def process(self, kit: 'AnalysisKit') -> None:
ad = kit.get_analyzer_data(self)
delta = 100000
parts = math.floor((((len(kit.sofa.text) + delta) - 1)) / delta)
if (parts == 0):
parts = 1
cur = 0
next_pos = 0
t = kit.first_token
first_pass3182 = True
while True:
if first_pass3182: first_pass3182 = False
else: t = t.next0_
if (not (t is not None)): break
if (t.begin_char > next_pos):
next_pos += delta
cur += 1
if (not self._on_progress(cur, parts, kit)):
break
at = GoodAttrToken.try_parse(t, None, True, True)
if (at is None):
continue
attr = at._create_attr()
if (attr is None):
t = at.end_token
continue
rt = ReferentToken(attr, at.begin_token, at.end_token)
rt.referent = ad.register_referent(attr)
kit.embed_token(rt)
t = (rt)
def process(self, kit : 'AnalysisKit') -> None:
""" Основная функция выделения объектов
Args:
container:
lastStage:
"""
ad = kit.getAnalyzerData(self)
t = kit.first_token
first_pass3149 = True
while True:
if first_pass3149: first_pass3149 = False
else: t = t.next0_
if (not (t is not None)): break
tt = t
tok = UriAnalyzer.__m_schemes.tryParse(t, TerminParseAttr.NO)
if (tok is not None):
i = (tok.termin.tag)
tt = tok.end_token
if (tt.next0_ is not None and tt.next0_.isChar('(')):
tok1 = UriAnalyzer.__m_schemes.tryParse(tt.next0_.next0_, TerminParseAttr.NO)
if ((tok1 is not None and tok1.termin.canonic_text == tok.termin.canonic_text and tok1.end_token.next0_ is not None) and tok1.end_token.next0_.isChar(')')):
tt = tok1.end_token.next0_
if (i == 0):
if ((tt.next0_ is None or ((not tt.next0_.isCharOf(":|") and not tt.is_table_control_char)) or tt.next0_.is_whitespace_before) or tt.next0_.whitespaces_after_count > 2):
continue
t1 = tt.next0_.next0_
while t1 is not None and t1.isCharOf("/\\"):
t1 = t1.next0_
if (t1 is None or t1.whitespaces_before_count > 2):
continue
ut = UriItemToken.attachUriContent(t1, False)
if (ut is None):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557(tok.termin.canonic_text.lower(), ut.value)), UriReferent)
rt = ReferentToken(ad.registerReferent(ur), t, ut.end_token)
rt.begin_token = Utils.ifNotNull(UriAnalyzer.__siteBefore(t.previous), t)
if (rt.end_token.next0_ is not None and rt.end_token.next0_.isCharOf("/\\")):
rt.end_token = rt.end_token.next0_
kit.embedToken(rt)
t = (rt)
continue
if (i == 10):
tt = tt.next0_
if (tt is None or not tt.isChar(':')):
continue
tt = tt.next0_
while tt is not None:
if (tt.isCharOf("/\\")):
pass
else:
break
tt = tt.next0_
if (tt is None):
continue
if (tt.isValue("WWW", None) and tt.next0_ is not None and tt.next0_.isChar('.')):
tt = tt.next0_.next0_
if (tt is None or tt.is_newline_before):
continue
ut = UriItemToken.attachUriContent(tt, True)
if (ut is None):
continue
if (len(ut.value) < 4):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557(tok.termin.canonic_text.lower(), ut.value)), UriReferent)
rt = ReferentToken(ad.registerReferent(ur), t, ut.end_token)
rt.begin_token = Utils.ifNotNull(UriAnalyzer.__siteBefore(t.previous), t)
if (rt.end_token.next0_ is not None and rt.end_token.next0_.isCharOf("/\\")):
rt.end_token = rt.end_token.next0_
kit.embedToken(rt)
t = (rt)
continue
if (i == 2):
if (tt.next0_ is None or not tt.next0_.isChar('.') or tt.next0_.is_whitespace_before):
continue
if (tt.next0_.is_whitespace_after and tok.termin.canonic_text != "WWW"):
continue
ut = UriItemToken.attachUriContent(tt.next0_.next0_, True)
if (ut is None):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557("http", ut.value)), UriReferent)
rt = ReferentToken(ur, t, ut.end_token)
rt.begin_token = Utils.ifNotNull(UriAnalyzer.__siteBefore(t.previous), t)
if (rt.end_token.next0_ is not None and rt.end_token.next0_.isCharOf("/\\")):
rt.end_token = rt.end_token.next0_
kit.embedToken(rt)
t = (rt)
continue
if (i == 1):
sch = tok.termin.canonic_text
ut = None
if (sch == "ISBN"):
ut = UriItemToken.attachISBN(tt.next0_)
if ((ut is None and t.previous is not None and t.previous.isChar('(')) and t.next0_ is not None and t.next0_.isChar(')')):
tt0 = t.previous.previous
while tt0 is not None:
if (tt0.whitespaces_after_count > 2):
break
if (tt0.is_whitespace_before):
ut = UriItemToken.attachISBN(tt0)
if (ut is not None and ut.end_token.next0_ != t.previous):
ut = (None)
break
tt0 = tt0.previous
elif ((sch == "RFC" or sch == "ISO" or sch == "ОКФС") or sch == "ОКОПФ"):
ut = UriItemToken.attachISOContent(tt.next0_, ":")
elif (sch == "ГОСТ"):
ut = UriItemToken.attachISOContent(tt.next0_, "-.")
elif (sch == "ТУ"):
if (tok.chars.is_all_upper):
ut = UriItemToken.attachISOContent(tt.next0_, "-.")
if (ut is not None and (ut.length_char < 10)):
ut = (None)
else:
ut = UriItemToken.attachBBK(tt.next0_)
if (ut is None):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560(ut.value, sch)), UriReferent)
if (ut.begin_char < t.begin_char):
rt = ReferentToken(ur, ut.begin_token, t)
if (t.next0_ is not None and t.next0_.isChar(')')):
rt.end_token = t.next0_
else:
rt = ReferentToken(ur, t, ut.end_token)
if (t.previous is not None and t.previous.isValue("КОД", None)):
rt.begin_token = t.previous
if (ur.scheme.startswith("ОК")):
UriAnalyzer.__checkDetail(rt)
kit.embedToken(rt)
t = (rt)
if (ur.scheme.startswith("ОК")):
while t.next0_ is not None:
if (t.next0_.is_comma_and and (isinstance(t.next0_.next0_, NumberToken))):
pass
else:
break
ut = UriItemToken.attachBBK(t.next0_.next0_)
if (ut is None):
break
ur = (Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560(ut.value, sch)), UriReferent))
rt = ReferentToken(ur, t.next0_.next0_, ut.end_token)
UriAnalyzer.__checkDetail(rt)
kit.embedToken(rt)
t = (rt)
continue
if (i == 3):
t0 = tt.next0_
while t0 is not None:
if (t0.isCharOf(":|") or t0.is_table_control_char or t0.is_hiphen):
t0 = t0.next0_
else:
break
if (t0 is None):
continue
ut = UriItemToken.attachSkype(t0)
if (ut is None):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560(ut.value.lower(), ("skype" if tok.termin.canonic_text == "SKYPE" else tok.termin.canonic_text))), UriReferent)
rt = ReferentToken(ur, t, ut.end_token)
kit.embedToken(rt)
t = (rt)
continue
if (i == 4):
t0 = tt.next0_
if (t0 is not None and ((t0.isChar(':') or t0.is_hiphen))):
t0 = t0.next0_
if (t0 is None):
continue
ut = UriItemToken.attachIcqContent(t0)
if (ut is None):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560(ut.value, "ICQ")), UriReferent)
rt = ReferentToken(ur, t, t0)
kit.embedToken(rt)
t = (rt)
continue
if (i == 5 or i == 6):
t0 = tt.next0_
has_tab_cel = False
is_iban = False
first_pass3150 = True
while True:
if first_pass3150: first_pass3150 = False
else: t0 = t0.next0_
if (not (t0 is not None)): break
if ((((t0.isValue("БАНК", None) or t0.morph.class0_.is_preposition or t0.is_hiphen) or t0.isCharOf(".:") or t0.isValue("РУБЛЬ", None)) or t0.isValue("РУБ", None) or t0.isValue("ДОЛЛАР", None)) or t0.isValue("№", None) or t0.isValue("N", None)):
pass
elif (t0.is_table_control_char):
has_tab_cel = True
elif (t0.isCharOf("\\/") and t0.next0_ is not None and t0.next0_.isValue("IBAN", None)):
is_iban = True
t0 = t0.next0_
elif (t0.isValue("IBAN", None)):
is_iban = True
elif (isinstance(t0, TextToken)):
npt = NounPhraseHelper.tryParse(t0, NounPhraseParseAttr.NO, 0)
if (npt is not None and npt.morph.case_.is_genitive):
t0 = npt.end_token
continue
break
else:
break
if (t0 is None):
continue
ur2 = None
ur2begin = None
ur2end = None
t00 = t0
val = t0.getSourceText()
if (str.isdigit(val[0]) and ((((i == 6 or tok.termin.canonic_text == "ИНН" or tok.termin.canonic_text == "БИК") or tok.termin.canonic_text == "ОГРН" or tok.termin.canonic_text == "СНИЛС") or tok.termin.canonic_text == "ОКПО"))):
if (t0.chars.is_letter):
continue
if (Utils.isNullOrEmpty(val) or not str.isdigit(val[0])):
continue
if (t0.length_char < 9):
tmp = io.StringIO()
print(val, end="", file=tmp)
ttt = t0.next0_
first_pass3151 = True
while True:
if first_pass3151: first_pass3151 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.whitespaces_before_count > 1):
