def __calcNewlineBetweenCoef(fr1: 'FragToken', fr2: 'FragToken') -> int: if (fr1.newlines_after_count > 1): return 1 tt = fr1.begin_token while tt is not None and tt.end_char <= fr1.end_char: if (BracketHelper.canBeStartOfSequence(tt, False, False)): br = BracketHelper.tryParse(tt, BracketParseAttr.CANBEMANYLINES, 100) if (br is not None and br.end_char >= fr2.begin_char): return -1 tt = tt.next0_ t = fr1.end_token if (t.isCharOf(":;.")): return 1 if ((isinstance(t, TextToken)) and ((t.morph.class0_.is_preposition or t.morph.class0_.is_conjunction))): return -1 t1 = fr2.begin_token if (isinstance(t1, TextToken)): if (t1.chars.is_all_lower): return -1 if (BracketHelper.canBeStartOfSequence(t1, False, False)): if (t.chars.is_all_lower): return -1 elif (isinstance(t1, NumberToken)): if (t.chars.is_all_lower): return -1 if (t.chars.is_all_lower): if (fr2.end_token.isChar(';')): return -1 return 0
def __checkDetail(rt : 'ReferentToken') -> None: if (rt.end_token.whitespaces_after_count > 2 or rt.end_token.next0_ is None): return if (rt.end_token.next0_.isChar('(')): br = BracketHelper.tryParse(rt.end_token.next0_, BracketParseAttr.NO, 100) if (br is not None): (rt.referent).detail = MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO) rt.end_token = br.end_token
def __tryAttachSpeciality(t : 'Token', key_word_before : bool) -> 'TitleItemToken': if (t is None): return None susp = False if (not key_word_before): if (not t.is_newline_before): susp = True val = None t0 = t dig_count = 0 for i in range(3): nt = Utils.asObjectOrNull(t, NumberToken) if (nt is None): break if (nt.typ != NumberSpellingType.DIGIT or nt.morph.class0_.is_adjective): break if (val is None): val = io.StringIO() if (susp and t.length_char != 2): return None digs = nt.getSourceText() dig_count += len(digs) print(digs, end="", file=val) if (t.next0_ is None): break t = t.next0_ if (t.isCharOf(".,") or t.is_hiphen): if (susp and (i < 2)): if (not t.isChar('.') or t.is_whitespace_after or t.is_whitespace_before): return None if (t.next0_ is not None): t = t.next0_ if (val is None or (dig_count < 5)): return None if (dig_count != 6): if (not key_word_before): return None else: Utils.insertStringIO(val, 4, '.') Utils.insertStringIO(val, 2, '.') tt = t.next0_ first_pass3126 = True while True: if first_pass3126: first_pass3126 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.is_newline_before): break br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100) if (br is not None): tt = br.end_token t = tt continue t = tt return TitleItemToken._new2501(t0, t, TitleItemToken.Types.SPECIALITY, Utils.toStringStringIO(val))
def tryParse(t: 'Token', prev: 'TransItemToken', after_conj: bool, attach_high: bool = False) -> 'TransItemToken': res = TransItemToken.__TryParse(t, prev, after_conj, attach_high) if (res is None): return None if (res.typ == TransItemToken.Typs.NAME): br = BracketHelper.tryParse(res.end_token.next0_, BracketParseAttr.NO, 100) if (br is not None and br.begin_token.isChar('(')): alt = MiscHelper.getTextValueOfMetaToken(br, GetTextAttr.NO) if (MiscHelper.canBeEqualCyrAndLatSS(res.value, alt)): res.alt_value = alt res.end_token = br.end_token return res
def _addName(self, begin: 'Token', end: 'Token') -> 'Termin': if (BracketHelper.canBeStartOfSequence(begin, True, False)): br = BracketHelper.tryParse(begin, BracketParseAttr.NO, 100) if (br is not None and br.end_token == end): begin = begin.next0_ end = end.previous val = MiscHelper.getTextValue( begin, end, Utils.valToEnum( (GetTextAttr.KEEPREGISTER) | (GetTextAttr.KEEPQUOTES), GetTextAttr)) if (val is None): return None if (val.endswith(".") and not val.endswith("..")): val = val[0:0 + len(val) - 1].strip() self.addSlot(TitlePageReferent.ATTR_NAME, val, False, 0) return Termin(val.upper())
def tryParse(t: 'Token', prev: 'WeaponItemToken', after_conj: bool, attach_high: bool = False) -> 'WeaponItemToken': res = WeaponItemToken.__TryParse(t, prev, after_conj, attach_high) if (res is None): npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is not None and npt.noun.begin_char > npt.begin_char): res = WeaponItemToken.__TryParse(npt.noun.begin_token, prev, after_conj, attach_high) if (res is not None): if (res.typ == WeaponItemToken.Typs.NOUN): str0_ = npt.getNormalCaseText(None, True, MorphGender.UNDEFINED, False) if (str0_ == "РУЧНОЙ ГРАНАТ"): str0_ = "РУЧНАЯ ГРАНАТА" if ((Utils.ifNotNull(str0_, "")).endswith(res.value)): if (res.alt_value is None): res.alt_value = str0_ else: str0_ = str0_[0:0 + len(str0_) - len(res.value)].strip() res.alt_value = "{0} {1}".format( str0_, res.alt_value) res.begin_token = t return res return None if (res.typ == WeaponItemToken.Typs.NAME): br = BracketHelper.tryParse(res.end_token.next0_, BracketParseAttr.NO, 100) if (br is not None and br.isChar('(')): alt = MiscHelper.getTextValueOfMetaToken(br, GetTextAttr.NO) if (MiscHelper.canBeEqualCyrAndLatSS(res.value, alt)): res.alt_value = alt res.end_token = br.end_token return res
def __tryParse(t : 'Token', lev : int) -> 'BookLinkToken': if (t is None or lev > 3): return None if (t.isChar('[')): re = BookLinkToken.__tryParse(t.next0_, lev + 1) if (re is not None and re.end_token.next0_ is not None and re.end_token.next0_.isChar(']')): re.begin_token = t re.end_token = re.end_token.next0_ return re if (re is not None and re.end_token.isChar(']')): re.begin_token = t return re if (re is not None): if (re.typ == BookLinkTyp.SOSTAVITEL or re.typ == BookLinkTyp.EDITORS): return re br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): if ((isinstance(br.end_token.previous, NumberToken)) and (br.length_char < 30)): return BookLinkToken._new346(t, br.end_token, BookLinkTyp.NUMBER, MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO)) t0 = t if (isinstance(t, ReferentToken)): if (isinstance(t.getReferent(), PersonReferent)): return BookLinkToken.tryParseAuthor(t, FioTemplateType.UNDEFINED) if (isinstance(t.getReferent(), GeoReferent)): return BookLinkToken._new343(t, t, BookLinkTyp.GEO, t.getReferent()) if (isinstance(t.getReferent(), DateReferent)): dr = Utils.asObjectOrNull(t.getReferent(), DateReferent) if (len(dr.slots) == 1 and dr.year > 0): return BookLinkToken._new346(t, t, BookLinkTyp.YEAR, str(dr.year)) if (dr.year > 0 and t.previous is not None and t.previous.is_comma): return BookLinkToken._new346(t, t, BookLinkTyp.YEAR, str(dr.year)) if (isinstance(t.getReferent(), OrganizationReferent)): org0_ = Utils.asObjectOrNull(t.getReferent(), OrganizationReferent) if (org0_.kind == OrganizationKind.PRESS): return BookLinkToken._new343(t, t, BookLinkTyp.PRESS, org0_) if (isinstance(t.getReferent(), UriReferent)): uri = Utils.asObjectOrNull(t.getReferent(), UriReferent) if ((uri.scheme == "http" or uri.scheme == "https" or uri.scheme == "ftp") or uri.scheme is None): return BookLinkToken._new343(t, t, BookLinkTyp.URL, uri) tok_ = BookLinkToken.__m_termins.tryParse(t, TerminParseAttr.NO) if (tok_ is not None): typ_ = Utils.valToEnum(tok_.termin.tag, BookLinkTyp) ok = True if (typ_ == BookLinkTyp.TYPE or typ_ == BookLinkTyp.NAMETAIL or typ_ == BookLinkTyp.ELECTRONRES): if (t.previous is not None and ((t.previous.isCharOf(".:[") or t.previous.is_hiphen))): pass else: ok = False if (ok): return BookLinkToken._new346(t, tok_.end_token, typ_, tok_.termin.canonic_text) if (typ_ == BookLinkTyp.ELECTRONRES): tt = tok_.end_token.next0_ first_pass2751 = True while True: if first_pass2751: first_pass2751 = False else: tt = tt.next0_ if (not (tt is not None)): break if ((isinstance(tt, TextToken)) and not tt.chars.is_letter): continue if (isinstance(tt.getReferent(), UriReferent)): return BookLinkToken._new343(t, tt, BookLinkTyp.ELECTRONRES, tt.getReferent()) break if (t.isChar('/')): res = BookLinkToken._new346(t, t, BookLinkTyp.DELIMETER, "/") if (t.next0_ is not None and t.next0_.isChar('/')): res.end_token = t.next0_ res.value = "//" if (not t.is_whitespace_before and not t.is_whitespace_after): coo = 3 no = True tt = t.next0_ while tt is not None and coo > 0: vvv = BookLinkToken.tryParse(tt, lev + 1) if (vvv is not None and vvv.typ != BookLinkTyp.NUMBER): no = False break tt = tt.next0_; coo -= 1 if (no): return None return res if ((isinstance(t, NumberToken)) and (t).int_value is not None and (t).typ == NumberSpellingType.DIGIT): res = BookLinkToken._new346(t, t, BookLinkTyp.NUMBER, str((t).value)) val = (t).int_value if (val >= 1930 and (val < 2030)): res.typ = BookLinkTyp.YEAR if (t.next0_ is not None and t.next0_.isChar('.')): res.end_token = t.next0_ elif ((t.next0_ is not None and t.next0_.length_char == 1 and not t.next0_.chars.is_letter) and t.next0_.is_whitespace_after): res.end_token = t.next0_ elif (isinstance(t.next0_, TextToken)): term = (t.next0_).term if (((term == "СТР" or term == "C" or term == "С") or term == "P" or term == "S") or term == "PAGES"): res.end_token = t.next0_ res.typ = BookLinkTyp.PAGES res.value = str((t).value) return res if (isinstance(t, TextToken)): term = (t).term if (((((((term == "СТР" or term == "C" or term == "С") or term == "ТОМ" or term == "T") or term == "Т" or term == "P") or term == "PP" or term == "V") or term == "VOL" or term == "S") or term == "СТОР" or t.isValue("PAGE", None)) or t.isValue("СТРАНИЦА", "СТОРІНКА")): tt = t.next0_ while tt is not None: if (tt.isCharOf(".:~")): tt = tt.next0_ else: break if (isinstance(tt, NumberToken)): res = BookLinkToken._new345(t, tt, BookLinkTyp.PAGERANGE) tt0 = tt tt1 = tt tt = tt.next0_ first_pass2752 = True while True: if first_pass2752: first_pass2752 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.isCharOf(",") or tt.is_hiphen): if (isinstance(tt.next0_, NumberToken)): tt = tt.next0_ res.end_token = tt tt1 = tt continue break res.value = MiscHelper.getTextValue(tt0, tt1, GetTextAttr.NO) return res if ((term == "M" or term == "М" or term == "СПБ") or term == "K" or term == "К"): if (t.next0_ is not None and t.next0_.isCharOf(":;")): re = BookLinkToken._new345(t, t.next0_, BookLinkTyp.GEO) return re if (t.next0_ is not None and t.next0_.isCharOf(".")): res = BookLinkToken._new345(t, t.next0_, BookLinkTyp.GEO) if (t.next0_.next0_ is not None and t.next0_.next0_.isCharOf(":;")): res.end_token = t.next0_.next0_ elif (t.next0_.next0_ is not None and (isinstance(t.next0_.next0_, NumberToken))): pass elif (t.next0_.next0_ is not None and t.next0_.next0_.is_comma and (isinstance(t.next0_.next0_.next0_, NumberToken))): pass else: return None return res if (term == "ПЕР" or term == "ПЕРЕВ" or term == "ПЕРЕВОД"): tt = t if (tt.next0_ is not None and tt.next0_.isChar('.')): tt = tt.next0_ if (tt.next0_ is not None and ((tt.next0_.isValue("C", None) or tt.next0_.isValue("С", None)))): tt = tt.next0_ if (tt.next0_ is None or tt.whitespaces_after_count > 2): return None re = BookLinkToken._new345(t, tt.next0_, BookLinkTyp.TRANSLATE) return re if (term == "ТАМ" or term == "ТАМЖЕ"): res = BookLinkToken._new345(t, t, BookLinkTyp.TAMZE) if (t.next0_ is not None and t.next0_.isValue("ЖЕ", None)): res.end_token = t.next0_ return res if (((term == "СМ" or term == "CM" or term == "НАПР") or term == "НАПРИМЕР" or term == "SEE") or term == "ПОДРОБНЕЕ" or term == "ПОДРОБНО"): res = BookLinkToken._new345(t, t, BookLinkTyp.SEE) t = t.next0_ first_pass2753 = True while True: if first_pass2753: first_pass2753 = False else: t = t.next0_ if (not (t is not None)): break if (t.isCharOf(".:") or t.isValue("ALSO", None)): res.end_token = t continue if (t.isValue("В", None) or t.isValue("IN", None)): res.end_token = t continue vvv = BookLinkToken.__tryParse(t, lev + 1) if (vvv is not None and vvv.typ == BookLinkTyp.SEE): res.end_token = vvv.end_token break break return res if (term == "БОЛЕЕ"): vvv = BookLinkToken.__tryParse(t.next0_, lev + 1) if (vvv is not None and vvv.typ == BookLinkTyp.SEE): vvv.begin_token = t return vvv no = MiscHelper.checkNumberPrefix(t) if (isinstance(no, NumberToken)): return BookLinkToken._new345(t, no, BookLinkTyp.N) if (((term == "B" or term == "В")) and (isinstance(t.next0_, NumberToken)) and (isinstance(t.next0_.next0_, TextToken))): term2 = (t.next0_.next0_).term if (((term2 == "Т" or term2 == "T" or term2.startswith("ТОМ")) or term2 == "TT" or term2 == "ТТ") or term2 == "КН" or term2.startswith("КНИГ")): return BookLinkToken._new345(t, t.next0_.next0_, BookLinkTyp.VOLUME) if (t.isChar('(')): if (((isinstance(t.next0_, NumberToken)) and (t.next0_).int_value is not None and t.next0_.next0_ is not None) and t.next0_.next0_.isChar(')')): num = (t.next0_).int_value if (num > 1900 and num <= 2040): if (num <= datetime.datetime.now().year): return BookLinkToken._new346(t, t.next0_.next0_, BookLinkTyp.YEAR, str(num)) if (((isinstance(t.next0_, ReferentToken)) and (isinstance(t.next0_.getReferent(), DateReferent)) and t.next0_.next0_ is not None) and t.next0_.next0_.isChar(')')): num = (t.next0_.getReferent()).year if (num > 0): return BookLinkToken._new346(t, t.next0_.next0_, BookLinkTyp.YEAR, str(num)) return None
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 __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 process(self, kit: 'AnalysisKit') -> None: ad = kit.getAnalyzerData(self) models = TerminCollection() objs_by_model = dict() obj_by_names = TerminCollection() t = kit.first_token first_pass3158 = True while True: if first_pass3158: first_pass3158 = False else: t = t.next0_ if (not (t is not None)): break its = WeaponItemToken.tryParseList(t, 10) if (its is None): continue rts = self.__tryAttach(its, False) if (rts is not None): for rt in rts: rt.referent = ad.registerReferent(rt.referent) kit.embedToken(rt) t = (rt) for s in rt.referent.slots: if (s.type_name == WeaponReferent.ATTR_MODEL): mod = str(s.value) for k in range(2): if (not str.isdigit(mod[0])): li = [] wrapli2638 = RefOutArgWrapper(None) inoutres2639 = Utils.tryGetValue( objs_by_model, mod, wrapli2638) li = wrapli2638.value if (not inoutres2639): li = list() objs_by_model[mod] = li if (not rt.referent in li): li.append(rt.referent) models.addStr(mod, li, None, False) if (k > 0): break brand = rt.referent.getStringValue( WeaponReferent.ATTR_BRAND) if (brand is None): break mod = "{0} {1}".format(brand, mod) elif (s.type_name == WeaponReferent.ATTR_NAME): obj_by_names.add( Termin._new117(str(s.value), rt.referent)) if (len(objs_by_model) == 0 and len(obj_by_names.termins) == 0): return t = kit.first_token first_pass3159 = True while True: if first_pass3159: first_pass3159 = False else: t = t.next0_ if (not (t is not None)): break br = BracketHelper.tryParse(t, BracketParseAttr.NO, 10) if (br is not None): toks = obj_by_names.tryParse(t.next0_, TerminParseAttr.NO) if (toks is not None and toks.end_token.next0_ == br.end_token): rt0 = ReferentToken( Utils.asObjectOrNull(toks.termin.tag, Referent), br.begin_token, br.end_token) kit.embedToken(rt0) t = (rt0) continue if (not ((isinstance(t, TextToken)))): continue if (not t.chars.is_letter): continue tok = models.tryParse(t, TerminParseAttr.NO) if (tok is None): if (not t.chars.is_all_lower): tok = obj_by_names.tryParse(t, TerminParseAttr.NO) if (tok is None): continue if (not tok.is_whitespace_after): if (tok.end_token.next0_ is None or not tok.end_token.next0_.isCharOf(",.)")): if (not BracketHelper.isBracket(tok.end_token.next0_, False)): continue tr = None li = Utils.asObjectOrNull(tok.termin.tag, list) if (li is not None and len(li) == 1): tr = li[0] else: tr = (Utils.asObjectOrNull(tok.termin.tag, Referent)) if (tr is not None): tit = WeaponItemToken.tryParse(tok.begin_token.previous, None, False, True) if (tit is not None and tit.typ == WeaponItemToken.Typs.BRAND): tr.addSlot(WeaponReferent.ATTR_BRAND, tit.value, False, 0) tok.begin_token = tit.begin_token rt0 = ReferentToken(tr, tok.begin_token, tok.end_token) kit.embedToken(rt0) t = (rt0) continue
