gh-80620: Support negative timestamps on windows in `time.gmtime`, `time.localtime`, and `datetime` module by gesslerpd · Pull Request #143463 · python/cpython

gesslerpd · 2026-01-06T03:31:28Z

Previously, negative timestamps (representing dates before 1970-01-01) were not supported on Windows due to platform limitations. The changes introduce a fallback implementation using the Windows FILETIME API, allowing negative timestamps to be correctly handled in both UTC and local time conversions. Additionally, related test code is updated to remove Windows-specific skips and error handling, ensuring consistent behavior across platforms.

This may have some overlap and/or impact on the proposed changes in #134461

Issue: Inconsistencies with datetime.fromtimestamp(t) when t < 0 #80620

Other related issues:

vstinner

I would prefer to change datetime in a separated PR, and restrict this change to time.gmtime() and time.localtime() changes.

You should document time.gmtime() and time.localtime() changes in a NEWS entry.

vstinner · 2026-01-08T14:55:36Z

Python/pytime.c

+    /* Calculate day of year using Windows FILETIME difference */
+    // SYSTEMTIME st_jan1 = {st_result.wYear, 1, 0, 1, 0, 0, 0, 0};
+    // FILETIME ft_jan1, ft_date;
+    // if (!SystemTimeToFileTime(&st_jan1, &ft_jan1) ||
+    //     !SystemTimeToFileTime(&st_result, &ft_date)) {
+    //     PyErr_SetFromWindowsErr(0);
+    //     return -1;
+    // }
+    // ULARGE_INTEGER jan1, date;
+    // jan1.LowPart = ft_jan1.dwLowDateTime;
+    // jan1.HighPart = ft_jan1.dwHighDateTime;
+    // date.LowPart = ft_date.dwLowDateTime;
+    // date.HighPart = ft_date.dwHighDateTime;
+    // /* Convert 100-nanosecond intervals to days */
+    // LONGLONG days_diff = (date.QuadPart - jan1.QuadPart) / (24LL * 60 * 60 * HUNDRED_NS_PER_SEC);
+
+    // tm->tm_yday = (int)days_diff;


Please remove commented/dead code.

Do you think that yday is important to have? Doesn't impact datetime AFAIK but it does show up in struct_time objects returned from gmtime/localtime. For completeness with respect to time module it could be put back in.

If possible, it would be better to fill tm_yday as well.

Restored tm_yday calculation, I think the windows API version was a bit heavy so simplified it by adding custom util function.

vstinner · 2026-01-08T15:02:51Z

test_time.test_mktime() can be updated: except (OverflowError, OSError) is no longer needed if I'm right.

gesslerpd · 2026-01-08T15:33:32Z

test_time.test_mktime() can be updated: except (OverflowError, OSError) is no longer needed if I'm right.

time.mktime still doesn't support negative timestamps with this change, gmtime and localtime do as side effect, was mostly making these changes for datetime.

I was going to tweak callsites so it doesn't impact time module at all but thought I'd leave that choice to the review.

vstinner · 2026-01-08T15:40:54Z

I was going to tweak callsites so it doesn't impact time module at all but thought I'd leave that choice to the review.

Oh, it's a good thing to implement the new feature in the time module.

Existing test_time test:

    def test_mktime(self):
        # Issue #1726687
        for t in (-2, -1, 0, 1):
            try:
                tt = time.localtime(t)
            except (OverflowError, OSError):
                pass
            else:
                self.assertEqual(time.mktime(tt), t)

If localtime() is modified to support negative timestamp, but mktime() can still fail, the test should be modified to something like:

        for t in (-2, -1, 0, 1):
            tt = time.localtime(t)
            try:
                t2 = time.mktime(tt)
            except (OverflowError, OSError):
                pass
            else:
                self.assertEqual(t2, t)

Lib/test/test_time.py

vstinner

Thanks for additional test on leap and non-leap years.

Python/pytime.c

vstinner

LGTM. I just have minor coding style suggestions on comments.

Python/pytime.c

Co-authored-by: Victor Stinner <vstinner@python.org>

gesslerpd · 2026-01-14T22:54:19Z

Suggestions applied, thanks again for review!

vstinner · 2026-01-15T09:51:27Z

Merged, thanks. This was a long awaited feature :-)

bedevere-bot · 2026-01-15T09:59:38Z

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot x86 Debian Installed with X 3.x (no tier) has failed when building commit f5685a2.

