repo stringclasses 1
value | instance_id stringlengths 20 22 | problem_statement stringlengths 126 60.8k | merge_commit stringlengths 40 40 | base_commit stringlengths 40 40 |
|---|---|---|---|---|
python/cpython | python__cpython-128873 | # Unsused argument in `_PyCode_Quicken`
`_PyCode_Quicken` in `specialize.c` accepts `PyObject *consts` but does not use it. Not sure which type of issue to use, so creating blank one.
<!-- gh-linked-prs -->
### Linked PRs
* gh-128873
<!-- /gh-linked-prs -->
| e6c76b947b2d0708cea1d5e090d02e79fda1cc26 | ecd2f84555bf494007420ba5a37adf8d1c23e856 |
python/cpython | python__cpython-128864 | # [C API] Deprecate private functions which have a public replacement
Hi,
Python 3.13 and 3.14 added public functions to replace private functions. For example, a public function `PyBytes_Join()` was added to replace the private function `_PyBytes_Join()`.
I propose to deprecate the private functions which have a pu... | 9012fa741d55419dc77c5c191794eb93e71ae9a4 | 470a0a68ebbbb4254f1a3e8e22cce0c3a0827055 |
python/cpython | python__cpython-128926 | # Make `_Py_TryIncref` public as an unstable API as `PyUnstable_TryIncref()`
# Feature or enhancement
We should make `_Py_TryIncref` public as function with the following signature:
```c
int PyUnstable_TryIncref(PyObject *op);
```
**EDIT**: Renamed to `PyUnstable_TryIncref` in accordance with Victor's suggestion.
... | d23f5701ad2a64c7877bb37258420c417357423e | 7dd0a7e52ee832559b89d5ccba732c8e91260df8 |
python/cpython | python__cpython-128941 | # JIT: implement memory pystats collection
# Feature or enhancement
### Proposal:
In order to improve the JIT memory allocation, we need to know how much memory it uses.
Specifically it would be useful to know the following metrics:
* code total memory
* data total memory
* trampoline total memory
* padding total m... | 567394517a10c9a9f3af25a31009589ae2c50f1b | e6c76b947b2d0708cea1d5e090d02e79fda1cc26 |
python/cpython | python__cpython-128841 | # IPv6 address parsing doesn't limit buffer size
# Bug report
### Bug description:
IPv6 addresses have a maximum length (8 colon-separated parts) but the current implementation doesn't limit the length. Similar issue to https://github.com/django/django/commit/ca2be7724e1244a4cb723de40a070f873c6e94bf
### CPython ver... | 47f1161d3a2bec52b5b5e952150141709c247da2 | 973b8f69d31b4d86c37e0b7194a209f4f2efff06 |
python/cpython | python__cpython-128817 | # test_doctests failure when running tests with -werror on platforms without readline (win32, wasi)
# Bug report
### Bug description:
```
======================================================================
FAIL: test_lineendings (test.test_doctest.test_doctest)
Doctest: test.test_doctest.test_doctest.test_lineend... | 599be687ec7327c30c6469cf743aa4ee9e82232d | 1a1056d394a489d229b26c06c7c79aa9c06696f6 |
python/cpython | python__cpython-131703 | # Deprecate PyComplexObject.cval and soft-deprecate _Py_c*() API
### Proposal:
Suggested by @vstinner in https://github.com/python/cpython/pull/124829#issuecomment-2588284712.
I would agree, as these routines could be useful (and it seems, they are used by few projects) to implement mathematical functions just like ... | 79f7c67bf607f9d4f39a299aa9a1a4f7b317617c | 5ffb89420c3fb41cb6cc795c9f983a8b5cc207ab |
python/cpython | python__cpython-128808 | # Add marking phase to free-threaded cyclic GC
# Feature or enhancement
### Proposal:
Like gh-126491 but implemented for the free-threaded version of the cyclic GC. This is likely a performance win for mostly the same reasons. In the free-threaded version, the marking phase avoids iterating over the mimalloc arena... | 080f444a58dae1bb76a87ce746a09fd499792cfb | 8d8b854824c4723d7c5924f1d5c6a397ea7214a5 |
python/cpython | python__cpython-128971 | # Unexpected location of naked exception wrapped in except* block
# Bug report
### Bug description:
When an `ExceptionGroup` is implicitly constructed by catching a naked exception in an `except*` block, the traceback of the `ExceptionGroup` seems to point to the stackframe above where it is logically created:
```p... | c39ae8922bad3e5ceeafa05891536c1584b6f3db | 9e52e553f4a906c120f807e940891f7325011b67 |
python/cpython | python__cpython-129184 | # Python 3.14.0a3 regression: creating venv with `--system-site-packages` doesn't work (system packages are not detected)
# Bug report
### Bug description:
venv created with `--system-site-packages` in Python 3.14.0a3 doesn't really see the site packages. This happens also with the current main.
Bisected to c... | a549f439384b4509b25639337ffea21c2e55d452 | 99849ee0d3ebcddc97b6aeaf389f43a12f541068 |
python/cpython | python__cpython-128726 | # configure warnings as errors in the test suite
# Feature or enhancement
### Proposal:
It's very easy (especially when deprecating something) to introduce a lot of noisy test output with new warnings, eg: https://github.com/python/cpython/issues/127949#issuecomment-2558227488
I think it would be useful to either c... | 76856ae1659dbba066e51f353b31cb7cf0a8a5fe | 939df0f9f6a76e0916b3ba53841e1413ab90952e |
python/cpython | python__cpython-128764 | # Race on tp_version_tag under free threading
# Bug report
### Bug description:
TSAN reports a data race for the following program in a free-threaded build:
```python
import collections.abc
import concurrent.futures
import functools
import threading
num_threads = 8
def closure(b, x):
b.wait()
for _ in range(1... | 6e1e78054060ad326f26dd8dbf12adfedbb52883 | d0ecbdd838034c1f061e9b4e9b54a900141458c3 |
python/cpython | python__cpython-128735 | # ResourceWarning in urllib tests
Many `urllib` tests produce resource warnings. For example (there are much more cases):
```console
$ ./python -We -m test -vuall test_urllib test_urllib2 test_urllib2_localnet test_urllib_response test_urllibnet
...
