def _read_vtt(self, input_file):
"""
Read from WebVTT file
"""
lines = input_file.readlines()
# ignore the first line containing "WEBVTT" and the following blank line
i = 2
while i < len(lines):
line = lines[i].strip()
if len(line) > 0:
identifier_index = int(line)
identifier = "f" + str(identifier_index).zfill(6)
i += 1
if i < len(lines):
line = lines[i].strip()
timings = line.split(" --> ")
if len(timings) == 2:
begin = gf.time_from_hhmmssmmm(timings[0])
end = gf.time_from_hhmmssmmm(timings[1])
fragment_lines = []
while (i + 1 < len(lines)) and (len(line) > 0):
i += 1
line = lines[i].strip()
if len(line) > 0:
fragment_lines.append(line)
# should never happen, but just in case...
if len(fragment_lines) == 0:
fragment_lines = [""]
text_fragment = TextFragment(identifier=identifier, lines=fragment_lines)
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
i += 1
def _read_smil(self, input_file):
"""
Read from SMIL file.
Limitations:
1. parses only <par> elements, in order
2. timings must have hh:mm:ss.mmm or ss.mmm format (autodetected)
3. both clipBegin and clipEnd attributes of <audio> must be populated
"""
smil_ns = "{http://www.w3.org/ns/SMIL}"
contents = input_file.read()
root = etree.fromstring(contents.encode("utf-8"))
for par in root.iter(smil_ns + "par"):
for child in par:
if child.tag == (smil_ns + "text"):
identifier = gf.split_url(child.get("src"))[1]
elif child.tag == (smil_ns + "audio"):
begin = gf.time_from_hhmmssmmm(child.get("clipBegin"))
if begin is None:
begin = gf.time_from_ssmmm(child.get("clipBegin"))
end = gf.time_from_hhmmssmmm(child.get("clipEnd"))
if end is None:
end = gf.time_from_ssmmm(child.get("clipEnd"))
# TODO read text from additional text_file?
text = u""
text_fragment = TextFragment(identifier=identifier, lines=[text])
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
def _read_srt(self, input_file):
"""
Read from SRT file
"""
lines = input_file.readlines()
i = 0
while i < len(lines):
line = lines[i].strip()
if len(line) > 0:
identifier_index = int(line)
identifier = "f" + str(identifier_index).zfill(6)
i += 1
if i < len(lines):
line = lines[i].strip()
timings = line.split(" --> ")
if len(timings) == 2:
begin = gf.time_from_hhmmssmmm(timings[0], decimal_separator=",")
end = gf.time_from_hhmmssmmm(timings[1], decimal_separator=",")
fragment_lines = []
while (i + 1 < len(lines)) and (len(line) > 0):
i += 1
line = lines[i].strip()
if len(line) > 0:
fragment_lines.append(line)
# should never happen, but just in case...
if len(fragment_lines) == 0:
fragment_lines = [""]
text_fragment = TextFragment(identifier=identifier, lines=fragment_lines)
sm_fragment = SyncMapFragment(text_fragment, begin, end)
self.append(sm_fragment)
i += 1
def parse(self, input_text, syncmap):
"""
Read from SMIL file.
Limitations:
1. parses only ``<par>`` elements, in order
2. timings must have ``hh:mm:ss.mmm`` or ``ss.mmm`` format (autodetected)
3. both ``clipBegin`` and ``clipEnd`` attributes of ``<audio>`` must be populated
"""
from lxml import etree
smil_ns = "{http://www.w3.org/ns/SMIL}"
root = etree.fromstring(gf.safe_bytes(input_text))
for par in root.iter(smil_ns + "par"):
for child in par:
if child.tag == (smil_ns + "text"):
identifier = gf.safe_unicode(gf.split_url(child.get("src"))[1])
elif child.tag == (smil_ns + "audio"):
begin_text = child.get("clipBegin")
if ":" in begin_text:
begin = gf.time_from_hhmmssmmm(begin_text)
else:
begin = gf.time_from_ssmmm(begin_text)
end_text = child.get("clipEnd")
if ":" in end_text:
end = gf.time_from_hhmmssmmm(end_text)
else:
end = gf.time_from_ssmmm(end_text)
# TODO read text from additional text_file?
self._add_fragment(
syncmap=syncmap,
identifier=identifier,
lines=[u""],
begin=begin,
end=end
)
def test_time_from_hhmmssmmm(self):
tests = [
[None, 0.000],
["", 0.000],
["23:45.678", 0.000], # no 2 ":"
["3:45.678", 0.000], # no 2 ":"
["45.678", 0.000], # no 2 ":"
["5.678", 0.000], # no 2 ":"
["5", 0.000], # no 2 ":"
["00:00:01", 0.000], # no "."
["1:23:45.678", 5025.678], # tolerate this (?)
["1:2:45.678", 3765.678], # tolerate this (?)
["1:23:4.678", 4984.678], # tolerate this (?)
["1:23:4.", 4984.000], # tolerate this (?)
["00:00:00.000", 0.000],
["00:00:12.000", 12.000],
["00:00:12.345", 12.345],
["00:01:00.000", 60],
["00:01:23.000", 83.000],
["00:01:23.456", 83.456],
["01:00:00.000", 3600.000],
["01:00:12.000", 3612.000],
["01:00:12.345", 3612.345],
["01:23:00.000", 4980.000],
["01:23:45.000", 5025.000],
["01:23:45.678", 5025.678],
]
for test in tests:
self.assertEqual(gf.time_from_hhmmssmmm(test[0]), test[1])
def test_time_from_hhmmssmmm(self):
tests = [
(None, TimeValue("0.000")),
("", TimeValue("0.000")),
("23:45.678", TimeValue("0.000")), # no 2 ":"
("3:45.678", TimeValue("0.000")), # no 2 ":"
("45.678", TimeValue("0.000")), # no 2 ":"
("5.678", TimeValue("0.000")), # no 2 ":"
("5", TimeValue("0.000")), # no 2 ":"
("00:00:01", TimeValue("0.000")), # no "."
("1:23:45.678", TimeValue("5025.678")), # tolerate this (?)
("1:2:45.678", TimeValue("3765.678")), # tolerate this (?)
("1:23:4.678", TimeValue("4984.678")), # tolerate this (?)
("1:23:4.", TimeValue("4984.000")), # tolerate this (?)
("00:00:00.000", TimeValue("0.000")),
("00:00:12.000", TimeValue("12.000")),
("00:00:12.345", TimeValue("12.345")),
("00:01:00.000", TimeValue("60")),
("00:01:23.000", TimeValue("83.000")),
("00:01:23.456", TimeValue("83.456")),
("01:00:00.000", TimeValue("3600.000")),
("01:00:12.000", TimeValue("3612.000")),
("01:00:12.345", TimeValue("3612.345")),
("01:23:00.000", TimeValue("4980.000")),
("01:23:45.000", TimeValue("5025.000")),
("01:23:45.678", TimeValue("5025.678")),
]
for test in tests:
self.assertEqual(gf.time_from_hhmmssmmm(test[0]), test[1])
def test_time_from_hhmmssmmm(self):
tests = [
[None, 0.000],
["", 0.000],
["23:45.678", 0.000], # no 2 ":"
["3:45.678", 0.000], # no 2 ":"
["45.678", 0.000], # no 2 ":"
["5.678", 0.000], # no 2 ":"
["5", 0.000], # no 2 ":"
["00:00:01", 0.000], # no "."
["1:23:45.678", 5025.678], # tolerate this (?)
["1:2:45.678", 3765.678], # tolerate this (?)
["1:23:4.678", 4984.678], # tolerate this (?)
["1:23:4.", 4984.000], # tolerate this (?)
["00:00:00.000", 0.000],
["00:00:12.000", 12.000],
["00:00:12.345", 12.345],
["00:01:00.000", 60],
["00:01:23.000", 83.000],
["00:01:23.456", 83.456],
["01:00:00.000", 3600.000],
["01:00:12.000", 3612.000],
["01:00:12.345", 3612.345],
["01:23:00.000", 4980.000],
["01:23:45.000", 5025.000],
["01:23:45.678", 5025.678],
]
for test in tests:
self.assertEqual(gf.time_from_hhmmssmmm(test[0]), test[1])
def test_time_from_hhmmssmmm(self):
tests = [
(None, TimeValue("0.000")),
("", TimeValue("0.000")),
("23:45.678", TimeValue("0.000")), # no 2 ":"
("3:45.678", TimeValue("0.000")), # no 2 ":"
("45.678", TimeValue("0.000")), # no 2 ":"
("5.678", TimeValue("0.000")), # no 2 ":"
("5", TimeValue("0.000")), # no 2 ":"
("00:00:01", TimeValue("0.000")), # no "."
("1:23:45.678", TimeValue("5025.678")), # tolerate this (?)
("1:2:45.678", TimeValue("3765.678")), # tolerate this (?)
("1:23:4.678", TimeValue("4984.678")), # tolerate this (?)
("1:23:4.", TimeValue("4984.000")), # tolerate this (?)
("00:00:00.000", TimeValue("0.000")),
("00:00:12.000", TimeValue("12.000")),
("00:00:12.345", TimeValue("12.345")),
("00:01:00.000", TimeValue("60")),
("00:01:23.000", TimeValue("83.000")),
("00:01:23.456", TimeValue("83.456")),
("01:00:00.000", TimeValue("3600.000")),
("01:00:12.000", TimeValue("3612.000")),
("01:00:12.345", TimeValue("3612.345")),
("01:23:00.000", TimeValue("4980.000")),
("01:23:45.000", TimeValue("5025.000")),
("01:23:45.678", TimeValue("5025.678")),
]
for test in tests:
self.assertEqual(gf.time_from_hhmmssmmm(test[0]), test[1])
def read_properties(self, audio_file_path):
"""
Read the properties of an audio file
and return them as a dictionary.
Example: ::
d["index"]=0
d["codec_name"]=mp3
d["codec_long_name"]=MP3 (MPEG audio layer 3)
d["profile"]=unknown
d["codec_type"]=audio
d["codec_time_base"]=1/44100
d["codec_tag_string"]=[0][0][0][0]
d["codec_tag"]=0x0000
d["sample_fmt"]=s16p
d["sample_rate"]=44100
d["channels"]=1
d["channel_layout"]=mono
d["bits_per_sample"]=0
d["id"]=N/A
d["r_frame_rate"]=0/0
d["avg_frame_rate"]=0/0
d["time_base"]=1/14112000
d["start_pts"]=0
d["start_time"]=0.000000
d["duration_ts"]=1545083190
d["duration"]=109.487188
d["bit_rate"]=128000
d["max_bit_rate"]=N/A
d["bits_per_raw_sample"]=N/A
d["nb_frames"]=N/A
d["nb_read_frames"]=N/A
d["nb_read_packets"]=N/A
d["DISPOSITION:default"]=0
d["DISPOSITION:dub"]=0
d["DISPOSITION:original"]=0
d["DISPOSITION:comment"]=0
d["DISPOSITION:lyrics"]=0
d["DISPOSITION:karaoke"]=0
d["DISPOSITION:forced"]=0
d["DISPOSITION:hearing_impaired"]=0
d["DISPOSITION:visual_impaired"]=0
d["DISPOSITION:clean_effects"]=0
d["DISPOSITION:attached_pic"]=0
:param string audio_file_path: the path of the audio file to analyze
:rtype: dict
:raises: TypeError: if ``audio_file_path`` is None
:raises: OSError: if the file at ``audio_file_path`` cannot be read
:raises: FFPROBEParsingError: if the call to ``ffprobe`` does not produce any output
