Thanks, for my purposes right now, extending the memory heap size on the command line with:
java -Xmx1000m -jar josm-latest.jar seems to work great! Thanks, Sam On 3/24/09, Andy Allan <gravityst...@gmail.com> wrote: > On Tue, Mar 24, 2009 at 7:17 PM, Sam Vekemans > <acrosscanadatra...@gmail.com> wrote: >> Hi, >> Some of my sample files are greater than 2 megs, and so i have a hard >> time trying to load the file onto JOSM.? >> >> My approach for uploading all that data once i get it into JOSM is >> simple. >> >> I would select the 1/2 of it, then 1/2 again and delete. Thus, left >> with only 1/4 of the data for the upload process. Once it all get >> uploaded.. i would open the file again (make the other one invisable), >> and only select another 1/4 of the data, then upload again. It wont >> matter if the areas overlapped, as the system recognizes that the data >> already exists. ... it's the same method as if your doing alot of >> editing... and you dont want to get a timeout error, so you only >> upload a part of your edits at a time. >> >> Does anyone know if there is a different way to preview these large >> .osm files before uploading? > > You might want to look into bulk-upload.pl - which is much better for > handling huge uploads. If you hop on over to the dev list and ask > about running imports from a local server (i.e. one in the same > datacentre as the OSM db server) you'll get reasonable performance > benefits from lower latency between each node/way/relation > transaction. > > Cheers, > Andy > _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