test_geturl (test.test_urllib.urlopen_DataTests.test_geturl) ... ... | 5ace71713b03cb37d829f50c849a8bb8a518738d | cb72feb8a30edb448bc8dc31330ed7420279d7c2 |
python/cpython | python__cpython-128733 | # ResourceWarning in test_robotparser
# Bug report
```console
$ ./python -We -m test -vuall test_robotparser
...
test_read_404 (test.test_robotparser.NetworkTestCase.test_read_404) ... Warning -- Unraisable exception
Exception ignored in: <function _TemporaryFileCloser.__del__ at 0x7ff389434dd0>
Traceback (most recen... | 5e65a1acc0b630397f1d190aed279114e6e99612 | f6c61bf2d7d8b66ccd9f16e723546bdcc251a3d0 |
python/cpython | python__cpython-128730 | # RuntimeWarning in test_unittest
```console
$ ./python -We -m test -vuall test_unittest -m test_deprecation_of_return_val_from_test_async_method
...
test_deprecation_of_return_val_from_test_async_method (test.test_unittest.test_case.Test_TestCase.test_deprecation_of_return_val_from_test_async_method) ... Warning -... | cb72feb8a30edb448bc8dc31330ed7420279d7c2 | 30268b5d2fbb1a5e6c876f0cdc4cbdb5d93315e6 |
python/cpython | python__cpython-128721 | # Top-level code environment docs are outdated for Python 3.13+
# Bug report
### Bug description:
Ref: https://docs.python.org/3/library/__main__.html
There is a script and entry-point example which has a conditional upon `__main__.__file__` and also a remark like:
> Note that in this case the `__main__` s... | a4760ef8e5463116b3076e0f5e3c38b314f7b20f | 01d91500ca5b40bd84ef27fdf1f858a25c602bdb |
python/cpython | python__cpython-128741 | # Race between Py_SetRecursionLimit and _Py_LeaveRecursiveCallPy under free-threading
# Bug report
### Bug description:
Under Python 3.13 at commit 65da5db28a39cb0d9ee311206bdc8745b482c56f, we see the following TSAN race:
```python
import concurrent.futures
import functools
import sys
import threading
num_threads ... | f6c61bf2d7d8b66ccd9f16e723546bdcc251a3d0 | ff39e3ff7bebc9d700d89c5cd22145db2c879cf2 |
python/cpython | python__cpython-129016 | # Race in func_get_annotations under free-threading
# Bug report
### Bug description:
With CPython 3.13 from commit 65da5db28a39cb0d9ee311206bdc8745b482c56f TSAN reports the following race, which looks right: the initialization of `p->func_annotations` is racy.
Reproducer:
```python
import concurrent.futures
impor... | 55f17b77c305be877ac856d6426b13591cbc7fc8 | 63f0406d5ad25a55e49c3903b29407c50a17cfd5 |
python/cpython | python__cpython-128854 | # Windows 10 Mimetypes for .webp return empty
# Bug report
### Bug description:
on Windows 10 [`guess_file_type()`](https://github.com/python/cpython/blob/3.13/Lib/mimetypes.py#L151) checking `.webp` file return `tuple('', None)` not `tuple('image/webp', None)`
The problem is `db.read_windows_registry()` over... | 303043f5062c1e7ffb7907abde61dbf13c98f8e9 | 3402e133ef26736296c07992266a82b181a5d532 |
python/cpython | python__cpython-128701 | # sysconfig.get_platform(): specify result for Windows ARM64
https://docs.python.org/3/library/sysconfig.html#sysconfig.get_platform currently says for Windows:
> Windows will return one of:
> - win-amd64 (64bit Windows on AMD64, aka x86_64, Intel64, and EM64T)
> - win32 (all others - specifically, sys.platform... | 553cdc6d6856c1b4539a45eb90d0769f7c629355 | 1b39b502d33c68f52fd775c4e6c2174baddd40bd |
python/cpython | python__cpython-128702 | # `./python.exe -m test test_types test_inspect` fails as `(env changed)`
# Bug report
```
» ./python.exe -m test test_types test_inspect
Using random seed: 1107570136
0:00:00 load avg: 3.01 Run 2 tests sequentially in a single process
0:00:00 load avg: 3.01 [1/2] test_types
0:00:00 ... | 65b484db97b170af0f0e50655c4dcc6b9b1af8e9 | 2fcdc8488c32d18f4567f797094068a994777f16 |
python/cpython | python__cpython-128693 | # `threading.local()` access should scale well from multiple threads
Currently, accessing a `threading.local()` from multiple threads doesn't scale well because of reference count contention on the shared `_thread._local` object. We should use deferred reference counting on `_thread._local` to avoid this bottleneck.
... | c1417487e98e270d614965ed78ff9439044b65a6 | 087bb48acac997c06e69dae25bae2dd75194b980 |
python/cpython | python__cpython-128708 | # `LOAD_CONST_IMMORTAL` doesn't survive being instrumented and de-instrumented.
# Bug report
### Bug description:
When instrumenting, instructions are first de-optimized to their base instruction.
When instrumentation is removed we assume that specialization will restore the specialized instruction.
However, that do... | ddd959987c557beaf823b681bf5e5e573ad657ac | 29fe8072cf404b891dde9c1d415095edddbe19de |
python/cpython | python__cpython-128674 | # `typing.get_type_hints` is not fully covered
<img width="1338" alt="Снимок экрана 2025-01-09 в 15 23 03" src="https://github.com/user-attachments/assets/027336ba-bcbf-4156-b3bf-7b657fc74b63" />
I will send a PR with a new test.
<!-- gh-linked-prs -->
### Linked PRs
* gh-128674
<!-- /gh-linked-prs -->
| 43ac9f505903ba806aa6a5d93e6a67beb04bebc4 | ea39c8b08d8f025273bfa5b7a95f7b5984dc1e86 |
python/cpython | python__cpython-128663 | # `typing.evaluate_forward_ref` is not fully covered
Right now it is:
<img width="1341" alt="Снимок экрана 2025-01-09 в 11 36 03" src="https://github.com/user-attachments/assets/6a44ec6e-3012-46ed-9e90-cac2e6a6d767" />
I have a PR ready.
<!-- gh-linked-prs -->
### Linked PRs
* gh-128663
* gh-128930
<!-- /gh-linke... | b725297cee9e5608b709f3c7291d974c97f68fff | 43ac9f505903ba806aa6a5d93e6a67beb04bebc4 |
python/cpython | python__cpython-128886 | # Race in py_digest_by_name, _hashopenssl.c under free-threading
# Bug report
### Bug description:
I built cpython (3.13 branch) with free-threading and TSAN. The following python code from time to time reports TSAN warnings:
```python
import hashlib
import concurrent.futures
import threading
def _hash_string(hash... | c1f352bf0813803bb795b796c16040a5cd4115f2 | 5ce70ad129d2e34a09f8ae6ee0542f4f996fb8ec |
python/cpython | python__cpython-128651 | # incorrect statement in partial documentation
# Documentation
Now:
```rst
...; without :data:`!Placeholder`, only the
first positional argument can be pre-filled.
```
Should be:
```rst
...; without :data:`!Placeholder`, only the chosen number of
leading positional arguments can be pre-filled.
```
<... | 8af57810946c216b3e18c94c8f0ee3c0c96566a9 | 7dc41ad6a7826ffc675f088972de96624917696e |
python/cpython | python__cpython-128647 | # GzipFile.readinto reads full file before copying into the provided buffer
# Bug report
### Bug description:
gzip.GzipFile uses the BufferedIOBase implementation of `.readinto()`, which simply calls `.read` and copies the result into a buffer. This negates the purpose of using `.readinto()` at all.
This may be con... | 72e5b25efb580fb1f0fdfade516be90d90822164 | 78790811989ab47319e2ee725e0c435b3cdd21ab |
python/cpython | python__cpython-129596 | # Significant Configparser Performance Regression
# Bug report
### Bug description:
Hello, @jaraco! The following commit https://github.com/python/cpython/commit/019143fecbfc26e69800d28d2a9e3392a051780b slows down `ConfigParser.read()` from 2 to 6 times.
`_Line` is created many times when reading a file and t... | cd6abe27a2582786da7b4b8bb008910563f31735 | 071820113f11b8f6a21f98652d0840e10268114c |
python/cpython | python__cpython-128744 | # NULL ptr deref in _PyCode_ConstantKey when compiling code
# Crash report
### What happened?
Unfortunately it's a slightly large minimal reproducer. You can use `xxd -r` to go from the hexdump to the actual binary.
```
~/p/cpython ❯❯❯ xxd ~/Downloads/clusterfuzz-testcase-minimized-fuzz_pycompile-5092056728403968
... | 891c61c1fa480928dd60cce8bbc8764630c95025 | d1db43c139121202898e2d75df43ed2eb06a8470 |
python/cpython | python__cpython-128630 | # Add Py_PACK_VERSION macros
Add these to limited API:
* `Py_PACK_FULL_VERSION(x, y, z, level, serial)`: pack a version number from components
into the format used by `Py_VERSION_HEX` and `Py_LIMITED_API`.
For example, `Py_PACK_FULL_VERSION(3, 14, 0, 0xA, 1)` evaluates to 0x030E00A1.
* `Py_PACK_VERSION(x, ... | 1439b81928f1b52c5a0ac7fd81fdd66afd5f72da | 4685401845ba3e2ab8c9f4a9a10aa2969b11985f |
python/cpython | python__cpython-128628 | # Emscripten: Use wasm-gc for call trampolines instead of JS type reflection
Followup to: #121698. WebAssembly in the browser is soon gaining stack switching support, but we're not allowed to stack switch through JS frames. The stack switching support used to require the wasm-js-type-reflection proposal, so in #121698... | d0ecbdd838034c1f061e9b4e9b54a900141458c3 | 5e65a1acc0b630397f1d190aed279114e6e99612 |
python/cpython | python__cpython-128618 | # `test_typing.test_readonly_inheritance` does not work correctly
# Bug report
https://github.com/python/cpython/blob/cdfb8bc93a4d8c06d2404ba2d243937ba209438c/Lib/test/test_typing.py#L8904-L8921
Notice that it always tests `Child1` and never `Child2`
Refs https://github.com/python/cpython/commit/df4784b3b7519d137ca... | 971a52b5495e3d596e599faa1f31d4671897026d | 474e419792484d1c16e7d9c99b7bf144136b9307 |
python/cpython | python__cpython-128616 | # Pickling of `ParamSpecArgs` and `ParamSpecKwargs` is not covered
<img width="519" alt="Снимок экрана 2025-01-08 в 12 17 31" src="https://github.com/user-attachments/assets/8e7e8e45-0849-46ee-af3b-ea044f72364c" />
I also tried to remove these lines and executed (just to be sure):
```
./python.exe -m test tes... | 74a517181a9bb65a1f6da149af7427a9fcb3add3 | 95cd9c669cdc7718198addb1abb49941a2c61fae |
python/cpython | python__cpython-128614 | # `typing.Concatenate` is not fully covered
Coverage reports:
<img width="1324" alt="Снимок экрана 2025-01-08 в 11 59 22" src="https://github.com/user-attachments/assets/667eb627-c4ff-4106-be5c-ad6e4d10400a" />
I think that we need to cover these lines in tests.
PR is ready.
<!-- gh-linked-prs -->
### Linked PRs... | eb26e170695f15714b5e2ae0c0b83aa790c97869 | 971a52b5495e3d596e599faa1f31d4671897026d |
python/cpython | python__cpython-128498 | # Default to stdout isatty for colour detection instead of stderr
# Bug report
### Bug description:
The terminal colour detection defaults to checking `stderr`'s `isatty`, rather than `stdout`.
As suggested at https://github.com/python/cpython/issues/128317#issuecomment-2569566697, let's default to checking `stdout... | 6f167d71347de6717d9f6b64026e21f23d41ef0b | a42915979796565635dcfcd45d2b62b3e5e46eba |
python/cpython | python__cpython-129063 | # Cancellation leaks out of asyncio.TaskGroup on 3.12 when using eager tasks
# Bug report
### Bug description:
```python
import sys
import asyncio
if sys.platform == "win32":
EventLoop = asyncio.ProactorEventLoop
else:
EventLoop = asyncio.SelectorEventLoop
async def demo():
class MyError(Exception):
... | ed6934e71e55d398df8263f4697f58e4a3815f69 | ab61d3f4303d14a413bc9ae6557c730ffdf7579e |
python/cpython | python__cpython-128718 | # A new tail-calling interpreter for significantly better interpreter performance
# Feature or enhancement
## Proposal
Prior discussion at: https://github.com/faster-cpython/ideas/issues/642
I propose adding a tail-calling interpreter to CPython for significantly better performance on compilers that support it.
Th... | cb640b659e14cb0a05767054f95a9d25787b472d | 555dc50c811e3e9ebdc30a1d511cf48a32666d6f |
python/cpython | python__cpython-128604 | # tkinter widget instance name sometimes duplicated on inherited class
# Bug report
### Bug description:
If you run de following code:
```python
import tkinter as tk
class Label2(tk.Label):
def __init__(self, top, text):
tk.Label.__init__(self, top, text=text, bg='Yellow')
w = tk.Tk()
... | da8825ea95a7096bb4f933d33b212a94ade10f6e | 402b91da87052878b4e7e8946ba91bdf4ee4bebe |
python/cpython | python__cpython-128560 | # Asyncio should not import typing at runtime
# Feature or enhancement
### Proposal:
Currently, asyncio.timeout imports final, Type, and Optional from typing.
final has no impact on type checkers, as they use the typeshed for the standard library, and type and optional can be written without importing from typing. ... | 7363476b6405e3d288a61282aa7bc6aca9c2114d | 61c1a2478e6da8dc6dbdce4d6ac66b03d5963710 |
python/cpython | python__cpython-128553 | # TaskGroup's task's exception's participate in a reference cycle, via TaskGroup.create_task -> task -> exception and loop.create_task -> task -> exception
# Bug report
### Bug description:
the following test fails when run with eager tasks
```python
async def test_exception_refcycles_parent_task(self):
... | 61b9811ac6843e22b5896ef96030d421b79cd892 | 6ea04da27036eaa69d65150148bb8c537d9beacf |
python/cpython | python__cpython-130569 | # Find a way to handle branch events in `async for`
When instrumenting regular `for` loops, we instrument the `FOR_ITER` instruction as the `LEFT` branch and the `POP_ITER` instruction as the `RIGHT` branch.
This doesn't work for `async for` as it uses exceptions to exit the loop.
Raising an exception is the docum... | 2a18e80695ac1f05c95ea3b1cfe29defd45e2f71 | fda056e64bdfcac3dd3d13eebda0a24994d83cb8 |
python/cpython | python__cpython-128554 | # `NOT_TAKEN` instructions need to be added after optimization
The `NOT_TAKEN` instruction is added during code gen, but can be left when jumps are removed by the optimizer.
We should probably add the `NOT_TAKEN` instructions in the assembler after optimization, as suggested in https://github.com/python/cpython/pul... | 2434fd2d50b8b770585ad5949a664e4bbab4bde1 | b6c919b674549d519efbbc378588ca6d3efc7242 |
python/cpython | python__cpython-128521 | # Update the docstring of `untokenize()` to match the docs
# Documentation
In https://github.com/python/cpython/issues/128031, it was deemed too risky to change `untokenize()` to more closely match the original input string. We should update the docstring of `untokenize` to reflect that and prevent confusion in the... | aef52ca8b334ff90e8032da39f4d06e7b5130eb9 | a62ba52f1439c1f878a3ff9b8544caf9aeef9b90 |
python/cpython | python__cpython-128845 | # Test BOLT builds in CI
# Feature or enhancement
As part of https://github.com/python/cpython/issues/101525, we should run BOLT optimizations in CI.
It looks like optimizations aren't tested in CI right now, just on the build bots? I could use some guidance on the best place for this.
<!-- gh-linked-prs -->
### Li... | 9ed7bf2cd783b78b28f18abc090f43467a28f4aa | 3829104ab412a47bf3f36b8c133c886d2cc9a6d4 |
python/cpython | python__cpython-128510 | # Add an API for determining if an object is immortal
# Feature or enhancement
### Proposal:
Immortal objects might lead to some unexpected results for users that are (unfortunately) messing with reference count details, as seen in #127191; but currently, there's no way to determine what objects are actually immorta... | 9ba281d871c4df3a3ac4cb7896d24ba0d42751a3 | 60a85415aeb5a8be54b3c412d19a7444bf5ac757 |
python/cpython | python__cpython-128506 | # Upgrade doctest CI to Ubuntu-24.04
# Feature or enhancement
### Proposal:
As title, since most of CI run in `ubuntu-24.04`, and Doctest CI was pinned to `ubuntu-22.04` according to #125236. Maybe we can upgrade it now.
### Has this already been discussed elsewhere?
No response given
### Links to previous disc... | 2228e92da31ca7344a163498f848973a1b356597 | ae23a012e6c8aadc4a588101cbe7bc86ede45627 |
python/cpython | python__cpython-131633 | # Race in _ctypes_alloc_callback and libffi under free-threading
# Bug report
### Bug description:
I built libffi with TSAN and cpython (3.13 branch) with free-threading and TSAN. The following python code reports TSAN warnings:
```python
import ctypes
import concurrent.futures
import threading
def te... | 04d4aacaacbe26146e9b5ff6b0043f86f3c543d4 | 9c4fb92e126000650a8efb4d27c4e3d1af82f71e |
python/cpython | python__cpython-128484 | # Missing documentation for `traceback.FrameSummary.end_{col,line}no` attributes
These attributes were added in Python 3.11, as part of [PEP-657](https://peps.python.org/pep-0657/), and I believe they're intended to be public API. If so, adding them to the `__init__` docstring would be nice!
