Remove duplicate subtitles in a .srt file

remove-srt-duplications.py

"""Remove duplicate subtitles in a .srt file.

- What does the standard srt file look like?
- See https://www.3playmedia.com/2017/03/08/create-srt-file/
- What does the duplicate subtitle look like?
- The duplicate subtitle is something like the addition of its previous and next subtitle.
- Which match pattern has been used in this script?
- Line 85, 87
- What does the output srt file look like?
- [original_file_name]_changed.srt

Use Case
--------
From

#########################################
22
00:01:47,800 --> 00:01:49,880
Well, you've probably heard the word bit.
23
00:01:49,880 --> 00:01:49,920
# This is a duplicate subtitle.
- We live in the information age
Well, you've probably heard the word bit.
24
00:01:49,920 --> 00:01:51,720
- We live in the information age
#########################################

To

#########################################
22
00:01:47,800 --> 00:01:49,880
Well, you've probably heard the word bit.
23
00:01:49,880 --> 00:01:49,920
24
00:01:49,920 --> 00:01:51,720
- We live in the information age
#########################################
"""
import sys


def entries_generator(srt_file):
    """Generate a entries queue.
    input:
        srt_file: The original filename. [*.srt]
    output:
        entries: A queue generator.
    """
    with open(srt_file, 'r') as srt:
        while True:
            # read lines in order
            number_in_sequence = srt.readline()
            timecode = srt.readline()
            # whether it's the end of the file.
            if not number_in_sequence:
                break
            # put all subtitles seperated by newline into a list.
            subtitles = []
            while True:
                subtitle = srt.readline()
                # whether it's the end of a entry.
                if subtitle == '\n':
                    break
                subtitles.append(subtitle)
            yield number_in_sequence, timecode, subtitles


def replace_duplicationss(entries):
    """Replace the duplicate subtitles with empty string inplace.
    input:
        entries: The entries generator.
    output:
        entries: A modified entries.
    """
    count_for_duplications = 0
    # save the previous subtitles then can be used to make comparsion.
    previous_subtitles = ['']
    for number_in_sequence, timecode, subtitles in entries:
        # whether the present subtitles match the preset pattern.
        # the last line of each other is same.
        if subtitles[-1] == previous_subtitles[-1]:
            # except the normal repeated subtitle.
            if not subtitles == previous_subtitles:
                count_for_duplications += 1
                # used to check out which duplicate subtitles have been
                # detected.
                print('######################################################')
                print(f'Current number in sequence: {number_in_sequence}')
                print(f'Duplicate subtitle:')
                print(f"{''.join(subtitles)}")
                print(f'The previous one:')
                print(f"{''.join(previous_subtitles)}")
                print('######################################################')
                subtitles = ['']
        else:
            previous_subtitles = subtitles

        yield number_in_sequence, timecode, subtitles
    print(f'Total {count_for_duplications} duplications have been fuund out.')


if __name__ == '__main__':
    srt_file = sys.argv[1]
    entries = entries_generator(srt_file)

    with open(srt_file[:-4] + '_changed.srt', 'w') as f:
        for number_in_sequence, timecode, subtitles in replace_duplicationss(entries):
            f.write(number_in_sequence)
            f.write(timecode)
            for subtitle in subtitles:
                f.write(subtitle)
            f.write('\n')