Ned Deily added the comment:
This is a duplicate of Issue18458. The problem is due to an incompatible update
to the GNU readline compatibility layer of the editline shared library
(libedit) in OS X 10.9. The fix for this will be in the upcoming Python 2.7.6
and 3.3.3 maintenance releases
Ned Deily added the comment:
Robert, I appreciate your effort but I do not think it is good practice to
recommend downloading and installing binary bits from the bug tracker from an
uploaded file. Also, at a cursory examination, the readline.so supplied is not
exactly a drop-in replacement
Ned Deily added the comment:
This is a duplicate of Issue18458. Until 2.7.6 is available for download, a
workaround is to disable readline processing by moving _readline.so as
suggested in the other issue.
--
nosy: +ned.deily
resolution: -> duplicate
stage: -> committed/re
Changes by Ned Deily :
Added file: http://bugs.python.org/file32324/patch_readline_issue_18458.sh
___
Python tracker
<http://bugs.python.org/issue18458>
___
___
Python-bug
Ned Deily added the comment:
Until the updated readline module is available in python.org 2.7.6 and 3.3.3
maintenance releases, here is a script that will check python.org versions of
Python 2.7.x, 3.2.x, 3.3.x, and 3.4.x for interactive crashes and will patch
around the problem as needed
New submission from Ned Deily:
On OS X 10.9 Mavericks, there is a problem with Aqua Cocoa Tk 8.5 (8.5.15 or
earlier) when run in 64-bit mode, the default for most applications. The
symptoms are that Tk widgets may not be properly drawn or updated on the screen
unless some HID activity takes
Ned Deily added the comment:
Presumably you are running on OS X 10.9 Mavericks. If so, this is a duplicate
of Issue18458. Please see the available workaround documented there until
Python 2.7.6 is released.
--
nosy: +ned.deily
resolution: -> duplicate
stage: -> committed/re
Ned Deily added the comment:
Barry: yes, 2.6 is affected. See discussion on python-dev.
--
___
Python tracker
<http://bugs.python.org/issue18458>
___
___
Pytho
Ned Deily added the comment:
I built Python at 3fba718e46e5 on both 10.8 and 10.9 which was just before your
changes for this issue went in, Mark.
On 10.8: test_cmath fails and:
>>> math.log1p(-0.0)
0.0
And, indeed, on 10.9, it passes and:
>>> mat
Ned Deily added the comment:
"nobody with Xcode 5 can build C extensions for python.org Python installations"
Andrew, can you elaborate on what problems you or others are seeing? Xcode 5
should not be an issue for users of the current python.org Pythons. When
building extension m
Ned Deily added the comment:
Committed to default for release in 3.4.0. With the consent of the release
managers, this change has been backported to 2.7 (for release in 2.7.6) and 3.3
(for 3.3.3) in support of Issue15663.
--
resolution: -> fixed
stage: patch review -> com
Ned Deily added the comment:
Thanks for the report. The issue of setup.py not looking in an OS X SDK for
zlib.h was recently fixed in currently active branches: f3a2dab1623b (default),
2d8d5fea6194 (3.3), c207ac413457 (2.7).
By the way, there has been a lot of confusion about the
New submission from Ned Deily:
On OS X 10.8 or 10.9 when using Xcode 5 and attempting to build an extension
module using a universal Python that includes a PPC architecture (such as from
one of the python.org 32-bit-only installers), the extension module build can
fail with numerous compiler
Ned Deily added the comment:
The problem is fixed in 6cfb0f2815ce for release in 2.7.6 and e62a64507913 for
3.3.3. It was previously fixed in 3.4.0a4 by 73532f33fbb3.
--
resolution: -> fixed
stage: -> committed/rejected
status: open -&g
Changes by Ned Deily :
--
resolution: -> duplicate
stage: -> committed/rejected
status: open -> closed
superseder: -> interactive interpreter crashes and test_readline fails on OS X
10.9 Mavericks due to libedit update
___
Python tr
Changes by Ned Deily :
--
nosy: +rhettinger
___
Python tracker
<http://bugs.python.org/issue19414>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ned Deily added the comment:
With 3.3.3rc1 and Tcl/Tk 8.4, I'm now seeing on OS X 10.5 with ActiveTcl 8.4.20:
==
ERROR: test_merge (test.test_tcl.Tc
Ned Deily added the comment:
An update: release candidates for Python 3.3.3 and 2.7.6 are now available,
including binary installers for OS X, that include this fix and fully support
OS X 10.9.
