Re: [gpsd-dev] Configuring GitLab CI (was Upcoming migration to GitLab)

[Gary E. Miller]

Yo Hal!

On Mon, 24 Jun 2019 12:44:41 -0700
Hal Murray <address@hidden> wrote:

> address@hidden said:
> > I get those, a lot, on systems where do not run dbus.  The test
> > should still pass.  Maybe change that to a warning?   
> 
> Why is a test trying to connect to dbus?

The regress-driver starts gpsd on a socket, feeds the socect the test
data, and gets the JSON back from gpsd.

The test gpsd is started normally, except for the socket and port.
Part of normal gpsd startup is to try to connecto to the local dbus.

> There should be a run-time no-dbus switch so the tests can run
> clean.  Or two lines of code in the test to check the UID?

Patches welcome.  Never been a real problem before.  Only a cosmetic
annoyance.  I got lots of real problems to solve.

> There is a configure time switch.  If the rest of the dbus test
> doesn't do anything interesting, you can rebuild with
> "dbus_export=no".

But then you are no longer testing the gpsd as the user wants to build
and use it.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can’t measure it, you can’t improve it." - Lord Kelvin

pgpdYR_td2eAf.pgp
Description: OpenPGP digital signature