def spellcheckword(word):
check = Word(word)
if check.correct == False:
try:
return check.variants[0]
except IndexError:
return word
else:
return word
def spell_word(word: str):
try:
gr = morph.parse(word)[0].tag.POS
if gr not in ['PREP', 'CONJ', 'PRCL', 'INTJ', 'NPRO'
] and len(word) > 2:
speller_w = Word(word)
return word if speller_w.correct else speller_w.spellsafe
return word
except:
return word
def speller(text):
correct = text
try:
for word in set(text.split(" ")):
if len(word) >= 2:
check = Word(word)
if not check.correct and check.spellsafe:
correct = text.replace(word, check.spellsafe)
except Exception:
pass
finally:
return correct
def check_text(word_list):
misspelt_words = []
for word in word_list:
check = Word(word)
val = check.correct
if not val:
print('Unrecognized word: ' + word)
if check.spellsafe == None:
print('Suggested edits: ' + str(check.spellsafe))
else:
print('Did you mean: ' + str(check.variants))
def correct(txt):
text = ""
for ae in txt.split(" "):
check = Word(ae)
try:
if check.correct:
text += ae + " "
else:
text += check.variants[0] + " "
except:
text += ae + " "
if text.endswith(" "):
text = text[:-1]
return text
def automatic_typo_correction(text):
"""
Автоматическое исправление опечаток в тексте используя pyaspeller.
:param text: строка содержащая текст
:return: строка после обработки
"""
list_word = text.split()
result_text = ""
for word in list_word:
check = Word(word)
result_check = check.spellsafe
if result_check:
result_text += result_check + " "
else:
result_text += word + " "
return result_text
def clean_typos(self, text):
result = []
for w in text.split():
dict_w = self.check_dict(w)
if dict_w:
result.append(dict_w)
else:
check = Word(w)
if check.correct \
or len(check.variants) == 0:
correct_word = w
else:
correct_word = check.variants[0]
result.append(correct_word)
self.update_dict(w, correct_word)
return ' '.join(result)
def correctSpelling(line):
stop_sybols = [
u'/', u'\\', u'№', u':', u'1', u'2', u'3', u'4', u'5', u'6', u'7',
u'8', u'9', u'0', u'–'
]
checked_words = []
for word in line.split():
if not any(st in word for st in stop_sybols):
try:
check = Word(word)
if not check.correct:
if check.spellsafe:
checked_words.append(
check.spellsafe.translate({ord(u"-"): " "}))
else:
checked_words.append(word)
except:
pass
return " ".join(checked_words)
def spellchecking(cleantweet):
list_word = cleantweet.split()
result_string = []
morph = pymorphy2.MorphAnalyzer()
for word in list_word:
check = Word(word)
if not check.correct:
if (len(check.variants)) == 0:
pass
else:
print(word)
print(check.variants)
word = check.variants[0]
else:
pass
word = morph.parse(word)[0].normal_form
word = str(word)
result_string.append(word)
return " ".join(result_string)
def check_word(one_word):
check = Word(one_word)
if not check.correct:
print('Incorrect word: ' + one_word)
print('Did you mean: ' + check.variants)
raw = [line.rstrip() for line in f.readlines()]
with open('./by_wd/test.norm', 'r') as f:
gold_norm = [line.rstrip() for line in f.readlines()]
with open('./by_wd/{}'.format(run), 'r') as f:
csmt_norm = [line.rstrip() for line in f.readlines()]
w = open('./0_results/aligned_{}.tsv'.format(run), 'w')
for raw, gold, csmt in zip(raw, gold_norm, csmt_norm):
raw_words = raw.split()
gold_words = gold.split()
csmt_words = csmt.split()
csmt_spellcheck = []
for word in csmt_words:
sp = Word(word)
checked = sp.spellsafe
if checked:
csmt_spellcheck.append(checked)
else:
csmt_spellcheck.append(word)
# if len(gold_words) == len(csmt_words):
for line in itertools.zip_longest(raw_words, gold_words, csmt_words,
csmt_spellcheck):
w.write('{}\t{}\t{}\t{}\n'.format(*line))
w.close()
from pyaspeller import Word
# f = open('./diff_sizes/norm.txt', 'r')
#
# for line in f:
# for word in line.rstrip().split():
# sp = Word(word)
# checked = sp.spellsafe
# if checked:
# print(word)
for i in range(10):
sp = Word('кашаю')
checked = sp.spellsafe
print(checked)
def main():
# the ciphertext
ciphertext = 'RQH YDULDWLRQ WR WKH VWDQGDUG FDHVDU FLSKHU LV ZKHQ WKH DOSKDEHW LV "NHBHG" EB XVLQJ D ZRUG. LQ WKH ' \
'WUDGLWLRQDO YDULHWB, RQH FRXOG ZULWH WKH DOSKDEHW RQ WZR VWULSV DQG MXVW PDWFK XS WKH VWULSV DIWHU ' \
'VOLGLQJ WKH ERWWRP VWULS WR WKH OHIW RU ULJKW. WR HQFRGH, BRX ZRXOG ILQG D OHWWHU LQ WKH WRS URZ DQG ' \
'VXEVWLWXWH LW IRU WKH OHWWHU LQ WKH ERWWRP URZ. IRU D NHBHG YHUVLRQ, RQH ZRXOG QRW XVH D VWDQGDUG ' \
'DOSKDEHW, EXW ZRXOG ILUVW ZULWH D ZRUG (RPLWWLQJ GXSOLFDWHG OHWWHUV) DQG WKHQ ZULWH WKH UHPDLQLQJ OHWWHUV ' \
'RI WKH DOSKDEHW. IRU WKH HADPSOH EHORZ, L XVHG D NHB RI "UXPNLQ.FRP" DQG BRX ZLOO VHH WKDW WKH SHULRG LV ' \
'UHPRYHG EHFDXVH LW LV QRW D OHWWHU. BRX ZLOO DOVR QRWLFH WKH VHFRQG "P" LV QRW LQFOXGHG EHFDXVH WKHUH ZDV ' \
'DQ P DOUHDGB DQG BRX FDQ\'W KDYH GXSOLFDWHV. '
# empty array for storing alphabet i.e. possible keys
alphabet = ''
# build the alphabet, list of possible keys
for one in range(97, 123):
letter = chr(one).upper()
alphabet = alphabet + letter
# check to see if alphabet populates correctly
print(alphabet)
# start decrypting
# for every index value in the alphabet (0 - 25)
for key in range(len(alphabet)):
# reset the plaintext string on each round
plaintext = ''
# look at each value in the ciphertext
for symbol in ciphertext:
# if the value exists in the alphabet
if symbol in alphabet:
# get the index of the symbol in the alphabet
num = alphabet.find(symbol)
# try every variation of the index from 0-25
num = num - key # key here is the index of the outer for loop
# if the index goes out of bounds i.e. < 0 or < A
if num < 0:
# reset the value to start at Z again
num = num + len(alphabet)
# update the plaintext with the indexed value from alphabet
plaintext = plaintext + alphabet[num]
# otherwise the symbol is not in alphabet i.e. white space
else:
# therefore just append the symbol to the plaintext unchanged
plaintext = plaintext + symbol
# split the plaintext to try identify english words
words = re.split("[^a-zA-Z]+", plaintext)
# look at the first two words in the string
check1 = Word(words[0])
check2 = Word(words[1])
# if they are not english words
if check1.correct is False:
if check2.correct is False:
# print("not in dictionary")
continue
else:
print("FOUND IT!")
# print the value of the key and plaintext
print("The key is: " + str(key) + "\nPlaintext: " + plaintext)
# no need to check other keys
break
body[url] += ' ' + div.text
except:
continue
print(
f'Слов для проверки на странице {url}: {len(get_words(body[url]))}\n')
# #### Проверка
for url in body:
errors = {}
print(url)
for clean_word in get_words(body[url]):
try:
check = Word(clean_word)
if check.correct == False:
if clean_word not in errors:
errors[clean_word] = {}
errors[clean_word]['variants'] = check.variants
errors[clean_word]['count'] = 1
else:
errors[clean_word]['count'] += 1
except Exception as e:
print(f'Что-то пошло не так: {e}, слово: {clean_word}')
continue
pprint(errors)
print('\n')
browser.quit()