http://www.python.org/download/releases/3.3.3/
http://www.python.org/download/releases/2.7.6
Ned Deily added the comment:
Update: release candidate ("rc1") packages for Python 3.3.3 and 2.7.6 have been
created. They are available here:
http://www.python.org/download/releases/3.3.3/
http://www.python.org/download/releases/2.7.6/
For 3.3.3 and 2.7.6, the python.org OS X 64-
New submission from Ned Deily:
Using current IDLE in 2.7 tip (as in 2.7.6rc1), selecting a file to open in the
IDLE menu or on the command line cause IDLE to either (depending on the
platform and invocation) crash or hang due to an uncaught exception in
idlelib/IOBinding.py:
/usr/local/bin
Changes by Ned Deily :
--
assignee: ronaldoussoren ->
resolution: -> duplicate
stage: -> committed/rejected
status: open -> closed
superseder: -> pydoc - stall when requesting a list of available modules in
the online help utility
___
Ned Deily added the comment:
Mark, I'm not sure I understand what you saw but the patch script will cause a
Python crash as part of its testing so that is to be expected. You should not
have to run the script using sudo. This script also only applies to Pythons
installed from python.or
Changes by Ned Deily :
--
components: +Build, Windows -Extension Modules
nosy: +loewis
___
Python tracker
<http://bugs.python.org/issue19450>
___
___
Python-bug
Ned Deily added the comment:
No, this does not fit the signature of Issue18458, which is fixed in 2.7.6rc1
anyway. On the other hand, it is not likely that this is a problem in Python
or its standard library. Pandas uses numpy and other third-party packages that
contain C code. It's
Ned Deily added the comment:
Googling around a bit, there was another report of this problem that includes
OS X crash stack traces. The crash there occurred in umath.so, a part of Numpy
and that leads to a now-resolved bug in Numpy.
https://github.com/pydata/pandas/issues/5396
https
Changes by Ned Deily :
--
nosy: +ned.deily, ronaldoussoren
___
Python tracker
<http://bugs.python.org/issue19478>
___
___
Python-bugs-list mailing list
Unsub
Ned Deily added the comment:
I don't have any experience using OS X sandboxing yet but I wonder whether
there are other instances of semaphores or other resources used elsewhere in
the standard library that would benefit from a common sol
Ned Deily added the comment:
Here are my review comments (sorry, no Rietveld):
ensurepip/__init__.py
-
Why have the separate _run_pip function? It would be simpler and less
confusing to just merge run_pip into bootstrap and eliminate the extra for
loop. Both depend on
Ned Deily added the comment:
I suppose the changes could be isolated to just ensurepip if it used a
temporary dir for the scripts and then moved the appropriate ones itself to the
standard scripts directory.
--
___
Python tracker
<h
Ned Deily added the comment:
Just for the record, here's what the installed ./lib/python3.4/site-packages
looks like after ensurepip has run:
-rw-r--r-- 1 nad pyd 119 Nov 1 15:20 README
drwxr-x--- 2 nad pyd 204 Nov 1 15:21 __pycache__/
drwxr-x--- 3 nad pyd 170 N
Ned Deily added the comment:
Without digging deeper, I'd be a little cautious about using a test involving
sys.executable and shutil.which. sys.executable in the past at least was not
always what you might think in certain cases with OS X framework ins
Ned Deily added the comment:
It's reproducible on OS X as well with a 32-bit Python 2.7.5 and a 64-bit
Python 2.7.6rc1. However, the example works OK if I start IDLE with no
subprocess (-n).
