def open_ass(file):
try:
with open(file, "r", encoding="gbk") as f:
doc = ass.parse(f)
return doc
except:
pass
try:
with open(file, "r", encoding="big5") as f:
doc = ass.parse(f)
return doc
except:
pass
try:
with open(file, "r", encoding="utf-8") as f:
doc = ass.parse(f)
return doc
except:
pass
try:
with open(file, 'rb') as f:
input_bytes = f.read()
result = chardet.detect(input_bytes)
conf = result['confidence']
res_encoding = result['encoding']
if conf > 0.95:
with open(file, "r", encoding=res_encoding) as f:
doc = ass.parse(f)
return doc
except:
pass
return None
def test_parse_encoding(self):
with self.test_ass.open("r", encoding='utf_8') as f:
with self.assertRaises(ValueError):
ass.parse(f)
with self.test_ass.open("r", encoding='ascii') as f:
with self.assertRaises(ValueError):
ass.parse(f)
def test_dump_encoding(self):
for encoding in ('utf_8_sig', 'utf-8-sig'):
with self.test_ass.open("r", encoding=encoding) as f:
doc = ass.parse(f)
with self.test_ass.open("r", encoding=encoding.upper()) as f:
doc = ass.parse(f)
import tempfile
with tempfile.TemporaryFile(mode='w', encoding='utf_8') as f:
with pytest.warns(UserWarning):
doc.dump_file(f)
def load_subs(subsfile):
'''
Loads up and parses subtitles from subsfile and returns subsdigest object.
'''
with open(subsfile, encoding='utf_8_sig') as f:
subsdata = subdigest.Subtitles(ass.parse(f), subsfile)
return subsdata
def loadSubsAss(self, file):
with open(file, "r") as f:
self.subs = ass.parse(f).events
self.clear()
g = 0
# for i in self.subs:
# if i.style not in "ED,OP,staff": #*Default": #
# self.insertItem(g, i.text.decode('utf_8'))
# g = g + 1
for i in self.subs:
self.insertItem(g, i.text.decode('utf_8'))
g = g + 1
for i in xrange(self.count()):
self.item(i).setFont(QFont('Meiryo', 16)) # MS Mincho
i = self.subs[0]
self.currentSubStart = i.start.total_seconds(
) * 1000 # + i.start.microseconds
self.currentSubEnd = i.end.total_seconds(
) * 1000 # + i.end.microseconds
self.currentRow = 0
i = self.subs[1]
self.nextSubStart = i.start.total_seconds(
) * 1000 # + i.start.microseconds
self.nextSubEnd = i.end.total_seconds() * 1000 # + i.end.microseconds
def read_dialogues(self): subs = ass.parse(self.contents.splitlines()) buffer = '' for quote in subs.events: if isinstance(quote, ass.document.Dialogue): buffer += quote.text + '\n' return unicode(buffer)
def edit_cn_ass_subtitle_style(subtitle_file):
"""
为ass格式的中文字幕设置style
:param subtitle_file:
:return:
"""
with open(subtitle_file, "r") as f:
subtitle = ass.parse(f)
print(subtitle.styles)
# subtitle.styles[0].fontname = '方正黑体_GBK'
subtitle.styles[0].fontname = '黑体'
subtitle.styles[0].fontsize = 21
subtitle.styles[0].primary_color = '&H00FFFFFF'
subtitle.styles[0].secondary_color = '&HF0000000'
subtitle.styles[0].outline_color = '&H006C3300'
subtitle.styles[0].back_color = '&H00000000'
subtitle.styles[0].bold = -1
subtitle.styles[0].border_style = 1
subtitle.styles[0].outline = 2
subtitle.styles[0].shadow = 1
subtitle.styles[0].alignment = 2
with open(subtitle_file, "w") as f:
subtitle.dump_file(f)
if os.path.exists(subtitle_file):
return subtitle_file
def main():
file_path = sys.argv[1]
with open(file_path, 'r') as fp:
doc = ass.parse(fp)
for i, event in enumerate(doc.events[:]):
if event.TYPE == 'Dialogue' and event.effect == "fx":
doc.events[i] = None # set to None to remove later
continue
if (event.TYPE == 'Comment' and event.effect == "karaoke"
or event.TYPE == 'Dialogue' and not event.effect):
if "Kanji" in event.style:
comment_event = ass.document.Comment(**event.fields)
comment_event.effect = "karaoke"
new_event = ass.document.Dialogue(**event.fields)
new_event.effect = "fx"
# Cannot use str.join because tags might be inbetween text
parts = re.split(r"(\{[^\}]*\})", new_event.text)
