def __str__(self) -> str:
if (self.countries is None):
return self.pref
res = Utils.newStringIO(self.pref)
for c in self.countries:
print(" {0}".format(c), end="", file=res, flush=True)
return Utils.toStringStringIO(res)
def __str__(self) -> str:
res = Utils.newStringIO(self.value)
if (self.is_noun_phrase):
print(" NounPrase", end="", file=res)
if (self.is_denomination):
print(" Denom", end="", file=res)
if (self.is_in_dictionary):
print(" InDictionary", end="", file=res)
if (self.is_after_conjunction):
print(" IsAfterConjunction", end="", file=res)
if (self.is_std_tail):
print(" IsStdTail", end="", file=res)
if (self.is_std_name):
print(" IsStdName", end="", file=res)
if (self.is_ignored_part):
print(" IsIgnoredPart", end="", file=res)
if (self.preposition is not None):
print(" IsAfterPreposition '{0}'".format(self.preposition),
end="",
file=res,
flush=True)
print(" {0} ({1})".format(str(self.chars), self.getSourceText()),
end="",
file=res,
flush=True)
return Utils.toStringStringIO(res)
def attach_url(t0: 'Token') -> 'UriItemToken':
srv = UriItemToken.attach_domain_name(t0, True, False)
if (srv is None):
return None
txt = Utils.newStringIO(srv.value)
t1 = srv.end_token
if (t1.next0_ is not None and t1.next0_.is_char(':')
and (isinstance(t1.next0_.next0_, NumberToken))):
t1 = t1.next0_.next0_
print(":{0}".format(t1.value), end="", file=txt, flush=True)
elif ((srv.value == "vk.com" and t1.next0_ is not None
and t1.next0_.is_hiphen) and t1.next0_.next0_ is not None):
t1 = t1.next0_.next0_
dat = UriItemToken.__attach_uri_content(t1, ".-_+%", False)
if (dat is not None):
t1 = dat.end_token
print("/{0}".format(dat.value), end="", file=txt, flush=True)
t = t1.next0_
while t is not None:
if (t.is_whitespace_before):
break
if (not t.is_char('/')):
break
if (t.is_whitespace_after):
t1 = t
break
dat = UriItemToken.__attach_uri_content(t.next0_, ".-_+%", False)
if (dat is None):
t1 = t
break
t1 = dat.end_token
t = t1
print("/{0}".format(dat.value), end="", file=txt, flush=True)
t = t.next0_
if ((t1.next0_ is not None and t1.next0_.is_char('?')
and not t1.next0_.is_whitespace_after)
and not t1.is_whitespace_after):
dat = UriItemToken.__attach_uri_content(t1.next0_.next0_,
".-_+%=&", False)
if (dat is not None):
t1 = dat.end_token
print("?{0}".format(dat.value), end="", file=txt, flush=True)
if ((t1.next0_ is not None and t1.next0_.is_char('#')
and not t1.next0_.is_whitespace_after)
and not t1.is_whitespace_after):
dat = UriItemToken.__attach_uri_content(t1.next0_.next0_, ".-_+%",
False)
if (dat is not None):
t1 = dat.end_token
print("#{0}".format(dat.value), end="", file=txt, flush=True)
i = 0
i = 0
while i < txt.tell():
if (str.isalpha(Utils.getCharAtStringIO(txt, i))):
break
i += 1
if (i >= txt.tell()):
return None
return UriItemToken._new2706(t0, t1, Utils.toStringStringIO(txt))
def __str__(self) -> str:
if (Utils.isNullOrEmpty(self.term)):
return "Null"
str0_ = self.term
if (self.char_info.is_all_lower):
str0_ = str0_.lower()
elif (self.char_info.is_capital_upper and len(str0_) > 0):
str0_ = "{0}{1}".format(self.term[0], self.term[1:].lower())
elif (self.char_info.is_last_lower):
str0_ = "{0}{1}".format(self.term[0:0+len(self.term) - 1], self.term[len(self.term) - 1:].lower())
if (self.word_forms is None):
return str0_
res = Utils.newStringIO(str0_)
for l_ in self.word_forms:
print(", {0}".format(str(l_)), end="", file=res, flush=True)
return Utils.toStringStringIO(res)
def __str__(self) -> str:
res = Utils.newStringIO(Utils.ifNotNull(self.caption, self.name))
if (self.upper_bound > 0 or self.lower_bound > 0):
if (self.upper_bound == 0):
print("[{0}..*]".format(self.lower_bound),
end="",
file=res,
flush=True)
elif (self.upper_bound == self.lower_bound):
print("[{0}]".format(self.upper_bound),
end="",
file=res,
flush=True)
else:
print("[{0}..{1}]".format(self.lower_bound, self.upper_bound),
end="",
file=res,
flush=True)
return Utils.toStringStringIO(res)
def to_string(self,
short_variant: bool,
lang: 'MorphLang' = None,
lev: int = 0) -> str:
res = Utils.newStringIO(self.template)
vals = list()
for s in self.slots:
if (s.type_name == MeasureReferent.ATTR_VALUE):
if (isinstance(s.value, str)):
val = Utils.asObjectOrNull(s.value, str)
if (val == "NaN"):
val = "?"
