[Python-checkins] gh-74468: [tarfile] Fix incorrect name attribute of ExFileObject (GH-102424)

ethanfurman webhook-mailer at python.org
Mon Mar 27 19:21:15 EDT 2023


https://github.com/python/cpython/commit/56d055a0d81a809e4ff8e1d56756a3bf32317efb
commit: 56d055a0d81a809e4ff8e1d56756a3bf32317efb
branch: main
author: Oleg Iarygin <oleg at arhadthedev.net>
committer: ethanfurman <ethan at stoneleaf.us>
date: 2023-03-27T16:21:07-07:00
summary:

gh-74468: [tarfile] Fix incorrect name attribute of ExFileObject (GH-102424)

Co-authored-by: Simeon Visser <svisser at users.noreply.github.com>

files:
A Misc/NEWS.d/next/Library/2023-03-04-20-58-29.gh-issue-74468.Ac5Ew_.rst
M Lib/tarfile.py
M Lib/test/test_tarfile.py

diff --git a/Lib/tarfile.py b/Lib/tarfile.py
index d686435d90ad..b733195c9c56 100755
--- a/Lib/tarfile.py
+++ b/Lib/tarfile.py
@@ -601,12 +601,12 @@ class _FileInFile(object):
        object.
     """
 
-    def __init__(self, fileobj, offset, size, blockinfo=None):
+    def __init__(self, fileobj, offset, size, name, blockinfo=None):
         self.fileobj = fileobj
         self.offset = offset
         self.size = size
         self.position = 0
-        self.name = getattr(fileobj, "name", None)
+        self.name = name
         self.closed = False
 
         if blockinfo is None:
@@ -703,7 +703,7 @@ class ExFileObject(io.BufferedReader):
 
     def __init__(self, tarfile, tarinfo):
         fileobj = _FileInFile(tarfile.fileobj, tarinfo.offset_data,
-                tarinfo.size, tarinfo.sparse)
+                tarinfo.size, tarinfo.name, tarinfo.sparse)
         super().__init__(fileobj)
 #class ExFileObject
 
diff --git a/Lib/test/test_tarfile.py b/Lib/test/test_tarfile.py
index 75b60e9a50e7..39f6f499c818 100644
--- a/Lib/test/test_tarfile.py
+++ b/Lib/test/test_tarfile.py
@@ -479,6 +479,13 @@ def test_length_zero_header(self):
             with tarfile.open(support.findfile('recursion.tar')) as tar:
                 pass
 
+    def test_extractfile_name(self):
+        # gh-74468: TarFile.name must name a file, not a parent archive.
+        file = self.tar.getmember('ustar/regtype')
+        with self.tar.extractfile(file) as fobj:
+            self.assertEqual(fobj.name, 'ustar/regtype')
+
+
 class MiscReadTestBase(CommonReadTest):
     def requires_name_attribute(self):
         pass
diff --git a/Misc/NEWS.d/next/Library/2023-03-04-20-58-29.gh-issue-74468.Ac5Ew_.rst b/Misc/NEWS.d/next/Library/2023-03-04-20-58-29.gh-issue-74468.Ac5Ew_.rst
new file mode 100644
index 000000000000..8fad551f3a4e
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-03-04-20-58-29.gh-issue-74468.Ac5Ew_.rst
@@ -0,0 +1,3 @@
+Attribute name of the extracted :mod:`tarfile` file object now holds
+filename of itself rather than of the archive it is contained in.
+Patch by Oleg Iarygin.



More information about the Python-checkins mailing list