(prompted by checking ... | 051f0e5683fec3840fa7fc99723741dd2d701eae | 37145cb89fe806377a5e9ed1fdac92dd3a5df2c0 |
python/cpython | python__cpython-128475 | # asyncio.staggered.staggered_race can log NameError: cannot access free variable 'exceptions' where it is not associated with a value in enclosing scope
# Bug report
### Bug description:
https://github.com/Chia-Network/chia-blockchain/actions/runs/12586550910/job/35084132471#step:16:1813
```
______ test_long_reorg... | ec91e1c2762412f1408b0dfb5d281873b852affe | cf0b2da1e6947aa15be119582c2017765ab46863 |
python/cpython | python__cpython-128511 | # BOLT fails with `--enabled-shared` on LLVM 19 with "unable to get new address"
# Bug report
### Bug description:
As originally reported in https://github.com/astral-sh/python-build-standalone/pull/463
> LLVM 19.x has a hard error when handling PIC compiled functions containing computed gotos. It appears pri... | 24b147a19b360c49cb1740aa46211d342aaa071f | 953b49e5468d02afaddadc2307f4763422078603 |
python/cpython | python__cpython-128457 | # BOLT uses deprecated option `-reorder-functions=hfsort+`
# Bug report
### Bug description:
Building with BOLT optimizations emits the following warning
```
BOLT-WARNING: '-reorder-functions=hfsort+' is deprecated, please use '-reorder-functions=cdsort' instead
```
From the usage at https://github.com/python/cpyt... | 4974bbdb290b61a2d8860de490fde1228c296753 | befcfdfdaba15ecae38739ecabebd8046c1b1977 |
python/cpython | python__cpython-129478 | # New warning in the main (b4f799b1e7): ./Modules/socketmodule.c
# Bug report
### Bug description:
./configure -q && make -s
give me
```
[...]
./Modules/socketmodule.c: In function ‘sock_accept’:
./Modules/socketmodule.c:474:23: warning: ‘ctx.result’ may be used uninitialized [-Wmaybe-uninitialized]
474 | # defin... | c07ac3c86a8f5021e27bf2c27c6bf0a25229d846 | 10ee2d9d3bcde27c75f179214ad41c00e4852a7a |
python/cpython | python__cpython-128450 | # Skip Windows CI for `configure`/`Makefile` changes
# Feature or enhancement
There should be no need to run the Windows CI for \*nix build system changes. Suggesting we skip the Windows CI for `configure`/Make related changes.
<!-- gh-linked-prs -->
### Linked PRs
* gh-128450
* gh-130434
* gh-130435
* gh-131702
* g... | b05fa90b21dd01bb836285cdd41920320b09e681 | 06ac157c53046f4fcad34383ef131f773085f3d5 |
python/cpython | python__cpython-128455 | # Bolt instrumentation missing `-update-debug-sections`
# Bug report
### Bug description:
The following warning is displayed during BOLT instrumentation
```
BOLT-WARNING: debug info will be stripped from the binary. Use -update-debug-sections to keep it.
```
The flag is provided during "apply"
https://github.com/... | b60044b838f3ea97395cd6f3adbd5330356fc273 | 0cafa97932c6574734bbaa07180bbd5a762b01a6 |
python/cpython | python__cpython-128429 | # Support Nil and Max UUID formats from RFC 9562
# Feature or enhancement
### Proposal:
RFC 9562 defines Nil and Max UUID formats:
- https://www.rfc-editor.org/rfc/rfc9562.html#name-nil-uuid
- https://www.rfc-editor.org/rfc/rfc9562.html#name-max-uuid
I've often had a need to use these as placeholders, sentinels or... | 8ec76d90340287eb3587f0ae388bbfe158fb28d8 | d8e16ef7037ac254e4799d2991c7fc3fe576c545 |
python/cpython | python__cpython-128428 | # Free-threading C API HOWTO should mention PySequenceFast API
Right now [this page](https://docs.python.org/3/howto/free-threading-extensions.html#general-api-guidelines) says that macros that access the internal state of data structures are thread-unsafe, but because `PySequenceFast` is a function I completely misse... | e7adeecc2b318505eb53bc779320f028be40cccc | 8eebe4e6d02bb4ad3f1ca6c52624186903dce893 |
python/cpython | python__cpython-128728 | # data races in frame inspection and tracebacks
Run the tests with https://github.com/python/cpython/pull/128147 and tsan enabled:
```console
env TSAN_OPTIONS=suppressions={$PWD}/Tools/tsan/suppressions_free_threading.txt ./python -m test test_asyncio -F
```
TSAN Warnings:
<details>
```
WARNING: ThreadSani... | 75214f87f1ddd1d7b9a5b663a9c688b1bd41c098 | bf64a582f00a030fee11b7f89c6a02ea967990ca |
python/cpython | python__cpython-128410 | # Document `pdb.post_mortem()` accepts exceptions on Python 3.13+
# Documentation
Some change in Python 3.13 (maybe 5f3433f210d25d366fcebfb40adf444c8f46cd59 or related) allows `pdb.post_mortem()` to accept an exception object, as well as a traceback object. This could be useful to use for chained exception access, ... | 1b39b502d33c68f52fd775c4e6c2174baddd40bd | 802556abfa008abe0bdd78e6f9e18bef71db90c1 |
python/cpython | python__cpython-128422 | # Crash when using _Py_DumpTracebackThreads
# Crash report
### What happened?
```python
# triggering program
# based on test.test_faulthandler.FaultHandlerTests.test_dump_traceback_threads
import faulthandler
from threading import Thread, Event
class Waiter(Thread):
def __init__(self):
Thread.__init_... | c9356feef28e6dfc4dc32830d3427a5ae0e426e2 | a626f9a67b76e5fe69677afd5f8317d8c61de8de |
python/cpython | python__cpython-132218 | # Suggest `async with` when `with` finds no `__enter__`/`__exit__`
# Bug report
### Bug description:
(This is *not* an asyncio bug! I am just using `asyncio.TaskGroup()` as an example.)
```python
import asyncio
def foo():
with asyncio.TaskGroup() as g: # BUG: should be `async with`
pass
```
This curren... | 8a9c6c4d16a746eea1e000d6701d1c274c1f331b | 95800fe6e719c829acf52fbb00198135b78719b4 |
python/cpython | python__cpython-130235 | # PEP 667 + PEP 709 segfaults from accessing closure variables bound by any inlined comprehensions
# Bug report
### Bug description:
The segfault described here is related to PEP-667 and PEP-709.
