gh-121021: Support writing extended timestamps for zipfile

Introduce `with_ext_timestamps` parameter to `ZipFile` and
`ZipInfo.from_file()` to fill extra data with extended timestamps.

Author: danny0838

Doc/library/zipfile.rst

@@ -178,7 +178,7 @@ ZipFile objects
 
 .. class:: ZipFile(file, mode='r', compression=ZIP_STORED, allowZip64=True, \
                    compresslevel=None, *, strict_timestamps=True, \
-                   metadata_encoding=None)
+                   with_ext_timestamps=False, metadata_encoding=None)
 
    Open a ZIP file, where *file* can be a path to a file (a string), a
    file-like object or a :term:`path-like object`.
@@ -227,6 +227,9 @@ ZipFile objects
    Similar behavior occurs with files newer than 2107-12-31,
    the timestamp is also set to the limit.
 
+   The *with_ext_timestamps* controls whether to fill the extended timestamps
+   when writing files to the archive.
+
    When mode is ``'r'``, *metadata_encoding* may be set to the name of a codec,
    which will be used to decode metadata such as the names of members and ZIP
    comments.
@@ -285,6 +288,9 @@ ZipFile objects
       Added support for specifying member name encoding for reading
       metadata in the zipfile's directory and file headers.
 
+   .. versionchanged:: next
+      Added the *with_ext_timestamps* keyword-only parameter.
+
 
 .. method:: ZipFile.close()
 
@@ -885,7 +891,8 @@ There is one classmethod to make a :class:`ZipInfo` instance for a filesystem
 file:
 
 .. classmethod:: ZipInfo.from_file(filename, arcname=None, *, \
-                                   strict_timestamps=True)
+                                   strict_timestamps=True, \
+                                   with_ext_timestamps=False)
 
    Construct a :class:`ZipInfo` instance for a file on the filesystem, in
    preparation for adding it to a zip file.
@@ -902,6 +909,10 @@ file:
    Similar behavior occurs with files newer than 2107-12-31,
    the timestamp is also set to the limit.
 
+   Setting ``with_ext_timestamps=True`` fills the file's extended timestamps
+   to the extra data, which allows other ZIP tools to recover the timestamp
+   more accurately when extracting the file.
+
    .. versionadded:: 3.6
 
    .. versionchanged:: 3.6.2
@@ -910,6 +921,9 @@ file:
    .. versionchanged:: 3.8
       Added the *strict_timestamps* keyword-only parameter.
 
+   .. versionchanged:: next
+      Added the *with_ext_timestamps* keyword-only parameter.
+
 
 Instances have the following methods and attributes:
 

Lib/test/test_zipfile/test_core.py

@@ -676,6 +676,91 @@ def test_add_file_after_2107(self):
             zinfo = zipfp.getinfo(TESTFN)
             self.assertEqual(zinfo.date_time, (2107, 12, 31, 23, 59, 59))
 
+    def test_add_file_with_ext_timestamp(self):
+        """Check that calling ZipFile.write() sets extra data according to
+        with_ext_timestamps parameter."""
+        mtime = 946684800.123456
+        mtime_ns = 946684800_123456789
+        atime_ns = 946684800_987654321
+        ctime_ns = 946684800_555555555
+
+        with mock.patch('os.stat_result.st_mtime', mtime), \
+             mock.patch('os.stat_result.st_mtime_ns', mtime_ns), \
+             mock.patch('os.stat_result.st_atime_ns', atime_ns), \
+             mock.patch('os.stat_result.st_ctime_ns', ctime_ns):
+
+            # with_ext_timestamps=False (default)
+            with zipfile.ZipFile(TESTFN2, "w") as zipfp:
+                zipfp.write(TESTFN)
+
+            with zipfile.ZipFile(TESTFN2) as zipfp:
+                zinfo = zipfp.infolist()[0]
+
+            self.assertEqual(zinfo.extra, b'')
+
+            # with_ext_timestamps=True
+            with zipfile.ZipFile(TESTFN2, "w", with_ext_timestamps=True) as zipfp:
+                zipfp.write(TESTFN)
+
+            with zipfile.ZipFile(TESTFN2) as zipfp:
+                zinfo = zipfp.infolist()[0]
+
+            self.assertEqual(zinfo.date_time[0], 2000)
+
+            # NTFS Extra Field (0x000a)
+            delta = 116444736000000000
+            ntfs_field = struct.unpack_from('<HHLHHQQQ', zinfo.extra)
+            self.assertEqual(ntfs_field, (
+                0x000a, 32,
+                0, 0x0001, 24,
+                mtime_ns // 100 + delta,
+                atime_ns // 100 + delta,
+                ctime_ns // 100 + delta,
+            ))
+
+            # Extended timestamp (0x5455)
+            ut_field = struct.unpack_from('<HHBL', zinfo.extra, struct.calcsize('<HHLHHQQQ'))
+            self.assertEqual(ut_field, (0x5455, 5, 1, int(mtime)))
+
+    def test_add_file_with_ext_timestamp_after_2038(self):
+        """Extended timestamp field should exist for a timestamp after
+        2038-01-19T03:14:07Z."""
+        mtime = 2147483648.123456  # 2038-01-19T03:14:08.123456Z
+
+        with mock.patch('os.stat_result.st_mtime', mtime):
+            with zipfile.ZipFile(TESTFN2, "w", strict_timestamps=False,
+                                 with_ext_timestamps=True) as zipfp:
+                zipfp.write(TESTFN)
+
+            with zipfile.ZipFile(TESTFN2) as zipfp:
+                zinfo = zipfp.infolist()[0]
+
+            self.assertEqual(zinfo.date_time[0], 2038)
+
+            # Extended timestamp (0x5455)
+            ntfs_field_len = struct.calcsize('<HHLHHQQQ')
+            ut_field = struct.unpack_from('<HHBL', zinfo.extra, ntfs_field_len)
+            self.assertEqual(ut_field, (0x5455, 5, 1, int(mtime)))
+
+    def test_add_file_with_ext_timestamp_after_2106(self):
+        """Extended timestamp field should not exist for a timestamp after
+        2106-02-07T06:28:15Z."""
+        mtime = 4294967296.123456  # 2106-02-07T06:28:16.123456Z
+
+        with mock.patch('os.stat_result.st_mtime', mtime):
+            with zipfile.ZipFile(TESTFN2, "w", strict_timestamps=False,
+                                 with_ext_timestamps=True) as zipfp:
+                zipfp.write(TESTFN)
+
+            with zipfile.ZipFile(TESTFN2) as zipfp:
+                zinfo = zipfp.infolist()[0]
+
+            self.assertEqual(zinfo.date_time[0], 2106)
+
+            # Only an NTFS Extra Field (0x000a) exists
+            ntfs_field_len = struct.calcsize('<HHLHHQQQ')
+            self.assertEqual(len(zinfo.extra), ntfs_field_len)
+
 
 @requires_zlib()
 class DeflateTestsWithSourceFile(AbstractTestsWithSourceFile,

Lib/zipfile/__init__.py

@@ -629,7 +629,8 @@ def _decodeExtra(self, filename_crc):
             extra = extra[ln+4:]
 
     @classmethod
-    def from_file(cls, filename, arcname=None, *, strict_timestamps=True):
+    def from_file(cls, filename, arcname=None, *, strict_timestamps=True,
+                  with_ext_timestamps=False):
         """Construct an appropriate ZipInfo for a file on the filesystem.
 
         filename should be the path to a file or directory on the
@@ -665,6 +666,27 @@ def from_file(cls, filename, arcname=None, *, strict_timestamps=True):
         else:
             zinfo.file_size = st.st_size
 
+        if with_ext_timestamps:
+            extras = {}
+
+            # NTFS Extra Field (0x000a)
+            delta = 116444736000000000
+            ft_mtime = st.st_mtime_ns // 100 + delta
+            ft_atime = st.st_atime_ns // 100 + delta
+            ft_ctime = st.st_ctime_ns // 100 + delta
+            ntfs_tag = struct.pack('<LHHQQQ', 0, 0x0001, 24, ft_mtime, ft_atime, ft_ctime)
+            extras[0x000a] = struct.pack('<HH', 0x000a, len(ntfs_tag)) + ntfs_tag
+
+            # Extended timestamp (0x5455)
+            # According to libzip's doc, the timestamps should be 4-byte
+            # unsigned integers:
+            # https://libzip.org/specifications/extrafld.txt
+            mtime = int(st.st_mtime)
+            if 0 <= mtime <= 0xFFFF_FFFF:
+                extras[0x5455] = struct.pack('<HHBL', 0x5455, 5, 0x01, mtime)
+
+            zinfo.extra = b''.join(extras.values()) + _Extra.strip(zinfo.extra, extras)
+
         return zinfo
 
     def _for_archive(self, archive):
@@ -1896,7 +1918,8 @@ class ZipFile:
     _ignore_invalid_names = False
 
     def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=True,
-                 compresslevel=None, *, strict_timestamps=True, metadata_encoding=None):
+                 compresslevel=None, *, strict_timestamps=True,
+                 with_ext_timestamps=False, metadata_encoding=None):
         """Open the ZIP file with mode read 'r', write 'w', exclusive create
         'x', or append 'a'."""
         if mode not in ('r', 'w', 'x', 'a'):
@@ -1915,6 +1938,7 @@ def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=True,
         self.pwd = None
         self._comment = b''
         self._strict_timestamps = strict_timestamps
+        self._with_ext_timestamps = with_ext_timestamps
         self.metadata_encoding = metadata_encoding
 
         # Check that we don't try to write with nonconforming codecs
@@ -2526,7 +2550,8 @@ def write(self, filename, arcname=None,
             )
 
         zinfo = ZipInfo.from_file(filename, arcname,
-                                  strict_timestamps=self._strict_timestamps)
+                                  strict_timestamps=self._strict_timestamps,
+                                  with_ext_timestamps=self._with_ext_timestamps)
 
         if zinfo.is_dir():
             zinfo.compress_size = 0

Misc/NEWS.d/next/Core_and_Builtins/2026-06-30-18-07-15.gh-issue-121021.pcgjVP.rst

@@ -0,0 +1,4 @@
+Introduce ``with_ext_timestamps`` parameter to
+:class:`~zipfile.ZipFile` and :meth:`~zipfile.ZipInfo.from_file` to
+fill extra data with the file's extended timestamps when writing a
+file to the archive.