The process of creating map involves applying rules and styles to spatial datasets which produces a final raster image. To get started you should first :doc:`install Mapnik library </getting_started>`.
This tutorial will focus on creating a simple Python script in which we will use Mapnik library to render a map. Although you can use any text file editor to edit Python scripts, you should use one which you feel most comfortable with. If you never used Python, you should read some introductory tutorial.
Note
Create a single directory which will contain all of tutorial files, i.e mapnik_tutorial
There can be no map without data, fortunately there are a large number of resources available online. For this tutorial we are going to use a simple dataset available for download at http://mapnik-utils.googlecode.com/svn/data/world_borders.zip. You should download this file and unpack it to your tutorial directory.
After unpacking you should have 4 new files which define a Shape file dataset:
- world_borders.shp - actual geometry of the features
- world_borders.shx - positional index of feature geometry
- world_borders.dbf - attributes of features
- world_borders.prj - optional file, defines projection in well-known text format
Hint
Mapnik comes with a shapeindex utility which creates spatial index for Shape file datasets. Spatial index speeds up data access, especially for large datasets. shapeindex reference
Additional free datasets are provided on this services:
- some service
- some other service
Open up your text editor and start editing a new file. For now just
copy/paste following code and save your file, i.e. tutorial01.py.
import mapnik
map = mapnik.Map(600,300,"+proj=latlong +datum=WGS84")
map.background = mapnik.Color('steelblue')
mapnik.render_to_file(map,'world.png', 'png')After you saved the file you should execute it. Open up a terminal, navigate to the folder with saved file and spatial data and run:
$ python tutorial01.py
Running this Python script should produce a new image file
worild.png which can be viewed in any image viewer. Now let's
examine what have we done in more detail.
On the first code line we import mapnik module to current Python
interpreter environment. This enables us to access Mapnik library in
Python code. To create a map we need to create a new mapnik.Map
object (insert link to reference). So on the line 3 we create a
mapnik.Map object with 3 parameters: width, height and spatial
reference system (SRS) definition, and assign it a name map.
Note
Mapnik accepts any projection that Proj.4 handles. See http://spatialreference.org for the proj4 strings for various projections. For example will give you the exact parameters for EPSG 4326.
You can read more about spatial reference systems and projections in :doc:`introducition to GIS </introduction_to_GIS>`
In our case data is in EPSG 4326 spatial reference system and
matching Proj.4 string is +proj=latlong +datum=WGS84. On the
next line we assign map object attribute background (insert link
to reference) a new mapnik.Color object (insert link to
reference).
Color object accepts 'named' colours, #rrggbb, #rgb, rgb(r%,g%,b%) or rgba(r%,g%,b%,a%) format. Named color steelblue is equivalent to hex triplet #4682b4 or rgb(70, 130, 180).
Last step, on the line 6, saves current map object to file
world.png of type png using mapnik.render_to_file function
(insert link to reference). Final result should be a blank image with
steelblue background, 600px width and 300px height in PNG format,
similar to this one:
And that's it you've just created your first map, even though it doesn't actually use any spatial data.