Change by Steve Dower :
--
keywords: +patch
pull_requests: +28034
stage: needs patch -> patch review
pull_request: https://github.com/python/cpython/pull/29799
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
For future reference, in case someone stumbles over this issue, here's how I
tested this change:
* set $env:IncludeUWP="true"
* did a regular PCbuild/build.bat
* generated a Store layout with "./python.bat PC/layout --preset-appx --copy
./o
Steve Dower added the comment:
This is because we use libffi as a DLL and you can't statically reference
dynamically loaded addresses on Windows.
You could change that format table to be initialised on first use, or resolve
the FFI type lazily. Or change ctypes to statically link l
Steve Dower added the comment:
New changeset 4841e694ee1686f8c933ddfcb8c854915867ce17 by Steve Dower in branch
'main':
bpo-45901: Fixes argument passing when invoking .py files directly through the
Store package (GH-29799)
https://github.com/python/cpyt
Steve Dower added the comment:
This change modified the audit event 'code.__name__', which requires a
deprecation period (all events are public API, as per PEP 578).
We need to revert that part of the change. I don't think we need to add a new
event to report the qualnam
Steve Dower added the comment:
Correction: the event is `code.__new__`
--
___
Python tracker
<https://bugs.python.org/issue44530>
___
___
Python-bugs-list mailin
Change by Steve Dower :
--
pull_requests: +28041
stage: resolved -> patch review
pull_request: https://github.com/python/cpython/pull/29809
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
New changeset 133fb267f437f9bd568738d0e0e59e574fc82b02 by Miss Islington (bot)
in branch '3.9':
bpo-45901: Fixes argument passing when invoking .py files directly through the
Store package (GH-29799)
https://github.com/python/cpyt
Steve Dower added the comment:
Status update on this: I owe everyone a perf comparison of the before/after
with this change.
I don't particularly want to block on a regression unless it's significant
(honestly still have no idea what to expect), but open to others' thoughts
Change by Steve Dower :
--
assignee: -> steve.dower
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python
Steve Dower added the comment:
New changeset db55f3fabafc046e4fca907210ced4ce16bf58d6 by Steve Dower in branch
'main':
bpo-44530: Reverts a change to the 'code.__new__' audit event (GH-29809)
https://github.com/python/cpython/commit/db55f3fabafc046e4fca
Steve Dower added the comment:
It's one data point (well, statistics over 1000 points), but it looks like it's
actually a slight improvement in performance over the previous code on Windows
:)
before after
min 23.103 22.154
25% 25.069 23.59925
50% 25.8125 24.2715
75
Steve Dower added the comment:
Nice catch, thanks!
Did you start the CLA process already? (I personally would exempt this PR from
it as "not sufficiently creative", but I'm not sure I'm allowed to do that
anymore...)
--
__
Steve Dower added the comment:
Basically unchanged on Debian/WSL as well.
There's a new conflict arisen, so I'll resolve that and then merge.
--
___
Python tracker
<https://bugs.python.o
Steve Dower added the comment:
New changeset 734ed3538314c9651ae64d5e2e0e98aae3aec17b by David Federman in
branch 'main':
bpo-45931: Prevent Directory.Build.props/targets from leaking from directories
above the repo when building on Windows (GH-29854)
https://github.com/pyth
Steve Dower added the comment:
Looks like it's gone through. Thanks!
--
nosy: -miss-islington
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
versions: +Python 3.10, Python 3.9
___
Python t
Steve Dower added the comment:
An easy workaround for people encountering this is to set the PlatformToolset
environment variable to v143 before building:
$env:PlatformToolset = "v143"
(or in CMD: set PlatformToolset=v143)
Alternatively, you should be able to install the v1
Steve Dower added the comment:
New changeset 99fcf1505218464c489d419d4500f126b6d6dc28 by Steve Dower in branch
'main':
bpo-45582: Port getpath[p].c to Python (GH-29041)
https://github.com/python/cpython/commit/99fcf1505218464c489d419d4500f1
Steve Dower added the comment:
It's done! Those were some of the hardest memory leaks I've had to track down,
but it should be clear now.
