Ronald Oussoren added the comment:
Todays version of the patch adds more tests, especially for exceptions in
__getdescriptor__ during the lookup of special methods. Those tests were added
because the C code couldn't get exceptions other than AttributeError before
this patch and can get
Ronald Oussoren added the comment:
What's your plan for installers for Py3.6?
In a world where backward compatibility is not an issue I'd definitely advocate
trying to move to some kind off .app as the installation.
That is: have a {SomeName}.app that contains the entire Python in
Ronald Oussoren added the comment:
GCC has a pragma and function attributes for changing the optimisation flags,
that could be used to disable the tail call optimazation here.
Something like the following (using a pragma):
#pragma GCC push_options
#pragma GCC optimize ("-fno-optimize-si
Changes by Ronald Oussoren :
--
nosy: +ronaldoussoren
___
Python tracker
<http://bugs.python.org/issue24844>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ronald Oussoren added the comment:
This appears to be an issue for an external library
<https://pypi.python.org/pypi/ctypeslib>, although one hosted at svn.python.org.
--
nosy: +ronaldoussoren
___
Python tracker
<http://bugs.python.org/i
Changes by Ronald Oussoren :
--
nosy: +theller
___
Python tracker
<http://bugs.python.org/issue24880>
___
___
Python-bugs-list mailing list
Unsubscribe:
Ronald Oussoren added the comment:
W.r.t. install_name_tool: macholib (PyPI project) contains functionality to do
this work in pure python.
I do not propose to include macholib in the stdlib, it is too special purpose
and not everyone that has provided patches is a CPython contributor
Ronald Oussoren added the comment:
W.r.t. the bus error on OSX: see issue #18049, the default stack size on OSX is
too small to reach the default recursion limit.
--
___
Python tracker
<http://bugs.python.org/issue25
New submission from Ronald Oussoren:
On OSX 10.4 or later the following functions are available:
size_t pthread_get_stacksize_np(pthread_t);
void* pthread_get_stackaddr_np(pthread_t);
These functions could be used to implement PyOS_CheckStack on OSX. The
advantage of this would be that it
Ronald Oussoren added the comment:
If I read the issue correctly the OSX 10.11 SDK doesn't include OpenSSL
headers.
There's not a lot we can do about that beyond documenting this in the build
instructions.
It might be interesting to look into using Apple's crypto APIs inst
Ronald Oussoren added the comment:
Hi,
This appears to be a description of an issue for
<https://pypi.python.org/pypi/virtualenv>, not a component of CPython or its
standard library.
The issue tracker for virtualenv is at
<https://github.com/pypa/virtualenv/issues>.
---
Ronald Oussoren added the comment:
Apple no longer ships headers for OpenSSL in OSX SDK as shipped with Xcode 7.1,
and I'd be surprised if that changes (without having any inside knowledge about
this).
W.r.t. adding configure option for OpenSSL: That would be fine although that
make
Ronald Oussoren added the comment:
Victor: read(2) and recv(2) are defined as returning all available data when
interrupted by a signal and might therefore not return an error. The same is
true for write(2): it returns the number of bytes written when there are any
bytes written (instead of
New submission from Ronald Oussoren:
https://emptysqua.re/blog/getaddrinfo-deadlock/ claims that getaddrinfo may
deadlock when using threads and fork on (amongst others) OSX due to using a
global lock.
That lock is used when getaddrinfo is believed to be not thread safe, see the
relevant
Ronald Oussoren added the comment:
#25924 is related to this, I filed this after reading the blog post. The lock
might not be necessary on OSX, and possibly on the other systems as well.
Yury: resetting the lock in the child should be safe because after the fork the
child only has a single
Ronald Oussoren added the comment:
The build fails while building the "Python Launcher" application. This is an
Objective-C project.
If GCC doesn't support the system headers on OSX 10.11 there's nothing the
Python project can do about that.
--
n
Ronald Oussoren added the comment:
The makefiles use CC throughout and don't look at and OBJC variable. Is that
variable a standard way to specify an ObjC compiler in makefiles?
Make's default ruleset for compiling .m files uses $(CC) and doesn't look at
$(OBJC).
I wouldn&#x
Ronald Oussoren added the comment:
The lack of o_cloexec could be caused by the SDK used to build the binary
installer. AFAIK _DARWIN_C_LEVEL is set automatically based on the deployment
target.
Ronald
Sent from my iPad
> On 11 feb. 2016, at 23:44, STINNER Victor wrote:
>
>
Ronald Oussoren added the comment:
This appears to be an issue with the QGis application and not a problem with
Python itself.
See the following URL for support on QGis:
<http://www.qgis.org/en/site/forusers/support.html>
Regards,
Ronald Oussoren added the comment:
The text appears to be correct as it is. What is says is that __init__ must not
return any value other than None and that is correct, you will get an exception
when you return a value that is not None.
>>> class C():
...def __init__(self):
Ronald Oussoren added the comment:
Why does this patch execv() the interpreter to set options? I'd expect it to be
possible to get the same result by updating the argument parsing code in
Py_Main.
--
nosy: +ronaldoussoren
___
Python tracker
Ronald Oussoren added the comment:
How can you create plist files that contain UID values using Apple's APIs?
The plist library is meant to be interoperable with files created using Apple's
APIs for creating plist files, and I didn't find an API that created UID valu
Ronald Oussoren added the comment:
Patch looks good to me.
