def __init__(self, index=0, start=None, end=None, text='', position=''):
try:
self.index = int(index)
except (TypeError, ValueError): # try to cast as int, but it's not mandatory
self.index = index
self.start = SubRipTime.coerce(start or 0)
self.end = SubRipTime.coerce(end or 0)
self.position = str(position)
self.text = str(text)
def __init__(self, index=0, start=None, end=None, text='', position=''):
try:
self.index = int(index)
except (TypeError, ValueError): # try to cast as int, but it's not mandatory
self.index = index
self.start = SubRipTime.coerce(start or 0)
self.end = SubRipTime.coerce(end or 0)
self.position = str(position)
self.text = str(text)
def write_into(self, output_file, eol=None):
"""
write_into(output_file [, eol])
Serialize current state into `output_file`.
`output_file` -> Any instance that respond to `write()`, typically a
file object
"""
output_eol = eol or self.eol
for item in self:
string_repr = str(item)
if output_eol != '\n':
string_repr = string_repr.replace('\n', output_eol)
output_file.write(string_repr)
# Only add trailing eol if it's not already present.
# It was kept in the SubRipItem's text before but it really
# belongs here. Existing applications might give us subtitles
# which already contain a trailing eol though.
if not string_repr.endswith(2 * output_eol):
output_file.write(output_eol)
def write_into(self, output_file, eol=None):
"""
write_into(output_file [, eol])
Serialize current state into `output_file`.
`output_file` -> Any instance that respond to `write()`, typically a
file object
"""
output_eol = eol or self.eol
for item in self:
string_repr = str(item)
if output_eol != '\n':
string_repr = string_repr.replace('\n', output_eol)
output_file.write(string_repr)
# Only add trailing eol if it's not already present.
# It was kept in the SubRipItem's text before but it really
# belongs here. Existing applications might give us subtitles
# which already contain a trailing eol though.
if not string_repr.endswith(2 * output_eol):
output_file.write(output_eol)
except ImportError:
from UserList import UserList
from itertools import chain
from copy import copy
from pysrt.srtexc import Error
from pysrt.srtitem import SubRipItem
from pysrt.compat import str
BOMS = ((codecs.BOM_UTF32_LE, 'utf_32_le'),
(codecs.BOM_UTF32_BE, 'utf_32_be'),
(codecs.BOM_UTF16_LE, 'utf_16_le'),
(codecs.BOM_UTF16_BE, 'utf_16_be'),
(codecs.BOM_UTF8, 'utf_8'))
CODECS_BOMS = dict((codec, str(bom, codec)) for bom, codec in BOMS)
BIGGER_BOM = max(len(bom) for bom, encoding in BOMS)
class SubRipFile(UserList, object):
"""
SubRip file descriptor.
Provide a pure Python mapping on all metadata.
SubRipFile(items, eol, path, encoding)
items -> list of SubRipItem. Default to [].
eol -> str: end of line character. Default to linesep used in opened file
if any else to os.linesep.
path -> str: path where file will be saved. To open an existant file see
def test_idempotence(self):
vtt = SubRipItem.from_string(self.vtt)
self.assertEqual(str(vtt), self.vtt)
item = SubRipItem.from_string(self.coordinates)
self.assertEqual(str(item), self.coordinates)
def test_serialization(self):
self.assertEqual(str(self.item), self.string)
def test_compare_from_string_and_from_path(self):
unicode_content = codecs.open(self.utf8_path, encoding='utf_8').read()
iterator = zip(pysrt.open(self.utf8_path),
pysrt.from_string(unicode_content))
for file_item, string_item in iterator:
self.assertEqual(str(file_item), str(string_item))
def __str__(self):
if self.ordinal < 0:
# Represent negative times as zero
return str(SubRipTime.from_ordinal(0))
return self.TIME_PATTERN % tuple(self)
def __str__(self):
if self.ordinal < 0:
# Represent negative times as zero
return str(SubRipTime.from_ordinal(0))
return self.TIME_PATTERN % tuple(self)
except ImportError:
from UserList import UserList
from itertools import chain
from copy import copy
from pysrt.srtexc import Error
from pysrt.srtitem import SubRipItem
from pysrt.compat import str
BOMS = ((codecs.BOM_UTF32_LE, 'utf_32_le'),
(codecs.BOM_UTF32_BE, 'utf_32_be'),
(codecs.BOM_UTF16_LE, 'utf_16_le'),
(codecs.BOM_UTF16_BE, 'utf_16_be'),
(codecs.BOM_UTF8, 'utf_8'))
CODECS_BOMS = dict((codec, str(bom, codec)) for bom, codec in BOMS)
BIGGER_BOM = max(len(bom) for bom, encoding in BOMS)
class SubRipFile(UserList, object):
"""
SubRip file descriptor.
