Steve Dower added the comment:
Yeah, that's the sole buildbot currently running VS 2015. I'm expecting to have
more after VS 2015 RC is released, since that will be "basically finished".
Until then, I'm also regularly building with the latest internal versions and
tra
Steve Dower added the comment:
It looks like #1088716 is where the change was made a bit over 10 years ago.
Adding Paul to see if he has any further insight, but I'm pretty sure at that
point there was no 64-bit Python (there certainly wasn't a 64-bit Windows that
people were u
Steve Dower added the comment:
I assume you're referring to normal_startup and startup_nosite in perf.py at
h.p.o/benchmarks? Handy to know about (I need to explore our top-level repos
more often, obviously), but probably still not going to measure time in the
Windows PE loader as accur
New submission from Steve Dower:
The installer for 3.5 will include an option to install debug binaries
alongside the release ones, so we should update extending/windows.rst to
explain that and generally describe how to build an extension. It should also
refer to python3.dll at least once
Steve Dower added the comment:
This is probably also a good place to refer to the changes in #22980 and how
they can help when building extensions outside of distutils.
--
___
Python tracker
<http://bugs.python.org/issue23
Changes by Steve Dower :
--
resolution: -> fixed
stage: patch review -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.or
Steve Dower added the comment:
Updated patch, and hopefully it will make it into review this time.
I deliberately excluded the image for the using/windows.rst documentation but
you can see it at http://imgur.com/CdQaBmp
--
Added file: http://bugs.python.org/file37838/23260_3.diff
Changes by Steve Dower :
--
nosy: +larry
___
Python tracker
<http://bugs.python.org/issue23260>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Steve Dower:
When we completely switch Windows builds over to VC14, we're going to encounter
some new assert dialogs from the CRT. These won't appear in release builds, but
because the buildbots run tests against debug builds they will get in the way.
A numbe
Steve Dower added the comment:
I updated and tested your patch - basically making posixmodule.c use the new
_Py_stat_struct throughout instead of the Win32 definition. The tests run fine
on my VC14 machine.
Should _Py_fstat be conditioned on Py_LIMITED_API in some way, since it'
Changes by Steve Dower :
Added file: http://bugs.python.org/file37845/py_fstat_2.patch
___
Python tracker
<http://bugs.python.org/issue23152>
___
___
Python-bugs-list m
Steve Dower added the comment:
Huh, okay, looks like the patch still isn't sufficient (I forgot to put -uall
when testing it...) - there are calls in fileio.c that need changing too.
I'll try and do a thorough sweep of calls to fstat and post another patch in
the next
Changes by Steve Dower :
--
nosy: +steve.dower
___
Python tracker
<http://bugs.python.org/issue22286>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
Looks like it was only the _io module needing more updates. New patch attached.
--
Added file: http://bugs.python.org/file37888/py_fstat_3.patch
___
Python tracker
<http://bugs.python.org/issue23
Steve Dower added the comment:
This is fixed in VS2015 last time I tried it. I've proposed workarounds for
this before, but it's never affected any official build do there's never been
any traction. The fstat patch is normally the check that I use.
I don't know what Tomasz
Steve Dower added the comment:
It's not supported for building Python, is what I meant. I don't know what the
status is for Microsoft support of VS2012, but I'd expect it's security fix
only by now.
Being able to support building Python with alternative compilers is a task
Steve Dower added the comment:
I'll make the change for Windows.
--
___
Python tracker
<http://bugs.python.org/issue23212>
___
___
Python-bugs-list m
Changes by Steve Dower :
--
resolution: -> not a bug
status: open -> closed
___
Python tracker
<http://bugs.python.org/issue23258>
___
___
Python-bugs-list
Changes by Steve Dower :
--
versions: -Python 3.5
___
Python tracker
<http://bugs.python.org/issue20983>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Steve Dower :
--
versions: -Python 3.5
___
Python tracker
<http://bugs.python.org/issue19955>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
Definitely a dup, though I don't have the number handy. There's a patch on the
other issue waiting for a mimetypes maintainer to step up.
--
___
Python tracker
<http://bugs.python.o
Steve Dower added the comment:
Looks like a test ordering issue. When test_imageop runs before test_py3kwarn,
the warning has presumably already been shown once and so isn't triggered when
the test is looking for it (compare build 3092 linked above with 3093, which
passed).
