Gentoo Archives: gentoo-council

From: Brian Harring <ferringb@×××××.com>
To: gentoo-council@l.g.o
Cc: gentoo-pms@l.g.o
Subject: [gentoo-council] eapi3 riders
Date: Mon, 14 Dec 2009 11:11:23
Message-Id: 20091214110923.GE6344@hrair
1 Just nudging y'all to see if there is any complaints w/ slipping a few
2 riders in w/ eapi3 (aka prefix).
3
4 Specifically,
5
6 1) punting AA (no ebuild/eclass uses it, although
7 suppression of it is required for a few ebuilds due to env conflicts
8 w/ their build framework)
9
10 2) punting KV and it's friends- (check|get)_KV and
11 KV_(major|micro|minor|to_int). The only ebuilds/eclasses aware of
12 these vars seem to be glibc <2.5-r3 for nptl checks. This one is
13 potentially arguable although shifting it into an eclass would
14 definitely suffice if it were actually needed.
15
16 3) the fun one. mtime preservation (bug 264130). Exact wording is
17 still being tweaked (mostly screwing w/ double
18 negatives/contradictions), but it looks like the brewha on that one is
19 finally quieted down. I can reiterate the specifics of why it's
20 needed if needs be, just ask.
21
22 The reason I'm suggesting these be slipped in is pretty
23 straightforward- for the first two, it's just punting some vars/code
24 that are dead from the ebuild spec. For the last one, paludis has
25 mtime preservation code (disabled, and I've not tested it to see if
26 it's sane), pkgcore has a compliant implementation (does second level
27 resolution for merges), and portage is en route (second level is
28 doable, they're just trying to preserve NS where possible).
29
30 So... basically minimal work on the PM standpoint for mtime, and a
31 pretty useful gain for pkgs.
32
33 Thoughts? Additionally, anyother minor cleanups folks can think of?
34 ~harring

Replies

Subject Author
Re: [gentoo-council] eapi3 riders Ulrich Mueller <ulm@g.o>
Re: [gentoo-council] eapi3 riders Mike Frysinger <vapier@g.o>