def createNickname(pr : 'PersonReferent', t : 'Token') -> 'Token': """ Выделить кличку Args: pr(PersonReferent): t(Token): начальный токен Returns: Token: если не null, то последний токен клички, а в pr запишет саму кличку """ has_keyw = False is_br = False first_pass3097 = True while True: if first_pass3097: first_pass3097 = False else: t = t.next0_ if (not (t is not None)): break if (t.is_hiphen or t.is_comma or t.isCharOf(".:;")): continue if (t.morph.class0_.is_preposition): continue if (t.isChar('(')): is_br = True continue if ((t.isValue("ПРОЗВИЩЕ", "ПРІЗВИСЬКО") or t.isValue("КЛИЧКА", None) or t.isValue("ПСЕВДОНИМ", "ПСЕВДОНІМ")) or t.isValue("ПСЕВДО", None) or t.isValue("ПОЗЫВНОЙ", "ПОЗИВНИЙ")): has_keyw = True continue break if (not has_keyw or t is None): return None if (BracketHelper.isBracket(t, True)): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): ni = MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO) if (ni is not None): pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0) t = br.end_token tt = t.next0_ first_pass3098 = True while True: if first_pass3098: first_pass3098 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.is_comma_and): continue if (not BracketHelper.isBracket(tt, True)): break br = BracketHelper.tryParse(tt, BracketParseAttr.NO, 100) if (br is None): break ni = MiscHelper.getTextValue(br.begin_token.next0_, br.end_token.previous, GetTextAttr.NO) if (ni is not None): pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0) tt = br.end_token t = tt if (is_br and t.next0_ is not None and t.next0_.isChar(')')): t = t.next0_ return t else: pli = PersonItemToken.tryAttachList(t, None, PersonItemToken.ParseAttr.NO, 10) if (pli is not None and ((len(pli) == 1 or len(pli) == 2))): ni = MiscHelper.getTextValue(pli[0].begin_token, pli[len(pli) - 1].end_token, GetTextAttr.NO) if (ni is not None): pr.addSlot(PersonReferent.ATTR_NICKNAME, ni, False, 0) t = pli[len(pli) - 1].end_token if (is_br and t.next0_ is not None and t.next0_.isChar(')')): t = t.next0_ return t return None
def __TryParse(t: 'Token', prev: 'TransItemToken', after_conj: bool, attach_high: bool = False) -> 'TransItemToken': if (t is None): return None t1 = t if (t1.isChar(',')): t1 = t1.next0_ if (t1 is not None and t1.isValue("ПРИНАДЛЕЖАТЬ", "НАЛЕЖАТИ")): t1 = t1.next0_ if (isinstance(t1, ReferentToken)): if (t1.getReferent().type_name == "ORGANIZATION"): return TransItemToken._new2521(t, t1, TransItemToken.Typs.ORG, t1.getReferent(), t1.morph) route = False if (t1 is not None and ((t1.isValue("СЛЕДОВАТЬ", "СЛІДУВАТИ") or t1.isValue("ВЫПОЛНЯТЬ", "ВИКОНУВАТИ")))): t1 = t1.next0_ route = True if (t1 is not None and t1.morph.class0_.is_preposition): t1 = t1.next0_ if (t1 is not None and ((t1.isValue("РЕЙС", None) or t1.isValue("МАРШРУТ", None)))): t1 = t1.next0_ route = True if (isinstance(t1, ReferentToken)): if (isinstance(t1.getReferent(), GeoReferent)): geo_ = Utils.asObjectOrNull(t1.getReferent(), GeoReferent) if (geo_.is_state or geo_.is_city): tit = TransItemToken._new2522(t, t1, TransItemToken.Typs.ROUTE, list()) tit.route_items.append(geo_) t1 = t1.next0_ first_pass3132 = True while True: if first_pass3132: first_pass3132 = False else: t1 = t1.next0_ if (not (t1 is not None)): break if (t1.is_hiphen): continue if (t1.morph.class0_.is_preposition or t1.morph.class0_.is_conjunction): continue geo_ = (Utils.asObjectOrNull(t1.getReferent(), GeoReferent)) if (geo_ is None): break if (not geo_.is_city and not geo_.is_state): break tit.route_items.append(geo_) tit.end_token = t1 if (len(tit.route_items) > 1 or route): return tit elif ((isinstance(t1.getReferent(), DateReferent)) and (t1.whitespaces_before_count < 3)): tit = TransItemToken._new2523(t, t1, TransItemToken.Typs.DATE, t1.getReferent()) if (t1.next0_ is not None): if (t1.next0_.isValue("В", None) and t1.next0_.next0_ is not None and t1.next0_.next0_.isChar('.')): tit.end_token = t1.next0_.next0_ elif (t1.next0_.isValue("ВЫП", None) or t1.next0_.isValue("ВЫПУСК", None)): tit.end_token = t1.next0_ if (t1.next0_.next0_ is not None and t1.next0_.next0_.isChar('.')): tit.end_token = t1.next0_.next0_ return tit if (isinstance(t, TextToken)): num = MiscHelper.checkNumberPrefix(t) if (num is not None): tit = TransItemToken.__attachRusAutoNumber(num) if (tit is None): tit = TransItemToken._attachNumber(num, False) if (tit is not None): tit.begin_token = t return tit tok = TransItemToken.M_ONTOLOGY.tryParse(t, TerminParseAttr.NO) if (tok is None and ((t.isValue("С", None) or t.isValue("C", None) or t.isValue("ЗА", None)))): tok = TransItemToken.M_ONTOLOGY.tryParse( t.next0_, TerminParseAttr.NO) if (tok is None and BracketHelper.isBracket(t, True)): tok1 = TransItemToken.M_ONTOLOGY.tryParse( t.next0_, TerminParseAttr.NO) if (tok1 is not None and BracketHelper.isBracket( tok1.end_token.next0_, True)): tok = tok1 tok.begin_token = t tok.end_token = tok.end_token.next0_ tok.begin_token = t elif (tok1 is not None): tt = Utils.asObjectOrNull(tok1.termin, TransItemToken.TransTermin) if (tt.typ == TransItemToken.Typs.BRAND): tok = tok1 tok.begin_token = t if (tok is None and t.isValue("МАРКА", None)): res1 = TransItemToken.__TryParse(t.next0_, prev, after_conj, False) if (res1 is not None): if (res1.typ == TransItemToken.Typs.NAME or res1.typ == TransItemToken.Typs.BRAND): res1.begin_token = t res1.typ = TransItemToken.Typs.BRAND return res1 if (tok is not None): tt = Utils.asObjectOrNull(tok.termin, TransItemToken.TransTermin) if (tt.typ == TransItemToken.Typs.NUMBER): tit = TransItemToken.__attachRusAutoNumber( tok.end_token.next0_) if (tit is None): tit = TransItemToken._attachNumber( tok.end_token.next0_, False) if (tit is not None): tit.begin_token = t return tit else: return None if (tt.is_doubt and not attach_high): if (prev is None or prev.typ != TransItemToken.Typs.NOUN): if ((prev is not None and prev.typ == TransItemToken.Typs.BRAND and tt.typ == TransItemToken.Typs.BRAND) and Utils.compareStrings( tt.canonic_text, prev.value, True) == 0): pass else: return None if (tt.canonic_text == "СУДНО"): if ((((tok.morph.number) & (MorphNumber.PLURAL))) != (MorphNumber.UNDEFINED)): if (not BracketHelper.canBeStartOfSequence( tok.end_token.next0_, False, False)): return None tit = TransItemToken._new2524(tok.begin_token, tok.end_token, tt.kind, tt.typ, tt.is_doubt, tok.chars, tok.morph) tit.value = tt.canonic_text if (tit.typ == TransItemToken.Typs.NOUN): tit.value = tit.value.lower() else: tit.value = tit.value.upper() return tit if (tok is None and t.morph.class0_.is_adjective): npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is not None and len(npt.adjectives) > 0): state_ = None tt = t first_pass3133 = True while True: if first_pass3133: first_pass3133 = False else: tt = tt.next0_ if (not (tt is not None and tt.previous != npt.end_token)): break tok = TransItemToken.M_ONTOLOGY.tryParse( tt, TerminParseAttr.NO) if (tok is None and state_ is None): state_ = tt.kit.processReferent("GEO", tt) if (tok is not None and tok.end_token == npt.end_token): if ((tok.termin).typ == TransItemToken.Typs.NOUN): tit = TransItemToken._new2524( t, tok.end_token, (tok.termin).kind, TransItemToken.Typs.NOUN, (tok.termin).is_doubt, tok.chars, npt.morph) tit.value = (tok.termin).canonic_text.lower() tit.alt_value = npt.getNormalCaseText( None, False, MorphGender.UNDEFINED, False).lower() if (LanguageHelper.endsWithEx( tit.alt_value, "суд", "суда", None, None)): if (not BracketHelper.canBeStartOfSequence( tok.end_token.next0_, False, False)): continue if (state_ is not None): if ((state_.referent).is_state): tit.state = state_ return tit if (t is not None and t.isValue("КЛАСС", None) and t.next0_ is not None): br = BracketHelper.tryParse(t.next0_, BracketParseAttr.NO, 100) if (br is not None): return TransItemToken._new2526( t, br.end_token, TransItemToken.Typs.CLASS, MiscHelper.getTextValueOfMetaToken(br, GetTextAttr.NO)) nt = Utils.asObjectOrNull(t, NumberToken) if (nt is not None): if (prev is None or nt.typ != NumberSpellingType.DIGIT): return None if (prev.typ == TransItemToken.Typs.BRAND): return TransItemToken.__attachModel(t, False, prev) else: return None res = TransItemToken.__attachRusAutoNumber(t) if ((res) is not None): if (not res.is_doubt): return res if (prev is not None and prev.typ == TransItemToken.Typs.NOUN and prev.kind == TransportKind.AUTO): return res if (prev is not None and ((prev.typ == TransItemToken.Typs.BRAND or prev.typ == TransItemToken.Typs.MODEL))): return res t1 = t if (t.is_hiphen): t1 = t.next0_ if (prev is not None and prev.typ == TransItemToken.Typs.BRAND and t1 is not None): tit = TransItemToken.__attachModel(t1, True, prev) if (tit is not None): tit.begin_token = t return tit if (prev is not None and ((prev.typ == TransItemToken.Typs.NOUN or after_conj))): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None and br.is_quote_type): tit = TransItemToken.tryParse(br.begin_token.next0_, prev, after_conj, False) if (tit is not None and tit.end_token.next0_ == br.end_token): if (not tit.is_doubt or tit.typ == TransItemToken.Typs.BRAND): tit.begin_token = br.begin_token tit.end_token = br.end_token return tit s = MiscHelper.getTextValueOfMetaToken(br, GetTextAttr.NO) if (not Utils.isNullOrEmpty(s) and (len(s) < 30)): chars_ = 0 digs = 0 un = 0 for c in s: if (not Utils.isWhitespace(c)): if (str.isalpha(c)): chars_ += 1 elif (str.isdigit(c)): digs += 1 else: un += 1 if (((digs == 0 and un == 0 and t.next0_.chars.is_capital_upper)) or prev.kind == TransportKind.SHIP or prev.kind == TransportKind.SPACE): return TransItemToken._new2526( br.begin_token, br.end_token, TransItemToken.Typs.NAME, s) if (digs > 0 and (chars_ < 5)): return TransItemToken._new2526( br.begin_token, br.end_token, TransItemToken.Typs.MODEL, s.replace(" ", "")) if (prev is not None and (((prev.typ == TransItemToken.Typs.NOUN or prev.typ == TransItemToken.Typs.BRAND or prev.typ == TransItemToken.Typs.NAME) or prev.typ == TransItemToken.Typs.MODEL))): tit = TransItemToken.__attachModel( t, prev.typ != TransItemToken.Typs.NAME, prev) if (tit is not None): return tit if (((prev is not None and prev.typ == TransItemToken.Typs.NOUN and prev.kind == TransportKind.AUTO) and (isinstance(t, TextToken)) and t.chars.is_letter) and not t.chars.is_all_lower and (t.whitespaces_before_count < 2)): pt = t.kit.processReferent("PERSON", t) if (pt is None): tit = TransItemToken._new2529(t, t, TransItemToken.Typs.BRAND) tit.value = (t).term return tit if (((prev is not None and prev.typ == TransItemToken.Typs.NOUN and ((prev.kind == TransportKind.SHIP or prev.kind == TransportKind.SPACE)))) or after_conj): if (t.chars.is_capital_upper): ok = True npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is not None and len(npt.adjectives) > 0): ok = False else: rt = t.kit.processReferent("PERSON", t) if (rt is not None): ok = False if (t.getMorphClassInDictionary().is_proper_surname): if (not t.morph.case_.is_nominative): ok = False if (ok): t1 = t tt = t.next0_ while tt is not None: if (tt.whitespaces_before_count > 1): break if (tt.chars != t.chars): break tit = TransItemToken.tryParse(tt, None, False, False) if ((tit) is not None): break t1 = tt tt = tt.next0_ s = MiscHelper.getTextValue(t, t1, GetTextAttr.NO) if (s is not None): res1 = TransItemToken._new2530( t, t1, TransItemToken.Typs.NAME, True, s) if (not t1.is_newline_after): br = BracketHelper.tryParse( t1.next0_, BracketParseAttr.NO, 100) if (br is not None): res1.end_token = br.end_token res1.alt_value = res1.value res1.value = MiscHelper.getTextValueOfMetaToken( br, GetTextAttr.NO) return res1 return None
def tryParseList(t: 'Token', max_count: int = 10) -> typing.List['TransItemToken']: tr = TransItemToken.tryParse(t, None, False, False) if (tr is None): return None if ((tr.typ == TransItemToken.Typs.ORG or tr.typ == TransItemToken.Typs.NUMBER or tr.typ == TransItemToken.Typs.CLASS) or tr.typ == TransItemToken.Typs.DATE): return None tr0 = tr res = list() res.append(tr) t = tr.end_token.next0_ if (tr.typ == TransItemToken.Typs.NOUN): while t is not None: if (t.isChar(':') or t.is_hiphen): pass else: break t = t.next0_ and_conj = False while t is not None: if (max_count > 0 and len(res) >= max_count): break if (tr0.typ == TransItemToken.Typs.NOUN or tr0.typ == TransItemToken.Typs.ORG): if (t.is_hiphen and t.next0_ is not None): t = t.next0_ tr = TransItemToken.tryParse(t, tr0, False, False) if (tr is None): if (BracketHelper.canBeEndOfSequence(t, True, None, False) and t.next0_ is not None): if (tr0.typ == TransItemToken.Typs.MODEL or tr0.typ == TransItemToken.Typs.BRAND): tt1 = t.next0_ if (tt1 is not None and tt1.is_comma): tt1 = tt1.next0_ tr = TransItemToken.tryParse(tt1, tr0, False, False) if (tr is None and (isinstance(t, ReferentToken))): rt = Utils.asObjectOrNull(t, ReferentToken) if (rt.begin_token == rt.end_token and (isinstance(rt.begin_token, TextToken))): tr = TransItemToken.tryParse(rt.begin_token, tr0, False, False) if (tr is not None and tr.begin_token == tr.end_token): tr.begin_token = tr.end_token = t if (tr is None and t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): tr = TransItemToken.tryParse(t.next0_, tr0, False, False) if (tr is not None): if (tr.typ != TransItemToken.Typs.NUMBER): tr = (None) elif (tr.end_token.next0_ is not None): tr.begin_token = t if (tr.end_token.next0_.isChar(')')): tr.end_token = tr.end_token.next0_ if (tr is None): tt = br.end_token.next0_ if (tt is not None and tt.is_comma): tt = tt.next0_ tr = TransItemToken.tryParse(tt, tr0, False, False) if (tr is not None and tr.typ == TransItemToken.Typs.NUMBER): pass else: tr = (None) if (tr is None and t.is_hiphen): if (tr0.typ == TransItemToken.Typs.BRAND or tr0.typ == TransItemToken.Typs.MODEL): tr = TransItemToken.tryParse(t.next0_, tr0, False, False) if (tr is None and t.is_comma): if ((tr0.typ == TransItemToken.Typs.NAME or tr0.typ == TransItemToken.Typs.BRAND or tr0.typ == TransItemToken.Typs.MODEL) or tr0.typ == TransItemToken.Typs.CLASS or tr0.typ == TransItemToken.Typs.DATE): tr = TransItemToken.tryParse(t.next0_, tr0, True, False) if (tr is not None): if (tr.typ == TransItemToken.Typs.NUMBER): pass else: tr = (None) if (tr is None): if (tr0.typ == TransItemToken.Typs.NAME): if (t.isChar(',')): tr = TransItemToken.tryParse(t.next0_, tr0, True, False) elif (t.morph.class0_.is_conjunction and t.is_and): tr = TransItemToken.tryParse(t.next0_, tr0, True, False) and_conj = True if (tr is not None): if (tr.typ != TransItemToken.Typs.NAME): break tr.is_after_conjunction = True if (tr is None): break if (t.is_newline_before): if (tr.typ != TransItemToken.Typs.NUMBER): break res.append(tr) if (tr.typ == TransItemToken.Typs.ORG and tr0.typ == TransItemToken.Typs.NOUN): pass else: tr0 = tr t = tr.end_token if (and_conj): break t = t.next0_ i = 0 while i < (len(res) - 1): if (res[i].typ == TransItemToken.Typs.MODEL and res[i + 1].typ == TransItemToken.Typs.MODEL): res[i].end_token = res[i + 1].end_token res[i].value = "{0}{1}{2}".format( res[i].value, ('-' if res[i].end_token.next0_ is not None and res[i].end_token.next0_.is_hiphen else ' '), res[i + 1].value) del res[i + 1] i -= 1 i += 1 if ((len(res) > 1 and res[0].typ == TransItemToken.Typs.BRAND and res[1].typ == TransItemToken.Typs.MODEL) and res[1].length_char == 1 and not ((isinstance(res[1].begin_token, NumberToken)))): return None return res
def getNameEx(begin: 'Token', end: 'Token', cla: 'MorphClass', mc: 'MorphCase', gender: 'MorphGender' = MorphGender.UNDEFINED, ignore_brackets_and_hiphens: bool = False, ignore_geo_referent: bool = False) -> str: if (end is None or begin is None): return None if (begin.end_char > end.begin_char and begin != end): return None res = io.StringIO() prefix = None t = begin first_pass2809 = True while True: if first_pass2809: first_pass2809 = False else: t = t.next0_ if (not (t is not None and t.end_char <= end.end_char)): break if (res.tell() > 1000): break if (t.is_table_control_char): continue if (ignore_brackets_and_hiphens): if (BracketHelper.isBracket(t, False)): if (t == end): break if (t.isCharOf("(<[")): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None and br.end_char <= end.end_char): tmp = ProperNameHelper.getNameEx( br.begin_token.next0_, br.end_token.previous, MorphClass.UNDEFINED, MorphCase.UNDEFINED, MorphGender.UNDEFINED, ignore_brackets_and_hiphens, False) if (tmp is not None): if ((br.end_char == end.end_char and br.begin_token.next0_ == br.end_token.previous and not br.begin_token.next0_.chars.is_letter) and not ((isinstance( br.begin_token.next0_, ReferentToken)))): pass else: print(" {0}{1}{2}".format( t.getSourceText(), tmp, br.end_token.getSourceText()), end="", file=res, flush=True) t = br.end_token continue if (t.is_hiphen): if (t == end): break elif (t.is_whitespace_before or t.is_whitespace_after): continue tt = Utils.asObjectOrNull(t, TextToken) if (tt is not None): if (not ignore_brackets_and_hiphens): if ((tt.next0_ is not None and tt.next0_.is_hiphen and (isinstance(tt.next0_.next0_, TextToken))) and tt != end and tt.next0_ != end): if (prefix is None): prefix = tt.term else: prefix = "{0}-{1}".format(prefix, tt.term) t = tt.next0_ if (t == end): break else: continue s = None if (cla.value != (0) or not mc.is_undefined or gender != MorphGender.UNDEFINED): for wff in tt.morph.items: wf = Utils.asObjectOrNull(wff, MorphWordForm) if (wf is None): continue if (cla.value != (0)): if ((((wf.class0_.value) & (cla.value))) == 0): continue if (not mc.is_undefined): if (((wf.case_) & mc).is_undefined): continue if (gender != MorphGender.UNDEFINED): if ((((wf.gender) & (gender))) == (MorphGender.UNDEFINED)): continue if (s is None or wf.normal_case == tt.term): s = wf.normal_case if (s is None and gender != MorphGender.UNDEFINED): for wff in tt.morph.items: wf = Utils.asObjectOrNull(wff, MorphWordForm) if (wf is None): continue if (cla.value != (0)): if ((((wf.class0_.value) & (cla.value))) == 0): continue if (not mc.is_undefined): if (((wf.case_) & mc).is_undefined): continue if (s is None or wf.normal_case == tt.term): s = wf.normal_case if (s is None): s = tt.term if (tt.chars.is_last_lower and tt.length_char > 2): s = tt.getSourceText() for i in range(len(s) - 1, -1, -1): if (str.isupper(s[i])): s = s[0:0 + i + 1] break if (prefix is not None): delim = "-" if (ignore_brackets_and_hiphens): delim = " " s = "{0}{1}{2}".format(prefix, delim, s) prefix = (None) if (res.tell() > 0 and len(s) > 0): if (str.isalnum(s[0])): ch0 = Utils.getCharAtStringIO(res, res.tell() - 1) if (ch0 == '-'): pass else: print(' ', end="", file=res) elif (not ignore_brackets_and_hiphens and BracketHelper.canBeStartOfSequence( tt, False, False)): print(' ', end="", file=res) print(s, end="", file=res) elif (isinstance(t, NumberToken)): if (res.tell() > 0): if (not t.is_whitespace_before and Utils.getCharAtStringIO( res, res.tell() - 1) == '-'): pass else: print(' ', end="", file=res) nt = Utils.asObjectOrNull(t, NumberToken) if ((t.morph.class0_.is_adjective and nt.typ == NumberSpellingType.WORDS and nt.begin_token == nt.end_token) and (isinstance(nt.begin_token, TextToken))): print((nt.begin_token).term, end="", file=res) else: print(nt.value, end="", file=res) elif (isinstance(t, MetaToken)): if ((ignore_geo_referent and t != begin and t.getReferent() is not None) and t.getReferent().type_name == "GEO"): continue s = ProperNameHelper.getNameEx( (t).begin_token, (t).end_token, cla, mc, gender, ignore_brackets_and_hiphens, ignore_geo_referent) if (not Utils.isNullOrEmpty(s)): if (res.tell() > 0): if (not t.is_whitespace_before and Utils.getCharAtStringIO( res, res.tell() - 1) == '-'): pass else: print(' ', end="", file=res) print(s, end="", file=res) if (t == end): break if (res.tell() == 0): return None return Utils.toStringStringIO(res)
