def _dd(self, file_name, offset, size, extension, output_file_name=None):
'''
Extracts a file embedded inside the target file.
@file_name - Path to the target file.
@offset - Offset inside the target file where the embedded file begins.
@size - Number of bytes to extract.
@extension - The file exension to assign to the extracted file on disk.
@output_file_name - The requested name of the output file.
Returns the extracted file name.
'''
total_size = 0
# Default extracted file name is <hex offset>.<extension>
default_bname = "%X" % offset
if self.max_size and size > self.max_size:
size = self.max_size
if not output_file_name or output_file_name is None:
bname = default_bname
else:
# Strip the output file name of invalid/dangerous characters (like file paths)
bname = os.path.basename(output_file_name)
fname = unique_file_name(bname, extension)
try:
# Open the target file and seek to the offset
fdin = self.config.open_file(file_name, length=size, offset=offset)
# Open the output file
try:
fdout = BlockFile(fname, 'w')
except KeyboardInterrupt as e:
raise e
except Exception as e:
# Fall back to the default name if the requested name fails
fname = unique_file_name(default_bname, extension)
fdout = BlockFile(fname, 'w')
while total_size < size:
(data, dlen) = fdin.read_block()
if not data:
break
else:
fdout.write(str2bytes(data[:dlen]))
total_size += dlen
# Cleanup
fdout.close()
fdin.close()
except KeyboardInterrupt as e:
raise e
except Exception as e:
raise Exception(
"Extractor.dd failed to extract data from '%s' to '%s': %s" %
(file_name, fname, str(e)))
return fname
def _dd(self, file_name, offset, size, extension, output_file_name=None):
'''
Extracts a file embedded inside the target file.
@file_name - Path to the target file.
@offset - Offset inside the target file where the embedded file begins.
@size - Number of bytes to extract.
@extension - The file exension to assign to the extracted file on disk.
@output_file_name - The requested name of the output file.
Returns the extracted file name.
'''
total_size = 0
# Default extracted file name is <hex offset>.<extension>
default_bname = "%X" % offset
if self.max_size and size > self.max_size:
size = self.max_size
if not output_file_name or output_file_name is None:
bname = default_bname
else:
# Strip the output file name of invalid/dangerous characters (like file paths)
bname = os.path.basename(output_file_name)
fname = unique_file_name(bname, extension)
try:
# Open the target file and seek to the offset
fdin = self.config.open_file(file_name, length=size, offset=offset)
# Open the output file
try:
fdout = BlockFile(fname, 'w')
except KeyboardInterrupt as e:
raise e
except Exception as e:
# Fall back to the default name if the requested name fails
fname = unique_file_name(default_bname, extension)
fdout = BlockFile(fname, 'w')
while total_size < size:
(data, dlen) = fdin.read_block()
if not data:
break
else:
fdout.write(str2bytes(data[:dlen]))
total_size += dlen
# Cleanup
fdout.close()
fdin.close()
except KeyboardInterrupt as e:
raise e
except Exception as e:
raise Exception("Extractor.dd failed to extract data from '%s' to '%s': %s" % (file_name, fname, str(e)))
binwalk.core.common.debug("Carved data block 0x%X - 0x%X from '%s' to '%s'" % (offset, offset+size, file_name, fname))
return fname
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
output_directory = os.path.join(os.path.dirname(path), unique_file_name('_' + os.path.basename(path), extension='extracted'))
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
# Set the initial base extraction directory for later determining the level of recusion
if not self.base_recursion_dir:
self.base_recursion_dir = os.path.realpath(output_directory) + os.path.sep
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
outdir = os.path.join(basedir, '_' + basename)
output_directory = unique_file_name(outdir, extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
# Set the initial base extraction directory for later determining the level of recusion
# TODO: This is no longer needed since self.directory has the same information. Update code accordingly.
if not self.base_recursion_dir:
self.base_recursion_dir = os.path.realpath(output_directory) + os.path.sep
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
output_directory = os.path.join(
os.path.dirname(path),
unique_file_name('_' + os.path.basename(path),
extension='extracted'))
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
# Set the initial base extraction directory for later determining the level of recusion
if not self.base_recursion_dir:
self.base_recursion_dir = os.path.realpath(
output_directory) + os.path.sep
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
outdir = os.path.join(basedir, '_' + basename)
output_directory = unique_file_name(outdir, extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
