(match.group(1), match.group(2).upper()),
name="CM_uppercase_after_dot"),
# remove spaces before punctuation
NReProcessor(re.compile(r'(?u)(?:(?<=^)|(?<=\w)) +([!?.,](?![!?.,]))'),
r"\1",
name="CM_punctuation_space"),
]
post_processors = empty_line_post_processors
class RemoveTags(SubtitleModification):
identifier = "remove_tags"
description = "Remove all style tags"
exclusive = True
modifies_whole_file = True
long_description = """\
Removes all possible style tags from the subtitle, such as font, bold, color etc.
"""
def modify(self, content, debug=False, parent=None, **kwargs):
for entry in parent.f:
# this actually plaintexts the entry and by re-assigning it to plaintext, it replaces \n with \N again
entry.plaintext = entry.plaintext
registry.register(CommonFixes)
registry.register(RemoveTags)
# coding=utf-8
import logging
from subzero.modification.mods import SubtitleModification
from subzero.modification import registry
logger = logging.getLogger(__name__)
class ChangeFPS(SubtitleModification):
identifier = "change_FPS"
description = "Change the FPS of the subtitle"
exclusive = True
advanced = True
modifies_whole_file = True
long_description = """\
Re-syncs the subtitle to the framerate of the current media file.
"""
def modify(self, content, debug=False, parent=None, **kwargs):
fps_from = kwargs.get("from")
fps_to = kwargs.get("to")
parent.f.transform_framerate(float(fps_from), float(fps_to))
registry.register(ChangeFPS)
("dark-red", "#800000"),
("dark-green", "#008000"),
("dark-yellow", "#808000"),
("dark-blue", "#000080"),
("dark-magenta", "#800080"),
("dark-cyan", "#008080"),
("dark-grey", "#808080"),
])
class Color(SubtitleModification):
identifier = "color"
description = "Change the color of the subtitle"
exclusive = True
advanced = True
colors = COLOR_MAP
long_description = """\
Adds the requested color to every line of the subtitle. Support depends on player.
"""
def modify(self, content, debug=False, parent=None, **kwargs):
color = self.colors.get(kwargs.get("name"))
if color:
return u'<font color="%s">%s</font>' % (color, content)
return content
registry.register(Color)
class ShiftOffset(SubtitleModification):
identifier = "shift_offset"
description = "Change the timing of the subtitle"
exclusive = False
advanced = True
args_mergeable = True
modifies_whole_file = True
long_description = "Adds or substracts a certain amount of time from the whole subtitle to match your media"
@classmethod
def merge_args(cls, args1, args2):
new_args = dict((key, int(value)) for key, value in args1.iteritems())
for key, value in args2.iteritems():
if key in new_args:
new_args[key] += int(value)
else:
new_args[key] = int(value)
return new_args
def modify(self, content, debug=False, parent=None, **kwargs):
parent.f.shift(h=int(kwargs.get("h", 0)),
m=int(kwargs.get("m", 0)),
s=int(kwargs.get("s", 0)),
ms=int(kwargs.get("ms", 0)))
registry.register(ShiftOffset)
identifier = "shift_offset"
description = "Change the timing of the subtitle"
exclusive = False
advanced = True
args_mergeable = True
modifies_whole_file = True
long_description = "Adds or substracts a certain amount of time from the whole subtitle to match your media"
@classmethod
def merge_args(cls, args1, args2):
new_args = dict((key, int(value)) for key, value in args1.iteritems())
for key, value in args2.iteritems():
if not int(value):
continue
if key in new_args:
new_args[key] += int(value)
else:
new_args[key] = int(value)
return dict(filter(lambda (k, v): bool(v), new_args.iteritems()))
def modify(self, content, debug=False, parent=None, **kwargs):
parent.f.shift(h=int(kwargs.get("h", 0)), m=int(kwargs.get("m", 0)), s=int(kwargs.get("s", 0)),
ms=int(kwargs.get("ms", 0)))
registry.register(ShiftOffset)
name="CM_RTL_reverse")
]
split_upper_re = re.compile(ur"(\s*[.!?♪\-]\s*)")
class FixUppercase(SubtitleModification):
identifier = "fix_uppercase"
description = "Fixes all-uppercase subtitles"
modifies_whole_file = True
exclusive = True
order = 41
only_uppercase = True
apply_last = True
long_description = "Some subtitles are in all-uppercase letters. This at least makes them readable."
