Re: [Fwd: Re: Dump/load changes data?]

Michael Price <mprice@atl.lmco.com> writes:
> Well, I've seen developers "veto" changes going into a patch release
> because they weren't regressions. Since this bug has been present
> forever its not really a regression.
>
> So perhaps now is a good time for everyone to clarify for me exactly
> what should and should not go into a patch release. Did I miss the
> section in the docs that covers this issue?

I think we shouldn't be so strict that we only allow regression fixes,
especially not before 1.0. Maybe a good guideline is:

  1. Fixes for major bugs that were not known at the time of the
     release. This always includes segfaults, but may also include
     other kinds of severe bugs.

  2. Any other non-code changes that seem significant (for example,
     documentation changes), if you're making a patch release due to
     reason #1 anyway.

Use your judgement for the rest. Obviously, new features and
enhancements should be avoided for patch releases.

How does that sound?
-K

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Mar 21 21:50:10 2003