Socket Memory Leak

[Martin v. Löwis]

William Trenker <wtrenker at shaw.ca> writes:

> When I look at the Python patches on Sourceforge I can't see the
> patch that fixes problem 783312.  This is probably because I don't
> understand the process of what happens between the time a fix is
> implemented in CVS and the time it shows up as a patch on the patch
> list.  Can someone enlighten me on this process?

The process is the other way 'round. Patches contributed to the SF
patch list may, after some consideration, be integrated into the CVS.
Changes integrated into the CVS need to be posted to the patches
tracker, unless either
a) the patch author has no write access to the CVS, or
b) the patch author is uncertain as to whether the patch is correct.

> Also, when a significant bug like this is identified and fixed, is
> this announced on a list somewhere so we can incorporate critical
> fixes like this?

There is no procedure to do that. Instead, the CVS revisions of the
files being changed are recorded in the SF bug report, and the bug
report number is recorded in the CVS commit message, and in the CVS
version of the Misc/NEWS file. That is the place where you should get
a list of all significant changes.

> Of course this might all be documented on the Python site and I've
> overlooked it.  A link to such documentation would be greatly
> appreciated.

I recommend you take a close look at the CVS. It is all inside.

Regards,
Martin