a-simple-spatial-database

This is an attempt to share the work that I have done, by developing a web interface to browse spatial data in our team.

It is largely based on Leaflet, a great open source JavaScript library.

The web interface shows a map, on which you can display each dataset as a layer. When you select data in the map, the results are listed below it, dynamically.

The idea is to keep it simple: only a basic Linux web server is needed. The interface consists in HTML and JavaScript, and GeoJSON files for the database. So no PHP, no SQL, etc. Also, to facilitate installation and avoid compatibility issues, the external libraries used in the code are provided in this repository (in the external directory).

A demo is available > here <

External libraries (provided in this repository)

• Leaflet
• Turf.js
• jQuery
• jQuery UI

Leaflet plugins:

• Leaflet.draw
• leaflet-bing-layer
• Leaflet.Coordinates
• Leaflet.GestureHandling
• Leaflet-hash

Database

You need to provide one GeoJSON file per dataset. By dataset is meant a set of homogeneous data. For example, one file listing images of one satellite (or multiple satellites, as long as they share the same family of parameters), one file listing GPS stations, etc.

A very handy tool, ogr2ogr, is available in GDAL for converting between formats (including GeoJSON).

About the GeoJSON files

• Additional requirements:

  • For the moment, among the types of geometry proposed in GeoJSON, four are available: Point, LineString, MultiLineString and Polygon.
  • The properties member of each Feature object must contain a type property. Its value must be the same for each object. For example, it could be the name of the satellite or "GPS station".

• "Special" properties (optional):

  • date : if the data have a date information, using date as the property name will allow to filter the results by date.
  • url : if you provide a link to download the data, using url as the property name will make the URL prettier.
  • kmz : if you offer KMZ files for download, using kmz as the property name will make the URL prettier too.
  • preview : if you want to display a preview image, using preview as the property name will make the preview prettier.
  • If you want to display the coordinates of the four corners of an image, use the following property names: lon_tl (for top left), lat_tl, lon_tr, lat_tr, lon_br, lat_br, lon_bl, lat_bl.
  • When you click on an object on the map, a popup displays the type property, and also the date and/or name properties, if they exist in the GeoJSON file.

(Four GeoJSON files, one for each type of geometry available, are provided in the data directory, as examples.)

Installation

1. Download the files and move them to a dedicated directory, in the DocumentRoot of your web server (typically, you will create this directory in /var/www/html/).

2. Place the GeoJSON files in the data directory (four files are provided as examples).

3. Run the createindex.py initialization script. This will generate index.html, which is the core part of the interface.

createindex.py [-h] [--bing_maps_key BING_MAPS_KEY] [--no_openstreetmap] [--title TITLE] [--hide_footer]

Optional arguments:

-h, --help                          show this help message and exit
--bing_maps_key BING_MAPS_KEY       your Bing Maps API key (if not provided, OpenStreetMap will be used)
--no_openstreetmap                  disable OpenStreetMap
--title TITLE                       to customize the web page title
--hide_footer                       hide the footer containing information on credits and license

And this is it!

Access index.html through your web server to make use of the interface. In a local environment, the URL should look like: http://localhost/a-simple-spatial-database/index.html

Important

For safety reasons, normally the createindex.py script self-destructs at the end of its execution. This is to avoid potential security breaches on the web server. However, you may find it convenient to keep it, for testing or if you need to run it again, to update the interface or to add a new dataset, for example. Therefore, the script will ask for your preference.

In case you don't authorize the script to delete itself, consider deleting it manually or moving it to a safe location (i.e. not accessible to the client).

In any case, you can always download it again from here.