lseek() always returns 0 for character pseudo-devices like
`/dev/urandom` (for other non-regular files, e.g. `/dev/stdin`, it
always returns -1, to which CPython reacts by raising appropriate
exceptions). They are thus technically seekable despite not having seek
semantics.
When calling read() on e.g. an instance of `io.BufferedReader` that
wraps such a file, `BufferedReader` reads ahead, filling its buffer,
creating a discrepancy between the number of bytes read and the internal
`tell()` always returning 0, which previously resulted in e.g.
`BufferedReader.tell()` or `BufferedReader.seek()` being able to return
positions < 0 even though these are supposed to be always >= 0.
Invariably keep the return value non-negative by returning
max(former_return_value, 0) instead, and add some corresponding tests.
(cherry picked from commit 26800cf25a)
Co-authored-by: 6t8k <58048945+6t8k@users.noreply.github.com>
Immediate merits:
* eliminate complex workarounds for 'z' format support
(NOTE: mpdecimal recently added 'z' support, so this becomes
efficient in the long term.)
* fix 'z' format memory leak
* fix 'z' format applied to 'F'
* fix missing '#' format support
Suggested and prototyped by Stefan Krah.
Fixes gh-114563, gh-91060
(cherry picked from commit 72340d15cd)
Co-authored-by: John Belmonte <john@neggie.net>
Co-authored-by: Stefan Krah <skrah@bytereef.org>
On Alpine Linux it could leave some field non-initialized.
(cherry picked from commit d22c066b80)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
gh-104522: Fix OSError raised when run a subprocess (GH-114195)
Only set filename to cwd if it was caused by failed chdir(cwd).
_fork_exec() now returns "noexec:chdir" for failed chdir(cwd).
(cherry picked from commit e2c097ebde)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
Co-authored-by: Robert O'Shea <PurityLake@users.noreply.github.com>
This avoids impact on later parts of the application which may be able to do things they otherwise shouldn't.
(cherry picked from commit de4ced54eb)
Co-authored-by: Steve Dower <steve.dower@python.org>
(cherry picked from commit c31be58da8)
Co-authored-by: AN Long <aisk@users.noreply.github.com>
Co-authored-by: Alex Waygood <Alex.Waygood@Gmail.com>
Co-authored-by: Pieter Eendebak <pieter.eendebak@gmail.com>
Co-authored-by: Erlend E. Aasland <erlend.aasland@protonmail.com>
io.TextIOWrapper was dropping the internal decoding buffer
during read() and write() calls.
(cherry picked from commit 73c9326563)
Co-authored-by: Zackery Spytz <zspytz@gmail.com>
The length field of StgDictObject for Structure class contains now
the total number of items in ffi_type_pointer.elements (excluding
the trailing null).
The old behavior of using the number of elements in the parent class can
cause the array to be truncated when it is copied, especially when there
are multiple layers of subclassing.
(cherry picked from commit 5f3cc90a12)
Co-authored-by: Jeffrey Kintscher <49998481+websurfer5@users.noreply.github.com>
Fix the same issue of PR #112604 on PPC64LE platform
Refactor tests to make easier to add more platfroms if needed.
(cherry picked from commit 6644ca45cd)
Change-Id: I31730a3ebe558570ce1d7a3b26db8392f18d1770
* Fix a crash when pass UINT_MAX.
* Fix an integer overflow on 64-bit non-Windows platforms.
(cherry picked from commit 0daf555c6f)
Co-authored-by: Zackery Spytz <zspytz@gmail.com>
First fix resolve situation when pyexpat module (which contains expat_CAPI
capsule) deallocates before _elementtree, so we need to hold a strong
reference to pyexpat module to.
Second fix resolve situation when module state is deallocated before
deallocation of XMLParser instances, which uses module state to clear
some stuff.
(cherry picked from commit 894f0e573d)
Co-authored-by: Kirill Podoprigora <kirill.bast9@mail.ru>
Set MAX_STRUCT_SIZE to 32 in stgdict.c when on Arm platforms.
This because on Arm platforms structs with at most 4 elements of any
floating point type values can be passed through registers. If the type
is double the maximum size of the struct is 32 bytes.
On x86-64 Linux, it's maximum 16 bytes hence we need to differentiate.
