[Flightgear-devel] Route Manager leg properties changed from 3.2 to 3.5

[<ric...@gm...>]

I've been looking at the route manager a bit more closely this afternoon, 
trying to track down a couple of things that I thought I had noticed were
different from previous versions but wasn't quite sure.

Given a route of EGNM-32 POL@7000 MCT@4000 EGGP-27, I ran the following
Nasal script:

###
var route = "autopilot/route-manager/route";
var n = getprop(route, "num");

for (var i = 0; i < n; i += 1) {
    var id = getprop(route, "wp["~i~"]/id");
    var leg = getprop(route, "wp["~i~"]/leg-distance-nm");
    var brg = getprop(route, "wp["~i~"]/leg-bearing-true-deg");
    var dar = getprop(route, "wp["~i~"]/distance-along-route-nm");
    var alt = getprop(route, "wp["~i~"]/altitude-ft");
    printf("%-8s %.2fnm %.0f %.2fnm %.0fft", id, leg, brg, dar, alt);
}
###

With 3.2.0, I get this:

EGNM-32  17.51nm 247 0.00nm -10000ft
POL      23.94nm 194 17.51nm 7000ft
MCT      20.56nm 266 41.45nm 4000ft
EGGP-27  0.00nm  266 62.01nm -10000ft

With 3.5.0, I get this:

EGNM-32  0.00nm  318 0.00nm -10000ft
POL      17.46nm 241 17.46nm 7000ft
MCT      23.84nm 194 41.30nm 4000ft
EGGP-27  20.47nm 266 61.76nm -10000ft

(I used the same fgdata directory by temporarily setting the version
back to 3.2.0 to fool the dependency check).

So it looks like the leg distance has been redefined as the distance _to_ the
waypoint, rather than the distance from the waypoint to the next. Similarly
the bearing is now the bearing inbound rather than the bearing outbound.

Was that intentional? It makes more sense to me as it is now, but it will 
break things in aircraft that get leg distances and leg bearings from the 
route manager.  

Another thing I noticed was that the "autopilot/route-manager/total-distance"
property is now calculated as a spurious large value when the departure and
destination airports are entered. It is only calculated correctly when the route is activated. 
In 3.2 it was calculated before activation. Again, is that a bug or intentional?

Are the small difference in distances anything to investigate further, bearing mind I 
was using the exact same fgdata? 

Richard