--
___
Python tracker
<http://bugs.py
Ned Deily added the comment:
With Cocoa Tk 8.5.15 or Cocoa Tk 8.6.1 on OS X 10.8.5, test_widgets.ButtonTest
crashes Tk:
test_image (tkinter.test.test_tkinter.test_widgets.ButtonTest) ... 2013-11-03
01:52:53.498 pytest_10.8[82465:f07] *** Assertion failure in -[NSBitmapImageRep
Ned Deily added the comment:
OS X 10.6 Snow Leopard shipped with a new version of Tk 8.5 that is mostly
unusable with IDLE and other Tkinter applications. If you are using the
Apple-supplied system Python on OS X, you should install a new version of
Python that allows use of a different
Ned Deily added the comment:
Double-clicking on a .py may not be opening the file with the right version of
IDLE. OS X maintains the associations between file types, based on file
characteristics like the extension (e.g. ".py"), and applications. To select
exactly which applicati
Ned Deily added the comment:
My apologies! What I just wrote is accurate but not the problem you are seeing.
I forgot that there is a problem with IDLE's Open command that was discovered
after the release of 2.7.6rc1. See Issue19426 It will be fixed in the final
release of 2.7.6
Ned Deily added the comment:
P.S. If you feel comfortable with using the command line, you *could* install
the fix for Issue19426 yourself into 2.7.6rc1 if you can't wait for the final
release. For the OS X python.org installers, something like this should work
from a user login
Ned Deily added the comment:
LGTM
--
___
Python tracker
<http://bugs.python.org/issue19481>
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.pyth
New submission from Ned Deily:
Running test_idle is supposed to cause the tests in Lib/idlelib/idle_test to be
run. Unfortunately, under 2.7, when test.regrtest is used as the test runner,
either directly or via "make test" (which buildbots use), the idlelib tests are
currentl
Ned Deily added the comment:
Terry, sorry the patch was confusing. Just FYI, if you are using "hg import"
particularly when reviewing patches, you should always be using "hg import
--no-commit" as explained in the devguide to avoid unintended changes. Whether
or not a me
New submission from Ned Deily:
Piet van Oostrum (on python-dev and elsewhere) wrote:
I tried to install matplotlib 1.3.1 on the release candidates of Python 2.7.6
and 3.3.3.
I am on Mac OS X 10.6.8.
Although the installation gave no problems, there is a problem with Tcl/Tk.
The new Pythons
Ned Deily added the comment:
That's unfortunate. It looks like matplotlib on OS X (darwin) has cpp code
that interfaces directly with Tk and, hence, needs Tk headers. Further, for OS
X (darwin), its setup.py assumes that Tkinter is linked with Tcl and Tk
installed as frameworks in /Li
Ned Deily added the comment:
Pickling for the RPC protocol between the GUI process and the interpreter
subprocess, which would explain why there is no problem when running idle -n
(no subproces)?
--
___
Python tracker
<http://bugs.python.
Ned Deily added the comment:
Thanks for the report. The link is now fixed.
--
nosy: +ned.deily
resolution: -> fixed
stage: -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs.python.or
Ned Deily added the comment:
This is a duplicate of Issue19426. The fix for it will be in 2.7.6 final.
--
nosy: +ned.deily
resolution: -> duplicate
stage: needs patch -> committed/rejected
status: open -> closed
superseder: -> Opening a file in IDLE causes a c
Ned Deily added the comment:
P.S. For users comfortable with the command line, there is an procedure
documented here for applying the fix to 2.7.6rc1 (2.7.6 final is expected
soon): http://bugs.python.org/issue19484#msg202062
--
___
Python tracker
Ned Deily added the comment:
Due to incompatibilities with some key third-party projects as documented in
Issue 19490 and the urgency of getting new maintenance releases out, the best
course of action is to revert built-in Tcl/Tk support for 3.3.3 and 2.7.6.
