[mkgmap-dev] New exit_hints branch

Fri Dec 21 22:11:07 GMT 2012

Good drawing, that's exactly what I meant. Still I'm not totally convinced.
If both E and e do have a motorway_junction node, why shouldn't the code be
applied to both motorway_links? Now I saw the typical Garmin behaviour on e
to show the name of the lower grade road at the end of the motorway_link,
which wasn't very informative.

Cherio, Johan

2012/12/21 WanMil <wmgcnfg at web.de>

> > This afternoon in live testing I found what seems to be a bug. Exits
> > from a parallel lane, tagged as motorway_link, didn't show up with the
> > exit_hint. Maybe the code does not know to which motorway_link of the
> > y-branch the exit_hint has to be put on. Is it possible to assign the
> > exit_hint to the motorway_link which is going to the right?
> >
>
> I set the exit_hint only on exits directly from motorways because hints
> on parallel lines should be created from the destination tags.
>
> So have a look on the following junction:
>
> ===E====================================
>      \                              /
>       -------e----------------------
>               \
>                \destination=Berlin
>
> === motorway
> --- motorway_link
> E   exit node on motorway
> e   exit node on motorway_link
>
> So and exit_hint is created only for E because after you have passed E
> the Garmin should tell you the correct destination which is usually
> written on the signs in the motorway junction (in Germany...).
> If an exit hint would also be created for e you would see the
> destination only after passing e and that's too late.
> The motorway junctions if have analyzed did have the same tags on E and
> e so there is no additional information.
>
> WanMil
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20121221/381d9268/attachment.html 