parts = (f"{c}\\N" if not part.startswith("{") else c
for part in parts for c in part)
new_event.text = "".join(parts)[:-2]
doc.events[i] = comment_event
doc.events.append(new_event)
# remove events flagged for deletion
doc.events = filter(None, doc.events)
with open(file_path + "_", 'w') as fp:
doc.dump_file(fp)
def edit_cn_ass_subtitle_style(subtitle_file):
"""
为ass格式的中文字幕设置style
:param subtitle_file:
:return:
"""
with open(subtitle_file, "r") as f:
subtitle = ass.parse(f)
print(subtitle.styles)
# subtitle.styles[0].fontname = '方正黑体_GBK'
subtitle.styles[0].fontname = '黑体'
subtitle.styles[0].fontsize = 21
subtitle.styles[0].primary_color = '&H00FFFFFF'
subtitle.styles[0].secondary_color = '&HF0000000'
subtitle.styles[0].outline_color = '&H006C3300'
subtitle.styles[0].back_color = '&H00000000'
subtitle.styles[0].bold = -1
subtitle.styles[0].border_style = 1
subtitle.styles[0].outline = 2
subtitle.styles[0].shadow = 1
subtitle.styles[0].alignment = 2
with open(subtitle_file, "w") as f:
subtitle.dump_file(f)
if os.path.exists(subtitle_file):
return subtitle_file
def get_timestamps(dialogue):
'''
Reads 'chptr' comments from dialogue subs to find timestamps of OP, ED and Eyecatch.
Chapters must be named in the following format:
Opening
Part B
Ending
...
Returns a dict of timestamps and frame numbers.
'''
op = ec = ed = opsync = ecsync = edsync = None
with open(dialogue, encoding='utf_8_sig') as f:
dialogue_data = ass.parse(f)
for event in dialogue_data.events:
if not isinstance(event, ass.line.Comment) and not event.name == 'chptr':
continue
if 'opening' in event.text.lower():
opsync = event.start
op = math.ceil(opsync.total_seconds() * 23.976)
if 'ending' in event.text.lower():
edsync = event.start
ed = math.ceil(edsync.total_seconds() * 23.976)
if 'part b' in event.text.lower(): # Part B / Eyecatch
ecsync = event.start
ec = math.ceil(ecsync.total_seconds() * 23.976)
return {'op': op, 'ec': ec, 'ed': ed, 'opsync': opsync, 'ecsync': ecsync, 'edsync': edsync}
def test_dump_encoding(self):
with self.test_ass.open("r", encoding='utf_8_sig') as f:
doc = ass.parse(f)
import tempfile
with tempfile.TemporaryFile(mode='w', encoding='utf_8') as f:
with self.assertWarns(UserWarning):
doc.dump_file(f)
def test_parse_dump(self):
with self.test_ass.open("r", encoding='utf_8_sig') as f:
contents = f.read()
doc = ass.parse(StringIO(contents))
out = StringIO()
doc.dump_file(out)
assert out.getvalue().strip() == contents.strip()
def test_parse_dump(self):
with open("test.ass", "r") as f:
contents = f.read()
doc = ass.parse(StringIO(contents))
out = StringIO()
doc.dump_file(out)
self.assertEqual(out.getvalue().strip(), contents.strip())
def test_parse_dump(self):
with open("test.ass", "r") as f:
contents = f.read()
doc = ass.parse(StringIO(contents))
out = StringIO()
doc.dump_file(out)
self.assertEqual(out.getvalue().strip(), contents.strip())
def fix_ssa_start_end(basename):
with open(f'{basename}.ssa') as f:
d = ass.parse(f)
prev = d.events[0]
for i in d.events[1:]:
i.start = prev.end + datetime.timedelta(seconds=2)
prev = i
with open(f'{basename}.ssa', 'w') as f:
d.dump_file(f)
def main():
parser = argparse.ArgumentParser(description="Automatically merge fonts used in a Matroska file.")
parser.add_argument('mkv', help="""
Video where the fonts will go. Must be a Matroska file.
""")
parser.add_argument('subtitles', nargs="+", help="""
Subtitles (can be several) containing fonts to be merged. Must be an ASS file.
""")
parser.add_argument('--mkvmerge', metavar="path", help="""
Path to mkvmerge.exe if not in variable environments.
""")
parser.add_argument('--fontfolder', metavar="path", help="""
Add a file with fonts to use.
""")
parser.add_argument('--output', '-o', metavar="path", help="""
Destination path of the Matroska merged file.
""")
args = parser.parse_args()
if args.mkvmerge is None and not distutils.spawn.find_executable("mkvmerge.exe"):
return print(Fore.RED + "fontmerge.py: error: mkvmerge in not in your environnements variable, add it or specify the path to mkvmerge.exe with --mkvmerge." + Fore.WHITE)
if not is_mkv(args.mkv):
return print(Fore.RED + "fontmerge.py: error: the file on mkv is not a Matroska file."+ Fore.WHITE)
if not is_ass(args.subtitles):
return print(Fore.RED + "fontmerge.py: error: the file on ass is not an ASS file." + Fore.WHITE)
if not is_writable(args.mkv):
return print(Fore.RED + "fontmerge.py: error: unable to create the Matroska file." + Fore.WHITE)
if args.output is not None :
if not is_dir(os.path.dirname(args.output)):
return print(Fore.RED + "fontmerge.py: error: output path is not a valid folder." + Fore.WHITE)
if args.fontfolder is not None :
if not is_dir(args.fontfolder):
return print(Fore.RED + "fontmerge.py: error: font path is not a directory." + Fore.WHITE)
else:
if not contains_fonts(args.fontfolder):
print(Fore.RED + "fontmerge.py: error: font path does not contain any fonts." + Fore.WHITE)
args.fontfolder = None
fonts_path = []
installedFonts = get_installed_fonts(args.fontfolder)
for assf in args.subtitles:
with open(assf, 'r', encoding='utf_8_sig') as f:
subtitles = [(os.path.basename(assf), ass.parse(f))]
fonts_path.extend(get_used_font_path(subtitles, installedFonts))
if len(fonts_path) != len(dict.fromkeys(fonts_path)):
print("Some fonts are duplicate. Removing.")
fonts_path = list(dict.fromkeys(fonts_path))
merge(args.mkv, fonts_path, args.mkvmerge, args.output)
def main():
parser = argparse.ArgumentParser()
for group in _filter_groups:
arggroup = parser.add_argument_group(group.__name__)
for name, member in inspect.getmembers(group):
if getattr(member, '_filter', False):
generate_argument(arggroup, member)
parser.add_argument("-i",
"--input",
help="Specify input file (default: stdin)")
parser.add_argument("-o",
"--output",
help="Specify output file (default: stdout)")
parser.add_argument("--in-place",
action="store_true",
help="Perform operations in place")
args = parser.parse_args()
if args.input is None or args.input == '-':
filename = None
sub_obj = ass.parse(codecs.getreader('utf-8-sig')(sys.stdin.buffer))
else:
filename = args.input
with open(args.input, 'r', encoding='utf-8-sig') as f:
sub_obj = ass.parse(f)
sub_obj = Subtitles(sub_obj, filename)
for func, filter_args in getattr(args, 'chain', []):
filt = getattr(sub_obj, func)
sub_obj = filt(*filter_args)
if args.in_place and args.input is not None:
args.output = args.input
if args.output is None or args.output == '-':
sys.stdout.buffer.write(str(sub_obj).encode('utf-8-sig'))
else:
with open(args.output, 'w', encoding="utf-8-sig") as f:
f.write(str(sub_obj))
def tryOpen(f):
"""Open a .ass file with a bunch of encodings until one works."""
if not os.path.isfile(f):
raise Exception('Invalid file: %s' % (f,))
lastException = None
for encoding in ('utf-8', 'utf-16', 'utf-16-le'):
try:
return ass.parse(open(f, 'r', encoding=encoding).readlines())
except Exception as e:
lastException = e
raise Exception('Cannot find suitable encoding for file %s. Last error: %s' % (f, lastException))
def get_dialogues(self): if self.subs_format == 'ass': f = open(self.subs, 'r') subs = ass.parse(f) return [Dialogue(event.text, self.to_ms(event.start), self.to_ms(event.end)) for event in subs.events] elif self.subs_format == 'srt': subs = pysrt.open(self.subs) return [Dialogue(sub.text, self.to_ms_srt(sub.start), self.to_ms_srt(sub.end)) for sub in subs]
def get_dialogues(file_name):
with open(file_name, "r") as rawFile:
doc = ass.parse(rawFile)
dialogues = []
for event in doc.events:
dialogues.append({
"text": regex_process(event.text),
"start": timedelta_to_ass(event.start),
"end": timedelta_to_ass(event.end),
"file": os.path.abspath(file_name)[:-4]
})
return dialogues
def main(args):
with open(args.filename, 'r') as f:
doc = ass.parse(f)
known = set(style.name for style in doc.styles)
used = set(event.style for event in doc.events if event.text.strip() != '')
unknown = used - known
if unknown:
raise LookupError('Some styles are not defined: {}'.format(unknown))
# remove unused styles
doc.styles = [style for style in doc.styles if style.name in used]
# rename styles to random names
if args.rename:
prefix = hex(random.getrandbits(128))[-8:]
style_names = {}
_new_names = set()
for style in doc.styles:
while True:
new_name = hex(random.getrandbits(128))[-8:]
if new_name not in _new_names:
_new_names.add(new_name)
break
style_names[style.name] = '{}-{}'.format(prefix, new_name)
for style in doc.styles:
style.name = style_names[style.name]
# clean up events and remove comments or empty texts.
events = doc.events
events = (clean_text(event, args.quotes) for event in events)
events = (event for event in events if event.text != ''
and event.TYPE.lower() == 'dialogue')
doc.events = list(events)
# rename the styles used in events.
if args.rename:
for event in doc.events:
event.style = style_names[event.style]
# remove all the third party tools headers (Aegisub etc.)
fields = doc.fields.items()
fields = ((k, v) for k, v in fields if k in desired_fields)
doc.fields = dict(fields)
if args.title is not None:
doc.fields['Title'] = args.title
# dump script
with open(args.output, 'w') as f:
doc.dump_file(f)
def main(args):
with open(args.filename, 'r') as f:
doc = ass.parse(f)
known = set(style.name for style in doc.styles)
used = set(event.style for event in doc.events if event.text.strip() != '')
unknown = used - known
if unknown:
raise LookupError('Some styles are not defined: {}'.format(unknown))
# remove unused styles
doc.styles = [style for style in doc.styles if style.name in used]
# rename styles to random names
if args.rename:
prefix = hex(random.getrandbits(128))[-8:]
style_names = {}
_new_names = set()
for style in doc.styles:
while True:
new_name = hex(random.getrandbits(128))[-8:]
if new_name not in _new_names:
_new_names.add(new_name)
break
style_names[style.name] = '{}-{}'.format(prefix, new_name)
for style in doc.styles:
style.name = style_names[style.name]
# clean up events and remove comments or empty texts.
events = doc.events
events = (clean_text(event, args.quotes) for event in events)
events = (event for event in events
if event.text != '' and event.TYPE.lower() == 'dialogue')
doc.events = list(events)
# rename the styles used in events.
if args.rename:
for event in doc.events:
event.style = style_names[event.style]
# remove all the third party tools headers (Aegisub etc.)
fields = doc.fields.items()
fields = ((k, v) for k, v in fields if k in desired_fields)
doc.fields = dict(fields)
if args.title is not None:
doc.fields['Title'] = args.title
# dump script
with open(args.output, 'w') as f:
doc.dump_file(f)
def gomen(subfile, times):
if times < 0:
print "Then why are you running this script?"
sys.exit()
print "Opening subtitle file..."
with open(subfile, "r") as subf:
subs = ass.parse(subf)
try:
subs.fields["Title"] = "Gomen"
except:
pass
fullstr = ""
indexes_to_replace = []
for dialogue_line in subs.events:
if type(dialogue_line) is ass.document.Dialogue and dialogue_line.text.find("(") == -1:
indexes_to_replace.append(subs.events.index(dialogue_line))
if dialogue_line.text.find("\N") != -1:
for separ in dialogue_line.text.split("\N"):
fullstr += separ + "*"
fullstr += "^"
else:
fullstr += dialogue_line.text + "^"
translations_to_perform = [lang_list[random.randint(0, len(lang_list) - 1)] for x in range(times)]
en_to_mystery = translate(fullstr, "en", translations_to_perform[0])
if times > 1:
for x in range(times - 1):
en_to_mystery = translate(en_to_mystery, translations_to_perform[x], translations_to_perform[x + 1])
if "^" in en_to_mystery:
translated_list = translate(en_to_mystery, translations_to_perform[times - 1], "en").split("^")
else:
print "Our string got messed up in translation."
sys.exit()
translated_parsed_list = [replace(item, "*", "\N") for item in translated_list]
x = 0
try:
for index in indexes_to_replace:
subs.events[index].text = translated_parsed_list[x]
x += 1
except:
print "Indexes are messed up but whatever. That just makes it better, right?"
if (
subs.events[0].text.count("\\x") > 2
or subs.events[1].text.count("\\x") > 2
or subs.events[2].text.count("\\x") > 2
or subs.events[3].text.count("\\x") > 2
):
print "Translation failed. Our string got screwed up. Trying again..."
else:
print "Finished. Writing to file."
with open("subs_translated.ass", "w") as subs_out:
subs.dump_file(subs_out)
def parse(file_path: str) -> Episode:
with open(file_path, encoding='utf-8-sig') as file:
serialized_ass = ass.parse(file)
title = serialized_ass.fields.get('Title')
production_code = int(serialized_ass.fields.get('Original Script'))
episode = Episode(production_code, title=title)
for event in serialized_ass.events:
line = Line(production_code, event.start, event.end, event.name,
event.text)
episode.add_line(line)
return episode
def parse(filename):
segments = []
if filename.endswith(".ass"):
with open(filename, "r") as f:
doc = ass.parse(f)
for event in doc.events:
if ignore_ass_event(event):
continue
seg = Segment(event.start, event.end, event.tags_stripped())
segments.append(seg)
else:
raise Exception("Invalid file format")
return segments
def tryOpen(f):
"""Open a .ass file with a bunch of encodings until one works."""
if not os.path.isfile(f):
raise Exception('Invalid file: %s' % (f, ))
lastException = None
for encoding in ('utf-8', 'utf-16', 'utf-16-le'):
try:
return ass.parse(open(f, 'r', encoding=encoding).readlines())
except Exception as e:
lastException = e
raise Exception(
'Cannot find suitable encoding for file %s. Last error: %s' %
(f, lastException))
def extractTextFromSubtitles(fileName):
tracks = getSubtitleTracks(fileName)
output = ""
for track in tracks:
srtName = exportSRT(fileName, track)
lines = []
with open(srtName,"r") as f:
doc = ass.parse(f)
for event in doc.events:
lines.append(cleanLine(event.text))
combined = "\n".join(lines)
if "in" in combined or "to" in combined or "for" in combined:
output += combined
return output
def index_subtitle(index_writer, object_id, file, format):
if format not in SUPPORTED_SUBTITLE_FORMATS:
raise ESubtitleFormatNotSupported(format)
try:
if format == SRT:
subtitles = srt.parse(file)
_index_srt(index_writer, object_id, subtitles)
else:
doc = ass.parse(file)
_index_ass(index_writer, object_id, doc.events)
index_writer.commit()
except Exception as e:
index_writer.cancel()
raise e
def main():
used_command = sys.argv.pop(0)
doc_title = None
rename_styles = False
global_sync = 0
files = []
while sys.argv:
arg = sys.argv.pop(0)
if arg in ('-h', '--help'):
usage(used_command)
sys.exit(0)
elif arg == '-r':
rename_styles = True
elif arg == '-t':
doc_title = sys.argv.pop(0)
elif arg == '-g':
global_sync = int(sys.argv.pop(0)) * 1000
elif arg == '-s':
file_sync = int(sys.argv.pop(0)) * 1000
filename = sys.argv.pop(0)
files.append((file_sync, filename))
else:
files.append((None, arg))
docs = []
for sync, filename in files:
if sync is None:
sync = global_sync
print('', file=sys.stderr)
print('File: {}'.format(filename), file=sys.stderr)
print(' Sync: {}'.format(sync), file=sys.stderr)
print('', file=sys.stderr)
with open(filename, 'r') as f:
doc = ass.parse(f)
if sync != 0:
doc = shift(doc, sync)
docs.append(doc)
merged_docs = merge(*docs, rename_styles=rename_styles)
if doc_title is not None:
merged_docs.fields['Title'] = doc_title
merged_docs.dump_file(sys.stdout)
def main(path):
path = path or "."