I think I experienced a couple of similar but not same segfaults, and this here is just one of a few of them. I mi... | ccf17323c218a2fdcf7f4845d3eaa74ebddefa44 | 5f00501a940a0fb97870e70066fb301909320388 |
python/cpython | python__cpython-128389 | # `pyrepl` on Windows: add Ctrl+← and Ctrl+→ word-skipping and other keybindings
# Feature or enhancement
### Proposal:
Currently, `_pyrepl/windows_console.py` is very limited compared to `_pyrepl/unix_console.py`, for example, it doesn't support the `Ctrl`+`←` and `Ctrl`+`→` word-skipping keybindings (see also #119... | 688f3a0d4b94874ff6d72af3baafd8bbf911153e | baf65715fc9002e43cd0e1010b8dba9b4c84d503 |
python/cpython | python__cpython-128378 | # `test_cmd_line.test_non_interactive_output_buffering` fails when `PYTHONUNBUFFERED` is enabled
# Bug report
### Bug description:
Reproducer:
```bash
➜ cpython git:(main) ✗ export PYTHONUNBUFFERED=1
➜ cpython git:(main) ✗ python3 -m test test_cmd_line
Using random seed: 2481773783
Raised RLIMIT_NOFILE: ... | 30268b5d2fbb1a5e6c876f0cdc4cbdb5d93315e6 | 22a442181d5f1ac496dac08e24fd852de404882f |
python/cpython | python__cpython-128445 | # Performance of instrumented `FOR_ITER` is poor as `DISABLE` doesn't remove the instrumentation.
Disabling the "not taken" branch of an instrumented `FOR_ITER` does not de-instrument the `FOR_ITER`, it only de-instruments the `NOT_TAKEN` which prevents specialization and JIT compilation of the `FOR_ITER`. It is only ... | f826beca0cedb8e4b92896544c75fd0d9dcb0446 | b9c693dcca01537eee1ef716ffebc632be37594b |
python/cpython | python__cpython-130724 | # `test_concurrent_futures.test_wait.ThreadPoolWaitTests.test_timeout` flakes frequently on free-threaded Windows x64
# Bug report
### Bug description:
The following test fails frequently, but not consistently:
```
test_timeout (test.test_concurrent_futures.test_wait.ThreadPoolWaitTests.test_timeout) ... FAI... | 5221d9ce0e2beb0fe04bed072e1bb448fd522882 | cfa0b1dc375e63cde28e61a47108c645b0e74834 |
python/cpython | python__cpython-128361 | # Migrate from `PyGILState_Check` to thread state checks internally
# Feature or enhancement
### Proposal:
In some older areas of the codebase, we have the following to ensure that the caller holds the GIL (or really, an [attached thread state](https://github.com/python/cpython/pull/127990)):
```c
assert(PyGILState... | 4d0a6595a06c554c57ebd90ee64ff4c2bec239b8 | c6b570e5e3b214d2038645c5fa7806e0fb3f7dcd |
python/cpython | python__cpython-128359 | # `LDFLAGS` and `LIBS` not used consistently in `configure`
# Bug report
### Bug description:
Following up on #128322, I noticed that the same pattern of setting `LDFLAGS` instead of `LIBS` is used frequently. Additionally, there is not consistency in the ordering of the added libraries, e.g., `LIBS="$LIBS ..."` vs ... | b75ed951d4de8ba85349d80c8e7f097b3cd6052f | 4974bbdb290b61a2d8860de490fde1228c296753 |
python/cpython | python__cpython-128379 | # Docs: decimal.BasicContext etc. are not classes
# Documentation
The `decimal` module exports three constants `BasicContext`, `ExtendedContext` and `DefaultContext`, which are instances of `Context`:
```
$ python3
Python 3.13.1 (main, Dec 20 2024, 10:15:12) [Clang 14.0.0 (clang-1400.0.29.202)] on darwin
Type... | 60c65184695a3eab766b3bc26fc99f695deb998f | c810ed7c8e0a7464d19700ba1c8668a406f1c042 |
python/cpython | python__cpython-128344 | # GIL is not properly disabled for _freeze_module.vcxproj
# Bug report
### Bug description:
The _freeze_module.vcxproj file is looking for `#undef Py_GIL_DISABLED`, but that was changed to `/* #define Py_GIL_DISABLED 1 */` some time ago, so the build system ends up looking to replace a line that doesn't exist, t... | c26bed1160978fe8b1844878b8123778e47870c6 | a13460ac4427892be613bf0e2988f5a2b2364602 |
python/cpython | python__cpython-128343 | # Timeout unit isn't specified in subprocess docstrings
# Documentation
Timeout unit isn't specified in subprocess docstrings. For example here:
```
def run(*popenargs,
input=None, capture_output=False, timeout=None, check=False, **kwargs):
"""Run command with arguments and return a CompletedProcess... | dafe7a44630aa32bb411cceb45c7b7df725e3fe3 | 47d2cb8eb7595df5940225867dbb66b6dd59413a |
python/cpython | python__cpython-128331 | # Terminal settings are not correctly restored on repl exit
# Bug report
### Bug description:
I am working on updating the version of Python that ships in [OmniOS](https://omnios.org) -- an illumos distribution -- from 3.12 to 3.13 and have encountered a problem with terminal settings not being properly restored... | 0b15d9c0d2d30c7d3f17ebb90dd822ef32f977cc | ec91e1c2762412f1408b0dfb5d281873b852affe |
python/cpython | python__cpython-128322 | # Statically linked `sqlite3` fails checks due to missing `-lm`
# Bug report
### Bug description:
On Linux, if I build a static version of sqlite3 (e.g., `./configure --disable-shared && make` from a sqlite3 tarball)
You can see that `./configure` for CPython fails the SQLite checks (added back in https://git... | 81376fef76a53fb79893bfa9c9db18d97c228fbe | ef63cca494571f50906baae1d176469a3dcf8838 |
python/cpython | python__cpython-128318 | # Highlight today in colour in calendar's CLI output
# Feature or enhancement
In Python 3.13 we added colour output to the new REPL, tracebacks and doctest, and in 3.14 to unittest and `test.regrtest`, which can also be controlled with the `PYTHON_COLORS`, `NO_COLOR` and `FORCE_COLOR` environment variables:
* https:... | f21af186bf21c1c554209ac67d78d3cf99f7d7c0 | fa985bee6189aabac1c329f2de32aa9a4e88e550 |
python/cpython | python__cpython-128768 | # asyncio eager tasks' names are set too late
# Bug report
### Bug description:
consider:
```python
import asyncio
async def main():
names = []
async with asyncio.TaskGroup() as tg:
async def append_name():
names.append(asyncio.current_task().get_name())
tg.create_task(append_n... | 38a99568763604ccec5d5027f0658100ad76876f | 6c914bf85cc3a07e22e4618d0b5607c158e3cf66 |
python/cpython | python__cpython-128306 | # add eager_start parameter to loop.create_task
# Feature or enhancement
### Proposal:
We want to be able to opt in and opt out of eager tasks anywhere in the asyncio call stack, eg in some asyncio library, and not rely on some other asyncio framework setting the default.
