[Python-Dev] Ignoring SIGPIPE (was: [Python-checkins] python/dist/src/Modules socketmodule.c, 1.298, 1.299) (original) (raw)

Neal Norwitz neal at metaslash.com
Sat Aug 21 01:40:05 CEST 2004

On Fri, Aug 20, 2004 at 03:06:21PM +1000, Dave Cole wrote:

Andrew McNamara wrote: >>>Patch #1003700: Add socketpair function to socket module. >> >>The docstring (below) states the arguments are the same as socket(). >>However, in sockinitobj() line 2496, the family is initialized to >>AFINET. I think the #if defined(AFUNIX) code above should be >>removed and family should be initialized to AFINET. > >I talked Dave into this - AFUNIX is typically the only address family >that is valid for socketpair(). Using AFINET under linux and OS X >results in EOPNOTSUPP.

Would a change to the documentation and docstring be sufficient it it explained that the default family is AFUNIX if defined for the platform, AFINET otherwise?

That would make me happy. I just don't want the doc/docstring to lie. I don't have a problem with it defaulting to AF_UNIX.

>>I don't think the #ifdef SIGPIPE code is correct. If the user >>installed a signal handler calling signal() will remove it. I >>think the call to signal() should be removed. > >I agree, but I think that was copied verbatim from elsewhere in >socketmodule.c, so it was left with the aim of being bug for bug >compatible with socket.socket().

Indeed. I was not going to change something as fundamental as the existing signal handling. The onyl sensible thing to do was to copy the behaviour already in place when you create a socket the "normal" way.

You are correct. There are two places that call signal(SIGPIPE, SIG_IGN); I find this surprising.

Guido, you added this code in version 1.4 from 1991.
You do remember the reason why, don't you? It's only 13 years ago. :-) I'm hesitant to remove the old calls, but I'm not sure I want it added to new calls if it's not necessary.

Any opinions?

Neal

More information about the Python-Dev mailing list