69 lines
1.8 KiB
Python
69 lines
1.8 KiB
Python
|
import pysrt
|
||
|
|
from pysubparser import parser
|
||
|
|
import datetime
|
||
|
|
from datetime import datetime
|
||
|
|
import time
|
||
|
|
from itertools import groupby
|
||
|
|
|
||
|
|
#subtitles = parser.parse("/home/user/mnf/project/MNF/media/videos/Subtitling_Videos/german/azure_subtitle.srt")
|
||
|
|
srt_data = pysrt.open("/home/user/mnf/project/MNF/media/videos/Subtitling_Videos/german/azure_subtitle.srt")
|
||
|
|
#print(srt_data)
|
||
|
|
subs = []
|
||
|
|
|
||
|
|
|
||
|
|
for i in srt_data:
|
||
|
|
|
||
|
|
if i.start.minutes == 0:
|
||
|
|
text = i.text
|
||
|
|
start = i.start.seconds
|
||
|
|
end = i.end.seconds
|
||
|
|
#print(type(start))
|
||
|
|
subs.append(((start, end), text))
|
||
|
|
else:
|
||
|
|
text = i.text
|
||
|
|
start_s = i.start.seconds
|
||
|
|
print(type(start_s))
|
||
|
|
start_m = i.start.minutes
|
||
|
|
start = str(start_m) + "." + str(start_s)
|
||
|
|
start = start_m + start_s/60
|
||
|
|
print(type(start))
|
||
|
|
end_s = i.end.seconds
|
||
|
|
end_m = i.end.minutes
|
||
|
|
end = end_m + end_s/60
|
||
|
|
print(type(start))
|
||
|
|
subs.append(((float(start), float(end)), text))
|
||
|
|
|
||
|
|
print(subs)
|
||
|
|
|
||
|
|
"""
|
||
|
|
|
||
|
|
with open("/home/user/mnf/project/MNF/media/videos/Subtitling_Videos/german/azure_subtitle.srt") as f:
|
||
|
|
|
||
|
|
res = [list(g) for b,g in groupby(f, lambda x: bool(x.strip())) if b]
|
||
|
|
|
||
|
|
#print(res)
|
||
|
|
|
||
|
|
from collections import namedtuple
|
||
|
|
|
||
|
|
Subtitle = namedtuple('Subtitle', 'number start end content')
|
||
|
|
|
||
|
|
subs = []
|
||
|
|
|
||
|
|
texting_list = []
|
||
|
|
for sub in res:
|
||
|
|
if len(sub) >= 3: # not strictly necessary, but better safe than sorry
|
||
|
|
sub = [x.strip() for x in sub]
|
||
|
|
number, start_end, *content = sub # py3 syntax
|
||
|
|
start, end = start_end.split(' --> ')
|
||
|
|
start_1 = start.split(',')[0]
|
||
|
|
end_1 = end.split(',')[0]
|
||
|
|
start = start_1
|
||
|
|
end = end_1
|
||
|
|
start_n = time.strptime(start,"%H:%M:%S")
|
||
|
|
end_n = time.strptime(end,"%H:%M:%S")
|
||
|
|
texting_list.append(((start_n,end_n), content[0]))
|
||
|
|
subs.append(Subtitle(number, start, end, content))
|
||
|
|
|
||
|
|
print(texting_list)
|
||
|
|
|
||
|
|
"""
|