"Mark Phippard" <markphip_at_gmail.com> writes:
> Yes, I knew you said that but I had gotten a little confused by some
> of your other comments where it seemed like you were evaluation what
> to include on a commit by commit basis.
Sure, I am advocating that kind of evaluation. If a change is
trivial/obvious/simple, then there's no reason it needs to restart the
soak.
The mechanism by which we include those changes while not including
other ones is an implementation detail. There are various ways to do
it, they all work fine, and this thread doesn't need to go into them --
that's why we have a release manager (yay).
> I should really probably read the archives before I make this point,
> but I know we have had previous releases where 4-5 weeks into the
> cycle we merged non-trivial fixes and did not do a full soak. I do
> not want to open a debate on our policies, which I generally agree
> with, but let's be honest here. If these fixes go into a 1.5.1 we
> will release them with essentially no soak period at all.
That's a very good point -- I guess it just reaffirms the fact that
point releases are supposed to be for conservative fixes anyway, not big
things. But remember, anything that goes into 1.5.1 will also have been
"soaked" on trunk for that long. Also, we can optionally soak a point
release for as long as we want; some of the pressure is off then anyway,
because the .0 will already be out.
I don't want to be a mindless stickler for rules here. I'll just review
those changes in detail and see if afterwards I'm still disturbed by
anything. Hope others will be doing same.
> Yes for merge, but no for the log -g/blame -g options. As Mike said,
> this property is versioned and those commands operate on what happened
> to the property in the history of the repository.
Mmm, right, gotcha.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-05-19 23:36:20 CEST