SSCS - Submit Your Appeal

Background:

Anyone who disagrees with a decision about their entitlement to benefits has the right to appeal against that decision. The first step is asking the Department for Work and Pensions to look at the decision again. This is known as requesting ‘Mandatory Reconsideration’. If they still disagree, they can appeal to the Social Security and Child Support tribunal.

Should an appellant wish to appeal online this node.js web application allows them to do so. The application takes the appellant on a journey, presenting a single question per page (GDS guidelines), at the end of their journey we present an appeal summary page, allowing the user to edit their answers or sign and submit.

Dependencies

• Docker
• Redis

Development

Config

Redis is required to run the application. You can either install it or use a docker image (See docker section below for instructions). Install Redis: download, extract and build:

http://download.redis.io/redis-stable.tar.gz
tar xvzf redis-stable.tar.gz
cd redis-stable
make

Sanity check:

make test

Add this to your path:

/usr/local/bin

Copy over both the Redis server and the command line interface:

sudo cp src/redis-server /usr/local/bin/
sudo cp src/redis-cli /usr/local/bin/

Start redis:

redis-server

Install npm dependencies:

yarn install

Generate cookie banner content:

./node_modules/gulp/bin/gulp.js default

Bring up SYA in a new terminal window:

yarn dev

View the application:

https://localhost:3000

Docker

If you would like to view the application without having to setup Redis you can do so via Docker.

We use the Dockerfile and docker-compose.yml to create a container to bring up the app which includes Redis.

Build the node.js Dockerfile containing SYA into a local image:

docker build -t hmcts/submit-your-appeal:latest .

Bring up the container:

docker-compose up sya

View the application:

https://localhost:3000

If you prefer to run the application natively but still use docker for Redis you can do so by running: docker-compose up redis

You would then start the application by running yarn dev as above

End-to-end testing

Ensure both SYA (from one of the methods above) and the API are up (including docker dependencies run through CFTLib). At present these tests do not run within Docker, therefore, open a new terminal window.

Functional tests: We have split our functional tests into two. Firstly we have tests for entire journeys through the form:

yarn test:functional

However, in order to get this command to run properly you currently have to remove --grep @functional tag from the test:functional script in package.json.

Secondly, we have tests for various pages in the form:

yarn test:e2e-pages

Functional test batches:

To improve reliability running the functional tests locally you can run them in batches using the following command

yarn test:functional:batches

If you wish to increase the speed of the tests, you can decrease the wait time between each action, the default is set to 500ms. Override this by setting the env var E2E_WAIT_FOR_ACTION_VALUE, I find 50ms works fine. Use this together with batching like so:

E2E_WAIT_FOR_ACTION_VALUE=50 yarn test:functional:batches

Smoke tests:

yarn test:smoke

Unit tests

yarn test

Code coverage

yarn test:coverage

Security scan of installed packages

yarn test:audit

Useful docker commands

List images

docker images

List containers

docker ps

Execute an interactive bash shell on the container

docker exec -it <container id> sh