Fetching latest commit…
Cannot retrieve the latest commit at this time.
|Failed to load latest commit information.|
===== ebgeo ===== The eb map system. The ebgeo package provides for two main bits of functionality: 1. A `maps' subpackage for rendering map tiles with Mapnik, and 2. A `utils' subpackage which is a collection of useful modules when working with geospatial data, including clustering. Requirements ============ Mapnik version 0.6.0 or greater -- http://mapnik.org/ TileCache 2.10 -- http://tilecache.org/ processing 0.50 or greater (optional) -- http://pypi.python.org/pypi/processing Background ========== You should be familiar with the mapping toolkit Mapnik before getting started with this package. Installation of Mapnik is non-trivial, although it has gotten easier over time. You'll also need to provide your own styles for the map tiles, which means creating Mapnik style XML. Cascadenik_ provides a CSS-like syntax that compiles to Mapnik XML; you may find this an friendy way to get started creating styles: .. _Cascadenik: http://code.google.com/p/mapnik-utils/wiki/Cascadenik You'll also be responsible for your own source geospatial data. This is a database of features (i.e., individual roads, parks, bodies of water, etc.) with attributes (i.e., road types, areas of parks, names and labels, etc.) that Mapnik queries and, combined with styles defined in the Mapnik XML format, represents in the graphical output of the rendering engine. There are two main freely available sources of streets data, TIGER/Line_ (produced by the Census for U.S. streets) and OpenStreetMap_ (a user-contributed database of streets world-wide). .. _TIGER/Line: http://www.census.gov/geo/www/tiger/ .. _OpenStreetMap: http://www.openstreetmap.org/ Geospatial data --------------- Mapnik provides a number of different ways to access geospatial data, from shapefiles to PostGIS databases. The ebgeo code assumes that your data is stored in a PostGIS database. It also makes some assumptions about which layers and attribute field names are defined. The ebgeo.maps.mapserver module source code should be your reference for this information; refer to the draw_map() method. Installation ============ After downloading the ebgeo tarball and unpacking, put the ebgeo package on your Python path. Then create a directory called "mapnik" within your `sys.prefix` directory (which you can get by running "python -c 'import sys; print sys.prefix'"). This is the directory where you should place your Mapnik style XML files. Rendering map tiles =================== The `render_tiles' executable iterates over a number of zoom levels and renders the map tiles contained within. Usage: render_tiles [<options>] /path/to/config <layername> Options: -h, --help show this help message and exit -s START, --start=START starting zoom level -t STOP, --stop=STOP ending zoom level -c CITY, --city=CITY only render this city's tiles (use city slug) -v, --verbose -D, --debug -n NUM_PROCS, --num-procs=NUM_PROCS number of render processes (defaults to 1) The /path/to/config argument is a path to a TileCache config. Please see the TileCache docs for the syntax, which is a .ini-type file. The <layername> argument is a string that should match a defined layer section in the TileCache config file and the Mapnik style XML filename. The -s and -t options indicate the starting and stopping zoom level for rendering. The starting, -s, option is inclusive: in other words, if the value of -s is 1, the rendering will start at level 1. However, the stopping, -t, option is exclusive: in other words, if the value of -t is 10, the rendering will stop upon completing level 9. The number of zoom levels is defined in the TileCache config. The -c option allows you to constrain tile rendering to the bounding box of a city defined in your application. The value of -c is the slug that represents that city. (See ebpub.metros) The -n options allows you to spread rendering over multiple CPUs, utilizing the great Python package `processing.' Dynamic map tiles ================= There's an alternate way to view map tiles other than rendering them first and serving them statically. ebgeo.maps.views has a get_tile() view that will render individual tiles and serve them directly to the requesting browser. See ebgeo.maps.urls for the URLconf that wires up a URL request to the get_tile() view.