On 10/8/05, Max Bowsher <maxb@ukf.net> wrote:
> kfogel@collab.net wrote:
> > Mark Phippard <markp@softlanding.com> writes:
> >> Pretty cool.
> >>
> >> It looks like for partial committers it is not including the actual
> >> commits
> >> made by those committers unless they are also mentioned in the log
> >> message.
> >> Often times when partial committers have a patch approved they commit it
> >> themselves. Most/many of those do not appear to show up.
> >>
> >> Perhaps that was intentional?
> >
> > Nope, not intentional nor desired (though a known bug, I'd noticed it
> > before). Partial committers are already on the radar screen, so to
> > speak, so it's not *so* critical that every little thing they've done
> > show up in the list. However, it would be a real improvement if it
> > would include those commits. I hope to get around to that soon (of
> > course, if you've got time to make a patch, so much the better!).
>
> The patch is simple - enclosed below.
>
> However, it actually makes the statistics *less* useful at present.
> Longstanding partial committers who have done many commits within their
> domain, so this pulls many partial committers to the top of the list,
> inaccurately, since a commit within their partial domain isn't an indication
> to consider expanding their commit access.
>
> What we need, is a way to filter commits against the commit domain of each
> committer, but currently the commit domains are defined in ad-hoc English,
> and are totally non-machine-parseable.
When partial committers commit nontrivial patches outside their
domain, there's usually an "Approved by" header.
Cheers,
David
--
David James -- http://www.cs.toronto.edu/~james
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 8 13:39:20 2005