for f in list_dir(path):
if f and f.endswith('.ass'):
if f == 'ok':
continue
print ''
print f
with open(f, 'r') as f_:
doc = ass.parse(f_)
print ' fn:', doc.styles[0].fontname
print ' fs:', doc.styles[0].fontsize
print ' bold:', doc.styles[0].bold
print ' shadow:', doc.styles[0].shadow
print ' border_style:', doc.styles[0].border_style
print ' primary_color: ', doc.styles[0].primary_color
for i in range(len(doc.styles)):
doc.styles[i].fontname = 'Microsoft YaHei Light'
doc.styles[i].fontsize = 13
doc.styles[i].bold = -1
doc.styles[i].shadow = 0
doc.styles[i].outline = 0
doc.styles[i].border_style = 0
doc.styles[i].primary_color = Color(r=0xff,
g=0xff,
b=0xff,
a=0x30)
ok_path = os.path.join(path, 'ok')
if not os.path.exists(ok_path):
os.mkdir(ok_path)
new_path = f.replace(path, os.path.join(path, 'ok'))
print new_path
with open(new_path, 'w') as nf:
doc.dump_file(nf)
with open(new_path, 'r') as nf:
txt = nf.read().replace(en_font_name,
'Microsoft YaHei Light').replace(
en_font_size, 'fs11')
with open(new_path, 'w') as nf:
nf.write(txt)
def edit_two_lang_style(subtitle_file):
"""
为ass格式的双语字幕设置style
:param subtitle_file:
:return:
"""
with open(subtitle_file, "r") as f:
subtitle = ass.parse(f)
print(subtitle.styles)
# {\fn宋体\fs20\shad2\4a&H50&\3c&HFF8000&\4c&HFF8000&}
# Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour,
# OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut,
# ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow,
# Alignment, MarginL, MarginR, MarginV, Encoding
# Style: Default,方正黑体_GBK,21,&H00FFFFFF,&HF0000000,&H006C3300,
# &H00000000,-1,0,0,0,100,100,0,0,1,2,1,2,5,5,5,134
# subtitle.styles[0].fontname = '方正黑体_GBK'
subtitle.styles[0].fontname = '黑体'
subtitle.styles[0].fontsize = 21
subtitle.styles[0].primary_color = '&H00FFFFFF'
subtitle.styles[0].secondary_color = '&HF0000000'
subtitle.styles[0].outline_color = '&H006C3300'
subtitle.styles[0].back_color = '&H00000000'
subtitle.styles[0].bold = -1
subtitle.styles[0].border_style = 1
subtitle.styles[0].outline = 2
subtitle.styles[0].shadow = 1
subtitle.styles[0].alignment = 2
for events in subtitle.events:
utf8string = events.text.decode("utf-8")
events.text = utf8string.replace(
r'\N',
r'\N{\fn黑体\fs14\b0\c&HFFFFFF&\3c&H2F2F2F&\4c&H000000&}'
)
with open(subtitle_file, "w") as f:
subtitle.dump_file(f)
if os.path.exists(subtitle_file):
return subtitle_file
def edit_two_lang_style(subtitle_file):
"""
为ass格式的双语字幕设置style
:param subtitle_file:
:return:
"""
with open(subtitle_file, "r") as f:
subtitle = ass.parse(f)
print(subtitle.styles)
# {\fn宋体\fs20\shad2\4a&H50&\3c&HFF8000&\4c&HFF8000&}
# Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour,
# OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut,
# ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow,
# Alignment, MarginL, MarginR, MarginV, Encoding
# Style: Default,方正黑体_GBK,21,&H00FFFFFF,&HF0000000,&H006C3300,
# &H00000000,-1,0,0,0,100,100,0,0,1,2,1,2,5,5,5,134
# subtitle.styles[0].fontname = '方正黑体_GBK'
subtitle.styles[0].fontname = '黑体'
subtitle.styles[0].fontsize = 21
subtitle.styles[0].primary_color = '&H00FFFFFF'
subtitle.styles[0].secondary_color = '&HF0000000'
subtitle.styles[0].outline_color = '&H006C3300'
subtitle.styles[0].back_color = '&H00000000'
subtitle.styles[0].bold = -1
subtitle.styles[0].border_style = 1
subtitle.styles[0].outline = 2
subtitle.styles[0].shadow = 1
subtitle.styles[0].alignment = 2
for events in subtitle.events:
utf8string = events.text.decode("utf-8")
events.text = utf8string.replace(
r'\N', r'\N{\fn黑体\fs14\b0\c&HFFFFFF&\3c&H2F2F2F&\4c&H000000&}')
with open(subtitle_file, "w") as f:
subtitle.dump_file(f)
if os.path.exists(subtitle_file):
return subtitle_file
def AssWeTran(SourceFileName):
SourceFile = open(SourceFileName, 'rb')
FileEnco = chardet.detect(SourceFile.read()).get('encoding')
SourceFile.close()
# 按文件编码重新打开文件
SourceFile = codecs.open(SourceFileName, 'rb', FileEnco)
TargetFile = codecs.open(SourceFileName + '.srt', 'w', 'utf-8')
AssFile = ass.parse(SourceFile)
TimeDic = dict()
for event in AssFile.events:
if not re.match('//', event.text) and not event.text == '':
StartTime = str(event.start)
EndTime = str(event.end)
StartTime = '0' + StartTime
StartTime = StartTime[:-3]
StartTime = re.sub(r'\.', ',', StartTime)
EndTime = '0' + EndTime
EndTime = EndTime[:-3]
EndTime = re.sub(r'\.', ',', EndTime)
Time = StartTime + ' --> ' + EndTime
SubTitle = re.sub(r'{[^}]*}', '', event.text)
SubTitle = re.sub(r'\\N', '\n', SubTitle)
if not Time in TimeDic.keys():
TimeDic[Time] = SubTitle
else:
TimeDic[Time] = TimeDic[Time] + '\n' + SubTitle
i = 1
for key, value in TimeDic.items():
TargetFile.write(str(i))
TargetFile.write('\n')
TargetFile.write(key)
TargetFile.write('\n')
TargetFile.write(value)
TargetFile.write('\n')
TargetFile.write('\n')
i = i + 1
TargetFile.close()
return TargetFile.name
def merge_file(
self,
other_file: argparse.FileType(encoding='utf-8-sig')) -> Subtitles:
"""Append the styles and event lines from another file."""
f = ass.parse(other_file)
self.sub_file.events.extend(f.events)
existing_styles = {
style.name: style.dump()
for style in self.sub_file.styles
}
for style in f.styles:
if style.name in existing_styles and style.dump(
) != existing_styles[style.name]:
print(
f"Warning: Ignoring style {style.name} from "
f"{other_file.name}.",
file=sys.stderr)
continue
self.sub_file.styles.append(style)
return self
def _import_file(self, imp_definition, styles, events, fields):
# strip extradata info
fname = re.sub(r"^\{=\d+(,\d+)*\}", "", imp_definition.text)
shifted = imp_definition.effect == 'import-shifted'
self._ms_count += 1
self._ms_files[self._ms_count] = fname
path = pathlib.Path(self.filename).parent / fname
with path.open(encoding='utf-8-sig') as f:
imp = ass.parse(f)
if shifted:
try:
sync_line = next(line for line in imp.events
if line.effect == 'sync')
except StopIteration:
raise ValueError(f"No sync line in {fname}")
time_diff = imp_definition.start - sync_line.start
for line in imp.events:
line.start += time_diff
line.end += time_diff
for line in imp.events:
line.layer += imp_definition.layer
line.style = f"{self._ms_count}${line.style}"
events.append(line)
for style in imp.styles:
style.name = f"{self._ms_count}${style.name}"
styles[style.name] = style
for field, value in imp.fields.items():
if field in fields and fields[field] != value:
logging.warning(
f"Ignoring conflicting value {value} for {field} from {fname}"
)
else:
fields[field] = value
def load_files():
subtitles = []
videos = []
for filename in os.listdir('.'):
if filename.lower().endswith('.ass'):
with open(filename, 'r') as f:
subtitles.append(Ass(filename, ass.parse(f)))
elif filename.lower().endswith(('.mkv', '.mp4')):
proc = Popen(
['ffprobe', '-i', filename, '-v', 'quiet', '-print_format',
'json', '-show_streams'], stdout=PIPE, stderr=DEVNULL)
out, err = proc.communicate()
proc.wait()
data = json.loads(out.decode('ascii'))
videos.append(Video(filename, data))
videos.sort()
subtitles.sort()
videos = Collection(videos, 'Video files')
subtitles = Collection(subtitles, 'ASS files')
return subtitles, videos
def ass_parser(filename):
f = codecs.open(filename, 'r', 'utf8')
doc = ass.parse(f)
f.close()
print("length of event is: {}".format(len(doc.events)))
doc.events.sort(key=lambda x:x.start)
for event in doc.events:
event.text = re.sub("{.*?}", "", event.text).replace("\\N", "\n")
srtname = filename[:-3] + 'srt'
f = codecs.open(srtname, 'w', 'utf8')
for index, event in enumerate(doc.events):
f.write("%d\n"%(index + 1))
f.write("%s --> %s\n" % (time_string(event.start), time_string(event.end)))
f.write(event.text)
f.write("\n\n")
f.close()
def readAss(fname):
if str(fname).endswith('.ass'):
with open(fname, 'r', encoding='utf-8') as f:
doc = ass.parse(f)
dml = []
for evt in doc.events:
dm = DanMu()
dm.createByAss(evt, doc.play_res_x, doc.play_res_y)
dml.append(dm)
return doc.styles, dml
elif str(fname).endswith(".xml"):
tmpfile = fname + "_tmp.ass"
print('请自行转换')
raise
#readAss('dm.ass')
def main():
parser = argparse.ArgumentParser(
description="Automatically merge fonts used in a Matroska file.")
parser.add_argument('subtitles',
help="""
Subtitles containing fonts to be merged. Must be an ASS file. """)
parser.add_argument('mkv',
help="""
Video where the fonts will go. Must be a Matroska file.
