Filter repeated captions on SRT files

srt_unique.py

# pip install srt
import argparse

import srt


parser = argparse.ArgumentParser()
parser.add_argument("input_filename")
parser.add_argument("output_filename")
args = parser.parse_args()

with open(args.input_filename) as fobj:
    new_subtitles = []
    last_text, start, last_stop, counter = None, None, None, 1
    for caption in srt.parse(fobj):
        text = caption.content
        if last_text is None or last_text != text:
            if last_text != text and last_text is not None:
                new_subtitles.append(
                    srt.Subtitle(index=counter, start=start, end=last_stop, content=last_text)
                )
                counter += 1
                last_printed = (start, last_stop, last_text)
            start = caption.start
            last_text = text
            last_stop = caption.end
        elif last_text == text:
            last_stop = caption.end
    if last_printed != (start, last_stop, last_text):
        new_subtitles.append(
            srt.Subtitle(index=counter, start=start, end=last_stop, content=last_text)
        )
        counter += 1
with open(args.output_filename, mode="w") as fobj:
    fobj.write(srt.compose(new_subtitles))