As I mentioned in the commit message, this change attempts to preserve every
known quirk. However, I think these ought to be streamlined across
Steve Dower added the comment:
Unless someone can find an authoring bug under Tools/msi, I don't think we can
do anything about this. We rely on the built-in Windows support for adding and
removing these variables, and if that's broken, we can't do anything with our
current s
Steve Dower added the comment:
New changeset d9301703fb1086cafbd730c17e3d450a192485d6 by Crowthebird in branch
'main':
bpo-45816: Support building with VS 2022 (v143 toolset) on Windows (GH-29577)
https://github.com/python/cpython/commit/d9301703fb1086cafbd730c17e3d45
Change by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Change by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
It probably all needs straightening out:
* Microsoft Visual Studio 2015 *or later* can be used to build the project files
* Microsoft Visual C++ 14.3 is what we use to build the official releases for
the main branch (3.11)
* earlier branches may still be using
Steve Dower added the comment:
New changeset 7d7c91a8e8c0bb04105a21a17d1061ffc1c04d80 by neonene in branch
'main':
bpo-45582: Add a NOT operator to the condition in getpath_isxfile (GH-29906)
https://github.com/python/cpython/commit/7d7c91a8e8c0bb04105a21a17d1061
Change by Steve Dower :
--
pull_requests: -27916
___
Python tracker
<https://bugs.python.org/issue44844>
___
___
Python-bugs-list mailing list
Unsubscribe:
Change by Steve Dower :
--
pull_requests: +28132
pull_request: https://github.com/python/cpython/pull/29908
___
Python tracker
<https://bugs.python.org/issue44
Steve Dower added the comment:
@Ray I posted a PR that works fine on Windows, would you be able to try it on
Linux? The executables seem to have different names, so I'm a little concerned
about the change, but I think it should be al
Steve Dower added the comment:
Not sure if Mark and Martin are part of an expert group, but neither are active
in this area these days.
For the original question, I have no idea why the command can't use a builtin
name, so hard to say whether we'd change it. A summary of why it
Steve Dower added the comment:
New changeset af1db4eb555e02d2bff3476f99f7a653764203b0 by neonene in branch
'main':
bpo-45582: Fix getpath_isxfile() and test_embed on Windows (GH-29930)
https://github.com/python/cpython/commit/af1db4eb555e02d2bff3476f99f7a6
Steve Dower added the comment:
They should all be under "-32" when in roaming appdata, because otherwise
having 32-bit and 64-bit installs side-by-side will conflict.
If these fields were newly added to distutils then should probably fix them to
match sysconfig. If they'r
Steve Dower added the comment:
Are you launching "py" from inside MSYS2? Or a regular prompt?
What does the output of "py -0p" show? (It should list all the discovered
installs)
We might need someone who understands how MSYS2 works h
Steve Dower added the comment:
So if I've understood you, there are three Python installs on your machine. Two
of them are Windows builds, which are listed in the registry and discovered by
py.exe -0p. The third is the MSYS2 build, which is *not* listed in the registry
and is
Steve Dower added the comment:
("2" and "3" are just arbitrary numbers... substitute however many Python
runtimes you actually have, or "N" and "N+1" if you prefer.)
--
___
Python trac
Change by Steve Dower :
--
pull_requests: +28173
pull_request: https://github.com/python/cpython/pull/29948
___
Python tracker
<https://bugs.python.org/issue45
Steve Dower added the comment:
New changeset b7ef27bc084665ce58d89fc69530c6f9d2d37754 by Steve Dower in branch
'main':
bpo-45582: Ensure PYTHONHOME still overrides detected build prefixes (GH-29948)
https://github.com/python/cpython/commit/b7ef27bc084665ce58d89fc69530c6
Steve Dower added the comment:
Yes, it's been fixed.
The debate about whether to test in CI with C assertions enabled or not can
continue somewhere else.
--
resolution: -> fixed
stage: test needed -> resolved
status: open -> closed
Change by Steve Dower :
--
pull_requests: +28205
pull_request: https://github.com/python/cpython/pull/29979
___
Python tracker
<https://bugs.python.org/issue45
Steve Dower added the comment:
New changeset b0b30862796e97b3f0ee358bcc61d21f0cc98441 by Steve Dower in branch
'main':
bpo-45582: Write empty pybuilddir.txt on Windows to allow relocatable build
directories (GH-29979)
https://github.com/python/cpyt
New submission from Steve Dower :
Before, a venv would include the DLLs directory in sys.path. Since issue45582,
it does not, and so anything that relies on a standard library native module
cannot run.
--
assignee: steve.dower
components: Windows
messages: 408033
nosy: paul.moore
Steve Dower added the comment:
+RM
--
nosy: +pablogsal
___
Python tracker
<https://bugs.python.org/issue46015>
___
___
Python-bugs-list mailing list
Unsub
Change by Steve Dower :
--
keywords: +patch
pull_requests: +28215
stage: needs patch -> patch review
pull_request: https://github.com/python/cpython/pull/29992
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
New changeset 7778116c2f573edf320bd55301137a968e4339d8 by Steve Dower in branch
'main':
bpo-46015: Fixes calculation of sys.path in a venv on Windows (GH-29992)
https://github.com/python/cpython/commit/7778116c2f573edf320bd55301137a
Change by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
New submission from Steve Dower :
If a C runtime's math functions set errno to ERANGE, we assume it is a valid
underflow if fabs(result) < 1.0.
However, because expm1 includes a -1.0, it underflows towards -1.0. This fails
the above check, and so if a runtime's expm1() sets E
Change by Steve Dower :
--
keywords: +patch
pull_requests: +28220
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/29997
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
I considered just switching to <2.0, but wasn't sure if I would be breaking
some other unspoken behaviour there. But you're right, it's really just
detecting underflow vs. overflow, so that's a much simpler way to check.
I've
Steve Dower added the comment:
I've also got no idea how to write a test for this, given that it's a very thin
wrapper around a platform's C runtime library. Our existing test discovered
when the library changed behaviour to start setting errno, which is probably
the
Steve Dower added the comment:
Lines 500-504 are the ones that trigger it. Apparently there are no tests in
that file for straight exp(), but the equivalent tests for those would return
0.0 and suppress ERANGE too.
--
___
Python tracker
<ht
Steve Dower added the comment:
New changeset 3363e1cb05d0d19ed172ea63606d8cb6268747fc by Steve Dower in branch
'main':
bpo-46018: Ensure that math.expm1 does not raise on underflow (GH-29997)
https://github.com/python/cpython/commit/3363e1cb05d0d19ed172ea63606d8c
Change by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
That's a shame.
If someone has the actual file (I don't), they can publish it elsewhere. Or if
one of our team is prepared to validate it (I'm not) we can overwrite what's
there.
Otherwise, the fix is to use the latest version of Python,
Steve Dower added the comment:
The problem is that Cygwin (on which MinGW and Git Bash are based) tries to
read the contents of the executable file, rather than just executing it. The
type of link used by Windows here cannot be read as a normal file.
The provided workaround is the best we
Steve Dower added the comment:
I'm not totally averse to just removing the link pragmas completely, but since
that's got significant compatibility impact, I'm happy with this approach.
Let's shorten the name though. "PY_NO_
Steve Dower added the comment:
New changeset d593881505c1f4acfd17f41312b27cc898451816 by Steve Dower (jsnklln)
in branch 'master':
bpo-38723: Pdb._runscript should use io.open_code() instead of open() (GH-17127)
https://github.com/python/cpyt
Steve Dower added the comment:
Thanks for the PR!
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
versions: +Python 3.8
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
My latest push to PR 16967 is a significant change to what ntpath.realpath will
return for broken symlinks, but I think it's the right change.
Basically, if the OS fully resolves the path, great! We'll return that (and
handle \\?\ stripping)
If the
New submission from Steve Dower :
See
https://dev.azure.com/Python/cpython/_build/results?buildId=53812&view=ms.vss-test-web.build-test-results-tab
Traceback (most recent call last):
File "/Users/runner/runners/2.160.0/work/1/s/Lib/unittest/case.py", line 60,
in testPartExec
New submission from Steve Lorimer :
There is a race condition when closing/joining a pool with maxtasksperchild=1
Illustrative example:
```
#!/usr/bin/env python3
import os
import time
import logging
import multiprocessing.pool
def run_task(i):
print(f'[{os.getpid()}] task({i}) com
Steve Dower added the comment:
I'm not sure - I try the same thing (on 3.6, 3.7 and 3.8) and get different
results:
>>> import pathlib
>>> p = pathlib.Path(".", "a", "b", "c")
>>> p.resolve(strict=False)
WindowsPath(
Steve Dower added the comment:
If "a" exists in the current directory, I get this (correct) result:
>>> import pathlib
>>> p = pathlib.Path(".", "a", "b", "c")
>>> p.resolve(strict=False)
WindowsPath('/a/b
Change by Steve Dower :
--
keywords: +patch
pull_requests: +16665
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/16828
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
I just added a PR that will produce ARM64 releases (as embeddable, nuget and
Microsoft Store packages).
