This change addresses part of issue 4336.
Change endheaders() to take an optional message_body argument
that is sent along with the headers. Change xmlrpclib and
httplib's other methods to use this new interface.
It is more efficient to make a single send() call, which should
get the entire client request into one packet (assuming it is
smaller than the MTU) and will avoid the long pause for delayed
ack following timeout.
Also:
- Add a comment about the buffer size for makefile().
- Extract _set_content_length() method and fix whitespace issues there.
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r67384 | benjamin.peterson | 2008-11-25 16:13:31 -0600 (Tue, 25 Nov 2008) | 4 lines
don't duplicate calls to start_tree()
RefactoringTool.pre_order values now holds a list of the fixers while pre_order_mapping holds the dict.
........
r67386 | benjamin.peterson | 2008-11-25 16:44:52 -0600 (Tue, 25 Nov 2008) | 1 line
#4423 fix_imports was still replacing usage of a module if attributes were being used
........
r67387 | benjamin.peterson | 2008-11-25 16:47:54 -0600 (Tue, 25 Nov 2008) | 1 line
fix broken test
........
r67389 | benjamin.peterson | 2008-11-25 17:13:17 -0600 (Tue, 25 Nov 2008) | 1 line
remove compatibility code; we only cater to 2.5+
........
r67390 | benjamin.peterson | 2008-11-25 22:03:36 -0600 (Tue, 25 Nov 2008) | 1 line
fix#3994; the usage of changed imports was fixed in nested cases
........
r67392 | benjamin.peterson | 2008-11-26 11:11:40 -0600 (Wed, 26 Nov 2008) | 1 line
simpilfy and comment fix_imports
........
r67399 | benjamin.peterson | 2008-11-26 11:47:03 -0600 (Wed, 26 Nov 2008) | 1 line
remove more compatibility code
........
r67400 | benjamin.peterson | 2008-11-26 12:07:41 -0600 (Wed, 26 Nov 2008) | 1 line
set svn:ignore
........
r67403 | benjamin.peterson | 2008-11-26 13:11:11 -0600 (Wed, 26 Nov 2008) | 1 line
wrap import
........
r67404 | benjamin.peterson | 2008-11-26 13:29:49 -0600 (Wed, 26 Nov 2008) | 1 line
build the fix_imports pattern in compile_pattern, so MAPPING can be changed and reflected in the pattern
........
r67405 | benjamin.peterson | 2008-11-26 14:01:24 -0600 (Wed, 26 Nov 2008) | 1 line
stop ugly messages about runtime errors being from printed
........
r67426 | benjamin.peterson | 2008-11-28 16:01:40 -0600 (Fri, 28 Nov 2008) | 5 lines
don't replace a module name if it is in the middle of a attribute lookup
This fix also stops module names from being replaced if they are not in an attribute lookup.
........
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r67183 | benjamin.peterson | 2008-11-10 21:51:33 -0600 (Mon, 10 Nov 2008) | 1 line
handle 'import x as y' in fix_imports; this still needs more work...
........
r67191 | benjamin.peterson | 2008-11-11 17:24:51 -0600 (Tue, 11 Nov 2008) | 1 line
super() is good
........
r67371 | benjamin.peterson | 2008-11-24 16:02:00 -0600 (Mon, 24 Nov 2008) | 1 line
don't blow up in the metaclass fixer when assignments in the class statement aren't simple
........
class.
These methods can be overridden. The tests verify that the overridden
methods are called, and that changes to the connection have a visible
effect on the request.
Changed semantic of _fileio.FileIO's close() method on file objects with closefd=False.
The file descriptor is still kept open but the file object behaves like a closed file.
The FileIO object also got a new readonly attribute closefd.
Approved by Barry
Backport of r67106 from the py3k branch
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66985 | benjamin.peterson | 2008-10-20 16:43:46 -0500 (Mon, 20 Oct 2008) | 1 line
no need to use nested try, except, finally
........
r67170 | benjamin.peterson | 2008-11-08 12:28:31 -0600 (Sat, 08 Nov 2008) | 1 line
fix#4271: fix_imports didn't recognize imports with parenthesis (ie from x import (a, b))
........
r67173 | benjamin.peterson | 2008-11-08 17:42:08 -0600 (Sat, 08 Nov 2008) | 1 line
consolidate test
........
r67177 | benjamin.peterson | 2008-11-09 21:52:52 -0600 (Sun, 09 Nov 2008) | 1 line
let the metclass fixer handle complex assignments in the class body gracefully
........
r67178 | benjamin.peterson | 2008-11-10 15:26:43 -0600 (Mon, 10 Nov 2008) | 1 line
the metaclass fixers shouldn't die when bases are not a simple name
........
r67179 | benjamin.peterson | 2008-11-10 15:29:58 -0600 (Mon, 10 Nov 2008) | 1 line
allow the fix_import pattern to catch from imports with parenthesis
........
does not return an iterator for the 4th and 5th items.
