Welcome to Kepler’s documentation!¶
Kepler automates several aspects of the GIS data workflow process.
Installation¶
Kepler relies on GDAL to perform a few necessary steps in processing GeoTIFFs. You will need to make sure you have the libgdal, libgdal-dev and gdal-bin packages installed on your system before installing the python dependencies. The version of the system GDAL should match the version of the python bindings specified in the requirements.txt file. Make a note of where the header files have been installed. This should usually be /usr/include/gdal, but you can check with:
$ gdal-config --cflags
Then, from the project root:
$ env CPLUS_INCLUDE_PATH=/usr/include/gdal \
> C_INCLUDE_PATH=/usr/include/gdal \
> pip install -r requirements.txt
Replace /usr/include/gdal with wherever the header files are on your system.
Configuration¶
Kepler uses environment variables to manage application configuration. These should be placed in a .env file at the root of the project. The following variables should be set:
SQLALCHEMY_DATABASE_URI |
Location of sqlite database |
GEOSERVER_PUBLIC_URL |
URL for public GeoServer instance |
GEOSERVER_RESTRICTED_URL |
URL for secure GeoServer instance |
GEOSERVER_WORKSPACE |
Name of workspace for data |
GEOSERVER_DATASTORE |
Name of datastore for shapefiles |
SOLR_URL |
URL for GeoBlacklight Solr instance |
SWORD_SERVICE_URL |
URL for DSpace SWORD service |
SWORD_SERVICE_USERNAME |
Username for SWORD service |
SWORD_SERVICE_PASSWORD |
Password for SWORD service |
UUID_NAMESPACE |
The namespace used in generating the version 5 UUID |
SECRET_KEY |
Randomly generated key |
OAI_ORE_URL |
URL for OAI ORE DSpace service |
GEOSERVER_AUTH_USER |
Username for GeoServer REST service |
GEOSERVER_AUTH_PASS |
Password for GeoServer REST service |
Running the Application Locally¶
The application can be run locally using either the heroku toolbelt or honcho. For example, from project root:
$ honcho start
Running the Tests¶
You should have python binaries for 2.7, 3.3 and 3.4 available on your system. Install tox and run:
$ tox
Building the Documentation¶
In order to generate the documentation you have to make sure Sphinx is installed in a virtualenv along with all the kepler dependencies. One possibility is to create a different virtualenv for generating the docs. For example:
$ cd /path/to/kepler
$ mkvirtualenv kepler-docs
(kepler-docs)$ pip install Sphinx
(kepler-docs)$ env CPLUS_INCLUDE_PATH=/usr/local/include \
C_INCLUDE_PATH=/usr/local/include pip install -r requirements.txt
(kepler-docs)$ cd docs
(kepler-docs)$ make html
This will place the built documents in kepler/docs/_build/. The built docs can be uploaded to the project page using ghp-import. It is recommended to install ghp-import using pipsi:
$ pipsi install ghp-import
$ cd /path/to/kepler
$ ghp-import -n docs/_build/html
$ git co gh-pages
$ git push origin gh-pages
