[Idle-dev] [Fwd: Re: Idlefork]

[Stephen M. Gava]

Moving this discussion on to idle-dev.

-----Forwarded Message-----

From: Stephen M. Gava <elguavas@users.sourceforge.net>
To: Kurt B. Kaiser <kbk@shore.net>
Cc: Raymond Hettinger <python@rcn.com>, Chui Tey <teyc@users.sourceforge.net>, Guido van Rossum <guido@python.org>
Subject: Re: Idlefork
Date: 12 Jun 2002 09:37:04 +1000

> I saw the printing checkin and sync'd up to it.
> 
> You checked it in on the MAIN branch, I'm not sure if that that was
> your intention.  

Yeah I noticed that MAIN hadn't changed so I just committed there.

> I'm planning on doing a checkin today which will pull
> the rug out from under the MAIN branch, so I'd recommend that you
> 
> cvs up -r DS_RPC_BRANCH
> 
> to get over there and do further non-GvR-rpc related checkins on that
> branch.

I guess I should do that. My feelings about the branching arrangement
were a bit mixed, but I was thinking of just committing my own stuff
straight to MAIN whenever you had indicated that it was in at least a
semi-stable state. The thing is, if DS_RPC_BRANCH is meant to be a
relatively stable branch that users can do checkouts from, then I don't
know if I should be committing further config stuff there, since I'm
just as liable to be breaking things as anybody else. Maybe my thinking
isn't being particularly clear on this. I thought you mentioned at one
stage that maybe all new work should continue on MAIN and DS_RPC_BRANCH
should be left for stable checkout and historical comparison purposes,
in which case I wondered why did we even need to branch since the idle
tree was already tagged before that point. I meant to ask you about that
when we were still discussing the branching but then I got distracted.
Probably I'm failing to see some important point entirely.

> I've been pulling my hair out for the past week on the RPC stuff but I
> see light at the end of the tunnel.  Don't _think_ it's the train.

*L* Hopefully if it is it'll stop to pick you up.

BTW, while I'm on the subject of getting distracted and forgetting to do
stuff, I also meant to mention _everyone_ that not only is python 2.2.x
or better required to work on (or even run) idlefork at the moment, but
that Guido has made a clear statement that we are not to bother at all
about writing code especially to be compatible with earlier python
versions. If there are useful 2.2.x + constructs for the job at hand
then use them. If and when idlefork becomes the new idle it will be to
accompany the then current and future python versions. Older python
versions will have the previous idle versions. 

Stephen.

-- 
Stephen M. Gava  <elguavas@users.sourceforge.net>
IDLEfork ( http://idlefork.sourceforge.net )  " just like IDLE, only
crunchy "