def _parse(self, lines):
self._compute_blocks(lines)
for block in self.blocks:
if self._is_cue_block(block):
caption = self._parse_cue_block(block)
self.captions.append(caption)
elif self._is_comment_block(block):
continue
elif self._is_style_block(block):
if self.captions:
raise MalformedFileError(
'Style block defined after the first cue in line {}.'.
format(block.line_number))
style = Style()
style.lines = block.lines[1:]
self.styles.append(style)
else:
if len(block.lines) == 1:
raise MalformedCaptionError(
'Standalone cue identifier in line {}.'.format(
block.line_number))
else:
raise MalformedCaptionError(
'Missing timing cue in line {}.'.format(
block.line_number + 1))
def _parse(self, lines):
c = None
for index, line in enumerate(lines):
if self._should_skip_line(line, index, c): # allow child classes to skip lines based on the content
continue
if self._is_timeframe_line(line):
try:
start, end = self._parse_timeframe_line(line)
except MalformedCaptionError as e:
raise MalformedCaptionError('{} in line! {}'.format(e, index + 1))
c = Caption(start, end)
elif line:
if c is None:
raise MalformedCaptionError('Caption missing timeframe in line {}.'.format(index + 1))
else:
c.add_line(line)
else:
if c is None:
continue
if not c.lines:
raise MalformedCaptionError('Caption missing text in line {}.'.format(index + 1))
self.captions.append(c)
c = None
if c is not None and c.lines:
self.captions.append(c)
def _parse_timeframe_line(self, line):
"""Parse timeframe line and return start and end timestamps."""
tf = self._validate_timeframe_line(line)
if not tf:
raise MalformedCaptionError('Invalid time format')
return tf.group(1), tf.group(2)
def _parse_timestamp(self, timestamp):
res = re.match(TIMESTAMP_PATTERN, timestamp)
if not res:
raise MalformedCaptionError('Invalid timestamp: {}'.format(timestamp))
values = list(map(lambda x: int(x) if x else 0, res.groups()))
return self._to_seconds(*values)
def _parse_timestamp(self, timestamp):
res = re.match(TIMESTAMP_PATTERN, timestamp)
if not res:
raise MalformedCaptionError(
'Invalid timestamp: {}'.format(timestamp))
return self._to_seconds(
int(res.group(1)), # hours
int(res.group(2)), # minutes
int(res.group(3)), # seconds
int(res.group(4)) # milliseconds
)
def _should_skip_line(self, line, index, caption):
is_header_title = index == 0 and line == 'WEBVTT'
is_header_data = len(self.captions) == 0 and caption is None
if self.expect_timeframe:
raise MalformedCaptionError(
'Caption missing timeframe in line {}.'.format(index + 1))
is_cue_identifier = caption is None and not is_header_title and not is_header_data
if is_cue_identifier:
self.expect_timeframe = True
return is_header_title or is_header_data or is_cue_identifier
def _parse_cue_block(self, block):
caption = Caption()
cue_timings = None
for line_number, line in enumerate(block.lines):
if self._is_cue_timings_line(line):
if cue_timings is None:
try:
cue_timings = self._parse_timeframe_line(line)
except MalformedCaptionError as e:
raise MalformedCaptionError('{} in line {}'.format(
e, block.line_number + line_number))
else:
raise MalformedCaptionError(
'--> found in line {}'.format(block.line_number +
line_number))
elif line_number == 0:
caption.identifier = line
else:
caption.add_line(line)
caption.start = cue_timings[0]
caption.end = cue_timings[1]
return caption