What do you need to do:

Don't panic.
Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/1244/builds/7242) and take a look at the build logs.
Check if the failure is related to this commit (f5685a2) or if it is a false positive.
If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/#/builders/1244/builds/7242

Summary of the results of the build (if available):

==

Click to see traceback logs

Traceback (most recent call last):
  File �[35m"/buildbot/buildarea/3.x.ware-debian-x86.installed/build/target/lib/python3.15/test/test_time.py"�[0m, line �[35m208�[0m, in �[35mtest_gmtime�[0m
    res = time.gmtime(t)
�[1;35mOverflowError�[0m: �[35mtimestamp out of range for platform time_t�[0m

bedevere-bot · 2026-01-15T10:01:01Z

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot ARM Raspbian 3.x (tier-3) has failed when building commit f5685a2.

What do you need to do:

Don't panic.
Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/424/builds/12631) and take a look at the build logs.
Check if the failure is related to this commit (f5685a2) or if it is a false positive.
If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/#/builders/424/builds/12631

Failed tests:

test_time

Failed subtests:

test_gmtime - test.test_time.TimeTestCase.test_gmtime

Summary of the results of the build (if available):

==

Click to see traceback logs

Traceback (most recent call last):
  File "/var/lib/buildbot/workers/3.x.gps-raspbian.nondebug/build/Lib/test/test_time.py", line 208, in test_gmtime
    res = time.gmtime(t)
OverflowError: timestamp out of range for platform time_t

vstinner · 2026-01-15T10:22:15Z

Oh, test_gmtime() fails on 32-bit systems: I wrote #143861 to fix the test.

bedevere-bot · 2026-01-15T10:34:46Z

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot ARM64 macOS 3.x (tier-2) has failed when building commit f5685a2.

What do you need to do:

Don't panic.
Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
Go to the page of the buildbot that failed (https://buildbot.python.org/#/builders/725/builds/12561) and take a look at the build logs.
Check if the failure is related to this commit (f5685a2) or if it is a false positive.
If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/#/builders/725/builds/12561

Failed tests:

test_urllib2net

Summary of the results of the build (if available):

==

Click to see traceback logs

remote: Enumerating objects: 15, done.        
remote: Counting objects:   7% (1/13)        
remote: Counting objects:  15% (2/13)        
remote: Counting objects:  23% (3/13)        
remote: Counting objects:  30% (4/13)        
remote: Counting objects:  38% (5/13)        
remote: Counting objects:  46% (6/13)        
remote: Counting objects:  53% (7/13)        
remote: Counting objects:  61% (8/13)        
remote: Counting objects:  69% (9/13)        
remote: Counting objects:  76% (10/13)        
remote: Counting objects:  84% (11/13)        
remote: Counting objects:  92% (12/13)        
remote: Counting objects: 100% (13/13)        
remote: Counting objects: 100% (13/13), done.        
remote: Compressing objects:   9% (1/11)        
remote: Compressing objects:  18% (2/11)        
remote: Compressing objects:  27% (3/11)        
remote: Compressing objects:  36% (4/11)        
remote: Compressing objects:  45% (5/11)        
remote: Compressing objects:  54% (6/11)        
remote: Compressing objects:  63% (7/11)        
remote: Compressing objects:  72% (8/11)        
remote: Compressing objects:  81% (9/11)        
remote: Compressing objects:  90% (10/11)        
remote: Compressing objects: 100% (11/11)        
remote: Compressing objects: 100% (11/11), done.        
remote: Total 15 (delta 2), reused 2 (delta 2), pack-reused 2 (from 2)        
From https://github.com/python/cpython
 * branch                    main       -> FETCH_HEAD
Note: switching to 'f5685a266b252455e03ef8e6055eaf4007ec749d'.

You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.

If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:

  git switch -c <new-branch-name>

Or undo this operation with:

  git switch -

Turn off this advice by setting config variable advice.detachedHead to false

HEAD is now at f5685a266b2 gh-80620: Support negative timestamps on windows in `time.gmtime`, `time.localtime`, and `datetime` module (#143463)
Switched to and reset branch 'main'

make: *** [buildbottest] Error 2

…e`, `time.localtime`, and `datetime` module (python#143463) Previously, negative timestamps (representing dates before 1970-01-01) were not supported on Windows due to platform limitations. The changes introduce a fallback implementation using the Windows FILETIME API, allowing negative timestamps to be correctly handled in both UTC and local time conversions. Additionally, related test code is updated to remove Windows-specific skips and error handling, ensuring consistent behavior across platforms. Co-authored-by: Victor Stinner <vstinner@python.org>

gesslerpd added 2 commits January 5, 2026 21:10

init

c8eab29

remove hacks for previous windows negative timestamp behavior

8ec0eca

gesslerpd requested review from abalkin and pganssle as code owners January 6, 2026 03:31

bedevere-app bot mentioned this pull request Jan 6, 2026

Inconsistencies with datetime.fromtimestamp(t) when t < 0 #80620

Closed

bedevere-app bot added the awaiting review label Jan 6, 2026

gesslerpd changed the title ~~gh-80620: Support negative timestamps on windows for datetime.datetime~~ gh-80620: Support negative timestamps on windows in datetime module Jan 6, 2026

gesslerpd added 2 commits January 5, 2026 22:24

update datetime tests

3d9f1a2

add blurb

64c70c3

gesslerpd force-pushed the windows-datetime-pre-epoch branch from 0b9b5e8 to 64c70c3 Compare January 6, 2026 04:24

vstinner reviewed Jan 8, 2026

View reviewed changes

gesslerpd added 2 commits January 8, 2026 17:14

restore tm_yday calculation, improve time test

4f8c2ae

update blurb

d58444c

gesslerpd changed the title ~~gh-80620: Support negative timestamps on windows in datetime module~~ gh-80620: Support negative timestamps on windows in time.gmtime, time.localtime, and datetime module Jan 8, 2026

vstinner reviewed Jan 9, 2026

View reviewed changes

Lib/test/test_time.py Show resolved Hide resolved

gesslerpd added 3 commits January 9, 2026 12:02

add gmtime pre-epoch test

97c1bfb

test yday jumps

2224f68

fix tm_wday copy/paste mistake

09a3dac

vstinner reviewed Jan 12, 2026

View reviewed changes

Python/pytime.c Outdated Show resolved Hide resolved

Python/pytime.c Outdated Show resolved Hide resolved

Python/pytime.c Outdated Show resolved Hide resolved

review fixes

d711f28

vstinner approved these changes Jan 14, 2026

View reviewed changes

Python/pytime.c Outdated Show resolved Hide resolved

Python/pytime.c Outdated Show resolved Hide resolved

bedevere-app bot added awaiting merge and removed awaiting review labels Jan 14, 2026

Apply suggestions from code review

a34079a

Co-authored-by: Victor Stinner <vstinner@python.org>

vstinner merged commit f5685a2 into python:main Jan 15, 2026
47 checks passed

bedevere-app bot removed the awaiting merge label Jan 15, 2026

vstinner mentioned this pull request Jan 15, 2026

gh-80620: Support negative values in fromtimestamp on Windows using 0 + timedelta #134461

Closed

hugovk mentioned this pull request Mar 10, 2026

Negative timestamp failure during DST on Windows #145731

Closed

Uh oh!

Conversation

gesslerpd commented Jan 6, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vstinner left a comment

Choose a reason for hiding this comment

Uh oh!

vstinner Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

gesslerpd Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

vstinner Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

gesslerpd Jan 8, 2026

Choose a reason for hiding this comment

Uh oh!

vstinner commented Jan 8, 2026

Uh oh!

gesslerpd commented Jan 8, 2026

Uh oh!

vstinner commented Jan 8, 2026

Uh oh!

Uh oh!

vstinner left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

vstinner left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

gesslerpd commented Jan 14, 2026

Uh oh!

Uh oh!

vstinner commented Jan 15, 2026

Uh oh!

bedevere-bot commented Jan 15, 2026

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Uh oh!

bedevere-bot commented Jan 15, 2026

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Uh oh!

vstinner commented Jan 15, 2026

Uh oh!

bedevere-bot commented Jan 15, 2026

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

gesslerpd commented Jan 6, 2026 •

edited

Loading