Hi Paul, > > What you are proposing is a combination of a bot and an import. This is > unusual and it will likely take longer to do the required community > consultation. Additionally you are proposing using some new software. I > would expect the consultation to take a *minimum* of 2-4 weeks. >
Oh well, I'm afraid it's inevitable, but since OpenSeaMap is it's infancy it has all time at it hands :-) > You are proposing to delete your import and reimport it regularly. Why are > you not updating any changed objects? > It's sure not what is the best way to do it, since the original NOAA data doesn't have any unique ids in it. Maybe hash of lat, lon, object type, not sure if it's unique though. > How do you intend to handle when a user edits a point? Your current plan > would create a new duplicate point. > That's true. > You are writing new software that handles the upload to the API. What > testing have you done against the dev API? > I debugged it against the dev API and then ran against the main one. The only problem I had is that I couldn't find the dev server for overpass XAPI. Any review of OsmUploader class is greatly appreciated, fortunately it's quite small piece of code, so should not bore the reviewer. > Additionally, although not critical to the import, you state that this > import is for the "OpenSeaMap" database. OpenSeaMap is just one of many > consumers of the OpenStreetMap database. There is no part of the DB that > "belongs" to OpenSeaMap. > Good point, I updated wiki, so it doesn't use the term OpenSeaMap database, but rather to OpenSeaMap map. -- Sailor Jerry _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
