prehistoric-systems/the-chomsky-index
1
0
Fork 0
mirror of https://github.com/jasons-gh/the-chomsky-index.git synced 2026-03-11 08:54:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Use pysubs2 library to convert from SRT to CNT for non-repeating files

Browse source
This commit is contained in:
Jason Lobo 2020-10-27 19:55:25 +00:00
parent 023af34d8f
commit e2a1ea1805
1 changed files with 5 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
convert.py
View file

@ -1,5 +1,5 @@
import pathlib
import srt
import pysubs2
from datetime import datetime
from pathlib import Path
@ -43,15 +43,14 @@ for srt_file in [x for x in Path(__file__).parent.glob('**/*.srt') if x.is_file(
i += 1
else:
print(srt_file.name + ' is a nonrepeating file')
subtitle_generator = srt.parse(file_object)
subtitles = list(subtitle_generator)
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.content.replace('\n', ' ')) + '\n')
lines.append(str(i.text.replace('\n', ' ').replace(r'\N', ' ')) + '\n')
times = []
for i in subtitles:
times.append(str(int(i.start.total_seconds())) + '\n')
times.append(str(int(i.start // 1000)) + '\n')
# Make .en.cnt with lines then times