compile(). This was due to left-over special-casing before UTF-8 became the
default source encoding.
Closes issue #3574. Thanks to Victor Stinner for help with the patch.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r66814 | martin.v.loewis | 2008-10-06 17:15:40 +0200 (Mo, 06 Okt 2008) | 3 lines
Issue #4014: Don't claim that Python has an Alpha release status, in addition
to claiming it is Mature.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r66686 | martin.v.loewis | 2008-09-30 00:09:07 +0200 (Di, 30 Sep 2008) | 5 lines
Issue #3965: Allow repeated calls to turtle.Screen, by making it a
true singleton object.
Reviewed by Gregor Lingl.
........
Removed itertools usage from Lib/traceback.py, because itertools
is extension module, so maybe unavailable on build process.
(Lib/_dummy_thread.py uses Lib/traceback.py)
Reviewed by Amaury Forgeot d'Arc.
with only a file pointer when it called fp_setreadl() which expected a file
path. Changed fp_setreadl() to use either a file path or file descriptor
(derived from the file pointer) to fix the issue.
Closes issue 3594.
Reviewed by Antoine Pitrou and Benjamin Peterson.
could be easily encountered under Windows in debug mode when exercising
the recursion limit checking code, due to bogus handling of recursion
limit when USE_STACKCHEK was enabled.
Reviewed by Amaury Forgeot d'Arc on IRC.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r66171 | amaury.forgeotdarc | 2008-09-03 01:19:56 +0200 (mer., 03 sept. 2008) | 9 lines
Issue 2975: when compiling multiple extension modules with visual studio 2008
from the same python instance, some environment variables (LIB, INCLUDE)
would grow without limit.
Tested with these statements:
distutils.ccompiler.new_compiler().initialize()
print os.environ['LIB']
But I don't know how to turn them into reliable unit tests.
........
passed in as the message to use for a warning. Fixed along with making the code
more robust against other errors where return values were not checked.
Closes issue 3639.
Code review by Benjamin Peterson.
Python 2.6 one, since the intention is to keep an unified 2.x/3.x
codebase.
The Python code is automatically translated using "2to3". Please, do not
update this code in Python 3.0 by hand. Update the 2.6 one and then do
"2to3".
exception afterwards (for a subsequent parameter), the user code will
not call PyBuffer_Release() and memory will leak.
Reviewed by Amaury Forgeot d'Arc.
an 'except' clause), the exception __context__ would be reset to None.
This crases the interpreter if this precisely happens inside PyErr_SetObject.
- now the __context__ is properly preserved
- in any case, PyErr_SetObject now saves the current exc_value in a local variable, to
avoid such crashes in the future.
Reviewer: Antoine Pitrou.
match Python 2.5 speed despite the __instancecheck__ / __subclasscheck__
mechanism. In the process, fix a bug where isinstance() and issubclass(),
when given a tuple of classes as second argument, were looking up
__instancecheck__ / __subclasscheck__ on the tuple rather than on each
type object.
Reviewed by Benjamin Peterson and Raymond Hettinger.
The conversion to the unicode API was incorrect, it should use bytes.
repr is a bad variable name. The use is overloaded, but I'll leave
that to fix later.
R=Brett
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall
valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
./python -E -tt ./Lib/test/regrtest.py test_pickletools
svn+ssh://pythondev@svn.python.org/python/trunk
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall (both debug and opt modes)
........
r66006 | neal.norwitz | 2008-08-23 22:04:52 -0700 (Sat, 23 Aug 2008) | 25 lines
Fix:
* crashes on memory allocation failure found with failmalloc
* memory leaks found with valgrind
* compiler warnings in opt mode which would lead to invalid memory reads
* problem using wrong name in decimal module reported by pychecker
Update the valgrind suppressions file with new leaks that are small/one-time
leaks we don't care about (ie, they are too hard to fix).
