* Do not PUSH/POP traceback or type to the stack as part of exc_info
* Remove exc_traceback and exc_type from _PyErr_StackItem
* Add to what's new, because this change breaks things like Cython
* bpo-44893: Implement EntryPoint as simple class and deprecate tuple access in favor of attribute access. Syncs with importlib_metadata 4.8.1.
* Apply refactorings found in importlib_metadata 4.8.2.
Functions signal(), getsignal(), pthread_sigmask(), sigpending(),
sigwait() and valid_signals() were omitted.
If __all__ is not defined all non-builtin functions should have
correct __module__.
This parameter was accidentally removed when fixing
https://bugs.python.org/issue45129, this reverts the unnecessary changes
there.
Co-authored-by: Andrew Svetlov <andrew.svetlov@gmail.com>
`@dataclass` in 3.10 prohibits using list, dict, or set as default values. It does this to avoid the mutable default problem. This test is both too strict, and not strict enough. Too strict, because some immutable subclasses should be safe, and not strict enough, because other mutable types should be prohibited. With this change applied, `@dataclass` now uses unhashability as a proxy for mutability: if objects aren't hashable, they're assumed to be mutable.
The line numbers of actually calling the decorator functions of
functions and classes was wrong (as opposed to loading them, were they
have been correct previously too).
Co-authored-by: Łukasz Langa <lukasz@langa.pl>
This defines VPATH differently in PGO instrumentation builds, to account for a different default output directory. It also adds sys._vpath on Windows to make the value available to sysconfig so that it can be used in tests.
When Python is embedded in other applications, it is not easy to determine which version of Python is being used. This change exposes the Python version as part of the API data. Tools like Austin (https://github.com/P403n1x87/austin) can benefit from this data when targeting applications like uWSGI, as the Python version can then be inferred systematically by looking at the exported symbols rather than relying on unreliable pattern matching or other hacks (like remote code execution etc...).
Automerge-Triggered-By: GH:pablogsal
* Make generator, coroutine and async gen structs all the same size.
* Store interpreter frame in generator (and coroutine). Reduces the number of allocations neeeded for a generator from two to one.
The getpath.py file is frozen at build time and executed as code over a namespace. It is never imported, nor is it meant to be importable or reusable. However, it should be easier to read, modify, and patch than the previous code.
This commit attempts to preserve every previously tested quirk, but these may be changed in the future to better align platforms.
Modify the ``EnumType.__dir__()`` and ``Enum.__dir__()`` to ensure
that user-defined methods and methods inherited from mixin classes always
show up in the output of `help()`. This change also makes it easier for
IDEs to provide auto-completion.
Rename PyConfig.no_debug_ranges to PyConfig.code_debug_ranges and
invert the value.
Document -X no_debug_ranges and PYTHONNODEBUGRANGES env var in
PyConfig.code_debug_ranges documentation.
* Inlined code from variance functions
* Added helper functions for the float square root of a fraction
* Call helper functions
* Add blurb
* Fix over-specified test
* Add a test for the _sqrt_frac() helper function
* Increase the tested range
* Add type hints to the internal function.
* Fix test for correct rounding
* Simplify ⌊√(n/m)⌋ calculation
Co-authored-by: Mark Dickinson <dickinsm@gmail.com>
* Add comment and beef-up tests
* Test for zero denominator
* Add algorithmic references
* Add test for the _isqrt_frac_rto() helper function.
* Compute the 109 instead of hard-wiring it
* Stronger test for _isqrt_frac_rto()
* Bigger range
* Bigger range
* Replace float() call with int/int division to be parallel with the other code path.
* Factor out division. Update proof link. Remove internal type declaration
Co-authored-by: Mark Dickinson <dickinsm@gmail.com>
* [bpo-43137](): webbrowser: Prefer gio open over gvfs-open
gvfs-open(1) was superseded by gio(1) in 2015, and removed from GNOME
releases in 2018. Debian and its derivatives like Ubuntu currently still
have a compatibility shim for gvfs-open, but we plan to remove it.
webbrowser prefers xdg-settings and xdg-open over gvfs-open, so this
will only have any practical effect on systems where the xdg-utils
package is not installed.
Note that we don't check for GNOME_DESKTOP_SESSION_ID before using gio.
gio does the right thing on any desktop environment that follows
freedesktop.org specifications, similar to xdg-settings, so it's
unnecessary to guard in this way. GNOME_DESKTOP_SESSION_ID was deprecated
in 2008 and removed from upstream gnome-session in 2018 (it's still
present in Debian/Ubuntu for backward compatibility, but probably
shouldn't be). The replacement way to detect a desktop environment is
the XDG_CURRENT_DESKTOP environment variable, which is a colon-separated
sequence where the first item is the current desktop environment and the
second and subsequent items (if present) are other desktop environments
that it resembles or is based on.
Resolves:
* [bpo-43137](): webbrowser: Never invoke gnome-open
gnome-open was part of GNOME 2, which was superseded in around 2010 and
is unmaintained. The replacement was gvfs-open, which was subsequently
replaced by gio(1) (as used in the previous commit).