:raises: FFPROBEPathError: if the path to the ``ffprobe`` executable cannot be called
:raises: FFPROBEUnsupportedFormatError: if the file has a format not supported by ``ffprobe``
"""
# test if we can read the file at audio_file_path
if audio_file_path is None:
self.log_exc(u"The audio file path is None", None, True, TypeError)
if not gf.file_can_be_read(audio_file_path):
self.log_exc(u"Input file '%s' cannot be read" % (audio_file_path), None, True, OSError)
# call ffprobe
arguments = [self.rconf[RuntimeConfiguration.FFPROBE_PATH]]
arguments.extend(self.FFPROBE_PARAMETERS)
arguments.append(audio_file_path)
self.log([u"Calling with arguments '%s'", arguments])
try:
proc = subprocess.Popen(
arguments,
stdout=subprocess.PIPE,
stdin=subprocess.PIPE,
stderr=subprocess.PIPE
)
(stdoutdata, stderrdata) = proc.communicate()
proc.stdout.close()
proc.stdin.close()
proc.stderr.close()
except OSError as exc:
self.log_exc(u"Unable to call the '%s' ffprobe executable" % (self.rconf[RuntimeConfiguration.FFPROBE_PATH]), exc, True, FFPROBEPathError)
self.log(u"Call completed")
# check there is some output
if (stdoutdata is None) or (len(stderrdata) == 0):
self.log_exc(u"ffprobe produced no output", None, True, FFPROBEParsingError)
# decode stdoutdata and stderrdata to Unicode string
try:
stdoutdata = gf.safe_unicode(stdoutdata)
stderrdata = gf.safe_unicode(stderrdata)
except UnicodeDecodeError as exc:
self.log_exc(u"Unable to decode ffprobe out/err", exc, True, FFPROBEParsingError)
# dictionary for the results
results = {
self.STDOUT_CHANNELS : None,
self.STDOUT_CODEC_NAME : None,
self.STDOUT_DURATION : None,
self.STDOUT_SAMPLE_RATE : None
}
# scan the first audio stream the ffprobe stdout output
# TODO more robust parsing
# TODO deal with multiple audio streams
for line in stdoutdata.splitlines():
if line == self.STDOUT_END_STREAM:
self.log(u"Reached end of the stream")
break
elif len(line.split("=")) == 2:
key, value = line.split("=")
results[key] = value
self.log([u"Found property '%s'='%s'", key, value])
try:
self.log([u"Duration found in stdout: '%s'", results[self.STDOUT_DURATION]])
results[self.STDOUT_DURATION] = TimeValue(results[self.STDOUT_DURATION])
self.log(u"Valid duration")
except:
self.log_warn(u"Invalid duration")
results[self.STDOUT_DURATION] = None
# try scanning ffprobe stderr output
for line in stderrdata.splitlines():
match = self.STDERR_DURATION_REGEX.search(line)
if match is not None:
self.log([u"Found matching line '%s'", line])
results[self.STDOUT_DURATION] = gf.time_from_hhmmssmmm(line)
self.log([u"Extracted duration '%.3f'", results[self.STDOUT_DURATION]])
break
if results[self.STDOUT_DURATION] is None:
self.log_exc(u"No duration found in stdout or stderr. Unsupported audio file format?", None, True, FFPROBEUnsupportedFormatError)
# return dictionary
self.log(u"Returning dict")
return results
def read_properties(self, audio_file_path):
"""
Read the properties of an audio file
and return them as a dictionary.
