convert.py

import pysubs2
from datetime import datetime
from pathlib import Path


# Files auto-generated by YT have each subtitle line repeating 3 times, once at
# the end of a pair, once by itself and once at the beginning of a pair.
# But sometimes only the first case (end of a pair) exists, which are lines
# 3 + 10i.


def convert():
    for srt_file in [x for x in Path(__file__).parent.glob('**/*.srt') if x.is_file()]:
        with open(srt_file, 'r', encoding='utf-8') as file_object:
            contents = file_object.readlines()
    
            # Check for repeating lines from auto-generated subtitles (use lines 3+10i)
            # or non-repeating subtitles from a manual upload (parse using srt library)
    
            if (
                len(contents) > 40 and
                ((contents[3].strip() != '' and contents[3] == contents[7]
                                            and contents[7] == contents[12]) or
                (contents[13].strip() != '' and contents[13] == contents[17]
                                            and contents[17] == contents[22]) or
                (contents[23].strip() != '' and contents[23] == contents[27]
                                            and contents[27] == contents[32]) or
                (contents[33].strip() != '' and contents[33] == contents[37]
                                            and contents[37] == contents[42]))
                ):
                print(srt_file.name + ' is a repeating file')

                lines = []
                i = 0
                while 3+10*i < len(contents):
                    lines.append(contents[3+10*i])
                    i += 1

                times = []
                i = 0
                while i < len(lines):
                    t = datetime.strptime(contents[1+10*i][:8], '%H:%M:%S')
                    times.append(str(t.hour*3600 + t.minute*60 + t.second) + '\n')
                    i += 1

            else:
                print(srt_file.name + ' is a nonrepeating file')
                subtitles = pysubs2.load(srt_file, encoding="utf-8")
                lines = []
                for i in subtitles:
                    lines.append(str(i.text.replace('\n', ' ').replace(r'\N', ' ')) + '\n')
                times = []
                for i in subtitles:
                    times.append(str(int(i.start // 1000)) + '\n')

            # Make .en.cnt with lines then times

            with open(str(Path(__file__).parent) + '\\' + srt_file.name[:-7] + '.en.cnt', 'a') as cnt_file:
                for i in lines:
                    cnt_file.write(i)

            with open(str(Path(__file__).parent) + '\\' + srt_file.name[:-7] + '.en.cnt', 'a') as cnt_file:
                for i in times:
                    cnt_file.write(str(i))