* [bpo-43137](): webbrowser: Don't run gvfs-open on GNOME
gvfs-open was deprecated in 2015 and removed in 2018. The replacement
is gio(1) (as used in a previous commit).
GNOME_DESKTOP_SESSION_ID was deprecated in 2008 and removed in 2018.
The replacement is XDG_CURRENT_DESKTOP (as mentioned in a previous
commit).
---
To test this on a typical modern Linux system, it is necessary to disable the `xdg-settings` and `xdg-open` code paths, for example with this hack:
<details><summary>Hack to disable use of xdg-settings and xdg-open</summary>
```diff
diff --git a/Lib/webbrowser.py b/Lib/webbrowser.py
index 3244f206aa..8f6c09d1d2 100755
--- a/Lib/webbrowser.py
+++ b/Lib/webbrowser.py
@@ -459,7 +459,7 @@ def open(self, url, new=0, autoraise=True):
def register_X_browsers():
# use xdg-open if around
- if shutil.which("xdg-open"):
+ if 0 and shutil.which("xdg-open"):
register("xdg-open", None, BackgroundBrowser("xdg-open"))
# Opens an appropriate browser for the URL scheme according to
@@ -549,7 +549,7 @@ def register_standard_browsers():
# Prefer X browsers if present
if os.environ.get("DISPLAY") or os.environ.get("WAYLAND_DISPLAY"):
try:
- cmd = "xdg-settings get default-web-browser".split()
+ cmd = "false xdg-settings get default-web-browser".split()
raw_result = subprocess.check_output(cmd, stderr=subprocess.DEVNULL)
result = raw_result.decode().strip()
except (FileNotFoundError, subprocess.CalledProcessError, PermissionError, NotADirectoryError) :
```
</details>
I haven't attempted to assess which of the specific web browsers such as Galeon are still extant, and which ones disappeared years ago. They could almost certainly be cleaned up, but that's beyond the scope of this PR.
* bpo-45514: Apply changes from importlib_resources@a3ef4128c6
* Mark legacy functions as deprecated in the docs and link to the migration docs in importlib_resources docs.
* Apply changes from importlib_resources@329ae9d5f2c.
* Indicate importlib.resources as a module.
Co-authored-by: Filipe Laíns <lains@riseup.net>
* Make internal APIs that take PyFrameConstructor take a PyFunctionObject instead.
* Add reference to function to frame, borrow references to builtins and globals.
* Add COPY_FREE_VARS instruction to allow specialization of calls to inner functions.
Special handling is needed, because for non-slots dataclasses the instance attributes are not set: reading from a field just references the class's attribute of the same name, which contains the default value. But this doesn't work for classes using __slots__: they don't read the class's attribute. So in that case (and that case only), initialize the instance attribute. Handle this for both normal defaults, and for fields using default_factory.
Some of the tests in test_queue had a race condition in which a
non-sentinel value could be enqueued after the final sentinel value
leading to not all the inputs being processed (and test failures).
This changes feed() to enqueue a sentinel once the inputs are exhausted,
which guarantees that the final queued object is a sentinel. This
requires the number of feeder threads to match the number of consumer
threads, but that's already the case in the relevant tests.
Unlike the other locks reinitialized by _PyRuntimeState_ReInitThreads,
the "interpreters.main->id_mutex" is not freed by _PyRuntimeState_Fini
and should not force the default raw allocator.
Continue with the improvement of the library netrc
Original work and report Xiang Zhang <angwerzx@126.com>
* 📜🤖 Added by blurb_it.
Co-authored-by: blurb-it[bot] <43283697+blurb-it[bot]@users.noreply.github.com>
Remove the asyncore and asynchat modules, deprecated in Python
3.6: use the asyncio module instead.
Remove the smtpd module, deprecated in Python 3.6: the aiosmtpd
module can be used instead, it is based on asyncio.
* Remove asyncore, asynchat and smtpd documentation
* Remove test_asyncore, test_asynchat and test_smtpd
* Rename Lib/asynchat.py to Lib/test/support/_asynchat.py
* Rename Lib/asyncore.py to Lib/test/support/_asyncore.py
* Rename Lib/smtpd.py to Lib/test/support/_smtpd.py
* Remove DeprecationWarning from private _asyncore, _asynchat and
_smtpd modules
* _smtpd: remove deprecated properties
* bpo-45512: Raise sqlite3.Connection.__init__ is called with bad isolation level
* Also explicitly test allowed isolation levels
* Use subTest for better error messages if something goes wrong
* Update Lib/test/test_sqlite3/test_dbapi.py
Co-authored-by: Dong-hee Na <donghee.na92@gmail.com>
Co-authored-by: Dong-hee Na <donghee.na92@gmail.com>
Remove the --findleaks command line option of regrtest: use the
--fail-env-changed option instead. Since Python 3.7, it was a
deprecated alias to the --fail-env-changed option.
This gains 10% or more in startup time for `python -c pass` on UNIX-ish systems.
The Makefile.pre.in generating code builds on Eric's work for bpo-45020, but the .c file generator is new.
Windows version TBD.