(cherry picked from commit bc68f4a4ab)
Restore `subprocess`'s intended use of `vfork()` by default for performance on Linux;
also fixes the behavior of `extra_groups=[]` which was unintentionally broken in 3.12.0:
Fixed a performance regression in 3.12's :mod:`subprocess` on Linux where it
would no longer use the fast-path ``vfork()`` system call when it could have
due to a logic bug, instead falling back to the safe but slower ``fork()``.
Also fixed a security bug introduced in 3.12.0. If a value of ``extra_groups=[]``
was passed to :mod:`subprocess.Popen` or related APIs, the underlying
``setgroups(0, NULL)`` system call to clear the groups list would not be made
in the child process prior to ``exec()``.
The security issue was identified via code inspection in the process of
fixing the first bug. Thanks to @vain for the detailed report and
analysis in the initial bug on Github.
(cherry picked from commit 9fe7655c6c)
+ Reword NEWS for the bugfix/security release. (mentions the assigned CVE number)
Co-authored-by: Gregory P. Smith [Google LLC] <greg@krypto.org>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
gh-105967: Work around a macOS bug, limit zlib C library crc32 API calls to 1gig (GH-112615)
Work around a macOS bug, limit zlib crc32 calls to 1GiB.
Without this, `zlib.crc32` and `binascii.crc32` could produce incorrect
results on multi-gigabyte inputs depending on the macOS version's Apple
supplied zlib implementation.
(cherry picked from commit 4eddb4c9d9)
Co-authored-by: Gregory P. Smith <greg@krypto.org>
gh-112105: Make completer delims work on libedit (gh-112106)
(cherry picked from commit 2df26d8348)
Co-authored-by: Tian Gao <gaogaotiantian@hotmail.com>
The existence of background threads running on a subinterpreter was preventing interpreters from getting properly destroyed, as well as impacting the ability to run the interpreter again. It also affected how we wait for non-daemon threads to finish.
We add PyInterpreterState.threads.main, with some internal C-API functions.
(cherry-picked from commit 1dd9dee45d)
This change makes sure sys.path[0] is set properly for subinterpreters. Before, it wasn't getting set at all.
This change does not address the broader concerns from gh-109853.
(cherry-picked from commit a040a32ea2)
* [3.12] gh-112358: Fix Python 3.12 regression with subclassing struct.Struct. (GH-112424)
Revert commit c8c0afc713 (PR GH-94532),
which moved `struct.Struct` initialisation from `Struct.__init__` to `Struct.__new__`.
This caused issues with code in the wild that subclasses `struct.Struct`..
(cherry picked from commit 9fe60340d7)
Co-authored-by: Mark Dickinson <dickinsm@gmail.com>
* Remove unrelated test
In non-debug more the check for the "errors" argument is skipped,
and then PyUnicode_AsUTF8() can fail, but its result was not checked.
Co-authored-by: Victor Stinner <vstinner@python.org>
(cherry picked from commit 9302f05f9a)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
* Fix crash when encoding is not string or None.
* Fix crash when both line_buffering and write_through raise exception
when converted ti int.
* Add a number of tests for constructor and reconfigure() method
with invalid arguments.
(cherry picked from commit ee06fffd38)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
gh-111777: Fix assertion errors on incorrectly still-tracked GC object destruction (GH-111778)
In PyObject_GC_Del, in Py_DEBUG mode, when warning about GC objects that
were not properly untracked before starting destruction, take care to
untrack the object _before_ warning, to avoid triggering a GC run and
causing the problem the code tries to warn about. Also make sure to save and
restore any pending exceptions, which the warning would otherwise clobber or
trigger an assertion error on.
(cherry picked from commit ce6a533c4b)
Co-authored-by: T. Wouters <thomas@python.org>
Fix undefined behaviour in datetime.time.fromisoformat() (GH-111982)
Fix undefined behaviour in datetime.time.fromisoformat() when parsing a string without a timezone. 'tzoffset' is not assigned to by parse_isoformat_time if it returns 0, but time_fromisoformat then passes tzoffset to another function, which is undefined behaviour (even if the function in question does not use the value).
(cherry picked from commit 21615f77b5)
Co-authored-by: T. Wouters <thomas@python.org>
* [3.12] gh-110395: invalidate open kqueues after fork (GH-110517)
Invalidate open select.kqueue instances after fork as the fd will be invalid in the child.
(cherry picked from commit a6c1c04d4d)
Co-authored-by: Davide Rizzo <sorcio@gmail.com>
* move assert to after the child dying
this is in `main` via https://github.com/python/cpython/pull/111816/files
