feat: update code from upstream 3.14.1

Author: Rogdham

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/), and this project
 adheres to [Semantic Versioning](https://semver.org/).
 
+## Unreleased
+
+### :rocket: Added
+
+- Update code with CPython 3.14.1 version
+
 ## [1.1.0] - 2025-11-23
 
 [1.1.0]: https://github.com/rogdham/backports.zstd/releases/tag/v1.1.0

src/python/backports/zstd/tarfile.py

@@ -364,7 +364,7 @@ def __init__(self, name, mode, comptype, fileobj, bufsize,
             fileobj = _StreamProxy(fileobj)
             comptype = fileobj.getcomptype()
 
-        self.name     = name or ""
+        self.name     = os.fspath(name) if name is not None else ""
         self.mode     = mode
         self.comptype = comptype
         self.fileobj  = fileobj
@@ -2726,6 +2726,9 @@ def makelink_with_filter(self, tarinfo, targetpath,
                 return
             else:
                 if os.path.exists(tarinfo._link_target):
+                    if os.path.lexists(targetpath):
+                        # Avoid FileExistsError on following os.link.
+                        os.unlink(targetpath)
                     os.link(tarinfo._link_target, targetpath)
                     return
         except symlink_exception:

src/python/backports/zstd/zipfile/__init__.py

@@ -262,7 +262,7 @@ def is_zipfile(filename):
         else:
             with open(filename, "rb") as fp:
                 result = _check_zipfile(fp)
-    except OSError:
+    except (OSError, BadZipFile):
         pass
     return result
 
@@ -272,9 +272,6 @@ def _handle_prepended_data(endrec, debug=0):
 
     # "concat" is zero, unless zip was concatenated to another file
     concat = endrec[_ECD_LOCATION] - size_cd - offset_cd
-    if endrec[_ECD_SIGNATURE] == stringEndArchive64:
-        # If Zip64 extension structures are present, account for them
-        concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator)
 
     if debug > 2:
         inferred = concat + offset_cd
@@ -286,33 +283,49 @@ def _EndRecData64(fpin, offset, endrec):
     """
     Read the ZIP64 end-of-archive records and use that to update endrec
     """
-    try:
-        fpin.seek(offset - sizeEndCentDir64Locator, 2)
-    except OSError:
-        # If the seek fails, the file is not large enough to contain a ZIP64
+    offset -= sizeEndCentDir64Locator
+    if offset < 0:
+        # The file is not large enough to contain a ZIP64
         # end-of-archive record, so just return the end record we were given.
         return endrec
-
+    fpin.seek(offset)
     data = fpin.read(sizeEndCentDir64Locator)
     if len(data) != sizeEndCentDir64Locator:
-        return endrec
+        raise OSError("Unknown I/O error")
     sig, diskno, reloff, disks = struct.unpack(structEndArchive64Locator, data)
     if sig != stringEndArchive64Locator:
         return endrec
 
     if diskno != 0 or disks > 1:
         raise BadZipFile("zipfiles that span multiple disks are not supported")
 
-    # Assume no 'zip64 extensible data'
-    fpin.seek(offset - sizeEndCentDir64Locator - sizeEndCentDir64, 2)
+    offset -= sizeEndCentDir64
+    if reloff > offset:
+        raise BadZipFile("Corrupt zip64 end of central directory locator")
+    # First, check the assumption that there is no prepended data.
+    fpin.seek(reloff)
+    extrasz = offset - reloff
     data = fpin.read(sizeEndCentDir64)
     if len(data) != sizeEndCentDir64:
-        return endrec
+        raise OSError("Unknown I/O error")
+    if not data.startswith(stringEndArchive64) and reloff != offset:
+        # Since we already have seen the Zip64 EOCD Locator, it's
+        # possible we got here because there is prepended data.
+        # Assume no 'zip64 extensible data'
+        fpin.seek(offset)
+        extrasz = 0
+        data = fpin.read(sizeEndCentDir64)
+        if len(data) != sizeEndCentDir64:
+            raise OSError("Unknown I/O error")
+    if not data.startswith(stringEndArchive64):
+        raise BadZipFile("Zip64 end of central directory record not found")
+
     sig, sz, create_version, read_version, disk_num, disk_dir, \
         dircount, dircount2, dirsize, diroffset = \
         struct.unpack(structEndArchive64, data)
-    if sig != stringEndArchive64:
-        return endrec
+    if (diroffset + dirsize != reloff or
+        sz + 12 != sizeEndCentDir64 + extrasz):
+        raise BadZipFile("Corrupt zip64 end of central directory record")
 
     # Update the original endrec using data from the ZIP64 record
     endrec[_ECD_SIGNATURE] = sig
@@ -322,6 +335,7 @@ def _EndRecData64(fpin, offset, endrec):
     endrec[_ECD_ENTRIES_TOTAL] = dircount2
     endrec[_ECD_SIZE] = dirsize
     endrec[_ECD_OFFSET] = diroffset
+    endrec[_ECD_LOCATION] = offset - extrasz
     return endrec
 
 
@@ -355,7 +369,7 @@ def _EndRecData(fpin):
         endrec.append(filesize - sizeEndCentDir)
 
         # Try to read the "Zip64 end of central directory" structure
-        return _EndRecData64(fpin, -sizeEndCentDir, endrec)
+        return _EndRecData64(fpin, filesize - sizeEndCentDir, endrec)
 
     # Either this is not a ZIP file, or it is a ZIP file with an archive
     # comment.  Search the end of the file for the "end of central directory"
@@ -379,8 +393,7 @@ def _EndRecData(fpin):
         endrec.append(maxCommentStart + start)
 
         # Try to read the "Zip64 end of central directory" structure
-        return _EndRecData64(fpin, maxCommentStart + start - filesize,
-                             endrec)
+        return _EndRecData64(fpin, maxCommentStart + start, endrec)
 
     # Unable to find a valid end of central directory structure
     return None
@@ -2142,7 +2155,7 @@ def _write_end_record(self):
                                    " would require ZIP64 extensions")
             zip64endrec = struct.pack(
                 structEndArchive64, stringEndArchive64,
-                44, 45, 45, 0, 0, centDirCount, centDirCount,
+                sizeEndCentDir64 - 12, 45, 45, 0, 0, centDirCount, centDirCount,
                 centDirSize, centDirOffset)
             self.fp.write(zip64endrec)
 

tests/test/support/__init__.py

@@ -68,7 +68,7 @@
     "BrokenIter",
     "in_systemd_nspawn_sync_suppressed",
     "run_no_yield_async_fn", "run_yielding_async_fn", "async_yield",
-    "reset_code",
+    "reset_code", "on_github_actions"
     ]
 
 
@@ -1366,6 +1366,7 @@ def reset_code(f: types.FunctionType) -> types.FunctionType:
     f.__code__ = f.__code__.replace()
     return f
 
+on_github_actions = "GITHUB_ACTIONS" in os.environ
 
 #=======================================================================
 # Check for the presence of docstrings.

tests/test/test_tarfile.py

@@ -854,6 +854,57 @@ def test_next_on_empty_tarfile(self):
         with tarfile.open(fileobj=fd, mode="r") as tf:
             self.assertEqual(tf.next(), None)
 
+    def _setup_symlink_to_target(self, temp_dirpath):
+        target_filepath = os.path.join(temp_dirpath, "target")
+        ustar_dirpath = os.path.join(temp_dirpath, "ustar")
+        hardlink_filepath = os.path.join(ustar_dirpath, "lnktype")
+        with open(target_filepath, "wb") as f:
+            f.write(b"target")
+        os.makedirs(ustar_dirpath)
+        os.symlink(target_filepath, hardlink_filepath)
+        return target_filepath, hardlink_filepath
+
+    def _assert_on_file_content(self, filepath, digest):
+        with open(filepath, "rb") as f:
+            data = f.read()
+        self.assertEqual(sha256sum(data), digest)
+
+    @unittest.skipUnless(
+        hasattr(os, "link"), "Missing hardlink implementation"
+    )
+    @os_helper.skip_unless_symlink
+    def test_extract_hardlink_on_symlink(self):
+        """
+        This test verifies that extracting a hardlink will not follow an
+        existing symlink after a FileExistsError on os.link.
+        """
+        with os_helper.temp_dir() as DIR:
+            target_filepath, hardlink_filepath = self._setup_symlink_to_target(DIR)
+            with tarfile.open(tarname, encoding="iso8859-1") as tar:
+                tar.extract("ustar/regtype", DIR, filter="data")
+                tar.extract("ustar/lnktype", DIR, filter="data")
+                self._assert_on_file_content(target_filepath, sha256sum(b"target"))
+                self._assert_on_file_content(hardlink_filepath, sha256_regtype)
+
+    @unittest.skipUnless(
+        hasattr(os, "link"), "Missing hardlink implementation"
+    )
+    @os_helper.skip_unless_symlink
+    def test_extractall_hardlink_on_symlink(self):
+        """
+        This test verifies that extracting a hardlink will not follow an
+        existing symlink after a FileExistsError on os.link.
+        """
+        with os_helper.temp_dir() as DIR:
+            target_filepath, hardlink_filepath = self._setup_symlink_to_target(DIR)
+            with tarfile.open(tarname, encoding="iso8859-1") as tar:
+                tar.extractall(
+                    DIR, members=["ustar/regtype", "ustar/lnktype"], filter="data",
+                )
+                self._assert_on_file_content(target_filepath, sha256sum(b"target"))
+                self._assert_on_file_content(hardlink_filepath, sha256_regtype)
+
+
 class MiscReadTest(MiscReadTestBase, unittest.TestCase):
     test_fail_comp = None
 
@@ -1750,6 +1801,16 @@ def test_file_mode(self):
         finally:
             os.umask(original_umask)
 
+    def test_pathlike_name(self):
+        expected_name = os.path.abspath(tmpname)
+        tarpath = os_helper.FakePath(tmpname)
+
+        for func in (tarfile.open, tarfile.TarFile.open):
+            with self.subTest():
+                with func(tarpath, self.mode) as tar:
+                    self.assertEqual(tar.name, expected_name)
+                os_helper.unlink(tmpname)
+
 
 class GzipStreamWriteTest(GzipTest, StreamWriteTest):
     def test_source_directory_not_leaked(self):

tests/test/test_zipfile/_path/test_path.py

@@ -274,7 +274,7 @@ def test_pathlike_construction(self, alpharep):
         zipfile_ondisk = self.zipfile_ondisk(alpharep)
         pathlike = FakePath(str(zipfile_ondisk))
         root = zipfile.Path(pathlike)
-        root.root.close()  # gh-137589
+        root.root.close()
 
     @pass_alpharep
     def test_traverse_pathlike(self, alpharep):
@@ -373,7 +373,7 @@ def test_root_on_disk(self, alpharep):
         root = zipfile.Path(self.zipfile_ondisk(alpharep))
         assert root.name == 'alpharep.zip' == root.filename.name
         assert root.stem == 'alpharep' == root.filename.stem
-        root.root.close()  # gh-137589
+        root.root.close()
 
     @pass_alpharep
     def test_suffix(self, alpharep):
@@ -577,11 +577,11 @@ def test_pickle(self, alpharep, path_type, subpath):
         zipfile_ondisk = path_type(str(self.zipfile_ondisk(alpharep)))
         root = zipfile.Path(zipfile_ondisk, at=subpath)
         saved_1 = pickle.dumps(root)
-        root.root.close()  # gh-137589
+        root.root.close()
         restored_1 = pickle.loads(saved_1)
         first, *rest = restored_1.iterdir()
         assert first.read_text(encoding='utf-8').startswith('content of ')
-        restored_1.root.close()  # gh-137589
+        restored_1.root.close()
 
     @pass_alpharep
     def test_extract_orig_with_implied_dirs(self, alpharep):
@@ -593,7 +593,7 @@ def test_extract_orig_with_implied_dirs(self, alpharep):
         # wrap the zipfile for its side effect
         zipfile.Path(zf)
         zf.extractall(source_path.parent)
-        zf.close()  # gh-137589
+        zf.close()
 
     @pass_alpharep
     def test_getinfo_missing(self, alpharep):