[3.14] gh-154325: Skip test_add_file_after_2107() if the file system rejects the timestamp (GH-154328) (GH-154332)
Some file systems (UFS and ZFS on illumos) reject timestamps that do not
fit in 32 bits with EOVERFLOW instead of raising OverflowError.
(cherry picked from commit 63a2709b700d5532b8925caca5bc8b9437e7a1f3)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
[3.14] gh-103925: Fix csv.Sniffer for a quoted field ending a CRLF line (GH-103926) (GH-154323)
"$" does not match before "\r" even in the MULTILINE mode, so such a
field was not found and the delimiter was guessed from character
frequencies instead, which could give a letter.
(cherry picked from commit 70f7c6c0f2ddfd3b447946f1b926776b2a344703)
Co-authored-by: Zhou Wei <lilaboc.cn@gmail.com> Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154307: Fix TemporaryDirectory cleanup on DragonFly BSD (GH-154309) (GH-154312)
On DragonFly BSD, removing a file or directory with the UF_NOUNLINK flag
fails with EISDIR (IsADirectoryError) instead of EPERM, so the cleanup did
not reset the flags. Handle IsADirectoryError the same as PermissionError.
(cherry picked from commit 1c1088b1da5a7484b7b04e90ccc47aa362e709eb)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154308: Clear file flags in os_helper.rmtree() (GH-154310) (GH-154315)
On BSD systems a test may leave behind files or directories with flags
such as UF_IMMUTABLE or UF_NOUNLINK set, which prevent the directory from
being modified or removed. Clear the flags before removing, so cleaning
up a temporary directory does not fail.
(cherry picked from commit 1f649fecb645d70b9c48268839e4177d8fe4e1d1)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154291: Fix socket.has_dualstack_ipv6() on DragonFly BSD (GH-154292) (GH-154300)
On DragonFly BSD setsockopt(IPPROTO_IPV6, IPV6_V6ONLY, 0) succeeds without
actually clearing the flag, so has_dualstack_ipv6() wrongly returned True.
Verify with getsockopt() that IPV6_V6ONLY was cleared.
(cherry picked from commit c7bbf040e721781be8f11a3b0716220835d4de2f)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154283: Make threading.get_native_id() unique across processes on DragonFly (GH-154287) (GH-154297)
On DragonFly BSD lwp_gettid() is only unique within a process (the main
thread is always LWP 1), so combine it with the process id, like Solaris.
(cherry picked from commit 53597df11e2d9f01ade2529c147df24b49685878)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154289: Fix test_socket.testGetServBy on DragonFly (GH-154290) (GH-154295)
DragonFly lists "echo 4/ddp" before "echo 7/tcp" in /etc/services, like
FreeBSD, so avoid the "echo" service there too.
(cherry picked from commit 96059f9446510f594c4148d53bd1a4e3532af697)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-154258: Fix mmap.resize() crash on NetBSD growing a shared anonymous mapping (GH-154259) (GH-154263)
NetBSD mremap() returns a mapping whose grown region is not backed when
growing a shared anonymous mapping, so accessing it crashes. Reject it
with ValueError, as is already done on Linux.
(cherry picked from commit 542b98293108a84fa5d904fb6dcf8bfb5080ec93)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
[3.14] gh-127049: fix race condition in asyncio signalling an unrelated process with ThreadedChildWatcher (GH-153810) (#154221)
gh-127049: fix race condition in asyncio signalling an unrelated process with ThreadedChildWatcher (GH-153810)
(cherry picked from commit 2875d1dc913d2d7810465da4bc3899305da5abc7)
[3.14] gh-154046: docs: fix dead link to codahale timing attacks article in secrets.rst (GH-154047) (#154214)
gh-154046: docs: fix dead link to codahale timing attacks article in secrets.rst (GH-154047)
(cherry picked from commit e8042329b6bb2472b16645c559e4223486b85965)
[3.14] gh-119710: fix asyncio Process.wait() to finish on process exit and not wait for closing of pipes (GH-151983) (#154171)
gh-119710: fix asyncio Process.wait() to finish on process exit and not wait for closing of pipes (GH-151983)
(cherry picked from commit f2521324e6bdf331460abec7c2a88e2d3736b91c)
[3.14] gh-153864: Fix curses window.insch() for non-ASCII characters on a wide build (GH-153865) (GH-154152)
On a wide build, winsch() does not locale-decode a byte above 127, unlike
waddch(), so insch()/mvinsch() stored the wrong character for a non-ASCII
byte of an 8-bit locale ('€' 0xA4 under ISO-8859-15 became U+00A4 '¤').