I haven
Steve Dower added the comment:
It's not supported. You'll need to get the Python 2.7 source code and rebuild
the binaries under Debug.
Python 3.5 will probably have the option to download and install debug versions
of the binaries, but Python 2.7 won't
Steve Dower added the comment:
You'll also need to change your project to use the release version of the C
Runtime library and undefine _DEBUG throughout, otherwise you'll get conflicts
in things like memory allocators and alignment. It's not quite as simple as
choos
Steve Dower added the comment:
Afraid not. The closest you can get is building in Release with full debug
symbols and no optimisations, which should al lest get you decent debugging.
However, you won't get the extra memory check patterns or assertions throughout
your
Steve Dower added the comment:
Got about as many reviews as I expected, but the builds have been going fine
and I want this checked in before alpha, so in it goes :)
--
resolution: -> fixed
stage: -> resolved
status: open -> closed
_
Steve Dower added the comment:
As Andrew says, the patch is fine for the bug, but the PyCFunction_New function
is certainly part of the stable ABI and needs to be restored.
--
___
Python tracker
<http://bugs.python.org/issue21
Steve Dower added the comment:
We're already tracking this at #19143, where I've got a future proofed version
of the function. Thanks for the prod though, I'll try and get some action going
again.
--
resolution: -> duplicate
status: open -> closed
superseder: -
Steve Dower added the comment:
(Was reminded about this by #23417)
Any word on the back compat issues in platform.py? Can we make the version in
3.5 only work with Vista+, or do I need to handle XP in case someone takes the
file onto an earlier Python version
Steve Dower added the comment:
Including the one shipped in the stdlib? Python 3.5 no longer supports XP.
--
___
Python tracker
<http://bugs.python.org/issue19
Steve Dower added the comment:
@Marc-Andre: Okay, I'll make it easier to backport. You're listed as maintainer
- would you like me to post updates here and let you merge them in?
@Mark - yep, I'm aware of that. That's how I know that 3.5 doesn't support XP
anymore :)
Steve Dower added the comment:
You should be able to apply the patch to 3.5.0a1 and test it that way easily
enough. Other than the doc changes, it's just .py files. (Building 3.5 with
earlier VS versions is also okay right now.)
--
___
P
Steve Dower added the comment:
How do pip & co. handle the change? I assume they just look in sysconfig and
don't think too much about the path...
--
___
Python tracker
<http://bugs.python.or
Steve Dower added the comment:
Have you previously run ``build.bat -e`` from the command line? That's the best
way to make sure you're ready to work inside VS.
(And technically it's allowed to happen, since VS 2013 is not supported and our
build can do whatever it likes :) But
Steve Dower added the comment:
These are warnings about implicit downcasting in a 64-bit build. They're not
fatal to the build or execution (as far as we know), and as nice as it would be
to fix them all, "good" fixes may introduce behaviour changes (for example, new
overflow
Steve Dower added the comment:
Adding 'socket' nosy list, as Antoine may have a stronger opinion about
changing _socket to avoid these warnings - they aren't all in Windows-exclusive
code. Or I'm happy to add the suppression to the build if that's the better
option.
Steve Dower added the comment:
The newer attachment should handle back to Windows 2000. Unfortunately, I have
no machines to test this on, but all of the API calls existed at least on XP
and the version numbers are correct.
It runs correctly back to Python 2.7, works on 2.6 but gives the
Steve Dower added the comment:
Patch attached.
Interesting, once you declare Vista as the minimum API set, WinSock claims to
support CMSG_LEN which enables sendmsg() et al support in _socket. Since the
WinSock API bears little relation to the POSIX(?) API, it doesn't build, so
I&#x
Steve Dower added the comment:
Anyone interested in reviewing this patch?
--
assignee: -> steve.dower
stage: -> patch review
type: -> crash
___
Python tracker
<http://bugs.python.or
Steve Dower added the comment:
Attached a patch with options 1 and 2 implemented - both are fairly trivial.
Any thoughts/preferences?
--
assignee: -> steve.dower
keywords: +patch
Added file: http://bugs.python.org/file38104/23314 Options.pa
Steve Dower added the comment:
It may be useful if you want an error rather than the 32-bit version, though
there are other ways to check that if it's critical and it probably is better
in the script.
