Re: [gpsd-dev] Fw: [gpsd-commit-watch] [SCM] GPSD branch, master, updated. dev-3.19a-744-g4b1c33b

[Gary E. Miller]

Yo Fred!

On Tue, 28 May 2019 18:11:10 -0700 (PDT)
Fred Wright <address@hidden> wrote:

> On Tue, 28 May 2019, Gary E. Miller wrote:
> >
> >    "This fix, first added in commit 9af0ab644, was inappropriately
> >    removed by commit 6e2842d3e.  It now has a more detailed comment
> >    explaining why it's not redundant with the other filtering."
> >
> > How about fixing this case for ALL clients and not just xgps?  
> 
> It was already fixed for all clients (see the original commit).  It
> only got unfixed for xgps, so it only needed to be refixed for xgps.

How can it possibly be fixed for all clients but one?  There are a
lot of gpsd clients not in our repo.  It should be fixed in libgps_json
just before SKY is sent.

Otherwise you end up with a lot of slightly different fixes, many of
them slightly wrong.

> > The only place you could be with a SBAS bird at 0/0 would be
> > somewhere south east of Easter Island .  
> 
> The filter isn't SBAS-specific, since that would make it more
> complicated.

Really?  Just test for gnssid=1

> But even a nav satellite at 0/0 wouldn't be actually
> used by most receivers, due to the masking angle, so the potential
> false positive is fairly benign.

Usually.  But gpsd is used in space, then the elevation mask is set to
-10 or -25.

> Requiring *both* el and az to be
> zero is to minimize this.

Good thing, otherwise triggers much too often.

Can you send us a raw capture of the GPS misbehaving?

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

pgp8sd9EkJKJF.pgp
Description: OpenPGP digital signature