def tryParseList(t: 'Token', max_count: int = 10) -> typing.List['WeaponItemToken']: tr = WeaponItemToken.tryParse(t, None, False, False) if (tr is None): return None if (tr.typ == WeaponItemToken.Typs.CLASS or tr.typ == WeaponItemToken.Typs.DATE): return None tr0 = tr res = list() if (len(tr.__inner_tokens) > 0): res.extend(tr.__inner_tokens) if (res[0].begin_char > tr.begin_char): res[0].begin_token = tr.begin_token res.append(tr) t = tr.end_token.next0_ if (tr.typ == WeaponItemToken.Typs.NOUN): while t is not None: if (t.isChar(':') or t.is_hiphen): pass else: break t = t.next0_ and_conj = False first_pass3155 = True while True: if first_pass3155: first_pass3155 = False else: t = t.next0_ if (not (t is not None)): break if (max_count > 0 and len(res) >= max_count): break if (t.isChar(':')): continue if (tr0.typ == WeaponItemToken.Typs.NOUN): if (t.is_hiphen and t.next0_ is not None): t = t.next0_ tr = WeaponItemToken.tryParse(t, tr0, False, False) if (tr is None): if (BracketHelper.canBeEndOfSequence(t, True, None, False) and t.next0_ is not None): if (tr0.typ == WeaponItemToken.Typs.MODEL or tr0.typ == WeaponItemToken.Typs.BRAND): tt1 = t.next0_ if (tt1 is not None and tt1.is_comma): tt1 = tt1.next0_ tr = WeaponItemToken.tryParse(tt1, tr0, False, False) if (tr is None and (isinstance(t, ReferentToken))): rt = Utils.asObjectOrNull(t, ReferentToken) if (rt.begin_token == rt.end_token and (isinstance(rt.begin_token, TextToken))): tr = WeaponItemToken.tryParse(rt.begin_token, tr0, False, False) if (tr is not None and tr.begin_token == tr.end_token): tr.begin_token = tr.end_token = t if (tr is None and t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): tt = br.end_token.next0_ if (tt is not None and tt.is_comma): tt = tt.next0_ tr = WeaponItemToken.tryParse(tt, tr0, False, False) if (tr is not None and tr.typ == WeaponItemToken.Typs.NUMBER): pass else: tr = (None) if (tr is None and t.is_hiphen): if (tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL): tr = WeaponItemToken.tryParse(t.next0_, tr0, False, False) if (tr is None and t.is_comma): if ((tr0.typ == WeaponItemToken.Typs.NAME or tr0.typ == WeaponItemToken.Typs.BRAND or tr0.typ == WeaponItemToken.Typs.MODEL) or tr0.typ == WeaponItemToken.Typs.CLASS or tr0.typ == WeaponItemToken.Typs.DATE): tr = WeaponItemToken.tryParse(t.next0_, tr0, True, False) if (tr is not None): if (tr.typ == WeaponItemToken.Typs.NUMBER): pass else: tr = (None) if (tr is None): break if (t.is_newline_before): if (tr.typ != WeaponItemToken.Typs.NUMBER): break if (len(tr.__inner_tokens) > 0): res.extend(tr.__inner_tokens) res.append(tr) tr0 = tr t = tr.end_token if (and_conj): break i = 0 while i < (len(res) - 1): if (res[i].typ == WeaponItemToken.Typs.MODEL and res[i + 1].typ == WeaponItemToken.Typs.MODEL): res[i].end_token = res[i + 1].end_token res[i].value = "{0}{1}{2}".format( res[i].value, ('-' if res[i].end_token.next0_ is not None and res[i].end_token.next0_.is_hiphen else ' '), res[i + 1].value) del res[i + 1] i -= 1 i += 1 return res
def __analizeProfit(self, bfi: 'BusinessFactItem') -> 'ReferentToken': if (bfi.end_token.next0_ is None): return None t0 = bfi.begin_token t1 = bfi.end_token typ = t1.getNormalCaseText(None, True, MorphGender.UNDEFINED, False).lower() org0_ = None org0_ = (Utils.asObjectOrNull(t1.next0_.getReferent(), OrganizationReferent)) t = t1 if (org0_ is not None): t = t.next0_ else: rt = t.kit.processReferent(OrganizationAnalyzer.ANALYZER_NAME, t.next0_) if (rt is not None): org0_ = (Utils.asObjectOrNull(rt.referent, OrganizationReferent)) t = rt.end_token dt = None sum0_ = None t = t.next0_ first_pass2774 = True while True: if first_pass2774: first_pass2774 = False else: t = t.next0_ if (not (t is not None)): break if (t.isChar('.')): break if (t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): t = br.end_token continue if ((((t.morph.class0_.is_verb or t.isValue("ДО", None) or t.is_hiphen) or t.isValue("РАЗМЕР", None) or t.isValue("РОЗМІР", None))) and t.next0_ is not None and (isinstance(t.next0_.getReferent(), MoneyReferent))): if (sum0_ is not None): break sum0_ = (Utils.asObjectOrNull(t.next0_.getReferent(), MoneyReferent)) t = t.next0_ t1 = t continue r = t.getReferent() if ((isinstance(r, DateRangeReferent)) or (isinstance(r, DateReferent))): if (dt is None): dt = r t1 = t elif ((isinstance(r, OrganizationReferent)) and org0_ is None): org0_ = (Utils.asObjectOrNull(r, OrganizationReferent)) t1 = t if (sum0_ is None): return None if (org0_ is None): tt = t0.previous while tt is not None: if (tt.isChar('.')): break b0 = Utils.asObjectOrNull(tt.getReferent(), BusinessFactReferent) if (b0 is not None): org0_ = (Utils.asObjectOrNull(b0.who, OrganizationReferent)) break org0_ = Utils.asObjectOrNull(tt.getReferent(), OrganizationReferent) if ((org0_) is not None): break tt = tt.previous if (org0_ is None): return None bfr = BusinessFactReferent._new436(bfi.base_kind) bfr.who = org0_ bfr.typ = typ bfr.addSlot(BusinessFactReferent.ATTR_MISC, sum0_, False, 0) if (dt is not None): bfr.when = dt else: self.__findDate(bfr, bfi.begin_token) return ReferentToken(bfr, t0, t1)
def parse(t0: 'Token', lev_: int) -> 'MailLine': if (t0 is None): return None res = MailLine(t0, t0) pr = True t = t0 first_pass3027 = True while True: if first_pass3027: first_pass3027 = False else: t = t.next0_ if (not (t is not None)): break if (t.is_newline_before and t0 != t): break res.end_token = t if (t.is_table_control_char or t.is_hiphen): continue if (pr): if ((isinstance(t, TextToken)) and t.isCharOf(">|")): res.lev += 1 else: pr = False tok = MailLine.M_FROM_WORDS.tryParse(t, TerminParseAttr.NO) if (tok is not None and tok.end_token.next0_ is not None and tok.end_token.next0_.isChar(':')): res.typ = MailLine.Types.FROM t = tok.end_token.next0_ continue if (isinstance(t, ReferentToken)): r = t.getReferent() if (r is not None): if ((((isinstance(r, PersonReferent)) or (isinstance(r, GeoReferent)) or (isinstance(r, AddressReferent))) or r.type_name == "PHONE" or r.type_name == "URI") or (isinstance(r, PersonPropertyReferent)) or r.type_name == "ORGANIZATION"): res.refs.append(r) if (res.typ == MailLine.Types.UNDEFINED): t = t0 while t is not None and (t.end_char < res.end_char): if (not t.is_hiphen and t.chars.is_letter): break t = t.next0_ ok = 0 nams = 0 oth = 0 last_comma = None first_pass3028 = True while True: if first_pass3028: first_pass3028 = False else: t = t.next0_ if (not (t is not None and (t.end_char < res.end_char))): break if (isinstance(t.getReferent(), PersonReferent)): nams += 1 continue if (isinstance(t, TextToken)): if (not t.chars.is_letter): last_comma = t continue tok = MailLine.M_HELLO_WORDS.tryParse( t, TerminParseAttr.NO) if (tok is not None): ok += 1 t = tok.end_token continue if (t.isValue("ВСЕ", None) or t.isValue("ALL", None) or t.isValue("TEAM", None)): nams += 1 continue pit = PersonItemToken.tryAttach( t, None, PersonItemToken.ParseAttr.NO, None) if (pit is not None): nams += 1 t = pit.end_token continue oth += 1 if ((oth) > 3): if (ok > 0 and last_comma is not None): res.end_token = last_comma oth = 0 break if ((oth < 3) and ok > 0): res.typ = MailLine.Types.HELLO if (res.typ == MailLine.Types.UNDEFINED): ok_words = 0 if (t0.isValue("HAVE", None)): pass t = t0 first_pass3029 = True while True: if first_pass3029: first_pass3029 = False else: t = t.next0_ if (not (t is not None and t.end_char <= res.end_char)): break if (not ((isinstance(t, TextToken)))): continue if (t.isChar('<')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): t = br.end_token continue if (not t.is_letters or t.is_table_control_char): continue tok = MailLine.M_REGARD_WORDS.tryParse(t, TerminParseAttr.NO) if (tok is not None): ok_words += 1 while t is not None and t.end_char <= tok.end_char: t.tag = (tok.termin) t = t.next0_ t = tok.end_token if ((isinstance(t.next0_, TextToken)) and t.next0_.morph.case_.is_genitive): t = t.next0_ first_pass3030 = True while True: if first_pass3030: first_pass3030 = False else: t = t.next0_ if (not (t.end_char <= res.end_char)): break if (t.morph.class0_.is_conjunction): continue npt1 = NounPhraseHelper.tryParse( t, NounPhraseParseAttr.NO, 0) if (npt1 is None): break if (not npt1.morph.case_.is_genitive): break while t.end_char < npt1.end_char: t.tag = (t) t = t.next0_ t.tag = (t) continue if ((t.morph.class0_.is_preposition or t.morph.class0_.is_conjunction or t.morph.class0_.is_misc) or t.isValue("C", None)): continue if ((ok_words > 0 and t.previous is not None and t.previous.is_comma) and t.previous.begin_char > t0.begin_char and not t.chars.is_all_lower): res.end_token = t.previous break npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is None): if ((res.end_char - t.end_char) > 10): ok_words = 0 break tok = MailLine.M_REGARD_WORDS.tryParse(npt.end_token, TerminParseAttr.NO) if (tok is not None and (isinstance(npt.end_token, TextToken))): term = (npt.end_token).term if (term == "ДЕЛ"): tok = (None) if (tok is None): if (npt.noun.isValue("НАДЕЖДА", None)): t.tag = (t) elif (ok_words > 0 and t.isValue("NICE", None) and ((res.end_char - npt.end_char) < 13)): t.tag = (t) else: ok_words = 0 break ok_words += 1 while t is not None and t.end_char <= tok.end_char: t.tag = (tok.termin) t = t.next0_ t = tok.end_token if (ok_words > 0): res.typ = MailLine.Types.BESTREGARDS if (res.typ == MailLine.Types.UNDEFINED): t = t0 while t is not None and (t.end_char < res.end_char): if (not ((isinstance(t, TextToken)))): break elif (not t.is_hiphen and t.chars.is_letter): break t = t.next0_ if (t is not None): if (t != t0): pass if (((t.isValue("ПЕРЕСЫЛАЕМОЕ", None) or t.isValue("ПЕРЕАДРЕСОВАННОЕ", None))) and t.next0_ is not None and t.next0_.isValue("СООБЩЕНИЕ", None)): res.typ = MailLine.Types.FROM res.must_be_first_line = True elif ((t.isValue("НАЧАЛО", None) and t.next0_ is not None and ((t.next0_.isValue("ПЕРЕСЫЛАЕМОЕ", None) or t.next0_.isValue("ПЕРЕАДРЕСОВАННОЕ", None)))) and t.next0_.next0_ is not None and t.next0_.next0_.isValue("СООБЩЕНИЕ", None)): res.typ = MailLine.Types.FROM res.must_be_first_line = True elif (t.isValue("ORIGINAL", None) and t.next0_ is not None and ((t.next0_.isValue("MESSAGE", None) or t.next0_.isValue("APPOINTMENT", None)))): res.typ = MailLine.Types.FROM res.must_be_first_line = True elif (t.isValue("ПЕРЕСЛАНО", None) and t.next0_ is not None and t.next0_.isValue("ПОЛЬЗОВАТЕЛЕМ", None)): res.typ = MailLine.Types.FROM res.must_be_first_line = True elif (((t.getReferent() is not None and t.getReferent().type_name == "DATE")) or ((t.isValue("IL", None) and t.next0_ is not None and t.next0_.isValue("GIORNO", None))) or ((t.isValue("ON", None) and (isinstance(t.next0_, ReferentToken)) and t.next0_.getReferent().type_name == "DATE"))): has_from = False has_date = t.getReferent() is not None and t.getReferent( ).type_name == "DATE" if (t.is_newline_after and (lev_ < 5)): res1 = MailLine.parse(t.next0_, lev_ + 1) if (res1 is not None and res1.typ == MailLine.Types.HELLO): res.typ = MailLine.Types.FROM next0__ = MailLine.parse(res.end_token.next0_, lev_ + 1) if (next0__ is not None): if (next0__.typ != MailLine.Types.UNDEFINED): next0__ = (None) tmax = res.end_char if (next0__ is not None): tmax = next0__.end_char br1 = None while t is not None and t.end_char <= tmax: if (t.isValue("ОТ", None) or t.isValue("FROM", None)): has_from = True elif ( t.getReferent() is not None and ((t.getReferent().type_name == "URI" or (isinstance(t.getReferent(), PersonReferent))))): if (t.getReferent().type_name == "URI" and has_date): if (br1 is not None): has_from = True next0__ = (None) if (t.previous.isChar('<') and t.next0_ is not None and t.next0_.isChar('>')): t = t.next0_ if (t.next0_ is not None and t.next0_.isChar(':')): t = t.next0_ if (t.is_newline_after): has_from = True next0__ = (None) t = t.next0_ while t is not None and t.end_char <= res.end_char: if (t.isValue("HA", None) and t.next0_ is not None and t.next0_.isValue("SCRITTO", None)): has_from = True break elif (((t.isValue("НАПИСАТЬ", None) or t.isValue("WROTE", None))) and ((res.end_char - t.end_char) < 10)): has_from = True break t = t.next0_ if (has_from): res.typ = MailLine.Types.FROM if (next0__ is not None and t.end_char >= next0__.begin_char): res.end_token = next0__.end_token break elif (br1 is None and not t.isChar('<') and BracketHelper.canBeStartOfSequence( t, True, False)): br1 = BracketHelper.tryParse( t, BracketParseAttr.NO, 100) if (br1 is not None): t = br1.end_token t = t.next0_ else: has_uri = False while t is not None and (t.end_char < res.end_char): if (t.getReferent() is not None and ((t.getReferent().type_name == "URI" or (isinstance(t.getReferent(), PersonReferent))))): has_uri = True elif (t.isValue("ПИСАТЬ", None) and has_uri): if (t.next0_ is not None and t.next0_.isChar('(')): if (has_uri): res.typ = MailLine.Types.FROM break t = t.next0_ return res
