Data Submission Tool

Getting started

To set up development environment clone this repo and create a .env file.

cp env.template .env

Modify values in .env before continuing.

List of important variables:

• ODIC_*:

  update required settings. eps. important are OIDC_RP_CLIENT_ID and OIDC_RP_CLIENT_SECRET.

• GMAPS_API_KEY:

  google maps api key

• FRONTEND_DEV_MODE:

  set to True to load development css and js files.

Building UI

The UI (js and css code) lives inside the frontend folder. There is a Makefile which helps compiling the clojure code into js.

cd frontend
make <target>

UI make targets

• clean:

  remove all compiled js output

• figwheel:

  build the deveploment version of the UI and keep watching for changes to rebuild UI automatically when changes are written to the source code.

• build:

  build the development version of the UI

• copy-static:

  copy updated static files from node_modules (generated via make build) into resources folder.

• release:

  build the production / release version of the UI

Django App

The development environment is fully dockerised and uses docker-compose with local host bind mounts (./volumes folder) for data storage. If a full reset of the dev env is required, stop all containers and delete the ./volumes folder.

# start db in detached mode
docker-compose up -d db
# build initial image for data-submission-tool
docker-compose build web
# start django app in debug mode
docker-compose up web

The django app should now be accessible at http://localhost:8000 . All code and UI changes on host system should be detected by the django app inside the container and are immediately available; usually without restartin the django app.

To run the cron job which updates and ingests data feeds into the db run the cron container.

# run in a separate terminal or use -d to sapwn in detached mode.
docker-compose up cron

If an interactive python debugger is required the django container nees to be started slightly different.

# stop existing conatiner if running
docker-compose stop web
# start django container (this allows dropping into an interactive pdb session)
docker-compose run --rm --service-ports web

# if required: in a 2nd terminal install any additional pkgs required for debugging
# e.g.:
RUN_CID=$(docker ps --filter "label=com.docker.compose.oneoff=True" --filter "label=com.docker.compose.service=web" --format '{{.ID}}')
docker exec -ti -u 0 $RUN_CID pip3 install pyreadline ipdb

Migrations can be created in a similar fashion:

# stop existing conatiner if running
docker-compose stop web
# start django container (this allows dropping into an interactive pdb session)
docker-compose run --rm --service-ports web python3 manage.py makemigrations

Release

A Makefile is provided to simplify building and tagging final images. Please make sure that the UI has been built in release mode as described above.

Make targets:

• build:

  build and tag the image with :latest and a tag derived from current git version.

• up:

  run built images without mounting any source files. This allows fora quick test whether the images are working correctly. This setup is accessible at http://localhost:8001

• push:

  push :latest and a tag derived from git to docker hub