Patches welcome :)
--
___
Python tr
Steve Dower added the comment:
Wow, I didn't expect that so quickly :)
I'll check these out as soon as I can, but they look okay from a quick glance
on my phone.
--
___
Python tracker
<http://bugs.python.o
Steve Dower added the comment:
_Py_verifyfd has to go away, unfortunately. It requires inside knowledge of the
exact CRT version, and with VC14 the CRT will automatically upgrade so that
we're always using the latest.
I've gotten a function added to the CRT to make it unnecessary f
Steve Dower added the comment:
Do you mean pip.exe? Does running 'python.exe -m pip install ...' work for you?
--
___
Python tracker
<http://bugs.python.o
Steve Dower added the comment:
EBADF will still be returned; _PyVerify_fd is only there to prevent the
assertion dialogs in debug builds. Release builds will not need _PyVerify_fd at
all (though it's public, so it will remain, but it won't be necessary to
protect calls into the
Steve Dower added the comment:
In that case, this should be reported to the setuptools project, since they are
responsible for creating the pip.exe launcher when pip is installed.
--
___
Python tracker
<http://bugs.python.org/issue21
Steve Dower added the comment:
To be clearer (while still respecting the confidentiality agreement I'm under),
previously this code would (if _DEBUG) display an assertion dialog and
(regardless of _DEBUG) terminate the process:
close(fd); // succeeds, assuming a good fd
clo
Steve Dower added the comment:
Other than my one query on the review, code_ssize_t_2.patch.patch looks good to
me.
--
nosy: +steve.dower
___
Python tracker
<http://bugs.python.org/issue18
Steve Dower added the comment:
int_overflows.patch looks good to me.
I really appreciate that these patches have been done properly too - I've seen
far too much code where people just throw in casts to silence the warnings.
This is why I like working on P
Changes by Steve Dower :
--
nosy: +steve.dower
___
Python tracker
<http://bugs.python.org/issue9566>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
I don't even know what pgen is or why I'd want to build it...
If running it updates a file that would trigger a rebuild, then I'd really like
it to copy what _freeze_importlib currently does as far as writing to a
temporary file and conditionally
Steve Dower added the comment:
expanduser() sounds like the better choice here.
--
___
Python tracker
<http://bugs.python.org/issue20523>
___
___
Python-bug
Steve Dower added the comment:
Already reported at https://github.com/pypa/pip/issues/1889
--
resolution: -> third party
status: open -> closed
___
Python tracker
<http://bugs.python.org/i
Steve Dower added the comment:
I can't reproduce this easily. Could you delete the externals/tcltk folder and
the Debug_VC12 folders under tk.../win, tix.../win and tcl-core.../win, rebuild
and capture the output from the VS Output window? Apparently something in one
of these ext
Steve Dower added the comment:
Hmm, so maybe it only happens after you build the 64-bit version first. But
apparently it's more complex than that, since that doesn't repro either. When I
get some more time I'll run though and double check all the rebuild conditions,
but it has
Steve Dower added the comment:
Windows is probably the only platform that detects when _freeze_importlib makes
a change and aborts the build :)
I'd be interested to see what the diff is though. My guess is some of the
recent marshal changes may have affected the o
Steve Dower added the comment:
It's possible, but the eol extension is the best way to handle it.
Doing the comparison is relatively easy (we can just strip all '\r' and '\n'
characters and compare), but if it has changed, it's much harder to update the
new file
Steve Dower added the comment:
I don't think there are any tests for the launcher at all, though it would be
fairly simple to write a Python script that runs ``py -c "import sys;
print(sys.prefix)"`` and checks the output.
The patch looks fine to me, once I noticed that v
Steve Dower added the comment:
The changeset will skip updating importlib.h if the only change is line
endings, but if there are other changes it will be updated to a CRLF file. This
will be fine if the eol extension is enabled, and h.p.o will reject pushes that
have CRLF line endings
Steve Dower added the comment:
I can't say why 2.7 doesn't use _fseeki64, but 3.5 certainly does.
Possibly it's a significant change of behaviour that would break backwards
compatibility? Making a currently working call raise new exceptions is
certainly worth double-
Steve Dower added the comment:
Thanks. I always forget about those.
--
___
Python tracker
<http://bugs.python.org/issue23437>
___
___
Python-bugs-list mailin
Steve Dower added the comment:
I guess it's okay, but I have literally zero knowledge or experience with the
msilib module. Martin is still maintainer for that, as far as I know.
--
___
Python tracker
<http://bugs.python.org/is
Steve Dower added the comment:
There's nothing wrong with that line, it's probably the Tk or Tcl build
failing. Zach may have some ideas, or I'll play with it some more in the next
few days.
--
___
Python tracker
<http://bugs.pyt
Steve Dower added the comment:
Is there a difference if you do open(..., 'w')? It's a different enough
operation that it may have a different error.
--
___
Python tracker
<http://bugs.pyt
New submission from Steve Dower:
This file is necessary for _sqlite3 to be importable with the debug binaries.
--
assignee: steve.dower
components: Installation, Windows
messages: 236204
nosy: steve.dower, tim.golden, zach.ware
priority: normal
severity: normal
status: open
title
Steve Dower added the comment:
Patch needs updating, and I'd expect pyport.h to start by checking if
HAVE_PY_SSIZE_T is defined. Otherwise, we may try to re-typedef Py_ssize_t.
--
___
Python tracker
<http://bugs.python.org/is
Steve Dower added the comment:
Last call before I let the buildbots be the reviewers :)
--
priority: normal -> critical
___
Python tracker
<http://bugs.python.org/issu
Steve Dower added the comment:
The docs are pretty clear about using Popen instead of os.popen, and people
using popen have likely worked around it already, so we're more likely to break
them by changing it now.
I vote to close.
--
___
P
Changes by Steve Dower :
--
resolution: -> fixed
status: open -> closed
___
Python tracker
<http://bugs.python.org/issue14559>
___
___
Python-bugs-list
Steve Dower added the comment:
Not going to happen. The desktop is for users, not installers, and apps
installing shortcuts all over it is one of the reasons we got the Start Screen
in Windows 8 (where only users can add items - no way for installers to do it).
Besides, such an option would
Steve Dower added the comment:
Not sure if this still affects 2.7, but it doesn't affect 3.5 anymore.
If one of the buildbot owners confirms that 2.7 is fine, I'll close.
--
___
Python tracker
<http://bugs.python.o
Steve Dower added the comment:
Sounds good to me.
--
resolution: -> wont fix
status: open -> closed
___
Python tracker
<http://bugs.python.org/i
Steve Dower added the comment:
I'm not interested in fixing the 2.7 installer, but patches are welcome.
--
___
Python tracker
<http://bugs.python.org/i
Steve Dower added the comment:
I haven't looked closely at the rest, btw. Will try and find time today or
tomorrow.
--
___
Python tracker
<http://bugs.python.org/is
Steve Dower added the comment:
The installer changes look fine, though there shouldn't be PATH related strings
in exe/, so there's probably a separate issue there. Wix doesn't really have
any editors that aren't just XML editors.
We probably want to update the Content
Steve Dower added the comment:
Either "application/zip" or "application/x-zip-compressed", I'm not sure
exactly what the difference is, but the default .zip association has the latter.
The CRC error may be your machine or it may be because we're currently running
Steve Dower added the comment:
Looks like the .rc files should actually be UCS-2, since that's how the strings
are going to be stored into the executables.
If I rename the .h file to .h_ (and change the encoding to UCS-2 with BOM),
will it be ignored by argument c
Steve Dower added the comment:
Digging around the likely encodings to be running on Windows machines, it looks
like 0xA9 is always the right symbol, so that change should be fine.
--
___
Python tracker
<http://bugs.python.org/issue23
Steve Dower added the comment:
The caller to getfilesize is only using it to check whether it's small enough
to load the file into memory all at once, so "too big" is an okay response
(that function is in marshal.c and not used anywhere else).
The error label just retu
Steve Dower added the comment:
Attached a patch to 3.5 that resolves this, and I'll backport to 3.4. I haven't
got access to Windows 7 or 8 right now to test it, but it's fine on Vista
without the patch and 8.1 with the patch.
It'd be great if people could help check exac
Steve Dower added the comment:
Having run some more tests, it may be that the only regular problem is in the
test for inherited file descriptors.
I've attached a patch for tf_inherit_check.py that will prevent assert dialogs.
It's not pretty, but it avoids touching the interpreter
Steve Dower added the comment:
Thanks, good catch.
I'm not entirely sure how we ended up in this state, since the background is
actually the system colour and the text is currently hardcoded, but I'll go
through and make sure that the system colour indexes are used throughout.
Changes by Steve Dower :
--
resolution: -> fixed
stage: -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.org/issue24293>
___
Steve Dower added the comment:
I understood it only disallowed complaining about breaking changes without a
deprecation cycle :)
I'm sorry I didn't realize you were away. If you have examples of how
subclassing this class (and not just CCompiler) is useful and does something
tha
Steve Dower added the comment:
That change was in for beta 1, so we would have noticed if we didn't get Mac
builds.
Something else has changed, probably some headers.
--
___
Python tracker
<http://bugs.python.org/is
Steve Dower added the comment:
Those macros are only included if Py_BUILD_CORE is defined, regardless of
platform (see Include/pyport.h).
Is it possible that's being undefined somehow?
--
___
Python tracker
<http://bugs.python.org/is
Steve Dower added the comment:
Looking at a `grep PY_CORE_CFLAGS`, that sounds reasonable to me.
I assumed that all core files were already being compiled with Py_BUILD_CORE
(they certainly are for Windows), so this seems to be an oversight for
timemodule.c
Steve Dower added the comment:
When do we build timemodule.c outside of core?
--
___
Python tracker
<http://bugs.python.org/issue24244>
___
___
Python-bugs-list m
New submission from Steve Dower:
For 3.5 32-bit installs we register in a different key to avoid collisions with
64-bit installs, but now the 3.4 launcher can't find us.
If you have installed 3.4 for all users (likely) and then install 3.5
just-for-me (also likely), then you'll c
Steve Dower added the comment:
I didn't mention it initially, but the harm it can do is overriding newer
versions. The 3.5 and later installers have proper versioning built in, but I
don't think that's possible with the 3.4 installer as well (due to MSI
limitations, not any de
Steve Dower added the comment:
It might be nice to have a test that tries to import all of the stable ABI at
build time so that the buildbots break if someone adds a new API and doesn't
think about what they've done.
As an aside, it'd be nice for the script to take the version
New submission from Steve Dower:
(This is a reminder to myself, unless someone gets a patch in sooner than I get
to it.)
Currently if you click "Customize" in the installer and then Next/Install
without changing any settings, you get a default per-machine installation.
However, if
Steve Dower added the comment:
Reopening this because the lack of a background colour means that the progress
text accumulates while installed. I'll get to this sooner or later, but b2 is
probably going to have this issue.
--
resolution: fixed ->
stage: resolved ->
sta
Changes by Steve Dower :
--
resolution: -> fixed
stage: -> resolved
___
Python tracker
<http://bugs.python.org/issue24317>
___
___
Python-bugs-list
Changes by Steve Dower :
--
status: open -> closed
___
Python tracker
<http://bugs.python.org/issue24317>
___
___
Python-bugs-list mailing list
Unsubscrib
Changes by Steve Dower :
--
resolution: -> fixed
stage: -> resolved
status: open -> closed
___
Python tracker
<http://bugs.python.org/issue24293>
___
Steve Dower added the comment:
Already fixed this in issue24293, which is what caused it. Unfortunately, I
didn't get it done in time for beta 2.
Thanks for the report though.
--
resolution: -> duplicate
stage: -> resolved
status: open -> closed
superseder: ->
Changes by Steve Dower :
--
nosy: -steve.dower
___
Python tracker
<http://bugs.python.org/issue16991>
___
___
Python-bugs-list mailing list
Unsubscribe:
Steve Dower added the comment:
Going from void to int is fine for when the return value is ignored. However,
those who check the return value will see garbage on earlier versions.
As always, safest to add a new function here. Or make the stable version
continue to be void - it won't a
Steve Dower added the comment:
Which revision? Can you post a hash or issue number instead - in my repo r46496
is from a couple of years ago...
--
___
Python tracker
<http://bugs.python.org/issue24
4701 - 4800 of 5794 matches
Mail list logo
