SkyLines has quite many dependencies and is not always easy to install for new developers. Please don't hesitate to ask for help if you hit any roadblocks.
The production server is running the Debian Linux operating system and most of our developers use Ubuntu or Debian too. We recommend to use either one of those systems for development, but it may also be possible to make it work on OS X or Windows.
There is also a Docker environment for SkyLines. This makes it possible to run a containerized SkyLines development environment. More information can be found in INSTALL.docker.md.
Since SkyLines is written and based on Python you
should install it if you don't have it yet. SkyLines is currently targeting
the 2.7 branch of Python. You can check your version by running python --version
from the command line.
All the necessary Python libraries are installed using the pip
tool. If you
don't have it yet please install it using e.g. sudo apt-get install python-pip
on Ubuntu/Debian. More information about pip can be found at
http://www.pip-installer.org/.
Now you can install the python dependencies by calling:
$ sudo pip install pipenv
$ pipenv install --dev
Note: You might have to install the additional Ubuntu/Debian packages
libpq-dev
, python-dev
and g++
for the psycopg2
dependency.
Furthermore the XCSoar tools might require additional dependencies (see
XCSoar tools section)
The SkyLines backend is relying on the open source database
PostgreSQL and its
PostGIS 2.x extension, that provides it with
geospatial functionality. The fuzzystrmatch
extension is also needed which
is provided by the postgresql-contrib
package on Debian/Ubuntu.
To install PostGIS you should follow the instructions at http://postgis.net/install or http://trac.osgeo.org/postgis/wiki/UsersWikiInstall (for Debian/Ubuntu). Please note that you will need at least version 2.0 of PostGIS for SkyLines.
Once PostGIS is installed you should create a database user for yourself and a database for SkyLines roughly like this:
# change to the postgres user
$ sudo su - postgres
# create a database user account for yourself
$ createuser -s <your username>
# create skylines database with yourself as the owner
$ createdb skylines -O <your username>
# install PostGIS extensions into the PostgreSQL database
$ psql -d skylines -c 'CREATE EXTENSION postgis;'
# install fuzzystrmatch extension into the database
$ psql -d skylines -c 'CREATE EXTENSION fuzzystrmatch;'
Note: As of PostGIS 2.1.3, out-of-db rasters and all raster drivers are disabled by default. In order to re-enable these, you need to set the following environment variables: POSTGIS_GDAL_ENABLED_DRIVERS=GTiff and POSTGIS_ENABLE_OUTDB_RASTERS=1 in the server environment. For more information see the PostGIS manual and your distribution's documentation.
After creating the database you have to create the necessary tables and indices
by calling pipenv run ./manage.py db create
from the the command line.
Since the XCSoar project already has much of the code
implemented that is necessary for flight analysis, it makes sense to reuse that
code where applicable. SkyLines is using XCSoar as a python library. This library
is built and installed by the xcsoar
python package. To build this library you
might have to install additional libraries like libcurl
and freetype
, which
can be installed on Debian/Ubuntu by executing
apt-get install libcurl4-openssl-dev libfreetype6-dev
. Please have a look into
the XCSoar documentation if you need more help with the building process.
If the above steps are completed you should be able to run a base version of SkyLines locally now:
$ pipenv run ./manage.py runserver
(The following chapters are optional!)
Since an empty database is boring, you should at least load the airports from
the Welt2000 into the database by
calling (the commit
flag indicates that any data should be written to the
database):
$ pipenv run ./manage.py import welt2000 --commit
SkyLines can use Celery with
Redis as broker for asynchronous tasks like in-depth
analysis of flights. Celery is one of SkyLines requirements and will be
installed by pip, but you need to get Redis on your own. On Debian, all you
need is to install the redis-server
package:
$ apt-get install redis-server
To run the Celery worker, call
$ pipenv run ./manage.py celery runworker