You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
sendmsg() and recvmsg() for C socket module #41944
socketmodule.diff.tar.bz2: New patches to socketmodule.{c,h} to implement recvmsg and sendmsg in an abstraction.
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
assignee=Noneclosed_at=<Date2009-07-27.02:39:18.047>created_at=<Date2005-05-03.11:19:55.000>labels= ['extension-modules', 'type-feature']
title='sendmsg() and recvmsg() for C socket module'updated_at=<Date2009-07-27.02:39:18.036>user='https://bugs.python.org/hwundram'
The following tar.bz2 contains a feature enhancement
proposal for the socket module to create wrappings of
the recvmsg and sendmsg primitives. These are sometimes
useful, especially when dealing with ancillary
messages, such as communicating the process, user and
group id of a remote process to a server for proper
identification.
These functions have so far been available in a
separate module (python-eunuchs from the twisted
project), but as they are standard POSIX and have been
available on *BSD, AIX, Linux and several other OSs for
some time they should be integrated into the core
socket module.
The attached file contains patches for configure.in the
socket module, socket.py and pyconfig.h.in. I've yet to
write a test-suite, and the documentation could use
some reworking.
That's really great, it's a shame this patch didn't get much interest. Do you feel like adding test to this, it would probably psuh for its inclusion ?
beazley commented on Dec 24, 2008
beazleymannequin
Mannequin
More actions
Bump. This functionality seems to be needed if anyone is going to be
messing around with advanced features of IPv6. As it stands, the socket
module in Python 2.6/3.0 is incomplete without this.
beazley commented on Dec 27, 2008
beazleymannequin
Mannequin
More actions
Just a followup comment to note that adding support for
sendmsg()/recvmsg() is what you need to do "file descriptor passing"
between processes on Unix---another technique for writing network servers.
The patch doesn't apply cleanly to the current trunk (which is
understandable, given the patch is from 2005). Also, some tests would be
welcome as Thomas said. And bonus points for documentation :)
Activity
hwundram commentedon May 3, 2005
The following tar.bz2 contains a feature enhancement
proposal for the socket module to create wrappings of
the recvmsg and sendmsg primitives. These are sometimes
useful, especially when dealing with ancillary
messages, such as communicating the process, user and
group id of a remote process to a server for proper
identification.
These functions have so far been available in a
separate module (python-eunuchs from the twisted
project), but as they are standard POSIX and have been
available on *BSD, AIX, Linux and several other OSs for
some time they should be integrated into the core
socket module.
The attached file contains patches for configure.in the
socket module, socket.py and pyconfig.h.in. I've yet to
write a test-suite, and the documentation could use
some reworking.
therve commentedon Mar 6, 2007
That's really great, it's a shame this patch didn't get much interest. Do you feel like adding test to this, it would probably psuh for its inclusion ?
beazley commentedon Dec 24, 2008
Bump. This functionality seems to be needed if anyone is going to be
messing around with advanced features of IPv6. As it stands, the socket
module in Python 2.6/3.0 is incomplete without this.
beazley commentedon Dec 27, 2008
Just a followup comment to note that adding support for
sendmsg()/recvmsg() is what you need to do "file descriptor passing"
between processes on Unix---another technique for writing network servers.
pitrou commentedon Dec 27, 2008
The patch doesn't apply cleanly to the current trunk (which is
understandable, given the patch is from 2005). Also, some tests would be
welcome as Thomas said. And bonus points for documentation :)
ezio-melotti commentedon Jul 27, 2009
Closing this as duplicate of bpo-6560.