Open bhousel opened 5 years ago
Thanks @b-jazz - I opened this new issue for it.
I'm not sure how the user did it, but will check the drawing code and try to think what might have caused it.
As far as I can tell from the node ids, the user drew the ways like this (circle indicates where the coincident nodes ended up):
Ah. I didn’t know that the direction is indicated by decreasing node ids. Good to know.
Someone could easily do this by disconnecting the ways at that node.
I suppose it could be accomplished by disconnecting, but with the frequency it happens, I think something else is going on. I wouldn't imagined a seasoned mapper would do something like this for this particular changeset, and an inexperienced mapper is unlikely to disconnect them either since they probably wouldn't understand why they would want to do that.
And if they did disconnect them on purpose, without moving them afterwards, shouldn't iD coalesce the two nodes into one (in the case of like objects at the very least)?
Actually in this case I don't think it was disconnected - one of the two nodes is still connected and one is not.
(original screenshot was confusing because the white vertex was drawn on top of the grey one)
I might have had a similar issue recently which I did not understand. I dont remember doing anything special with this area when I added it.
If this is a similar case, maybe it helps with debugging. Could it be, that in some cases "double click to end the new-cycle" could add duplicate nodes right next to eachother? That could explain my area-case; but also the "way" case above?
Here's an example of a single way being created recently with a modern version of iD that has three nodes all in the same location. No intersecting with other ways. Just three nodes all piled on top of each other.
Here is a similar situation with duplicate nodes. I'm not sure if you want this as a new bug or simply as comments on the old bug, so I'll start here.
https://www.openstreetmap.org/changeset/69938949
The user created a footway from north to south, and then started a new footway from the midpoint of that way and went east to the existing Echo Heights Drive. The starting point for the east/west footway has a node with the identical lat/long of the north/south footway. Both of the footways were created in the same changeset.
Originally posted by @b-jazz in https://github.com/openstreetmap/iD/issues/4433#issuecomment-490125720