Gentoo Archives: gentoo-dev

From: Jeremy Olexa <darkside@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] Gentoo Prefix: on EPREFIX, ED and EROOT inside ebuilds
Date: Fri, 20 Nov 2009 01:44:02
Message-Id: 4B05F3F3.9020902@gentoo.org
In Reply to: Re: [gentoo-dev] Gentoo Prefix: on EPREFIX, ED and EROOT inside ebuilds by Denis Dupeyron
1 Some questions answered. snipped the rest.
2
3 Denis Dupeyron wrote:
4 > 2009/10/18 Tomáš Chvátal <scarabeus@g.o>:
5 >> Why on earth portage simply does not detect the prefix enviroment is being run
6 >> and then INTERNALY switch D->ED and other variables.
7 >
8 > If that means we can get away without touching ebuilds, apart from
9 > changing their EAPI variable, then that's absolutely what we have to
10 > do. I'd like things to be done the right way though (see below).
11
12 When you change econf to do --prefix=${EPREFIX}/usr then you cannot
13 simply s/D/ED/ for everything. I hope this makes sense when you think
14 about it. ;)
15
16 src_install() {
17 emake DESTDIR="${D}" install || die
18 mv "${ED}"/usr/bin/{,exuberant-}ctags || die
19 }
20
21 But then again, some ebuilds need no changing once you fix econf to do
22 the work, which is nice.
23
24 >
25 > On Fri, Nov 13, 2009 at 4:43 AM, Ulrich Mueller <ulm@g.o> wrote:
26 >> However, there is need for additional discussion. From the council
27 >> meeting log I could extract the following open questions:
28 >
29 > It would be preferable for the discussion to happen on this list
30 > before the meeting or we'll end up postponing again due to having more
31 > questions coming up at that time.
32
33 We are willing to talk, but it always seems like the Council is "not
34 prepared" no matter what we do. Hope everyone involved can change that.
35
36 >
37 >> 2. Should the Prefix team be allowed to do the necessary changes to
38 >> ebuilds themselves, or should it be done by the respective
39 >> maintainers?
40 >
41 > I think here it's obvious that anybody who is an ebuild dev and sees
42 > anything to fix (prefix or else) is encouraged to go ahead and do it,
43 > as we've always done. The recommendation is and will always be to talk
44 > to the current maintainers out of politeness and to be extra careful
45 > (i.e. usually letting the maintainers do it) in case of
46 > system/tricky/exotic package. We don't give full cvs access to the
47 > whole tree to all ebuild devs for nothing.
48
49 It is quite obvious that we are not trying to make trouble. Talk is
50 cheap, so we prefer that. But, we see no need to ask permission to add
51 ~prefix keywords, same as other arch teams.
52
53 Currently, 'repoman -d full' will fail in some packages. We are fixing this.
54
55 > Also I think it would be nice if somebody took care of a portage
56 > patch, since I hear it is rather simple. Fabian again? Or Zac? Any
57 > other volunteers?
58 >
59 > I would prefer to have all the pieces in places before the next
60 > meeting so that we can vote on the real thing and have prefix
61 > implemented the right way before the end of the year.
62
63 portage devs and prefix devs have agreed that it is rather 'easy' to
64 merge the prefix-portage branch. Just waiting.. ;) We have access to
65 check into the portage repo, so this should not hold anything up
66 regarding any decisions.
67
68 >
69 >> 6. (Any question that I've missed?)
70
71 >> How are scripts using #!shebangs going to work?
72 >> You write an ebuild, and you DEPEND upon >=foo-3, because the build
73 >> process includes some foo code. The foo code is executed via
74 >> scripts using #!/usr/bin/foo. Normally, this is fine.
75 >> But on prefix, /usr/bin/foo might be a crappy, OS X mangled foo-2
76 >> that's no good. So even though you've got the foo-3 dep met, it'll be
77 >> met in /opt/Gentoo/blah, so your package will fail.
78
79 The prefix-portage branch has a nice feature that fixes shebangs
80 automatically to be ${EPREFIX}/foo instead of /foo. It has even caught
81 some Gentoo Linux bugs.
82
83 >
84 >> How are ebuilds to be marked as supporting prefix or not?
85 > (Here I'm guessing that changing the EAPI variable will do)
86
87 Gentoo Prefix has keywords. So if EAPI 3 has ED/EROOT support but the
88 ebuild doesn't use them then the ebuild does not need an EAPI bump. In
89 this case, please rephrase your question to be "How are ebuilds to be
90 marked as working on a prefix arch or not?" and then it is clear that it
91 is the same as Gentoo Linux.
92
93 >
94 >> Why is there only a single permitted installation path?
95 > (I'm under the impression this is a limitation of the windows
96 > installer but not of prefix itself. So patching the installer would
97 > fix that)
98
99 My installation path on my 6-8 prefix arches is in my NFS home. If you
100 are referring to the Windows special installation package, well..that is
101 just a "stage4" installer with binary packages. The windows installer is
102 no where near the heart of Gentoo Prefix, instead it is a product of
103 Gentoo Prefix and a convenience factor offered by another Prefix dev. It
104 showcases the possibilities quite well, IMO.
105
106 You can set EPREFIX to anything. One of our users even set it to "/" -
107 which we don't endorse but it is possible. :)
108
109 >
110 >> What exactly is expected from a prefix-compliant package manager to
111 >> support full prefix installs, as opposed to just supporting installs
112 >> to / with prefix-aware ebuilds?
113 > (The PMS patch should answer that)
114 >
115 > Denis.
116 >

Replies