Finding country from coordinates using Python

Question

I have tweets with coordinates. I'm trying to bucket the tweets by country using the coordinate information. I know that I can achieve this using some hosted web services but all of these have API limits that I am sure to run into.

I started writing this code learning from another resource about osgeo library.

import osgeo.ogr

shapefile = osgeo.ogr.Open("geodata/worldborders/TM_WORLD_BORDERS-0.3.shp")
layer = shapefile.GetLayer(0)

for i in range(layer.GetFeatureCount()):
    feature = layer.GetFeature(i)
    countryCode = feature.GetField("ISO3")
    countryName = feature.GetField("NAME")
    print countryCode, countryName

    geometry = feature.GetGeometryRef()

The geometry variable has a member function called Contains that can take another Geometry and check whether the other is contained inside itself.

I have a coordinate value from Twitter data in the following format: [45.29680725, -75.92681408]. However an example on osgeo showcases creation of a point like this:

point = osgeo.ogr.Geometry(osgeo.ogr.wkbPoint)
point.SetPoint(0, 474595, 4429281)

How do I convert my coordinates into the above-required format? If I do manage to convert, is this the right approach to mapping twitter coordinates to countries?

Answer 1

This achieves finding country given coordinates: https://github.com/che0/countries

It uses a shapefile containing world borders loaded using python bindings for gdal.

Example:

 import countries

 cc = countries.CountryChecker('TM_WORLD_BORDERS-0.3.shp')
 print cc.getCountry(countries.Point(49.7821, 3.5708)).iso

The world borders shapefile can be obtained from http://thematicmapping.org/downloads/world_borders.php.

Answer 2

Given that you only need to find what country the tweets are in I guess the VMAP0 data is sufficient.

As for Python libraries I would recommend looking at Shapely, which supports the Shapely library, a turorial on point-in-polygon can be found here. As for reading in the country polygons from the vmap0-data (which i presume are polygons in shapefiles), you should look at ogr.