subversion-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Branko Čibej <>
Subject Re: [PATCH] Introduce per-instance filesystem UUIDs
Date Sat, 09 Aug 2014 23:42:49 GMT
On 09.08.2014 21:53, Evgeny Kotkov wrote:
>> The idea of introducing a separate repository UUID that is "guaranteed" to
>> be unique has been discussed a couple times, tough maybe not on-list. I
>> agree that we need something like this, for the reasons you state below.
> [...]
>> I suggest you create a branch and commit this there, so that people can
>> actually test with different scenarios. If and when you feel that the branch
>> is ready, just initiate a merge-vote on this list — see the recent vote for
>> merging the svn-auth-x509 branch as an example.
>> I think the change is serious enough that it merits being developed on a
>> branch, especially since I'd really like to stabilize trunk and begin the
>> 1.9 release cycle. This could mean that your change wouldn't make it into
>> 1.9.
> I am sorry, but did you look at the patch?  Seriously, why do we need a branch
> for a change with a core part that fits into a screen?  And what is wrong with
> testing different scenarios after "svn patch"-ing the trunk?
> We know that there is a problem, and here is a patch that solves it.  It is not
> like I would send some sort of a work-in-progress to this mailing list with a
> [PATCH] tag in the title; it's ready to be committed.  Based on my judgement,
> this patch is good, and I actually wanted to commit this directly.  However,
> I also want to avoid any possible kind of bias here, and be sure that a change
> like this does not remain unnoticed.
> Do you have any sort of objections for this patch as solution for the defined
> problem?

See Stefan's mail. It may be a small change in terms of the size of the
patch, but it's a major new feature of the repository back-end. We
decided a while ago that we prefer to develop and validate those on
branches, especially now that we're (hopefully) close to a release.
Clearly working on a branch is easier than hoping that the current patch
is complete (which it obviously isn't).

-- Brane

Branko Čibej | Director of Subversion
WANdisco | Realising the impossibilities of Big Data

View raw message