OS Open Data Shapefile to KML

In early April the Ordnance Survey released a whole set of previously expensive geographic data into the Open Data initiative. The data is often in ESRI shapefile format as that is pretty much the norm for GIS software. That’s great if you are taking the data into a GIS workstation, but what if you want to turn the data into XML or JSON for use in a web service for example?

That was the problem I faced last week when working on the MP Finder application that should soon hit the Apple app store.

My specific goal was to turn the shapefile data into KML as an intermediate format which I could then write a perl parser for. KML is great as an intermediate format because I can visualise it in Google Earth to check that the export has gone OK and of course it is a structured document expressed in XML – so it’s perfect for machine parsing.

It turned out to be not quite as easy as I first thought.

The issue was the sheer length of some of the polygons in the shapefile, and more specifically northern Scotland which has a crinkly coastline to rival that of Norway. This was causing a lot of software to either crash, or silently dump the datapoints over a certain threshold. Not good.

The solution was to use the snappily named “ogr2ogr” tool which forms part of the toolset compiled by Frank Warmerdam and released as FWtools.

My first attempt was to use the simple command;

ogr2ogr -f KML output.kml input.shp

And whilst at first glance this seemed to have worked, the data points in the polygons were certainly not latitude and longitude. The problem was the map projections were not matching and after a bit of research and a big clue from stack overflow, I corrected the projection using the new command;

ogr2ogr -t_srs EPSG:4326 -f KML output.kml input.shp

Now we have a correctly projected KML file of the original shapefile data. You can test it by viewing it in Google Earth and more importantly, now that it is in KML it becomes trivial to write a parser to extract the data you might require … but that will be the subject of another post altogether.

Share this:
  • Print
  • Digg
  • del.icio.us
  • Facebook
  • Google Bookmarks
  • LinkedIn
  • MySpace
  • StumbleUpon
  • Twitter
  • Yahoo! Buzz

This post was written by:

- who has written 22 posts on Mindsizzlers.


Contact the author

Leave a Reply

Advert

For more information about our services…

Contact Us

Wordle of the Day

Image from software at http://wordle.net
Data by Web Trends Now

Categories