def tryAttach(t: 'Token', prev: 'OrgItemNameToken', ext_onto: bool, first: bool) -> 'OrgItemNameToken': if (t is None): return None if (t.isValue("ОРДЕНА", None) and t.next0_ is not None): npt = NounPhraseHelper.tryParse(t.next0_, NounPhraseParseAttr.NO, 0) if (npt is not None): t1 = npt.end_token if (((t1.isValue("ЗНАК", None) or t1.isValue("ДРУЖБА", None))) and (t1.whitespaces_after_count < 2)): npt = NounPhraseHelper.tryParse(t1.next0_, NounPhraseParseAttr.NO, 0) if (npt is not None): t1 = npt.end_token return OrgItemNameToken._new1684(t, t1, True) if (t.next0_.getMorphClassInDictionary().is_proper_surname): return OrgItemNameToken._new1684(t, t.next0_, True) ppp = t.kit.processReferent("PERSON", t.next0_) if (ppp is not None): return OrgItemNameToken._new1684(t, ppp.end_token, True) if ((t.whitespaces_after_count < 2) and BracketHelper.canBeStartOfSequence( t.next0_, True, False)): br = BracketHelper.tryParse(t.next0_, BracketParseAttr.NEARCLOSEBRACKET, 10) if (br is not None and (br.length_char < 40)): return OrgItemNameToken._new1684(t, br.end_token, True) if (first and t.chars.is_cyrillic_letter and t.morph.class0_.is_preposition): if (not t.isValue("ПО", None) and not t.isValue("ПРИ", None)): return None res = OrgItemNameToken.__TryAttach(t, prev, ext_onto) if (res is None): if (ext_onto): if (((isinstance(t.getReferent(), GeoReferent))) or (((isinstance(t, TextToken)) and not t.isChar(';')))): return OrgItemNameToken._new1688(t, t, t.getSourceText()) return None if (prev is None and not ext_onto): if (t.kit.ontology is not None): ad = Utils.asObjectOrNull( t.kit.ontology._getAnalyzerData( OrganizationAnalyzer.ANALYZER_NAME), OrganizationAnalyzer.OrgAnalyzerData) if (ad is not None): tok = ad.org_pure_names.tryParse(t, TerminParseAttr.NO) if (tok is not None and tok.end_char > res.end_char): res.end_token = tok.end_token if (prev is not None and not ext_onto): if ((prev.chars.is_all_lower and not res.chars.is_all_lower and not res.is_std_tail) and not res.is_std_name): if (prev.chars.is_latin_letter and res.chars.is_latin_letter): pass elif (OrgItemNameToken.__m_std_nouns.tryParse( res.begin_token, TerminParseAttr.NO) is not None): pass else: return None if ((res.end_token.next0_ is not None and not res.end_token.is_whitespace_after and res.end_token.next0_.is_hiphen) and not res.end_token.next0_.is_whitespace_after): tt = Utils.asObjectOrNull(res.end_token.next0_.next0_, TextToken) if (tt is not None): if (tt.chars == res.chars or tt.chars.is_all_upper): res.end_token = tt res.value = "{0}-{1}".format(res.value, tt.term) if ((res.end_token.next0_ is not None and res.end_token.next0_.is_and and res.end_token.whitespaces_after_count == 1) and res.end_token.next0_.whitespaces_after_count == 1): res1 = OrgItemNameToken.__TryAttach(res.end_token.next0_.next0_, prev, ext_onto) if (res1 is not None and res1.chars == res.chars and OrgItemTypeToken.tryAttach( res.end_token.next0_.next0_, False, None) is None): if (not ((res1.morph.case_) & res.morph.case_).is_undefined): res.end_token = res1.end_token res.value = "{0} {1} {2}".format( res.value, ("ТА" if res.kit.base_language.is_ua else "И"), res1.value) tt = res.begin_token while tt is not None and tt.end_char <= res.end_char: if (OrgItemNameToken.__m_std_nouns.tryParse( tt, TerminParseAttr.NO) is not None): res.std_org_name_nouns += 1 tt = tt.next0_ if (OrgItemNameToken.__m_std_nouns.tryParse(res.end_token, TerminParseAttr.NO) is not None): cou = 1 non = False et = res.end_token if (not OrgItemNameToken.__isNotTermNoun(res.end_token)): non = True br = False tt = res.end_token.next0_ first_pass3045 = True while True: if first_pass3045: first_pass3045 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.is_table_control_char): break if (tt.isChar('(')): if (not non): break br = True continue if (tt.isChar(')')): br = False et = tt break if (not ((isinstance(tt, TextToken)))): break if (tt.whitespaces_before_count > 1): if (tt.newlines_before_count > 1): break if (tt.chars != res.end_token.chars): break if (tt.morph.class0_.is_preposition or tt.is_comma_and): continue dd = tt.getMorphClassInDictionary() if (not dd.is_noun and not dd.is_adjective): break npt2 = NounPhraseHelper.tryParse(tt, NounPhraseParseAttr.NO, 0) if (npt2 is None): if (dd == MorphClass.ADJECTIVE): continue break if (OrgItemNameToken.__m_std_nouns.tryParse( npt2.end_token, TerminParseAttr.NO) is None): break if (npt2.end_token.chars != res.end_token.chars): break if ((npt2.end_token.isValue("УПРАВЛЕНИЕ", None) or npt2.end_token.isValue("ИНСТИТУТ", None) or npt2.end_token.isValue("УПРАВЛІННЯ", None)) or npt2.end_token.isValue("ІНСТИТУТ", None) or tt.previous.isValue("ПРИ", None)): rt = tt.kit.processReferent( OrganizationAnalyzer.ANALYZER_NAME, tt) if (rt is not None): break cou += 1 tt = npt2.end_token if (not OrgItemNameToken.__isNotTermNoun(tt)): non = True et = tt if (non and not br): res.std_org_name_nouns += cou res.end_token = et return res
def tryParse(t: 'Token', loc_onto: 'IntOntologyCollection') -> 'NamedItemToken': if (t is None): return None if (isinstance(t, ReferentToken)): r = t.getReferent() if ((r.type_name == "PERSON" or r.type_name == "PERSONPROPERTY" or (isinstance(r, GeoReferent))) or r.type_name == "ORGANIZATION"): return NamedItemToken._new1635(t, t, r, t.morph) return None typ = NamedItemToken.__m_types.tryParse(t, TerminParseAttr.NO) nam = NamedItemToken.__m_names.tryParse(t, TerminParseAttr.NO) if (typ is not None): if (not ((isinstance(t, TextToken)))): return None res = NamedItemToken._new1636(typ.begin_token, typ.end_token, typ.morph, typ.chars) res.kind = (Utils.valToEnum(typ.termin.tag, NamedEntityKind)) res.type_value = typ.termin.canonic_text if ((nam is not None and nam.end_token == typ.end_token and not t.chars.is_all_lower) and (Utils.valToEnum( nam.termin.tag, NamedEntityKind)) == res.kind): res.name_value = nam.termin.canonic_text res.is_wellknown = True return res if (nam is not None): if (nam.begin_token.chars.is_all_lower): return None res = NamedItemToken._new1636(nam.begin_token, nam.end_token, nam.morph, nam.chars) res.kind = (Utils.valToEnum(nam.termin.tag, NamedEntityKind)) res.name_value = nam.termin.canonic_text ok = True if (not t.is_whitespace_before and t.previous is not None): ok = False elif (not t.is_whitespace_after and t.next0_ is not None): if (t.next0_.isCharOf(",.;!?") and t.next0_.is_whitespace_after): pass else: ok = False if (ok): res.is_wellknown = True res.type_value = (Utils.asObjectOrNull(nam.termin.tag2, str)) return res adj = MiscLocationHelper.tryAttachNordWest(t) if (adj is not None): if (adj.morph.class0_.is_noun): if (adj.end_token.isValue("ВОСТОК", None)): if (adj.begin_token == adj.end_token): return None re = NamedItemToken._new1638(t, adj.end_token, adj.morph) re.kind = NamedEntityKind.LOCATION re.name_value = MiscHelper.getTextValue( t, adj.end_token, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE) re.is_wellknown = True return re return None if (adj.whitespaces_after_count > 2): return None if ((isinstance(adj.end_token.next0_, ReferentToken)) and (isinstance(adj.end_token.next0_.getReferent(), GeoReferent))): re = NamedItemToken._new1638(t, adj.end_token.next0_, adj.end_token.next0_.morph) re.kind = NamedEntityKind.LOCATION re.name_value = MiscHelper.getTextValue( t, adj.end_token.next0_, GetTextAttr.FIRSTNOUNGROUPTONOMINATIVE) re.is_wellknown = True re.ref = adj.end_token.next0_.getReferent() return re res = NamedItemToken.tryParse(adj.end_token.next0_, loc_onto) if (res is not None and res.kind == NamedEntityKind.LOCATION): s = adj.getNormalCaseText(MorphClass.ADJECTIVE, True, res.morph.gender, False) if (s is not None): if (res.name_value is None): res.name_value = s.upper() else: res.name_value = "{0} {1}".format( s.upper(), res.name_value) res.type_value = (None) res.begin_token = t res.chars = t.chars res.is_wellknown = True return res if (t.chars.is_capital_upper and not MiscHelper.canBeStartOfSentence(t)): npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is not None and len(npt.adjectives) > 0): test = NamedItemToken.tryParse(npt.noun.begin_token, loc_onto) if (test is not None and test.end_token == npt.end_token and test.type_value is not None): test.begin_token = t tmp = io.StringIO() for a in npt.adjectives: s = a.getNormalCaseText(MorphClass.ADJECTIVE, True, test.morph.gender, False) if (tmp.tell() > 0): print(' ', end="", file=tmp) print(s, end="", file=tmp) test.name_value = Utils.toStringStringIO(tmp) test.chars = t.chars if (test.kind == NamedEntityKind.LOCATION): test.is_wellknown = True return test if ((BracketHelper.isBracket(t, True) and t.next0_ is not None and t.next0_.chars.is_letter) and not t.next0_.chars.is_all_lower): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): res = NamedItemToken(t, br.end_token) res.is_in_bracket = True res.name_value = MiscHelper.getTextValue( t, br.end_token, GetTextAttr.NO) nam = NamedItemToken.__m_names.tryParse( t.next0_, TerminParseAttr.NO) if (nam is not None and nam.end_token == br.end_token.previous): res.kind = (Utils.valToEnum(nam.termin.tag, NamedEntityKind)) res.is_wellknown = True res.name_value = nam.termin.canonic_text return res if (((isinstance(t, TextToken)) and t.chars.is_letter and not t.chars.is_all_lower) and t.length_char > 2): res = NamedItemToken._new1638(t, t, t.morph) str0_ = (t).term if (str0_.endswith("О") or str0_.endswith("И") or str0_.endswith("Ы")): res.name_value = str0_ else: res.name_value = t.getNormalCaseText(None, False, MorphGender.UNDEFINED, False) res.chars = t.chars if (((not t.is_whitespace_after and t.next0_ is not None and t.next0_.is_hiphen) and (isinstance(t.next0_.next0_, TextToken)) and not t.next0_.next0_.is_whitespace_after) and t.chars.is_cyrillic_letter == t.next0_.next0_.chars.is_cyrillic_letter): res.end_token = t.next0_.next0_ t = res.end_token res.name_value = "{0}-{1}".format( res.name_value, t.getNormalCaseText(None, False, MorphGender.UNDEFINED, False)) return res return None
def process(self, kit: 'AnalysisKit') -> None: ad = kit.getAnalyzerData(self) is_lit_block = 0 refs_by_num = dict() rts = [] t = kit.first_token first_pass2754 = True while True: if first_pass2754: first_pass2754 = False else: t = t.next0_ if (not (t is not None)): break if (t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None and br.length_char > 70 and (br.length_char < 400)): if (br.is_newline_after or ((br.end_token.next0_ is not None and br.end_token.next0_.isCharOf(".;")))): rts = BookLinkAnalyzer.__tryParse( t.next0_, False, br.end_char) if (rts is not None and len(rts) >= 1): if (len(rts) > 1): rts[1].referent = ad.registerReferent( rts[1].referent) kit.embedToken(rts[1]) (rts[0].referent).book = Utils.asObjectOrNull( rts[1].referent, BookLinkReferent) if (rts[0].begin_char == rts[1].begin_char): rts[0].begin_token = rts[1] if (rts[0].end_char == rts[1].end_char): rts[0].end_token = rts[1] rts[0].begin_token = t rts[0].end_token = br.end_token (rts[0].referent).typ = BookLinkRefType.INLINE rts[0].referent = ad.registerReferent( rts[0].referent) kit.embedToken(rts[0]) t = (rts[0]) continue if (not t.is_newline_before): continue if (is_lit_block <= 0): tt = BookLinkToken.parseStartOfLitBlock(t) if (tt is not None): is_lit_block = 5 t = tt continue rts = BookLinkAnalyzer.__tryParse(t, is_lit_block > 0, 0) if (rts is None or (len(rts) < 1)): is_lit_block -= 1 if ((is_lit_block) < 0): is_lit_block = 0 continue is_lit_block += 1 if ((is_lit_block) > 5): is_lit_block = 5 if (len(rts) > 1): rts[1].referent = ad.registerReferent(rts[1].referent) kit.embedToken(rts[1]) (rts[0].referent).book = Utils.asObjectOrNull( rts[1].referent, BookLinkReferent) if (rts[0].begin_char == rts[1].begin_char): rts[0].begin_token = rts[1] if (rts[0].end_char == rts[1].end_char): rts[0].end_token = rts[1] re = Utils.asObjectOrNull(rts[0].referent, BookLinkRefReferent) re = (Utils.asObjectOrNull(ad.registerReferent(re), BookLinkRefReferent)) rts[0].referent = (re) kit.embedToken(rts[0]) t = (rts[0]) if (re.number is not None): li = [] wrapli385 = RefOutArgWrapper(None) inoutres386 = Utils.tryGetValue(refs_by_num, re.number, wrapli385) li = wrapli385.value if (not inoutres386): li = list() refs_by_num[re.number] = li li.append(re) t = kit.first_token first_pass2755 = True while True: if first_pass2755: first_pass2755 = False else: t = t.next0_ if (not (t is not None)): break if (not ((isinstance(t, TextToken)))): continue rt = BookLinkAnalyzer.__tryParseShortInline(t) if (rt is None): continue re = Utils.asObjectOrNull(rt.referent, BookLinkRefReferent) li = [] wrapli387 = RefOutArgWrapper(None) inoutres388 = Utils.tryGetValue(refs_by_num, Utils.ifNotNull(re.number, ""), wrapli387) li = wrapli387.value if (not inoutres388): continue i = 0 while i < len(li): if (t.begin_char < li[i].occurrence[0].begin_char): break i += 1 if (i >= len(li)): continue re.book = li[i].book if (re.pages is None): re.pages = li[i].pages re.typ = BookLinkRefType.INLINE re = (Utils.asObjectOrNull(ad.registerReferent(re), BookLinkRefReferent)) rt.referent = (re) kit.embedToken(rt) t = (rt)
def tryAttachOrg(t : 'Token', can_be_cyr : bool=False) -> 'ReferentToken': from pullenti.ner.org.internal.OrgItemNameToken import OrgItemNameToken if (t is None): return None br = False if (t.isChar('(') and t.next0_ is not None): t = t.next0_ br = True if (isinstance(t, NumberToken)): if ((t).typ == NumberSpellingType.WORDS and t.morph.class0_.is_adjective and t.chars.is_capital_upper): pass else: return None else: if (t.chars.is_all_lower): return None if ((t.length_char < 3) and not t.chars.is_letter): return None if (not t.chars.is_latin_letter): if (not can_be_cyr or not t.chars.is_cyrillic_letter): return None t0 = t t1 = t0 nam_wo = 0 tok = None geo_ = None add_typ = None first_pass3043 = True while True: if first_pass3043: first_pass3043 = False else: t = t.next0_ if (not (t is not None)): break if (t != t0 and t.whitespaces_before_count > 1): break if (t.isChar(')')): break if (t.isChar('(') and t.next0_ is not None): if ((isinstance(t.next0_.getReferent(), GeoReferent)) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')): geo_ = (Utils.asObjectOrNull(t.next0_.getReferent(), GeoReferent)) t = t.next0_.next0_ continue typ = OrgItemTypeToken.tryAttach(t.next0_, True, None) if ((typ is not None and typ.end_token.next0_ is not None and typ.end_token.next0_.isChar(')')) and typ.chars.is_latin_letter): add_typ = typ t = typ.end_token.next0_ continue if (((isinstance(t.next0_, TextToken)) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')) and t.next0_.chars.is_capital_upper): t = t.next0_.next0_ t1 = t continue break tok = OrgItemEngItem.tryAttach(t, can_be_cyr) if (tok is None and t.isCharOf(".,") and t.next0_ is not None): tok = OrgItemEngItem.tryAttach(t.next0_, can_be_cyr) if (tok is None and t.next0_.isCharOf(",.")): tok = OrgItemEngItem.tryAttach(t.next0_.next0_, can_be_cyr) if (tok is not None): if (tok.length_char == 1 and t0.chars.is_cyrillic_letter): return None break if (t.is_hiphen and not t.is_whitespace_after and not t.is_whitespace_before): continue if (t.isCharOf("&+") or t.is_and): continue if (t.isChar('.')): if (t.previous is not None and t.previous.length_char == 1): continue elif (MiscHelper.canBeStartOfSentence(t.next0_)): break if (not t.chars.is_latin_letter): if (not can_be_cyr or not t.chars.is_cyrillic_letter): break if (t.chars.is_all_lower): if (t.morph.class0_.is_preposition or t.morph.class0_.is_conjunction): continue if (br): continue break mc = t.getMorphClassInDictionary() if (mc.is_verb): if (t.next0_ is not None and t.next0_.morph.class0_.is_preposition): break if (t.next0_ is not None and t.next0_.isValue("OF", None)): break if (isinstance(t, TextToken)): nam_wo += 1 t1 = t if (tok is None): return None if (t0 == tok.begin_token): br2 = BracketHelper.tryParse(tok.end_token.next0_, BracketParseAttr.NO, 100) if (br2 is not None): org1 = OrganizationReferent() if (tok.short_value is not None): org1.addTypeStr(tok.short_value) org1.addTypeStr(tok.full_value) nam1 = MiscHelper.getTextValue(br2.begin_token, br2.end_token, GetTextAttr.NO) if (nam1 is not None): org1.addName(nam1, True, None) return ReferentToken(org1, t0, br2.end_token) return None org0_ = OrganizationReferent() te = tok.end_token if (tok.is_bank): t1 = tok.end_token if (tok.full_value == "company" and (tok.whitespaces_after_count < 3)): tok1 = OrgItemEngItem.tryAttach(tok.end_token.next0_, can_be_cyr) if (tok1 is not None): t1 = tok.end_token tok = tok1 te = tok.end_token if (tok.full_value == "company"): if (nam_wo == 0): return None nam = MiscHelper.getTextValue(t0, t1, GetTextAttr.IGNOREARTICLES) if (nam == "STOCK" and tok.full_value == "company"): return None alt_nam = None if (Utils.isNullOrEmpty(nam)): return None if (nam.find('(') > 0): i1 = nam.find('(') i2 = nam.find(')') if (i1 < i2): alt_nam = nam tai = None if ((i2 + 1) < len(nam)): tai = nam[i2:].strip() nam = nam[0:0+i1].strip() if (tai is not None): nam = "{0} {1}".format(nam, tai) if (tok.is_bank): org0_.addTypeStr(("bank" if tok.kit.base_language.is_en else "банк")) org0_.addProfile(OrgProfile.FINANCE) if ((t1.next0_ is not None and t1.next0_.isValue("OF", None) and t1.next0_.next0_ is not None) and t1.next0_.next0_.chars.is_latin_letter): nam0 = OrgItemNameToken.tryAttach(t1.next0_, None, False, False) if (nam0 is not None): te = nam0.end_token else: te = t1.next0_.next0_ nam = MiscHelper.getTextValue(t0, te, GetTextAttr.NO) if (isinstance(te.getReferent(), GeoReferent)): org0_._addGeoObject(Utils.asObjectOrNull(te.getReferent(), GeoReferent)) elif (t0 == t1): return None else: if (tok.short_value is not None): org0_.addTypeStr(tok.short_value) org0_.addTypeStr(tok.full_value) if (Utils.isNullOrEmpty(nam)): return None org0_.addName(nam, True, None) if (alt_nam is not None): org0_.addName(alt_nam, True, None) res = ReferentToken(org0_, t0, te) t = te while t.next0_ is not None: if (t.next0_.isCharOf(",.")): t = t.next0_ else: break if (t.whitespaces_after_count < 2): tok = OrgItemEngItem.tryAttach(t.next0_, can_be_cyr) if (tok is not None): if (tok.short_value is not None): org0_.addTypeStr(tok.short_value) org0_.addTypeStr(tok.full_value) res.end_token = tok.end_token if (geo_ is not None): org0_._addGeoObject(geo_) if (add_typ is not None): org0_.addType(add_typ, False) if (not br): return res t = res.end_token if (t.next0_ is None or t.next0_.isChar(')')): res.end_token = t.next0_ else: return None return res