With the knowledge gained, I will
Ned Deily added the comment:
After further investigation and deliberation, I believe that we need to revert
the built-in Tcl/Tk from the OS X installers for 2.7.6 and 3.3.3. The
contributing factors:
1. As implemented, we now know that the built-in Tcl/Tk breaks source installs
and/or
Ned Deily added the comment:
Piet, yes, I've been thinking of how to do that. Unfortunately, it can only be
a hint since, in the case of an "installer" Python, there is no guarantee that
the header files on the build machine are available on the installed machine in
the same l
Ned Deily added the comment:
It is a duplicate. Fixed in 3.3.3 and 2.7.6.
--
resolution: -> duplicate
stage: -> committed/rejected
status: open -> closed
superseder: -> interactive interpreter crashes and test_readline fails on OS X
10.9 Mavericks due to li
Ned Deily added the comment:
P.S. See that issue for a workaround. Also release candidate installers for
3.3.3 and 2.7.6 are now available with final releases very soon.
--
___
Python tracker
<http://bugs.python.org/issue19
Ned Deily added the comment:
Looking at the compileall module, it appears that -f and options other than -b
have never (or, at least, for a long time, not) been supported when directories
or files are not supplied and it defaults to . Note
the call in main() to compile_path vs those to
Ned Deily added the comment:
Update: ActiveTcl 8.5.15.1 is now available and includes the fix for the 10.9
refresh problem documented here.
Unfortunately, the built-in versions of Tcl/Tk 8.5 included with the
pre-release python.org OS X 64-bit/32-bit x86-64/i386 installers for 3.3.3rc1
and
Ned Deily added the comment:
I wouldn't trust the NEWS items. I think the only reliable thing to do is diff
each file, unfortunately.
--
nosy: +ned.deily
___
Python tracker
<http://bugs.python.org/is
Changes by Ned Deily :
--
nosy: +orsenthil
___
Python tracker
<http://bugs.python.org/issue17518>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Ned Deily :
--
nosy: +orsenthil
___
Python tracker
<http://bugs.python.org/issue19547>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ned Deily added the comment:
b1244046f37a appears to have broken buildbots. See, for example:
http://buildbot.python.org/all/builders/AMD64%20Ubuntu%20LTS%203.x/builds/2984
--
___
Python tracker
<http://bugs.python.org/issue19
Changes by Ned Deily :
--
nosy: +ncoghlan
___
Python tracker
<http://bugs.python.org/issue19548>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ned Deily added the comment:
Thanks for the report. As of the 2.7.6 release, bz2-compressed 2.7.x tarballs
are no longer produced. The link on the download page now points to the
xz-compressed tarball, which offers better compression than bz2.
--
nosy: +ned.deily
resolution
Ned Deily added the comment:
If nobody else gets to it first, I'll do this in the next couple of days.
--
nosy: +ned.deily
___
Python tracker
<http://bugs.python.org/is
Ned Deily added the comment:
Issue1584 has implemented ways to specify non-standard Tcl and Tk include file
and library locations, both via ./configure and as "make" arguments. Since
there are many other issues open regarding specific aspects of setup.py and
since, in nine years,
New submission from Ned Deily:
In Issue 19490, Piet van Oostrum suggested:
"I think future versions of Python should add the relevant information about
how they are linked to Tcl/Tk in sysconfig. This would include the path of the
include files, the shared libraries and the tcl files.
Ned Deily added the comment:
As proposed above, the built-in Tcl/Tk support has been reverted from the OS X
installers for 3.3.3rc2 and for 2.7.6 final. Issue15663 will continue to track
changes for 3.4.0; the implementation there will have to change for 3.4.0b1.
