is_letter
and tt.length_char > 1)
and (((tt.chars.is_all_upper or mc.is_adverb
or mc.is_undefined) or mc.is_adjective))):
uu = UnitToken.tryParseList(tt, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
t = tt
t1 = t
if (len(internals_) == 0):
name_.end_token = tt
continue
if (tt.is_comma):
continue
if (tt.isChar('.')):
if (not MiscHelper.canBeStartOfSentence(tt.next0_)):
continue
uu = UnitToken.tryParseList(tt.next0_, add_units, False)
if (uu is not None):
if (uu[0].length_char > 2 or len(uu) > 1):
units = uu
t = uu[len(uu) - 1].end_token
t1 = t
break
break
tt = npt2.end_token
t = tt
t1 = t
if (len(internals_) > 0):
pass
elif (t.isValue("ПРЕДЕЛ", None) or t.isValue("ГРАНИЦА", None)
or t.isValue("ДИАПАЗОН", None)):
pass
elif (t.chars.is_letter):
name_.end_token = t1
t1 = t1.next0_
first_pass3038 = True
while True:
if first_pass3038: first_pass3038 = False
else: t1 = t1.next0_
if (not (t1 is not None)): break
if (t1.is_table_control_char):
pass
elif (t1.isCharOf(":,_")):
www = NumbersWithUnitToken._tryParseWHL(t1.next0_)
if (www is not None):
whd = www
t = www.end_token
t1 = t
continue
elif (t1.is_hiphen and t1.is_whitespace_after
and t1.is_whitespace_before):
pass
else:
break
if (t1 is None):
return None
mts = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, not0_)
if (mts is None):
return None
mt = mts[0]
if (name_.begin_token.morph.class0_.is_preposition):
name_.begin_token = name_.begin_token.next0_
if (len(mts) > 1 and len(internals_) == 0):
if (len(mt.units) == 0):
if (units is not None):
for m in mts:
m.units = units
res1 = MeasureToken._new1511(t0, mts[len(mts) - 1].end_token,
name_.morph, True)
res1.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
k = 0
while k < len(mts):
ttt = MeasureToken._new1506(mts[k].begin_token,
mts[k].end_token, mts[k])
if (whd is not None):
nams = Utils.asObjectOrNull(whd.tag, list)
if (k < len(nams)):
ttt.name = nams[k]
res1.internals.append(ttt)
k += 1
tt1 = res1.end_token.next0_
if (tt1 is not None and tt1.isChar('±')):
nn = NumbersWithUnitToken._tryParse(tt1, add_units, True,
False)
if (nn is not None and nn.plus_minus_percent):
res1.end_token = nn.end_token
res1.nums = nn
return res1
if (not mt.is_whitespace_before):
if (mt.begin_token.previous is None):
return None
if (mt.begin_token.previous.isCharOf(":),")
or mt.begin_token.previous.is_table_control_char):
pass
else:
return None
if (len(mt.units) == 0 and units is not None):
mt.units = units
if (mt.div_num is not None and len(units) > 1
and len(mt.div_num.units) == 0):
i = 1
while i < len(units):
if (units[i].pow0_ == -1):
j = i
while j < len(units):
mt.div_num.units.append(units[j])
units[j].pow0_ = (-units[j].pow0_)
j += 1
del mt.units[i:i + len(units) - i]
break
i += 1
if ((minmax < 0) and mt.single_val is not None):
mt.from_val = mt.single_val
mt.from_include = True
mt.single_val = (None)
if (minmax > 0 and mt.single_val is not None):
mt.to_val = mt.single_val
mt.to_include = True
mt.single_val = (None)
if (len(mt.units) == 0):
units = UnitToken.tryParseList(mt.end_token.next0_, add_units,
True)
if (units is None):
if (can_units_absent):
pass
else:
return None
else:
mt.units = units
res = MeasureToken._new1513(t0, mt.end_token, name_.morph, internals_)
