Gentoo Archives: gentoo-dev

From: Mart Raudsepp <leio@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] autotools.eclass: automatically move configure.in to configure.ac
Date: Sun, 11 Jun 2017 20:02:16
Message-Id: 1497211321.24558.7.camel@gentoo.org
In Reply to: Re: [gentoo-dev] autotools.eclass: automatically move configure.in to configure.ac by William Hubbs
1 Ühel kenal päeval, P, 11.06.2017 kell 13:08, kirjutas William Hubbs:
2 > On Sun, Jun 11, 2017 at 07:14:52PM +0300, Mart Raudsepp wrote:
3 > > Ühel kenal päeval, P, 11.06.2017 kell 11:12, kirjutas William
4 > > Hubbs:
5 > > > On Sun, Jun 11, 2017 at 05:35:53PM +0200, Kristian Fiskerstrand
6 > > > wrote:
7 > > > > On 06/11/2017 05:24 PM, David Seifert wrote:
8 > > > > > > We can always patch the eclass at that point if that is
9 > > > > > > still a
10 > > > > > > big
11 > > > > > > concern, but I fundamentally agree with William on this,
12 > > > > > > starting
13 > > > > > > point
14 > > > > > > should be fixing it upstream, so can start with a tracking
15 > > > > > > bug
16 > > > > > > on
17 > > > > > > affected packages.
18 > > > > > >
19 > > > > >
20 > > > > > Here's a deal: you can start filing + fixing them.
21 > > > > >
22 > > > >
23 > > > > The [tracker] is already started, it was determined to add QA
24 > > > > warning
25 > > > > with info on filing upstream bugs in [bug 426262] and the
26 > > > > proper
27 > > > > solution is again iterated in [bug 546614], so its not like
28 > > > > this is
29 > > > > a
30 > > > > new approach that is being suggested, but sure, we should all
31 > > > > file
32 > > > > bugs
33 > > > > when we encounter them.
34 > > > >
35 > > > > References:
36 > > > > [tracker] https://bugs.gentoo.org/show_bug.cgi?id=530632
37 > > > >
38 > > > > [bug 426262]
39 > > > > https://bugs.gentoo.org/show_bug.cgi?id=426262
40 > > > >
41 > > > > [bug 546614]
42 > > > > https://bugs.gentoo.org/show_bug.cgi?id=546614
43 > > >
44 > > > It seems that the proper fix to this, even for a package that
45 > > > won't
46 > > > do
47 > > > the fix upstream is to use WANT_AUTOCONF in the ebuild to force
48 > > > the
49 > > > version of autoconf you need.
50 > >
51 > > No. It appears you don't know how WANT_AUTOCONF works.
52 >
53 >  
54 >  From the eclass:
55 >
56 > # @ECLASS-VARIABLE: WANT_AUTOCONF
57 > # @DESCRIPTION:
58 > # The major version of autoconf your package needs
59 >
60 > That leads me to believe that you can set WANT_AUTOCONF="someversion"
61 > in your ebuild and your package will use that version of autoconf.
62 >
63 > If that's wrong, it needs to be fixed and that's a different bug
64 > entirely, but it doesn't change my position on adding this particular
65 > change to autotools.eclass.
66
67 It is the major version, as documented. Yes, it could specify what
68 these valid versions currently are, as they really are:
69
70 none
71 2.1
72 2.5
73 latest
74
75 Currently latest equals 2.5 and is the default.
76
77 In practice, none means not to add an autoconf atom to DEPEND (even
78 with the default AUTOTOOLS_AUTO_DEPEND=yes) - if ebuild/eclass
79 inheriting autotools.eclass handles its own exact autoconf depend atom
80 (eautoconf will get called in eautoreconf regardless). Only main tree
81 consumer of this appears to be gtk-sharp-module.eclass that adds its
82 own autoconf/automake atoms when needed, because eautoreconf is
83 conditional by variables used by that eclass and it needed autoconf
84 2.61 or newer, not just 2.59.
85
86 2.1 means autoconf:2.1
87
88 2.5 and latest currently means >=autoconf-2.59
89
90 Patches welcome to documentation, I suppose.
91
92
93 It is usually a bad sign if you need to change it away from latest for
94 some reason in the ebuild and ideally they'd all be the default
95 (latest).
96
97 There was no configure.ac before autoconf-2.50, only configure.in, and
98 thus a warning doesn't make sense. The real warning here is the need
99 for WANT_AUTOCONF=2.1
100
101
102 HTH,
103 Mart

Replies