This is a static archive of our old OpenStreetMap Help Site. Please post any new questions and answers at community.osm.org.

Osmosis - Only want water…

0

Greetings,

I'm digging into planet files and creating extracts for the first time, and I'd like to create a regional extract which contains only water elements - lakes, ponds, rivers, streams, and coastlines... etc.

I've created an Osmosis filter which has a bounding box which works well, as well as...

--way-key-value keyValueList=natural.water,natural.wetland,natural.coastline,waterway

There are no other filters present... I was expecting to get output which had ONLY those elements above in them. I do get my water data, but in addition I'm getting lots and lots of extraneous nodes. Many match up with roads and intersections, others are power transmission poles, etc.

How can I filter OUT these extra nodes, while keeping in that which is associated with the water items I'm searching for?

Any help is appreciated! -Greg

asked 25 Apr '13, 21:13

greggerm's gravatar image

greggerm
45225
accept rate: 0%

edited 25 Apr '13, 21:14


One Answer:

3

Though you seem to have found the solution for your problem I'd like to mention osmfiler. It should do for your problem and as C-program is several times faster than osmosis.

Probably you also want to retrieve just waterways (or what you need) from the Overpass API. So you'd get only the data you want - no filtering for areas and tags afterwards.

answered 27 Apr '13, 11:29

malenki's gravatar image

malenki
4.7k24683
accept rate: 6%

1

Malenki - great tips. I've been able to use Overpass to export information directly to JOSM which contains the data I need.

(Ultimately, I'm trying to download all water data into JOSM so I can see what's MISSING, and then draw it in based on a Bing overlay)

Thanks once again - your solution with Overpass is precisely what I needed.

(29 Apr '13, 15:48) greggerm

Glad I could help. :) It would be nice if you'd mark this question as solved.

(29 Apr '13, 21:46) malenki

Source code available on GitHub .