if (((not t0.is_whitespace_before and t0.previous is not None
and t0 == name_.begin_token) and t0.previous.is_hiphen
and not t0.previous.is_whitespace_before)
and (isinstance(t0.previous.previous, TextToken))):
name_.begin_token = res.begin_token = name_.begin_token.previous.previous
res.name = MiscHelper.getTextValueOfMetaToken(
name_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE)
res.nums = mt
for u in res.nums.units:
if (u.keyword is not None):
if (u.keyword.begin_char >= res.begin_char):
res.reliable = True
res.__parseInternals(add_units)
if (len(res.internals) > 0 or not can_be_set):
return res
t1 = res.end_token.next0_
if (t1 is not None and t1.is_comma_and):
t1 = t1.next0_
mts1 = NumbersWithUnitToken.tryParseMulti(t1, add_units, False, False)
if ((mts1 is not None and len(mts1) == 1 and
(t1.whitespaces_before_count < 3)) and len(mts1[0].units) > 0
and not UnitToken.canBeEquals(mts[0].units, mts1[0].units)):
res.is_set = True
res.nums = (None)
res.internals.append(
MeasureToken._new1506(mt.begin_token, mt.end_token, mt))
res.internals.append(
MeasureToken._new1506(mts1[0].begin_token, mts1[0].end_token,
mts1[0]))
res.end_token = mts1[0].end_token
return res
def __try_parse_en(first: 'Token', typ: 'NounPhraseParseAttr',
max_char_pos: int) -> 'NounPhraseToken':
if (first is None):
return None
items = None
has_article = False
has_prop = False
has_misc = False
if (first.previous is not None
and first.previous.morph.class0_.is_preposition
and (first.whitespaces_before_count < 3)):
has_prop = True
t = first
first_pass3048 = True
while True:
if first_pass3048: first_pass3048 = 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 (not t.chars.is_latin_letter):
break
if (t != first and t.whitespaces_before_count > 2):
if ((((typ) & (NounPhraseParseAttr.MULTILINES))) !=
(NounPhraseParseAttr.NO)):
pass
elif (MiscHelper.is_eng_article(t.previous)):
pass
else:
break
tt = Utils.asObjectOrNull(t, TextToken)
if (t == first and tt is not None):
if (MiscHelper.is_eng_article(tt)):
has_article = True
continue
if (isinstance(t, ReferentToken)):
if ((((typ) & (NounPhraseParseAttr.REFERENTCANBENOUN))) == (
NounPhraseParseAttr.NO)):
break
elif (tt is None):
break
if ((t.is_value("SO", None) and t.next0_ is not None
and t.next0_.is_hiphen) and t.next0_.next0_ is not None):
if (t.next0_.next0_.is_value("CALL", None)):
t = t.next0_.next0_
continue
mc = t.get_morph_class_in_dictionary()
if (mc.is_conjunction or mc.is_preposition):
break
if (mc.is_pronoun or mc.is_personal_pronoun):
if ((((typ) & (NounPhraseParseAttr.PARSEPRONOUNS))) == (
NounPhraseParseAttr.NO)):
break
elif (mc.is_misc):
if (t.is_value("THIS", None) or t.is_value("THAT", None)):
has_misc = True
if ((((typ) & (NounPhraseParseAttr.PARSEPRONOUNS))) == (
NounPhraseParseAttr.NO)):
break
is_adj = False
if (((has_article or has_prop or has_misc)) and items is None):
pass
elif (isinstance(t, ReferentToken)):
pass
else:
if (not mc.is_noun and not mc.is_adjective):
if (mc.is_undefined and has_article):
pass
elif (items is None and mc.is_undefined
and t.chars.is_capital_upper):
pass
elif (mc.is_pronoun):
pass
elif (tt.term.endswith("EAN")):
is_adj = True