vals.append(val)
elif (isinstance(s.value, Referent)):
vals.append(s.value.to_string(True, lang, 0))
for i in range(res.tell() - 1, -1, -1):
ch = Utils.getCharAtStringIO(res, i)
if (not str.isdigit(ch)):
continue
j = ((ord(ch)) - (ord('1')))
if ((j < 0) or j >= len(vals)):
continue
Utils.removeStringIO(res, i, 1)
Utils.insertStringIO(res, i, vals[j])
print(self.out_units(lang), end="", file=res)
if (not short_variant):
nam = self.get_string_value(MeasureReferent.ATTR_NAME)
if (nam is not None):
print(" - {0}".format(nam), end="", file=res, flush=True)
for s in self.slots:
if (s.type_name == MeasureReferent.ATTR_REF
and (isinstance(s.value, MeasureReferent))):
print(" / {0}".format(s.value.to_string(True, lang, 0)),
end="",
file=res,
flush=True)
ki = self.kind
if (ki != MeasureKind.UNDEFINED):
print(" ({0})".format(Utils.enumToString(ki).upper()),
end="",
file=res,
flush=True)
return Utils.toStringStringIO(res)
def toString(self, short_variant : bool, lang : 'MorphLang'=None, lev : int=0) -> str:
res = Utils.newStringIO(self.template)
vals = list()
for s in self.slots:
if (s.type_name == MeasureReferent.ATTR_VALUE):
if (isinstance(s.value, str)):
vals.append(Utils.asObjectOrNull(s.value, str))
elif (isinstance(s.value, Referent)):
vals.append((s.value).toString(True, lang, 0))
for i in range(res.tell() - 1, -1, -1):
ch = Utils.getCharAtStringIO(res, i)
if (not str.isdigit(ch)):
continue
j = ((ord(ch)) - (ord('1')))
if ((j < 0) or j >= len(vals)):
continue
Utils.removeStringIO(res, i, 1)
Utils.insertStringIO(res, i, vals[j])
uu = self.units
if (len(uu) > 0):
print(uu[0].toString(True, lang, 0), end="", file=res)
i = 1
while i < len(uu):
pow0_ = uu[i].getStringValue(UnitReferent.ATTR_POW)
if (not Utils.isNullOrEmpty(pow0_) and pow0_[0] == '-'):
print("/{0}".format(uu[i].toString(True, lang, 1)), end="", file=res, flush=True)
if (pow0_ != "-1"):
print("<{0}>".format(pow0_[1:]), end="", file=res, flush=True)
else:
print("*{0}".format(uu[i].toString(True, lang, 0)), end="", file=res, flush=True)
i += 1
if (not short_variant):
nam = self.getStringValue(MeasureReferent.ATTR_NAME)
if (nam is not None):
print(" - {0}".format(nam), end="", file=res, flush=True)
for s in self.slots:
if (s.type_name == MeasureReferent.ATTR_REF and (isinstance(s.value, MeasureReferent))):
print(" / {0}".format((s.value).toString(True, lang, 0)), end="", file=res, flush=True)
ki = self.kind
if (ki != MeasureKind.UNDEFINED):
print(" ({0})".format(Utils.enumToString(ki).upper()), end="", file=res, flush=True)
return Utils.toStringStringIO(res)
def __str__(self) -> str:
res = Utils.newStringIO(Utils.ifNotNull(self.normal_case, ""))
if (self.normal_full is not None
and self.normal_full != self.normal_case):
print("\\{0}".format(self.normal_full),
end="",
file=res,
flush=True)
if (res.tell() > 0):
print(' ', end="", file=res)
print(super().__str__(), end="", file=res)
s = (None if self.misc is None else str(self.misc))
if (not Utils.isNullOrEmpty(s)):
print(" {0}".format(s), end="", file=res, flush=True)
if (self.undef_coef > (0)):
print(" (? {0})".format(self.undef_coef),
end="",
file=res,
flush=True)
return Utils.toStringStringIO(res)
def correct_word_by_morph(self, word: str) -> str:
vars0_ = list()
tmp = Utils.newStringIO(len(word))
ch = 1
while ch < len(word):