I've also opened Issue
Changes by Ned Deily :
--
nosy: +georg.brandl
___
Python tracker
<http://bugs.python.org/issue19570>
___
___
Python-bugs-list mailing list
Unsubscribe:
Changes by Ned Deily :
--
assignee: -> ned.deily
___
Python tracker
<http://bugs.python.org/issue19553>
___
___
Python-bugs-list mailing list
Unsubscrib
Changes by Ned Deily :
--
keywords: +patch
Added file: http://bugs.python.org/file32646/issue19553.patch
___
Python tracker
<http://bugs.python.org/issue19
Ned Deily added the comment:
Here's a patch for review. It implements:
* a new configure option:
--with(out)-ensurepip=[=upgrade]
"install" or "upgrade" using bundled pip
* a new makefile macro ENSUREPIP that can be supplied to a
Changes by Ned Deily :
Removed file: http://bugs.python.org/file32646/issue19553.patch
___
Python tracker
<http://bugs.python.org/issue19553>
___
___
Python-bugs-list m
Changes by Ned Deily :
Added file: http://bugs.python.org/file32647/issue19553.patch
___
Python tracker
<http://bugs.python.org/issue19553>
___
___
Python-bugs-list mailin
Ned Deily added the comment:
I think the most serious problem is 3 since it directly affects the end user.
It would be good to have that fixed someway for 3.4.0b1. Resolutions to the
other items could wait.
--
___
Python tracker
<h
Ned Deily added the comment:
1. Yes, I think so.
2. OK
3. Ah, I was just using the default 1.4.1 from PyPI. With current dev from
github, the result is now pip3.4 like the wheel install. So once 1.5.x is
released, this shouldn't be a problem.
4. Sorry, I was imprecise. For "-m
New submission from Ned Deily:
There are currently various differences in the files installed by "make
altinstall" and "make install" for OS X framework (./configure
--enable-framework) and/or OS X universal (--with-universal-archs)
configurations. Some of the difference
Changes by Ned Deily :
--
keywords: +patch
Added file:
http://bugs.python.org/file32698/issue19649_bin_install_targets.patch
___
Python tracker
<http://bugs.python.org/issue19
Ned Deily added the comment:
Works for me using the python.org 3.3.3 64-/32-bit installer. Are you using
that Python 3.3.3? Please show the results of typing the following commamds
and substituting for "python3.3" whatever command name you are using that
causes the segfault.
py
Ned Deily added the comment:
There's something odd going on with that buildbot. The failures are from
Serhly's new tests. For 10.4 Tiger, the buildbot should be running Tk 8.4. It
does seem to have a third-party Tcl and Tk installed in /Library/Frameworks,
which is good, but the
Changes by Ned Deily :
--
nosy: +db3l
___
Python tracker
<http://bugs.python.org/issue19654>
___
___
Python-bugs-list mailing list
Unsubscribe:
https://mail.pyth
Ned Deily added the comment:
Christian is correct. /System/Library/Frameworks is the location of the Apple
supplied system Pythons and further Apple does not ship any version of Python
3. So you should figure out how your /System/Library got altered and restore it
to release state
Changes by Ned Deily :
--
keywords: +patch
nosy: +sbt
___
Python tracker
<http://bugs.python.org/issue19675>
___
___
Python-bugs-list mailing list
Unsubscribe:
New submission from Ned Deily:
http://buildbot.python.org/all/builders/AMD64%20Snow%20Leop%203.x/builds/568/steps/test/logs/stdio
==
ERROR: test_race (test.test_logging.HandlerTest
Changes by Ned Deily :
--
resolution: -> fixed
stage: patch review -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs.python.or
Ned Deily added the comment:
The proposed patch with minor changes is now pushed. I believe all of the
review points have either been resolved or are covered by separate pip issues
with the exception of item 8. For item 6 (permissions not being forced), I
decided that there are so many
Changes by Ned Deily :
--
nosy: +ronaldoussoren
___
Python tracker
<http://bugs.python.org/issue19677>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ned Deily added the comment:
>From the crash reports, it appears that you were trying to use IDLE with the
>Apple-supplied Tcl and Tk frameworks. OS X 10.6.x was the first release of the
>newer Aqua Cocoa Tk and it was very problematic. You should have seen a
>warning message whe
New submission from Ned Deily:
7b9235852b3b for Issue2927 moved and documents the previously undocumented
unescape function. Unfortunately, at least one third-party module, distlib,
depends on it. And distlib is now used by pip and needed for the ensurepip
feature of 3.4.