(sequence-like and mapping-like state)
A list is not an iterator...
Will backport to 2.6 and 2.5.
order to actually test what it was supposed to test, i.e. that the code
in the __del__ method did not crash. Use instead the new helper
test_support.captured_output().
Turn them into unit tests, and correct an obvious typo:
(("a", "b") ("c", "d") ("e", "f"))
compiles even with the missing commas, but does not execute very well...
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66805 | benjamin.peterson | 2008-10-04 20:11:02 -0500 (Sat, 04 Oct 2008) | 1 line
mention what the fixes directory is for
........
r66841 | benjamin.peterson | 2008-10-07 17:48:12 -0500 (Tue, 07 Oct 2008) | 1 line
use assertFalse and assertTrue
........
r66860 | benjamin.peterson | 2008-10-08 16:05:07 -0500 (Wed, 08 Oct 2008) | 1 line
instead of abusing the pattern matcher, use start_tree to find a next binding
........
r66884 | benjamin.peterson | 2008-10-13 15:50:30 -0500 (Mon, 13 Oct 2008) | 1 line
don't print tokens to stdout when -v is given
........
r66885 | benjamin.peterson | 2008-10-13 16:28:57 -0500 (Mon, 13 Oct 2008) | 1 line
add the -x option to disable fixers
........
r66886 | benjamin.peterson | 2008-10-13 16:33:53 -0500 (Mon, 13 Oct 2008) | 1 line
cut down on some crud
........
r66893 | benjamin.peterson | 2008-10-14 17:16:54 -0500 (Tue, 14 Oct 2008) | 1 line
add an optional set literal fixer
........
r66907 | benjamin.peterson | 2008-10-15 16:59:41 -0500 (Wed, 15 Oct 2008) | 1 line
don't write backup files by default
........
r66910 | benjamin.peterson | 2008-10-15 17:43:10 -0500 (Wed, 15 Oct 2008) | 1 line
add the -n option; it stops backupfiles from being written
........
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66707 | benjamin.peterson | 2008-09-30 18:27:10 -0500 (Tue, 30 Sep 2008) | 1 line
fix#4001: fix_imports didn't check for __init__.py before converting to relative imports
........
r66775 | collin.winter | 2008-10-03 12:08:26 -0500 (Fri, 03 Oct 2008) | 4 lines
Add an alternative iterative pattern matching system that, while slower, correctly parses files that cause the faster recursive pattern matcher to fail with a recursion error. lib2to3 falls back to the iterative matcher if the recursive one fails.
Fixes http://bugs.python.org/issue2532. Thanks to Nick Edds.
........
r66782 | benjamin.peterson | 2008-10-03 17:51:36 -0500 (Fri, 03 Oct 2008) | 1 line
add Victor Stinner's fixer for os.getcwdu -> os.getcwd #4023
........
FreeBSD 7's underlying malloc() is behaves differently to earlier versions
and seriously overcommits available memory on amd64. This may affect
other 64bit platforms in some circumstances, so the scale of the
problematic test is wound back.
Patch by Mark Dickinson, reviewed by Martin von Loewis.
calling fork() from a child thread. This disables that unit test (with a note
printed to stderr) on those platforms.
A caveat about buggy platforms is added to the os.fork documentation.
timer that did not return a float and a timer was still running when the
Profiler object was garbage collected.
Fixes issue 3895.
Code review by Benjamin Peterson.
A asyncore based mock ftp server is used to test the protocol.
This is all thanks to Giampaolo Rodola #3939
(Barry gave me permission to do this before final on IRC.)
Part of source_os2emx.patch in issue 3868:
Include/pystrcmp.h: OS/2 has same C APIs as Windows
Lib/test/test_io.py: OS/2 has same behaviour as Windows for this test
Reviewed by Amaury Forgeot d'Arc
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66470 | benjamin.peterson | 2008-09-15 18:29:43 -0500 (Mon, 15 Sep 2008) | 1 line
don't use os.linesep for newlines; it breaks tests on windows
........
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66191 | benjamin.peterson | 2008-09-03 17:00:52 -0500 (Wed, 03 Sep 2008) | 1 line
update the Grammar file after recent syntax changes
........
r66418 | benjamin.peterson | 2008-09-12 18:49:48 -0500 (Fri, 12 Sep 2008) | 1 line
a trival fix to get a few more print corner cases #2899
........
r66438 | benjamin.peterson | 2008-09-12 21:32:30 -0500 (Fri, 12 Sep 2008) | 5 lines
add Jack Diederich's fixer for metaclass syntax #2366
my contribution to this was adding a few tests and fixing a few bugs
I also reviewed it (Jack is a committer)
........
r66445 | benjamin.peterson | 2008-09-13 10:50:00 -0500 (Sat, 13 Sep 2008) | 1 line
add a few more tests concerning int literals and weird spacing
........
the __long__ slot is allowed to return either int or long, but the behaviour of
float objects should not change between 2.5 and 2.6.
