Gentoo Archives: gentoo-python

From: Maxim Koltsov <maksbotan@g.o>
To: Krzysztof Pawlik <nelchael@g.o>
Cc: Nikolaj Sjujskij <sterkrig@×××××××.com>, "Michał Górny" <mgorny@g.o>, gentoo-python <gentoo-python@l.g.o>
Subject: Re: [gentoo-python] python-distutils.eclass vs. python.eclass + distutils.eclass
Date: Sat, 26 May 2012 18:45:22
Message-Id: CAB_KkxxbTKZurYzi8bp1iBLaExp8JAZuC8KWtJgh=iaUt6DiRA@mail.gmail.com
In Reply to: Re: [gentoo-python] python-distutils.eclass vs. python.eclass + distutils.eclass by Krzysztof Pawlik
1 2012/5/26 Krzysztof Pawlik <nelchael@g.o>:
2 > On 26/05/12 15:09, Nikolaj Sjujskij wrote:
3 >> Den 2012-05-26 17:07:29 skrev Micha³ Górny <mgorny@g.o>:
4 >>
5 >>> On Sat, 26 May 2012 17:01:26 +0400
6 >>> "Nikolaj Sjujskij" <sterkrig@×××××××.com> wrote:
7 >>>
8 >>>> > So I think the second part of this (x.y to x.y+1 transitions, in the
9 >>>> > Python world, are generally relatively smooth) invalidates your
10 >>>> > point in the first part: if the transitions are generally smooth,
11 >>>> > then yes, when Python 3.3 gets stabilized, I want all of my Python
12 >>>> > packages to be available from the 3.3 interpreter.
13 >>>>   Let's take a "stable" user who updates (`emerge --update --deep
14 >>>> --newuse @world`) his/her system regularly.
15 >>>> Python 3.3 is released, added to Portage tree and eventually unmasked.
16 >>>> PYTHON_TARGETS variable is changed to include 3.3. And suddenly
17 >>>> `emerge --newuse @world` on stable system suggests rebuilding of
18 >>>> every package using new eclass, because new (though disabled)
19 >>>> USE-flags was added. And when Python 3.3 is keyworded stable, hence
20 >>>> bringing new default PYTHON_TARGETS, user should now rebuild those
21 >>>> packages once more, but now, at least, not uselessly.
22 >>>>
23 >>>> Just yesterday I had www-servers/uwsgi recompiled because of changed
24 >>>> RUBY_TARGETS. And I even have no Ruby installed.
25 >>>
26 >>> I suggest you report a bug against portage and/or PMS.
27 >>  Excuse me, but I really fail to see how this could be their fault.
28 >
29 > Yes, you do. Let me explain: there was a thread some time ago about portage
30 > rebuilding package when new USE flag is introduced in ebuild that does not
31 > change enabled USE set, that's how it's related.
32
33 This was my first thought too: if just appeared flag is not set, there
34 is no sense to rebuild. But there is one possible case when this
35 assumption is false. Image that foo has support for bar and this
36 support was on by default and had no useflag. Suddenly package
37 maintainer decides, no matter why, make it optional and off by
38 default. He adds useflag to IUSE and it's not set by default. Then not
39 rebuilding it is *wrong* behavior.
40 Yes i know that this example is rather abstract and very unlikely to
41 happen, but we must consider all cases.
42
43 > --
44 > Krzysztof Pawlik  <nelchael at gentoo.org>  key id: 0xF6A80E46
45 > desktop-misc, java, vim, kernel, python, apache...
46 >

Replies