break
if (isinstance(ttt, NumberToken)):
print(ttt.getSourceText(), end="", file=tmp)
t0 = ttt
continue
if (ttt.is_hiphen or ttt.isChar('.')):
if (ttt.next0_ is None or not ((isinstance(ttt.next0_, NumberToken)))):
break
if (ttt.is_whitespace_after or ttt.is_whitespace_before):
break
continue
break
val = (None)
if (tmp.tell() == 20):
val = Utils.toStringStringIO(tmp)
elif (tmp.tell() == 9 and tok.termin.canonic_text == "БИК"):
val = Utils.toStringStringIO(tmp)
elif (((tmp.tell() == 10 or tmp.tell() == 12)) and tok.termin.canonic_text == "ИНН"):
val = Utils.toStringStringIO(tmp)
elif (tmp.tell() >= 15 and tok.termin.canonic_text == "Л/С"):
val = Utils.toStringStringIO(tmp)
elif (tmp.tell() >= 11 and ((tok.termin.canonic_text == "ОГРН" or tok.termin.canonic_text == "СНИЛС"))):
val = Utils.toStringStringIO(tmp)
elif (tok.termin.canonic_text == "ОКПО"):
val = Utils.toStringStringIO(tmp)
if (val is None):
continue
elif (not ((isinstance(t0, NumberToken)))):
if ((isinstance(t0, TextToken)) and is_iban):
tmp1 = io.StringIO()
t1 = None
ttt = t0
first_pass3152 = True
while True:
if first_pass3152: first_pass3152 = False
else: ttt = ttt.next0_
if (not (ttt is not None)): break
if (ttt.is_newline_before and ttt != t0):
break
if (ttt.is_hiphen):
continue
if (not ((isinstance(ttt, NumberToken)))):
if (not ((isinstance(ttt, TextToken))) or not ttt.chars.is_latin_letter):
break
print(ttt.getSourceText(), end="", file=tmp1)
t1 = ttt
if (tmp1.tell() >= 34):
break
if (tmp1.tell() < 10):
continue
ur1 = UriReferent._new2560(Utils.toStringStringIO(tmp1), tok.termin.canonic_text)
ur1.addSlot(UriReferent.ATTR_DETAIL, "IBAN", False, 0)
rt1 = ReferentToken(ad.registerReferent(ur1), t, t1)
kit.embedToken(rt1)
t = (rt1)
continue
if (not t0.isCharOf("/\\") or t0.next0_ is None):
continue
tok2 = UriAnalyzer.__m_schemes.tryParse(t0.next0_, TerminParseAttr.NO)
if (tok2 is None or not ((isinstance(tok2.termin.tag, int))) or (tok2.termin.tag) != i):
continue
t0 = tok2.end_token.next0_
while t0 is not None:
if (t0.isCharOf(":N№")):
t0 = t0.next0_
elif (t0.is_table_control_char):
t0 = t0.next0_
t00 = t0
has_tab_cel = True
else:
break
if (not ((isinstance(t0, NumberToken)))):
continue
tmp = io.StringIO()
while t0 is not None:
if (not ((isinstance(t0, NumberToken)))):
break
print(t0.getSourceText(), end="", file=tmp)
t0 = t0.next0_
if (t0 is None or not t0.isCharOf("/\\,") or not ((isinstance(t0.next0_, NumberToken)))):
continue
val = Utils.toStringStringIO(tmp)
Utils.setLengthStringIO(tmp, 0)
ur2begin = t0.next0_
t0 = t0.next0_
while t0 is not None:
if (not ((isinstance(t0, NumberToken)))):
break
if (t0.whitespaces_before_count > 4 and tmp.tell() > 0):
break
print(t0.getSourceText(), end="", file=tmp)
ur2end = t0
t0 = t0.next0_
ur2 = (Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557(tok2.termin.canonic_text, Utils.toStringStringIO(tmp))), UriReferent))
if (len(val) < 5):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560(val, tok.termin.canonic_text)), UriReferent)
rt = ReferentToken(ur, t, (t0 if ur2begin is None else ur2begin.previous))
if (has_tab_cel):
rt.begin_token = t00
if (ur.scheme.startswith("ОК")):
UriAnalyzer.__checkDetail(rt)
ttt = t.previous
first_pass3153 = True
while True:
if first_pass3153: first_pass3153 = False
else: ttt = ttt.previous
if (not (ttt is not None)): break
if (ttt.is_table_control_char):
break
if (ttt.morph.class0_.is_preposition):
continue
