| 1 |
On Tue, 2007-04-17 at 21:50 +0200, Stefan Schweizer wrote: |
| 2 |
> It would be cool to implement a keywording@g.o alias just to |
| 3 |
> assign those bugs to so that we maintainers do not need to see them. |
| 4 |
|
| 5 |
While you may not find them useful, there have been 3 recent occasions |
| 6 |
of user requesting things get keyworded that I maintained on |
| 7 |
architectures where the packages didn't work. I don't know what these |
| 8 |
users did, but on all three occasions, I managed to step in and stop |
| 9 |
breakage from hitting the tree *because* I was in the chain of |
| 10 |
assignment/CC. |
| 11 |
|
| 12 |
I see no problem with some fake alias for keywording, provided the |
| 13 |
maintainers were still contacted first to allow them to say whether a |
| 14 |
package is indeed ready for stabilization. Remember, not all |
| 15 |
stabilization/keywording bugs come from other developers/maintainers. |
| 16 |
|
| 17 |
> > Once the last remaining arch has completed the bug, it is up to them to |
| 18 |
> > close it. They know it's up to them to close it since the bug is |
| 19 |
> > assigned directly to them. |
| 20 |
> |
| 21 |
> In my opinion the last architecture should also remove the old ebuild |
| 22 |
> they have just made obsolete by stabling/keywording the new version, |
| 23 |
> since they commit to the directory anyway. |
| 24 |
|
| 25 |
This only works on cases where the older ebuild isn't in another SLOT |
| 26 |
and nothing else requires it. Yes, it *should* be cool to do this, but |
| 27 |
I think cleaning up packages/ebuilds is something best left to the |
| 28 |
maintainer. You're always welcome to say something along the lines of |
| 29 |
"last architecture to stable, please remove $ebuild when you're done" on |
| 30 |
the bugs in question. |
| 31 |
|
| 32 |
-- |
| 33 |
Chris Gianelloni |
| 34 |
Release Engineering Strategic Lead |
| 35 |
Alpha/AMD64/x86 Architecture Teams |
| 36 |
Games Developer/Council Member/Foundation Trustee |
| 37 |
Gentoo Foundation |
Archives