Bugfix releases (RE: profiler results for __getattr__, am I reading this correctly ? )

[Tim Peters]

[Anthony Baxter]
> ...
> It [PEP 6] still doesn't cover the process of deciding what would go
> in the patch release...

It would help if you worked with the PEP author (Aahz) until it was clear to
both of you, else it will stay muddy forver.  I believe his rabid <wink>
intent was "no new features!", but I agree the PEP doesn't really say that.

> I know that I'd be happier upgrading to 2.1.2 than 2.2 in the next
> couple of months

That's wise, but mostly because 2.2 won't be released before the next couple
of months is history.

> - and given the peasants-are-revolting sentiment of c.l.py at the
> moment, I doubt I'm the only one...

A curious thing about Open Source is that demand doesn't seem to create
supply, at least not in the way that casually mentioning you'd really like
to get wasted-- and have some spare cash --magically attracts eager crack
dealers to your front door.  Satisfying this particular demand is quite
doable, but The Usual Suspects aren't going to volunteer (they're overloaded
without it).

> How would this sound as a first cut at a what-goes-in-and-what-stays:
>
>    a) bugfixes.

Yes.

>    b) no new modules, no new packages.

Ruled out via "no new features".

>    c) no new methods on classes, unless they're bugfixes.

Ditto.

>    d) if the bug fix relies on some 2.2-ism, then it's either rewritten,
>       or ignored and relnoted

Sure.

Note that a discusssion of which 2.1.1 bugs count as critical is current on
Python-Dev.  If you're comfortable enough with CVS to manage the mechanics
of working on a branch, the developers will give you all the advice you can
stomach.