Fuzzy String Matching and Removal

fuzzy_remove.py

# fuzzy_removal will scan through the text and replace the text with replacement if the text is similar to the target
# default replacement is <Patient Name>
# default similarity score is 80
def fuzzy_removal(
    text: str,
    target: str,
    replacement: str='<Patient Name>',
    min_score: int=80,
    debug: bool=False,
) -> str:
    final_texts = ''
    target_length = len(target.split(' '))

    words = text.split(' ')
    full_length = len(words)

    # iterate through texts list and form a string of words with length equal to target_length plus 2
    # compare the string with target_name and print match if matched
    i: int = 0
    if debug:
        print("text=%s" % text)
        print("Full length = %d" % full_length)
    final_texts = text
    while i < full_length:
        breakout = False
        for increment_length in range(0, target_length + 2):
            final_pos = i + increment_length
            if final_pos > full_length:
                final_pos = full_length
            text = ' '.join(words[i:final_pos])
            score = fuzz.ratio(text, target)
            if debug:
                print("comparing (%s) with (%s), score = %d" % (text, target, score))
            if score > min_score:
                if debug:
                    print('Match: %s' % text)
                i = final_pos
                final_texts = final_texts.replace(text, replacement)

                # this is to break out of the for loop and continue with the while loop
                breakout = True
                break

        if breakout:
            continue

        i += 1

    return final_texts