Gentoo Archives: gentoo-portage-dev

From: Alec Warner <antarus@g.o>
To: gentoo-portage-dev@l.g.o
Cc: Joakim Tjernlund <Joakim.Tjernlund@××××××××.com>, "mgorny@g.o" <mgorny@g.o>, "grobian@g.o" <grobian@g.o>
Subject: Re: [gentoo-portage-dev] In what phase are file "merged"?
Date: Fri, 06 Aug 2021 20:53:50
Message-Id: CAAr7Pr9zwB6Rs6=6vVHHpVkY9x6+dbRWeD1WfYUqAy291HxF=g@mail.gmail.com
In Reply to: Re: [gentoo-portage-dev] In what phase are file "merged"? by Ulrich Mueller
1 On Thu, Aug 5, 2021 at 9:23 PM Ulrich Mueller <ulm@g.o> wrote:
2 >
3 > >>>>> On Fri, 06 Aug 2021, Joakim Tjernlund wrote:
4 >
5 > > On Wed, 2021-06-23 at 13:33 +0200, Michał Górny wrote:
6 > >> On Wed, 2021-06-23 at 12:40 +0200, Ulrich Mueller wrote:
7 > >> > I don't think that the ebuild can rely on any particular status of
8 > >> > the new package in pkg_*rm (of the old package), or the status of
9 > >> > the old package in pkg_*inst (of the new package).
10 > >>
11 > >> I would even say that it can't rely on the particular status of the
12 > >> old package in any case, if it's meant to be removed. In particular,
13 > >> its dependencies can be unmerged before the package itself.
14 >
15 > > Stubled ove this mail again and noticed "its dependencies can be
16 > > unmerged before the package itself" stmt. That does not make sense to
17 > > me. Deps should be unmerged after any pkg that depends on them?
18 >
19 > A popular workflow is "emerge -c -p" followed by "emerge -C" on entries
20 > of the list shown. IIUC emerge -C doesn't do any dependency resolution,
21 > therefore ebuilds cannot rely on any removal order.
22
23 Not quite sure I follow. Let's assume I have A -> B -> C.
24
25 Is it legal for A to call a binary packaged in A in A's pkg_prerm?
26 If yes, then B and C have to be on the livefs at least until A's
27 pkg_prerm has run; right?; otherwise if we unmerged B or C before then
28 we might break A's binaries?
29
30 -A
31
32 >
33 > Ulrich