From e39e12911969003502283ca849358b0f9d360f08 Mon Sep 17 00:00:00 2001 From: TrellixVulnTeam Date: Wed, 21 Dec 2022 07:37:27 +0000 Subject: [PATCH] Adding tarfile member sanitization to extractall() --- .../tasks/calculate_black_spots.py | 21 ++++++++++++++++++- .../tasks/generate_training_input.py | 21 ++++++++++++++++++- 2 files changed, 40 insertions(+), 2 deletions(-) diff --git a/app/black_spots/tasks/calculate_black_spots.py b/app/black_spots/tasks/calculate_black_spots.py index bdacdecb..c1e2a20b 100644 --- a/app/black_spots/tasks/calculate_black_spots.py +++ b/app/black_spots/tasks/calculate_black_spots.py @@ -116,7 +116,26 @@ def calculate_black_spots(history_length=datetime.timedelta(days=5 * 365 + 1), r shp_tar = RoadSegmentsShapefile.objects.get(uuid=segments_shp_uuid).shp_tgz.path with tarfile.open(shp_tar, "r:gz") as tar: # TODO: Extract only the combined segments file, not the entire tarball - tar.extractall(tar_output_dir) + 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, tar_output_dir) logger.info("Performing blackspot calculations") segments_path = os.path.join(tar_output_dir, 'segments', COMBINED_SEGMENTS_SHP_NAME) diff --git a/app/black_spots/tasks/generate_training_input.py b/app/black_spots/tasks/generate_training_input.py index f088cf5c..6e7d855e 100644 --- a/app/black_spots/tasks/generate_training_input.py +++ b/app/black_spots/tasks/generate_training_input.py @@ -149,7 +149,26 @@ def get_data(records_csv_uuid, segments_shp_uuid, road_srid): try: # TODO: Extract only the combined segments file, not the entire tarball with tarfile.open(shp_tar, 'r:gz') as tar: - tar.extractall(tar_output_dir) + 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, tar_output_dir) segments_path = os.path.join(tar_output_dir, 'segments', COMBINED_SEGMENTS_SHP_NAME) with fiona.open(segments_path) as segments: