Re: [PATCH v3 0/7] New remote-bzr remote helper

From: Felipe Contreras <felipe.contreras@gmail.com>
Date: Nov 28, 2012, 03:05 UTC
Message-ID: <CAMP44s23FfxQG6v+dfRGWHzxpWZ5BSGpAm=8Q8Bmj04tDwmwKw@mail.gmail.com>
In-Reply-To: <7vpq2yihaq.fsf@alter.siamese.dyndns.org>

On Wed, Nov 28, 2012 at 3:37 AM, Junio C Hamano <gitster@pobox.com> wrote:

Show 23 quoted lines

> Felipe Contreras <felipe.contreras@gmail.com> writes:
>
>>> At this point, both have been cooking for a week or more in 'next',
>>> there is no existing users, they are on the fringe so breakages in
>>> them won't negatively affect anybody anyway.  So it doesn't matter
>>> much if they are merged to 'master' and then fixed up with follow up
>>> patches after that, or fixed up with follow up patches while they
>>> are in 'next', as they won't be rewound and restarted from scratch.
>>
>> The fixes are affecting some people, that's why I did them. Some were
>> reported here in the mailing list, and some in my github's clone:
>>
>> https://github.com/felipec/git/issues?page=1&state=closed
>
> Are you talking about -hg or -bzr or both?
>
> In any case, I am mostly concerned about *my* next release, whose
> rc0 will be tagged sometime this week or the next week.
>
> People who have been bitten by bugs from *your* tree or versions in
> 'next' do not count.  When I said "no existing users", I was talking
> about the end users who need rock solid stable "releases" because
> tagged versions are the only ones they use.

If users you call "fringe" have noticed these compatibility issues, chances are your "existing users" are going to catch them as well.

Those issues were fixed right away, but I didn't sent them because I wanted things to settle. I didn't see that v2 landed in next until now.

Show 8 quoted lines

> If the code of these topics is still in flux and needs constant
> fixes, probably it is a better idea to cook them longer in 'next',
> skipping the upcoming 1.8.1 release.  If we are going to go that
> route, we can drop the v2 fc/remote-bzr and queue v3 when we rewind
> the tip of 'next' after 1.8.1 release (by that time you may have v4
> of the series, but then we can skip v3).  Is that more preferrable
> than rushing these topics forward before they are ready for general
> audience?

They are not in constant flux, that's why I haven't send any new re-rolls since v3, which was on November 11. I've been using v3 for baseline since them, and the rest of the patches I've sent on top of that.

In fact, these particular fixes were already sent on November 13 (on top of v3): http://article.gmane.org/gmane.comp.version-control.git/209558

On November 10 Jeff threatened to to merge v2 to next on the "What's cooking", and I told him I was about to sent a re-roll, he acknowledged the same day, and I sent the next one.

Since v3 remote-bzr hasn't been on flux.

Now, what you do is up to you, but I think v3 plus the two patches I sent on nov 13, and just resent today should be safe. That being said, I don't use remote-bzr really, and I don't know how many people have been using it, so I have no idea how ready it really is. If I were you I would just merge v3 to next, or revert v2 and merge v3, and then apply the two patches on top. Or if you want, revert v2, wait for 1.8.1, and then merge v3. Either way it's doubtful there will be a v4 (if there are next patches they will be on top of v3, as they have been for quite some time now), so it's not clear what "existing users" will gain by that.

I'm confident about remote-hg though.

Cheers.

-- 
Felipe Contreras