Adrià Mercader -http://amercader.net (@amercader)
A GeoJSON REST service for georeferenced Wikipedia articles, built with MapFish.
It was built around the georeferenced Wikipedia articles dataset that can be downloaded from this website:
http://amercader.net/dev/wikipedia
You will need to import the dataset to a PostGIS database (or any other geodatabase engine supported by GeoAlchemy, though this has not been tested).
Create and activate a virtual environment:
virtualenv --no-site-packages wikigeolinks
cd wikigeolinks
source bin/activate
easy_install pip
Install the source (will also install the other requirements):
pip install -e git+git://github.com/amercader/wikigeolinks.git#egg=wikigeolinks
Create a configuration file:
cd src/wikigeolinks
paster make-config wikigeolinks development.ini
Edit the database connection string to point to your database containing the georeferenced articles:
sqlalchemy.url = postgresql://<user_name>:<password>@<server>/<database>
Serve the application with the following command:
paster serve --reload development.ini
You should get a GeoJSON response visiting the following URL:
http://localhost:5000/articles
Georeferenced articles are returned as GeoJSON Features:
GET http://<server>/articles/<id_or_title>
E.g.:
GET http://<server>/articles/1234
GET http://<server>/articles/United%20States%20Naval%20Observatory
{
geometry: {
type: "Point",
coordinates: [
-77.066946,
38.921473
]
},
id: 1234,
type: "Feature",
bbox: [
-77.066946,
38.921473,
-77.066946,
38.921473
],
properties: {
links_count: 7,
title: "United States Naval Observatory"
}
}
You can use both the numeric id or the article title as indentifiers.
You can perform various queries using the MapFish Protocol on the main endpoint:
GET http://<server>/articles?<query>
-
Search by title:
GET http://<server>/articles?title__ilike=%tarrag%&attrs=id,title,links_count&queryable=title&order_by=links_count&dir=desc&limit=30 -
Search by location:
GET http://<server>/articles?lon=-1.60&lat=54.98&tolerance=0.5&order_by=links_count&dir=desc&limit=30 -
Search by bounding box:
GET http://<server>/articles?bbox=5,50,7,60
All these queries return a FeatureCollection of georeferenced articles.
{
type: "FeatureCollection",
features: [
{
geometry: {
type: "Point",
coordinates: [
1.25,
41.13333333333333
]
},
id: 47441,
type: "Feature",
bbox: [
1.25,
41.13333333333333,
1.25,
41.13333333333333
],
properties: {
links_count: 13,
title: "Camp de Tarragona"
}
},
{
geometry: {
type: "Point",
coordinates: [
1.2740638888888889,
41.19218055555555
]
},
id: 172375,
type: "Feature",
bbox: [
1.2740638888888889,
41.19218055555555,
1.2740638888888889,
41.19218055555555
],
properties: {
links_count: 9,
title: "Camp de Tarragona railway station"
}
},
...
}