On 10 February 2010 16:32, Ian Dees <[email protected]> wrote: > On Wed, Feb 10, 2010 at 9:34 AM, SteveC <[email protected]> wrote: > >> any way the importer can >> >> a) go back and fix all the millions of duplicated nodes >> >> b) not do it in future? > > > This is my fault, Steve. I ran the import but the bulk upload script I used > had problems with the way I had created the OSM files (they assumed that > negative IDs were globally unique across all primitives, I assumed that they > were unique across a primitive type). I tried removing the empty nodes, but > I obviously missed some (most) of them. > > I don't really have time to go back and fix my mistake (just started a new > job), but it won't happen in the future (at least from me). > > PS: Perhaps someone could start a thread on osm-dev to brainstorm ideas for > a "bulk import/change" application/website that does something like what the > French import people did to find areas that might have already been mapped > and show an overlay of what the new data would look like. Bulk imports like > the NHD stuff could happen through that... > > I think it is a good idea to start that discussion. In addition, I will update the page for the French import to reflect the changes that were made in the SQL queries. I hope a good discussion will avoid many half baked import that end up being nasty to fix afterwards.
Emilie Laffray
_______________________________________________ Talk-us mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk-us
