[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Nmh-workers] nmh on openbsd...
From: |
David Levine |
Subject: |
Re: [Nmh-workers] nmh on openbsd... |
Date: |
Thu, 17 Apr 2014 23:36:33 -0400 |
William wrote:
> In preparing an update to the OpenBSD port to 1.6-RC1, we found it
> failed to compile uip/rcvtty due to this commit:
>
> http://git.savannah.gnu.org/cgit/nmh.git/patch/?id=6443ed24f6c3b41186636a0b0f
> ea93ebe9452047
>
> It results in:
>
> cc -O2 -pipe -o uip/rcvtty uip/rcvtty.o uip/scansbr.o sbr/libmh.a -ltermca
> p 2>&1 | sed -e 's/: w\(arning: s.*() is .* misused, please use\)/: W\1/'
> sbr/libmh.a(sbr_libmh_a-m_getfld.o)(.text+0x6d5): In function `m_unknown':
> sbr/libmh.a(sbr_libmh_a-mts.o)(.text+0x5be): In function `LocalName':
> sbr/libmh.a(sbr_libmh_a-m_maildir.o)(.text+0x13e): In function `m_maildir':
> uip/rcvtty.o(.text+0x6ef): In function `main':
> : undefined reference to `alert'
> collect2: ld returned 1 exit status
>
> How we hit this needs a bit of explanation.
Even with that explanation, I don't understand how that
commit broke that link. First let me ask, is
HAVE_GETUTXENT #defined in config.h? If not, why is there a call
to alert()? All appearances of it are protected by
#if HAVE_GETUTXENT.
> After the failed discussion here on nmh-workers about utmp support, in
> which we attempted to resurrect it with this tiny patch:
>
> http://lists.nongnu.org/archive/html/nmh-workers/2013-01/msg00018.html
Ken responded in detail:
http://lists.nongnu.org/archive/html/nmh-workers/2013-01/msg00020.html
Can we continue the discussion from the points that he raised?
Thanks,
David