Re: Potential BTS improvements

[Colin Watson <cjwatson@debian.org>]

On Sat, Mar 27, 2004 at 09:08:55AM -0500, Raul Miller wrote:
> On Sat, Mar 27, 2004 at 02:49:38PM +0100, Adrian Bunk wrote:
> > But to know whether the bug affects the version in testing, cases like
> > closing by sending a mail to <version>-done@b.d.o have to be correctly
> > handled (e.g. "not a bug" is different from "already fixed in my upload
> > waiting in incoming").
> 
> As I understand it, this has already been implemented, so maybe this
> discussion should change to an examination of the new system.
> 
> However, I believe the right way to deal with this is:  Bugs are reported
> against a specific package version.  We should assume that future package
> versions have the bug until we reach the package version where we have
> a report that the bug is fixed.
> 
> Obviously, there will be cases where this isn't precisely accurate.
> Where that's a problem, the right thing to do is introduce more bug
> reports (and possibly bug close messages) to provide the additional
> information.
> 
> But, once again, it's my understanding that this has already been
> implemented, though I don't about deployment.

Yes, it has been implemented and will be deployed once I figure out how
to move the requisite data from ftp-master to bugs reliably in the face
of unreliable networks. (rsync could really do with a --move option ...)

I suggest that discussions about this wait until deployment, as the
current discussions are going over quite old ground.

Cheers,

-- 
Colin Watson                                  [cjwatson@flatline.org.uk]