[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gpsd-dev] chrpath dependency
From: |
Greg Troxel |
Subject: |
Re: [gpsd-dev] chrpath dependency |
Date: |
Sat, 23 Nov 2013 15:09:04 -0500 |
User-agent: |
Gnus/5.130006 (Ma Gnus v0.6) Emacs/23.4 (berkeley-unix) |
"Eric S. Raymond" <address@hidden> writes:
>> Maybe the solution is to make chrpath a hard dependency, scons fails
>> if it is not detected, and get rid of all the static stuff in
>> SConstruct
>
> I tried that at one point. It made some distribution packagers very
> unhappy.
Was that Debian? From the NetBSD viewpoint, chrpath is a hard
dependency, because (as I understand it) without it either the
in-work-tree versions will wrongly link against installed libs to run
the test, or the installed versions will be wrong.
Is the demand not to use chrpath becuase -R is banned anyway?
Or something else?
One could make chrpath a hard dependency causing failure, unless someone
passes chrpath=no. That should make normal users not silently lose
(output buried in the build is effectively silent), and let people that
object to chrpath omit it easily.
pgpvi4b0IRlJ0.pgp
Description: PGP signature
Re: [gpsd-dev] chrpath dependency, Gerry Creager - NOAA Affiliate, 2013/11/25
Re: [gpsd-dev] chrpath dependency, Greg Troxel, 2013/11/22