From 3aba935d3540ccf4620c903b5d7f324e7c292e05 Mon Sep 17 00:00:00 2001 From: TrellixVulnTeam Date: Wed, 21 Dec 2022 02:40:22 +0000 Subject: [PATCH] Adding tarfile member sanitization to extractall() --- .../Lib/tarfile.py" | 21 ++++++++++++++++++- .../Lib/test/test_tarfile.py" | 21 ++++++++++++++++++- 2 files changed, 40 insertions(+), 2 deletions(-) diff --git "a/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/tarfile.py" "b/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/tarfile.py" index 30150ec..1c513f9 100755 --- "a/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/tarfile.py" +++ "b/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/tarfile.py" @@ -2527,7 +2527,26 @@ def main(): if is_tarfile(src): with TarFile.open(src, 'r:*') as tf: - tf.extractall(path=curdir) + def is_within_directory(directory, target): + + abs_directory = os.path.abspath(directory) + abs_target = os.path.abspath(target) + + prefix = os.path.commonprefix([abs_directory, abs_target]) + + return prefix == abs_directory + + def safe_extract(tar, path=".", members=None, *, numeric_owner=False): + + for member in tar.getmembers(): + member_path = os.path.join(path, member.name) + if not is_within_directory(path, member_path): + raise Exception("Attempted Path Traversal in Tar File") + + tar.extractall(path, members, numeric_owner=numeric_owner) + + + safe_extract(tf, path=curdir) if args.verbose: if curdir == '.': msg = '{!r} file is extracted.'.format(src) diff --git "a/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/test/test_tarfile.py" "b/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/test/test_tarfile.py" index bee00c5..970ef01 100644 --- "a/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/test/test_tarfile.py" +++ "b/cpython-3.8\345\256\236\351\252\214\344\273\243\347\240\201/Lib/test/test_tarfile.py" @@ -612,7 +612,26 @@ def test_extractall_pathlike_name(self): with support.temp_dir(DIR), \ tarfile.open(tarname, encoding="iso8859-1") as tar: directories = [t for t in tar if t.isdir()] - tar.extractall(DIR, directories) + def is_within_directory(directory, target): + + abs_directory = os.path.abspath(directory) + abs_target = os.path.abspath(target) + + prefix = os.path.commonprefix([abs_directory, abs_target]) + + return prefix == abs_directory + + def safe_extract(tar, path=".", members=None, *, numeric_owner=False): + + for member in tar.getmembers(): + member_path = os.path.join(path, member.name) + if not is_within_directory(path, member_path): + raise Exception("Attempted Path Traversal in Tar File") + + tar.extractall(path, members, numeric_owner=numeric_owner) + + + safe_extract(tar, DIR, directories) for tarinfo in directories: path = DIR / tarinfo.name self.assertEqual(os.path.getmtime(path), tarinfo.mtime)