Code review comment for lp:~ian-clatworthy/bzr/filtering-revert-bug

On Mon, 2009-12-14 at 08:27 +0000, Martin Pool wrote:
> Hi,
>
> As part of the SRU process, we need to attribute changes in the branch
> to bugs. However, there was no bug for this particular change. For
> things going into the stable series, please (everyone) make sure there
> is one.

I've replied on-list. I don't think this, as currently proposed, will
really do what the Ubuntu folk are looking for.

If you see https://wiki.ubuntu.com/StableReleaseUpdates, you will see
that 'new upstreams' are explicitly rejected as reasons for SRU's - and
that is perhaps where the tension is coming from.

The wiki page does not explicitly say, because its written from an
Ubuntu dev perspective, but the bugs referenced there may need to be
*Ubuntu* bugs, not upstream bugs. However,
https://wiki.ubuntu.com/LandscapeUpdates and
https://lists.ubuntu.com/archives/ubuntu-devel-announce/2009-March/000550.html have more detail on a blanket exception prepare for landscape-client, and they want upstream bugs. So its possible that things could go either way. Note that landscape also needs a branch per bug, and so forth.

All that said, I suspect that its mostly paranoia and the changes are
not examined in the degree of detail that policy suggests might be the
case. (though I am sure they are examined).

-Rob