def OnESPacket(current_pid, packet, header_size):
"""
Callback invoked on the successful extraction of an Elementary Stream packet from the
Transport Stream file packets.
:param current_pid: The TS Program ID for the TS packets this info originated from
:param packet: The ENTIRE ES packet, header and payload-- which may have been assembled
from multiple TS packet payloads.
:param header_size: Size of the header in bytes (characters in the string). Provided to more
easily separate the packet into header and payload.
:return: None
"""
global pid
global VERBOSE
global SILENT
global elapsed_time_s
if pid >= 0 and current_pid != pid:
return
try:
payload = ES.get_pes_payload(packet)
f = list(payload)
#f = bytearray(payload)
data_group = DataGroup(f)
if not data_group.is_management_data():
#We now have a Data Group that contains caption data.
#We take out its payload, but this is further divided into 'Data Unit' structures
caption = data_group.payload()
#iterate through the Data Units in this payload via another generator.
for data_unit in next_data_unit(caption):
#we're only interested in those Data Units which are "statement body" to get CC data.
if not isinstance(data_unit.payload(), StatementBody):
continue
#okay. Finally we've got a data unit with CC data. Feed its payload to the custom
if pid < 0 and VERBOSE and not SILENT:
pid = current_pid
print("Found Closed Caption data in PID: " + str(pid))
print("Will now only process this PID to improve performance.")
#formatter function above. This dumps the basic text to stdout.
cc = formatter(data_unit.payload().payload(), elapsed_time_s)
if cc and VERBOSE:
#according to best practice, always deal internally with UNICODE, and encode to
#your encoding of choice as late as possible. Here, i'm encoding as UTF-8 for
#my command line.
#DECODE EARLY, ENCODE LATE
print(cc.encode('utf-8'))
except EOFError:
pass
except Exception, err:
if VERBOSE and not SILENT and pid >= 0:
print("Exception thrown while handling DataGroup in ES. This may be due to many factors"
+ "such as file corruption or the .ts file using as yet unsupported features.")
traceback.print_exc(file=sys.stdout)
def OnESPacket(current_pid, packet, header_size):
"""
Callback invoked on the successful extraction of an Elementary Stream packet from the
Transport Stream file packets.
:param current_pid: The TS Program ID for the TS packets this info originated from
:param packet: The ENTIRE ES packet, header and payload-- which may have been assembled
from multiple TS packet payloads.
:param header_size: Size of the header in bytes (characters in the string). Provided to more
easily separate the packet into header and payload.
:return: None
"""
global pid
global VERBOSE
global SILENT
global elapsed_time_s
global ass
global infilename
global outfilename
global tmax
global time_offset
if pid >= 0 and current_pid != pid:
return
try:
payload = ES.get_pes_payload(packet)
f = list(payload)
data_group = DataGroup(f)
if not data_group.is_management_data():
#We now have a Data Group that contains caption data.
#We take out its payload, but this is further divided into 'Data Unit' structures
caption = data_group.payload()
#iterate through the Data Units in this payload via another generator.
for data_unit in next_data_unit(caption):
#we're only interested in those Data Units which are "statement body" to get CC data.
if not isinstance(data_unit.payload(), StatementBody):
continue
if not ass:
v = not SILENT
ass = ASSFormatter(tmax=tmax, video_filename=outfilename, verbose=v)
ass.format(data_unit.payload().payload(), elapsed_time_s)
# this code used to sed the PID we're scanning via first successful ARIB decode
# but i've changed it below to draw present CC language info form ARIB
# management data. Leaving this here for reference.
#if pid < 0 and not SILENT:
# pid = current_pid
# print("Found Closed Caption data in PID: " + str(pid))
# print("Will now only process this PID to improve performance.")
else:
# management data
management_data = data_group.payload()
numlang = management_data.num_languages()
if pid < 0 and numlang > 0:
for language in range(numlang):
if not SILENT:
print("Closed caption management data for language: "
+ management_data.language_code(language)
+ " available in PID: " + str(current_pid))
print("Will now only process this PID to improve performance.")
pid = current_pid
except EOFError:
pass
except FileOpenError as ex:
# allow IOErrors to kill application
raise ex
except Exception, err:
if not SILENT and pid >= 0:
print("Exception thrown while handling DataGroup in ES. This may be due to many factors"
+ "such as file corruption or the .ts file using as yet unsupported features.")
traceback.print_exc(file=sys.stdout)
def OnESPacket(current_pid, packet, header_size):
"""
Callback invoked on the successful extraction of an Elementary Stream packet from the
Transport Stream file packets.
:param current_pid: The TS Program ID for the TS packets this info originated from
:param packet: The ENTIRE ES packet, header and payload-- which may have been assembled
from multiple TS packet payloads.
:param header_size: Size of the header in bytes (characters in the string). Provided to more
easily separate the packet into header and payload.
:return: None
"""
global pid
global VERBOSE
global SILENT
global elapsed_time_s
global ass
global infilename
global outfilename
global tmax
global time_offset
if pid >= 0 and current_pid != pid:
return
try:
payload = ES.get_pes_payload(packet)
f = list(payload)
data_group = DataGroup(f)
if not data_group.is_management_data():
#We now have a Data Group that contains caption data.
#We take out its payload, but this is further divided into 'Data Unit' structures
caption = data_group.payload()
#iterate through the Data Units in this payload via another generator.
for data_unit in next_data_unit(caption):
#we're only interested in those Data Units which are "statement body" to get CC data.
if not isinstance(data_unit.payload(), StatementBody):
continue
if not ass:
v = not SILENT
ass = ASSFormatter(tmax=tmax, video_filename=outfilename, verbose=v)
ass.format(data_unit.payload().payload(), elapsed_time_s)
# this code used to sed the PID we're scanning via first successful ARIB decode
# but i've changed it below to draw present CC language info form ARIB
# management data. Leaving this here for reference.
#if pid < 0 and not SILENT:
# pid = current_pid
# print("Found Closed Caption data in PID: " + str(pid))
# print("Will now only process this PID to improve performance.")
else:
# management data
management_data = data_group.payload()
numlang = management_data.num_languages()
if pid < 0 and numlang > 0:
for language in range(numlang):
if not SILENT:
print("Closed caption management data for language: "
+ management_data.language_code(language)
+ " available in PID: " + str(current_pid))
print("Will now only process this PID to improve performance.")
pid = current_pid
except EOFError:
pass
except FileOpenError as ex:
# allow IOErrors to kill application
raise ex
except Exception, err:
if not SILENT and pid >= 0:
print("Exception thrown while handling DataGroup in ES. This may be due to many factors"
+ "such as file corruption or the .ts file using as yet unsupported features.")
traceback.print_exc(file=sys.stdout)