Converts the currently set .ass subtitle to .srt and sets it as the subtitle for the current video in XBMC for Xbox

gistfile1.txt

"""
    ass2srt.py
    Converts the currently set .ass subtitle to .srt and sets it as the subtitle for the current video
"""
import re
import xbmc
from os.path import basename, splitext
from datetime import time
import traceback
import codecs

encoding='utf-8'
try:
    from chardet.universaldetector import UniversalDetector
except ImportError:
    UniversalDetector=False
    print 'ass2srt.py: could not load UniversalDetector, defaulting encoding to %s' % encoding

class SubripLine:

    FORMAT=None
    
    @classmethod
    def set_format(cls,line):
        """Sets static FORMAT dictionary
           Raises if the fields do not contain:
           Start
           End
           Text
        """
        print 'ass2srt.py: field is %s' % line
        fields=[field.strip() for field in line.split(',')]
        if not set(['Start', 'End', 'Text']) <= set(fields): #if (start, end, text) not a subset of fields
            raise ValueError('ass2srt.py: Fields do not contain Start, End, or Text')
        cls.FORMAT=dict(zip(fields, range(0,len(fields))))
        print 'ass2srt.py: Format is %s' % cls.FORMAT        
    
    @staticmethod
    def create_time_from_timestamp(timestamp):
        """Parses timestamp and returns a datetime.time object
           Example timestamp: 0:20:46.090
        """
        m=[int(digit) for digit in re.findall(r'(\d):(\d\d):(\d\d)\.(\d\d)',timestamp)[0]]
        return time(m[0], m[1], m[2], microsecond=m[3]*1000)
    
    def __init__(self, line):
        FORMAT=self.__class__.FORMAT
        line=re.sub(r'{.*?}','',line) #strip formatting
        line=re.sub(r'\\N','\n',line) #convert newlines to real newline
        line=re.sub(r'\\.','',line) #strip control characters
        #need to limit splits or else the commas in the dialogue will also be split. -1 needed as split limits to maxsplit+1
        tokens=[token.strip() for token in line.split(',', len(FORMAT)-1)]
        self.text=tokens[FORMAT['Text']]
        self.start_time=self.create_time_from_timestamp(tokens[FORMAT['Start']])
        self.end_time=self.create_time_from_timestamp(tokens[FORMAT['End']])

    def __str__(self):
        """Returns string in format of:
           start_time --> end_time\ntext\n
           ex.
           00:00:02,500 -->  00:00:04,970
           Fairy Tail
           
           does not add index
        """
        start_time=self.start_time.strftime('%H:%M:%S,') + \
                    str('%03d' % (self.start_time.microsecond/100))
        end_time=self.end_time.strftime('%H:%M:%S,') + \
                    str('%03d' % (self.end_time.microsecond/100))
        return '%s -->  %s\n%s\n' % (start_time, end_time, self.text)

ass_subtitle='special://temp/%s' % xbmc.Player().getSubtitles()
OUTPUT='Z:\\sp_%s.srt' % splitext(basename(ass_subtitle))[0]
# ass_subtitle='C:\\Users\\Viqar Samad\\Desktop\\gogakumono\\subs\\yukimeimeri.ass'
# OUTPUT='C:\\Users\\Viqar Samad\\Desktop\\%s.srt' % splitext(basename(ass_subtitle))[0]
print 'ass2srt.py: converting %s' % ass_subtitle

if UniversalDetector:
    f=open(ass_subtitle, 'r')
    detector = UniversalDetector()
    for line in f:
        detector.feed(line)
        if detector.done: break
    detector.close()
    f.close()
    encoding=detector.result['encoding']
    print 'ass2srt.py: determined encoding to be %s' % encoding

try:
    f=codecs.open(ass_subtitle, 'r' , encoding)
except LookupError:
    encoding='utf-8' #most likely choice
    f=codecs.open(ass_subtitle, 'r' , encoding)

line=f.readline()

while len(line) != 0 and line.strip() != '[Events]':
    line=f.readline()
if len(line) == 0:
    raise EOFError('ass2srt.py: reached eof, did not find [Events]')

line=f.readline().strip() #line after [Events] containing format
SubripLine.set_format(line)

subtitles=[]
line=f.readline()#main dialogue
while len(line) != 0 and not line.startswith('['): #not end of file or not the start of another section
    line.strip()#strip newline, can't do it before loop or else length will be 0 for a blank line
    if not line.startswith('Dialogue'):
        print 'ass2srt.py: skipping line %r' % line
        line=f.readline()
        continue
    try:
        subtitles.append(SubripLine(line))
    except:
        err='ass2srt.py: failure occured'
        if len(subtitles)>0:
            err+=', previous line \n%s' % subtitles[-1]
        print err
        traceback.print_exc()
    line=f.readline()
f.close()
print 'ass2srt.py: finished parsing'

subtitles.sort(cmp=lambda a,b: cmp(a.start_time,b.start_time)) #ass subtitles are not guaranteed  to be chronologically sorted
print 'ass2srt.py: sorted subtitles'

srt_subs='\n'.join(['%d\n%s' % (i+1, line) for i, line in enumerate(subtitles)])
print 'ass2srt.py: printing to %s; length:%d' % (OUTPUT, len(srt_subs))

subfile=codecs.open(OUTPUT, 'w', 'utf-8', buffering=0)
subfile.write(srt_subs)
subfile.close()

xbmc.Player().setSubtitles(OUTPUT)