Section author: Matt Windsor <matt.windsor@ury.org.uk>
LASS is a non-trivial project to install. Work is constantly being carried out to make the project easier to deploy, but at this early stage installation requires a large amount of manual intervention.
LASS carries a pip requirements file as requirements.txt; you can use this to pull the dependencies and sub-apps in.
Some things that LASS uses are written in languages other than Python, but nonetheless must be installed into any virtualenv LASS runs in in order for LASS to work. At time of writing, these include SQLite (for running unit tests) and Xapian (search backend).
Generally, scripts to install these into a virtualenv are available in the init_scripts directory as Bourne shell scripts.
These will not work outside of a virtualenv; in those cases, you will need to use your operating system’s native means of installing these dependencies.
The LASS distribution does not include templates; these will eventually be available from the same locations as the code distribution, or you can roll your own if you choose.
Similarly, static files (CSS, images etc.) are not provided - you will almost certainly want to make your own statics as the ones used at URY constitute URY’s unique station branding.
A “vanilla” set of CSS/LESS stylesheets may be provided in the future.
You will need to populate the database with some initial data. Where possible, the data is provided as fixtures to load with Django’s loaddata command.
The filler show must be created in order to allow the schedule filling algorithms to work properly.
The provided schedules/fixtures/filler_show.json fixture gives an example configuration for the filler show. Note that this fixture is incomplete: it does not approve or assign a creator to the show location, and the primary keys will likely need to be adjusted.
In addition, filler_show.json does not contain show metadata; you will need to add that manually for the time being.