|
I have a recent planet pbf and am finding duplicate node Ids in some regions. Can someone explain to me why this is happening? The duplicates always have different versions and changeset_ids. One of the duplicates (not always the newest or oldest) always has empty lat long data. What is going on? Why do these null rows not show up in a PGSnapShot load? What are these artifacts of? They don't seem numerous enough to be any sort of actual change history. Example: id version user_id tstamp changeset_id 77875 2 131476 2009-07-08 14:52:45.0000000 1776041 Has lat long data 77875 51 1227748 2017-12-11 08:43:34.0000000 54543387 Has empty lat long data |
|
You don't have duplicate node IDs. You have a node ID in the first line, and a relation id in the second line. Each OSM object type (nodes, ways, relations) has its own number range. Node #77875 has nothing to do with relation #77875. Apparently you are mis-interpreting a relation as a node; little surprise, then, that the relation has no latitude and longitude. 1
Right you are. My SqlServer port of osmosis was writing relations to the wrong table sometimes. Good catch, thank you.
(27 Nov '18, 01:03)
bwoods
@bwoods: meta: Please could you mark an answer as accepted if it solves your problem.
(27 Nov '18, 19:26)
aseerel4c26 ♦
|