### Has this already been discussed e... | 08d7687094c6acb8c2ea1925a292a94ce1246c82 | c4cc5d58aee6a3be55a95efee6ec25d5774f7b5f |
python/cpython | python__cpython-128284 | # Bug in `xml.dom.xmlbuilder.DOMBuilder.parse()`
# Bug report
### Bug description:
I believe there is a bug in `xml.dom.xmlbuilder.parse()`.
`xml.dom.xmlbuilder` is not documented by python, we don't have any tests for
them, and I can't find evidence on github of these classes being in use. That's
okay, bec... | 6ea04da27036eaa69d65150148bb8c537d9beacf | 145276a072ae058bac42ee43a4235cd4eda2726b |
python/cpython | python__cpython-128280 | # Build failure on NetBSD: Incorrect number of arguments to 'pthread_setname_np' in _threadmodule.c
# Bug report
### Bug description:
```sh
./configure --with-pydebug && make -j3
```
Output:
```sh
gcc -pthread -fno-strict-overflow -Wsign-compare -g -Og -Wall -O2 -std=c11 -Wextra -Wno-unused-parameter -Wno-missing... | 492b224b991cd9027f1bc6d9988d01e94f764992 | 2cf396c368a188e9142843e566ce6d8e6eb08999 |
python/cpython | python__cpython-128286 | # Make `socket` module thread safe
While working on https://github.com/python/cpython/issues/128002, I am seeing many tsan warning because of socket modules.
This issue tracks work for making it thread safe.
<!-- gh-linked-prs -->
### Linked PRs
* gh-128286
* gh-128304
* gh-128305
<!-- /gh-linked-prs -->
| e389d6c650ddacb55b08b657f1e4e9b1330f3455 | b2ac70a62ad1be8e037ce45ccf5f1b753ea5e64b |
python/cpython | python__cpython-128264 | # Remove un-necessary socket requirements in pdb async tests
### Proposal:
the pdb async tests should run coroutines using coro.send(None) because they don't need an async framework to run. This would remove asyncio from the tests and support running on emscripten/wasi
### Has this already been discussed elsewher... | aab51c3414bc815c4c31e8ef2a9003f4a546faa9 | aeb9b65aa26444529e4adc7d6e5b0d3dd9889ec2 |
python/cpython | python__cpython-128263 | # Classes with __slots__ ought to be specialized when constructing instances
# Feature or enhancement
### Proposal:
Right now simple classes have the `CALL_ALLOC_AND_ENTER_INIT` specialization when calling their constructor. This however, blocks classes with `__slots__`. I propose relaxing the requirement and allowi... | 7ef49074123511003c8b7f7f3ba2a4e05285e8dc | dafe7a44630aa32bb411cceb45c7b7df725e3fe3 |
python/cpython | python__cpython-95888 | # Remove un-necessary socket requirements in `contextlib` async tests
### Proposal:
the contextlib async tests should run coroutines using coro.send(None) because they don't need an async framework to run. This would remove asyncio from the tests and support running on emscripten/wasi
<!-- gh-linked-prs -->
### Li... | 81636d3bbd7f126692326bf957707e8a88c91739 | 76f1785657fde0ebee082b0df54da8f7c911c369 |
python/cpython | python__cpython-129488 | # In the new REPL, multiline input execution should fail fast
# Bug report
### Bug description:
To reproduce, simply paste this to the new REPL:
```py
raise Exception
print("But I'm powerful enough to bypass it")
```
Make sure what you are seeing is:
```pycon
>>> raise Exception
... print("But I'm ... | 7ed3dc6392613832f66c63507385b1da109cbf21 | 9f25c1f012c8d432a93bf2dcad5f19a64dc00d3c |
python/cpython | python__cpython-128228 | # `Doc/requirements-oldest-sphinx.txt` needs to be regenerated
# Bug report
### Bug description:
Due to https://github.com/python/cpython/security/dependabot/12 and https://github.com/python/cpython/security/dependabot/11, the dependency file must be regenerated (it's fine if those links go nowhere, only triager... | 7985d460c731b2c48419a33fc1820f9512bb6f21 | 7ed6c5c6961d0849f163d4d449fb36bae312b6bc |
python/cpython | python__cpython-128367 | # Race between ` PyUnicode_SET_UTF8` and `_PyUnicode_CheckConsistency`
# Bug report
### Bug description:
Reproducible by running :
1. `env CC=clang CXX=clang++ ./configure --disable-gil --with-thread-sanitizer --with-pydebug && make -j`
2. `env TSAN_OPTIONS=suppressions=${PWD}/Tools/tsan/suppressions_free_threadin... | 8eebe4e6d02bb4ad3f1ca6c52624186903dce893 | c9356feef28e6dfc4dc32830d3427a5ae0e426e2 |
python/cpython | python__cpython-128202 | # ``test_pdb`` raises a ``DeprecationWarning``
# Bug report
### Bug description:
```python
eclips4@nixos ~/p/p/cpython (main)> ./python -m test -v test_pdb -m test_pdb_next_command_for_coroutine
== CPython 3.14.0a3+ (heads/main:180d417e9f9, Jan 1 1980, 00:00:00) [GCC 13.3.0]
== Linux-6.6.63-x86_64-with-glibc2.40 lit... | d9ed42bc00c74b3150be2a0eb28da03e01dffcc7 | 418114c139666f33abff937e40ccbbbdce15bc39 |
python/cpython | python__cpython-128199 | # Add missing error checks for usages of `PyIter_Next`
# Bug report
### Bug description:
Not all usages of `PyIter_Next` in the codebase comply with specified behavior: https://github.com/python/cpython/blob/180d417e9f9456defd4c5b53cae678c318287921/Objects/abstract.c#L2905-L2919.
One example would be: https://github... | 5c814c83cdd3dc42bd9682106ffb7ade7ce6b5b3 | 81636d3bbd7f126692326bf957707e8a88c91739 |
python/cpython | python__cpython-128203 | # Add `_REPLACE_WITH_TRUE` to the tier 2 optimizer
# Feature or enhancement
### Proposal:
`_REPLACE_WITH_TRUE` is not currently in the tier 2 optimizer's known list, resulting in some missed opportunities.
