Hi,
Jonas Häggqvist wrote:
Well. Maybe. This changed caused nodes to not be deleted from my
database, since I was applying a bbox the change was getting discarded
since the bbox check failed.
So either the lon/lat *is* needed, or the bbox check should be skipped
for the delete action. Someone with more smarts than me should decide
which option is smarter.
This is a good observation. I hadn't thought of that at all and have
reverted the change in question for now. (The best solution would
probably be to assume that a node is at -999,-999 if it has no
coordinates or so but I wasn't sure if that would have any side effects.)
Applying bboxed diffs in slim mode is a bit of a problem anyway since it
will fill up your _ways and _rels tables with objects outside your bbox
(issue discussed here several times) and you'd have to run a delete
command every so often to get rid of them.
This optimisation/problem does not exist in the 64-bit branch that got
promoted to the current version recently.
An oversight on my part, I should have added that to the branch too. But
now it saves me the revert ;)
Bye
Frederik
--
Frederik Ramm ## eMail [email protected] ## N49°00'09" E008°23'33"
_______________________________________________
dev mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/dev
