Re: [Gpsbabel-misc] Route implementations

[P. R. <ro...@vo...>]

Robert,

> OTOH, converting arbitrary routes between formats is harder than
> it sounds and I don't think (and I'd love to be proven wrong via
> contributions of code) that GPSBabel will ever be as good at converting
> routes as it is at waypoints

I agree: the main problem I see in route treatment is the number of 
route points per single route for the various receivers.

As long as their contents are not relating to each other there won't be 
too much of a problem. But as soon as you try to manage "arbitrary 
length" routes like me and have to kind of combine them you are at a 
loss when switching to another target receiver type with differing 
wpt/rte count.

So, for strategic route treatment aims we'd need to discern route 
sequences that are internally linked, and such without that. The 
required information on that property is in the head of the user only, 
though.

But I'm not sure if you would follow me in general on that road since I 
seem to be somewhat alone with my usage of (multiple consecuting) routes 
for my tightly led motor bike tours.

>>My main argument towards Alex was, that the (internal) modality 
>>should not be exported to formats that don't need it, like GPX, 
>>as this adds potential traps when the user is working with them 
>>outside gpsbabel.

> With GPX, it's easy enough to merge the files together but with formats
> like Mapsend, it's impractical for a user to do in a text editor.

Hmm, maybe you missed my point: I was not arguing to do the merging 
myself by hand, but to rely on the automatic waypoint definition 
properties of the "combined formats": If you define route points you 
already provide the locations and symbols in the GPX statements. So no 
need to define them once again (or: before) in separate statements.

Another such format I presently know is the one Marco S. Hyman uses for 
his Garmin Utilities. And there you define the waypoint stuff inlined in 
the route definition as well, e.g.

[routes, n records]
**1 Route 1/4
au-1     47.752778   10.624722      0/1
re-2     47.752778   10.625000      0/2  re
hl-3     47.698611   10.603333      0/2  hl
...
li-28    47.382778   10.560833      0/2  li++
li-29    47.335833   10.539167      0/2  li
**2 Route 2/4
li-29    47.335833   10.539167      0/2  li
re-30    47.335556   10.539722      0/2  re
...
we-100    47.735278   10.619722      0/1  we
[end transfer, n/n records]

OTOH, Magellan route definitions must be split as there is no way of 
introducing the location information into the route point definition 
itself. In such formats modality cannot be helped, of course.


> Let's get back to the task at hand - Have you reviewed Alex's stuff
> and does is solve problems for you?   If not, what's wrong with it?

The only complaint I had was the (present) inability to use identical 
waypoints in multiple route definitions, as I do it as shown in the 
snippet above for the Garmin Utils.

In order to create a gapless route display on the receiver's screen I 
re-use the last waypoint of the former route as the first one of the next.

At the moment this is not reproduced by Alex' implementation, but he 
assured me that this would only require a smaller change.

	Peter
-- 
goethe GbR               Vollmer & Roosen GbR     Dr. Peter Roosen
www.goethe-gbr.de        www.linguadapt.de        www.peter-roosen.com
Tel. ++49 241 7091525    Tel. ++49 2451 971437    Tel. ++49 241 7091524
Fax  ++49 241 7091521    Fax  ++49 2451 971439    Fax  ++49 241 7091521

GPS: 50.7383 N, 6.0427 E