How do I replace words (in a txt file) that match my list of strings?

Question

• I am looking to understand what it takes to replace certain words from my .TXT file.
  • Those certain words would be strings in a censored_words list.

I was successful with a single-word replacement:

email = open('email.txt', 'r').read()

def single_string_replace(email):
    return email.replace('word1', 'REDACTED')

But I could not get a list of words to work "flawlessly". This is my attempt:

email = open('email.txt', 'r').read()
banned_words = ['word1', 'phrase one']

def list_replace(email):
    list_place = 0
    while list_place < len(banned_words):
        for word in banned_words:
            email = email.replace(word, 'REDACTED')
            list_place += 1
        return email

I am optimally looking to keep the .TXT files unchanged, and only seeing the changes by a print() statement such as

print(list_replace(email))

The issue that I am having is:

• Let's say I ban a word such as dog, and also ban the word hotdog. If 'dog' is first up in the list, then when 'hotdog' is being searched for ban, it doesn't find any.
• This creates 'hotREDACTED' instead of 'REDACTED'.
• Vice versa as well. If I want the word dog banned, but hotdog is okay, how can I make sure both of these above cases work without kinks?

As always has been, is, and shall be: all suggestions are welcome!

Thank you

Answer 1

you could use re.sub:

import re


email = open('email.txt', 'r').read()
banned_words = ['word1', 'phrase one']
pattern = '|'.join(f'\\b{w}\\b' for w in banned_words)

def list_replace(email):
    return re.sub(pattern, 'REDACTED', email)

print(list_replace(email))

Answer 2

Here is a function that replaces words. One could change the swlist in the function to add or delete more such stop words.

---

Function to replace text

from nltk import word_tokenize

def mask_word(with_sw):
    swlist = ['dog','cat']
    without_sw = ""
    char = 'nan'
    tokens = word_tokenize(with_sw)
    for char in tokens:
        if char in swlist:
            without_sw = without_sw + " " + "REDACTED"
        else:
            without_sw = without_sw + " " + char

    return(without_sw)    

An example usage is below

text = "this is a dog and hotdog test"

print(mask_word(text))

Output looks like this:

this is a REDACTED and hotdog test

Answer 3

Try it in this way

words = open('email.txt').read().split() #to get a list of words
words = [word.replace('\n','') for word in words] #removing all newlines if any
censored_words = ['ADD', 'YOUR', 'WORDS', 'HERE']

for word in words:
    if word in censored_words:
        print(word) #printing all the occurences of censored words