def capitalize(self, c):
return u"".join([s.capitalize() for s in split_upper_re.split(c)])
def modify(self, content, debug=False, parent=None, **kwargs):
for entry in parent.f:
entry.plaintext = self.capitalize(entry.plaintext)
registry.register(CommonFixes)
registry.register(RemoveTags)
registry.register(ReverseRTL)
registry.register(FixUppercase)
re.compile(ur'(?u)(^.*(?<=[A-ZÀ-Ž]{2})[A-ZÀ-Ž-_\s0-9"\']+:\s*)'),
"",
name="HI_before_colon"),
# text in brackets at start, after optional dash, before colon or at end of line
# fixme: may be too aggressive
NReProcessor(re.compile(
ur'(?um)(^-?\s?[([][A-zÀ-ž-_\s]{3,}[)\]](?:(?=$)|:\s*))'),
"",
name="HI_brackets_special"),
# all caps line (at least 4 consecutive uppercase chars)
NReProcessor(re.compile(ur'(?u)(^(?=.*[A-ZÀ-Ž]{4,})[A-ZÀ-Ž-_\s]+$)'),
"",
name="HI_all_caps"),
# dash in front
# NReProcessor(re.compile(r'(?u)^\s*-\s*'), "", name="HI_starting_dash"),
# all caps at start before new sentence
NReProcessor(re.compile(
ur'(?u)^(?=[A-ZÀ-Ž]{4,})[A-ZÀ-Ž-_\s]+\s([A-ZÀ-Ž][a-zà-ž].+)'),
r"\1",
name="HI_starting_upper_then_sentence"),
]
post_processors = empty_line_post_processors
registry.register(HearingImpaired)
data_dict = OCR_fix_data.get(parent.language.alpha3t)
if not data_dict:
logger.debug("No SnR-data available for language %s",
parent.language)
return
self.data_dict = data_dict
self.processors = self.get_processors()
def get_processors(self):
if not self.data_dict:
return []
return [
WholeLineProcessor(self.data_dict["WholeLines"],
name="SE_replace_line"),
MultipleWordReProcessor(self.data_dict["WholeWords"],
name="SE_replace_word"),
MultipleWordReProcessor(self.data_dict["BeginLines"],
name="SE_replace_beginline"),
MultipleWordReProcessor(self.data_dict["EndLines"],
name="SE_replace_endline"),
MultipleWordReProcessor(self.data_dict["PartialLines"],
name="SE_replace_partialline"),
MultipleLineProcessor(self.data_dict["PartialWordsAlways"],
name="SE_replace_partialwordsalways")
]
registry.register(FixOCR)
def modify(self, content, debug=False, parent=None, **kwargs):
for entry in parent.f:
# this actually plaintexts the entry and by re-assigning it to plaintext, it replaces \n with \N again
entry.plaintext = entry.plaintext
class ReverseRTL(SubtitleModification):
identifier = "reverse_rtl"
description = "Reverse punctuation in RTL languages"
exclusive = True
order = 50
languages = [Language("heb")]
long_description = """\
Some playback devices don't properly handle right-to-left markers for punctuation. Physically swap punctuation.
Applicable to languages: hebrew
"""
processors = [
# new? (?u)(^([\s.!?]*)(.+?)(\s*)(-?\s*)$); \5\4\3\2
#NReProcessor(re.compile(ur"(?u)((?=(?<=\b|^)|(?<=\s))([.!?-]+)([^.!?-]+)(?=\b|$|\s))"), r"\3\2",
# name="CM_RTL_reverse")
NReProcessor(re.compile(ur"(?u)(^([\s.!?]*)(.+?)(\s*)(-?\s*)$)"), r"\5\4\3\2",
name="CM_RTL_reverse")
]
registry.register(CommonFixes)
registry.register(RemoveTags)
registry.register(ReverseRTL)
name="CM_RTL_reverse")
]
split_upper_re = re.compile(ur"(\s*[.!?♪\-]\s*)")
class FixUppercase(SubtitleModification):
identifier = "fix_uppercase"
description = "Fixes all-uppercase subtitles"
modifies_whole_file = True
exclusive = True
order = 41
only_uppercase = True
apply_last = True
long_description = "Some subtitles are in all-uppercase letters. This at least makes them readable."
def capitalize(self, c):
return u"".join([s.capitalize() for s in split_upper_re.split(c)])
def modify(self, content, debug=False, parent=None, **kwargs):
for entry in parent.f:
entry.plaintext = self.capitalize(entry.plaintext)
registry.register(CommonFixes)
registry.register(RemoveTags)
registry.register(ReverseRTL)
registry.register(FixUppercase)
# coding=utf-8
import logging
from subzero.modification.mods import SubtitleModification
from subzero.modification import registry
logger = logging.getLogger(__name__)
class ChangeFPS(SubtitleModification):
identifier = "change_FPS"
description = "Change the FPS of the subtitle"
exclusive = True
advanced = True
modifies_whole_file = True
long_description = "Re-syncs the subtitle to the framerate of the current media file."
def modify(self, content, debug=False, parent=None, **kwargs):
fps_from = kwargs.get("from")
fps_to = kwargs.get("to")
parent.f.transform_framerate(float(fps_from), float(fps_to))
registry.register(ChangeFPS)
("dark-red", "#800000"),
("dark-green", "#008000"),
("dark-yellow", "#808000"),
("dark-blue", "#000080"),
("dark-magenta", "#800080"),
("dark-cyan", "#008080"),
("dark-grey", "#808080"),
])
class Color(SubtitleModification):
identifier = "color"
description = "Change the color of the subtitle"
exclusive = True
advanced = True
modifies_whole_file = True
apply_last = True
colors = COLOR_MAP
long_description = "Adds the requested color to every line of the subtitle. Support depends on player."