TBR=barry
TESTED=./python -E -tt ./Lib/test/regrtest.py -uall (both debug and opt modes)
in opt mode:
valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
./python -E -tt ./Lib/test/regrtest.py -uall,-bsddb,-compiler \
-x test_logging test_ssl test_multiprocessing
valgrind -q --leak-check=yes --suppressions=Misc/valgrind-python.supp \
./python -E -tt ./Lib/test/regrtest.py test_multiprocessing
for i in `seq 1 4000` ; do
LD_PRELOAD=~/local/lib/libfailmalloc.so FAILMALLOC_INTERVAL=$i \
./python -c pass
done
At least some of these fixes should probably be backported to 2.5.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r65868 | thomas.heller | 2008-08-19 21:25:04 +0200 (Di, 19 Aug 2008) | 3 lines
Fix a regression introduced by rev. 63792: ctypes function pointers
that are COM methods must have a boolean True value.
........
r65870 | thomas.heller | 2008-08-19 21:40:23 +0200 (Di, 19 Aug 2008) | 1 line
COM method code is windows specific
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r65681 | thomas.heller | 2008-08-14 21:10:48 +0200 (Do, 14 Aug 2008) | 4 lines
issue #3554: ctypes.string_at and ctypes.wstring_at must use the
pythonapi calling convention so that the GIL is held and error return
values are checked.
........
r65682 | thomas.heller | 2008-08-14 22:04:38 +0200 (Do, 14 Aug 2008) | 2 lines
Try to fix the test on 64-bit platforms.
........
r65684 | thomas.heller | 2008-08-14 22:19:18 +0200 (Do, 14 Aug 2008) | 2 lines
Disable the test until I have one that works.
........
Code contributed by Matt Giuca. quote() now encodes the input
before quoting, unquote() decodes after unquoting. There are
new arguments to change the encoding and errors settings.
There are also new APIs to skip the encode/decode steps.
[un]quote_plus() are also affected.
PyUnicode_AsStringAndSize -> _PyUnicode_AsStringAndSize to mark
them for interpreter internal use only.
We'll have to rework these APIs or create new ones for the
purpose of accessing the UTF-8 representation of Unicode objects
for 3.1.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64976 | thomas.heller | 2008-07-15 21:39:38 +0200 (Tue, 15 Jul 2008) | 3 lines
Issue #3313: Contrary to the man page, a failed dlopen() call does not
always set a dlerror() message.
........
r64977 | thomas.heller | 2008-07-15 21:44:25 +0200 (Tue, 15 Jul 2008) | 2 lines
Add Victor Stinner, he provided the patch for issue #3313.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64968 | thomas.heller | 2008-07-15 19:03:08 +0200 (Di, 15 Jul 2008) | 4 lines
Issue #3258: Fix an assertion error (in debug build) and a crash (in
release build) when the format string of a pointer to an incomplete
structure is created.
........
r64971 | thomas.heller | 2008-07-15 19:19:50 +0200 (Di, 15 Jul 2008) | 2 lines
NEWS entry for #issue 3258.
........
All the u* gettext variants were renamed to their
none u* variants, since there's no point in translating
to byte strings. I also killed off the unicode parameters
for install
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64688 | martin.v.loewis | 2008-07-03 14:51:14 +0200 (Do, 03 Jul 2008) | 9 lines
Patch #1622: Correct interpretation of various ZIP header fields.
Also fixes
- Issue #1526: Allow more than 64k files to be added to Zip64 file.
- Issue #1746: Correct handling of zipfile archive comments (previously
archives with comments over 4k were flagged as invalid). Allow writing
Zip files with archives by setting the 'comment' attribute of a ZipFile.
........
Removal of the Mac modules broke many of the Mac scripts (including
BuildApplet.py) so the building of the Python launcher and IDLE.app was
broken. I manually copied built versions of those apps into Mac. Everything else
which used Mac modules had to die.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64549 | brett.cannon | 2008-06-26 17:31:13 -0700 (Thu, 26 Jun 2008) | 7 lines
warnings.warn_explicit() did not have the proper TypeErrors in place to prevent
bus errors or SystemError being raised. As a side effect of fixing this, a bad
DECREF that could be triggered when 'message' and 'category' were both None was
fixed.
Closes issue 3211. Thanks JP Calderone for the bug report.
........
It consists of code from urllib, urllib2, urlparse, and robotparser.