--
___
Python tracker
<http://bugs.python.org/issue26709>
___
___
Python-bugs-list mailing list
Unsub
Ronald Oussoren added the comment:
I don't understand the explicit use of __eq__ in the bytes case. Why is that
needed?
--
___
Python tracker
<http://bugs.python.org/is
Ronald Oussoren added the comment:
Serhiy,
I slightly prefer the second patch, but either one would be fine.
The reason I asked about explicitly calling __eq__ is that this is an uncommon
pattern (at least for me).
Ronald
--
___
Python tracker
Ronald Oussoren added the comment:
There is a clear gap between two use cases for Python on OSX:
1) Integrate nicely with Apple stuff
2) Be useful as a development platform for software that will be deployed on
Linux machines.
Both use cases are valid, but have slightly different trade-offs
Ronald Oussoren added the comment:
On OSX the internal copy of libffi that's used is based on the one in PyObjC,
which in turn is based on the version of libffi on opensource.apple.com (IIRC
with some small patches that fix minor issues found by the PyObjC test
Ronald Oussoren added the comment:
1) I agree with Ned that the OSX behavior is not broken, it is different but
within spec. Python makes assumption about the format of locale names that
aren't universally valid.
2) We should be careful in using CFLocale. Those APIs are pa
Ronald Oussoren added the comment:
The test for type({}) is indeed wrong and should check for self._dict_type
instead. However, there needs to be a test before that change is made.
--
On the road, hence brief.
Op 24 apr. 2015 om 01:40 heeft Ned Deily het volgende
geschreven:
>
&g
Ronald Oussoren added the comment:
To react to myself: checking for self.dict_type might break users that pass in
a callable:
x = plistlib.load(fp, dict_type=lambda:{})
As Behdad memtioned testing that the type isn't list would be better:
if not isinstance(..., list):
That att
Ronald Oussoren added the comment:
I don't recall exactly why the universal build support is done the way it is,
adding it is too long ago.
The reason is likely no longer relevant with any reasonably up-to-date toolset.
The patch was created around the time x86 support was added to OS
Ronald Oussoren added the comment:
Note that these format characters can also be used outside of CPython.
--
nosy: +ronaldoussoren
___
Python tracker
<http://bugs.python.org/issue24
New submission from Ronald Oussoren:
The script below creates a basic PEP 420 style package with a single module in
it ('package.sub') and tries to import that module
With 3.5 the script runs without problems and prints 42 (as expected).
With a 3.5 (fresh checkout as of 2015-05-14
Ronald Oussoren added the comment:
According to [1] common symbols are not allowed in frameworks. I guess that's
why we added '-fno-common' to the linker flags.
[1]
https://developer.apple.com/library/mac/documentation/MacOSX/Conceptual/BPFrameworks/Concepts/FrameworkBinding.h
Changes by Ronald Oussoren :
Added file: http://bugs.python.org/file39384/issue-18378-py27.txt
___
Python tracker
<http://bugs.python.org/issue18378>
___
___
Python-bug
Changes by Ronald Oussoren :
Added file: http://bugs.python.org/file39385/issue-18378-py35.txt
___
Python tracker
<http://bugs.python.org/issue18378>
___
___
Python-bug
Ronald Oussoren added the comment:
Dimitry's patch looks good, I added my patch before checking if there already
is patch.
The only thing that might be cause discussion is when to accept 'UTF-8' as a
valid locale name. My patch only accepts in on OSX, while Dimitry'
Ronald Oussoren added the comment:
I can confirm that this fixes the problem I was having, both for the sample
code I posted and for the real codebase I extracted this from.
Thanks.
--
status: pending -> open
___
Python tracker
&l
Ronald Oussoren added the comment:
I'm in favour of using SOABI on OSX as well.
The build system currently uses "darwin" for the platform triplet, that's a bit
to minimal. I'd prefer to use "macosx" instead of "darwin" for the platform
na
Ronald Oussoren added the comment:
Ned: Keeping "darwin" as the platform tag is fine with me. It is slightly ugly
because it doesn't match the platform tag used by distutils/setuptools, but is
also something that most users won't use directly.
That said: there was some t
New submission from Ronald Oussoren:
PyObjC has an extension that's imported as "objc._objc". This works fine on
version upto the 3.5 beta (checkout from earlier today).
With that I get the following exception:
Python 3.5.0b1+ (default:7255af1a1c50+, May 25 2015, 11:46:
Ronald Oussoren added the comment:
The attached patch appears to fix the issue. This appears to be an off-by-one
error.
--
stage: -> needs patch
type: -> behavior
Added file: http://bugs.python.org/file39494/issue-24285.txt
___
Python t
Ronald Oussoren added the comment:
See issue 24285, I ran into the same issue as mention in msg244008. That issue
has a patch.
--
nosy: +ronaldoussoren
___
Python tracker
<http://bugs.python.org/issue24
Ronald Oussoren added the comment:
Does the upstream libffi work reliably on OSX by this time?
The bundled version of libffi was extracted from PyObjC, and that copy is
itself a patched version of the system libffi on OSX. The reason I don't use
the system libffi is both that I don&#
Ronald Oussoren added the comment:
Another reason to remove this feature: installing python 2.7.12 using the
installer on www.python.org breaks the system install of Python, likely because
of this issue.
I'm a bit sad that this feature has to go, but modern Python use (in parti
2401 - 2445 of 2445 matches
Mail list logo