# Set the initial base extraction directory for later determining the level of recusion
# TODO: This is no longer needed since self.directory has the same information. Update code accordingly.
if not self.base_recursion_dir:
self.base_recursion_dir = os.path.realpath(output_directory) + os.path.sep
return output_directory
def extract_v1(self):
for l in self.layout:
if l.filename is not None:
self.fd.seek(l.offset)
out_filename = unique_file_name(l.filename.decode())
with open(out_filename, 'w+b') as out_fd:
n, cz = 0, self.chunksize
while n < l.size:
n += cz
if n > l.size:
cz -= n - l.size
out_fd.write(str2bytes(self.fd.read(cz)))
if self.password_extractor is not None:
if l.type == CSP_LAYOUT_PASSWORD:
try:
self.password = check_output(
[self.password_extractor, l.filename])
except CalledProcessError as e:
warning(str(e))
elif l.type == CSP_LAYOUT_APPL:
if self.password is None:
warning('Password wasn\'t extracted')
continue
app_out_filename = splitext(out_filename)[0]
self.decrypt_v1(out_filename, l.size, app_out_filename)
os.unlink(out_filename)
self.untar(app_out_filename,
splitext(app_out_filename)[0])
def extract(self):
if not self.valid:
return False
for sec in self.sections:
if sec.type in [
SECTION_TYPE_ARCHIVE_TBZ2,
SECTION_TYPE_ENCRYPTED_ARCHIVE_TBZ2
]:
self.fd.seek(sec.data_offset)
if sec.type == SECTION_TYPE_ENCRYPTED_ARCHIVE_TBZ2:
decryptor = AES.new(self.key, AES.MODE_CBC, sec.vector)
out_filename = unique_file_name('{}_off{:X}.{}.tbz2'.format(
abspath(self.fd.path), sec.offset, sec.name))
with open(out_filename, 'w+b') as out_fd:
n, cz = 0, self.chunksize
while n < sec.data_size:
n += cz
if n > sec.data_size:
cz -= n - sec.data_size
chunk = str2bytes(self.fd.read(cz))
if sec.type == SECTION_TYPE_ENCRYPTED_ARCHIVE_TBZ2:
chunk = decryptor.decrypt(chunk)
out_fd.write(chunk)
self.untar(out_filename, splitext(out_filename)[0])
return True
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
# Make sure we put the initial extraction directory in the CWD
if self.directory is None:
self.directory = os.getcwd()
if basedir != self.directory:
# During recursive extraction, extracted files will be in subdirectories
# of the CWD. This allows us to figure out the subdirectory by simply
# splitting the target file's base directory on our known CWD.
#
# However, the very *first* file being scanned is not necessarily in the
# CWD, so this will raise an IndexError. This is easy to handle though,
# since the very first file being scanned needs to have its contents
# extracted to ${CWD}/_basename.extracted, so we just set the subdir
# variable to a blank string when an IndexError is encountered.
try:
subdir = basedir.split(self.directory)[1][1:]
except IndexError as e:
subdir = ""
else:
subdir = ""
if self.output_directory_override:
output_directory = os.path.join(self.directory, subdir,
self.output_directory_override)
else:
outdir = os.path.join(self.directory, subdir, '_' + basename)
output_directory = unique_file_name(outdir,
extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
self.output[path].directory = os.path.realpath(
output_directory) + os.path.sep
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
# Make sure we put the initial extraction directory in the CWD
if self.directory is None:
self.directory = os.getcwd()
if basedir != self.directory:
# During recursive extraction, extracted files will be in subdirectories
# of the CWD. This allows us to figure out the subdirectory by simply
# splitting the target file's base directory on our known CWD.
#
# However, the very *first* file being scanned is not necessarily in the
# CWD, so this will raise an IndexError. This is easy to handle though,
# since the very first file being scanned needs to have its contents
# extracted to ${CWD}/_basename.extracted, so we just set the subdir
# variable to a blank string when an IndexError is encountered.
try:
subdir = basedir.split(self.directory)[1][1:]
except IndexError as e:
subdir = ""
else:
subdir = ""
if self.output_directory_override:
output_directory = os.path.join(self.directory, subdir, self.output_directory_override)
else:
outdir = os.path.join(self.directory, subdir, '_' + basename)
output_directory = unique_file_name(outdir, extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
self.output[path].directory = os.path.realpath(output_directory) + os.path.sep
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
# Make sure we put the initial extracted file in the CWD
if self.directory is None:
if self.base_directory is None:
basedir = os.getcwd()
else:
basedir = self.base_directory
if not os.path.exists(basedir):
os.mkdir(basedir)
outdir = os.path.join(basedir, '_' + basename)
output_directory = unique_file_name(outdir, extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
# Set the initial base extraction directory for later determining the level of recusion
if self.directory is None:
self.directory = os.path.realpath(output_directory) + os.path.sep
self.output[path].directory = self.directory
return output_directory
def build_output_directory(self, path):
'''
Set the output directory for extracted files.
@path - The path to the file that data will be extracted from.
Returns None.
