Re: pkg/53673: Building Python 3.7 on Linux fails with error "pkg_create: can't stat" at stage "Generating post-install file lists"

["David H. Gutteridge" <david%gutteridge.ca@localhost>]

The following reply was made to PR pkg/53673; it has been noted by GNATS.

From: "David H. Gutteridge" <david%gutteridge.ca@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: thomas.orgis%uni-hamburg.de@localhost
Subject: Re: pkg/53673: Building Python 3.7 on Linux fails with error
 "pkg_create: can't stat" at stage "Generating post-install file lists"
Date: Mon, 25 Mar 2019 19:16:09 -0400

 On Mon, 25 Mar 2019 at 13:20:35 +0100, Dr. Thomas Orgis wrote:
 > I just hit this on my Linux box. There is no NIS support and python27
 > and python36 fail. Regarding this logic:
 > 
 > .  elif ${OPSYS} != "NetBSD" || exists(/usr/bin/ypcat)
 > 
 > Is this really what is intended? It seems more sensible to me when
 > changed to
 > 
 > .  elif ${OPSYS} != "NetBSD" && exists(/usr/bin/ypcat)
 > 
 > I actually have a Linux system where NIS is employed, and there
 > is /usr/bin/ypcat. Shouldn't the logic use that to decide if NIS is
 > built or not, instead of forking out at != NetBSD and happily assuming
 > that there is NIS?
 
 This has been the subject of more than one bug report, and a recent
 discussion on pkgsrc-users began here:
 http://mail-index.netbsd.org/pkgsrc-users/2019/02/22/msg028082.html
 
 A proposed alternate means of fixing the issue was logged in another
 PR:
 https://gnats.netbsd.org/cgi-bin/query-pr-single.pl?number=54056
 
 The absence of /usr/bin/ypcat isn't necessarily a reliable indicator
 that the nis module can't be built, e.g. on recent versions of Fedora.
 By the way, which Linux distribution(s) are you using?
 
 Dave