def __tryParse(t: 'Token', is_in_lit: bool, max_char: int = 0) -> typing.List['ReferentToken']: if (t is None): return None is_bracket_regime = False if (t.previous is not None and t.previous.isChar('(')): is_bracket_regime = True blt = BookLinkToken.tryParse(t, 0) if (blt is None): blt = BookLinkToken.tryParseAuthor(t, FioTemplateType.UNDEFINED) if (blt is None and not is_bracket_regime): return None t0 = t coef = 0 is_electr_res = False decree = None regtyp = BookLinkAnalyzer.RegionTyp.UNDEFINED num = None spec_see = None book_prev = None if (is_bracket_regime): regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS elif (blt.typ == BookLinkTyp.PERSON): if (not is_in_lit): return None regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS elif (blt.typ == BookLinkTyp.NUMBER): num = blt.value t = blt.end_token.next0_ if (t is None or t.is_newline_before): return None if (not t.is_whitespace_before): if (isinstance(t, NumberToken)): n = (t).value if ((((n == "3" or n == "0")) and not t.is_whitespace_after and (isinstance(t.next0_, TextToken))) and t.next0_.chars.is_all_lower): pass else: return None elif (not ((isinstance(t, TextToken))) or t.chars.is_all_lower): r = t.getReferent() if (isinstance(r, PersonReferent)): pass elif (is_in_lit and r is not None and r.type_name == "DECREE"): pass else: return None first_pass2757 = True while True: if first_pass2757: first_pass2757 = False else: t = t.next0_ if (not (t is not None)): break if (isinstance(t, NumberToken)): break if (not ((isinstance(t, TextToken)))): break if (BracketHelper.canBeStartOfSequence(t, True, False)): break if (not t.chars.is_letter): continue bbb = BookLinkToken.tryParse(t, 0) if (bbb is not None): if (bbb.typ == BookLinkTyp.TAMZE): spec_see = bbb t = bbb.end_token.next0_ break if (bbb.typ == BookLinkTyp.SEE): t = bbb.end_token continue break if (spec_see is not None and spec_see.typ == BookLinkTyp.TAMZE): coef += 1 max0_ = 1000 tt = t0 while tt is not None and max0_ > 0: if (isinstance(tt.getReferent(), BookLinkRefReferent)): book_prev = (tt.getReferent()).book break tt = tt.previous max0_ -= 1 blt1 = BookLinkToken.tryParseAuthor(t, FioTemplateType.UNDEFINED) if (blt1 is not None and blt1.typ == BookLinkTyp.PERSON): regtyp = BookLinkAnalyzer.RegionTyp.AUTHORS else: ok = False tt = t first_pass2758 = True while True: if first_pass2758: first_pass2758 = False else: tt = (None if tt is None else tt.next0_) if (not (tt is not None)): break if (tt.is_newline_before): break if (is_in_lit and tt.getReferent() is not None and tt.getReferent().type_name == "DECREE"): ok = True decree = tt break bbb = BookLinkToken.tryParse(tt, 0) if (bbb is None): continue if (bbb.typ == BookLinkTyp.ELECTRONRES): is_electr_res = True ok = True break if (bbb.typ == BookLinkTyp.DELIMETER): tt = bbb.end_token.next0_ if (BookLinkToken.tryParseAuthor( tt, FioTemplateType.UNDEFINED) is not None): ok = True break bbb = BookLinkToken.tryParse(tt, 0) if (bbb is not None): if (bbb.typ == BookLinkTyp.EDITORS or bbb.typ == BookLinkTyp.TRANSLATE or bbb.typ == BookLinkTyp.SOSTAVITEL): ok = True break if (not ok and not is_in_lit): if (BookLinkToken.checkLinkBefore(t0, num)): pass else: return None regtyp = BookLinkAnalyzer.RegionTyp.NAME else: return None res = BookLinkReferent() corr_authors = list() t00 = t blt00 = None start_of_name = None prev_pers_templ = FioTemplateType.UNDEFINED if (regtyp == BookLinkAnalyzer.RegionTyp.AUTHORS): first_pass2759 = True while True: if first_pass2759: first_pass2759 = False else: t = t.next0_ if (not (t is not None)): break if (max_char > 0 and t.begin_char >= max_char): break if (t.isCharOf(".;") or t.is_comma_and): continue if (t.isChar('/')): break if ((t.isChar('(') and t.next0_ is not None and t.next0_.isValue("EDS", None)) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')): t = t.next0_.next0_.next0_ break blt = BookLinkToken.tryParseAuthor(t, prev_pers_templ) if (blt is None and t.previous is not None and t.previous.is_and): blt = BookLinkToken.tryParseAuthor( t.previous, FioTemplateType.UNDEFINED) if (blt is None): if ((isinstance(t.getReferent(), OrganizationReferent)) and blt00 is not None): bbb2 = BookLinkToken.tryParse(t.next0_, 0) if (bbb2 is not None): if (bbb2.typ == BookLinkTyp.YEAR): res.addSlot(BookLinkReferent.ATTR_AUTHOR, t.getReferent(), False, 0) res.year = int(bbb2.value) coef += .5 t = bbb2.end_token.next0_ break if (blt.typ == BookLinkTyp.PERSON): tt2 = blt.end_token.next0_ bbb2 = BookLinkToken.tryParse(tt2, 0) if (bbb2 is not None): if (bbb2.typ == BookLinkTyp.YEAR): res.year = int(bbb2.value) coef += .5 blt.end_token = bbb2.end_token blt00 = (None) if (blt00 is not None and ((blt00.end_token.next0_ == blt.begin_token or blt.begin_token.previous.isChar('.')))): tt11 = blt.end_token.next0_ nex = BookLinkToken.tryParse(tt11, 0) if (nex is not None and nex.typ == BookLinkTyp.ANDOTHERS): pass else: if (tt11 is None): break if (tt11.isChar('/') and tt11.next0_ is not None and tt11.next0_.isChar('/')): break if (tt11.isChar(':')): break if ((str(blt).find('.') < 0) and str(blt00).find('.') > 0): break if ((isinstance(tt11, TextToken)) and tt11.chars.is_all_lower): break if (tt11.isCharOf(",.;") and tt11.next0_ is not None): tt11 = tt11.next0_ nex = BookLinkToken.tryParse(tt11, 0) if (nex is not None and nex.typ != BookLinkTyp.PERSON and nex.typ != BookLinkTyp.ANDOTHERS): break elif ( (blt00 is not None and blt00.person_template != FioTemplateType.UNDEFINED and blt.person_template != blt00.person_template) and blt.person_template == FioTemplateType.NAMESURNAME): if (blt.end_token.next0_ is None or not blt.end_token.next0_.is_comma_and): break if (BookLinkToken.tryParseAuthor( blt.end_token.next0_.next0_, FioTemplateType.UNDEFINED) is not None): pass else: break if (blt00 is None and blt.person_template == FioTemplateType.NAMESURNAME): tt = blt.end_token.next0_ if (tt is not None and tt.is_hiphen): tt = tt.next0_ if (isinstance(tt, NumberToken)): break BookLinkAnalyzer.__addAuthor(res, blt) coef += 1 t = blt.end_token if (isinstance(t.getReferent(), PersonReferent)): corr_authors.append( Utils.asObjectOrNull(t, ReferentToken)) blt00 = blt prev_pers_templ = blt.person_template start_of_name = blt.start_of_name if ((start_of_name) is not None): t = t.next0_ break continue if (blt.typ == BookLinkTyp.ANDOTHERS): coef += .5 t = blt.end_token.next0_ res.authors_and_other = True break break if (t is None): return None if ((t.is_newline_before and t != t0 and num is None) and res.findSlot( BookLinkReferent.ATTR_AUTHOR, None, True) is None): return None if (start_of_name is None): if (t.chars.is_all_lower): coef -= (1) if (t.chars.is_latin_letter and not is_electr_res and num is None): if (res.getSlotValue(BookLinkReferent.ATTR_AUTHOR) is None): return None tn0 = t tn1 = None uri = None next_num = None wrapnn393 = RefOutArgWrapper(0) inoutres394 = Utils.tryParseInt(Utils.ifNotNull(num, ""), wrapnn393) nn = wrapnn393.value if (inoutres394): next_num = str((nn + 1)) br = (BracketHelper.tryParse( t, Utils.valToEnum( (BracketParseAttr.CANCONTAINSVERBS) | (BracketParseAttr.CANBEMANYLINES), BracketParseAttr), 100) if BracketHelper.canBeStartOfSequence(t, True, False) else None) if (br is not None): t = t.next0_ pages = None first_pass2760 = True while True: if first_pass2760: first_pass2760 = False else: t = t.next0_ if (not (t is not None)): break if (max_char > 0 and t.begin_char >= max_char): break if (br is not None and br.end_token == t): tn1 = t break tit = TitleItemToken.tryAttach(t) if (tit is not None): if ((tit.typ == TitleItemToken.Types.TYP and tn0 == t and br is None) and BracketHelper.canBeStartOfSequence( tit.end_token.next0_, True, False)): br = BracketHelper.tryParse(tit.end_token.next0_, BracketParseAttr.NO, 100) if (br is not None): coef += (1) if (num is not None): coef += 1 tn0 = br.begin_token tn1 = br.end_token res.typ = tit.value.lower() t = br.end_token.next0_ break if (t.is_newline_before and t != tn0): if (br is not None and (t.end_char < br.end_char)): pass elif (not MiscHelper.canBeStartOfSentence(t)): pass else: if (t.newlines_before_count > 1): break if ((isinstance(t, NumberToken)) and num is not None and (t).int_value is not None): if (num == str(((t).int_value - 1))): break elif (num is not None): pass else: nnn = NounPhraseHelper.tryParse( t.previous, Utils.valToEnum( ((NounPhraseParseAttr.PARSEPREPOSITION) | (NounPhraseParseAttr.PARSEADVERBS) | (NounPhraseParseAttr.PARSENUMERICASADJECTIVE)) | (NounPhraseParseAttr.MULTILINES), NounPhraseParseAttr), 0) if (nnn is not None and nnn.end_char >= t.end_char): pass else: break if (t.isCharOf(".;") and t.whitespaces_after_count > 0): tit = TitleItemToken.tryAttach(t.next0_) if ((tit) is not None): if (tit.typ == TitleItemToken.Types.TYP): break stop = True words = 0 notwords = 0 tt = t.next0_ first_pass2761 = True while True: if first_pass2761: first_pass2761 = False else: tt = tt.next0_ if (not (tt is not None)): break blt0 = BookLinkToken.tryParse(tt, 0) if (blt0 is None): if (tt.is_newline_before): break if ((isinstance(tt, TextToken)) and not tt.getMorphClassInDictionary().is_undefined ): words += 1 else: notwords += 1 if (words > 6 and words > (notwords * 4)): stop = False break continue if ((blt0.typ == BookLinkTyp.DELIMETER or blt0.typ == BookLinkTyp.TRANSLATE or blt0.typ == BookLinkTyp.TYPE) or blt0.typ == BookLinkTyp.GEO or blt0.typ == BookLinkTyp.PRESS): stop = False break if (br is not None and br.end_token.previous.end_char > t.end_char): stop = False if (stop): break if (t == decree): t = t.next0_ break blt = BookLinkToken.tryParse(t, 0) if (blt is None): tn1 = t continue if (blt.typ == BookLinkTyp.DELIMETER): break if (((blt.typ == BookLinkTyp.MISC or blt.typ == BookLinkTyp.TRANSLATE or blt.typ == BookLinkTyp.NAMETAIL) or blt.typ == BookLinkTyp.TYPE or blt.typ == BookLinkTyp.VOLUME) or blt.typ == BookLinkTyp.PAGERANGE or blt.typ == BookLinkTyp.PAGES): coef += 1 break if (blt.typ == BookLinkTyp.GEO or blt.typ == BookLinkTyp.PRESS): if (t.previous.is_hiphen or t.previous.isCharOf(".;") or blt.add_coef > 0): break if (blt.typ == BookLinkTyp.YEAR): if (t.previous is not None and t.previous.is_comma): break if (blt.typ == BookLinkTyp.ELECTRONRES): is_electr_res = True break if (blt.typ == BookLinkTyp.URL): if (t == tn0 or t.previous.isCharOf(":.")): is_electr_res = True break tn1 = t if (tn1 is None and start_of_name is None): if (is_electr_res): uri_re = BookLinkReferent() rt0 = ReferentToken(uri_re, t00, t) rts0 = list() bref0 = BookLinkRefReferent._new389(uri_re) if (num is not None): bref0.number = num rt01 = ReferentToken(bref0, t0, rt0.end_token) ok = False while t is not None: if (t.is_newline_before): break blt0 = BookLinkToken.tryParse(t, 0) if (blt0 is not None): if (isinstance(blt0.ref, UriReferent)): uri_re.addSlot( BookLinkReferent.ATTR_URL, Utils.asObjectOrNull(blt0.ref, UriReferent), False, 0) ok = True t = blt0.end_token rt0.end_token = rt01.end_token = t t = t.next0_ if (ok): rts0.append(rt01) rts0.append(rt0) return rts0 if (decree is not None and num is not None): rts0 = list() bref0 = BookLinkRefReferent._new389(decree.getReferent()) if (num is not None): bref0.number = num rt01 = ReferentToken(bref0, t0, decree) t = decree.next0_ while t is not None: if (t.is_newline_before): break if (isinstance(t, TextToken)): if ((t).is_pure_verb): return None rt01.end_token = t t = t.next0_ rts0.append(rt01) return rts0 if (book_prev is not None): tt = t while tt is not None and ((tt.isCharOf(",.") or tt.is_hiphen)): tt = tt.next0_ blt0 = BookLinkToken.tryParse(tt, 0) if (blt0 is not None and blt0.typ == BookLinkTyp.PAGERANGE): rts0 = list() bref0 = BookLinkRefReferent._new389(book_prev) if (num is not None): bref0.number = num bref0.pages = blt0.value rt00 = ReferentToken(bref0, t0, blt0.end_token) rts0.append(rt00) return rts0 return None if (br is not None and ((tn1 == br.end_token or tn1 == br.end_token.previous))): tn0 = tn0.next0_ tn1 = tn1.previous if (start_of_name is None): while tn0 is not None: if (tn0.isCharOf(":,~")): tn0 = tn0.next0_ else: break while tn1 is not None and tn1.begin_char > tn0.begin_char: if (tn1.isCharOf(".;,:(~") or tn1.is_hiphen or tn1.isValue("РЕД", None)): pass else: break tn1 = tn1.previous nam = MiscHelper.getTextValue( tn0, tn1, Utils.valToEnum( (GetTextAttr.KEEPQUOTES) | (GetTextAttr.KEEPREGISTER), GetTextAttr)) if (start_of_name is not None): if (nam is None or (len(nam) < 3)): nam = start_of_name else: nam = "{0}{1}{2}".format( start_of_name, (" " if tn0.is_whitespace_before else ""), nam) if (nam is None): return None res.name = nam if (num is None and not is_in_lit): if (len(nam) < 20): return None coef -= (2) if (len(nam) > 500): coef -= (math.floor(len(nam) / 500)) if (is_bracket_regime): coef -= 1 if (len(nam) > 200): if (num is None): return None if (res.findSlot(BookLinkReferent.ATTR_AUTHOR, None, True) is None and not BookLinkToken.checkLinkBefore(t0, num)): return None en = 0 ru = 0 ua = 0 cha = 0 nocha = 0 chalen = 0 lt0 = tn0 lt1 = tn1 if (tn1 is None): if (t is None): return None lt0 = t0 lt1 = t tn1 = t.previous tt = lt0 while tt is not None and tt.end_char <= lt1.end_char: if ((isinstance(tt, TextToken)) and tt.chars.is_letter): if (tt.chars.is_latin_letter): en += 1 elif (tt.morph.language.is_ua): ua += 1 elif (tt.morph.language.is_ru): ru += 1 if (tt.length_char > 2): cha += 1 chalen += tt.length_char elif (not ((isinstance(tt, ReferentToken)))): nocha += 1 tt = tt.next0_ if (ru > (ua + en)): res.lang = "RU" elif (ua > (ru + en)): res.lang = "UA" elif (en > (ru + ua)): res.lang = "EN" if (nocha > 3 and nocha > cha and start_of_name is None): if (nocha > (math.floor(chalen / 3))): coef -= (2) if (res.lang == "EN"): tt = tn0.next0_ first_pass2762 = True while True: if first_pass2762: first_pass2762 = False else: tt = tt.next0_ if (not (tt is not None and (tt.end_char < tn1.end_char))): break if (tt.is_comma and tt.next0_ is not None and ((not tt.next0_.chars.is_all_lower or (isinstance(tt.next0_, ReferentToken))))): if (tt.next0_.next0_ is not None and tt.next0_.next0_.is_comma_and): if (isinstance(tt.next0_, ReferentToken)): pass else: continue nam = MiscHelper.getTextValue( tn0, tt.previous, Utils.valToEnum((GetTextAttr.KEEPQUOTES) | (GetTextAttr.KEEPREGISTER), GetTextAttr)) if (nam is not None and len(nam) > 15): res.name = nam break rt = ReferentToken(res, t00, tn1) authors = True edits = False br = (None) first_pass2763 = True while True: if first_pass2763: first_pass2763 = False else: t = t.next0_ if (not (t is not None)): break if (max_char > 0 and t.begin_char >= max_char): break if (BracketHelper.canBeStartOfSequence(t, False, False)): br = BracketHelper.tryParse(t, BracketParseAttr.CANBEMANYLINES, 100) if (br is not None and br.length_char > 300): br = (None) blt = BookLinkToken.tryParse(t, 0) if (t.is_newline_before and not t.isChar('/') and not t.previous.isChar('/')): if (blt is not None and blt.typ == BookLinkTyp.NUMBER): break if (t.previous.isCharOf(":")): pass elif (blt is not None and (( ((blt.typ == BookLinkTyp.DELIMETER or blt.typ == BookLinkTyp.PAGERANGE or blt.typ == BookLinkTyp.PAGES) or blt.typ == BookLinkTyp.GEO or blt.typ == BookLinkTyp.PRESS) or blt.typ == BookLinkTyp.N))): pass elif (num is not None and BookLinkToken.tryParseAuthor( t, FioTemplateType.UNDEFINED) is not None): pass elif (num is not None and blt is not None and blt.typ != BookLinkTyp.NUMBER): pass elif (br is not None and (t.end_char < br.end_char) and t.begin_char > br.begin_char): pass else: ok = False mmm = 50 tt = t.next0_ while tt is not None and mmm > 0: if (tt.is_newline_before): blt2 = BookLinkToken.tryParse(tt, 0) if (blt2 is not None and blt2.typ == BookLinkTyp.NUMBER and blt2.value == next_num): ok = True break if (blt2 is not None): if (blt2.typ == BookLinkTyp.PAGES or blt2.typ == BookLinkTyp.GEO or blt2.typ == BookLinkTyp.PRESS): ok = True break tt = tt.next0_ mmm -= 1 if (not ok): npt = NounPhraseHelper.tryParse( t.previous, Utils.valToEnum( ((NounPhraseParseAttr.MULTILINES) | (NounPhraseParseAttr.PARSEADVERBS) | (NounPhraseParseAttr.PARSEPREPOSITION)) | (NounPhraseParseAttr.PARSEVERBS) | (NounPhraseParseAttr.PARSEPRONOUNS), NounPhraseParseAttr), 0) if (npt is not None and npt.end_char >= t.end_char): ok = True if (not ok): break rt.end_token = t if (blt is not None): rt.end_token = blt.end_token if (t.isCharOf(".,") or t.is_hiphen): continue if (t.isValue("С", None)): pass if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and blt is not None and blt.typ == BookLinkTyp.EDITORS): edits = True t = blt.end_token coef += 1 continue if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and blt is not None and blt.typ == BookLinkTyp.SOSTAVITEL): edits = False t = blt.end_token coef += 1 continue if (regtyp == BookLinkAnalyzer.RegionTyp.FIRST and authors): blt2 = BookLinkToken.tryParseAuthor(t, prev_pers_templ) if (blt2 is not None and blt2.typ == BookLinkTyp.PERSON): prev_pers_templ = blt2.person_template if (not edits): BookLinkAnalyzer.