""")
parser.add_argument('--mkvmerge',
metavar="path",
help="""
Path to mkvmerge.exe if not in variable environments.
""")
args = parser.parse_args()
if not distutils.spawn.find_executable(
"mkvmerge.exe") and args.mkvmerge is None:
return print(
"fontmerge.py: error: mkvmerge in not in your environnements variable, add it or specify the path to mkvmerge.exe with --mkvmerge."
)
if not is_mkv(args.mkv):
return print(
"fontmerge.py: error: the file on -mkv is not a Matroska file.")
if not is_ass(args.subtitles):
return print("fontmerge.py: error: the file is not an Ass file.")
if not is_writable(args.mkv):
return print("fontmerge.py: error: unable to create the file.")
with open(args.subtitles, 'r', encoding='utf_8_sig') as f:
subtitles = [(os.path.basename(args.subtitles), ass.parse(f))]
fonts_path = get_used_font_path(subtitles)
merge(args.subtitles, args.mkv, fonts_path, args.mkvmerge)
return print("Successfully merging subtitles and mkv")
def ssaProcessor(fname: str):
if not os.path.isfile(fname):
print(f'Missing file: {fname}')
return
with open(fname, encoding='utf_8_sig') as f:
sub = ssa.parse(f)
for s in sub.styles:
transformColour(s.primary_color)
transformColour(s.secondary_color)
transformColour(s.outline_color)
transformColour(s.back_color)
for e in sub.events:
transformEvent(e)
output_fname = os.path.splitext(fname)
output_fname = output_fname[0] + '.hdr.ass'
with open(output_fname, 'w', encoding='utf_8_sig') as f:
sub.dump_file(f)
print(f'Wrote {output_fname}')
def sub_load(inp_fname):
'''
input = ass subtitle file
output = subtitle instance and dict with meta_data for easy access (for later)
'''
subs_stl = {}
#f = io.open(inp_fname, "rb")
#doc = ass.parse(f)
with open(inp_fname, "r", encoding='utf-8') as f:
doc = ass.parse(f)
subs = doc.events
for i in doc.styles:
stl_name = i.name
r = i.primary_color.r
g = i.primary_color.g
b = i.primary_color.b
a = i.primary_color.a
font_name = i.fontname
font_size = i.fontsize
subs_stl.update({str(stl_name): {'r':r, 'g':g, 'b':b, 'a':a,
'font_name':font_name,
'font_size':font_size}})
return(subs, subs_stl)
def 字数漢字本ass(名):
with open(名, 'r') as 本:
字幕 = (文.text for 文 in ass.parse(本).events)
return 字数漢字(字幕)
import argparse
import ass
import datetime
import io
import os
import re
import sys
parser = argparse.ArgumentParser()
parser.add_argument('--file', help='Subtitle file to operate on.', required=True)
parser.add_argument('--operation', help='"add" or "remove".', choices=('add', 'remove'), required=True)
parser.add_argument('--begin', help='Interval begin timestamp as a stringified float (number of seconds).')
parser.add_argument('--end', help='Interval end timestamp as a stringified float (number of seconds).')
args = parser.parse_args()
handle = open(args.file, 'r')
doc = ass.parse(handle.readlines())
handle.close()
begin = datetime.timedelta(seconds=float(args.begin))
end = datetime.timedelta(seconds=float(args.end))
duration = end - begin
assert begin <= end
if args.operation == 'add':
for e in list(doc.events):
if e.start > end: # \---int---/
e.start += duration # [---sub---]: Moved forward
e.end += duration
elif e.start >= begin and e.start <= end: # \-int-/
e.end += duration # [---sub---]: Extended by interval length
def parse(self, file_name, file_encoding):
if file_encoding.lower() != "utf-8":
file_name = get_utf_8_version(file_name, file_encoding)
with open(file_name) as f:
self._events = ass.parse(f).events
