Util-Linux package development
 help / color / mirror / Atom feed
From: Karel Zak <kzak@redhat.com>
To: Joel Holdsworth <joel@airwebreathe.org.uk>
Cc: Bernhard Voelker <mail@bernhard-voelker.de>,
	util-linux@vger.kernel.org,
	Joel Holdsworth <joel.holdsworth@vcatechnology.com>
Subject: Re: [PATCH] build-sys: cal requires both ncurses and tinfo
Date: Mon, 27 Jul 2015 11:55:07 +0200	[thread overview]
Message-ID: <20150727095507.GA6865@ws.net.home> (raw)
In-Reply-To: <55B276F1.5070706@airwebreathe.org.uk>

On Fri, Jul 24, 2015 at 06:33:37PM +0100, Joel Holdsworth wrote:
> Ok, so cal can build without ncurses and/or tinfo - fine ok.
> 
> The problem I have is with misc-utils/Makemodule.am:14
> 
> https://git.kernel.org/cgit/utils/util-linux/util-linux.git/tree/misc-utils/Makemodule.am?h=stable%2Fv2.26#n14
> 
> ...this is making it link with -ltinfo, which fails because I'm not
> providing it. Would it be more correct to make NCURSES_LIBS and
> TINFO_LIBS become empty when they're not available? 

It should be empty, see configure. It sets TINFO_LIBS by
PKG_CHECK_MODULES() or AC_CHECK_LIB().

> I'm slightly suspicious of my cross-build setup if configure is picking
> up "-ltinfo" dependencies from the host system, but even with
> '--disable-tinfo', TINFO_LIBS is still not set to "".

You want to use --without-tinfo (see configure --help):

$ ./configure &> /dev/null
$ grep "TINFO_LIBS =" Makefile
TINFO_LIBS = -ltinfo 

$ ./configure --without-tinfo &> /dev/null
$ grep "TINFO_LIBS =" Makefile
TINFO_LIBS =


> Also having UL_BUILD_INIT([cal], [yes]) to force cal to always be built
> seem ridiculous; in fact why should any of the tools be forced? The user
> should be able to mix and match whatever programs they want to have.

Well, the script uses [yes] because cal does not have any strict
dependence, so [check] does not have any sense here.

> Another related issue is that I was hoping to be able to do
> '--disable-all-programs', then '--enable-XXX' the ones I need one by one
> - but the disable-all flag seems to take precedence right now.

I don't think so. For example:

 ./configure --disable-all-programs --enable-unshare

works as expected. The problem is that we currently don't have
--enable- for all stuff, for example --enable-cal is missing.

Fixed now. I have added --disable-cal because it seems that cal(1) is
tricky enough due to compilation with/without ncurses/termpac/tinfo.

    Karel

-- 
 Karel Zak  <kzak@redhat.com>
 http://karelzak.blogspot.com

  parent reply	other threads:[~2015-07-27  9:55 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-23 22:56 [PATCH] build-sys: cal requires both ncurses and tinfo Joel Holdsworth
2015-07-24  6:23 ` Bernhard Voelker
2015-07-24  7:56   ` Karel Zak
2015-07-24 17:33     ` Joel Holdsworth
2015-07-24 20:09       ` Karel Zak
2015-07-27  9:55       ` Karel Zak [this message]
2015-07-24  7:51 ` Karel Zak

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20150727095507.GA6865@ws.net.home \
    --to=kzak@redhat.com \
    --cc=joel.holdsworth@vcatechnology.com \
    --cc=joel@airwebreathe.org.uk \
    --cc=mail@bernhard-voelker.de \
    --cc=util-linux@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox