Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The patch looks good. It may also fix issue9197, which fails in a similar
fashion
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue11321
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Can you provide a patch, or show how to do it concretely?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue11253
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
--with-wctype-functions was removed in 3.2 (see issue9210, r84752)
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue11309
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Nose works correctly for me with multiprocessing. In a directory, I have:
== run_nose.py =
from nose import main
if __name__ == '__main__':
main()
== test_me.py
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Ah, is 'nosetests' a .exe file? A frozen executable?
In this case, can you try the solution proposed at
http://docs.python.org/library/multiprocessing.html#multiprocessing.freeze_support
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Using multiprocessing on Windows can be different; please read
http://docs.python.org/library/multiprocessing.html#windows especially the part
named Safe importing of main module.
On Windows, fork() does not exist, so a new
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
It seems as if the multiprocessing module is starting new Windows
processes by duplicating the command line of the original process.
It does not. The spawned processes use the command::
python.exe -c 'from multiprocessing.forking
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
This has changed in python 3, and is even documented:
http://docs.python.org/dev/py3k/whatsnew/3.0.html#text-vs-data-instead-of-unicode-vs-8-bit
(6th bullet)
Python 2.x could not be changed, for compatibility reasons.
--
nosy
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The encoding and errors arguments are probably useful, but why would you use
the codecs module?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
And this code will crash a debug build:
class X(long):
... def __oct__(self):
... return 'foo'.upper()
...
'%o' % X()
Assertion failed: buf[sign] == '0', file ..\..\Objects\stringobject.c, line 4059
--
nosy
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I found that adding infile._CHUNK_SIZE = 20 makes the test much faster -
'only' 5 times slower than 2.7.
--
nosy: +amaury.forgeotdarc, pitrou
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
This script is indeed unusable. Here is a proper patch file, tested by running
it on a few files.
--
keywords: +patch
nosy: +amaury.forgeotdarc
Added file: http://bugs.python.org/file20681/pdeps.patch
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
It's a duplicate of issue7980.
One workaround is to call strptime() once in the main thread.
--
nosy: +amaury.forgeotdarc
resolution: - duplicate
status: open - closed
superseder: - time.strptime not thread safe
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
For this kind of question you'd better ask on the comp.lang.python newsgroup.
The bug tracker is not designed to get help.
A remark though: to really debug the RuntimeError a C debugger is needed (and a
debug build of Python). You'd
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Here are the files I used to reproduce the crash.
See build.sh for the build command, there may be better ways...
--
Added file: http://bugs.python.org/file20631/simpletest.zip
___
Python
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Could this be some bizzarre bug in the quote parser of cmd.exe?
Yes, this is even documented with cmd /?:
[...]If you specify /c or /k, cmd processes the remainder of string and
quotation marks are preserved only if all
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I don't reproduce the issue. Could you run python under gdb and give us the C
backtrace?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Python objects can have two textual representations: see the difference between
str() and repr():
http://docs.python.org/tutorial/inputoutput.html
And indeed, Python3 chose to drop the 'L' suffix.
--
nosy: +amaury.forgeotdarc
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
trace.py has a hack for this: it uses gc.get_referrers() to crawl back from the
code object to the encloding class.
I would not trust this gc.get_referrers() though; it is probably slow, and
there are even some crashers related
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
No, the issue can be fixed by better checking the arguments.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8036
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
No, your change is in the read_directory() function, which reads the whole
archive the first time it's used.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10955
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
What about tools that builds one .zip file for all modules, like py2exe?
A cp437 decoder is not so ugly to implement in C. It's just a charmap.
--
___
Python tracker rep...@bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I can't reproduce the error. Do you have a script that shows the issue?
What is the complete traceback?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Ah, got it. It's about threading.Timer, which looks like a class, but is
actually a function.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10968
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
It seems to be by design: from the documentation:
http://docs.python.org/py3k/library/threading.html
Timer is a subclass of Thread, and a Thread subclass should only override
the __init__() and run() methods.
What are you trying to do
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
We are only talking about bootstrap-time importing of encodings modules.
Again, the whole zip central directory is loaded on first import. If the zip
file contains non-ascii filenames, nothing can be imported
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
This looks similar to issue10614
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10972
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
can't we raise a RuntimeError on a deadlock?
Deadlock detection is difficult, and probably impossible if the involved locks
don't use the same underlying mechanism. (A lock can be a pthread object, a
file opened with os.O_EXCL
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
os.stat() follows symbolic links. You probably want to use os.lstat() instead:
http://docs.python.org/library/os.html#os.lstat
--
nosy: +amaury.forgeotdarc
resolution: - invalid
status: open - closed
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
About the patch: Break out of this dependency by assuming that the path to
the encodings module is ASCII-only.
The 'path' here is the entry inside the zip file (and does not include the
location of the zip file itself), so the comment
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
import child_directory
Then it's certainly an effect of the import lock:
http://docs.python.org/library/threading.html#importing-in-threaded-code
In your case, the first call to encode('utf8') indirectly imports utf8.py,
while
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
setdefault() is a method, its arguments are evaluated then the function is
called. This is not a bug, and this behavior cannot change.
If you are trying to cache the computation of a function, you should try
memoizing techniques, like
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
It's a feature. Subprocess output is binary data, not text; and since python3,
the string type is now what python2 called unicode!
Please read
http://docs.python.org/py3k/whatsnew/3.0.html#text-vs-data-instead-of-unicode-vs-8-bit
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
+1 for this kind of tests.
But I would not have their source in the official Modules directory. I expect
that there will be several tests of this kind, each one with different modules
to import, functions to run, global settings
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
In distutils.command.config, config().try_run(body) is probably what we need.
Now, I don't know how to use it...
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
What happens when the archive contains both 'foo' and '../foo'? They seem to be
extracted at the same place.
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
r78942 is quite large unfortunately.
But just patching _elementree.c::xmlparser_setevents() should be possible.
This would at least fix the invalid event tuple error.
--
___
Python tracker
New submission from Amaury Forgeot d'Arc amaur...@gmail.com:
A short crasher::
class X(object): pass
del X.__abstractmethods__
All versions since 2.6 are affected.
The fix is probably simple: the setter function type_set_abstractmethods() in
typeobject.c should check for a NULL value
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The script unicode2.py uses the console STD_OUTPUT_HANDLE iff
sys.stdout.fileno()==1.
But is it always the case? What about pythonw.exe?
Also some applications may redirect fd=1: I'm sure that py.test does this
http://pytest.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
What makes you think it should be a boolean?
in http://docs.python.org/py3k/c-api/arg.html#Py_BuildValue b means byte
and is processed as a tiny integer.
Now, that's true that Py_BuildValue could have a format for boolean values
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
This is the right place to ask... but it will be faster if someone provides a
patch.
--
keywords: +easy
stage: - needs patch
___
Python tracker rep...@bugs.python.org
http://bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Even if python.exe starts normally, py.test for example uses os.dup2() to
redirect the file descriptors 1 and 2 to temporary files. sys.stdout.fileno()
is still 1, the STD_OUTPUT_HANDLE did not change, but normal print() now goes
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Is it even possible? Each time I tried, the only solutions involved an external
program like Dependency Walker.
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The code is in Python/dynload_win.c
To load an extension module mymodule.pyd, Python calls
LoadLibrary('/path/to/mymodule.pyd'); when it returns NULL, the code calls
GetLastError() (which returns 126 in this case) then FormatMessage
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
- unit tests are needed.
- Py_LONG_LONG should be used instead of long long, and the cast removed.
- The patch assumes that longlong == int64, but is it true on all platforms?
--
nosy: +amaury.forgeotdarc
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
no, DWORD is a 32-bit unsigned integer
http://msdn.microsoft.com/en-us/library/aa383751(VS.85).aspx
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
r87733 in pyexpat.c modified a call to PyErr_Format with a %zi format code.
This format does not seem to be supported. %zd should be used instead.
--
nosy: +amaury.forgeotdarc
___
Python
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
It's not fixed. range() now uses the tp_index slot, in weakrefs this becomes:
WRAP_UNARY(proxy_index, PyNumber_Index)
and indeed PyNumber_Index does not accept strings.
But try with time.sleep() instead; here the line
WRAP_UNARY
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Btw, I have a failed assertion in the test suite, with time.ctime(1e12):
File: loctim64.c
Line: 78
Expression: (*ptime = _MAX__TIME64_T)
This is a recent py3k, compiled with VS2005
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Here is a renewed patch against py3k, with a test.
--
Added file: http://bugs.python.org/file20237/timet_64-2.patch
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8278
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The only concern is that the test probably fails on FAT
The patch does not show it, but I've put the test under a
'if get_file_system(support.TESTFN) == NTFS'
--
___
Python tracker rep
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Committed r87666 (py3k), I'll backport to 3.1 and 2.7.
--
resolution: accepted - fixed
status: open - pending
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8278
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Merged into release31-maint (r87668) and release27-maint (r87669)
--
status: pending - closed
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue8278
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Normally you should never call __del__, OTOH the issue is the same with a class
like::
class A:
def close(self):
self.close()
def __del__(self):
self.close()
The problem is not with _infinite_ recursion, though
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Precision: with new-styles classes (or py3k) the limit is
PyTrash_UNWIND_LEVEL-2. This does not change anything to the problem.
--
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
But there are probably working usages with unicode strings out there. For
example, I've seen code like struct.pack('6sHHBBB', 'GIF87a', ...)
Do you suggest to make this 3.1 code stop working in 3.2?
In any case, the 'c' format should
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Since the Release Manager agrees with the change, I withdraw my objection.
I have three remarks to the patch:
- Some examples in the documentation should be fixed:
http://docs.python.org/dev/py3k/library/struct.html#examples
pack('ci
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Indeed
--
resolution: - duplicate
status: open - closed
superseder: - Avoid parsing pyconfig.h and Makefile by autogenerating
extension module
___
Python tracker rep...@bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Tests now fail on windows XP:
http://www.python.org/dev/buildbot/all/builders/x86%20XP-4%203.x/builds/3874/steps/test/logs/stdio
os.symlink() may raise NotImplementedError, and test.support.can_symlink()
should catch
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
At this point a feature change seems unlikely, but it's not too late to emit a
DeprecationWarning.
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The patch looks good, I only have stylistic remarks:
- We normally don't use windows-specific types in CPython code. Please use int
instead of BOOL. And C variables are usually lowercase, even module globals. I
suggest something like
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
With a release build of python, I have a similar behavior on both 2.5 and 2.7;
the messages are different though:
2.5: WindowsError: [Error -2147483645] One or more arguments are invalid.
2.7: WindowsError: exception: breakpoint
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
PyPy has exactly the same issue. PyPy's workaround is to have a custom version
of sysconfig.py, but this is really a hack.
If these config_vars are really determined at compile time, IMO they should be
built in the interpreter
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
What is 64-bit safe should be 32-bit safe, not only 31-bit safe
Not here. Python uses signed size_t for various lengths and sizes.
On win32 this only gives you 31 bits...
--
nosy: +amaury.forgeotdarc
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Err... in r87339 there is a typo in all occurrences of 'Py_ReprEntr':
Py_ReprEnter of course!
--
nosy: +amaury.forgeotdarc
status: closed - open
___
Python tracker rep...@bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Why was sys.setsegfaultenabled() omitted?
It may be useful to disable the handler from a script
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Normally the filename is part of the error message, as you can see below.
It's possible that some operations omit it, though. Which function were you
calling?
open('foo')
Traceback (most recent call last):
File stdin, line 1
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The soabi tag could be useful on Windows as well, for example if some package
maintainer chooses another version of the compiler, or a custom version of
MSVCRT.
And FWIW, PyPy currently uses --soabi=pypy-1.4 on all platforms: import
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
On Windows, newlines in text files are always translated. Please read
http://docs.python.org/tutorial/inputoutput.html#reading-and-writing-files
--
nosy: +amaury.forgeotdarc
resolution: - invalid
status: open - closed
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Maybe a global registry... implemented with a WeakKeyDictionary?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue1705520
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I happen to have the same change in my workspace. Please apply this patch, it
fixes the test suite when compiled with VS8.0.
--
nosy: +amaury.forgeotdarc
___
Python tracker rep
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Thanks for the explanations! These instructions should be turned into a patch.
Some questions already:
- Why is the generated pyconfig.h not good enough? Could we modify the
./configure script instead?
- the empty.c should remain empty
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
As a consequence, extensions built for the official
binaries would crash in the MingW build, and vice versa
IMO this is a use case for the new soabi tag...
--
nosy: +amaury.forgeotdarc
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Sure - it actually builds a python.exe which is fully working for me.
If you need a proof, please let me know, I have no problem uploading
it somewhere...
This is useless, of course. The build must be reproducible.
Here is what I
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
IMO this is a use case for the new soabi tag...
Unfortunately not: these tags are not supported on Windows.
So it's a argument for adding (optional?) soabi tags on Windows!
--
___
Python
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The ZIP format specification mentions only cp437 and utf8:
http://www.pkware.com/documents/casestudies/APPNOTE.TXT see Apeendix D.
Do zip files created on Japanese Windows contain some information about the
encoding they use?
Or do
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Hmm, not sure.
Try with
set PYTHONPATH=.\a;.\b
sys.path will contains the entries:
current_dir\a
current_dir\b
a multiplication seems necessary.
--
nosy: +amaury.forgeotdarc
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
So the presence of os.symlink depends on some dynamic privilege?
It seems to me that it's the first time in Python. For example, os.chroot() is
always available on Unix, even when the user is not root. Of course the call
will fail
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The value of get_config_var(SO) is the same as before, something like
.cpython-32.so by default on Linux. (see, I just moved the line SO=.$SOABI$SO
at the bottom of the patch).
In the C file dynload_shlib.c, I chose the names SO_BASE
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Let's test Linux at least, then.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10262
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I agree with the intent: package maintainers, or people compiling python for
their own need, should have the right to choose the suffix used by extension
modules.
I suggest another option though, one that directly sets the SOABI used
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Note that the soabi name isn't fixed but changes with other configure options.
The default value, yes.
But my proposal of a --soabi option would not respect this. The caller is
responsible for changing the --soabi value when he adds
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Here is a first patch that adds the --with-soabi option.
--with-soabi=yes or --with-soabi sets SOABI to the computed value
(cpython-32m when I tried it). This is the default behavior on Linux.
--with-soabi=no or --without-soabi
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
A .pdb is the program database that contain debug info, i.e. the mapping
between assembler positions and source lines, the description of data
structures, and other things that are necessary to debug a program
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I don't agree. issubclass(1, list) has always raised an exception.
--
nosy: +amaury.forgeotdarc
resolution: - invalid
status: open - closed
___
Python tracker rep...@bugs.python.org
http
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
issubclass is for relationship between classes. Did you consider isinstance()
instead?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10569
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Looks good, except that win32_error(link, NULL) should be called instead of
posix_error(). errno is not guaranteed to be correctly set by the win32 api,
and GetLastError() ususally gives more accurate errors
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Careful: the ctypes module is much slower with pypy than with cpython.
In any case, a C extension module is preferred because the (platform-dependent)
details of the liblzma library are defined in a .h header file, not as a ctypes
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
The patch uses the ANSI version, and converts the filename from unicode to
bytes; this will fail for names that cannot be encoded with the mbcs codec.
All other functions in the posix module first try the Wide version of the win32
API
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
For the record, this was finally fixed with issue2862: gc.collect() now clears
the free-lists during the collection of the highest generation.
--
nosy: +amaury.forgeotdarc
___
Python tracker
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
issue9200 already proposes a similar change to str.is* methods.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10521
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
But a strict bugfix should fix something. Is there something that did not
work before, and will work after this patch? IOW, how do you compile
posixmodule.c with MinGW and does it produce a working module?
Now, I *am* interested
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I doubt this fix will be enough to fix compilation with mingw32.
But IMO it goes in the right direction.
With this patch, #ifdef MS_WINDOWS has the meaning of linked with some
version of MSVCRT. Do we agree on this, or is another
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Since the patch does not completely fix the mingw32 build, I suggest to apply
it only on 3.2, and continue to work on mingw32 support there.
Only after we will be able to discuss whether all the changes can be backported
to 2.7
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
I reproduce the problem, see the call stack below.
The issue is in the io module: the function _bufferedwriter_flush_unlocked()
runs with the file mutex already acquired, but it calls PyErr_CheckSignals()
which may run any code
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Thanks for the patch, it looks good, I have 2 remarks though:
- under the name name could be replaced by under the given name.
- The *search_function*'s refcount is incremented by this function.
This information is not useful
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
pypy does have a GIL!
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10437
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
Looks good to me, except for a 'stings' near the end of the patch.
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue7828
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
How do you use it?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue10444
Amaury Forgeot d'Arc amaur...@gmail.com added the comment:
How would you implement this? And would you turn a disk full error, for
example, into a ZipError as well?
--
nosy: +amaury.forgeotdarc
___
Python tracker rep...@bugs.python.org
http
701 - 800 of 2621 matches
Mail list logo