Utils.setLengthStringIO(tmp, 0)
print(word, end="", file=tmp)
Utils.setCharAtStringIO(tmp, ch, '*')
var = self.__check_corr_var(Utils.toStringStringIO(tmp),
self.m_root, 0)
if (var is not None):
if (not var in vars0_):
vars0_.append(var)
ch += 1
if (len(vars0_) == 0):
ch = 1
while ch < len(word):
Utils.setLengthStringIO(tmp, 0)
print(word, end="", file=tmp)
Utils.insertStringIO(tmp, ch, '*')
var = self.__check_corr_var(Utils.toStringStringIO(tmp),
self.m_root, 0)
if (var is not None):
if (not var in vars0_):
vars0_.append(var)
ch += 1
if (len(vars0_) == 0):
ch = 1
while ch < (len(word) - 1):
Utils.setLengthStringIO(tmp, 0)
print(word, end="", file=tmp)
Utils.removeStringIO(tmp, ch, 1)
var = self.__check_corr_var(Utils.toStringStringIO(tmp),
self.m_root, 0)
if (var is not None):
if (not var in vars0_):
vars0_.append(var)
ch += 1
if (len(vars0_) != 1):
return None
return vars0_[0]
def transliteral_correction(value: str,
prev_value: str,
always: bool = False) -> str:
""" Транслитеральная корректировка
Args:
value(str):
prev_value(str):
always(bool):
"""
pure_cyr = 0
pure_lat = 0
ques_cyr = 0
ques_lat = 0
udar_cyr = 0
y = False
udaren = False
i = 0
first_pass2897 = True
while True:
if first_pass2897: first_pass2897 = False
else: i += 1
if (not (i < len(value))): break
ch = value[i]
ui = UnicodeInfo.ALL_CHARS[ord(ch)]
if (not ui.is_letter):
if (ui.is_udaren):
udaren = True
continue
if (ui.is_apos and len(value) > 2):
return LanguageHelper.transliteral_correction(
value.replace("{0}".format(ch), ""), prev_value, False)
return value
if (ui.is_cyrillic):
if (LanguageHelper._m_cyr_chars.find(ch) >= 0):
ques_cyr += 1
else:
pure_cyr += 1
elif (ui.is_latin):
if (LanguageHelper._m_lat_chars.find(ch) >= 0):
ques_lat += 1
else:
pure_lat += 1
elif (LanguageHelper.__m_udar_chars.find(ch) >= 0):
udar_cyr += 1
else:
return value
if (ch == 'Ь' and ((i + 1) < len(value)) and value[i + 1] == 'I'):
y = True
to_rus = False
to_lat = False
if (pure_lat > 0 and pure_cyr > 0):
return value
if (((pure_lat > 0 or always)) and ques_cyr > 0):
to_lat = True
elif (((pure_cyr > 0 or always)) and ques_lat > 0):
to_rus = True
elif (pure_cyr == 0 and pure_lat == 0):
if (ques_cyr > 0 and ques_lat > 0):
if (not Utils.isNullOrEmpty(prev_value)):
if (LanguageHelper.is_cyrillic_char(prev_value[0])):
to_rus = True
elif (LanguageHelper.is_latin_char(prev_value[0])):
to_lat = True
if (not to_lat and not to_rus):
if (ques_cyr > ques_lat):
to_rus = True
elif (ques_cyr < ques_lat):
to_lat = True
if (not to_rus and not to_lat):
if (not y and not udaren and udar_cyr == 0):
return value
tmp = Utils.newStringIO(value)
i = 0
first_pass2898 = True
while True:
if first_pass2898: first_pass2898 = False
else: i += 1
if (not (i < tmp.tell())): break
if (Utils.getCharAtStringIO(tmp, i) == 'Ь'
and ((i + 1) < tmp.tell())
and Utils.getCharAtStringIO(tmp, i + 1) == 'I'):
Utils.setCharAtStringIO(tmp, i, 'Ы')
Utils.removeStringIO(tmp, i + 1, 1)
continue
cod = ord(Utils.getCharAtStringIO(tmp, i))
if (cod >= 0x300 and (cod < 0x370)):
Utils.removeStringIO(tmp, i, 1)
continue
if (to_rus):
ii = LanguageHelper._m_lat_chars.find(
Utils.getCharAtStringIO(tmp, i))
if (ii >= 0):
Utils.setCharAtStringIO(tmp, i,
LanguageHelper._m_cyr_chars[ii])
else:
ii = LanguageHelper.__m_udar_chars.find(
