def iter_tar_members(tp: tarfile.TarFile) -> Generator[tarfile.TarInfo, None, None]:
"""
This is a lazy alternative to TarInfo.getmembers() that only reads one tar item at a time.
We're reading the MTL file, which is almost always the first entry in the tar, and then
closing it, so we're avoiding skipping through the entirety of the tar.
"""
member = tp.next()
while member is not None:
yield member
member = tp.next()
def main(argv):
import getopt
def usage():
print('usage: %s [-b basedir] cmd [arg ...]' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'db:')
except getopt.GetoptError:
return usage()
debug = 0
basedir = 'tar'
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-b': basedir = v
tardb = TarDB(basedir)
if not args: return usage()
cmd = args.pop(0)
if cmd == 'create':
tardb.create()
elif cmd == 'import':
tardb.open()
for path in args:
tar = TarFile(path)
while True:
info = tar.next()
if info is None: break
fp = tar.fileobj
fp.seek(info.offset+BLOCKSIZE)
data = fp.read(info.size)
tardb.add_record(info, data)
tardb.flush()
tardb.close()
elif cmd == 'add':
tardb.open()
for path in args:
name = os.path.basename(path)
info = TarInfo(name)
with open(path, 'rb') as fp:
data = fp.read()
recno = tardb.add_record(info, data)
print(recno)
tardb.close()
elif cmd == 'get':
tardb.open()
for recno in args:
recno = int(recno)
(_, data) = tardb.get_recinfo(recno, True)
sys.stdout.buffer.write(data)
tardb.close()
elif cmd == 'getinfo':
tardb.open()
for recno in args:
recno = int(recno)
(info, _) = tardb.get_recinfo(recno, False)
print(info)
tardb.close()
else:
return usage()
return 0
def _open_archive_file(self, archive: TarFile, name: str) -> IO[bytes]:
while True:
member = archive.next()
if member is None:
break
if member.name == name:
fobj = archive.extractfile(member)
if fobj is None:
break
return fobj
# noinspection PyProtectedMember
raise ObjectDoesNotExistError(f'File {name} is missing in archive',
self._file_storage._driver, archive.name)
def _open_archive_file(self, archive: TarFile, name: str) -> IO[bytes]:
while True:
member = archive.next()
if member is None:
break
if member.name == name:
fobj = archive.extractfile(member)
if fobj is None:
break
return fobj
raise ObjectDoesNotExistError(
'File {} is missing in archive'.format(name),
self._file_storage._driver, archive.name)
def main(argv):
import getopt
def usage():
print('usage: %s [-b basedir] cmd [arg ...]' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'db:')
except getopt.GetoptError:
return usage()
debug = 0
basedir = 'msg'
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-b': basedir = v
if not args: return usage()
cmd = args.pop(0)
msgdb = MessageDB(basedir)
if cmd == 'create':
msgdb.create()
elif cmd == 'import':
msgdb.open()
for path in args:
tar = TarFile(path)
while True:
info = tar.next()
if info is None: break
fp = tar.fileobj
fp.seek(info.offset+BLOCKSIZE)
data = fp.read(info.size)
recno = msgdb.add_file(gzip2bytes(data))
print(recno)
msgdb.flush()
msgdb.close()
elif cmd == 'add':
msgdb.open()
for path in args:
with open(path, 'r') as fp:
data = fp.read()
recno = msgdb.add_file(data)
print(recno)
msgdb.close()
elif cmd == 'search':
msgdb.open()
for data in msgdb.search_text(args):
print(rmsp(data)[:80])
msgdb.close()
else:
return usage()
return 0
def read_binary_mask(self, tf: tarfile.TarFile) -> bm.BinaryMaskCollection:
log: Optional[Log] = None
masks: MutableSequence[bm.MaskData] = []
pixel_ticks: Optional[Mapping[Axes, ArrayLike[int]]] = None
physical_ticks: Optional[Mapping[Coordinates,
ArrayLike[Number]]] = None
while True:
tarinfo: Optional[tarfile.TarInfo] = tf.next()
if tarinfo is None:
break
# wrap it in a BytesIO object to ensure we never seek backwards.
extracted_fh = tf.extractfile(tarinfo)
if extracted_fh is None:
raise ValueError(f"Unable to extract file {tarinfo.name}")
byte_stream = io.BytesIO(extracted_fh.read())
if tarinfo.name == v0_0.LOG_FILENAME:
string_stream = codecs.getreader("utf-8")(byte_stream)
log = Log.decode(string_stream.read())
elif tarinfo.name == v0_0.PIXEL_TICKS_FILE:
pixel_ticks = pickle.load(byte_stream)
elif tarinfo.name == v0_0.PHYSICAL_TICKS_FILE:
physical_ticks = pickle.load(byte_stream)
elif tarinfo.name.startswith(v0_0.MASK_PREFIX):
mask_on_disk: v0_0.MaskOnDisk = pickle.load(byte_stream)
if not isinstance(mask_on_disk, v0_0.MaskOnDisk):
raise TypeError(
f"mask does not conform to expected mask structure")
masks.append(
bm.MaskData(mask_on_disk.binary_mask, mask_on_disk.offsets,
None))
else:
warnings.warn(
f"Unexpected file in binary mask collection {tarinfo.name}",
DataFormatWarning)
if pixel_ticks is None:
raise ValueError("pixel coordinates not found")
if physical_ticks is None:
raise ValueError("physical coordinates not found")
return bm.BinaryMaskCollection(pixel_ticks, physical_ticks, masks, log)
def get_site_and_version_from_backup(tar: tarfile.TarFile) -> Tuple[str, str]:
"""Get the first file of the tar archive. Expecting <site>/version symlink
for validation reasons."""
site_tarinfo = tar.next()
if site_tarinfo is None:
raise Exception("Failed to detect version of backed up site.")
try:
sitename, version_name = site_tarinfo.name.split("/", 1)
except ValueError:
raise Exception("Failed to detect version of backed up site. "
"Maybe the backup is from an incompatible version.")
if version_name == "version":
version = site_tarinfo.linkname.split("/")[-1]
else:
raise Exception("Failed to detect version of backed up site.")
return sitename, version
def main(argv):
import getopt
def usage():
print('usage: %s [-b basedir] cmd [arg ...]' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'db:')
except getopt.GetoptError:
return usage()
debug = 0
basedir = 'tar'
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-b': basedir = v
tardb = TarDB(basedir)
if not args: return usage()
cmd = args.pop(0)
if cmd == 'create':
tardb.create()
elif cmd == 'import':
tardb.open()
for path in args:
tar = TarFile(path)
while True:
info = tar.next()
if info is None: break
fp = tar.fileobj
fp.seek(info.offset + BLOCKSIZE)
data = fp.read(info.size)
tardb.add_record(info, data)
tardb.flush()
tardb.close()
elif cmd == 'add':
tardb.open()
for path in args:
name = os.path.basename(path)
info = TarInfo(name)
with open(path, 'rb') as fp:
data = fp.read()
recno = tardb.add_record(info, data)
print(recno)
tardb.close()
elif cmd == 'get':
tardb.open()
for recno in args:
recno = int(recno)
(_, data) = tardb.get_recinfo(recno, True)
sys.stdout.buffer.write(data)
tardb.close()
elif cmd == 'getinfo':
tardb.open()
for recno in args:
recno = int(recno)
(info, _) = tardb.get_recinfo(recno, False)
print(info)
tardb.close()
else:
return usage()
return 0