def modify(self, content, debug=False, parent=None, **kwargs):
color = self.colors.get(kwargs.get("name"))
if color:
for entry in parent.f:
entry.text = u'<font color="%s">%s</font>' % (color, entry.text)
registry.register(Color)
NReProcessor(re.compile(ur'(?u)([A-zÀ-ž][a-zà-ž]+)(I+)'),
lambda match: ur'%s%s' %
(match.group(1), "l" * len(match.group(2))),
name="CM_uppercase_i_in_word"),
# fix spaces in numbers (allows for punctuation: ,.:' (comma/dot only fixed if after space, those may be
# countdowns otherwise); don't break up ellipses
# fixme: maybe check whether it's a countdown (second part smaller than the first), otherwise handle default?
NReProcessor(re.compile(
r'(?u)([0-9]+[0-9:\']*(?<!\.\.)\s+(?!\.\.)[0-9,.:\']*(?=[0-9]+)[0-9,.:\'\s]+)(?=\s|$)'
),
lambda match: match.group(1).replace(" ", ""),
name="CM_spaces_in_numbers"),
# uppercase after dot
NReProcessor(re.compile(ur'(?u)((?:[^.\s])+\.\s+)([a-zà-ž])'),
lambda match: ur'%s%s' %
(match.group(1), match.group(2).upper()),
name="CM_uppercase_after_dot"),
# remove spaces before punctuation
NReProcessor(re.compile(r'(?u)(?:(?<=^)|(?<=\w)) +([!?.,](?![!?.,]))'),
r"\1",
name="CM_punctuation_space"),
]
post_processors = empty_line_post_processors
registry.register(CommonFixes)
# text in brackets at start, after optional dash, before colon or at end of line
# fixme: may be too aggressive
#NReProcessor(re.compile(ur'(?um)(^-?\s?[([][A-zÀ-ž-_\s]{3,}[)\]](?:(?=$)|:\s*))'), "",
# name="HI_brackets_special"),
# all caps line (at least 4 consecutive uppercase chars)
NReProcessor(re.compile(ur'(?u)(^(?=.*[A-ZÀ-Ž&+]{4,})[A-ZÀ-Ž-_\s&+]+$)'), "", name="HI_all_caps",
supported=lambda p: not p.only_uppercase),
# remove MAN:
NReProcessor(re.compile(ur'(?suxi)(.*MAN:\s*)'), "", name="HI_remove_man"),
# dash in front
# NReProcessor(re.compile(r'(?u)^\s*-\s*'), "", name="HI_starting_dash"),
# all caps at start before new sentence
NReProcessor(re.compile(ur'(?u)^(?=[A-ZÀ-Ž]{4,})[A-ZÀ-Ž-_\s]+\s([A-ZÀ-Ž][a-zà-ž].+)'), r"\1",
name="HI_starting_upper_then_sentence", supported=lambda p: not p.only_uppercase),
# remove music symbols
NReProcessor(re.compile(ur'(?u)(^%(t)s[*#¶♫♪\s]*%(t)s[*#¶♫♪\s]+%(t)s[*#¶♫♪\s]*%(t)s$)' % {"t": TAG}),
"", name="HI_music_symbols_only"),
]
post_processors = empty_line_post_processors
registry.register(HearingImpaired)
data_dict = OCR_fix_data.get(parent.language.alpha3t)
if not data_dict:
logger.debug("No SnR-data available for language %s", parent.language)
return
self.data_dict = data_dict
self.processors = self.get_processors()
def get_processors(self):
if not self.data_dict:
return []
return [
# remove broken HI tag colons (ANNOUNCER'., ". instead of :) after at least 3 uppercase chars
# don't modify stuff inside quotes
NReProcessor(re.compile(ur'(?u)(^[^"\'’ʼ❜‘‛”“‟„]*(?<=[A-ZÀ-Ž]{3})[A-ZÀ-Ž-_\s0-9]+)'
ur'(["\'’ʼ❜‘‛”“‟„]*[.,‚،⹁、;]+)(\s*)(?!["\'’ʼ❜‘‛”“‟„])'),
r"\1:\3", name="OCR_fix_HI_colons", supported=lambda p: not p.only_uppercase),
# fix F'bla
NReProcessor(re.compile(ur'(?u)(\bF)(\')([A-zÀ-ž]*\b)'), r"\1\3", name="OCR_fix_F"),
WholeLineProcessor(self.data_dict["WholeLines"], name="OCR_replace_line"),
MultipleWordReProcessor(self.data_dict["WholeWords"], name="OCR_replace_word"),
MultipleWordReProcessor(self.data_dict["BeginLines"], name="OCR_replace_beginline"),
MultipleWordReProcessor(self.data_dict["EndLines"], name="OCR_replace_endline"),
MultipleWordReProcessor(self.data_dict["PartialLines"], name="OCR_replace_partialline"),
MultipleLineProcessor(self.data_dict["PartialWordsAlways"], name="OCR_replace_partialwordsalways")
]
registry.register(FixOCR)
