On Sat, Sep 21, 2013 at 2:39 AM, André Pirard <a.pirard.pa...@gmail.com>wrote:
> On 2013-09-20 15:55, Marc Gemis wrote : > > I welcomed a new user, expecting that it was related to the Kreatos > hairdressers company: > > their reply: > > Dank voor de informatie. > Weet u een manier om (meerdere) nodes (in één keer) te uploaden naar > OpenStreetMap? Bij Google is dit via Google Places, met een XLS of XML > file. > Heb hiervoor gezocht op de website van OpenStreetMap, maar ik raak er niet > meteen wijs uit? > De informatie zou uit en MySQL database komen, dus een api met > rechtstreekse import/sync mogelijkheid hiervoor zou natuurlijk nog beter > zijn… > > > > --- > > How do I proceed ? Show them the OSM Api v0.6 ? + policy page ? Josm > upload ? Ask them the file and do the upload myself + verification of > individual points ? > > I was also wondering this week whether we could contact companies to ask > them to share their shops or fuel stations. What do you think about that ? > > > regards > > m > > > Hi Marc, > > I would create a single sample shop node in a new JOSM layer and save it > to an *.osm file. > A shop looks like this: > > <node id='-108978' action='modify' visible='true' lat='50. ...' lon='5. > ...'> > <tag k='name' v='Name of the shop' /> > <tag k='shop' v='bakery' /> > </node> > > Then I would > > - either edit that file, replicate the node and change whatever must > be changed > - or run a perl regexp command to transform another file to that > format > - reload the result in JOSM > - maybe check that everything is OK (esp. in the right place) or let > them do that later > - update OSM > - save the *.osm file again > > It can also be done with waypoints in a GPX file, but I know no GPX > extension to add OSM tags. > > Notes: > > - the IDs can be any negative number but must be different > - it's important to "save the *.osm file again": it will contain the > real IDs and it can help to make later mass updates > > If the problem is regexp, I can (most probably ;-) do it for you, but I'll > have to find how to generate the ID. > Send node sample and parseable flat text file (like csv). > > Cheers, > André > André, thanks for your input. This is the method I use to convert the pages of Onroerend Erfgoed of Wikipedia into an OSM file using Python. All items of 1 page end up in 1 osm file. I copy them 1-by-1 from that layer into the data layer, I use 2 layers in JOSM. On the data layer I merge them with the existing data (typically using CTRL-SHIFT-g with existing building outlines and some minor tweaks to the tags). Then I go back to the imported layer, delete the node and move on to the next. I delete the node in the imported layer to keep track of what I did. This is needed for larger towns where there more than a few protected items. I have a similar script that converts my GPX waypoints into an .osm file, but I'm not sure that using that file is much faster than using the regular JOSM tools. I also used the csv method before, for the monitoring stations of the Vlaamse Milieu Maatschappij. I think this method is easier for them, since the generation of a csv file from a database requires less programming than generating a .osm file. But I could have mentioned the method via direct osm-file generation as well. regards m
_______________________________________________ Talk-be mailing list Talk-be@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-be
