Change by Steve Dower :
--
keywords: +patch
pull_requests: +20892
stage: needs patch -> patch review
pull_request: https://github.com/python/cpython/pull/21748
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
Test failure on Windows. I'll take a look tomorrow.
==
FAIL: test_with_pip (test.test_venv.EnsurePipTest)
--
Traceback
New submission from Steve Dower :
At some point, Windows started displaying an authenticated certificate property
as the display name in UAC prompts, rather than the file properties.
Currently our SigningDescription is set to the build identifier, which is not a
nice display name. We should
Change by Steve Dower :
--
keywords: +patch
pull_requests: +20899
stage: needs patch -> patch review
pull_request: https://github.com/python/cpython/pull/21754
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
Yes, it only affects Windows OS.
On all other platforms, the python38._pth file is _always_ ignored. We have not
implemented this support for those platforms.
--
___
Python tracker
<https://bugs.python.
Steve Dower added the comment:
New changeset 777b611c8c5676b80898a429f71d28e59bddc49d by Steve Dower in branch
'master':
bpo-41492: Fixes the description appearing in UAC prompts on Windows (GH-21754)
https://github.com/python/cpython/commit/777b611c8c5676b80898a429f71d28
Steve Dower added the comment:
The issue above doesn't appear to repro on 3.9, so I guess master has started
leaking a file handle, presumably in zipimport.
I'll see what I can track down, but can't be sure I'll have enough time to get
it done for RC, so if anyone el
Steve Dower added the comment:
Okay, I've tracked it down to the new importlib.readers.ZipReader class keeping
the zip file open, presumably until it gets GC'd. This is used by certifi to
extract the CA certs from the whl when ensurepip is doing the self-install from
the mou
Steve Dower added the comment:
Added some test cases to the PR that directly trigger the issue, specifically
this one:
def test_entered_path_does_not_keep_open(self):
# This is what certifi does on import to make its bundle
# available for the process duration.
c
Change by Steve Dower :
--
pull_requests: +20919
pull_request: https://github.com/python/cpython/pull/21774
___
Python tracker
<https://bugs.python.org/issue41
Change by Steve Dower :
--
pull_requests: +20920
pull_request: https://github.com/python/cpython/pull/21775
___
Python tracker
<https://bugs.python.org/issue41
Steve Dower added the comment:
GitHub Actions has decided not to run CI today, so you'll have to look at Azure
Pipelines for the test failures:
https://dev.azure.com/Python/cpython/_build/results?buildId=67152&view=logs&j=c83831cd-3752-5cc7-2f01-8276919eb334&t=5a421c
Steve Dower added the comment:
New changeset 135de08128a76f49752ac57c316129500275e828 by Steve Dower in branch
'3.8':
bpo-41490: Update ensurepip to install pip 20.2.1 and setuptools 49.2.1
(GH-21775)
https://github.com/python/cpython/commit/135de08128a76f49752ac57c316129
Change by Steve Dower :
--
versions: -Python 3.8
___
Python tracker
<https://bugs.python.org/issue41490>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
New changeset 70e9243a55be9c32b41f2149cdfa3957f96f8471 by Steve Dower in branch
'3.9':
bpo-41490: Update ensurepip to install pip 20.2.1 and setuptools 49.2.1
(GH-21774)
https://github.com/python/cpython/commit/70e9243a55be9c32b41f2149cdfa39
Change by Steve Dower :
--
versions: -Python 3.9
___
Python tracker
<https://bugs.python.org/issue41490>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
The Python 3.8 and 3.3 installers will not conflict at all, as they are
completely different.
If that is your log file, we'll also need the file in your %TEMP% directory
that includes "core_JustForMe" in the name, as that's the subinsta
Steve Dower added the comment:
The trailing colon is unnecessary, and likely to cause more issues, but the
same thing occurs for Path('NUL').resolve()
This is probably best handled as issue37517, where you'll find more background
on the error messages.
But I believe that p
Steve Dower added the comment:
So it'll take a bit more rearranging to get everything to fit - I've attached
one proposal that shuffles the thanks to Mark Hammond down below the "enable
long path names" button. (It's grey text in the screenshot, but I've
Steve Dower added the comment:
Thanks, Mark. We shall not hold you solely responsible for the port any longer
;)
--
___
Python tracker
<https://bugs.python.org/issue41
Change by Steve Dower :
--
keywords: +patch
pull_requests: +20996
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/21871
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
New changeset 6444ca946984c638c67a72aac22fd6d3cc650c16 by Steve Dower in branch
'master':
bpo-41526: Fixed layout of final page of the installer (GH-21871)
https://github.com/python/cpython/commit/6444ca946984c638c67a72aac22fd6
Steve Dower added the comment:
Fixes merged, just need Łukasz to confirm it'll be in 3.9.0 and we can resolve
this.
--
nosy: +lukasz.langa
___
Python tracker
<https://bugs.python.org/is
Steve Dower added the comment:
This is normal, if obscure, behaviour. Pythonw starts without a console, and so
stdout is not connected to anything. As a result, you can't print (or pprint).
You'll need to set sys.stdout to something or provide a file if you want to
print output.
Steve Dower added the comment:
Thanks for the report! This should be in the next 3.9 RC.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
I'm inclined to agree that it should pass silently in this case, as if it were
printing with print() rather than .write().
What better meaning is there for sys.stdout == None than "no output"?
--
versions: +Python 3
Steve Dower added the comment:
Are you able to capture a process dump at the hang? I haven't seen this
anywhere else, and don't even know how to go about trying to reproduce it with
this information - Py_Initialize is called by every single Python process, so
there's somethin
Change by Steve Dower :
--
versions: -Python 3.5, Python 3.6, Python 3.7
___
Python tracker
<https://bugs.python.org/issue41376>
___
___
Python-bugs-list mailin
Steve Dower added the comment:
Assuming the fix for this was released, given lack of other information.
--
resolution: -> fixed
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
New submission from Steve Dower :
Currently, the user site packages on Windows is %APPDATA%\Python\PythonXY. This
can cause conflicts between the 32-bit and 64-bit versions of the runtime.
We should switch the pattern to Python{sys.winver}, which is XY or XY-32, the
same as elsewhere.
This
Steve Dower added the comment:
Any input from the asyncio experts?
I don't have an issue with handling the exception in this case, and hopefully
when someone is up to the task of dealing with the range of edge cases
throughout this loop implementation, hopefully they can get the orderi
Steve Dower added the comment:
> The problem is that Powershell has case-sensitive commands and pip is trying
> to call path completely in minor case
This is not normal configuration. I'm not sure what you've done here, but
please open a new issue with details about how
Steve Dower added the comment:
The curses module is currently not supported on Windows, and this patch is far
more than just updating the VS build. I've updated the title.
Refer to issue31951 for more context.
This will need more work than what's been done in the pull request,
Steve Dower added the comment:
Unfortunately, it's easier to change what we distribute, except that has
_always_ been "include" (based on a 2.4 installer I had laying around). And of
course we can't change the repo.
Still, case sensitivity is becoming more acceptable
Steve Dower added the comment:
Thanks, Ben!
--
___
Python tracker
<https://bugs.python.org/issue39010>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
> Could sigint_event be relocated to a platform-dependent extension of
> _PyRuntimeState that gets initialized and finalized independent of any module?
This, or leaving it where it is and making sure it gets initialised (it really
is about emulating c
Steve Dower added the comment:
I've considered DoSing myself a few times, but then change my mind and just
publish the release :)
A PR to change it to "(\d+\.\d+\.\d+)([a-zA-Z]+\d+)?$" would be fine, but is
not urgent. It certainly doesn't need to be backported, as thi
Change by Steve Dower :
--
keywords: +patch
pull_requests: +21185
stage: -> patch review
pull_request: https://github.com/python/cpython/pull/22098
___
Python tracker
<https://bugs.python.org/issu
Steve Dower added the comment:
> If it is otherwise possible to user-only install both 32 and 64 bit versions,
> then using the same site-packages strikes me a bug
It's very easy to install both 32 and 64-bit runtimes (and it doesn't matter if
they're user or system
Steve Dower added the comment:
So I found the "owner" of the HTML Help component in Windows (I put owner in
quotes because this tool is _very_ maintenance mode, so nothing short of a
critical vulnerability is going to be touched) and he helped me do some
debugging.
In short, the
Steve Dower added the comment:
New changeset dd18001c308fb3bb65006c91d95f6639583a3420 by Steve Dower in branch
'master':
bpo-41627: Distinguish 32 and 64-bit user site packages on Windows (GH-22098)
https://github.com/python/cpython/commit/dd18001c308fb3bb65006c91d95f66
Steve Dower added the comment:
> With the change in PR 22098, the 32-bit interpreter will install to a
> different location.
To clarify this, I meant the 32-bit interpreter will install *packages* to a
different location (when using the user scheme).
--
assignee: -> st
Steve Dower added the comment:
Thanks for the contribution! I think I only ever tested this with the 64-bit
package... :)
Could you make it so that we include the file under both names? Just for anyone
(like me) who has hard-coded the python.props name and parametrised the
directory name
Change by Steve Dower :
--
versions: +Python 3.10, Python 3.8, Python 3.9
___
Python tracker
<https://bugs.python.org/issue41744>
___
___
Python-bugs-list mailin
Steve Dower added the comment:
In general, if a bug here appears to be inactive, it's probably waiting on
someone to volunteer to move it forward. Often merely posting to a thread is
enough.
For this case, I think the best thing we can probably do is change the default
share mode for
Steve Dower added the comment:
Issue41490 can also be fixed by using FILE_SHARE_DELETE on all opened files
(and that's a release blocker, so we need to fix it somehow), and if DeleteFile
has been updated as you suggest then it might even help with the "pip replacing
its
Steve Dower added the comment:
The comment you quoted was referring to the NamedTemporaryFile(do_not_delete)
flag.
Yes, we'd have to reimplement the UCRT function using the system API.
Ultimately, it's not a great compatibility layer if you want to match POSIX
semantics and not
Steve Dower added the comment:
We'd CreateFile the file and then immediately pass it to _open_osfhandle, which
would keep the semantics the same apart from the share flags.
I'm not entirely against getting rid of O_TEXT support, but haven't taken the
time to think through t
Steve Dower added the comment:
Thanks! Just need a NEWS file (click Details next to the failed check for the
helper app). Something like "Fixes automatic import of props file when using
the Nuget package" would be good.
(Also posted
Steve Dower added the comment:
If you look at the PR into 3.9, it includes a test for this exact case. Start
by copying that into 3.10 and then make it pass and everything should be good :)
--
___
Python tracker
<https://bugs.python.
Steve Dower added the comment:
No objections. I'd guess it was a typo.
--
___
Python tracker
<https://bugs.python.org/issue41729>
___
___
Python-bugs-list m
Steve Dower added the comment:
New changeset 8f2b991eef062c22c429a96983c78be007180b66 by Zackery Spytz in
branch 'master':
bpo-41646: Mention path-like objects support in the docs for shutil.copy()
(GH-22208)
https://github.com/python/cpython/commit/8f2b991eef062c22c429a96983c78b
Steve Dower added the comment:
New changeset 7c11a9accabe3720940f334eb1226bb7bb9179c7 by Václav Slavík in
branch 'master':
bpo-41744: Package python.props with correct name in NuGet package (GH-22154)
https://github.com/python/cpython/commit/7c11a9accabe3720940f334eb1226b
Steve Dower added the comment:
Thanks!
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
Thanks!
--
assignee: -> steve.dower
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python
Steve Dower added the comment:
Works great! (By which I mean, doesn't work at all ;) )
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.pyth
New submission from Steve Dower :
When a socket object has a non-zero timeout, all send/recv/etc. calls are
preceded by a select() call before making the actual call. If there is no
timeout, the select call is skipped.
It appears that on Windows, select() will block until the previous
Steve Dower added the comment:
Yeah, this whole issue is basically because Burn (our installer tool) won't do
per-machine bundle registration even when it's only installing per-machine
packages. So the install block that it would normally throw up fails if you
use a different us
Steve Dower added the comment:
This warning is coming from pip, and it's because it has detected that your
system configuration doesn't match what many Python developers assume.
Specifically, that you've configured PATH to include both Python and any
package scripts you ha
Steve Dower added the comment:
Python's installer doesn't change that often ;)
Without some information about your scenario, I can't make a recommendation.
However, I will say that if you are including any of your own Python code with
your application, and especially any c
Change by Steve Dower :
--
resolution: -> fixed
stage: backport needed -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
> IMO, the most direct way to resolve the problem is by enabling "loader snaps"
> for python.exe via gflags and attaching a native debugger to the process ...
This is indeed the best way to go about solving it, so you can see why we don'
Steve Dower added the comment:
> would you see this as a note in
https://docs.python.org/3/library/ctypes.html?highlight=ctypes#loading-shared-libraries
Haven't looked at the PR, but it probably needs to be somewhere in the
import docs as well, to do with native extension modules
Steve Dower added the comment:
Thanks for doing the search :)
> -
https://docs.python.org/3/faq/windows.html#is-a-pyd-file-the-same-as-a-dll
Probably not here.
> - https://docs.python.org/3/using/windows.html#finding-modules
Perhaps it is best to put a new section here like wh
New submission from Steve Dower :
The "Install launcher for all users" option on the front page of the installer
should be updated to read "Install py.exe launcher for all users" to reduce
confusion.
However, we should also consider disabling this by default, so that the
Steve Dower added the comment:
Follow up thought: we could switch out the front page checkbox for "Upgrade
existing py.exe launcher (recommended)" when we detect an existing per-machine
install, and hide it entirely if we don't detect it. So at least non-admin
users have
Steve Dower added the comment:
I'm in favour of this change in principle, but would want to look at the PR
closely.
The biggest risk here is that we have to emulate GNU readline for
compatibility, which severely limits the data that can be passed through, and
also forces multiple enc
Steve Dower added the comment:
Sorry, but we don't have any way to update the current terminal process. We
already update the current shell, so if you start a new
terminal/Powershell/cmd/etc. instance it should have the updated variable.
The workaround you posted is fine. Directly a
Steve Dower added the comment:
Sounds good to me.
Pull requests welcome :)
--
___
Python tracker
<https://bugs.python.org/issue41853>
___
___
Python-bugs-list m
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
Change by Steve Dower :
--
resolution: -> fixed
stage: commit review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Change by Steve Dower :
--
pull_requests: +21445
pull_request: https://github.com/python/cpython/pull/22405
___
Python tracker
<https://bugs.python.org/issue39
Steve Dower added the comment:
I did a build for Windows and put up the PR taking it in and there seems to be
one (related) test failure:
==
FAIL: test_from (tkinter.test.test_tkinter.test_widgets.ScaleTest
Steve Dower added the comment:
(My PR is from a branch in the main repo, so if you want to just go in and fix
it, feel free.)
--
___
Python tracker
<https://bugs.python.org/issue39
Steve Dower added the comment:
> PR branches on the main repository are discouraged.
I had another reason, which was that I needed to trigger a signed build of the
separate Tcl and Tk binaries. This is much easier (and safer) to do from the
main repo than my own fork.
But yeah, in gene
Steve Dower added the comment:
I'm closing this as external, as the canonical source for platform tags is now
the packaging library.
Either pip and/or distlib should switch to using that library or the logic
included (which I'll note is considerably more complex - but also
Steve Dower added the comment:
mimetype uses the system database for MIME types, which on Windows is
application/x-zip-compressed, so this is by design.
If you have a need to specify a MIME type that does not match the system you're
on, you'll need your own logic.
--
Steve Dower added the comment:
I have no objection to bumping it further, provided someone has fixed
importlib.resources on 3.10 (i.e. my backported tests pass)
--
___
Python tracker
<https://bugs.python.org/issue41
New submission from Steve Dower :
Original report:
https://twitter.com/Steve_Casselman/status/1313564671652159488?s=20
A user received MSI error 1926 in a popup dialog, text reading:
> Could not set file security for file
> 'C:\Users\sc\AppData\Roaming\Microsoft\Installer\
Steve Dower added the comment:
Suspicious points:
* error 0 means success, so it shouldn't be popping up an error
* 'Installer\' is a directory, not a file
* %AppData%\Microsoft\Installer is normally readable by Everyone, but only
writable by Administrators, even in the user&#x
Steve Dower added the comment:
> I was able to reproduce the error dialog by changing the owner of the
> "Installer" folder to the current user and removing the two DACL entries that
> grant access to Administrators and SYSTEM.