The only missing feature is tkinter (and everything that depends on it). For
users who require that, I'd suggest grabbing the x86 installer from pytho
Steve Dower added the comment:
Thanks! Would you like to create a PR on GitHub for this? Or are you happy for
me to do it.
--
nosy: +steve.dower
___
Python tracker
<https://bugs.python.org/issue38
Steve Dower added the comment:
Actually, it looks like we need to add events for many of the _ctypes
functions, so I'll go through and do them.
--
___
Python tracker
<https://bugs.python.org/is
Change by Steve Dower :
--
assignee: -> steve.dower
___
Python tracker
<https://bugs.python.org/issue38622>
___
___
Python-bugs-list mailing list
Unsubscrib
Change by Steve Dower :
--
pull_requests: +16667
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/17158
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
I think the PR as it stands is a net improvement over where we currently are,
so I'll merge it. We can make more tweaks as necessary.
--
___
Python tracker
<https://bugs.python.org/is
Steve Dower added the comment:
New changeset abde52cd8e31830bfc06c5803221faae6172104a by Steve Dower in branch
'master':
bpo-38453: Ensure ntpath.realpath correctly resolves relative paths (GH-16967)
https://github.com/python/cpython/commit/abde52cd8e31830bfc06c5803221fa
Change by Steve Dower :
--
pull_requests: +16682
pull_request: https://github.com/python/cpython/pull/17174
___
Python tracker
<https://bugs.python.org/issue38
Steve Dower added the comment:
Thanks, Victor! That looks different from the one I was originally fixing (even
though the long/short path mismatch is there), so I'll take a look.
--
___
Python tracker
<https://bugs.python.org/is
Steve Dower added the comment:
New changeset 66c0f01f98fd6db0a4b1e789b9db9c7247a24ba9 by Steve Dower in branch
'3.8':
bpo-38453: Ensure ntpath.realpath correctly resolves relative paths (GH-16967)
https://github.com/python/cpython/commit/66c0f01f98fd6db0a4b1e789b9db9c
Change by Steve Dower :
--
pull_requests: +16691
pull_request: https://github.com/python/cpython/pull/17184
___
Python tracker
<https://bugs.python.org/issue38
Steve Dower added the comment:
New changeset 7c6130c8c36c941255365e5414c956fc919b8629 by Steve Dower in branch
'master':
bpo-38453: Ensure correct short path is obtained for test (GH-17184)
https://github.com/python/cpython/commit/7c6130c8c36c941255365e5414c956
Steve Dower added the comment:
At a minimum, it needs to be turned into a GitHub PR.
We've made some significant changes in this area in 3.8, so possibly the best
available code is now in shutil.rmtree (or shutil.copytree) rather than the
older patch
Steve Dower added the comment:
If the patches are for Tcl/Tk, then you should submit it to them.
I'm not a Tk maintainer, and I never want to be one, so frankly I'd be happier
to say it's not supported on ARM :)
But if it builds fine with our PCbuild/prepare_tcltk.bat script
Steve Dower added the comment:
I haven't debugged it, but I'm guessing we're not handling the trailing slash
properly when falling back to asking the parent directory for information.
Looking at the actual stat result for "C:\System Volume Information" vs.
Steve Dower added the comment:
Could you share just one of your .pyd files?
Without being able to see whether they are compiled incorrectly, it's hard to
be sure whether this is the same cause as before. It certainly looks like
distutils is still going to link corr
Steve Dower added the comment:
Closing this as I believe it's done, but happy to reopen if we find another
edge case (or start a new issue).
--
assignee: -> steve.dower
resolution: -> fixed
stage: patch review -> resolved
status:
Steve Dower added the comment:
It's a security issue because Python 3.8 says it will open files to be executed
with io.open_code() instead of open(). This allows a way to bypass that.
That said, this appears to be a fallback case, so I'm not hugely concerned. I
haven't quite
Steve Dower added the comment:
New changeset 00923c63995e34cdc25d699478f113de99a69df9 by Steve Dower in branch
'master':
bpo-38622: Add missing audit events for ctypes module (GH-17158)
https://github.com/python/cpython/commit/00923c63995e34cdc25d699478f113
Steve Dower added the comment:
> I hadn't realized that we'd made such a declaration WRT opening of code files
> in general.
It wasn't exactly a hugely publicised declaration :)
The relevant quote from PEP 578 is:
> All import and execution functionality involving
Change by Steve Dower :
--
pull_requests: +16742
pull_request: https://github.com/python/cpython/pull/17243
___
Python tracker
<https://bugs.python.org/issue38
Steve Dower added the comment:
Typically, as soon as I merge, I spot an edge case issue.
PySys_Audit(n, "O", a) is deliberately going to treat 'a' as the tuple of
arguments (when it is a tuple). This lets us simplify/optimise events where the
event arguments match th
Steve Dower added the comment:
New changeset dcf1f83de8678b09df5bd7d04ca5f4ef1cd02aca by Steve Dower in branch
'master':
bpo-38622: Ensure ctypes.PyObj_FromPtr audit event passes tuples as a single
argument (GH-17243)
https://github.com/python/cpyt
Change by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
New changeset de148f263fba75cd10d2cb010fe9c495cee4ec83 by Steve Dower in branch
'master':
bpo-33125: Add support for building and releasing Windows ARM64 packages
(GH-16828)
https://github.com/python/cpython/commit/de148f263fba75cd10d2cb010fe9c4
Steve Dower added the comment:
I've merged the change to master, but note that the release is not enabled, so
we aren't actually releasing ARM64 builds yet (though they will be compiled
[but not tested] in PRs - nothing should be different from other Windows builds
though). There
New submission from Steve Dower :
(Split out from issue33125)
We currently do not have ARM/ARM64 builds of Tcl/Tk for Windows, so we cannot
enable tkinter on this platform.
When builds of the dependencies become possible (in 8.6.10, apparently), we can
enable it again
Steve Dower added the comment:
I also filed issue38867 specifically for enabling Tkinter on Windows ARM.
--
___
Python tracker
<https://bugs.python.org/issue33
Steve Dower added the comment:
> Was pathlib forgotten here? Pathlib.home() is documented to return the same
> as expanduser("~") but it still prefers HOME instead of USERPROFILE.
Yes, it was forgotten (why doesn't it just use expanduser?). We should file a
new bug f
Steve Dower added the comment:
Thanks Terry, this is great!
Agree with everything apart from not liking "raise" - I think "raise an event"
is totally fine usage, and certainly better than "triggers an audit call"
(since it may not, if nobody's listeni
Steve Dower added the comment:
I would guess this is due to the updated libffi, and probably it's a case that
is not sufficiently tested.
Adding tests is the first step. It'll probably have to enumerate many parameter
types (in case there's something weird here like misalig
Steve Dower added the comment:
Yeah, this definitely relates to how Windows handles unqualified argv[0] in
CreateProcess. Though I thought we checked that out in another issue and
decided that "most" people are correctly using sys.executable here? (Or decided
that it was docum
Steve Dower added the comment:
As a general statement (I haven't taken the time yet to understand all the
intricacies of this particular issue), I would prefer to see CopyFile used
everywhere on Windows for the performance and reliability benefits. Perfect
POSIX emulation here doesn
Steve Dower added the comment:
Didn't we clean up this warning completely on Windows recently?
It apparently matters on POSIX to join() the child process, but on Windows you
aren't going to leak any resources by just forgetting about it, so I thought we
stopped tracking the
Steve Dower added the comment:
> `shutil.which` could be a direction, but still not enough since it’d break
> `flit install --python=py` because that’s give you the location of py.exe,
> not the actual interperter.
This would be fine if you still run the process to get its sys.e
Steve Dower added the comment:
> Perhaps generalize this as splitext(basename(sys.executable))[0] in
> order to support names other than "python" and "pythonw". It would have
> to handle quoting enough to ignore an initial double quote when parsing
> the execu
Steve Dower added the comment:
Does it depend on whether PY_SSIZE_T_CLEAN is defined?
--
nosy: +steve.dower
___
Python tracker
<https://bugs.python.org/issue38
Steve Dower added the comment:
New changeset e563a155be60fc0757914f87c8138f10de00bb16 by Steve Dower (Terry
Jan Reedy) in branch 'master':
bpo-38892: Improve docs for audit event (GH-17361)
https://github.com/python/cpython/commit/e563a155be60fc0757914f87c8138f
1301 - 1400 of 6138 matches
Mail list logo