Gentoo Archives: gentoo-user

From: Alan McKinnon <alan.mckinnon@×××××.com>
To: gentoo-user@l.g.o
Subject: Re: [gentoo-user] A portage nuisance
Date: Sat, 28 Oct 2017 19:59:32
Message-Id: b113737a-dc04-4e3a-34ee-7bfff6b019ce@gmail.com
In Reply to: Re: [gentoo-user] A portage nuisance by Wols Lists
1 On 28/10/2017 19:58, Wols Lists wrote:
2 > On 28/10/17 15:52, Andrew Savchenko wrote:
3 >> On Fri, 27 Oct 2017 14:58:13 +0100 Peter Humphrey wrote:
4 >>>> On Fri, 27 Oct 2017 12:52:54 -0000
5 >>>> Helmut Jarausch <jarausch@××××××.be> wrote:
6 >>>>
7 >>>>>> I have a problem with emerge for a long time.
8 >>>>>> Sometimes I need to (re-)emerge many packages like in an
9 >>>>>> emerge --emptytree @world
10 >>>>>>
11 >>>>>> Because I use several overlays, there are problems with a lot of
12 >>>>>> packages.
13 >>>>>> Unfortunately, emerge shows me just the first problem (like a missing
14 >>>>>> USE-flags) and then terminates.
15 >>>>>> Is there any means to let emerge go and report several (all) problems
16 >>>>>> which are independent of each other?
17 >>>>
18 >>>> EMERGE_DEFAULT_OPTS="--keep-going" ?
19 >> No, --keep-going allows to continue as long as possible after a
20 >> build failure. Helmut asks about dependecies resolution failures,
21 >> e.g. in some package REQUIRED_USE is not met, or circular
22 >> dependency occurs and so on.
23 >
24 > What I would like - a bit like --keep-going - is some option that tries
25 > again.
26 >
27 > When I do an "emerge -u" it sometimes blows up with this massive load of
28 > dependency failures. So what I end up doing is emerge a few packages
29 > that look like they're going to work, and then try my full update again.
30 > After several cycles through this, suddenly everything works.
31 >
32 > So my spec for what I would like is basically, as each package
33 > successfully resolves its dependencies, add it to a "try again" list. If
34 > the current list blows up in dependency hell, restart the emerge with
35 > just the packages in the "try again" list.
36
37
38 That is completely contrary to how portage is supposed to work and
39 produces an unpredictable result.
40
41 *You* know what you want and can do it, *you* are a human with a brain
42 who understands meaning.
43
44 Portage cannot do that, it is backed by silicon and has no concept of
45 meaning. So it has only one real choice - it can do it all or it does
46 not try.
47
48 I'm not surprised Zac never tried implementing partial graph resolution
49 for the very simple reason that if you try do it, you have no idea what
50 is going to be built. That is the opposite of what portage must deliver.
51 --
52 Alan McKinnon
53 alan.mckinnon@×××××.com

Replies

Subject Author
Re: [gentoo-user] A portage nuisance Anthony Youngman <antlists@××××××××××××.uk>