GeoSPaaS is a set of Python tools meant to help scientists deal with Earth Observation data. django-geo-spaas is the core of GeoSPaaS. It is composed of several GeoDjango apps which provide basic functionalities:
geospaas.catalog
: defines the structure of GeoSPaaS's catalog, which is used to store metadata about Earth Observation datasets. This is the core around which all GeoSPaaS functionalities are built.geospaas.vocabularies
: defines a structured way to assign properties to the datasets in the catalog. Based on and compatible with the GCMD standard.geospaas.base_viewer
: a basic web page which allows to view the contents of the catalog.geospaas.nansat_ingestor
: data ingester based on Nansat. Can be used to populate a catalog. The preferred tool for this is now django-geo-spaas-harvesting.geospaas.export_DIF
: app for exporting metadata to GCMD DIF format.
To make use of django-geo-spaas
, you need to install the apps in a GeoDjango project.
To set up such a project, you can follow the instructions here:
https://docs.djangoproject.com/en/5.0/ref/contrib/gis/tutorial/.
You can also go through the workshops in the following repository to learn more about GeoSPaaS and see some practical uses: https://github.com/nansencenter/geospaas-workshops.
Docker can be used to set up a GeoSPaaS project more easily.
You will find instructions below for a basic set-up.
- Install Docker
- Clone
django-geo-spaas
and change into that directory - Run
./build_container.sh
. That will do the following:
- build a Docker image with Nansat, Django and Django-Geo-SPaaS installed
- create a project directory
project
in the current dir - create the database and update vocabularies
- create a container with name
geospaas
with the current dir mounted to/src
. Thus the container sees all changes you do to the geospaas code.
- Run
docker start -i geospaas
. That will launchbash
inside thegeospaas
containter. From bash you can launch:python project/manage.py shell
. - If you want to mount more directories, you can run the following command:
docker create -it --name=geospaas \
-v `pwd`:/src \
-v /input/dir/on/host:/path/to/dir/in/container \
geospaas
And then run docker start -i geospaas
to go into the container and have access to commands.
6. Alretrnatively you can start the container in background:
docker start geospaas
and then execute commands in the runnning container:
docker exec -i geospaas project/manage.py shell
If you already have a project directory or if you are working on another app you can use the existing Docker image with Django-Geo-SPaaS. The image is already uploaded to Docker Hub, so no above steps are necessary. The workflow can be the following:
- Create a container with necessary directories mounted
- Start container in background
- Run Django commands from your host:
# create container named myapp from Docker image geospaas
docker run -it --rm --name=myapp \
-v /host/dir/myapp:/src \
-v /data/dir:/data \
nansencenter/geospaas
# create project dir in /host/dir/myapp
docker exec myapp django-admin startproject myproject
# update settings.py
INSTALLED_APPS = [
...
'django.contrib.gis',
'leaflet',
'django_forms_bootstrap',
'geospaas.base_viewer',
'geospaas.nansat_ingestor',
'geospaas.catalog',
'geospaas.vocabularies',
'myproject',
]
...
DATABASES = {
'default': {
'ENGINE': 'django.contrib.gis.db.backends.spatialite',
'NAME': os.path.join(BASE_DIR, 'geodjango.db'),
}
}
# update urls.py
from django.contrib import admin
from django.urls import path, include
urlpatterns = [
path('admin/', admin.site.urls),
path('tests/', include('geospaas.base_viewer.urls')),
]
# migrate database schema
docker exec myapp /src/myproject/manage.py migrate
# update vocabularies
docker exec myapp /src/myproject/manage.py update_vocabularies
# ingest metadata
docker exec -w /src/myproject myapp ./manage.py ingest /data/myfiles*.nc
The django-geo-spaas-harvesting app can be used to populate a GeoSPaaS catalog.