The old modules have all been removed. The new package has five
submodules: urllib.parse, urllib.request, urllib.response,
urllib.error, and urllib.robotparser. The urllib.request.urlopen()
function uses the url opener from urllib2.
Note that the unittests have not been renamed for the
beta, but they will be renamed in the future.
Joint work with Senthil Kumaran.
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64309 | amaury.forgeotdarc | 2008-06-16 21:12:42 +0200 (lun., 16 juin 2008) | 8 lines
Issue 3110: Crash with weakref subclass,
seen after a "import multiprocessing.reduction"
An instance of a weakref subclass can have attributes.
If such a weakref holds the only strong reference to the object,
deleting the weakref will delete the object. In this case,
the callback must not be called, because the ref object is being deleted!
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r64125 | benjamin.peterson | 2008-06-11 12:27:50 -0500 (Wed, 11 Jun 2008) | 2 lines
give the threading API PEP 8 names
........
The repr() of a string now contains printable Unicode characters unescaped.
The new ascii() builtin can be used to get a repr() with only ASCII characters in it.
PEP and patch were written by Atsuo Ishimoto.
with non-ascii characters.
This is the simple fix, which uses the FileSystemEncoding.
Replacing all the char* with unicode strings is a major rewrite,
and needs more thinking.
functions have been added to the subprocess module.
The fixer for this still needs to be written and proper Py3K deprecation
warnings for the functions that didn't make the transition need to be done in
2.6.
This is all part of trying to close issue #2872.
Merged revisions 63357 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63357 | alexandre.vassalotti | 2008-05-16 02:58:49 -0400 (Fri, 16 May 2008) | 2 lines
Changed references to the reprlib module to use its new name.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63208 | georg.brandl | 2008-05-13 15:04:54 -0400 (Tue, 13 May 2008) | 2 lines
#2831: add start argument to enumerate(). Patch by Scott Dial and me.
........
r63209 | marc-andre.lemburg | 2008-05-13 15:10:45 -0400 (Tue, 13 May 2008) | 3 lines
Remove leftovers from reverted setuptools checkin (they were added in r45525).
........
r63211 | georg.brandl | 2008-05-13 17:32:03 -0400 (Tue, 13 May 2008) | 2 lines
Fix a refleak in the _warnings module.
........
r63212 | andrew.kuchling | 2008-05-13 20:46:41 -0400 (Tue, 13 May 2008) | 1 line
List all the removes and renamed modules
........
r63214 | brett.cannon | 2008-05-13 21:09:40 -0400 (Tue, 13 May 2008) | 2 lines
Rewrap some lines in test_py3kwarn.
........
r63219 | georg.brandl | 2008-05-14 02:34:15 -0400 (Wed, 14 May 2008) | 2 lines
Add NEWS entry for #2831.
........
r63220 | neal.norwitz | 2008-05-14 02:47:56 -0400 (Wed, 14 May 2008) | 3 lines
Fix "refleak" by restoring the tearDown method removed by accident (AFAICT)
in r62788.
........
r63221 | georg.brandl | 2008-05-14 03:18:22 -0400 (Wed, 14 May 2008) | 2 lines
Fix another "refleak" by clearing the filters after test.
........
r63222 | neal.norwitz | 2008-05-14 03:21:42 -0400 (Wed, 14 May 2008) | 5 lines
Install the json package and tests as well as the lib2to3 tests
so the tests work when run from an install directory.
They are currently skipped on the daily runs (not from the buildbots)
for checking refleaks, etc.
........
r63256 | andrew.kuchling | 2008-05-14 21:10:24 -0400 (Wed, 14 May 2008) | 1 line
Note some removals and a rename
........
r63311 | brett.cannon | 2008-05-15 00:36:53 -0400 (Thu, 15 May 2008) | 2 lines
Add a snippet for the deprecation directive for docs.
........
r63313 | gregory.p.smith | 2008-05-15 00:56:18 -0400 (Thu, 15 May 2008) | 5 lines
disable the crashing test. I will also file a bug. This crash does
not appear to be a new bug, its just that the test coverage went up
recently exposing it. (I verified that by testing this test code on
an older Modules/_bsddb.c)
........
r63320 | georg.brandl | 2008-05-15 11:08:32 -0400 (Thu, 15 May 2008) | 2 lines
#2863: add gen.__name__ and add this name to generator repr().
