from babelsubs.generators.base import BaseGenerator, register
class TXTGenerator(BaseGenerator):
file_type = 'txt'
def __init__(self, subtitle_set, line_delimiter=u'\n\n', language=None):
"""
Generator is list of {'text': 'text', 'start': 'seconds', 'end': 'seconds'}
"""
self.subtitle_set = subtitle_set
self.line_delimiter = line_delimiter
self.language = language
def __unicode__(self):
output = []
for _, _, content, _ in self.subtitle_set.subtitle_items():
content and output.append(content.strip())
return self.line_delimiter.join(output)
register(TXTGenerator)
},
...
]
"""
file_type = 'json'
MAPPINGS = dict(linebreaks="<br>",
bold="<b>%s</b>",
italics="<i>%s</i>",
underline="<u>%s</u>",
quote_text=escape)
def __unicode__(self):
output = []
# FIXME: allow formatting tags
i = 1
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(
mappings=self.MAPPINGS):
output.append({
'start': from_ms,
'end': to_ms,
'text': content,
'position': i,
'meta': meta
})
i += 1
return json.dumps(output)
register(JSONGenerator)
if len(subtitle_sets) == 0:
raise TypeError(
"DFXPGenerator.merge_subtitles: No subtitles given")
if initial_ttml is None:
tt = SubtitleSet('').as_etree_node()
body = tt.find(TTML + 'body')
body.remove(body.find(TTML + 'div'))
else:
tt = initial_ttml
body = tt.find(TTML + 'body')
if body is None:
raise ValueError("no body tag")
# set the default language to blank. We will create a div for each
# subtitle set and set xml:lang on that.
tt.set(XML + 'lang', '')
# for each subtitle set we will append the body of tt
for i, subtitle_set in enumerate(subtitle_sets):
root_elt = subtitle_set.as_etree_node()
language_code = root_elt.get(XML + 'lang')
lang_div = etree.SubElement(body, TTML + 'div')
lang_div.set(XML + 'lang', language_code)
lang_div.extend(root_elt.find(TTML + 'body').findall(TTML + 'div'))
utils.indent_ttml(tt)
return etree.tostring(tt)
register(DFXPGenerator)
"end": 40300,
"text": "Hello there",
"position": 1
},
...
]
"""
file_type = 'json'
MAPPINGS = dict(linebreaks="<br>", bold="<b>%s</b>",
italics="<i>%s</i>", underline="<u>%s</u>",
quote_text=escape)
def __unicode__(self):
output = []
# FIXME: allow formatting tags
i = 1
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(mappings=self.MAPPINGS):
output.append({
'start': from_ms,
'end': to_ms,
'text': content,
'position': i,
'meta': meta
})
i += 1
return json.dumps(output)
register(JSONGenerator)
def __init__(self, subtitle_set, language=None):
super(WEBVTTGenerator, self).__init__(subtitle_set, language)
self.line_delimiter = '\r\n'
def __unicode__(self):
output = ['WEBVTT\n']
i = 1
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(mappings=self.MAPPINGS):
output.append(unicode(i))
output.append(u'%s --> %s' % (
self.format_time(from_ms),
self.format_time(to_ms)
))
output.append(content)
output.append(u'')
i += 1
return self.line_delimiter.join(output)[:-1]
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i.%03i" % (hours, minutes, seconds, milliseconds)
register(WEBVTTGenerator)
def __init__(self, subtitle_set, language=None):
super(HTMLGenerator, self).__init__(subtitle_set, language)
self.line_delimiter = '\r\n'
def __unicode__(self):
output = []
i = 1
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(mappings=self.MAPPINGS):
output.append(unicode(i))
output.append(u'%s --> %s' % (
self.format_time(from_ms),
self.format_time(to_ms)
))
output.append(content)
output.append(u'')
i += 1
return self.line_delimiter.join(output)
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i,%03i" % (hours, minutes, seconds, milliseconds)
register(HTMLGenerator)
quote_text=escape)
def __init__(self, subtitle_set, language=None):
super(HTMLGenerator, self).__init__(subtitle_set, language)
self.line_delimiter = '\r\n'
def __unicode__(self):
output = []
i = 1
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(
mappings=self.MAPPINGS):
output.append(unicode(i))
output.append(u'%s --> %s' %
(self.format_time(from_ms), self.format_time(to_ms)))
output.append(content)
output.append(u'')
i += 1
return self.line_delimiter.join(output)
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i,%03i" % (hours, minutes, seconds, milliseconds)
register(HTMLGenerator)
def __init__(self, subtitle_set, language=None):
super(SRTGenerator, self).__init__(subtitle_set, language)
self.line_delimiter = '\r\n'
def __unicode__(self):
output = []
i = 1
# FIX ME: allow formatting tags
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(
mappings=self.MAPPINGS):
output.append(unicode(i))
output.append(u'%s --> %s' %
(self.format_time(from_ms), self.format_time(to_ms)))
output.append(content)
output.append(u'')
i += 1
return self.line_delimiter.join(output)
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i,%03i" % (hours, minutes, seconds, milliseconds)
register(SRTGenerator)
"""Combine multiple subtitles sets into a single XML string.
"""
if len(subtitle_sets) == 0:
raise TypeError("DFXPGenerator.merge_subtitles: No subtitles given")
if initial_ttml is None:
tt = SubtitleSet('').as_etree_node()
body = tt.find(TTML + 'body')
body.remove(body.find(TTML + 'div'))
else:
tt = initial_ttml
body = tt.find(TTML + 'body')
if body is None:
raise ValueError("no body tag")
# set the default language to blank. We will create a div for each
# subtitle set and set xml:lang on that.
tt.set(XML + 'lang', '')
# for each subtitle set we will append the body of tt
for i, subtitle_set in enumerate(subtitle_sets):
root_elt = subtitle_set.as_etree_node()
language_code = root_elt.get(XML + 'lang')
lang_div = etree.SubElement(body, TTML + 'div')
lang_div.set(XML + 'lang', language_code)
lang_div.extend(root_elt.find(TTML + 'body').findall(TTML + 'div'))
utils.indent_ttml(tt)
return etree.tostring(tt)
register(DFXPGenerator)
MAPPINGS = dict(linebreaks="[br]")
def __init__(self, subtitles_set, line_delimiter=u'\r\n', language=None):
super(SBVGenerator, self).__init__(subtitles_set, line_delimiter,
language)
def __unicode__(self):
output = []
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(self.MAPPINGS):
start = self.format_time(from_ms)
end = self.format_time(to_ms)
output.append(u'%s,%s' % (start, end))
output.append(content.strip())
output.append(u'')
return self.line_delimiter.join(output)
def format_time(self, time):
if not time:
time = UNSYNCED_TIME_ONE_HOUR_DIGIT
seconds, milliseconds = divmod(int(time), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u'%01i:%02i:%02i.%03i' % (hours, minutes, seconds, milliseconds)
register(SBVGenerator)
else:
components = utils.milliseconds_to_time_clock_components(
milliseconds,
unsynced_val=utils.UNSYNCED_TIME_ONE_HOUR_DIGIT,
use_centiseconds=True)
return u'%(hours)i:%(minutes)02i:%(seconds)02i.%(centiseconds)02i' % components
def _clean_text(self, text):
return text.replace('\n', ' ')
def _content(self):
dl = self.line_delimiter
output = []
output.append(u'[Events]%s' % dl)
output.append(
u'Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text%s'
% dl)
tpl = u'Dialogue: 0,%s,%s,Default,,0000,0000,0000,,%s%s'
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(
self.MAPPINGS):
start = self.format_time(from_ms)
end = self.format_time(to_ms)
text = self._clean_text(content)
output.append(tpl % (start, end, text, dl))
return ''.join(output)
register(SSAGenerator)
def __init__(self, subtitle_set, language=None):
super(SRTGenerator, self).__init__(subtitle_set, language)
self.line_delimiter = '\r\n'
def __unicode__(self):
output = []
i = 1
# FIX ME: allow formatting tags
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(mappings=self.MAPPINGS):
output.append(unicode(i))
output.append(u'%s --> %s' % (
self.format_time(from_ms),
self.format_time(to_ms)
))
output.append(content)
output.append(u'')
i += 1
return self.line_delimiter.join(output)
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i,%03i" % (hours, minutes, seconds, milliseconds)
register(SRTGenerator)
def format_time(self, milliseconds):
if milliseconds is None:
components = utils.unsynced_time_components(one_hour_digit=True, uses_centiseconds=True)
else:
components = utils.milliseconds_to_time_clock_components(
milliseconds,
unsynced_val=utils.UNSYNCED_TIME_ONE_HOUR_DIGIT,
use_centiseconds=True)
return u'%(hours)i:%(minutes)02i:%(seconds)02i.%(centiseconds)02i' % components
def _clean_text(self, text):
return text.replace('\n', ' ')
def _content(self):
dl = self.line_delimiter
output = []
output.append(u'[Events]%s' % dl)
output.append(u'Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text%s' % dl)
tpl = u'Dialogue: 0,%s,%s,Default,,0000,0000,0000,,%s%s'
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(self.MAPPINGS):
start = self.format_time(from_ms)
end = self.format_time(to_ms)
text = self._clean_text(content)
output.append(tpl % (start, end, text, dl))
return ''.join(output)
register(SSAGenerator)
MAPPINGS = dict(linebreaks="[br]")
def __init__(self, subtitles_set, line_delimiter=u'\r\n', language=None):
super(SBVGenerator, self).__init__(subtitles_set, line_delimiter,
language)
def __unicode__(self):
output = []
for from_ms, to_ms, content, meta in self.subtitle_set.subtitle_items(
self.MAPPINGS):
start = self.format_time(from_ms)
end = self.format_time(to_ms)
output.append(u'%s,%s' % (start, end))
output.append(content.strip())
output.append(u'')
return self.line_delimiter.join(output)
def format_time(self, time):
if time is None:
time = UNSYNCED_TIME_ONE_HOUR_DIGIT
seconds, milliseconds = divmod(int(time), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u'%01i:%02i:%02i.%03i' % (hours, minutes, seconds, milliseconds)
register(SBVGenerator)
if sub.new_paragraph:
output.append(u'NOTE Paragraph')
output.append(u'')
output.append(self.format_cue_header(sub))
output.append(sub.text)
output.append(u'')
return self.line_delimiter.join(output)[:-1]
def format_cue_header(self, sub):
parts = []
parts.append(u'%s --> %s' % (
self.format_time(sub.start_time),
self.format_time(sub.end_time)
))
if sub.region == 'top':
parts.append('line:1')
return ' '.join(parts)
def format_time(self, milliseconds):
if milliseconds is None:
milliseconds = UNSYNCED_TIME_FULL
seconds, milliseconds = divmod(int(milliseconds), 1000)
minutes, seconds = divmod(seconds, 60)
hours, minutes = divmod(minutes, 60)
return u"%02i:%02i:%02i.%03i" % (hours, minutes, seconds, milliseconds)
register(WEBVTTGenerator)
