From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <gentoo-user+bounces-164987-garchives=archives.gentoo.org@lists.gentoo.org>
Received: from lists.gentoo.org (pigeon.gentoo.org [208.92.234.80])
by finch.gentoo.org (Postfix) with ESMTP id EBACE138CF8
for <garchives@archives.gentoo.org>; Tue, 23 Jun 2015 18:34:35 +0000 (UTC)
Received: from pigeon.gentoo.org (localhost [127.0.0.1])
by pigeon.gentoo.org (Postfix) with SMTP id 9D73C14106;
Tue, 23 Jun 2015 18:34:13 +0000 (UTC)
Received: from km35626.keymachine.de (km35626.keymachine.de [87.118.86.27])
(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))
(No client certificate requested)
by pigeon.gentoo.org (Postfix) with ESMTPS id 155C6140FB
for <gentoo-user@lists.gentoo.org>; Tue, 23 Jun 2015 18:34:11 +0000 (UTC)
Received: from localhost (localhost [127.0.0.1])
by km35626.keymachine.de (Postfix) with ESMTP id C053C1121F35
for <gentoo-user@lists.gentoo.org>; Tue, 23 Jun 2015 20:35:54 +0200 (CEST)
X-Virus-Scanned: Debian amavisd-new at km35626.keymachine.de.
Received: from km35626.keymachine.de ([127.0.0.1])
by localhost (km35626.keymachine.de [127.0.0.1]) (amavisd-new, port 10024)
with ESMTP id RqUUBAgMH2DB for <gentoo-user@lists.gentoo.org>;
Tue, 23 Jun 2015 20:35:53 +0200 (CEST)
Received: from grusum.endjinn.de (dslb-094-216-148-168.094.216.pools.vodafone-ip.de [94.216.148.168])
by km35626.keymachine.de (Postfix) with ESMTPSA id A1C681121EDC
for <gentoo-user@lists.gentoo.org>; Tue, 23 Jun 2015 20:35:53 +0200 (CEST)
Received: by grusum.endjinn.de (Postfix, from userid 500)
id AF10417004C; Tue, 23 Jun 2015 20:05:51 +0200 (CEST)
Date: Tue, 23 Jun 2015 20:06:16 +0200
From: David Haller <gentoo@dhaller.de>
To: gentoo-user@lists.gentoo.org
Subject: Re: [gentoo-user] necessary use flgas
Message-ID: <20150623180616.GA12427@grusum.endjinn.de>
Mail-Followup-To: gentoo-user@lists.gentoo.org
References: <CAO5-k+ojMQ5Z5kzoXPVCsNM=ke66PWugfkqR4bw1-jZ9GXN1vg@mail.gmail.com>
<20150623143851.7773394d@digimed.co.uk>
<20150623144558.GB16098@grusum.endjinn.de>
<20150623163150.3be30d2f@hactar.digimed.co.uk>
<20150623163532.GA3929@grusum.endjinn.de>
<20150623180802.79094832@hactar.digimed.co.uk>
Precedence: bulk
List-Post: <mailto:gentoo-user@lists.gentoo.org>
List-Help: <mailto:gentoo-user+help@lists.gentoo.org>
List-Unsubscribe: <mailto:gentoo-user+unsubscribe@lists.gentoo.org>
List-Subscribe: <mailto:gentoo-user+subscribe@lists.gentoo.org>
List-Id: Gentoo Linux mail <gentoo-user.gentoo.org>
X-BeenThere: gentoo-user@lists.gentoo.org
Reply-to: gentoo-user@lists.gentoo.org
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-15
Content-Disposition: inline
In-Reply-To: <20150623180802.79094832@hactar.digimed.co.uk>
User-Agent: Mutt/1.5.23 (2014-03-12)
X-Archives-Salt: ef2705df-0cf6-4b8a-9287-f20f2e33ef68
X-Archives-Hash: ad067dc1c7614a2cc074afafc4ff4f0f
Hello,
On Tue, 23 Jun 2015, Neil Bothwick wrote:
>On Tue, 23 Jun 2015 18:35:32 +0200, David Haller wrote:
>> >You do know that only the desktop profiles include cups as a default
>> >USE flag? The one I recommended does not.