$ python3.4 get
Ned Deily added the comment:
A distlib issue has also been opened.
https://bitbucket.org/pypa/distlib/issue/36/distlib-uses-undocumented-and-deprecated
--
nosy: +vinay.sajip
___
Python tracker
<http://bugs.python.org/issue19
Ned Deily added the comment:
Since the message is coming from Cocoa layers of OS X, this is undoubtedly a Tk
issue, rather than a Python issue. What version of Tcl/Tk are you using:
ActiveTcl 8.5.15.1? Suggest you open an issue on the Tk bug tracker and/or the
ActiveState bug tracker
Ned Deily added the comment:
LGTM, thanks!
--
stage: patch review -> commit review
___
Python tracker
<http://bugs.python.org/issue19688>
___
___
Python-
Ned Deily added the comment:
If it is acceptable for the "Remove" option to be somewhat unpredictable in the
case where pip or setuptools was already installed and not by the installer,
would "python -m pip uninstall --yes pip setuptools" work? If not, should a
new issue
Ned Deily added the comment:
FWIW, it looks like Path.glob() is behaving the same as the default shell (bash
3.2 on OS X 10.6.8):
Python 3.4.0a4+ (default:cce14bc9b675, Nov 22 2013, 13:01:47)
[GCC 4.2.1 (Apple Inc. build 5666) (dot 3)] on darwin
Type "help", "copyright", &
Ned Deily added the comment:
test_pathlib now passes. I'd say it's fixed until someone claims otherwise.
--
___
Python tracker
<http://bugs.python.o
Ned Deily added the comment:
This change seems to have broken the OS X 10.4 Tiger buildbot:
_ssl.c:2240: error: 'struct x509_store_st' has no member named 'param'
_ssl.c:2253: error: 'struct x509_store_st' has no member named 'param'
_ssl.c:2257: err
Ned Deily added the comment:
10.4 is *very* old:
$ /usr/bin/openssl version
OpenSSL 0.9.7l 28 Sep 2006
If you kept around that version of the headers and libs, you'd probably catch
most of the problems.
--
___
Python tracker
Ned Deily added the comment:
FTR, on OS X, there is no standard mechanism provided by the Apple-supplied
installer program to uninstall binary packages. Thus, most third-party binary
distributions for OS X that use the OS X installer can only be uninstalled
manually unless they provide some
Ned Deily added the comment:
test_widgets.ButtonTest is still causing a Cocoa Tk 8.5/8.6 crash which also
crashes Python while running the test suite. The test should be fixed or
disabled.
--
___
Python tracker
<http://bugs.python.org/issue19
Ned Deily added the comment:
This problem also breaks the 32-bit OS X installer build.
--
nosy: +larry
priority: normal -> release blocker
___
Python tracker
<http://bugs.python.org/iss
Changes by Ned Deily :
--
priority: release blocker ->
resolution: -> fixed
stage: needs patch -> committed/rejected
status: open -> closed
___
Python tracker
<http://bugs.python
Ned Deily added the comment:
Also, FTR, the "install or upgrade pip" option is currently enabled by default
for the OS X installers.
--
___
Python tracker
<http://bugs.python.o
Ned Deily added the comment:
See msg202006 above.
--
___
Python tracker
<http://bugs.python.org/issue19085>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ned Deily added the comment:
You should be able to find the group and group name using the "Directory
Utility" app. The easiest way to launch it is from a shell with:
open "/System/Library/CoreServices/Directory Utility.app"
There is a command line utility, dscl, but it
New submission from Ned Deily:
As of 3.4.0b1, the following test failures are observed with Tk 8.5 on OS X
(python.org 64-/32- installer with ActiveTcl 8.5.15.1 on OS X 10.9):
==
FAIL: test_bitmap
5001 - 5100 of 6927 matches
Mail list logo