........
r63324 | andrew.kuchling | 2008-05-15 16:07:39 -0400 (Thu, 15 May 2008) | 1 line
Import class from distutils.cmd, not .core, to avoid circular import
........
r63327 | alexandre.vassalotti | 2008-05-15 16:31:42 -0400 (Thu, 15 May 2008) | 2 lines
Fixed typo in a doctest of test_genexps.
........
r63332 | benjamin.peterson | 2008-05-15 18:34:33 -0400 (Thu, 15 May 2008) | 2 lines
add Mac modules to the list of deprecated ones
........
r63333 | benjamin.peterson | 2008-05-15 18:41:16 -0400 (Thu, 15 May 2008) | 2 lines
fix typos in whatsnew
........
r63348 | benjamin.peterson | 2008-05-15 22:24:49 -0400 (Thu, 15 May 2008) | 2 lines
make test_platform a bit more assertive (We'll see what the buildbots say.)
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63119 | benjamin.peterson | 2008-05-11 20:41:23 -0400 (Sun, 11 May 2008) | 2 lines
#2196 hasattr now allows SystemExit and KeyboardInterrupt to propagate
........
r63122 | benjamin.peterson | 2008-05-11 20:46:49 -0400 (Sun, 11 May 2008) | 2 lines
make message slightly more informative, so there's no chance of misunderstanding it
........
r63158 | ronald.oussoren | 2008-05-12 07:24:33 -0400 (Mon, 12 May 2008) | 5 lines
Remove references to platform 'mac'
The 'mac' platform (that is, os.name == 'mac') was used for the MacOS 9 port,
which is no longer supported (as of Python 2.4 IIRC).
........
r63159 | ronald.oussoren | 2008-05-12 07:31:05 -0400 (Mon, 12 May 2008) | 8 lines
MacOSX: remove dependency on Carbon package for urllib
This patch removes the dependency on the Carbon package from urllib.
The mac-specific code for getting proxy configuration is now writting in
Python using ctypes and uses the SystemConfiguration framework instead of
InternetConfig. Also provides a mac-specific implementation of proxy_bypass.
........
r63162 | eric.smith | 2008-05-12 10:00:01 -0400 (Mon, 12 May 2008) | 1 line
Added 'n' presentation type for integers.
........
r63164 | georg.brandl | 2008-05-12 12:26:52 -0400 (Mon, 12 May 2008) | 2 lines
#1713041: fix pprint's handling of maximum depth.
........
r63170 | georg.brandl | 2008-05-12 12:53:42 -0400 (Mon, 12 May 2008) | 2 lines
Fix parameter name for enumerate().
........
r63173 | georg.brandl | 2008-05-12 13:01:58 -0400 (Mon, 12 May 2008) | 2 lines
#2766: remove code without effect.
........
r63174 | georg.brandl | 2008-05-12 13:04:10 -0400 (Mon, 12 May 2008) | 3 lines
#2767: don't clear globs in run() call, since they could be needed in tearDown,
which clears them at the end.
........
r63175 | georg.brandl | 2008-05-12 13:14:51 -0400 (Mon, 12 May 2008) | 2 lines
#1760: try-except-finally is one statement since PEP 341.
........
r63186 | amaury.forgeotdarc | 2008-05-12 17:30:24 -0400 (Mon, 12 May 2008) | 2 lines
Sync code with documentation, and remove Win95 support in winsound module.
........
r63189 | amaury.forgeotdarc | 2008-05-12 18:21:39 -0400 (Mon, 12 May 2008) | 3 lines
Adapt test_pyclbr to the new version of urllib.py:
The new mac-specific functions must be ignored.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r62914 | skip.montanaro | 2008-05-08 20:45:00 -0400 (Thu, 08 May 2008) | 4 lines
Add an example about using NamedTemporaryFile() to replace mktemp(). I'm
unclear whether the verbatim text should have been indented or by how much.