Decode the byte with btowc() and insert it as a wide character, as addch()
effectively does.
[3.14] gh-139806: Mention pickle error changes in What's New in 3.14 (GH-154020) (GH-154162)
The gh-122311 changes made pickle.dump() and pickle.dumps() raise
PicklingError for some failures that previously raised AttributeError,
ImportError, ValueError or UnicodeEncodeError, depending on the
implementation.
Add an entry about this in the "Porting to Python 3.14" section.
(cherry picked from commit eb447081ed8182f825fa01953ec87cff41f3d585)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
[3.14] gh-153862: Fix curses window.inch() for non-ASCII characters on a wide build (GH-153863) (GH-154151)
On a wide build, winch() returns the low 8 bits of the character's code
point with no locale conversion, so inch()/mvinch() disagreed with instr()
for a non-ASCII character of an 8-bit locale ('€' under ISO-8859-15 gave
0xAC instead of 0xA4). Re-encode the cell to its locale byte via wctob(),
as ncurses does for getbkgd().
[3.14] gh-153926: Fix documented parameter name for `calendar.calendar` and `calendar.prcal` (GH-153927) (#153938)
gh-153926: Fix documented parameter name for `calendar.calendar` and `calendar.prcal` (GH-153927)
gh-153926: Fix documented parameter name for calendar.calendar and calendar.prcal
calendar.calendar() and calendar.prcal() are bound methods of the
module's TextCalendar instance, whose first parameter is named theyear,
but the documentation named it year, so the documented keyword raised
TypeError. Update the documentation to match the real signature, which
also matches the sibling month() and prmonth() functions.
(cherry picked from commit d12434b4661c6be7430002e66a54d920f7d4ca7f)
Combined backport of PRs from main branch, fixing free-threading data-races in itertools:
* gh-123471: make concurrent iteration over `itertools.cycle` safe under free-threading (gh-131212)
* gh-123471: Make itertools.product and itertools.combinations thread-safe (GH-132814)
* gh-123471: Make itertools.chain thread-safe (gh-135689)
* gh-123471: Make concurrent iteration over `itertools.permutations` and `itertools.combinations_with_replacement` thread-safe (gh-144402)
* gh-123471: make concurrent iteration over itertools.accumulate thread-safe (gh-144486)
* gh-123471: Make `itertools.zip_longest` safe in the FT build (gh-146033)
[3.14] gh-151644: Fix data race in sys.setdlopenflags/getdlopenflags under free-threading (gh-151768) (gh-153789)
In free-threading builds, concurrent calls to sys.getdlopenflags() and
sys.setdlopenflags() race on interp->imports.dlopenflags. Fix by
using FT_ATOMIC_LOAD_INT_RELAXED / FT_ATOMIC_STORE_INT_RELAXED in
_PyImport_GetDLOpenFlags and _PyImport_SetDLOpenFlags, consistent with
how analogous interpreter-state integer fields (lazy_imports_mode,
pystats_enabled) are protected.
Relaxed ordering is correct here: dlopenflags is a standalone config
integer with no ordering relationship to other memory.
[3.14] gh-85943: Fix BytesWarning in the struct format cache under -bb (GH-153627) (GH-153835)
Normalize bytes format strings to str before using them as the cache key,
so that equal str and bytes formats no longer collide and get compared.
(cherry picked from commit 190d2ffbb969ad571fd12d4988b37519091194d5)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-153404: Silently ignore non-decimal digits in robots.txt Crawl-delay and Request-rate (GH-153405) (GH-153831)
str.isdigit() returned True for non-decimal Unicode digits such as
U+00B2 SUPERSCRIPT TWO, which int() then rejected with ValueError.
(cherry picked from commit 597ec2d7e5f8bbd5a1eb9e6780444cae46cec3ea)
[3.14] gh-153550: Modernize the older prose sections of the tkinter documentation (GH-153647) (GH-153774)
Refresh the narrative "Tkinter life preserver" and "Handy reference"
sections to match the overhauled reference (gh-86726 and gh-153549).
* Rework "The packer" into a general "Geometry management" section
covering grid, pack and place, and cross-link it with the Grid, Pack
and Place reference classes.
* Rework "The window manager" and "Coupling widget variables", replacing
the dated App(Frame) examples with direct Tk() and ttk examples.
* Replace page-number citations to Ousterhout's book with links to the
relevant Tk man pages.
* Modernize the tkinter.ttk front matter and its Tcl/Tk links, and add
cross-references between the two documents.
