[OSRM-talk] When does OSRM preprocessing generate new node IDs (custom map data)?

[Patrick Niklaus]

Hey Armin,

Can you check if some of your ids exceed 2^33 - 1? We only allow 33bit per
node ID, since that easily fits all OSM ids.

If there are OSM ids that are not in your data that suggests a bug, OSRM
should preserve those. Do you have a small self-contained example we can
use to debug this?

Cheers,
Patrick

Am 15.01.2018 16:38 schrieb <eenuep at gmail.com>:

> Hi there,
>
> I hope this is the right place to ask this question:
>
> *_What I want to do:_*
> I am using the OSRM map-matching engine on custom data.
> To do this, I converted the data to an osm.xml file and import this data
> into OSRM (preprocessing).
>
> *_Unexpected behaviour:_*
> I noticed, that OSRM uses new node IDs when the raw data contains very
> large IDs.
> To avoid this, I made sure the custom data only contains IDs up to 10
> digits.
>
> Now I am still seeing that for some nodes, new IDs are generated:
> I.e. the IDs references in "annotation":{"nodes":[...] do not exist in
> the raw osm.xml file.
>
> *_My question:_*
> Is there a way to find out, in which cases this happens and why?
> *(e.g. some debug option which tells me "osm node id XXXX is too large /
> duplicated, generating new ID")*
>
> Best Regards,
> Armin
>
> _______________________________________________
> OSRM-talk mailing list
> OSRM-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/osrm-talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osrm-talk/attachments/20180115/8fa912fc/attachment.html>