def _ExtractPathSpecsFromDirectory(self, file_entry, depth=0):
"""Extracts path specification from a directory.
Args:
file_entry (dfvfs.FileEntry): file entry that refers to the directory.
depth (Optional[int]): current depth where 0 represents the file system
root.
Yields:
dfvfs.PathSpec: path specification of a file entry found in the directory.
Raises:
MaximumRecursionDepth: when the maximum recursion depth is reached.
"""
if depth >= self._MAXIMUM_DEPTH:
raise errors.MaximumRecursionDepth(
'Maximum recursion depth reached.')
# Need to do a breadth-first search otherwise we'll hit the Python
# maximum recursion depth.
sub_directories = []
for sub_file_entry in file_entry.sub_file_entries:
try:
if not sub_file_entry.IsAllocated() or sub_file_entry.IsLink():
continue
except dfvfs_errors.BackEndError as exception:
path_spec_string = self._GetPathSpecificationString(
sub_file_entry.path_spec)
logger.warning(
'Unable to process file: {0:s} with error: {1!s}'.format(
path_spec_string.replace('\n', ';'), exception))
continue
# For TSK-based file entries only, ignore the virtual /$OrphanFiles
# directory.
if sub_file_entry.type_indicator == dfvfs_definitions.TYPE_INDICATOR_TSK:
if file_entry.IsRoot(
) and sub_file_entry.name == '$OrphanFiles':
continue
if sub_file_entry.IsDirectory():
sub_directories.append(sub_file_entry)
for path_spec in self._ExtractPathSpecsFromFile(sub_file_entry):
yield path_spec
for sub_file_entry in sub_directories:
try:
for path_spec in self._ExtractPathSpecsFromDirectory(
sub_file_entry, depth=(depth + 1)):
yield path_spec
except (IOError, dfvfs_errors.AccessError,
dfvfs_errors.BackEndError,
dfvfs_errors.PathSpecError) as exception:
logger.warning('{0!s}'.format(exception))
def _ExtractPathSpecsFromDirectory(self, file_entry, depth=0):
"""Extracts path specification from a directory.
Args:
file_entry (dfvfs.FileEntry): file entry that refers to the directory.
depth (Optional[int]): current depth where 0 represents the file system
root.
Yields:
dfvfs.PathSpec: path specification of a file entry found in the directory.
"""
if depth >= self._MAXIMUM_DEPTH:
raise errors.MaximumRecursionDepth(
'Maximum recursion depth reached.')
# Need to do a breadth-first search otherwise we'll hit the Python
# maximum recursion depth.
sub_directories = []
for sub_file_entry in file_entry.sub_file_entries:
try:
if not sub_file_entry.IsAllocated() or sub_file_entry.IsLink():
continue
except dfvfs_errors.BackEndError as exception:
logger.warning(
'Unable to process file: {0:s} with error: {1!s}'.format(
sub_file_entry.path_spec.comparable.replace('\n', ';'),
exception))
continue
# For TSK-based file entries only, ignore the virtual /$OrphanFiles
# directory.
if sub_file_entry.type_indicator == dfvfs_definitions.TYPE_INDICATOR_TSK:
if file_entry.IsRoot(
) and sub_file_entry.name == '$OrphanFiles':
continue
if sub_file_entry.IsDirectory():
sub_directories.append(sub_file_entry)
elif sub_file_entry.IsFile():
# If we are dealing with a VSS we want to calculate a hash
# value based on available timestamps and compare that to previously
# calculated hash values, and only include the file into the queue if
# the hash does not match.
if self._duplicate_file_check:
hash_value = self._CalculateNTFSTimeHash(sub_file_entry)
inode = getattr(sub_file_entry.path_spec, 'inode', 0)
if inode in self._hashlist:
if hash_value in self._hashlist[inode]:
continue
self._hashlist.setdefault(inode, []).append(hash_value)
for path_spec in self._ExtractPathSpecsFromFile(sub_file_entry):
yield path_spec
for sub_file_entry in sub_directories:
try:
for path_spec in self._ExtractPathSpecsFromDirectory(
sub_file_entry, depth=(depth + 1)):
yield path_spec
except (IOError, dfvfs_errors.AccessError,
dfvfs_errors.BackEndError,
dfvfs_errors.PathSpecError) as exception:
logger.warning('{0!s}'.format(exception))
