[ gpsdrive ] help with help

[Hamish]

Hamish:
> > What does the command line -f flag actually do?
> >  -f, --force-position   force display of position even it is invalid

Guenther:
> hmm, that's a very old option, it doesn't really make sense to me...

Ah I get it now, it is talking about if the NMEA sentence fix is_valid
flag. I didn't realize nmea_handler.c was still being used.

> I think, we should remove it.

I notice that the variable is still used in convertRMC(), which is still
used too.

Any reason to continue to parse NMEA ourselves? If we let gpsd do that
for us then users of USB Sirf binary protocol and Garmin binary protocol
GPSs etc could use GpsDrive and we could simplify our end.
?

> > The mouse click help section (from the Help menu) says:
> > "Middle mouse button: Display position again"
> > what does that mean? does it still work?
>
> the code switches off posmode, when clicked.
> so it means "leave posmode", I guess...

oh, I had always used right-click for that.

> > If the middle mouse click was free, I think it might be nice if it
> > triggered the canvas to redraw with the map centered on the click
> > position.
>
> you mean switching to posmode with mouseclick?

no. nothing to do with posmode. often the detail of what intersection
is ahead eg is slightly off the screen. currently my method to re-center
the screen on the gps location is to zoom-in 16x then zoom-out back to
normal. That makes the gps in the middle and what I want to see on the
screen. I'd like to have the middle mouse button click redraw the sub-
tile red rectangle centered on the click pos, or as close as possible
if that would make the GPS position go more than (20+5)% from an edge.
(so it doesn't just swap back in that case)

> because in normal mode the map view is determined by the
> current position.

just sometimes map view can be in-view but off to a side.

 
> > Even better would be middle-drag to move it around as well. 
>
> a drag function would be really nice, yes. but I think to do this
> properly some significant changes to the map handling code have to
> be done.

ok.
(wish: GDAL from any geo datasource + real projection support too! :)

> > I am not sure if new bug reports get forwarded here or to the sf-svn
> > mailing list or nowhere; if not the new titles are:
>
> the sf-svn list only sends out commit messages from the svn, nothing
> else.
>
> bug reports should be added to the bugtracker on the sf-website. this
> would be the best way to track things, that can't be fixed very
> soon...
> or reported to this list.

my point was that I put some new bugs on the tracker but they do not
seem to be annouced so no one knows to look for them and they don't
get disccussed. No notification mechanism.


> > - --geometry command line control is broken
> this means?

details in the bug report,
  http://sourceforge.net/tracker/?group_id=148048


Hamish