Yeah, I'd assumed the most likely c
Steve Dower added the comment:
Yep, unless this directly impacts normal distutils usage, it's not going to be
fixed (regardless of the formal deprecation proposal, we already have a policy
of minimizing changes to distutils).
You could submit a patch to Bugzilla to switch th
Change by Steve Dower :
--
resolution: -> not a bug
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.org/issue41952>
___
___
Change by Steve Dower :
--
resolution: -> third party
stage: -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.org/issue41917>
___
___
Change by Steve Dower :
--
versions: +Python 3.10 -Python 2.7, Python 3.5, Python 3.6, Python 3.7
___
Python tracker
<https://bugs.python.org/issue28343>
___
___
Steve Dower added the comment:
I got the source repo update, and poked the PR to get it to re-run. If it's all
clear, I have no opposition to someone else doing the final merge.
--
___
Python tracker
<https://bugs.python.org/is
Steve Dower added the comment:
New changeset bfe6e03cd6931813dd61b50f5fdf7d8a8848f4cd by Erlend Egeberg
Aasland in branch 'master':
bpo-41557: Update Windows installer to use SQLite 3.33.0 (GH-21960)
https://github.com/python/cpython/commit/bfe6e03cd6931813dd61b50f5fdf7d
Steve Dower added the comment:
Thanks for figuring that out, Eryk.
Probably we should just update venv to do a realpath(sys._base_executable) to
handle the venv-from-symlinked-venv scenario.
Though I'd also be quite happy to just disallow that entirely (as we used to?).
If you e
Steve Dower added the comment:
This is due to how Windows resolves relative paths when creating a new process.
It *always* looks in the current application directory first, which with this
setup will be the original Python executable rather than the venv redirector.
The best fix is to pass
Steve Dower added the comment:
Thanks for doing the PR, Paul.
Honestly, I'd rather document in subprocess that Popen("python") is not a good
idea, and should use either sys.executable or shutil.which depending on what
you mean. This is a platform-independent gotcha, as it
Steve Dower added the comment:
> The bit that I *do* think is a venv gotcha is that it's entirely reasonable
> for a user to expect that if they run path\to\venv\Scripts\python.exe, then
> their Python script will be run by that executable.
This is still true, though, as muc
Steve Dower added the comment:
Yep, looks like the extra entry can just be removed. Guess we missed it in
issue4963
Care to submit a PR?
--
versions: +Python 3.10 -Python 3.7
___
Python tracker
<https://bugs.python.org/issue42
Steve Dower added the comment:
> I'm not asking that we guarantee any behaviour, or that we commit ourselves
> to anything. Just that we note that people ought not to be making a specific
> assumption
Which is a fine intent, it's just the running assumption is
Steve Dower added the comment:
Python 3.7.9 was a special case, as mentioned on the download page:
https://www.python.org/downloads/release/python-379/
--
___
Python tracker
<https://bugs.python.org/issue42
Steve Dower added the comment:
Yeah, looks done.
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<https://bugs.python.or
Steve Dower added the comment:
New changeset 985f0ab3ad5e8e9a8d7fc53026c38390b1f2b466 by Steve Dower in branch
'master':
bpo-39107: Updated Tcl and Tk to 8.6.10 in Windows installer (GH-22405)
https://github.com/python/cpython/commit/985f0ab3ad5e8e9a8d7fc53026c383
Steve Dower added the comment:
If you're building through setuptools, then it's a setuptools bug (they own
distutils going forward). https://github.com/pypa/setuptools
If you're not, please try with setuptools in the picture. It shouldn't require
any more than just
Steve Dower added the comment:
Proposed alternative based on Eryk's:
For maximum reliability, use a fully-qualified path for the executable.
To search for an unqualified name on :envvar:`PATH`, use
:meth:`shutil.which`. On all platforms, passing :data:`sys.execu
Change by Steve Dower :
--
pull_requests: +21738
pull_request: https://github.com/python/cpython/pull/22779
___
Python tracker
<https://bugs.python.org/issue41
2201 - 2300 of 5794 matches
Mail list logo
