|Main Archive Page > Month Archives > spamassassin-dev archives|
On Sep 14, 2011, at 11:05 AM, Darxus@chaosreigns.com wrote:
> On 09/14, Kevin A. McGrail wrote:
>> Here's my best recap and explanation because I think you are confusing the
>> branch issue too much.
> To summarize: Incrementing the version from 3.3.x to 3.4.x without
> branching svn qualifies as "branching".
> After re-reading this thread, I am hopeful that you are the only one who
> feels this way, and I would like you to please stop it :)
> Say this month, September, we do a release from trunk, call it 3.4.0, create
> an svn branch from trunk called 3.4 (due to status quo). Then in January,
> assuming we still want to do another release of what is in trunk at that
> time, with no major changes, and call it 3.4.1. What do we do with svn
> in January?
You wouldn't do that.
For a release, you branch trunk based on the major/minor release number, for instance 3.4.0 would branch to 3.4.
Then you would create the 3.4.0 release off that branch. If you later did a 3.4.1 release you would also release it from the 3.4 branch.
If you want to do a 3.5.0 release in January then you would branch to 3.5 and then do the 3.5.0 release from there.
> I guess the clearest I can be is to say: I'd like the releases
> to always come from trunk (either directly built from trunk, or svn
> branched around the time of release), unless a reason comes up not to.
> I don't care what the version numbers are, and I don't care (much)
> if you want to copy trunk to another svn branch around that time.
> So I guess I'm back to my last suggestion:
> September: Release from trunk, version 3.4.0, svn branched to 3.4.0 (not 3.4)
> January: Release from trunk, version 3.4.1, svn branched to 3.4.1
I'm -1 for this branching/release strategy.
> I don't think the svn branching is necessary, but if it makes some folks
> more comfortable, and we completely stop backporting patches, I don't
> think it's a problem.
> "The whole aim of practical politics is to keep the populace alarmed --
> and hence clamorous to be led to safety -- by menacing it with an endless
> series of hobgoblins, all of them imaginary." - H. L. Mencken