gpsd-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: regressions pass on NetBSD 9 / earmv7hfel (RPI3)


From: Greg Troxel
Subject: Re: regressions pass on NetBSD 9 / earmv7hfel (RPI3)
Date: Tue, 03 Aug 2021 19:00:12 -0400
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (berkeley-unix)

"Gary E. Miller" <gem@rellim.com> writes:

>> Sure, but pkgsrc rules are that you pick one of
>> 
>>   have an expressed dependency so that the prereq is always there
>> 
>>   don't depend on it, and force it off, so that when you build, you
>>   always get the same thing, not depending on whether something (not
>>   declared) is there or not
>
> I don't see the problem.  If you compile in dbus, it gets used if
> available.

I will look at how big dbus is.    If I add it as a dependency then
everybody who installs gpsd will end up installing dbus.

>> a fair point that perhaps gpsd should depend on dbus if lots of people
>> care.  But it was not depending and picking it up, which isn't good.
>
> That will never happen.  dbus is for desktops, most gpsd is not desktop.

It did happen.  I built the gpsd package, which doesn't have an explicit
dependency *in the packaging system* on dbus, on a system which happened
to have dbus installed, and the build system detected dbus and used it.
Not a big deal -- that's normal -- but if I have not configured the
package build to say it's required, it's a bug to have it picked up.

In pkgsrc, the default build of dbus depends on X11, by way of
dbus-launch.  Which is ok because as you say dbus is a desktop thing.  I
definitely don't want to add a dependency on X11 to gpsd.  Probably I
should add a build option to the package, but I may leave that to the
first person that wants it.

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]