........
r62915 | benjamin.peterson | 2008-05-08 20:50:40 -0400 (Thu, 08 May 2008) | 2 lines
reindent example
........
r62927 | georg.brandl | 2008-05-09 02:09:25 -0400 (Fri, 09 May 2008) | 2 lines
#2788: add .hgignore file.
........
r62928 | georg.brandl | 2008-05-09 02:10:43 -0400 (Fri, 09 May 2008) | 2 lines
#2781: fix function name.
........
r62929 | georg.brandl | 2008-05-09 02:18:27 -0400 (Fri, 09 May 2008) | 2 lines
Add a sentence to basicConfig() that is in the docstring.
........
r62930 | georg.brandl | 2008-05-09 02:26:54 -0400 (Fri, 09 May 2008) | 2 lines
Add another link to colorsys docs.
........
r62931 | georg.brandl | 2008-05-09 02:36:07 -0400 (Fri, 09 May 2008) | 2 lines
Add Kodos as a re reference.
........
r62932 | georg.brandl | 2008-05-09 02:39:58 -0400 (Fri, 09 May 2008) | 2 lines
Add a note about using reload().
........
r62933 | andrew.kuchling | 2008-05-09 07:46:05 -0400 (Fri, 09 May 2008) | 3 lines
Update planned release date.
Uncomment PEP 370 section.
Add some module items
........
r62934 | christian.heimes | 2008-05-09 08:19:09 -0400 (Fri, 09 May 2008) | 1 line
Add --user option to build_ext
........
r62948 | mark.dickinson | 2008-05-09 13:54:23 -0400 (Fri, 09 May 2008) | 3 lines
Issue #2487. math.ldexp(x, n) raised OverflowError when n was large and
negative; fix to return an (appropriately signed) zero instead.
........
r62949 | martin.v.loewis | 2008-05-09 14:21:55 -0400 (Fri, 09 May 2008) | 1 line
Use the CHM file name that Sphinx assigns.
........
Merged revisions 63247-63248 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63247 | georg.brandl | 2008-05-14 18:30:31 -0400 (Wed, 14 May 2008) | 2 lines
Update configparser docs for lowercasing rename.
........
r63248 | alexandre.vassalotti | 2008-05-14 18:44:22 -0400 (Wed, 14 May 2008) | 8 lines
Updated import statements to use the new `configparser` module name.
Updated the documentation to use the new name.
Revert addition of the stub entry for the old name.
Georg, I am reverting your changes since this commit should propagate
to py3k.
........
with chars outside the 7bit ascii (s# was already correct).
This is necessary to allow python run from a non-ASCII directory,
and seems enough on some platforms, probably where the default PyUnicode encoding (utf-8)
is also the default filesystem encoding.
Merged revisions 63132 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63132 | alexandre.vassalotti | 2008-05-11 22:11:22 -0400 (Sun, 11 May 2008) | 4 lines
Updated all import statements to use the new socketserver module name.
Renamed socketserver module in its own documentation.
Renamed documentation references.
........
Updated documentation to use new name.
Merged revisions 63077 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63077 | alexandre.vassalotti | 2008-05-11 15:39:48 -0400 (Sun, 11 May 2008) | 3 lines
Added stub for the Queue module to be renamed in 3.0.
Use the 3.0 module name to avoid spurious warnings.
........
Updated documentation.
Merged revisions 63042 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk
........
r63042 | alexandre.vassalotti | 2008-05-11 04:25:28 -0400 (Sun, 11 May 2008) | 5 lines
Added module stub for copy_reg renaming in 3.0.
Renamed copy_reg to copyreg in the standard library, to avoid
spurious warnings and ease later merging to py3k branch. Public
documentation remains intact.
........
svn+ssh://pythondev@svn.python.org/python/branches/py3k-ctypes-pep3118
........
r60059 | thomas.heller | 2008-01-18 22:17:05 +0100 (Fri, 18 Jan 2008) | 1 line
Implement pep3118 format strings for SimpleCData types.
........
r60108 | thomas.heller | 2008-01-19 22:56:12 +0100 (Sat, 19 Jan 2008) | 3 lines
Always use explicit endian specifiers for simple types, and a bug fix
too. Add unittest.