Reviewed by Benjamin Peterson
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r66173 | benjamin.peterson | 2008-09-02 18:57:48 -0500 (Tue, 02 Sep 2008) | 8 lines
A little 2to3 refactoring #3637
This moves command line logic from refactor.py to a new file called
main.py. RefactoringTool now merely deals with the actual fixers and
refactoring; options processing for example is abstracted out.
This patch was reviewed by Gregory P. Smith.
........
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.
catch_warnings(), and clean up the API.
While expanding the test suite, a bug was found where a warning about the
'line' argument to showwarning() was not letting functions with '*args' go
without a warning.
Closes issue 3602.
Code review by Benjamin Peterson.
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r65887 | benjamin.peterson | 2008-08-19 17:45:04 -0500 (Tue, 19 Aug 2008) | 1 line
allow the raw_input fixer to handle calls after the raw_input (ie. raw_input().split())
........
r65889 | benjamin.peterson | 2008-08-19 18:11:03 -0500 (Tue, 19 Aug 2008) | 1 line
no need for 2.4 compatibility now
........
r65967 | benjamin.peterson | 2008-08-21 18:43:37 -0500 (Thu, 21 Aug 2008) | 1 line
allow a Call to have no arguments
........
r65968 | benjamin.peterson | 2008-08-21 18:45:13 -0500 (Thu, 21 Aug 2008) | 1 line
add a fixer for sys.exc_info etc by Jeff Balogh #2357
........
r65981 | benjamin.peterson | 2008-08-22 15:41:30 -0500 (Fri, 22 Aug 2008) | 1 line
add a fixer to add parenthese for list and gen comps #2367
........
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.
Add some test coverage to this code. More tests should be added (TODO added).
R=Brett
TESTED=./python -E -tt ./Lib/test/regrtest.py test_multiprocessing
* 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/sandbox/trunk/2to3/lib2to3
........
r65876 | benjamin.peterson | 2008-08-19 15:54:52 -0500 (Tue, 19 Aug 2008) | 1 line
apply a fix I think will help Windows
........
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r65853 | benjamin.peterson | 2008-08-19 11:09:09 -0500 (Tue, 19 Aug 2008) | 1 line
apply a patch for #3131. this solves the problem for the moment, but we should do some refactoring to get display logic out of RefactoringTool
........
r65854 | benjamin.peterson | 2008-08-19 11:37:38 -0500 (Tue, 19 Aug 2008) | 1 line
another quick fix to get lib2to3 to work
........
This does have an unfortunate side-effect of silencing the warning for all
subsequent code that imports mimetools as well since the warning is only
executed upon the first import of mimetools.
svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........
r65397 | collin.winter | 2008-08-01 22:39:06 -0500 (Fri, 01 Aug 2008) | 5 lines
Patch #3480 by Nick Edds.
Dramatically simplifies the fix_imports pattern, resulting in a reduction of the test_all_fixers runtime from 122+ secs to 59 secs (a good predictor of 2to3 performance).
........
mimetools.
This has an unfortunate side-effect of potentially not letting any warning
about rfc822's deprecation be seen by user-visible code if rfc822 is not
imported before mimetools. This is because modules are cached in sys.modules
and thus do not have their deprecation triggered more than once. But this
silencing would have happened by other code that silences the use of mimetools
or rfc822 anyway in the stdlib or user code, and thus seems justified to be
done here.
commenting out True/False compatbility stuff, remove a use of apply(), and
remove a use of buffer() (just pulled the solution used in 3.0 which is direct
slicing).
long raise ValueError instead of returning 0. Also, change the error
message for conversion of an infinity to an integer, replacing 'long' by
'integer', so that it's appropriate for both long(float('inf')) and
int(float('inf')).
Vinay Sajip
Raymond Hettinger
Jeremy Hylton
Benjamin Peterson
Jesse Noller
Christian Heimes
Georg Brandl
Amaury Forgeot d'Arc
Josiah Carlson
Martin v. Löwis
Skip Montanaro
Brett Cannon
Hirokazu Yamamoto
Thomas Heller
Armin Rigo
Armin Ronacher
Andrew M. Kuchling
Barry Warsaw
Andrew MacIntyre
Gregory P. Smith
Bill Janssen
Jesus Cea
Mark Hammond
Nick Coghlan
Guido van Rossum
Facundo Batista
Antoine Pitrou
Marc-André Lemburg
Neal Norwitz
Robert Schuppenies
Mark Dickinson