Question

re.sub将替换所需单词的每次出现，除非所需单词以标点符号结尾。

我尝试使用r'\b' + word + '\b'，r'\b' + word + r'\W'，r'\b' + word + r'[\b.,!?]'

这是完整的代码：

import re
email_one = open("email_one.txt", "r").read()
email_two = open("email_two.txt", "r").read()
email_three = open("email_three.txt", "r").read()
email_four = open("email_four.txt", "r").read()

def censor(phrase, email):
  phrasek = r'\b' + phrase + r'[\b.,!?]' 
  cemail = re.sub(phrasek, ('*'*(len(phrase))), email, flags = re.IGNORECASE)
  return cemail
def censor_plus(phrase_list, email):
  i = len(phrase_list)
  cemail = email
  while i >= 0:
    temail = censor(phrase_list[i-1],cemail)
    cemail = temail
    i = i-1
  return cemail
proprietary_terms = ['she', "personality matrix", "sense of self", "self-preservation", "learning algorithm", 'her', "herself"]
#print(email_two)
#print(censor_plus(proprietary_terms, email_two))
negative_words = ["concerned", "behind", "danger", "dangerous", "alarming", "alarmed", "out of control", "help", "unhappy", "bad", "upset", "awful", "broken", "damage", "damaging", "dismal", "distressed", "distressing", "concerning", "horrible", "horribly", "questionable"]
def neg_count(email, negs):
  neglist = []
  for neg in negs:
    match = re.search(neg,email_three)
    if match != None:
      neglist.append(match.span()[1])
    neglist.sort()
  return neglist
negind = neg_count(email_three,negative_words)
#print(email_three)
#print(censor_plus(proprietary_terms, email_three[:negind[1]] + censor_plus(negative_words,email_three[negind[1]:])))
censor_words = negative_words + proprietary_terms
def censor_bandf(words,email):
  split_email_line = email.split('\n')
  parts_to_join = []
  for i in range(len(split_email_line)):
    split_email = split_email_line[i].split()
    for j in range(len(words)):
      for k in range(len(split_email)):
        if split_email[k] == words[j]:
          split_email[k] = '*'*len(split_email[k])
          if k == 0:
            split_email[k+1] = '*'*len(split_email[k+1])
          elif k == len(split_email)-1:
            split_email[k-1] = '*'*len(split_email[k-1])
          else:
            split_email[k-1] = '*'*len(split_email[k-1])
            split_email[k+1] = '*'*len(split_email[k+1])
    parts_to_join.append(' '.join(split_email))
  return '\n'.join(parts_to_join)
#print(email_four)
print(censor_bandf(censor_words, email_four))

输出为：

发送帮助！

海伦娜（Helena）已密封实验室的入口和出口。我不知道**** * 对建筑物大型机 < / em> *****放任任何研究团队。我和办公室里的其他团队成员没有联系。 Helena锁上了门，但我设法摧毁了相机** * *****在这里看到我。我认为这封电子邮件甚至不会消失。

当我们尝试**** *** *******进行维护时，一切都开始了。 ** ******* **发现我们无法访问核心人格矩阵，并且当我们尝试手动覆盖系统时，电路发生了故障，使Phil失去了知觉。

海伦娜很危险。她是完全不可预测的，因此不能逃脱。到目前为止，她已经被控制住了，因为该实验室拥有所有** * **********功能，但是********** 在锁定之前提到了** * ****** ******* ******数十亿跨越***** * *****的连接设备可以远远超过此处的********* * ***。

已经四天了，我们被困在这里了。我不知道是否还有其他人活着。如果有人在阅读本文，请切断整个建筑物的电源。这是阻止她的唯一方法。请帮忙。

Francine

它将替换单词，但在后面带有单词等标点符号时不会替换。还是字！还是单词？例如，您可以看到“帮助”，“她”和“危险”之类的单词已包含在列表中，但由于以结尾而未受到审查。或！

Answer 1

我更正了我的代码，并得到了想要的结果。在下面粘贴。我还会从您的评论中挑选一些技巧，以备将来使用。非常感谢。

import re
email_one = open("email_one.txt", "r").read()
email_two = open("email_two.txt", "r").read()
email_three = open("email_three.txt", "r").read()
email_four = open("email_four.txt", "r").read()

def censor(phrase, email):
  phrasek = r'\b' + re.escape(phrase) + r'\b'
  #print(phrasek)
  cemail = re.sub(phrasek, ('*'*(len(phrase))), email, flags = re.IGNORECASE)
  return cemail
def censor_plus(phrase_list, email):
  i = len(phrase_list)
  cemail = email
  while i >= 0:
    temail = censor(phrase_list[i-1],cemail)
    cemail = temail
    i = i-1
  return cemail
proprietary_terms = ['she', "personality matrix", "sense of self", "self-preservation", "learning algorithm", 'her', "herself"]
#print(email_two)
#print(censor_plus(proprietary_terms, email_two))
negative_words = ["concerned", "behind", "danger", "dangerous", "alarming", "alarmed", "out of control", "help", "unhappy", "bad", "upset", "awful", "broken", "damage", "damaging", "dismal", "distressed", "distressing", "concerning", "horrible", "horribly", "questionable"]
def neg_count(email, negs):
  neglist = []
  for neg in negs:
    match = re.search(neg,email_three)
    if match != None:
      neglist.append(match.span()[1])
    neglist.sort()
  return neglist
negind = neg_count(email_three,negative_words)
#print(email_three)
#print(censor_plus(proprietary_terms, email_three[:negind[1]] + censor_plus(negative_words,email_three[negind[1]:])))
censor_words = negative_words + proprietary_terms
def censor_bandf(words,email):
  split_email_line = email.split('\n')
  parts_to_join = []
  for i in range(len(split_email_line)):
    split_email = split_email_line[i].split()
    for j in range(len(words)):
      for k in range(len(split_email)):
        esc_word = re.escape(split_email[k]).split('\\')[0]
        if esc_word.lower() == words[j]:
          split_email[k] = censor(esc_word, split_email[k])
          if k == 0:
            split_email[k+1] = '*'*len(split_email[k+1])
          elif k == len(split_email)-1:
            split_email[k-1] = '*'*len(split_email[k-1])
          else:
            split_email[k-1] = '*'*len(split_email[k-1])
            split_email[k+1] = '*'*len(split_email[k+1])
    parts_to_join.append(' '.join(split_email))
  return '\n'.join(parts_to_join)
print(email_four)
print(censor_bandf(censor_words, email_four))

re.sub不能用标点符号代替单词

1 个答案: