merge feature

[karelv]

Hello,

I was looking for a dual subtitle feature. I didn't found.
But I found this fantastic package, I just would like to share my merge script here.
For me it works perfectly on my VLC player.
Perhaps it might be added to your command line tool.

Usage:
./subtitle_merge.py en.srt de.srt en_de.srt

Best regards,
Karel.

See the code below....

[byroot]

I would be happy to add useful features to the srt command.

But before that I need to understand exactly what your script intend to do.

It build hybrid subtitles with mixed languages, right ?

Can you provide me 2 test subtitles to see it in action ?

[karelv]

Ok, I've to admit that I was too quick in sharing my code.

So yes, indeed, I wanted to have 2 languages.
Currently I'm learning a new language, it is useful to listen and read the language you are learning while having also the subtitles in your mother language...

here is my new code and 2 input files as well as the result.
They are just 2 test - files

subtitle_merge.py

    import argparse
    from pysrt import SubRipFile, SubRipItem, SubRipTime

    parser = argparse.ArgumentParser(description='Merge 2 srt files.')

    parser.add_argument('fin', type=str, nargs=2,
                       help='input file')
    parser.add_argument(dest='fout', type=str, nargs=1,
                       help='the output file')

    args = parser.parse_args() 
    master = args.fin[0]
    slave = args.fin[1]
    result = args.fout[0]

    msubs = SubRipFile.open (master, encoding='iso-8859-1')
    ssubs = SubRipFile.open (slave, encoding='iso-8859-1')
    rsubs = SubRipFile.from_string ("", encoding='iso-8859-1')
    tp = SubRipTime()

    for msub in msubs:
        #"start before, ends before"
        for ssub in ssubs.slice (starts_after=tp.ordinal - 1, ends_before=msub.start.ordinal+1):
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=ssub.end, text=ssub.text))
            ssubs.remove (ssub) # remove as completely handled.

        tp = msub.start

        #"start before; ends before end of master sub"
        for ssub in ssubs.slice (starts_before=msub.start.ordinal, ends_before=msub.end.ordinal):
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=msub.start, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.start, end=ssub.end, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.start, end=ssub.end, text=msub.text))        
            ssubs.remove (ssub) # remove as completely handled.
            tp = ssub.end

        #"start before; ends after end of master sub"
        for ssub in ssubs.slice (starts_before=msub.start, ends_after=msub.end):
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=msub.start, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.start, end=msub.end, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.start, end=msub.end, text=msub.text))        
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.end, end=ssub.end, text=ssub.text))
            ssubs.remove (ssub) # remove as completely handled.
            tp = msub.end


        #"start during; ends before master"
        for ssub in ssubs.slice (starts_after=msub.start.ordinal-1, ends_before=msub.end.ordinal+1):
            if (tp != ssub.start):
                rsubs.append (SubRipItem(index=len(rsubs), start=tp, end=ssub.start, text=msub.text))            
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=ssub.end, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=ssub.end, text=msub.text))
            tp = ssub.end
            ssubs.remove (ssub) # remove as completely handled.

        #"start during; ends after master"
        for ssub in ssubs.slice (starts_before=msub.end, ends_after=msub.end):
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=msub.end, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=msub.end,   end=ssub.end, text=ssub.text))
            rsubs.append (SubRipItem(index=len(rsubs), start=ssub.start, end=msub.end, text=msub.text))
            ssubs.remove (ssub) # remove as completely handled.
            tp = msub.end

        if (tp != msub.end):
            if (tp.ordinal < msub.start.ordinal):
                tp = msub.start
            rsubs.append (SubRipItem(index=len(rsubs), start=tp, end=msub.end, text=msub.text))
            tp = msub.end

    rsubs.sort(key=lambda SubRipItem: SubRipItem.start.ordinal)

    for idx, rsub in enumerate(rsubs):
        rsub.index = idx
    rsubs.save (result, encoding='iso-8859-1')

master.srt

1
00:00:01,000 --> 00:00:03,000
Master sub title!

2
00:00:10,000 --> 00:00:15,000
First master sub.

3
00:00:20,000 --> 00:00:22,000
2nd master sub.

4
00:00:30,000 --> 00:00:40,000
3rd master sub.