elif (MiscHelper.is_eng_adj_suffix(tt.next0_)):
pass
else:
break
if (mc.is_verb):
if (t.next0_ is not None and t.next0_.morph.class0_.is_verb
and (t.whitespaces_after_count < 2)):
pass
elif (t.chars.is_capital_upper
and not MiscHelper.can_be_start_of_sentence(t)):
pass
elif ((t.chars.is_capital_upper and mc.is_noun and
(isinstance(t.next0_, TextToken)))
and t.next0_.chars.is_capital_upper):
pass
elif (isinstance(t, ReferentToken)):
pass
else:
break
if (items is None):
items = list()
it = NounPhraseItem(t, t)
if (mc.is_noun):
it.can_be_noun = True
if (mc.is_adjective or mc.is_pronoun or is_adj):
it.can_be_adj = True
items.append(it)
t = it.end_token
if (len(items) == 1):
if (MiscHelper.is_eng_adj_suffix(t.next0_)):
mc.is_noun = False
mc.is_adjective = True
t = t.next0_.next0_
if (items is None):
return None
noun = items[len(items) - 1]
res = NounPhraseToken(first, noun.end_token)
res.noun = (noun)
res.morph = MorphCollection()
for v in noun.end_token.morph.items:
if (v.class0_.is_verb):
continue
if (v.class0_.is_proper and noun.begin_token.chars.is_all_lower):
continue
if (isinstance(v, MorphWordForm)):
wf = MorphWordForm()
wf.copy_from_word_form(Utils.asObjectOrNull(v, MorphWordForm))
if (has_article and v.number != MorphNumber.SINGULAR):
wf.number = MorphNumber.SINGULAR
res.morph.add_item(wf)
else:
bi = MorphBaseInfo()
bi.copy_from(v)
if (has_article and v.number != MorphNumber.SINGULAR):
bi.number = MorphNumber.SINGULAR
res.morph.add_item(bi)
if (res.morph.items_count == 0 and has_article):
res.morph.add_item(
MorphBaseInfo._new192(MorphClass.NOUN, MorphNumber.SINGULAR))
i = 0
while i < (len(items) - 1):
res.adjectives.append(items[i])
i += 1
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 __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 __tryParseRu(first: 'Token', typ: 'NounPhraseParseAttr',
max_char_pos: int) -> 'NounPhraseToken':
if (first is None):
return None
items = None
adverbs = None
internal_noun_prase = None
conj_before = False
t = first
first_pass2788 = True
while True:
if first_pass2788: first_pass2788 = 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):
conj_before = True
if ((t.next0_ is not None and t.next0_.isCharOf("\\/")
and t.next0_.next0_ is not None)
and t.next0_.next0_.is_or):
t = t.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.getMorphClassInDictionary()
if (mc.is_proper_surname or mc.is_proper_secname):
break
conj_before = True
continue
elif (t.isChar('(')):
if (items is None):
return None
if ((((typ) & (NounPhraseParseAttr.IGNOREBRACKETS))) !=
(NounPhraseParseAttr.IGNOREBRACKETS)):
break
brr = BracketHelper.tryParse(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.tryParse(t, items, typ)
if (it is None or ((not it.can_be_adj and not it.can_be_noun))):
if ((((typ) & (NounPhraseParseAttr.PARSEADVERBS))) !=
(NounPhraseParseAttr.NO) and (isinstance(t, TextToken))
and t.morph.class0_.is_adverb):
if (items is None):
if (t.previous is not None
and t.previous.morph.class0_.is_preposition):
pass
else:
return None
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.getMorphClassInDictionary(
).is_verb and not it0.begin_token.previous.
getMorphClassInDictionary().is_adjective)
and not it0.begin_token.previous.
getMorphClassInDictionary().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):
break
items.append(it)
t = it.end_token
if (t.is_newline_after and not t.chars.is_all_lower):
mc = t.getMorphClassInDictionary()
if (mc.is_proper_surname):
break
if (t.morph.class0_.is_proper_surname and mc.is_undefined):
break
if (items is None):
return None
if (len(items) == 1 and items[0].can_be_adj):
and0_ = False
tt1 = items[0].end_token.next0_
first_pass2789 = True
while True:
if first_pass2789: first_pass2789 = 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.isValue("НО", None)
or tt1.isValue("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.__tryParseRu(tt2, typ, max_char_pos)
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.checkAccord(av, False)):
items[0].morph.addItem(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
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.tryParse(
tt1,
Utils.valToEnum((typ) | (NounPhraseParseAttr.PARSEPREPOSITION),
NounPhraseParseAttr), max_char_pos)
if (inp is not None):
tt1 = inp.end_token.next0_
npt1 = _NounPraseHelperInt.__tryParseRu(tt1, typ, max_char_pos)
if (npt1 is not None):
ok = True
for it in items:
if (not NounPhraseItem.tryAccordAdjAndNoun(
it,
Utils.asObjectOrNull(npt1.noun,
NounPhraseItem))):
ok = False
break
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.removeItems(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
return npt1
if (tt1 is not None and tt1.morph.class0_.is_noun):
it = NounPhraseItem.tryParse(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)
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.tryParse(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.tryAccordAdjAndNoun(it, items[0])):
npt1 = _NounPraseHelperInt.__tryParseRu(
it.begin_token, typ, max_char_pos)
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)
break
if (noun is None):
return None
res = NounPhraseToken(first, noun.end_token)
if (adverbs is not None):
for a in adverbs:
if (a.begin_char < noun.begin_char):
if (res.adverbs is None):
res.adverbs = list()
res.adverbs.append(a)
res.noun = (noun)
res.internal_noun = internal_noun_prase
for v in noun.noun_morph:
noun.morph.addItem(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:
v = vv
i = 0
while i < len(items):
ok = False
for av in items[i].adj_morph:
if (v.checkAccord(av, 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].tryAccordVar(v)):
ok = True
v = (Utils.asObjectOrNull(v.clone(),
NounPhraseItemTextVar))
v.number = MorphNumber.PLURAL
is_num_not = True
v.case_ = MorphCase()
for a in items[i].adj_morph:
v.case_ = (v.case_) | a.case_
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.addItem(v)
if (not is_num_not):
noun.morph = res.morph
i = 0
first_pass2790 = True
while True:
if first_pass2790: first_pass2790 = 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.checkAccord(av, 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.addItem(av)
if (stat 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.containsAttr("прев.", None)):
if ((((typ) &
(NounPhraseParseAttr.IGNOREADJBEST))) !=
(NounPhraseParseAttr.NO)):
err = True
if (wf.containsAttr("к.ф.", None)
and tt.morph.class0_.is_personal_pronoun):
return None
if (err):
continue
if (res.morph.case_.is_nominative):
v = MiscHelper.getTextValueOfMetaToken(items[i],
GetTextAttr.KEEPQUOTES)
if (not Utils.isNullOrEmpty(v)):
if (items[i].getNormalCaseText(
None, False, 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.addItem(wf)
else:
items[i].morph.insertItem(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_pass2791 = True
while True:
if first_pass2791: first_pass2791 = 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.addItem(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.isChar('(')):
pass
elif (te.is_comma):
zap += 1
elif (te.is_and):
and0_ += 1
if (i == (len(res.adjectives) - 2)):
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.getIndexerItem(0),
NounPhraseItemTextVar)
w2 = Utils.asObjectOrNull(adj.morph.getIndexerItem(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
wrapi1534 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l1, wrapi1534)
i1 = wrapi1534.value
wrapi2533 = RefOutArgWrapper(0)
Utils.tryGetValue(stat, l2, wrapi2533)
i2 = wrapi2533.value
if (i1 < i2):
adj.morph.removeItem(1)
adj.morph.insertItem(0, w2)
if (res.begin_token.getMorphClassInDictionary().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_pass2792 = True
while True:
if first_pass2792: first_pass2792 = False
else: tt = tt.previous
if (not (tt is not None)): break
if (tt.morph.class0_.is_adverb):
continue
if (tt.isCharOf(".;")):
break
if (tt.is_comma):
comma = True
continue
if (tt.isValue("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.checkAccord(it, False)):
if (res.morph.case_.is_instrumental):
return None
ews = Explanatory.findDerivates(
(it).normal_case, True,
tt.morph.language)
if (ews is not None):
for ew in ews:
if (ew.transitive > 0):
if (res.morph.case_.is_genitive
):
return None
if (ew.nexts is not None):
wrapcm535 = RefOutArgWrapper(
None)
inoutres536 = Utils.tryGetValue(
ew.nexts, "", wrapcm535)
cm = wrapcm535.value
if (inoutres536):
if (not (
(cm) & res.morph.case_
).is_undefined):
return None
break
if (res.begin_token == res.end_token):
mc = res.begin_token.getMorphClassInDictionary()
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.isValue("ВЕСЬ", None)):
pass
else:
return None
return res
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