Utils.getCharAtStringIO(tmp, i))
if (((ii)) >= 0):
Utils.setCharAtStringIO(
tmp, i, LanguageHelper.__m_udar_cyr_chars[ii])
elif (to_lat):
ii = LanguageHelper._m_cyr_chars.find(
Utils.getCharAtStringIO(tmp, i))
if (ii >= 0):
Utils.setCharAtStringIO(tmp, i,
LanguageHelper._m_lat_chars[ii])
else:
ii = LanguageHelper.__m_udar_chars.find(
Utils.getCharAtStringIO(tmp, i))
if (ii >= 0):
Utils.setCharAtStringIO(
tmp, i, LanguageHelper.__m_udar_cyr_chars[ii])
return Utils.toStringStringIO(tmp)
def __str__(self) -> str:
res = Utils.newStringIO(
("Null" if self.referent is None else str(self.referent)))
if (self.morph is not None):
print(" {0}".format(str(self.morph)), end="", file=res, flush=True)
return Utils.toStringStringIO(res)
def __str__(self) -> str:
res = Utils.newStringIO(self.term)
for l_ in self.morph.items:
print(", {0}".format(str(l_)), end="", file=res, flush=True)
return Utils.toStringStringIO(res)
def __doCrLfCorrection(self, txt: str) -> str:
""" Это анализ случаев принудительно отформатированного текста
Args:
txt(str):
"""
cou = 0
total_len = 0
i = 0
first_pass3166 = True
while True:
if first_pass3166: first_pass3166 = False
else: i += 1
if (not (i < len(txt))): break
ch = txt[i]
if ((ord(ch)) != 0xD and (ord(ch)) != 0xA):
continue
len0_ = 0
last_char = ch
j = (i + 1)
while j < len(txt):
ch = txt[j]
if ((ord(ch)) == 0xD or (ord(ch)) == 0xA):
break
elif ((ord(ch)) == 0x9):
len0_ += 5
else:
last_char = ch
len0_ += 1
j += 1
if (j >= len(txt)):
break
if (len0_ < 30):
continue
if (last_char != '.' and last_char != ':' and last_char != ';'):
next_is_dig = False
k = j + 1
while k < len(txt):
if (not Utils.isWhitespace(txt[k])):
if (str.isdigit(txt[k])):
next_is_dig = True
break
k += 1
if (not next_is_dig):
cou += 1
total_len += len0_
i = j
if (cou < 4):
return txt
total_len = math.floor(total_len / cou)
if ((total_len < 50) or total_len > 100):
return txt
tmp = Utils.newStringIO(txt)
i = 0
while i < tmp.tell():
ch = Utils.getCharAtStringIO(tmp, i)
len0_ = 0
last_char = ch
j = (i + 1)
while j < tmp.tell():
ch = Utils.getCharAtStringIO(tmp, j)
if ((ord(ch)) == 0xD or (ord(ch)) == 0xA):
break
elif ((ord(ch)) == 0x9):
len0_ += 5
else:
last_char = ch
len0_ += 1
j += 1
if (j >= tmp.tell()):
break
for jj in range(j - 1, -1, -1):
last_char = Utils.getCharAtStringIO(tmp, jj)
if (not Utils.isWhitespace(last_char)):
break
else:
jj = -1
not_single = False
jj = (j + 1)
if ((jj < tmp.tell())
and (ord(Utils.getCharAtStringIO(tmp, j))) == 0xD
and (ord(Utils.getCharAtStringIO(tmp, jj))) == 0xA):
jj += 1
while jj < tmp.tell():
ch = Utils.getCharAtStringIO(tmp, jj)
if (not Utils.isWhitespace(ch)):
break
if ((ord(ch)) == 0xD or (ord(ch)) == 0xA):
not_single = True
break
jj += 1
if (((not not_single and len0_ >
(total_len - 20) and (len0_ < (total_len + 10)))
and last_char != '.' and last_char != ':')
and last_char != ';'):
Utils.setCharAtStringIO(tmp, j, ' ')
self.crlf_corrected_count += 1
if ((j + 1) < tmp.tell()):
ch = Utils.getCharAtStringIO(tmp, j + 1)
if ((ord(ch)) == 0xA):
Utils.setCharAtStringIO(tmp, j + 1, ' ')
j += 1
i = (j - 1)
i += 1
return Utils.toStringStringIO(tmp)
def __calcTransliteralStatistics(txt: str, info: io.StringIO) -> int:
if (txt is None):
return 0
tmp = Utils.newStringIO(txt)
return SourceOfAnalysis.__doTransliteralCorrection(tmp, info)