5
00:00:45,000 --> 00:01:00,000
4th master sub.

slave.srt

1
00:00:01,000 --> 00:00:03,000
Slave sub title!

2
00:00:04,000 --> 00:00:05,500
First slave sub before 1st msub.

3
00:00:06,500 --> 00:00:08,000
Second slave sub before 1st msub.

4
00:00:09,000 --> 00:00:10,000
Third slave sub until 1st msub.

5
00:00:15,000 --> 00:00:17,000
Fourth slave sub right after 1st msub.

6
00:00:18,000 --> 00:00:24,000
Fifth slave sub @ before and after 2nd msub.

7
00:00:44,000 --> 00:00:46,000
Sixth slave sub @ 4th msub entry.

8
00:00:47,000 --> 00:00:48,000
Seventh slave sub during 4th msub.

9
00:00:49,000 --> 00:00:50,000
Eighth slave sub during 4th msub.

After ./subtitle_merge.py master.srt slave.srt merge.srt you get:

merge.srt

0
00:00:01,000 --> 00:00:03,000
Slave sub title!

1
00:00:01,000 --> 00:00:03,000
Master sub title!

2
00:00:04,000 --> 00:00:05,500
First slave sub before 1st msub.

3
00:00:06,500 --> 00:00:08,000
Second slave sub before 1st msub.

4
00:00:09,000 --> 00:00:10,000
Third slave sub until 1st msub.

5
00:00:10,000 --> 00:00:15,000
First master sub.

6
00:00:15,000 --> 00:00:17,000
Fourth slave sub right after 1st msub.

7
00:00:18,000 --> 00:00:20,000
Fifth slave sub @ before and after 2nd msub.

8
00:00:20,000 --> 00:00:22,000
Fifth slave sub @ before and after 2nd msub.

9
00:00:20,000 --> 00:00:22,000
2nd master sub.

10
00:00:22,000 --> 00:00:24,000
Fifth slave sub @ before and after 2nd msub.

11
00:00:30,000 --> 00:00:40,000
3rd master sub.

12
00:00:44,000 --> 00:00:45,000
Sixth slave sub @ 4th msub entry.

13
00:00:45,000 --> 00:00:46,000
Sixth slave sub @ 4th msub entry.

14
00:00:45,000 --> 00:00:46,000
4th master sub.

15
00:00:46,000 --> 00:00:47,000
4th master sub.

16
00:00:47,000 --> 00:00:48,000
Seventh slave sub during 4th msub.

17
00:00:47,000 --> 00:00:48,000
4th master sub.

18
00:00:48,000 --> 00:00:49,000
4th master sub.

19
00:00:49,000 --> 00:00:50,000
Eighth slave sub during 4th msub.

20
00:00:49,000 --> 00:00:50,000
4th master sub.

21
00:00:50,000 --> 00:01:00,000
4th master sub.

[karelv]

And as you can see I had a quick look to the flavors of git hub! :-)

[byroot]

And can you tell me if this simpler version is working for you ?

import argparse
from pysrt import SubRipFile, SubRipItem, SubRipTime

parser = argparse.ArgumentParser(description='Merge 2 srt files.')

parser.add_argument('fin', type=str, nargs=2,
                  help='input file')
parser.add_argument(dest='fout', type=str, nargs=1,
                  help='the output file')

args = parser.parse_args() 
master, slave = args.fin
result = args.fout[0]

msubs = SubRipFile.open(master, encoding='iso-8859-1')
ssubs = SubRipFile.open(slave, encoding='iso-8859-1')

rsubs = msubs + ssubs
rsubs.sort()
rsubs.clean_indexes()
rsubs.save (result, encoding='iso-8859-1')

[karelv]

This is what I did first, or at least something similar...., I'm relatively new to python, so I was adding one by one instead of '+', the sort I did like my shared code, as well as the 'clean_indexes'.

I was not happy with the output since I always get the a mixture: sometime first master then slave subtitles,
sometime the opposite,
sometime it needs up to 6 lines to display all the subtitles.

I like your solution more, it has cleaner code, it executes faster, but I prefer my version for its output.

Thanks for the challenge anyhow!

PS. I tried also your version, it suffer from the same issue, as my very initial version, at least in VLC player 2.0.1.