1from urllib.request import urlopen
 2import shutil
 3
 4from . import constants
 5from .video import Video
 6
 7
 8def get_subtitles(
 9        video_path: str, lang='eng', time_start='0:00', time_end='',
10        conf_threshold=65, sim_threshold=90, use_fullframe=False) -> str:
11    # download tesseract data file to ~/tessdata if necessary
12    fpath = constants.TESSDATA_DIR / '{}.traineddata'.format(lang)
13    if not fpath.is_file():
14        if lang == 'eng':
15            url = constants.ENG_URL
16        else:
17            url = constants.TESSDATA_URL.format(lang)
18        constants.TESSDATA_DIR.mkdir(parents=True, exist_ok=True)
19        with urlopen(url) as res, open(fpath, 'w+b') as f:
20            shutil.copyfileobj(res, f)
21
22    v = Video(video_path)
23    v.run_ocr(lang, time_start, time_end, conf_threshold, use_fullframe)
24    return v.get_subtitles(sim_threshold)
25
26
27def save_subtitles_to_file(
28        video_path: str, file_path='subtitle.srt', lang='eng',
29        time_start='0:00', time_end='', conf_threshold=65, sim_threshold=90,
30        use_fullframe=False) -> None:
31    with open(file_path, 'w+') as f:
32        f.write(get_subtitles(
33            video_path, lang, time_start, time_end, conf_threshold,
34            sim_threshold, use_fullframe))