........
r60112 | thomas.heller | 2008-01-19 23:25:14 +0100 (Sat, 19 Jan 2008) | 2 lines
Fully implement tp_asbuffer for pointer types.
........
r60261 | thomas.heller | 2008-01-24 22:01:29 +0100 (Thu, 24 Jan 2008) | 4 lines
Added shape and ndim field to StgDictObject. Implemented pep3118
format string, ndim, and shape for array types. Added a
buffer_info(type_or_object) for testing.
........
r60278 | thomas.heller | 2008-01-25 11:53:33 +0100 (Fri, 25 Jan 2008) | 2 lines
Implement pep3118 format strings for ctypes.Structure and ctypes.Union.
........
r60288 | thomas.heller | 2008-01-25 17:58:30 +0100 (Fri, 25 Jan 2008) | 2 lines
All ctypes types now use the same CData_GetBuffer function.
........
r60289 | thomas.heller | 2008-01-25 19:59:45 +0100 (Fri, 25 Jan 2008) | 2 lines
Fix format string for structures, and itemsize for arrays.
........
r60290 | thomas.heller | 2008-01-25 20:09:03 +0100 (Fri, 25 Jan 2008) | 2 lines
Implement to format string for function pointers.
........
r60292 | thomas.heller | 2008-01-25 20:32:20 +0100 (Fri, 25 Jan 2008) | 3 lines
Only structures with native packing implement the pep. Unions, or
packed structures do not.
........
r60293 | thomas.heller | 2008-01-25 20:34:31 +0100 (Fri, 25 Jan 2008) | 2 lines
Update the test.
........
r60295 | thomas.heller | 2008-01-25 20:44:41 +0100 (Fri, 25 Jan 2008) | 2 lines
Fixed a few XXX markers.
........
r60298 | thomas.heller | 2008-01-25 21:11:08 +0100 (Fri, 25 Jan 2008) | 1 line
Fix test for 64-bt platform.
........
r60299 | thomas.heller | 2008-01-25 21:34:11 +0100 (Fri, 25 Jan 2008) | 2 lines
Add test for the readonly bit.
........
r60384 | thomas.heller | 2008-01-28 08:45:04 +0100 (Mon, 28 Jan 2008) | 4 lines
Restructure the test so that it contains little endian format strings.
On big endian machines, the format strings are converted by replacing
'<' with '>'.
........
r60385 | thomas.heller | 2008-01-28 08:58:46 +0100 (Mon, 28 Jan 2008) | 1 line
Bugfix and test for explicit big and little endian types.
........
r60428 | thomas.heller | 2008-01-29 22:00:37 +0100 (Tue, 29 Jan 2008) | 1 line
Add comments to clarify the tests.
........
r62589 | thomas.heller | 2008-04-30 13:49:46 +0200 (Wed, 30 Apr 2008) | 1 line
Fix compiler warnings.
........
svn+ssh://pythondev@svn.python.org/python/trunk
........
r62499 | mark.dickinson | 2008-04-25 12:11:04 -0400 (Fri, 25 Apr 2008) | 7 lines
Issue #1496032. Add -mieee to BASECFLAGS on alpha, when gcc is
the compiler. This should(?) help to fix failures in test_math
and test_cmath on Linux/alpha.
Also add configure message reporting the result of uname -m, as
a debugging aid.
........
rewrite it so that it doesn't care about relative sizes of size_t,
long and long long.
The rewrite is modeled on PyLong_FromLong, instead of using
PyLong_FromByteArray; this makes the algorithm simpler and
more direct, and possibly also slightly faster.
Martin v. Löwis
Brett Cannon
Benjamin Peterson
Amaury Forgeot d'Arc
Guido van Rossum
Thomas Heller
Gerhard Häring
Barry Warsaw
Antoine Pitrou
Georg Brandl
Hirokazu Yamamoto
Gregory P. Smith
Guilherme Polo
Facundo Batista
Jesse Noller
Marc-André Lemburg
Jesus Cea
Neal Norwitz
Hye-Shik Chang
Nick Coghlan
Robert Schuppenies
Lars Gustäbel
Eric Smith
Mark Dickinson
Jeremy Hylton
Alexandre Vassalotti