Building a mad libs program in Python using regular expressions - python-3.x

Question

I'm a new Python programmer working through the book Automate the Boring Stuff with Python. One of the end-of-chapter projects is to build a mad libs program. Based on what has been introduced so far, I think that the author intends for me to use regular expressions.
Here is my code:
#! python3
#
# madlibs.py - reads a text file and let's the user add their own text
# anywhere the words ADJECTIVE, NOUN, ADVERB, or VERB appear in the text
# file.
import sys, re, copy
# open text file, save text to variable
if len(sys.argv) == 2:
print('Opening text file...')
textSource = open(sys.argv[1])
textContent = textSource.read()
textSource.close()
else:
print('Usage: madlibs.py <textSource>')
# locate instances of keywords
keywordRegex = re.compile(r'ADJECTIVE|NOUN|ADVERB|VERB', re.I)
matches = keywordRegex.findall(textContent)
# prompt user to replace keywords with their own input
answers = copy.copy(matches)
for i in range(len(answers)):
answers[i] = input()
# create a new text file with the end result
for i in range(len(matches)):
findMatch = re.compile(matches[i])
textContent = findMatch.sub(answers[i], textContent)
print(textContent)
textEdited = open('madlibbed.txt', 'w')
textEdited.write(textContent)
textEdited.close()
The input I'm using for textSource is a text file that reads:
This is the test source file. It has the keyword ADJECTIVE in it, as well as the keyword NOUN. Also, it has another instance of NOUN and then one of ADVERB.
My problem is that the findMatch.sub method is replacing both of the instances of NOUN at once. I understand that this is how the sub() method works, but I'm having trouble thinking of a simple way to work around it. How can I design this program so that it only targets and replaces one keyword at a time? I don't want all NOUNS to be replaced with the same word, but rather different words respective to the order in which the user types them.

Answer 1

All you need is to set a keyword argument count to sub, so that it will replace no more occurrences then you set.
textContent = findMatch.sub(answers[i], textContent, count=1)
for more details, see https://docs.python.org/3/library/re.html#re.sub

Answer 2

The thodnev's answer works, however you sometimes are better off by tokenizing the string first, and then building a new string with the parts.
If your string is:
textContent = 'This is the test source file. It has the keyword ADJECTIVE in it, as well as the keyword NOUN. Also, it has another instance of NOUN and then one of ADVERB.'
then you can use a re.finditer to do this:
for it in re.finditer(r'ADJECTIVE|NOUN|ADVERB|VERB', textContent):
print(it.span(), it.group())
gives
(49, 58) ADJECTIVE
(89, 93) NOUN
(128, 132) NOUN
(149, 155) ADVERB
You can use this information with substring to build a new string the way you want.