Provide a pure Python mapping on all metadata.
SubRipFile(items, eol, path, encoding)
items -> list of SubRipItem. Default to [].
eol -> str: end of line character. Default to linesep used in opened file
if any else to os.linesep.
path -> str: path where file will be saved. To open an existant file see
def test_compare_from_string_and_from_path(self):
unicode_content = codecs.open(self.utf8_path, encoding="utf_8").read()
iterator = zip(pysrt.open(self.utf8_path), pysrt.from_string(unicode_content))
for file_item, string_item in iterator:
self.assertEqual(str(file_item), str(string_item))
def test_idempotence(self):
vtt = SubRipItem.from_string(self.vtt)
self.assertEqual(str(vtt), self.vtt)
item = SubRipItem.from_string(self.coordinates)
self.assertEqual(str(item), self.coordinates)
def test_serialization(self):
self.assertEqual(str(self.item), self.string)
class SubRipItem(ComparableMixin):
"""
SubRipItem(index, start, end, text, position)
index -> int: index of item in file. 0 by default.
start, end -> SubRipTime or coercible.
text -> unicode: text content for item.
position -> unicode: raw srt/vtt "display coordinates" string
"""
ITEM_PATTERN = str('%s\n%s --> %s%s\n%s\n')
TIMESTAMP_SEPARATOR = '-->'
def __init__(self, index=0, start=None, end=None, text='', position=''):
try:
self.index = int(index)
except (TypeError,
ValueError): # try to cast as int, but it's not mandatory
self.index = index
self.start = SubRipTime.coerce(start or 0)
self.end = SubRipTime.coerce(end or 0)
self.position = str(position)
self.text = str(text)
@property
def duration(self):
return self.end - self.start
@property
def text_without_tags(self):
RE_TAG = re.compile(r'<[^>]*?>')
return RE_TAG.sub('', self.text)
@property
def characters_per_second(self):
characters_count = len(self.text_without_tags.replace('\n', ''))
try:
return characters_count / (self.duration.ordinal / 1000.0)
except ZeroDivisionError:
return 0.0
def __str__(self):
position = ' %s' % self.position if self.position.strip() else ''
return self.ITEM_PATTERN % (self.index, self.start, self.end, position,
self.text)
if is_py2:
__unicode__ = __str__
def __str__(self):
raise NotImplementedError('Use unicode() instead!')
def _cmpkey(self):
return (self.start, self.end)
def shift(self, *args, **kwargs):
"""
shift(hours, minutes, seconds, milliseconds, ratio)
Add given values to start and end attributes.
All arguments are optional and have a default value of 0.
"""
self.start.shift(*args, **kwargs)
self.end.shift(*args, **kwargs)
@classmethod
def from_string(cls, source):
return cls.from_lines(source.splitlines(True))
@classmethod
def from_lines(cls, lines):
if len(lines) < 2:
raise InvalidItem()
lines = [l.rstrip() for l in lines]
index = None
if cls.TIMESTAMP_SEPARATOR not in lines[0]:
index = lines.pop(0)
start, end, position = cls.split_timestamps(lines[0])
body = '\n'.join(lines[1:])
return cls(index, start, end, body, position)
@classmethod
def split_timestamps(cls, line):
timestamps = line.split(cls.TIMESTAMP_SEPARATOR)
if len(timestamps) != 2:
raise InvalidItem()
start, end_and_position = timestamps
end_and_position = end_and_position.lstrip().split(' ', 1)
end = end_and_position[0]
position = end_and_position[1] if len(end_and_position) > 1 else ''
return (s.strip() for s in (start, end, position))