Example: ::
d["index"]=0
d["codec_name"]=mp3
d["codec_long_name"]=MP3 (MPEG audio layer 3)
d["profile"]=unknown
d["codec_type"]=audio
d["codec_time_base"]=1/44100
d["codec_tag_string"]=[0][0][0][0]
d["codec_tag"]=0x0000
d["sample_fmt"]=s16p
d["sample_rate"]=44100
d["channels"]=1
d["channel_layout"]=mono
d["bits_per_sample"]=0
d["id"]=N/A
d["r_frame_rate"]=0/0
d["avg_frame_rate"]=0/0
d["time_base"]=1/14112000
d["start_pts"]=0
d["start_time"]=0.000000
d["duration_ts"]=1545083190
d["duration"]=109.487188
d["bit_rate"]=128000
d["max_bit_rate"]=N/A
d["bits_per_raw_sample"]=N/A
d["nb_frames"]=N/A
d["nb_read_frames"]=N/A
d["nb_read_packets"]=N/A
d["DISPOSITION:default"]=0
d["DISPOSITION:dub"]=0
d["DISPOSITION:original"]=0
d["DISPOSITION:comment"]=0
d["DISPOSITION:lyrics"]=0
d["DISPOSITION:karaoke"]=0
d["DISPOSITION:forced"]=0
d["DISPOSITION:hearing_impaired"]=0
d["DISPOSITION:visual_impaired"]=0
d["DISPOSITION:clean_effects"]=0
d["DISPOSITION:attached_pic"]=0
:param string audio_file_path: the path of the audio file to analyze
:rtype: dict
:raises: TypeError: if ``audio_file_path`` is None
:raises: OSError: if the file at ``audio_file_path`` cannot be read
:raises: FFPROBEParsingError: if the call to ``ffprobe`` does not produce any output
:raises: FFPROBEPathError: if the path to the ``ffprobe`` executable cannot be called
:raises: FFPROBEUnsupportedFormatError: if the file has a format not supported by ``ffprobe``
"""
# test if we can read the file at audio_file_path
if audio_file_path is None:
self.log_exc(u"The audio file path is None", None, True, TypeError)
if not gf.file_can_be_read(audio_file_path):
self.log_exc(u"Input file '%s' cannot be read" % (audio_file_path),
None, True, OSError)
# call ffprobe
arguments = [self.rconf[RuntimeConfiguration.FFPROBE_PATH]]
arguments.extend(self.FFPROBE_PARAMETERS)
arguments.append(audio_file_path)
self.log([u"Calling with arguments '%s'", arguments])
try:
proc = subprocess.Popen(arguments,
stdout=subprocess.PIPE,
stdin=subprocess.PIPE,
stderr=subprocess.PIPE)
(stdoutdata, stderrdata) = proc.communicate()
proc.stdout.close()
proc.stdin.close()
proc.stderr.close()
except OSError as exc:
self.log_exc(
u"Unable to call the '%s' ffprobe executable" %
(self.rconf[RuntimeConfiguration.FFPROBE_PATH]), exc, True,
FFPROBEPathError)
self.log(u"Call completed")
# check there is some output
if (stdoutdata is None) or (len(stderrdata) == 0):
self.log_exc(u"ffprobe produced no output", None, True,
FFPROBEParsingError)
# decode stdoutdata and stderrdata to Unicode string
try:
stdoutdata = gf.safe_unicode(stdoutdata)
stderrdata = gf.safe_unicode(stderrdata)
except UnicodeDecodeError as exc:
self.log_exc(u"Unable to decode ffprobe out/err", exc, True,
FFPROBEParsingError)
# dictionary for the results
results = {
self.STDOUT_CHANNELS: None,
self.STDOUT_CODEC_NAME: None,
self.STDOUT_DURATION: None,
self.STDOUT_SAMPLE_RATE: None
}
# scan the first audio stream the ffprobe stdout output
# TODO more robust parsing
# TODO deal with multiple audio streams
for line in stdoutdata.splitlines():
if line == self.STDOUT_END_STREAM:
self.log(u"Reached end of the stream")
break
elif len(line.split("=")) == 2:
key, value = line.split("=")
results[key] = value
self.log([u"Found property '%s'='%s'", key, value])
try:
self.log([
u"Duration found in stdout: '%s'",
results[self.STDOUT_DURATION]
])
results[self.STDOUT_DURATION] = TimeValue(
results[self.STDOUT_DURATION])
self.log(u"Valid duration")
except:
self.log_warn(u"Invalid duration")
results[self.STDOUT_DURATION] = None
# try scanning ffprobe stderr output
for line in stderrdata.splitlines():
match = self.STDERR_DURATION_REGEX.search(line)
if match is not None:
self.log([u"Found matching line '%s'", line])
results[self.STDOUT_DURATION] = gf.time_from_hhmmssmmm(
line)
self.log([
u"Extracted duration '%.3f'",
results[self.STDOUT_DURATION]
])
break
if results[self.STDOUT_DURATION] is None:
self.log_exc(
u"No duration found in stdout or stderr. Unsupported audio file format?",
None, True, FFPROBEUnsupportedFormatError)
# return dictionary
self.log(u"Returning dict")
return results