if (ttt.isValue("ОРГАНИЗАЦИЯ", None)):
continue
if (ttt.isValue("НОМЕР", None) or ttt.isValue("КОД", None)):
rt.begin_token = ttt
t = rt.begin_token
break
kit.embedToken(rt)
t = (rt)
if (ur2 is not None):
rt2 = ReferentToken(ur2, ur2begin, ur2end)
kit.embedToken(rt2)
t = (rt2)
continue
continue
if (t.isChar('@')):
u1s = UriItemToken.attachMailUsers(t.previous)
if (u1s is None):
continue
u2 = UriItemToken.attachDomainName(t.next0_, False, True)
if (u2 is None):
continue
for ii in range(len(u1s) - 1, -1, -1):
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2560("{0}@{1}".format(u1s[ii].value, u2.value).lower(), "mailto")), UriReferent)
b = u1s[ii].begin_token
t0 = b.previous
if (t0 is not None and t0.isChar(':')):
t0 = t0.previous
if (t0 is not None and ii == 0):
br = False
ttt = t0
first_pass3154 = True
while True:
if first_pass3154: first_pass3154 = False
else: ttt = ttt.previous
if (not (ttt is not None)): break
if (not ((isinstance(ttt, TextToken)))):
break
if (ttt != t0 and ttt.whitespaces_after_count > 1):
break
if (ttt.isChar(')')):
br = True
continue
if (ttt.isChar('(')):
if (not br):
break
br = False
continue
if (ttt.isValue("EMAIL", None) or ttt.isValue("MAILTO", None)):
b = ttt
break
if (ttt.isValue("MAIL", None)):
b = ttt
if ((ttt.previous is not None and ttt.previous.is_hiphen and ttt.previous.previous is not None) and ((ttt.previous.previous.isValue("E", None) or ttt.previous.previous.isValue("Е", None)))):
b = ttt.previous.previous
break
if (ttt.isValue("ПОЧТА", None) or ttt.isValue("АДРЕС", None)):
b = t0
ttt = ttt.previous
if (ttt is not None and ttt.isChar('.')):
ttt = ttt.previous
if (ttt is not None and ((t0.isValue("ЭЛ", None) or ttt.isValue("ЭЛЕКТРОННЫЙ", None)))):
b = ttt
if (b.previous is not None and b.previous.isValue("АДРЕС", None)):
b = b.previous
break
if (ttt.morph.class0_.is_preposition):
continue
rt = ReferentToken(ur, b, (u2.end_token if ii == (len(u1s) - 1) else u1s[ii].end_token))
kit.embedToken(rt)
t = (rt)
continue
if (not t.morph.language.is_cyrillic):
if (t.is_whitespace_before or ((t.previous is not None and t.previous.isCharOf(",(")))):
u1 = UriItemToken.attachUrl(t)
if (u1 is not None):
if (u1.is_whitespace_after or u1.end_token.next0_ is None or not u1.end_token.next0_.isChar('@')):
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557("http", u1.value)), UriReferent)
rt = ReferentToken(ur, u1.begin_token, u1.end_token)
rt.begin_token = Utils.ifNotNull(UriAnalyzer.__siteBefore(u1.begin_token.previous), u1.begin_token)
kit.embedToken(rt)
t = (rt)
continue
if ((isinstance(t, TextToken)) and not t.is_whitespace_after and t.length_char > 2):
if (UriAnalyzer.__siteBefore(t.previous) is not None):
ut = UriItemToken.attachUriContent(t, True)
if (ut is None or ut.value.find('.') <= 0 or ut.value.find('@') > 0):
continue
ur = Utils.asObjectOrNull(ad.registerReferent(UriReferent._new2557("http", ut.value)), UriReferent)
rt = ReferentToken(ur, t, ut.end_token)
rt.begin_token = UriAnalyzer.__siteBefore(t.previous)
if (rt.end_token.next0_ is not None and rt.end_token.next0_.isCharOf("/\\")):
rt.end_token = rt.end_token.next0_
kit.embedToken(rt)
t = (rt)
continue
if ((t.chars.is_latin_letter and not t.chars.is_all_lower and t.next0_ is not None) and not t.is_whitespace_after):
if (t.next0_.isChar('/')):
rt = UriAnalyzer.__TryAttachLotus(Utils.asObjectOrNull(t, TextToken))
if (rt is not None):
rt.referent = ad.registerReferent(rt.referent)
kit.embedToken(rt)
t = (rt)
continue