- Uniform usage of `test.support.requires_docstrings` instead of a
local check for `-OO`.
- Added `requires_docstrings` to a couple more methods that need it.
- Replaced a few instances of `test.test_pydoc` with `__name__` to allow
for different methods of running just this test file.
- Rewrote `test_server` to run faster and better test the server.
- Removed unused import.
- Removed unused locals.
- Minor whitespace cleanups.
In order to fix a bug in the 3.9 branch in #29394, more tests were added to
``test_functools.py`` to ensure that ``singledispatchmethod`` still correctly
wrapped a target method, even if the target method had already been wrapped by
multiple other decorators. This PR brings the new tests into the 3.11 and 3.10
branches as well.
Currently custom modules (the array set on PyImport_FrozenModules) replace all the frozen stdlib modules. That can be problematic and is unlikely to be what the user wants. This change treats the custom frozen modules as additions instead. They take precedence over all other frozen modules except for those needed to bootstrap the import system. If the "code" field of an entry in the custom array is NULL then that frozen module is treated as disabled, which allows a custom entry to disable a frozen stdlib module.
This change allows us to get rid of is_essential_frozen_module() and simplifies the logic for which frozen modules should be ignored.
https://bugs.python.org/issue45395
the current test depended on integer sets being iterated on in a certain
fixed order. That order is different on PyPy (insertion based) and could
change in CPython in the future in theory. Make the test robust against
a different iteration order by sorting.
The "freeze" tool has been part of the repo for a long time. However, it hasn't had any tests in the test suite to guard against regressions. We add such a test here. This is especially important as there has been a lot of change recently related to frozen modules, with more to come.
Note that as part of the test we build Python out-of-tree and install it in a temp dir.
https://bugs.python.org/issue45629
The doctest module raised an error if a docstring contained an example that
attempted to access a classmethod property. (Stacking '@classmethod' on top of
`@property` has been supported since Python 3.9; see
https://docs.python.org/3/howto/descriptor.html#class-methods.)
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
Use types.GenericAlias in inspect.formatannotation to correctly add
type arguments of builtin types to the string representation of
Signatures.
Co-authored-by: Martin Rückl <martin.rueckl@codecentric.de>
(original patch by eryksun)
Correctly hand various failure modes when resizing an mmap on Windows:
* Resizing a pagefile-backed mmap now creates a new mmap and copies data
* Attempting to resize when another mapping is held on the same file raises an OSError
* Attempting to resize a nametagged mmap raises an OSError if another mapping is held with the same nametag
The recently added PyConfig.stdlib_dir was being set with ".." entries. When __file__ was added for from modules this caused a problem on out-of-tree builds. This PR fixes that by normalizing "stdlib_dir" when it is calculated in getpath.c.
https://bugs.python.org/issue45506
This is a cross-platform check that the symbols are actually
exported in the ABI, not e.g. hidden in a macro.
Caveat: PyModule_Create2 & PyModule_FromDefAndSpec2 are skipped.
These aren't exported on some of our buildbots. This is a bug
(bpo-44133). This test now makes sure all the others don't regress.
Freelists for object structs can now be disabled. A new ``configure``
option ``--without-freelists`` can be used to disable all freelists
except empty tuple singleton. Internal Py*_MAXFREELIST macros can now
be defined as 0 without causing compiler warnings and segfaults.
Signed-off-by: Christian Heimes <christian@python.org>
Rename namespace package __loader__ class to be public.
Make the old name, i.e. _NamespaceLoader, an alias for the public name, for backward compatibility.
An object implementing the os.PathLike protocol can represent a file
system path as a str or bytes object.
Therefore, _infer_return_type function should infer os.PathLike[str]
object as str type and os.PathLike[bytes] object as bytes type.
Serhiy Storchaka
Andrew Svetlov
Kumar Aditya
Jason R. Coombs
andrei kulakov
Jakub Kulík
Christian Heimes
Mark Shannon
Irit Katriel
Yilei "Dolee" Yang
Ken Jin
AN Long
Vinay Sajip
Pablo Galindo Salgado
Jim Crist-Harif
Christian Sattler
Eric V. Smith
Steve Dower
Carl Friedrich Bolz-Tereick
Weipeng Hong
Yurii Karabas
neonene
Raymond Hettinger
Gabriele N. Tornetta
Kumar Aditya
Jacob Hayes
Ned Batchelder
Victor Stinner
James Gerity
180909
Vishal Pandey
Dennis Sweeney
Alex Waygood
Adrian Garcia Badaracco
Gideon
Erlend Egeberg Aasland
Sam Bull
180909
Simon McVittie
Miro Hrončok
Eric Snow
Nikita Sobolev
Logan Jones
Dylan Van Assche
Ruben Vorderman
Hugo van Kemenade
Dong-hee Na
Sam Gross
Emmanuel Arias
Brandt Bucher
Guido van Rossum
Zachary Ware
Chris Wesseling
Łukasz Langa
Tim Golden
Martin Rueckl
Petr Viktorin
Gregory P. Smith
Mark Dickinson
E-Paine
Kreus Amredes
Barry Warsaw
Kyungmin Lee