'''
# If we have not already created an output directory for this target file, create one now
if not has_key(self.extraction_directories, path):
basedir = os.path.dirname(path)
basename = os.path.basename(path)
# Make sure we put the initial extraction directory in the CWD
if self.directory is None:
self.directory = os.getcwd()
if basedir != self.directory:
subdir = basedir.split(self.directory)[1][1:]
else:
subdir = ""
if self.output_directory_override:
output_directory = os.path.join(self.directory, subdir, self.output_directory_override)
else:
outdir = os.path.join(self.directory, subdir, '_' + basename)
output_directory = unique_file_name(outdir, extension='extracted')
if not os.path.exists(output_directory):
os.mkdir(output_directory)
self.extraction_directories[path] = output_directory
self.output[path].directory = os.path.realpath(output_directory) + os.path.sep
# Else, just use the already created directory
else:
output_directory = self.extraction_directories[path]
return output_directory
def _dd(self, file_name, offset, size, extension, output_file_name=None):
'''
Extracts a file embedded inside the target file.
@file_name - Path to the target file.
@offset - Offset inside the target file where the embedded file begins.
@size - Number of bytes to extract.
@extension - The file exension to assign to the extracted file on disk.
@output_file_name - The requested name of the output file.
Returns the extracted file name.
'''
total_size = 0
# Default extracted file name is <displayed hex offset>.<extension>
default_bname = "%X" % (offset + self.config.base)
if self.max_size and size > self.max_size:
size = self.max_size
if not output_file_name or output_file_name is None:
bname = default_bname
else:
# Strip the output file name of invalid/dangerous characters (like file paths)
bname = os.path.basename(output_file_name)
fname = unique_file_name(bname, extension)
try:
# If byte swapping is enabled, we need to start reading at a swap-size
# aligned offset, then index in to the read data appropriately.
if self.config.swap_size:
adjust = offset % self.config.swap_size
else:
adjust = 0
offset -= adjust
# Open the target file and seek to the offset
fdin = self.config.open_file(file_name)
fdin.seek(offset)
# Open the output file
try:
fdout = BlockFile(fname, 'w')
except KeyboardInterrupt as e:
raise e
except Exception as e:
# Fall back to the default name if the requested name fails
fname = unique_file_name(default_bname, extension)
fdout = BlockFile(fname, 'w')
while total_size < size:
(data, dlen) = fdin.read_block()
if not data:
break
else:
total_size += (dlen-adjust)
if total_size > size:
dlen -= (total_size - size)
fdout.write(str2bytes(data[adjust:dlen]))
adjust = 0
# Cleanup
fdout.close()
fdin.close()
except KeyboardInterrupt as e:
raise e
except Exception as e:
raise Exception("Extractor.dd failed to extract data from '%s' to '%s': %s" % (file_name, fname, str(e)))
binwalk.core.common.debug("Carved data block 0x%X - 0x%X from '%s' to '%s'" % (offset, offset+size, file_name, fname))
return fname
def _dd(self, file_name, offset, size, extension, output_file_name=None):
'''
Extracts a file embedded inside the target file.
@file_name - Path to the target file.
@offset - Offset inside the target file where the embedded file begins.
@size - Number of bytes to extract.
@extension - The file exension to assign to the extracted file on disk.
@output_file_name - The requested name of the output file.
Returns the extracted file name.
'''
total_size = 0
# Default extracted file name is <displayed hex offset>.<extension>
default_bname = "%X" % (offset + self.config.base)
if self.max_size and size > self.max_size:
size = self.max_size
if not output_file_name or output_file_name is None:
bname = default_bname
else:
# Strip the output file name of invalid/dangerous characters (like
# file paths)
bname = os.path.basename(output_file_name)
fname = unique_file_name(bname, extension)
try:
# If byte swapping is enabled, we need to start reading at a swap-size
# aligned offset, then index in to the read data appropriately.
if self.config.swap_size:
adjust = offset % self.config.swap_size
else:
adjust = 0
offset -= adjust
# Open the target file and seek to the offset
fdin = self.config.open_file(file_name)
fdin.seek(offset)
# Open the output file
try:
fdout = BlockFile(fname, 'w')
except KeyboardInterrupt as e:
raise e
except Exception as e:
# Fall back to the default name if the requested name fails
fname = unique_file_name(default_bname, extension)
fdout = BlockFile(fname, 'w')
while total_size < size:
(data, dlen) = fdin.read_block()
if dlen < 1:
break
else:
total_size += (dlen - adjust)
if total_size > size:
dlen -= (total_size - size)
fdout.write(str2bytes(data[adjust:dlen]))
adjust = 0
# Cleanup
fdout.close()
fdin.close()
except KeyboardInterrupt as e:
raise e
except Exception as e:
raise Exception("Extractor.dd failed to extract data from '%s' to '%s': %s" %
(file_name, fname, str(e)))
binwalk.core.common.debug("Carved data block 0x%X - 0x%X from '%s' to '%s'" %
(offset, offset + size, file_name, fname))
return fname