To add it, here are the following steps:
1. Add a new bytecode definition `_REPLACE_WITH_TRUE` in `Python/opti... | 30efede33ca1fe32debbae93cc40b0e7e0b133b3 | d61542b5ff1fe64705e5ce1bcc53048f14098dba |
python/cpython | python__cpython-128193 | # Support HTTP digest authentication algorithm SHA-256 as per RFC-7617
# Feature or enhancement
### Proposal:
Support SHA-256 digest authentication following RFC7616 - https://datatracker.ietf.org/doc/html/rfc7616
CPython currently supports MD5 and SHA1 only.
Other possible methods include `SHA-256-ness`, `SHA-512... | f9a5a3a3ef34e63dc197156e9a5f57842859ca04 | 492b224b991cd9027f1bc6d9988d01e94f764992 |
python/cpython | python__cpython-128490 | # `ctypes` pointer writes are not thread safe
# Bug report
### Bug description:
Part of #127945.
`ctypes` C data objects have an internal pointer for what they're looking at (`b_ptr`). This field itself is generally fine to read non-atomically, because ctypes objects don't tend to overwrite the pointer that t... | 8dfc743a1dae499d3b116b81ef6aad78708c45bc | 5044c2245c1519e117ba3977a4c8f608785326e7 |
python/cpython | python__cpython-128680 | # Use of `ffi_type_complex_double` is unguarded
# Bug report
### Bug description:
e.g.
```
-std=c11 -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -Wstrict-prototypes -Werror=implicit-function-declaration -fvisibility=hidden -Werror=unguarded-availability-new -I./Include/internal -I./Include/inte... | d3b1bb228c951f8245f36ee28d9b175786522730 | 7239da75592081b6e8d0917a2cd2bf19907c8165 |
python/cpython | python__cpython-128151 | # Improve performances of `uuid.*` functions
# Feature or enhancement
The dedicated UUID constructors (e.g., `uuid.uuid4()`) generate bytes and pass them to the `UUID` constructor. However, the latter performs multiple and redundant checks. We can by-pass those checks since we are actually creating manually the UUI... | 6ff8f82f92a8af363b2bdd8bbaba5845eef430fc | 39fc7ef4fe211e8f7d3b5a6e392e475ecdfbce72 |
python/cpython | python__cpython-128165 | # 3.14 build error after recent commit: `pylifecycle.c:50:12: fatal error: os/log.h: No such file or directory`
# Bug report
### Bug description:
The recent commit https://github.com/python/cpython/commit/2041a95e68ebf6d13f867e214ada28affa830669 has broken the build of Python for a number of macOS versions, since `<... | e837a1f71e832ce8f551a6fac05e346f654457e0 | 24b147a19b360c49cb1740aa46211d342aaa071f |
python/cpython | python__cpython-128196 | # Race in PyUnicode_InternFromString under free-threading
# Bug report
### Bug description:
Here's a race reported by thread sanitizer that I haven't been able to find a small reproducer for, but it does look racy reading the code.
```
WARNING: ThreadSanitizer: data race (pid=1575489)
Read of size 4 at 0x7fb14614... | ae23a012e6c8aadc4a588101cbe7bc86ede45627 | b60044b838f3ea97395cd6f3adbd5330356fc273 |
python/cpython | python__cpython-128412 | # Race in bytes_hash under free-threading
# Bug report
### Bug description:
Run the following reproducer under Python 3.13.1t with free-threading enabled:
```python
import concurrent.futures
import functools
import threading
num_threads = 8
def closure(b, x):
b.wait()
x.__hash__()
with concurrent.futures.Thre... | 0706bab1c0985761cdbc07ab448f98c717276b36 | b49c68a1b3dfd2c2567c38b2d044c4a1c14a26a7 |
python/cpython | python__cpython-129975 | # Race on `_PyRuntime.signals.unhandled_keyboard_interrupt` when calling `eval()` concurrently in free-threading mode
# Bug report
### Bug description:
Run the following code under Python 3.13.1t with thread-sanitizer enabled:
```python
from collections import namedtuple
import concurrent.futures
import functools
i... | 451f291baaff918228ace4e8257be42737d7654a | aa284232013693103431fb672a0bcabac3d4200b |
python/cpython | python__cpython-128119 | # Improve performance of copy.copy
# Feature or enhancement
### Proposal:
Similar to the approached used for `copy.deepcopy` in #114266 we can simplifly the implementation of `copy.copy` and improve performance by checking on the type of the argument using a lookup.
### Has this already been discussed elsewhere?
... | 34b85ef26c44bbbd8a05d3e7dee6997b3623aab0 | 348012432155271cfbee71a78d0c27254fc230e2 |
python/cpython | python__cpython-128120 | # [PPC64LE buildbots] test_socket: VSOCK testStream(): recv fails with EACCES on Fedora 41
Example of failure: https://buildbot.python.org/#/builders/1467/builds/303/steps/6/logs/stdio
```
Traceback (most recent call last):
File ".../Lib/test/test_socket.py", line 550, in testStream
msg = self.conn.recv(10... | cbfe3023e46b544b80ea1a38a8c900c6fb881554 | 0974d7bb866062ed4aaa40f705d6cc4c294d99f1 |
python/cpython | python__cpython-128106 | # `test_strftime_y2k` fails while cross-compiling 3.14a3 for `x86_64_v2` and `x86_64_v3` on Linux
# Bug report
### Bug description:
The PGO test suite fails on the `test_datetime` tests; in particular, the `test_strftime_y2k` test cases. It looks like the padding is wrong:
```
cpython-3.14> AssertionError: '... | bb2dfadb9221fa3035fda42a2c153c831013e3d3 | bb73426cafb78691b249ffa50f1872ab6f899d4a |
python/cpython | python__cpython-128297 | # Race between _PyObject_GenericGetAttrWithDict and ensure_nonmanaged_dict under free-threading
# Bug report
### Bug description:
Under Python 3.13.1t build with thread-sanitizer:
```python
import concurrent.futures
import functools
import threading
num_threads = 100
def closure(b, o):
b.wait()
getattr(o, "fo... | 47d2cb8eb7595df5940225867dbb66b6dd59413a | fe4dd07a84ba423179a93ed84bdcc2b4c99b35a9 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.