On Mon, Jan 11, 2010 at 16:35, Emilie Laffray <emilie.laff...@gmail.com> wrote: > 2010/1/11 Ævar Arnfjörð Bjarmason <ava...@gmail.com> >> >> FWIW I tried just feeding the 16MB .osm file (it got a bit smaller) to >> bulk_upload.py. It reached the 50k node limit on the changeset and >> then died: >> >> http://api06.dev.openstreetmap.org/browse/changeset/2766 >> >> Time to check out some of the splitting tools... >> > > There is a sax version of the bulk_upload.py in the SVN repository. It > automatically breaks the data in different changesets. The only constraint > is that you need an unique id for each elements (nodes, ways, relations) > which may not be always the case. For example, you might have a way with an > id of -1 and a relation of -1.
That's not a problem with this data. I managed to upload the whole thing with JOSM though to the test server in 4 changesets: http://api06.dev.openstreetmap.org/user/ourFootPrints%20import/edits Here's an example of a way with some modified tags: http://api06.dev.openstreetmap.org/browse/way/55355 All of the ourfootprints:* tags are derived from the ourfootprints:raw_comment tag and munged by this super-ugly code: http://github.com/avar/ump2osm/commit/ba5964286e4d13fe6f3b2d6b87b123f0b3a453eb When I upload this for real I'm going to use something like these changeset tags: http://api06.dev.openstreetmap.org/browse/changeset/2766 Aside from doing some minor tweaking on highway classifications (e.g. motorway -> trunk) this dataset is pretty much ready for importing into the live database. Any comments on that? _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk