Closed as not planned
Description
Bug report
Bug description:
The documentation of tarfile._Stream says that "A stream-like object could be for example: sys.stdin, sys.stdout, a socket, a tape device etc.", but when using a pipe it raises tarfile.StreamError: seeking backwards is not allowed when the write end of the pipe is closed.
Not sure if this is a bug in the tarfile implementation or in my code.
The size of the tar file I used is 40M.
#!/usr/bin/env python
import os
import os.path
from pathlib import Path
import sys
import tarfile
from threading import Thread
BUF_SIZE = 8 * 1024
def tar_strip_components(file, n=0):
for info in file.getmembers():
components = Path(info.path).parts
info.path = Path(*components[1:])
yield info
def tar_unpack_thread(path, size, wr):
total_size = size
size = 0
with open(path, "rb") as f:
buf = bytearray(BUF_SIZE)
while True:
rate = (size / total_size) * 100
sys.stderr.write(f"{rate:.0f}%\r")
n = f.readinto(buf)
if n == 0:
break
size += n
data = buf[:n]
wr.write(data)
sys.stderr.write("done\n")
wr.close()
def tar_unpack(dest, path, size):
rd, wr = os.pipe()
rd = os.fdopen(rd, "rb")
wr = os.fdopen(wr, "wb")
t = Thread(target=tar_unpack_thread, args=(path, size, wr))
t.start()
# TODO: stream is not closed in case TarFile() raises an exception, but we
# don't care.
stream = tarfile._Stream(path, "r", "xz", rd, BUF_SIZE)
with tarfile.TarFile(path, "r", stream) as file:
members = tar_strip_components(file, 1)
file.extractall(dest, members)
rd.close()
t.join()
def main():
path = sys.argv[1]
size = os.stat(path).st_size
tar_unpack(".", path, size)
if __name__ == "__main__":
main()$ ./test.py archive.tar.xz
done
Traceback (most recent call last):
File "/home/manlio/src/python/pkg/mperillo/zig-installer/test/./test.py", line 67, in <module>
main()
File "/home/manlio/src/python/pkg/mperillo/zig-installer/test/./test.py", line 63, in main
tar_unpack(".", path, size)
File "/home/manlio/src/python/pkg/mperillo/zig-installer/test/./test.py", line 54, in tar_unpack
file.extractall(dest, members)
File "/usr/lib/python3.11/tarfile.py", line 2264, in extractall
self._extract_one(tarinfo, path, set_attrs=not tarinfo.isdir(),
File "/usr/lib/python3.11/tarfile.py", line 2327, in _extract_one
self._extract_member(tarinfo, os.path.join(path, tarinfo.name),
File "/usr/lib/python3.11/tarfile.py", line 2410, in _extract_member
self.makefile(tarinfo, targetpath)
File "/usr/lib/python3.11/tarfile.py", line 2453, in makefile
source.seek(tarinfo.offset_data)
File "/usr/lib/python3.11/tarfile.py", line 520, in seek
raise StreamError("seeking backwards is not allowed")
tarfile.StreamError: seeking backwards is not allowedIf I replace the use of tarfile.TarFile() with
while True:
buf = stream.read(BUF_SIZE)
if buf == b"":
sys.stderr.write("ok\n")
returnthe code works correctly.
Thanks.
CPython versions tested on:
3.11
Operating systems tested on:
Linux
Metadata
Metadata
Assignees
Projects
Status
Done