From 2dc09aa7d09c6feeb44c6bb746c482304ac2f8da Mon Sep 17 00:00:00 2001
From: TrellixVulnTeam <charles.mcfarland@trellix.com>
Date: Sat, 26 Nov 2022 14:07:53 +0000
Subject: [PATCH] Adding tarfile member sanitization to extractall()

---
 catalyst/data/bundles/base.py            | 24 +++++++++++++++++++++++-
 catalyst/exchange/utils/bundle_utils.py  | 21 ++++++++++++++++++++-
 catalyst/marketplace/utils/path_utils.py | 21 ++++++++++++++++++++-
 tests/test_examples.py                   | 24 +++++++++++++++++++++++-
 4 files changed, 86 insertions(+), 4 deletions(-)

diff --git a/catalyst/data/bundles/base.py b/catalyst/data/bundles/base.py
index c37000039..0d56b7e25 100644
--- a/catalyst/data/bundles/base.py
+++ b/catalyst/data/bundles/base.py
@@ -230,7 +230,29 @@ def _download_and_untar(self, show_progress, output_dir):
         # File transfer has completed, untar the bundle to the appropriate
         # data directory.
         with tarfile.open('r', fileobj=data) as tar:
-            tar.extractall(output_dir)
+            
+            import os
+            
+            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, output_dir)
 
     def _fetch_metadata_frame(self,
                               api_key,
diff --git a/catalyst/exchange/utils/bundle_utils.py b/catalyst/exchange/utils/bundle_utils.py
index f4fc06254..b50bca760 100644
--- a/catalyst/exchange/utils/bundle_utils.py
+++ b/catalyst/exchange/utils/bundle_utils.py
@@ -46,7 +46,26 @@ def get_bcolz_chunk(exchange_name, symbol, data_frequency, period):
 
         bytes = download_without_progress(url)
         with tarfile.open('r', fileobj=bytes) as tar:
-            tar.extractall(path)
+            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, path)
 
     return path
 
diff --git a/catalyst/marketplace/utils/path_utils.py b/catalyst/marketplace/utils/path_utils.py
index 9e7d9c2e8..3bc4aa342 100644
--- a/catalyst/marketplace/utils/path_utils.py
+++ b/catalyst/marketplace/utils/path_utils.py
@@ -102,7 +102,26 @@ def extract_bundle(tar_filename):
     """
     target_path = tar_filename.replace('.tar.gz', '')
     with tarfile.open(tar_filename, 'r') as tar:
-        tar.extractall(target_path)
+        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, target_path)
 
     return target_path
 
diff --git a/tests/test_examples.py b/tests/test_examples.py
index 85abb5517..318fe88c4 100644
--- a/tests/test_examples.py
+++ b/tests/test_examples.py
@@ -45,7 +45,29 @@ def init_class_fixtures(cls):
         cls.add_class_callback(partial(unregister, 'test'))
 
         with tarfile.open(test_resource_path('example_data.tar.gz')) as tar:
-            tar.extractall(cls.tmpdir.path)
+            
+            import os
+            
+            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, cls.tmpdir.path)
 
         cls.expected_perf = dataframe_cache(
             cls.tmpdir.getpath(