[Talk-ca] CanVec bug - splitting of long riverbank ways?
Brian Gamberg
bgamberg at nf.sympatico.ca
Wed Nov 14 04:59:40 GMT 2012
Canvec data appears to be produced with a maximum of 2K nodes per way. I assume that this was an attempt by the Canvec people to meet the max 2K node criteria of the OSM database. When the split occurs in a way that is a member of a multipolygon, there's not a problem. Otherwise, as Paul has pointed out, broken data may be uploaded inadvertently. In JOSM, a quick check for this is to use the search function with the string "nodes: 2000-".
On Monday, November 12, 2012 11:27 PM, Paul Norman <penorman at mac.com> wrote:
> I appear to of found a CanVec 10 bug in 094P13.0.osm
>
> There is a waterway=riverbank way (Dilly Creek) that is split in two at
> 59.7728641, 11219806992One way is 2k nodes, the other is 457 nodes.
>
> Ideally this would be split in two to form two separate areas. An alternate
> solution would be to not split them and have one way over 2k nodes. This
> will force the importer to fix it before uploading as opposed to right now
> where if someone doesn't notice they will upload broken data.
>
>
> _______________________________________________
> Talk-ca mailing list
> Talk-ca at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-ca
>
More information about the Talk-ca
mailing list