Gregory P. Smith [Sat, 23 Mar 2013 18:54:22 +0000 (11:54 -0700)]
Fixes issue #17488: Change the subprocess.Popen bufsize parameter default value
from unbuffered (0) to buffering (-1) to match the behavior existing code
expects and match the behavior of the subprocess module in Python 2 to avoid
introducing hard to track down bugs.
Gregory P. Smith [Sat, 23 Mar 2013 18:44:25 +0000 (11:44 -0700)]
Fixes issue #17488: Change the subprocess.Popen bufsize parameter default value
from unbuffered (0) to buffering (-1) to match the behavior existing code
expects and match the behavior of the subprocess module in Python 2 to avoid
introducing hard to track down bugs.
doko@ubuntu.com [Thu, 21 Mar 2013 20:31:41 +0000 (13:31 -0700)]
- Issue #16754: Fix the incorrect shared library extension on linux. Introduce
two makefile macros SHLIB_SUFFIX and EXT_SUFFIX. SO now has the value of
SHLIB_SUFFIX again (as in 2.x and 3.1). The SO macro is removed in 3.4.
doko@ubuntu.com [Thu, 21 Mar 2013 20:21:49 +0000 (13:21 -0700)]
- Issue #16754: Fix the incorrect shared library extension on linux. Introduce
two makefile macros SHLIB_SUFFIX and EXT_SUFFIX. SO now has the value of
SHLIB_SUFFIX again (as in 2.x and 3.1). The SO macro is removed in 3.4.
R David Murray [Thu, 21 Mar 2013 01:12:17 +0000 (21:12 -0400)]
Merge: #5713: Handle 421 error codes during sendmail by closing the socket.
This is a partial fix to the issue of servers disconnecting unexpectedly; in
this case the 421 says they are disconnecting, so we close the socket and
return the 421 in the appropriate error context.
Original patch by Mark Sapiro, updated by Kushal Das, with additional
tests by me.
R David Murray [Thu, 21 Mar 2013 00:36:14 +0000 (20:36 -0400)]
#5713: Handle 421 error codes during sendmail by closing the socket.
This is a partial fix to the issue of servers disconnecting unexpectedly; in
this case the 421 says they are disconnecting, so we close the socket and
return the 421 in the appropriate error context.
Original patch by Mark Sapiro, updated by Kushal Das, with additional
tests by me.
Gregory P. Smith [Wed, 20 Mar 2013 06:25:16 +0000 (23:25 -0700)]
Use a larger amount of data for tests such as the interrupted_write
tests that depend on filling up an OS pipe so that they work properly
on systems configured with large pipe buffers.
Gregory P. Smith [Tue, 19 Mar 2013 21:59:02 +0000 (14:59 -0700)]
Fixes Issue #17192: Update the ctypes module's libffi to v3.0.13. This
specifically addresses a stack misalignment issue on x86 and issues
on some more recent platforms.
R David Murray [Tue, 19 Mar 2013 17:56:01 +0000 (13:56 -0400)]
Merge: #17443: Fix buffering in IMAP4_stream.
In Python2 Popen uses *FILE objects, which wind up buffering even though
subprocess defaults to no buffering. In Python3, subprocess streams really
are unbuffered by default, but the imaplib code assumes read is buffered. This
patch uses the default buffer size from the io module to get buffered streams
from Popen.
Much debugging work and patch by Diane Trout.
The imap protocol is too complicated to write a test for this simple
change with our current level of test infrastructure.
R David Murray [Tue, 19 Mar 2013 17:52:33 +0000 (13:52 -0400)]
#17443: Fix buffering in IMAP4_stream.
In Python2 Popen uses *FILE objects, which wind up buffering even though
subprocess defaults to no buffering. In Python3, subprocess streams really
are unbuffered by default, but the imaplib code assumes read is buffered. This
patch uses the default buffer size from the io module to get buffered streams
from Popen.
Much debugging work and patch by Diane Trout.
The imap protocol is too complicated to write a test for this simple
change with our current level of test infrastructure.
R David Murray [Tue, 19 Mar 2013 06:32:35 +0000 (02:32 -0400)]
#17476: make allmethods actually return all methods.
This fixes a regression relative to Python2. (In 2, methods on a class were
unbound methods and matched the inspect queries being done, in 3 they are just
functions and so were missed).
This is an undocumented function that pydoc itself does not use, but
I found that numpy at least uses it in its documentation generator.
R David Murray [Tue, 19 Mar 2013 06:31:06 +0000 (02:31 -0400)]
#17476: make allmethods actually return all methods.
This fixes a regression relative to Python2. (In 2, methods on a class were
unbound methods and matched the inspect queries being done, in 3 they are just
functions and so were missed).
This is an undocumented function that pydoc itself does not use, but
I found that numpy at least uses it in its documentation generator.
Gregory P. Smith [Tue, 19 Mar 2013 00:11:20 +0000 (17:11 -0700)]
Fixes issue #17192: Update the ctypes module's libffi to v3.0.13. This
specifically addresses a stack misalignment issue on x86 and issues on
some more recent platforms.
Brett Cannon [Sun, 17 Mar 2013 22:48:16 +0000 (15:48 -0700)]
Issue #16880: _imp.load_dynamic() is not defined on a platform that
does not support dynamic loading (e.g. Atari), so make sure that imp
doesn't assume it always exists.