It depends on what is slowing down: retrieving the data via OverPass, parsing that file, drawing the individual pins ?
- retrieving the data can be avoided by caching in a GeoJSON file. I don't expect that clustering will help if this is the bottleneck - the two other parts depend on your machine, browser and JavaScript interpreter, although clustering helps improving the drawing, as one has to display less pins. Limiting the data to a certain viewport only helps when you start looking for POIs when you are zoomed in enough. If you don't do that, you'll see 30.000 markers when the whole world is viewed. On Wed, Oct 1, 2014 at 10:17 AM, SomeoneElse <li...@mail.atownsend.org.uk> wrote: > On 01/10/2014 02:00, Bryce Nesbitt wrote: > >> >> >> * Slowed down too much after about 200 points of interest (I need about >> 30,000 minimum for the current project) >> * Limited flexibility in designing the popup >> * Ways disappear on the map (they don't get a pin: zoom in on the map to >> see two hidden toilets). >> >> > It's not the same sort of thing as uMap, but for info the "switch2osm > leaflet example" does include some information about getting data from an > external source for the map viewport (so you don't try and display > everything all of the time): > > http://switch2osm.org/using-tiles/getting-started-with-leaflet/ > > You'd presumably need to do something similar to "thin out" the POIs made > available if there are "too many" in a given area. > > Cheers, > > Andy > > > _______________________________________________ > talk mailing list > talk@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk >
_______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