__addAuthor(res, blt2) coef += 1 t = blt2.end_token continue if (blt2 is not None and blt2.typ == BookLinkTyp.ANDOTHERS): if (not edits): res.authors_and_other = True coef += 1 t = blt2.end_token continue authors = False if (blt is None): continue if (blt.typ == BookLinkTyp.ELECTRONRES or blt.typ == BookLinkTyp.URL): is_electr_res = True if (blt.typ == BookLinkTyp.ELECTRONRES): coef += 1.5 else: coef += .5 if (isinstance(blt.ref, UriReferent)): res.addSlot(BookLinkReferent.ATTR_URL, Utils.asObjectOrNull(blt.ref, UriReferent), False, 0) elif (blt.typ == BookLinkTyp.YEAR): if (res.year == 0): res.year = int(blt.value) coef += .5 elif (blt.typ == BookLinkTyp.DELIMETER): coef += 1 if (blt.length_char == 2): regtyp = BookLinkAnalyzer.RegionTyp.SECOND else: regtyp = BookLinkAnalyzer.RegionTyp.FIRST elif ( (((blt.typ == BookLinkTyp.MISC or blt.typ == BookLinkTyp.TYPE or blt.typ == BookLinkTyp.PAGES) or blt.typ == BookLinkTyp.NAMETAIL or blt.typ == BookLinkTyp.TRANSLATE) or blt.typ == BookLinkTyp.PRESS or blt.typ == BookLinkTyp.VOLUME) or blt.typ == BookLinkTyp.N): coef += 1 elif (blt.typ == BookLinkTyp.PAGERANGE): pages = blt coef += 1 if (is_bracket_regime and blt.end_token.next0_ is not None and blt.end_token.next0_.isChar(')')): coef += (2) if (res.name is not None and res.findSlot(BookLinkReferent.ATTR_AUTHOR, None, True) is not None): coef = (10) elif (blt.typ == BookLinkTyp.GEO and ((regtyp == BookLinkAnalyzer.RegionTyp.SECOND or regtyp == BookLinkAnalyzer.RegionTyp.FIRST))): coef += 1 elif (blt.typ == BookLinkTyp.GEO and t.previous is not None and t.previous.isChar('.')): coef += 1 elif (blt.typ == BookLinkTyp.ANDOTHERS): coef += 1 if (authors): res.authors_and_other = True coef += blt.add_coef t = blt.end_token if ((coef < 2.5) and num is not None): if (BookLinkToken.checkLinkBefore(t0, num)): coef += (2) elif (BookLinkToken.checkLinkAfter(rt.end_token, num)): coef += (1) if (rt.length_char > 500): return None if (is_in_lit): coef += 1 if (coef < 2.5): if (is_electr_res and uri is not None): pass elif (coef >= 2 and is_in_lit): pass else: return None for rr in corr_authors: pits0 = PersonItemToken.tryAttachList( rr.begin_token, None, PersonItemToken.ParseAttr.CANINITIALBEDIGIT, 10) if (pits0 is None or (len(pits0) < 2)): continue if (pits0[0].typ == PersonItemToken.ItemType.VALUE): exi = False for i in range(len(rr.referent.slots) - 1, -1, -1): s = rr.referent.slots[i] if (s.type_name == PersonReferent.ATTR_LASTNAME): ln = Utils.asObjectOrNull(s.value, str) if (ln is None): continue if (ln == pits0[0].value): exi = True continue if (ln.find('-') > 0): ln = ln[0:0 + ln.find('-')] if (pits0[0].begin_token.isValue(ln, None)): del rr.referent.slots[i] if (not exi): rr.referent.addSlot(PersonReferent.ATTR_LASTNAME, pits0[0].value, False, 0) rts = list() bref = BookLinkRefReferent._new389(res) if (num is not None): bref.number = num rt1 = ReferentToken(bref, t0, rt.end_token) if (pages is not None): if (pages.value is not None): bref.pages = pages.value rt.end_token = pages.begin_token.previous rts.append(rt1) rts.append(rt) return rts
def 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 tryAttachRequisites(t: 'Token', cur: 'InstrumentParticipant', other: 'InstrumentParticipant', cant_be_empty: bool = False) -> 'ReferentToken': if (t is None or cur is None): return None err = 0 spec_chars = 0 rt = None t0 = t is_in_tab_cell = False cou = 0 tt = t.next0_ while tt is not None and (cou < 300): if (tt.is_table_control_char): is_in_tab_cell = True break tt = tt.next0_ cou += 1 first_pass3018 = True while True: if first_pass3018: first_pass3018 = False else: t = t.next0_ if (not (t is not None)): break if (t.begin_char == 8923): pass if (t.is_table_control_char): if (t != t0): if (rt is not None): rt.end_token = t.previous elif (not cant_be_empty): rt = ReferentToken(cur, t0, t.previous) break else: continue if ((t.isCharOf(":.") or t.isValue("М", None) or t.isValue("M", None)) or t.isValue("П", None)): if (rt is not None): rt.end_token = t continue pp = ParticipantToken.tryAttachToExist(t, cur, other) if (pp is not None): if (pp.referent != cur): break if (rt is None): rt = ReferentToken(cur, t, t) rt.end_token = pp.end_token err = 0 continue if (t.is_newline_before): iii = InstrToken.parse(t, 0, None) if (iii is not None): if (iii.typ == ILTypes.APPENDIX): break if (t.whitespaces_before_count > 25 and not is_in_tab_cell): if (t != t0): if (t.previous is not None and t.previous.isCharOf(",;")): pass elif (t.newlines_before_count > 1): break if ((isinstance(t.getReferent(), PersonReferent)) or (isinstance(t.getReferent(), OrganizationReferent))): if (not cur._containsRef(t.getReferent())): break if ((t.isCharOf(";:,.") or t.is_hiphen or t.morph.class0_.is_preposition) or t.morph.class0_.is_conjunction): continue if (t.isCharOf("_/\\")): spec_chars += 1 if ((spec_chars) > 10 and rt is None): rt = ReferentToken(cur, t0, t) if (rt is not None): rt.end_token = t continue if (t.is_newline_before and (isinstance(t, NumberToken))): break if (t.isValue("ОФИС", None)): if (BracketHelper.canBeStartOfSequence(t.next0_, True, False)): br = BracketHelper.tryParse(t.next0_, BracketParseAttr.NO, 100) if (br is not None): t = br.end_token continue if ((isinstance(t.next0_, TextToken)) and not t.next0_.chars.is_all_lower): t = t.next0_ continue r = t.getReferent() if ((((isinstance(r, PersonReferent)) or (isinstance(r, AddressReferent)) or (isinstance(r, UriReferent))) or (isinstance(r, OrganizationReferent)) or (isinstance(r, PhoneReferent))) or (isinstance(r, PersonIdentityReferent)) or (isinstance(r, BankDataReferent))): if (other is not None and other.findSlot(None, r, True) is not None): if (not ((isinstance(r, UriReferent)))): break if (rt is None): rt = ReferentToken(cur, t, t) if (cur.findSlot(InstrumentParticipant.ATTR_DELEGATE, r, True) is not None): pass else: cur.addSlot(InstrumentParticipant.ATTR_REF, r, False, 0) rt.end_token = t err = 0 else: if ((isinstance(t, TextToken)) and t.length_char > 1): err += 1 if (is_in_tab_cell and rt is not None): if (err > 300): break elif (err > 4): break return rt
def tryAttach(t : 'Token') -> 'TitleItemToken': tt = Utils.asObjectOrNull(t, TextToken) if (tt is not None): t1 = tt if (tt.term == "ТЕМА"): tit = TitleItemToken.tryAttach(tt.next0_) if (tit is not None and tit.typ == TitleItemToken.Types.TYP): t1 = tit.end_token if (t1.next0_ is not None and t1.next0_.isChar(':')): t1 = t1.next0_ return TitleItemToken._new2501(t, t1, TitleItemToken.Types.TYPANDTHEME, tit.value) if (tt.next0_ is not None and tt.next0_.isChar(':')): t1 = tt.next0_ return TitleItemToken(tt, t1, TitleItemToken.Types.THEME) if (tt.term == "ПО" or tt.term == "НА"): if (tt.next0_ is not None and tt.next0_.isValue("ТЕМА", None)): t1 = tt.next0_ if (t1.next0_ is not None and t1.next0_.isChar(':')): t1 = t1.next0_ return TitleItemToken(tt, t1, TitleItemToken.Types.THEME) if (tt.term == "ПЕРЕВОД" or tt.term == "ПЕР"): tt2 = tt.next0_ if (tt2 is not None and tt2.isChar('.')): tt2 = tt2.next0_ if (isinstance(tt2, TextToken)): if ((tt2).term == "C" or (tt2).term == "С"): tt2 = tt2.next0_ if (isinstance(tt2, TextToken)): return TitleItemToken(t, tt2, TitleItemToken.Types.TRANSLATE) if (tt.term == "СЕКЦИЯ" or tt.term == "SECTION" or tt.term == "СЕКЦІЯ"): t1 = tt.next0_ if (t1 is not None and t1.isChar(':')): t1 = t1.next0_ br = BracketHelper.tryParse(t1, BracketParseAttr.NO, 100) if (br is not None): t1 = br.end_token elif (t1 != tt.next0_): while t1 is not None: if (t1.is_newline_after): break t1 = t1.next0_ if (t1 is None): return None if (t1 != tt.next0_): return TitleItemToken(tt, t1, TitleItemToken.Types.DUST) t1 = (None) if (tt.isValue("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")): t1 = tt.next0_ elif (tt.morph.class0_.is_preposition and tt.next0_ is not None and tt.next0_.isValue("СПЕЦИАЛЬНОСТЬ", "СПЕЦІАЛЬНІСТЬ")): t1 = tt.next0_.next0_ elif (tt.isChar('/') and tt.is_newline_before): t1 = tt.next0_ if (t1 is not None): if (t1.isCharOf(":") or t1.is_hiphen): t1 = t1.next0_ spec = TitleItemToken.__tryAttachSpeciality(t1, True) if (spec is not None): spec.begin_token = t return spec sss = TitleItemToken.__tryAttachSpeciality(t, False) if (sss is not None): return sss if (isinstance(t, ReferentToken)): return None npt = NounPhraseHelper.tryParse(t, NounPhraseParseAttr.NO, 0) if (npt is not None): s = npt.getNormalCaseText(None, False, MorphGender.UNDEFINED, False) tok = TitleItemToken.M_TERMINS.tryParse(npt.end_token, TerminParseAttr.NO) if (tok is not None): ty = Utils.valToEnum(tok.termin.tag, TitleItemToken.Types) if (ty == TitleItemToken.Types.TYP): tit = TitleItemToken.tryAttach(tok.end_token.next0_) if (tit is not None and tit.typ == TitleItemToken.Types.THEME): return TitleItemToken._new2501(npt.begin_token, tit.end_token, TitleItemToken.Types.TYPANDTHEME, s) if (s == "РАБОТА" or s == "РОБОТА" or s == "ПРОЕКТ"): return None t1 = tok.end_token if (s == "ДИССЕРТАЦИЯ" or s == "ДИСЕРТАЦІЯ"): err = 0 ttt = t1.next0_ first_pass3125 = True while True: if first_pass3125: first_pass3125 = False else: ttt = ttt.next0_ if (not (ttt is not None)): break if (ttt.morph.class0_.is_preposition): continue if (ttt.isValue("СОИСКАНИЕ", "")): continue npt1 = NounPhraseHelper.tryParse(ttt, NounPhraseParseAttr.NO, 0) if (npt1 is not None and npt1.noun.isValue("СТЕПЕНЬ", "СТУПІНЬ")): ttt = npt1.end_token t1 = ttt continue rt = t1.kit.processReferent("PERSON", ttt) if (rt is not None and (isinstance(rt.referent, PersonPropertyReferent))): ppr = Utils.asObjectOrNull(rt.referent, PersonPropertyReferent) if (ppr.name == "доктор наук"): t1 = rt.end_token s = "ДОКТОРСКАЯ ДИССЕРТАЦИЯ" break elif (ppr.name == "кандидат наук"): t1 = rt.end_token s = "КАНДИДАТСКАЯ ДИССЕРТАЦИЯ" break elif (ppr.name == "магистр"): t1 = rt.end_token s = "МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ" break if (ttt.isValue("ДОКТОР", None) or ttt.isValue("КАНДИДАТ", None) or ttt.isValue("МАГИСТР", "МАГІСТР")): t1 = ttt npt1 = NounPhraseHelper.tryParse(ttt.next0_, NounPhraseParseAttr.NO, 0) if (npt1 is not None and npt1.end_token.isValue("НАУК", None)): t1 = npt1.end_token s = ("МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ" if ttt.isValue("МАГИСТР", "МАГІСТР") else ("ДОКТОРСКАЯ ДИССЕРТАЦИЯ" if ttt.isValue("ДОКТОР", None) else "КАНДИДАТСКАЯ ДИССЕРТАЦИЯ")) break err += 1 if ((err) > 3): break if (t1.next0_ is not None and t1.next0_.isChar('.')): t1 = t1.next0_ if (s.endswith("ОТЧЕТ") and t1.next0_ is not None and t1.next0_.isValue("О", None)): npt1 = NounPhraseHelper.tryParse(t1.next0_, NounPhraseParseAttr.PARSEPREPOSITION, 0) if (npt1 is not None and npt1.morph.case_.is_prepositional): t1 = npt1.end_token return TitleItemToken._new2501(npt.begin_token, t1, ty, s) tok1 = TitleItemToken.M_TERMINS.tryParse(t, TerminParseAttr.NO) if (tok1 is not None): t1 = tok1.end_token re = TitleItemToken(tok1.begin_token, t1, Utils.valToEnum(tok1.termin.tag, TitleItemToken.Types)) return re if (BracketHelper.canBeStartOfSequence(t, False, False)): tok1 = TitleItemToken.M_TERMINS.tryParse(t.next0_, TerminParseAttr.NO) if (tok1 is not None and BracketHelper.canBeEndOfSequence(tok1.end_token.next0_, False, None, False)): t1 = tok1.end_token.next0_ return TitleItemToken(tok1.begin_token, t1, Utils.valToEnum(tok1.termin.tag, TitleItemToken.Types)) return None
def tryAttach(t: 'Token', p1: 'InstrumentParticipant' = None, p2: 'InstrumentParticipant' = None, is_contract: bool = False) -> 'ParticipantToken': if (t is None): return None tt = t br = False if (p1 is None and p2 is None and is_contract): r1 = t.getReferent() if ((r1 is not None and t.next0_ is not None and t.next0_.is_comma_and) and (isinstance(t.next0_.next0_, ReferentToken))): r2 = t.next0_.next0_.getReferent() if (r1.type_name == r2.type_name): ttt = t.next0_.next0_.next0_ refs = list() refs.append(r1) refs.append(r2) first_pass3014 = True while True: if first_pass3014: first_pass3014 = False else: ttt = ttt.next0_ if (not (ttt is not None)): break if ((ttt.is_comma_and and ttt.next0_ is not None and ttt.next0_.getReferent() is not None) and ttt.next0_.getReferent().type_name == r1.type_name): ttt = ttt.next0_ if (not ttt.getReferent() in refs): refs.append(ttt.getReferent()) continue break first_pass3015 = True while True: if first_pass3015: first_pass3015 = False else: ttt = ttt.next0_ if (not (ttt is not None)): break if (ttt.is_comma or ttt.morph.class0_.is_preposition): continue if ((ttt.isValue("ИМЕНОВАТЬ", None) or ttt.isValue("ДАЛЬНЕЙШИЙ", None) or ttt.isValue("ДАЛЕЕ", None)) or ttt.isValue("ТЕКСТ", None)): continue if (ttt.isValue("ДОГОВАРИВАТЬСЯ", None)): continue npt = NounPhraseHelper.tryParse( ttt, NounPhraseParseAttr.NO, 0) if (npt is not None and npt.noun.isValue("СТОРОНА", None) and npt.morph.number != MorphNumber.SINGULAR): re = ParticipantToken._new1467( t, npt.end_token, ParticipantToken.Kinds.NAMEDASPARTS) re.parts = refs return re break if ((isinstance(r1, OrganizationReferent)) or (isinstance(r1, PersonReferent))): has_br = False has_named = False if (isinstance(r1, PersonReferent)): if (t.previous is not None and t.previous.isValue("ЛИЦО", None)): return None elif (t.previous is not None and ((t.previous.isValue("ВЫДАВАТЬ", None) or t.previous.isValue("ВЫДАТЬ", None)))): return None ttt = (t).begin_token while ttt is not None and (ttt.end_char < t.end_char): if (ttt.isChar('(')): has_br = True elif ((ttt.isValue("ИМЕНОВАТЬ", None) or ttt.isValue( "ДАЛЬНЕЙШИЙ", None) or ttt.isValue("ДАЛЕЕ", None)) or ttt.isValue("ТЕКСТ", None)): has_named = True elif ((ttt.is_comma or ttt.morph.class0_.is_preposition or ttt.is_hiphen) or ttt.isChar(':')): pass elif (isinstance(ttt, ReferentToken)): pass elif (has_br or has_named): npt = NounPhraseHelper.tryParse( ttt, NounPhraseParseAttr.REFERENTCANBENOUN, 0) if (npt is None): break if (has_br): if (npt.end_token.next0_ is None or not npt.end_token.next0_.isChar(')')): break if (not has_named): if (ParticipantToken.M_ONTOLOGY.tryParse( ttt, TerminParseAttr.NO) is None): break re = ParticipantToken._new1467( t, t, ParticipantToken.Kinds.NAMEDAS) re.typ = npt.getNormalCaseText(None, True, MorphGender.UNDEFINED, False) re.parts = list() re.parts.append(r1) return re ttt = ttt.next0_ has_br = False has_named = False end_side = None brr = None add_refs = None ttt = t.next0_ first_pass3016 = True while True: if first_pass3016: first_pass3016 = False else: ttt = ttt.next0_ if (not (ttt is not None)): break if ((isinstance(ttt, NumberToken)) and (isinstance(ttt.next0_, TextToken)) and (ttt.next0_).term == "СТОРОНЫ"): ttt = ttt.next0_ end_side = ttt if (ttt.next0_ is not None and ttt.next0_.is_comma): ttt = ttt.next0_ if (ttt.next0_ is not None and ttt.next0_.is_and): break if (brr is not None and ttt.begin_char > brr.end_char): brr = (None) if (BracketHelper.canBeStartOfSequence(ttt, False, False)): brr = BracketHelper.tryParse(ttt, BracketParseAttr.NO, 100) if (brr is not None and (brr.length_char < 7) and ttt.isChar('(')): ttt = brr.end_token brr = (None) continue elif ((ttt.isValue("ИМЕНОВАТЬ", None) or ttt.isValue( "ДАЛЬНЕЙШИЙ", None) or ttt.isValue("ДАЛЕЕ", None)) or ttt.isValue("ТЕКСТ", None)): has_named = True elif ((ttt.is_comma or ttt.morph.class0_.is_preposition or ttt.is_hiphen) or ttt.isChar(':')): pass elif (brr is not None or has_named): if (BracketHelper.canBeStartOfSequence( ttt, True, False)): ttt = ttt.next0_ npt = NounPhraseHelper.tryParse( ttt, NounPhraseParseAttr.REFERENTCANBENOUN, 0) typ22 = None if (npt is not None): ttt = npt.end_token if (npt.end_token.isValue("ДОГОВОР", None)): continue else: ttok = None if (isinstance(ttt, MetaToken)): ttok = ParticipantToken.M_ONTOLOGY.tryParse( (ttt).begin_token, TerminParseAttr.NO) if (ttok is not None): typ22 = ttok.termin.canonic_text elif (has_named and ttt.morph.class0_.is_adjective): typ22 = ttt.getNormalCaseText( MorphClass.ADJECTIVE, False, MorphGender.UNDEFINED, False) elif (brr is not None): continue else: break if (BracketHelper.canBeEndOfSequence( ttt.next0_, True, None, False)): ttt = ttt.next0_ if (brr is not None): if (ttt.next0_ is None): ttt = brr.end_token continue ttt = ttt.next0_ if (not has_named and typ22 is None): if (ParticipantToken.M_ONTOLOGY.tryParse( npt.begin_token, TerminParseAttr.NO) is None): break re = ParticipantToken._new1467( t, ttt, ParticipantToken.Kinds.NAMEDAS) re.typ = (Utils.ifNotNull( typ22, npt.getNormalCaseText(None, True, MorphGender.UNDEFINED, False))) re.parts = list() re.parts.append(r1) return re elif ((ttt.isValue("ЗАРЕГИСТРИРОВАННЫЙ", None) or ttt.isValue("КАЧЕСТВО", None) or ttt.isValue("ПРОЖИВАЮЩИЙ", None)) or ttt.isValue("ЗАРЕГ", None)): pass elif (ttt.getReferent() == r1): pass elif ( (isinstance(ttt.getReferent(), PersonIdentityReferent)) or (isinstance(ttt.getReferent(), AddressReferent))): if (add_refs is None): add_refs = list() add_refs.append(ttt.getReferent()) else: prr = ttt.kit.processReferent("PERSONPROPERTY", ttt) if (prr is not None): ttt = prr.end_token continue if (isinstance(ttt.getReferent(), GeoReferent)): continue npt = NounPhraseHelper.tryParse( ttt, NounPhraseParseAttr.NO, 0) if (npt is not None): if ((npt.noun.isValue("МЕСТО", None) or npt.noun.isValue("ЖИТЕЛЬСТВО", None) or npt.noun.isValue("ПРЕДПРИНИМАТЕЛЬ", None)) or npt.noun.isValue("ПОЛ", None) or npt.noun.isValue("РОЖДЕНИЕ", None)): ttt = npt.end_token continue if (ttt.is_newline_before): break if (ttt.length_char < 3): continue mc = ttt.getMorphClassInDictionary() if (mc.is_adverb or mc.is_adjective): continue if (ttt.chars.is_all_upper): continue break if (end_side is not None or ((add_refs is not None and t.previous is not None and t.previous.is_and))): re = ParticipantToken._new1467( t, Utils.ifNotNull(end_side, t), ParticipantToken.Kinds.NAMEDAS) re.typ = (None) re.parts = list() re.parts.append(r1) if (add_refs is not None): re.parts.extend(add_refs) return re too = ParticipantToken.M_ONTOLOGY.tryParse(t, TerminParseAttr.NO) if (too is not None): if ((isinstance(t.previous, TextToken)) and t.previous.isValue("ЛИЦО", None)): too = (None) if (too is not None and too.termin.tag is not None and too.termin.canonic_text != "СТОРОНА"): tt1 = too.end_token.next0_ if (tt1 is not None): if (tt1.is_hiphen or tt1.isChar(':')): tt1 = tt1.next0_ if (isinstance(tt1, ReferentToken)): r1 = tt1.getReferent() if ((isinstance(r1, PersonReferent)) or (isinstance(r1, OrganizationReferent))): re = ParticipantToken._new1467( t, tt1, ParticipantToken.Kinds.NAMEDAS) re.typ = too.termin.canonic_text re.parts = list() re.parts.append(r1) return re add_typ1 = (None if p1 is None else p1.typ) add_typ2 = (None if p2 is None else p2.typ) if (BracketHelper.canBeStartOfSequence(tt, False, False) and tt.next0_ is not None): br = True tt = tt.next0_ term1 = None term2 = None if (add_typ1 is not None and add_typ1.find(' ') > 0 and not add_typ1.startswith("СТОРОНА")): term1 = Termin(add_typ1) if (add_typ2 is not None and add_typ2.find(' ') > 0 and not add_typ2.startswith("СТОРОНА")): term2 = Termin(add_typ2) named = False typ_ = None t1 = None t0 = tt first_pass3017 = True while True: if first_pass3017: first_pass3017 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.morph.class0_.is_preposition and typ_ is not None): continue if (tt.isCharOf("(:)") or tt.is_hiphen): continue if (tt.is_table_control_char): break if (tt.is_newline_before and tt != t0): if (isinstance(tt, NumberToken)): break if ((isinstance(tt, TextToken)) and (isinstance(tt.previous, TextToken))): if (tt.previous.isValue((tt).term, None)): break if (BracketHelper.isBracket(tt, False)): continue tok = (ParticipantToken.M_ONTOLOGY.tryParse( tt, TerminParseAttr.NO) if ParticipantToken.M_ONTOLOGY is not None else None) if (tok is not None and (isinstance(tt.previous, TextToken))): if (tt.previous.isValue("ЛИЦО", None)): return None if (tok is None): if (add_typ1 is not None and ((MiscHelper.isNotMoreThanOneError(add_typ1, tt) or ((((isinstance(tt, MetaToken))) and (tt).begin_token.isValue(add_typ1, None)))))): if (typ_ is not None): if (not ParticipantToken.__isTypesEqual( add_typ1, typ_)): break typ_ = add_typ1 t1 = tt continue if (add_typ2 is not None and ((MiscHelper.isNotMoreThanOneError(add_typ2, tt) or ((((isinstance(tt, MetaToken))) and (tt).begin_token.isValue(add_typ2, None)))))): if (typ_ is not None): if (not ParticipantToken.__isTypesEqual( add_typ2, typ_)): break typ_ = add_typ2 t1 = tt continue if (tt.chars.is_letter): if (term1 is not None): tok1 = term1.tryParse(tt, TerminParseAttr.NO) if (tok1 is not None): if (typ_ is not None): if (not ParticipantToken.__isTypesEqual( add_typ1, typ_)): break typ_ = add_typ1 tt = tok1.end_token t1 = tt continue if (term2 is not None): tok2 = term2.tryParse(tt, TerminParseAttr.NO) if (tok2 is not None): if (typ_ is not None): if (not ParticipantToken.__isTypesEqual( add_typ2, typ_)): break typ_ = add_typ2 tt = tok2.end_token t1 = tt continue if (named and tt.getMorphClassInDictionary().is_noun): if (not tt.chars.is_all_lower or BracketHelper.isBracket(tt.previous, True)): if (DecreeToken.isKeyword(tt, False) is None): val = tt.getNormalCaseText( MorphClass.NOUN, True, MorphGender.UNDEFINED, False) if (typ_ is not None): if (not ParticipantToken.__isTypesEqual( typ_, val)): break typ_ = val t1 = tt continue if (named and typ_ is None and is_contract): if ((isinstance(tt, TextToken)) and tt.chars.is_cyrillic_letter and tt.chars.is_capital_upper): dc = tt.getMorphClassInDictionary() if (dc.is_undefined or dc.is_noun): dt = DecreeToken.tryAttach(tt, None, False) ok = True if (dt is not None): ok = False elif (tt.isValue("СТОРОНА", None)): ok = False if (ok): typ_ = (tt).getLemma() t1 = tt continue if (dc.is_adjective): npt = NounPhraseHelper.tryParse( tt, NounPhraseParseAttr.NO, 0) if (npt is not None and len(npt.adjectives) > 0 and npt.noun.getMorphClassInDictionary( ).is_noun): typ_ = npt.getNormalCaseText( None, True, MorphGender.UNDEFINED, False) t1 = npt.end_token continue if (tt == t): break if ((isinstance(tt, NumberToken)) or tt.isChar('.')): break if (tt.length_char < 4): if (typ_ is not None): continue break if (tok.termin.tag is None): named = True else: if (typ_ is not None): break if (tok.termin.canonic_text == "СТОРОНА"): tt1 = tt.next0_ if (tt1 is not None and tt1.is_hiphen): tt1 = tt1.next0_ if (not ((isinstance(tt1, NumberToken)))): break if (tt1.is_newline_before): break typ_ = "{0} {1}".format(tok.termin.canonic_text, (tt1).value) t1 = tt1 else: typ_ = tok.termin.canonic_text t1 = tok.end_token break tt = tok.end_token if (typ_ is None): return None if (not named and t1 != t and not typ_.startswith("СТОРОНА")): if (not ParticipantToken.__isTypesEqual(typ_, add_typ1) and not ParticipantToken.__isTypesEqual(typ_, add_typ2)): return None if (BracketHelper.canBeEndOfSequence(t1.next0_, False, None, False)): t1 = t1.next0_ if (not t.is_whitespace_before and BracketHelper.canBeStartOfSequence( t.previous, False, False)): t = t.previous elif (BracketHelper.canBeStartOfSequence(t, False, False) and BracketHelper.canBeEndOfSequence(t1.next0_, True, t, True)): t1 = t1.next0_ if (br and t1.next0_ is not None and BracketHelper.canBeEndOfSequence( t1.next0_, False, None, False)): t1 = t1.next0_ res = ParticipantToken._new1472( t, t1, (ParticipantToken.Kinds.NAMEDAS if named else ParticipantToken.Kinds.PURE), typ_) if (t.isChar(':')): res.begin_token = t.next0_ return res
def tryParseNumberWithPostfix(t: 'Token') -> 'NumberExToken': """ Выделение стандартных мер, типа: 10 кв.м. """ if (t is None): return None t0 = t is_dollar = None if (t.length_char == 1 and t.next0_ is not None): is_dollar = NumberHelper._isMoneyChar(t) if ((is_dollar) is not None): t = t.next0_ nt = Utils.asObjectOrNull(t, NumberToken) if (nt is None): if ((not ((isinstance(t.previous, NumberToken))) and t.isChar('(') and (isinstance(t.next0_, NumberToken))) and t.next0_.next0_ is not None and t.next0_.next0_.isChar(')')): toks1 = NumberExHelper._m_postfixes.tryParse( t.next0_.next0_.next0_, TerminParseAttr.NO) if (toks1 is not None and (Utils.valToEnum(toks1.termin.tag, NumberExType)) == NumberExType.MONEY): nt0 = Utils.asObjectOrNull(t.next0_, NumberToken) res = NumberExToken._new471(t, toks1.end_token, nt0.value, nt0.typ, NumberExType.MONEY, nt0.real_value, toks1.begin_token.morph) return NumberExHelper.__correctMoney( res, toks1.begin_token) tt = Utils.asObjectOrNull(t, TextToken) if (tt is None or not tt.morph.class0_.is_adjective): return None val = tt.term i = 4 first_pass2785 = True while True: if first_pass2785: first_pass2785 = False else: i += 1 if (not (i < (len(val) - 5))): break v = val[0:0 + i] li = NumberHelper._m_nums.tryAttachStr(v, tt.morph.language) if (li is None): continue vv = val[i:] lii = NumberExHelper._m_postfixes.tryAttachStr( vv, tt.morph.language) if (lii is not None and len(lii) > 0): re = NumberExToken._new472( t, t, str((li[0].tag)), NumberSpellingType.WORDS, Utils.valToEnum(lii[0].tag, NumberExType), t.morph) NumberExHelper.__correctExtTypes(re) return re break return None if (t.next0_ is None and is_dollar is None): return None f = nt.real_value t1 = nt.next0_ if (((t1 is not None and t1.isCharOf(",."))) or (((isinstance(t1, NumberToken)) and (t1.whitespaces_before_count < 3)))): tt11 = NumberHelper.tryParseRealNumber(nt, False) if (tt11 is not None): t1 = tt11.end_token.next0_ f = tt11.real_value if (t1 is None): if (is_dollar is None): return None elif ((t1.next0_ is not None and t1.next0_.isValue("С", "З") and t1.next0_.next0_ is not None) and t1.next0_.next0_.isValue("ПОЛОВИНА", None)): f += .5 t1 = t1.next0_.next0_ if (t1 is not None and t1.is_hiphen and t1.next0_ is not None): t1 = t1.next0_ det = False altf = f if (((isinstance(t1, NumberToken)) and t1.previous is not None and t1.previous.is_hiphen) and (t1).int_value == 0 and t1.length_char == 2): t1 = t1.next0_ if ((t1 is not None and t1.next0_ is not None and t1.isChar('(')) and (((isinstance(t1.next0_, NumberToken)) or t1.next0_.isValue("НОЛЬ", None))) and t1.next0_.next0_ is not None): nt1 = Utils.asObjectOrNull(t1.next0_, NumberToken) val = 0 if (nt1 is not None): val = nt1.real_value if (math.floor(f) == math.floor(val)): ttt = t1.next0_.next0_ if (ttt.isChar(')')): t1 = ttt.next0_ det = True if ((isinstance(t1, NumberToken)) and (t1).int_value is not None and (t1).int_value == 0): t1 = t1.next0_ elif ( ((((isinstance(ttt, NumberToken)) and ((ttt).real_value < 100) and ttt.next0_ is not None) and ttt.next0_.isChar('/') and ttt.next0_.next0_ is not None) and ttt.next0_.next0_.getSourceText() == "100" and ttt.next0_.next0_.next0_ is not None) and ttt.next0_.next0_.next0_.isChar(')')): rest = NumberExHelper.__getDecimalRest100(f) if ((ttt).int_value is not None and rest == (ttt).int_value): t1 = ttt.next0_.next0_.next0_.next0_ det = True elif ((ttt.isValue("ЦЕЛЫХ", None) and (isinstance(ttt.next0_, NumberToken)) and ttt.next0_.next0_ is not None) and ttt.next0_.next0_.next0_ is not None and ttt.next0_.next0_.next0_.isChar(')')): num2 = Utils.asObjectOrNull(ttt.next0_, NumberToken) altf = num2.real_value if (ttt.next0_.next0_.isValue("ДЕСЯТЫЙ", None)): altf /= (10) elif (ttt.next0_.next0_.isValue("СОТЫЙ", None)): altf /= (100) elif (ttt.next0_.next0_.isValue("ТЫСЯЧНЫЙ", None)): altf /= (1000) elif (ttt.next0_.next0_.isValue("ДЕСЯТИТЫСЯЧНЫЙ", None)): altf /= (10000) elif (ttt.next0_.next0_.isValue("СТОТЫСЯЧНЫЙ", None)): altf /= (100000) elif (ttt.next0_.next0_.isValue("МИЛЛИОННЫЙ", None)): altf /= (1000000) if (altf < 1): altf += val t1 = ttt.next0_.next0_.next0_.next0_ det = True else: toks1 = NumberExHelper._m_postfixes.tryParse( ttt, TerminParseAttr.NO) if (toks1 is not None): if ((Utils.valToEnum( toks1.termin.tag, NumberExType)) == NumberExType.MONEY): if (toks1.end_token.next0_ is not None and toks1.end_token.next0_.isChar(')')): res = NumberExToken._new473( t, toks1.end_token.next0_, nt.value, nt.typ, NumberExType.MONEY, f, altf, toks1.begin_token.morph) return NumberExHelper.__correctMoney( res, toks1.begin_token) res2 = NumberExHelper.tryParseNumberWithPostfix(t1.next0_) if (res2 is not None and res2.end_token.next0_ is not None and res2.end_token.next0_.isChar(')')): if (res2.int_value is not None): res2.begin_token = t res2.end_token = res2.end_token.next0_ res2.alt_real_value = res2.real_value res2.real_value = f NumberExHelper.__correctExtTypes(res2) if (res2.whitespaces_after_count < 2): toks2 = NumberExHelper._m_postfixes.tryParse( res2.end_token.next0_, TerminParseAttr.NO) if (toks2 is not None): if ((Utils.valToEnum( toks2.termin.tag, NumberExType) ) == NumberExType.MONEY): res2.end_token = toks2.end_token return res2 elif (nt1 is not None and nt1.typ == NumberSpellingType.WORDS and nt.typ == NumberSpellingType.DIGIT): altf = nt1.real_value ttt = t1.next0_.next0_ if (ttt.isChar(')')): t1 = ttt.next0_ det = True if (not det): altf = f if ((t1 is not None and t1.isChar('(') and t1.next0_ is not None) and t1.next0_.isValue("СУММА", None)): br = BracketHelper.tryParse(t1, BracketParseAttr.NO, 100) if (br is not None): t1 = br.end_token.next0_ if (is_dollar is not None): te = None if (t1 is not None): te = t1.previous else: t1 = t0 while t1 is not None: if (t1.next0_ is None): te = t1 t1 = t1.next0_ if (te is None): return None if (te.is_hiphen and te.next0_ is not None): if (te.next0_.isValue("МИЛЛИОННЫЙ", None)): f *= (1000000) altf *= (1000000) te = te.next0_ elif (te.next0_.isValue("МИЛЛИАРДНЫЙ", None)): f *= (1000000000) altf *= (1000000000) te = te.next0_ if (not te.is_whitespace_after and (isinstance(te.next0_, TextToken))): if (te.next0_.isValue("M", None)): f *= (1000000) altf *= (1000000) te = te.next0_ elif (te.next0_.isValue("BN", None)): f *= (1000000000) altf *= (1000000000) te = te.next0_ return NumberExToken._new474(t0, te, "", nt.typ, NumberExType.MONEY, f, altf, is_dollar) if (t1 is None or ((t1.is_newline_before and not det))): return None toks = NumberExHelper._m_postfixes.tryParse(t1, TerminParseAttr.NO) if ((toks is None and det and (isinstance(t1, NumberToken))) and (t1).value == "0"): toks = NumberExHelper._m_postfixes.tryParse( t1.next0_, TerminParseAttr.NO) if (toks is not None): t1 = toks.end_token if (not t1.isChar('.') and t1.next0_ is not None and t1.next0_.isChar('.')): if ((isinstance(t1, TextToken)) and t1.isValue( toks.termin.terms[0].canonical_text, None)): pass elif (not t1.chars.is_letter): pass else: t1 = t1.next0_ if (toks.termin.canonic_text == "LTL"): return None if (toks.begin_token == t1): if (t1.morph.class0_.is_preposition or t1.morph.class0_.is_conjunction): if (t1.is_whitespace_before and t1.is_whitespace_after): return None ty = Utils.valToEnum(toks.termin.tag, NumberExType) res = NumberExToken._new473(t, t1, nt.value, nt.typ, ty, f, altf, toks.begin_token.morph) if (ty != NumberExType.MONEY): NumberExHelper.__correctExtTypes(res) return res return NumberExHelper.__correctMoney(res, toks.begin_token) pfx = NumberExHelper.__attachSpecPostfix(t1) if (pfx is not None): pfx.begin_token = t pfx.value = nt.value pfx.typ = nt.typ pfx.real_value = f pfx.alt_real_value = altf return pfx if (t1.next0_ is not None and ((t1.morph.class0_.is_preposition or t1.morph.class0_.is_conjunction))): if (t1.isValue("НА", None)): pass else: nn = NumberExHelper.tryParseNumberWithPostfix(t1.next0_) if (nn is not None): return NumberExToken._new476(t, t, nt.value, nt.typ, nn.ex_typ, f, altf, nn.ex_typ2, nn.ex_typ_param) if (not t1.is_whitespace_after and (isinstance(t1.next0_, NumberToken)) and (isinstance(t1, TextToken))): term = (t1).term ty = NumberExType.UNDEFINED if (term == "СМХ" or term == "CMX"): ty = NumberExType.SANTIMETER elif (term == "MX" or term == "МХ"): ty = NumberExType.METER elif (term == "MMX" or term == "ММХ"): ty = NumberExType.MILLIMETER if (ty != NumberExType.UNDEFINED): return NumberExToken._new477(t, t1, nt.value, nt.typ, ty, f, altf, True) return None
def __tryAttachContractGround(t: 'Token', ip: 'InstrumentParticipant', can_be_passport: bool = False) -> 'Token': ok = False first_pass3021 = True while True: if first_pass3021: first_pass3021 = False else: t = t.next0_ if (not (t is not None)): break if (t.isChar(',') or t.morph.class0_.is_preposition): continue if (t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): t = br.end_token continue if (t.isValue("ОСНОВАНИЕ", None) or t.isValue("ДЕЙСТВОВАТЬ", None) or t.isValue("ДЕЙСТВУЮЩИЙ", None)): ok = True if (t.next0_ is not None and t.next0_.isChar('(')): br = BracketHelper.tryParse(t.next0_, BracketParseAttr.NO, 100) if (br is not None and (br.length_char < 10)): t = br.end_token continue dr = Utils.asObjectOrNull(t.getReferent(), DecreeReferent) if (dr is not None): ip.ground = dr return t pir = Utils.asObjectOrNull(t.getReferent(), PersonIdentityReferent) if (pir is not None and can_be_passport): if (pir.typ is not None and not "паспорт" in pir.typ): ip.ground = pir return t if (t.isValue("УСТАВ", None)): ip.ground = t.getNormalCaseText(MorphClass.NOUN, True, MorphGender.UNDEFINED, False) return t if (t.isValue("ДОВЕРЕННОСТЬ", None)): dts = DecreeToken.tryAttachList(t.next0_, None, 10, False) if (dts is None): has_spec = False ttt = t.next0_ first_pass3022 = True while True: if first_pass3022: first_pass3022 = False else: ttt = ttt.next0_ if (not (ttt is not None and ((ttt.end_char - t.end_char) < 200))): break if (ttt.is_comma): continue if (ttt.isValue("УДОСТОВЕРИТЬ", None) or ttt.isValue("УДОСТОВЕРЯТЬ", None)): has_spec = True continue dt = DecreeToken.tryAttach(ttt, None, False) if (dt is not None): if (dt.typ == DecreeToken.ItemType.DATE or dt.typ == DecreeToken.ItemType.NUMBER): dts = DecreeToken.tryAttachList( ttt, None, 10, False) break npt = NounPhraseHelper.tryParse( ttt, NounPhraseParseAttr.NO, 0) if (npt is not None): if (npt.end_token.isValue("НОТАРИУС", None)): ttt = npt.end_token has_spec = True continue if (ttt.getReferent() is not None): if (has_spec): continue break if (dts is not None and len(dts) > 0): t0 = t dr = DecreeReferent() dr.typ = "ДОВЕРЕННОСТЬ" for d in dts: if (d.typ == DecreeToken.ItemType.DATE): dr._addDate(d) t = d.end_token elif (d.typ == DecreeToken.ItemType.NUMBER): dr._addNumber(d) t = d.end_token else: break ad = t.kit.getAnalyzerDataByAnalyzerName( InstrumentAnalyzer.ANALYZER_NAME) ip.ground = ad.registerReferent(dr) rt = ReferentToken( Utils.asObjectOrNull(ip.ground, Referent), t0, t) t.kit.embedToken(rt) return rt ip.ground = "ДОВЕРЕННОСТЬ" return t break return None
