Matthias Urlichs wrote: > A third option would be to simply save the last 24 diffs and then > concatenate them appropriately. Some XML mangling may be required, > but that's certainly less of a processing burden than to go to the > database for information you already did extract True, that would also work. The reason I preferred using the same approach as hourly/minute diffs is that the code is there now and is very reliable, it just catches up again when the db comes online next time. Concatenating 24 hourly diffs to produce a daily diff will require either adding tasks to osmosis for this purpose or knocking up another shell script to invoke osmosis appropriately with a bunch of merge tasks.
_______________________________________________ talk mailing list [email protected] http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
