[issue9067] Use macros from pyctype.h

2011-03-30 Thread Andrej Krpic

Changes by Andrej Krpic :


--
nosy: +akrpic77

___
Python tracker 
<http://bugs.python.org/issue9067>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue10966] eliminate use of ImportError implicitly representing SkipTest

2011-03-30 Thread Andrej Krpic

Changes by Andrej Krpic :


--
nosy: +akrpic77

___
Python tracker 
<http://bugs.python.org/issue10966>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue7879] Too narrow platform check in test_datetime

2010-02-07 Thread Andrej Krpic

New submission from Andrej Krpic :

Windows doesn't accept negative timestamps (stated in the comment), yet checks 
is made against os.name instead of sys.platform.

patch fixes that, and also enables windows ce to pass on this test.

I think this is better than having os.name in ("nt", "ce").

--
components: Tests
files: test_datetime_win32_check.patch
keywords: patch
messages: 99026
nosy: akrpic77
severity: normal
status: open
title: Too narrow platform check in test_datetime
versions: Python 2.7
Added file: http://bugs.python.org/file16171/test_datetime_win32_check.patch

___
Python tracker 
<http://bugs.python.org/issue7879>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue7879] Too narrow platform check in test_datetime

2010-05-10 Thread Andrej Krpic

Changes by Andrej Krpic :


--
components: +Windows
type:  -> behavior

___
Python tracker 
<http://bugs.python.org/issue7879>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue7706] Missing #include guards

2010-01-15 Thread Andrej Krpic

New submission from Andrej Krpic :

Patch provides consistency with include guards already present in core.
This issue is somehow next step to #1495999 and #1492356.
These headers are missing from MSVC when targeting MS Windows CE .

HAVE_PROCESS_H already used in:
Modules/posixmodule.c, Python/thread_nt.h

HAVE_SYS_TYPES_H already used in:
Modules/io/_iomodule.c, Objects/fileobject.c, ...

HAVE_SIGNAL_H already in:
Modules/posixmodule.c

HAVE_ERRNO_H already in:
Include/Python.h, Python/ceval.c, ...
DONT_HAVE_ERRNO_H is not used anywhere

HAVE_FCNTL_H already in:
Modules/posixmodule.c, Modules/socketmodule.c

--
components: Interpreter Core
files: missing_ifdef_s.patch
keywords: patch
messages: 97809
nosy: akrpic77
severity: normal
status: open
title: Missing #include guards
type: compile error
versions: Python 2.7
Added file: http://bugs.python.org/file15891/missing_ifdef_s.patch

___
Python tracker 
<http://bugs.python.org/issue7706>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue8142] libffi update to 3.0.9

2010-05-24 Thread Andrej Krpic

Changes by Andrej Krpic :


--
nosy: +akrpic77
status: pending -> open

___
Python tracker 
<http://bugs.python.org/issue8142>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com



[issue17634] Win32: shutil.copy leaks file handles to child processes

2013-04-04 Thread Andrej Krpic

New submission from Andrej Krpic:

Function shutil.copy opens source file with open('rb') and destination file 
with open('wb') and then proceeds to copy the content.

If you create child process from the same python process during that time, that 
process will inherit open file handles.

This could lead to situation when parent process tries to delete the file, but 
child process prevents that by holding an open handle to the file (Windows XP, 
Windows 7).
 
This is not expected side effect for copy file operation.
Win32's native CopyFile API call doesn't leak file handles to child processes.

Python's open function behavior is defined by platform's implementation, which 
translates C function fopen with 'rb' and 'wb' modes to CRT open call without 
O_NOINHERIT flag, which creates inheritable Win32 HANDLEs.

Possible fix would be to add 'N' to mode string in calls to open function in 
shutil.copy.

--
components: Library (Lib), Windows
messages: 186041
nosy: akrpic77
priority: normal
severity: normal
status: open
title: Win32: shutil.copy leaks file handles to child processes
type: security
versions: Python 2.6, Python 2.7, Python 3.1, Python 3.2, Python 3.3

___
Python tracker 
<http://bugs.python.org/issue17634>
___
___
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com