def __correctTailAttributes(p : 'PersonReferent', t0 : 'Token') -> 'Token': res = t0 t = t0 if (t is not None and t.isChar(',')): t = t.next0_ born = False die = False if (t is not None and ((t.isValue("РОДИТЬСЯ", "НАРОДИТИСЯ") or t.isValue("BORN", None)))): t = t.next0_ born = True elif (t is not None and ((t.isValue("УМЕРЕТЬ", "ПОМЕРТИ") or t.isValue("СКОНЧАТЬСЯ", None) or t.isValue("DIED", None)))): t = t.next0_ die = True elif ((t is not None and t.isValue("ДАТА", None) and t.next0_ is not None) and t.next0_.isValue("РОЖДЕНИЕ", "НАРОДЖЕННЯ")): t = t.next0_.next0_ born = True while t is not None: if (t.morph.class0_.is_preposition or t.is_hiphen or t.isChar(':')): t = t.next0_ else: break if (t is not None and t.getReferent() is not None): r = t.getReferent() if (r.type_name == "DATE"): t1 = t if (t.next0_ is not None and ((t.next0_.isValue("Р", None) or t.next0_.isValue("РОЖДЕНИЕ", "НАРОДЖЕННЯ")))): born = True t1 = t.next0_ if (t1.next0_ is not None and t1.next0_.isChar('.')): t1 = t1.next0_ if (born): if (p is not None): p.addSlot(PersonReferent.ATTR_BORN, r, False, 0) res = t1 t = t1 elif (die): if (p is not None): p.addSlot(PersonReferent.ATTR_DIE, r, False, 0) res = t1 t = t1 if (die and t is not None): ag = NumberHelper.tryParseAge(t.next0_) if (ag is not None): if (p is not None): p.addSlot(PersonReferent.ATTR_AGE, str(ag.value), False, 0) t = ag.end_token.next0_ res = ag.end_token if (t is None): return res if (t.isChar('(')): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): t1 = t.next0_ born = False if (t1.isValue("РОД", None)): born = True t1 = t1.next0_ if (t1 is not None and t1.isChar('.')): t1 = t1.next0_ if (isinstance(t1, ReferentToken)): r = t1.getReferent() if (r.type_name == "DATERANGE" and t1.next0_ == br.end_token): bd = Utils.asObjectOrNull(r.getSlotValue("FROM"), Referent) to = Utils.asObjectOrNull(r.getSlotValue("TO"), Referent) if (bd is not None and to is not None): if (p is not None): p.addSlot(PersonReferent.ATTR_BORN, bd, False, 0) p.addSlot(PersonReferent.ATTR_DIE, to, False, 0) res = br.end_token t = res elif (r.type_name == "DATE" and t1.next0_ == br.end_token): if (p is not None): p.addSlot(PersonReferent.ATTR_BORN, r, False, 0) res = br.end_token t = res return res
def parse(t: 'Token', max_char: int = 0, prev: 'InstrToken' = None) -> 'InstrToken': is_start_of_line = False t00 = t if (t is not None): is_start_of_line = t00.is_newline_before while t is not None: if (t.is_table_control_char and not t.isChar(chr(0x1F))): if (t.is_newline_after and not is_start_of_line): is_start_of_line = True t = t.next0_ else: break if (t is None): return None if (t.is_newline_before): is_start_of_line = True t0 = t t1 = None has_word = False first_pass2988 = True while True: if first_pass2988: first_pass2988 = False else: t = t.next0_ if (not (t is not None)): break if (t.is_newline_before and t != t0): break if (max_char > 0 and t.begin_char > max_char): break if (is_start_of_line and t == t0): if (t.isValue("ГЛАВА", None)): next0__ = InstrToken.parse(t.next0_, 0, None) if (next0__ is not None and next0__.typ == ILTypes.PERSON): next0__.begin_token = t return next0__ tt = None if ((isinstance(t.getReferent(), PersonReferent)) or (isinstance(t.getReferent(), PersonPropertyReferent)) or (isinstance(t.getReferent(), InstrumentParticipant))): return InstrToken.__correctPerson( InstrToken._new1405(t00, t, ILTypes.PERSON, t)) is_ref = False if (isinstance(t.getReferent(), PersonPropertyReferent)): tt = t.next0_ is_ref = True elif (prev is not None and prev.typ == ILTypes.PERSON): rt = t.kit.processReferent(PersonAnalyzer.ANALYZER_NAME, t) if (rt is not None): if (isinstance(rt.referent, PersonReferent)): return InstrToken._new1406(t00, rt.end_token, ILTypes.PERSON) tt = rt.end_token.next0_ cou = 0 t11 = (None if tt is None else tt.previous) first_pass2989 = True while True: if first_pass2989: first_pass2989 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.is_table_control_char): continue re = tt.getReferent() if (isinstance(re, PersonReferent)): return InstrToken._new1405(t00, tt, ILTypes.PERSON, tt) if (isinstance(re, GeoReferent)): t11 = tt continue if (re is not None): break if (DecreeToken.isKeyword(tt, False) is not None): break if (tt.is_newline_before): cou += 1 if ((cou) > 4): break if (tt is None and is_ref): return InstrToken._new1405(t00, Utils.ifNotNull(t11, t), ILTypes.PERSON, t) dt = DecreeToken.tryAttach(t, None, False) if (dt is not None): if (dt.typ == DecreeToken.ItemType.TYP and not t.chars.is_all_lower): if (t != t0): break has_verb_ = False tt = dt.end_token while tt is not None: if (tt.is_newline_before): break elif ((isinstance(tt, TextToken)) and (tt).is_pure_verb): has_verb_ = True break tt = tt.next0_ if (not has_verb_): res2 = InstrToken._new1409( t0, dt.end_token, ILTypes.TYP, Utils.ifNotNull(dt.full_value, dt.value)) if (res2.value == "ДОПОЛНИТЕЛЬНОЕ СОГЛАШЕНИЕ" or res2.value == "ДОДАТКОВА УГОДА"): if (res2.begin_char > 500 and res2.newlines_before_count > 1): res2.typ = ILTypes.APPENDIX return res2 if (dt.typ == DecreeToken.ItemType.NUMBER): if (t != t0): break return InstrToken._new1409(t0, dt.end_token, ILTypes.REGNUMBER, dt.value) if (dt.typ == DecreeToken.ItemType.ORG): if (t != t0): break return InstrToken._new1411(t0, dt.end_token, ILTypes.ORGANIZATION, dt.ref, dt.value) if (dt.typ == DecreeToken.ItemType.TERR): if (t != t0): break re = InstrToken._new1411(t0, dt.end_token, ILTypes.GEO, dt.ref, dt.value) t1 = re.end_token.next0_ if (t1 is not None and t1.isChar(',')): t1 = t1.next0_ if (t1 is not None and t1.isValue("КРЕМЛЬ", None)): re.end_token = t1 elif ((t1 is not None and t1.isValue("ДОМ", "БУДИНОК") and t1.next0_ is not None) and t1.next0_.isValue("СОВЕТ", "РАД")): re.end_token = t1.next0_ if (t1.next0_.next0_ is not None and (isinstance( t1.next0_.next0_.getReferent(), GeoReferent))): re.end_token = t1.next0_.next0_ return re if (dt.typ == DecreeToken.ItemType.OWNER): if (t != t0): break if (dt.ref is not None and str(dt.ref.referent).startswith("агент")): dt = (None) if (dt is not None): res1 = InstrToken._new1411(t0, dt.end_token, ILTypes.PERSON, dt.ref, dt.value) return InstrToken.__correctPerson(res1) if (BracketHelper.canBeStartOfSequence(t, False, False)): br = BracketHelper.tryParse(t, BracketParseAttr.NO, 100) if (br is not None): t1 = br.end_token t = t1 continue if (t.next0_ is not None and BracketHelper.canBeEndOfSequence( t.next0_, False, None, False)): t1 = t.next0_ t = t1 continue if (isinstance(t, TextToken)): if (t.isChar('_')): t1 = t continue r = t.getReferent() if (isinstance(r, DateReferent)): tt = t if (tt.next0_ is not None and tt.next0_.isCharOf(",;")): tt = tt.next0_ if (not t.is_newline_before and not tt.is_newline_after): t1 = tt continue if (not has_word): return InstrToken._new1405(t, tt, ILTypes.DATE, t) if (t != t0): break has_word = True if (isinstance(r, InstrumentParticipant)): tt = (t).begin_token first_pass2990 = True while True: if first_pass2990: first_pass2990 = False else: tt = tt.next0_ if (not (tt is not None and (tt.end_char < t.end_char))): break rr = tt.getReferent() if (rr is None): continue if ((isinstance(rr, OrganizationReferent)) or (isinstance(rr, BankDataReferent)) or (isinstance(rr, UriReferent))): r = (None) break if ((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent)) or (isinstance(r, InstrumentParticipant))): if (t != t0): break if (isinstance(r, InstrumentParticipant)): pass res1 = InstrToken._new1405(t, t, ILTypes.PERSON, t) return InstrToken.__correctPerson(res1) if (isinstance(r, OrganizationReferent)): if (t != t0): break return InstrToken._new1405(t, t, ILTypes.ORGANIZATION, t) if (isinstance(r, DecreePartReferent)): dpr = Utils.asObjectOrNull(r, DecreePartReferent) if (dpr.appendix is not None): if (t.is_newline_before or is_start_of_line): if (t.is_newline_after or t.whitespaces_before_count > 30): return InstrToken._new1409(t, t, ILTypes.APPENDIX, "ПРИЛОЖЕНИЕ") ok = True tt = t.next0_ first_pass2991 = True while True: if first_pass2991: first_pass2991 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.is_newline_before): break npt = NounPhraseHelper.tryParse( tt, NounPhraseParseAttr.NO, 0) if (npt is not None): tt = npt.end_token continue ok = False break if (ok): return InstrToken._new1409(t, t, ILTypes.APPENDIX, "ПРИЛОЖЕНИЕ") if ((isinstance(r, DecreeReferent)) and (r).kind == DecreeKind.PUBLISHER and t == t0): res1 = InstrToken._new1406(t, t, ILTypes.APPROVED) tt = t.next0_ first_pass2992 = True while True: if first_pass2992: first_pass2992 = False else: tt = tt.next0_ if (not (tt is not None)): break if (tt.isCharOf(",;")): continue if ((isinstance(tt.getReferent(), DecreeReferent)) and (tt.getReferent()).kind == DecreeKind.PUBLISHER): res1.end_token = t else: break return res1 if (t.isValue("ЗА", None) and t.next0_ is not None and t.is_newline_before): rr = t.next0_.getReferent() if ((isinstance(rr, PersonReferent)) or (isinstance(rr, PersonPropertyReferent)) or (isinstance(rr, InstrumentParticipant))): if (t != t0): break res1 = InstrToken._new1405(t, t.next0_, ILTypes.PERSON, t.next0_) t = t.next0_.next0_ if ((isinstance(rr, InstrumentParticipant)) and t is not None): r = t.getReferent() if ((r) is not None): if ((isinstance(r, PersonReferent)) or (isinstance(r, PersonPropertyReferent))): res1.end_token = t res1.ref = (t) return res1 ii = 0 while ii < len(InstrToken._m_directives): if (t.isValue(InstrToken._m_directives[ii], None)): if (t.next0_ is not None and t.next0_.isValue("СЛЕДУЮЩЕЕ", "НАСТУПНЕ")): if (t != t0): break t11 = t.next0_ ok = False if (t11.next0_ is not None and t11.next0_.isCharOf(":.") and t11.next0_.is_newline_after): ok = True t11 = t11.next0_ if (ok): return InstrToken._new1409( t, t11, ILTypes.DIRECTIVE, InstrToken._m_directives_norm[ii]) if (t.is_newline_after or ((t.next0_ is not None and t.next0_.isChar(':') and t.next0_.is_newline_after))): if (t != t0): break if (not t.is_newline_before): if ((InstrToken._m_directives_norm[ii] != "ПРИКАЗ" and InstrToken._m_directives_norm[ii] != "ПОСТАНОВЛЕНИЕ" and InstrToken._m_directives_norm[ii] != "НАКАЗ") and InstrToken._m_directives_norm[ii] != "ПОСТАНОВУ"): break return InstrToken._new1409( t, (t if t.is_newline_after else t.next0_), ILTypes.DIRECTIVE, InstrToken._m_directives_norm[ii]) break ii += 1 if (t.is_newline_before and t.chars.is_letter and t.length_char == 1): for d in InstrToken._m_directives: t11 = MiscHelper.tryAttachWordByLetters(d, t, True) if (t11 is not None): if (t11.next0_ is not None and t11.next0_.isChar(':')): t11 = t11.next0_ return InstrToken._new1406(t, t11, ILTypes.DIRECTIVE) tte = ((t).begin_token if (isinstance(t, MetaToken)) else t) term = ((tte).term if isinstance(tte, TextToken) else None) if (is_start_of_line and not tte.chars.is_all_lower and t == t0): npt = NounPhraseHelper.tryParse(tte, NounPhraseParseAttr.NO, 0) if (npt is not None and ((term == "ПРИЛОЖЕНИЯ" or term == "ДОДАТКИ"))): # if (tte.Next != null && tte.Next.IsChar(':')) npt = (None) if (npt is not None and npt.morph.case_.is_nominative and (isinstance(npt.end_token, TextToken))): term1 = (npt.end_token).term if (((term1 == "ПРИЛОЖЕНИЕ" or term1 == "ДОДАТОК" or term1 == "МНЕНИЕ") or term1 == "ДУМКА" or term1 == "АКТ") or term1 == "ФОРМА" or term == "ЗАЯВКА"): tt1 = npt.end_token.next0_ dt1 = DecreeToken.tryAttach(tt1, None, False) if (dt1 is not None and dt1.typ == DecreeToken.ItemType.NUMBER): tt1 = dt1.end_token.next0_ elif (isinstance(tt1, NumberToken)): tt1 = tt1.next0_ elif ((isinstance(tt1, TextToken)) and tt1.length_char == 1 and tt1.chars.is_letter): tt1 = tt1.next0_ ok = True if (tt1 is None): ok = False elif (tt1.isValue("В", "У")): ok = False elif (tt1.isValue("К", None) and tt1.is_newline_before): return InstrToken._new1409(t, t, ILTypes.APPENDIX, term1) elif (not tt1.is_newline_before and InstrToken._checkEntered(tt1) is not None): ok = False elif (tt1 == t.next0_ and ((tt1.isChar(':') or ((tt1.isValue("НА", None) and term1 != "ЗАЯВКА"))))): ok = False if (ok): br = BracketHelper.tryParse( tt1, BracketParseAttr.NO, 100) if (br is not None): tt1 = br.end_token.next0_ if (br.end_token.next0_ is None or not br.end_token.is_newline_after or br.end_token.next0_.isCharOf(";,")): ok = False if (tt1 is not None and tt1.isValue( "ПРИЛОЖЕНИЕ", "ДОДАТОК")): ok = False if (prev is not None and prev.typ == ILTypes.APPENDIX): ok = False if (ok): cou = 0 ttt = tte.previous while ttt is not None and (cou < 300): if (ttt.is_table_control_char): if (not ttt.isChar(chr(0x1F))): if (ttt == tte.previous and ttt.isChar(chr(0x1E))): pass else: ok = False break ttt = ttt.previous cou += 1 if (ok): it1 = InstrToken1.parse(t, True, None, 0, None, False, 0, False) if (it1 is not None): if (it1.has_verb): ok = False if (ok and t.previous is not None): ttp = t.previous first_pass2993 = True while True: if first_pass2993: first_pass2993 = False else: ttp = ttp.previous if (not (ttp is not None)): break if (ttp.is_table_control_char and not ttp.isChar(chr(0x1F))): continue if (BracketHelper.isBracket(ttp, False) and not BracketHelper.canBeEndOfSequence( ttp, False, None, False)): continue if (ttp.isCharOf(";:")): ok = False break if ((ok and t.previous is not None and (t.newlines_before_count < 3)) and not t.is_newline_after): lines = 0 ttp = t.previous first_pass2994 = True while True: if first_pass2994: first_pass2994 = False else: ttp = ttp.previous if (not (ttp is not None)): break if (not ttp.is_newline_before): continue while ttp is not None and (ttp.end_char < t.begin_char): if (isinstance(ttp, NumberToken)): pass elif ((isinstance(ttp, TextToken)) and ttp.length_char > 1): if (ttp.isValue( "ПРИЛОЖЕНИЕ", "ДОДАТОК")): ok = False break else: break ttp = ttp.next0_ lines += 1 if ((lines) > 1): break if (ok and ((term1 != "ПРИЛОЖЕНИЕ" and term1 != "ДОДАТОК" and term1 != "МНЕНИЕ"))): if (t.newlines_before_count < 3): ok = False if (ok): return InstrToken._new1409(t, t, ILTypes.APPENDIX, term1) app = False if ((((term == "ОСОБОЕ" or term == "ОСОБЛИВЕ")) and t.next0_ is not None and t.next0_.isValue("МНЕНИЕ", "ДУМКА")) and t == t0 and is_start_of_line): app = True if ((((term == "ДОПОЛНИТЕЛЬНОЕ" or term == "ДОДАТКОВА")) and t.next0_ is not None and t.next0_.isValue("СОГЛАШЕНИЕ", "УГОДА")) and t == t0 and is_start_of_line): app = True if (app): tt = t.next0_ while tt is not None: if (tt.is_newline_before): break elif (tt.getMorphClassInDictionary() == MorphClass.VERB): app = False break tt = tt.next0_ if (app): return InstrToken._new1406(t, t.next0_, ILTypes.APPENDIX) if (not t.chars.is_all_lower and t == t0): tt = InstrToken._checkApproved(t) if (tt is not None): if (tt.next0_ is not None and (isinstance(tt.next0_.getReferent(), DecreeReferent))): return InstrToken._new1405(t, tt, ILTypes.APPROVED, tt.next0_.getReferent()) dt1 = DecreeToken.tryAttach(tt.next0_, None, False) if (dt1 is not None and dt1.typ == DecreeToken.ItemType.TYP): return InstrToken._new1406(t, tt, ILTypes.APPROVED) t1 = t is_start_of_line = False if (t1 is None): return None res = InstrToken._new1406(t00, t1, ILTypes.UNDEFINED) res.no_words = True t = t0 first_pass2995 = True while True: if first_pass2995: first_pass2995 = False else: t = t.next0_ if (not (t is not None and t.end_char <= t1.end_char)): break if (not ((isinstance(t, TextToken)))): if (isinstance(t, ReferentToken)): res.no_words = False continue if (not t.chars.is_letter): continue res.no_words = False if ((t).is_pure_verb): res.has_verb = True if (t0.isValue("ВОПРОС", "ПИТАННЯ") and t0.next0_ is not None and t0.next0_.isCharOf(":.")): res.typ = ILTypes.QUESTION return res