On Thursday 03 February 2005 01:21, Ciaran McCreesh wrote:
(B> On Wed, 2 Feb 2005 21:58:13 +0900 Jason Stubbs <[EMAIL PROTECTED]> wrote:
(B> | On Wednesday 02 February 2005 21:44, petre rodan wrote:
(B> | > Michael Cummings wrote:
(B> | > > To be frank, your initial email scared me away from upgrading to
(B> | > > -r15. Overlay's are how I keep my world sane (ok, that's a
(B> | > > stretch, but as far as Gentoo is concerned its valid).
(B> | >
(B> | > my thoughts exactly.
(B> |
(B> | All the more reason for you to test it, no?
(B>
(B> Not really. Testing on the live tree is a Bad Thing (TM).
(B
(BLet me explain this again. <sys-apps/portage-2.0.51-r15 uses PORDIR_OVERLAY
(Bfor dep checking even when committing. sys-apps/portage-2.0.51-r15 doesn't.
(BThat in itself, is a reason to upgrade.
(B
(BThis patch is meant to restore the usability of PORTDIR_OVERLAY when repoman
(Bis called from PORTDIR_OVERLAY, which it does. I asked for testing because of
(Bthe size of the patch. The norm is just to add the patch without testing
(B(beyond my own) and find out the bugs when the release hits the live tree.
(B
(B> Don't suppose you portage people have a fake tree and a set of regression
(B> tests for repoman?
(B
(BNo, there aren't yet. Nor will there be for some time. Until there is, I was
(Bhoping that the team would help to test things that are done specifically for
(Bthem. What is this? Always screaming for features but then complaining
(Bbecause they aren't delivered to your liking? Fine. If nobody (or very few)
(Bwants to test, I'll just add the patch to HEAD and forget about it for the
(B2.0.51 line so that it is tested through pre-releases.
(B
(BRegards,
(BJason Stubbs
(B
(B--
([email protected] mailing list