* Reconcile the "Packer options" and "Tk option data types" lists with
the reference, and fill in the empty Entry index list.
* Fix the malformed target-selection list in tkinter.dnd.
(cherry picked from commit 93beea7e5a3c5a03253e6812a06f2497505aeb30)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Separate tests for module-level API and for the MimeTypes class.
* Add tests for mimetypes.init() and MimeTypes() with knownfiles and with
explicitly passed files.
Compute probe coordinates from each widget's own realized geometry
instead of hardcoding pixels such as (5, 5) or width - 5, which land on
a different element, or miss it entirely, at a high display scaling.
(cherry picked from commit 87411d0742f070b1424c08b41cf3af49d377b215)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
[3.14] gh-153631: Move to `macos-26` runner for iOS (GH-153632) (#153674)
Switches iOS CI to use the macos-26 runner, and makes a small change to the iOS
build script to improve build stability. xcodebuild doesn't guarantee that the
CoreSimulatorService is running before starting a simulator. If the service
isn't running, xcodebuild reports that no simulators are available, and fails to
start the test app. However, simctl blocks until the simulator is available, and
simctl is used to evaluate the default simulator. So - the iOS build script now
unconditionally determines the default simulator, even if a specific simulator
is requested.
(cherry picked from commit 2dc1a91af801ea896e21f6c6e6cc41f57e8268d9)
[3.14] gh-153037: Make ZstdFile.__next__ raise io.UnsupportedOperation on non-readable files (GH-153045) (#153639)
gh-153037: Make ZstdFile.__next__ raise io.UnsupportedOperation on non-readable files (GH-153045)
Make `ZstdFile.__next__` raise `io.UnsupportedOperation` on non-readable files, consistent with other compression modules.
(cherry picked from commit ed716551e13d1e46a5cd17955657d64b8824626a)
[3.14] gh-90431: Document cancellation behavior of asyncio.wait() and asyncio.as_completed() (GH-153616) (#153618)
gh-90431: Document cancellation behavior of asyncio.wait() and asyncio.as_completed() (GH-153616)
(cherry picked from commit dd2faeb33d5fa0d2635b91ece6eaebd34c61408c)
[3.14] gh-152614: Add raises to `QueueShutDown` in `asyncio.Queue.put_no_wait` and `asyncio.Queue.get_nowait` documentation methods (GH-152681) (#153566)
gh-152614: Add raises to `QueueShutDown` in `asyncio.Queue.put_no_wait` and `asyncio.Queue.get_nowait` documentation methods (GH-152681)
(cherry picked from commit d1a5b54abaca2b443c7642ef3a7be8e1cd077ae3)
[3.14] gh-86726: Fix and improve tkinter documentation and docstrings (GH-153549) (GH-153552)
Correct inaccurate return-type and option descriptions in the tkinter
reference, and add missing module and class docstrings.
(cherry picked from commit 77cb7560c07feed0d5b144c0c1735e1620c96f51)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
[3.14] gh-126877: Fix the configure check for Tcl/Tk with optimizing compilers (GH-153543) (GH-153554)
The check assigned the addresses of Tcl_Init() and Tk_Init() to unused
variables, which optimizing compilers can eliminate, so it linked even
when the libraries were missing. Call the functions instead.
(cherry picked from commit ff0a9ae269fa38c21f01bdea1a7ff5d439c4f681)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Victor Stinner [Sat, 11 Jul 2026 09:47:03 +0000 (11:47 +0200)]
[3.14] gh-152785: Upgrade Ubuntu from 24.04 to 26.04 in GitHub Actions (#152717) (#153416) (#153526)
* [3.15] gh-152785: Upgrade Ubuntu from 24.04 to 26.04 in GitHub Actions (#152717) (#153416)
gh-152785: Upgrade Ubuntu from 24.04 to 26.04 in GitHub Actions (#152717)
* Replace "ubuntu-24.04" with "ubuntu-26.04" on most jobs.
* Replace "ubuntu-latest" with "ubuntu-26.04" for Cross build Linux.
* Replace "ubuntu-latest" with "ubuntu-slim" for small workloads.
* Update ".github/actionlint.yaml" to allow "ubuntu-26.04"
and "ubuntu-26.04-arm" images.
Backport changes:
* Keep Ubuntu 24.04 and llvm.sh to install LLVM 19 and LLVM 20.
* WASI uses ubuntu-26.04, not ubuntu-26.04-arm.
projects / thirdparty / Python / cpython.git / log