>>
>> Try compiling icedtea, libreoffice, scribus and whatnot without
>> pulling in cups, no matter the use-flags. And "sabotaging" the ebuild
>> and buildsystem to not use cups leads to failed builds, BT,TriedThat :((
>
>The OP wanted to set up a "snappier" system. Java and LibreOffice are
>not the first programs that spring to mind when I think snappier...
Well, I use LaTeX anyway for "office", but one needs stuff that mom is
used to, so one can talk her through stuff, eh? And Java? There's e.g.
tvbrowser and MediathekView. No need to talk about snappy but I
digressed from the OP anyway already. I just happily jumped on the
topic of leaving stuff out that one doesn't need. And generally stuff
you don't have installed cannot be attacked, esp. such ubiquitously
used stuff as cups (used by MacOS/iOS too), is an important attack
vector less installed.
Actually: mom uses libreoffice-calc to edit a .csv file, that is then
fed to a perl-script by me (via a couple of links on the XFCE/formerly
WinXP Desktop (via a .cmd batch) calling it differently), that
generates a LaTeX file using labels.sty that is fed to pdflatex and
spits out a PDF to be printed on labels, and even starting a
pdf-viewer to check it before printing ;) Worked just like a charm
for, ah, about 7+ years without any maintanece required, but
recentenly, mom must've borked up the charset on saving multiply,
probably due to changed defaults in libreoffice, looked like double
encoded utf8, but was borked even beyond that. Manually fixing it
turned out to be the least work. *Gah*. As mom wanted to weed out
outdated stuff anyway, she did it, but we talked about it and I'd had
done it.
Except from that, mom writes her letters and stuff with -writer, and
has been doing so since 199x (then with StarOffice).
>> Why does a GUI (a USER INTERFACE Toolkit fer f*** sake! Not a Printer
>> Interface!) _ALWAYS_ be able to print (if I interpret the AWT right)?
>> That's just dumb.
>
>No argument there.
*MEH* :) Yeah, there's a couple of "dumb" deps by upstream, that are
not configurable and not even easily patched out (I think I have one
where I could patch, but usually, it's too hardwired in, so to speak).
Makes you want to grab a fish (fresh from Lutetia), and slap the
culprit around the head with it ... And boy! Are we in for something
getting systemd hardwired as a dep ... *cringe*
>> [1] speaking of that: I noticed, that when I bork[2] an ebuild in
>> /usr/local/portage, it gets silently(!) ignored and the one from
>> /usr/portage is used. The only indication is the flag on the
>> package e.g. ::gentoo vs. ::local. Only once I move the gentoo
>> ebuilds into e.g. the .attic subfolder, emerge tells me what's
>> actually wrong with my ::local ebuild in /usr/local/portage. Have
>> I overlooked an option of emerge or is that a bug?
>
>ebuild /usr/local/portage/cat/pkg/pkg-x.y.ebuild merge
>
>will use the specific ebuild you give it.
Got to alias/script that! But it is a clumsy workaround. As your local
overlay (or any with a higher precedence) should override the base,
emerge should at least tell you about the problem with the overlay,
and then e.g. ask to emerge the base (/usr/portage), or abort. How
about it? I consider it a bug (unless I and Neil overlooked a switch
to emerge, and even then, I'd be for a different default of that, as
hey, if I do an overlay, I want to be told if I borked anything there,
not just almost quietly ignored, but anyways not told the actual
error, until I remove the /usr/portage version (again: missed option
to emerge?))
@all: What's your take on this? Have I (and Neil?) missed an option?
or has emerge a "sort of a bug"?
-dnh, who has not yet ever looked at emerge code, but guessing it
should not be much of a problem emitting the errors in the overlay
and some simple handling afterwards ;)
--
vi, pr.n.
A computer program designed to stress-test the use of modal bleeping.