|Main Archive Page > Month Archives > spamassassin-dev archives|
On 09/14, Kevin A. McGrail wrote:
> I think there is major confusion about what a branch is so this is
> less a comment on the ongoing discussion and more just clarifying
> the issue of branching from my point of view.
Yup, I'm definitely confused.
> >Why? What would we then do for the release scheduled for January?
> >Branch it again and call it 3.5?
> There would have to be a feature that justified a major release
> otherwise January would be a planned release in the branch (i.e.
> 3.4.1) assume
Wait, you're saying we can't (always) just release what's currently in
trunk because it's plenty stable and the latest we have available?
We have to release crufty 4 months old code again *just* because there
isn't a major change in trunk?
> >I guess I wouldn't really mind if people wanted to do a branch per
> >release. Branch trunk to 3.4.0 for this release, branch trunk to 3.4.1
> >in January. Does anybody object to that idea?
> 3.4.1 is not a branch. 3.4 is a branch. 3.3 is a branch, etc.
Why not? Aren't branch names arbitrary strings?
I admit I'm not all that familiar with the intricacies of source control
systems, but... can't subversion handle branching "trunk" to "3.4.0"? Or
is this some kind of rule, that you object to changing?
-- "Life is either a daring adventure or it is nothing at all." - Helen Keller http://www.ChaosReigns.com