Gentoo Archives: gentoo-dev

From: Brian Dolbec <dolsen@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] Re: Avoiding rebuilds
Date: Mon, 28 Jul 2014 15:16:36
Message-Id: 20140728081537.2f8a8199.dolsen@gentoo.org
In Reply to: [gentoo-dev] Re: Avoiding rebuilds by Martin Vaeth
1 On Mon, 28 Jul 2014 05:49:07 +0000 (UTC)
2 Martin Vaeth <martin@×××××.de> wrote:
3
4 > hasufell <hasufell@g.o> wrote:
5 > > Ulrich Mueller:
6 > >>
7 > >> I wonder if it wouldn't be saner to leave our revision syntax
8 > >> untouched.
9 >
10 > As already mentioned, -r1.1 is only one of several possible ways
11 > how to achieve the same aim; I am not speaking in favour for a
12 > particular method.
13 > The -r1.1 method has the advantage of being simple and transparent
14 > to the user and developer. Other approaches have other advantages:
15 >
16 > >> Instead, one could introduce a variable INSTALL_VERSION that would
17 >
18 > (It would have to be a variable stored in the metadata/ cache
19 > and thus also would only work with a new API, but these are only
20 > technical details.)
21 >
22 > >> default to ${PVR} but could be set to the version of a previous
23 > >> ebuild instead. The PM could compare it against INSTALL_VERSION in
24 > >> the VDB and skip build and installation if versions match.
25 >
26 > It should be a list and have empty default (*never* including the
27 > version itself), but these are also technical details.
28 > This solution would have the advantage that you could specify
29 > *full* versions and thus have even more fine-grained control when
30 > recompilations are necessary. One could also allow specify version
31 > ranges, slots, overlays, etc., perhaps even make the behaviour
32 > dependent of USE-flags, as you already mentioned, all
33 > similarl to current DEPEND syntax.
34 >
35 > The disadvantage is that it is slightly more work than -r1.1,
36 > less transparent, and easily overlooked to remove for a version bump,
37 > causing issues like these:
38 >
39 > > It will probably also cause confusion for comaintainers and
40 > > collaborators, especially when INSTALL_VERSION points to a version
41 > > that has already been removed.
42 >
43 >
44
45 I haven't had the energy to read all the mails over all the dynamic
46 deps thread...
47
48 the -r1.1 syntax has been in use by the prefix since early in it's
49 existence. I haven't kept track, but they may have finally done away
50 with it.
51
52 There are many other problems with using that syntax, namely most other
53 tools are not compatible with it, so more than just portage needs to be
54 modified. Adding that syntax to ebuilds will cause disruptions and bugs
55 for years to come.
56
57 So, please, forget about this syntax as a viable solution.
58
59 --
60 Brian Dolbec <dolsen>