Cilantro is a task runner designed to manage long running distributed jobs that operate on file system objects. It is written in Python (3.6+) and uses Celery and Flask.
-
Copy the
.env-defaultfile to.envand modify it. In most cases onlyUIDhas to be adjusted. The UID / GID of the current user can be read withid -u/id -gon UNIX systems. -
Install docker (Community Edition)
You might have to add the proper 3rd party PPA. Refer to the official documentation. Example for Ubuntu-based distributions: https://docs.docker.com/install/linux/docker-ce/ubuntu/#install-docker-ce
sudo apt install docker-ce
docker-compose build
docker-compose up
To stop the application run:
docker-compose stop
In order for the test to function properly you have to create some files with
.tif ending in the folder ./data/staging/foo.
The web service runs on port 5000. The following command will create a test task:
curl -XPOST http://localhost:5000/job/test/foo
You can then query the job status with the returned job_id:
curl http://localhost:5000/job/<job_id>
Flower is included in the docker config and is available for debugging under http://localhost:5555.
-
Start the application as described under Running the app with docker
-
Run the tests inside the dedicated docker container
docker exec cilantro_cilantro-test_1 python -m unittest -v -
Alternatively you can run the complete build script out of the build-scripts repository. After cloing the repo into your workspace, run the following command from within your cilantro directory.
../build-scripts/cilantro-build.sh
This will build, start and stop the docker infrastructure and run the tests.
On Linux hosts the tests will fail because the data directory created by docker not have the right permissions and the user account that runs the test can not access it. The easiest way to fix that is just to change the owner on the whole directory and subfolders.
sudo chown -R $(whoami):$(whoami) data/
After that re-run the tests and they may succeed.