Help With Fees - Staff App

Overview

This app is used by staff in the courts and tribunals to enter data regarding help with fees applications, record the decision, and collect statistics.

Project Standards

• Authentications via Devise / CanCanCan
• Rspec features, not cucumber
• Slim templating language
• JavaScript in preference to Coffeescript

Redis

We are using Redis for step back navigation link. You will need to run a redis-server in order for the application to work correctly. We are using mock_redis in rspec to make it work it tests.

Feature switching

Model FeatureSwitching have following attributes: feature_key, activation_time, office_id, enabled. When apply you can "schedule" a feature to be active from a certain date or for specific office. Feature switching table is managed manually from a rails console for now.

Active Feature flag for post UCD changes

Run this in your localhost to enable UCD changes: FeatureSwitching.create(feature_key: 'band_calculation', enabled: true)

N+1 queries debugging

There is a gem called Bullet. If you want to check N+1 queries in development mode, you can uncomment Bullet related lines in development.rb

Delayed CRON Jobs (delayed_cron_job)

We need to keep an eye on the results of DWP checks. When the API is down the service will disable benefit related applications. Then we re-run failed checks in 10 minutes intervals to see if the API is back again. We are not using standard CRON table because Kubernetes have a bug. So we are using delayed cron job that has the schedule in DB table. These tasks are checked by

rake jobs:work

that needs to run on the background, currently set up in rub.sh : rake jobs:work &

Delayed jobs for BenefitChecks

To set it up (if there is no record in DB) run this in rails console: BenefitCheckRerunJob.delay(cron: '*/10 * * * *').perform_now

Delayed jobs for DWP offline notification

Runs every 5 minutes DwpReportStatusJob.delay(cron: '*/5 * * * *').perform_now

Delayed jobs for HMRC data purge

Runs 10 minutes past midnight HmrcDataPurgeJob.delay(cron: '10 0 * * *').perform_now

Delayed jobs for Personal data purge

Runs every day at 1am PersonalDataPurgeJob.delay(cron: '0 1 * * *').perform_now

Delayed jobs for Abandoned Application data purge

Runs every day at 1am AbandonedApplicationPurgeJob.delay(cron: '0 1 * * *').perform_now

Delayed jobs for Old export file data purge

Runs every day at 2am OldFileExportPurgeJob.delay(cron: '0 2 * * *').perform_now

Front end library

You will need to install govuk-frontend library

npm install --save govuk-frontend

Mimemagic gem has a dependency so you need to install this on your machine first brew install shared-mime-info.

Creating initial user

There is a rake task that takes email, password and role

rake user:create

If you want to add any custom options, use the below as an example:

rake "user:create[[email protected], 12345678900, admin, name]"

Note: the quotes around the task are important!

Run an applicants report for a finacial year

There is a rake task that creates the report

rake "reports:applicants[2021, 2022]"

this will generate a zip file applicants-2021-2022-fy.csv.zip

Run raw data extract for longer timescale

There is a rake task that creates the report

rake "reports:raw_data_extract[2021-01-01, 2022-12-31]"

this will generate a zip file raw_data-1-1-2021-31-12-2022.csv.zip

Run tests in parallel

Follow the official guides to setup your local env

Run the specs in parallel

RAILS_ENV=test bundle exec rake parallel:spec

Run the cucumber features in parallel

CAPYBARA_SERVER_PORT=random bundle exec rake parallel:features

Cucumber test report

Cucumber report is enabled now. At the end of the test run you should see a link to a website. When you run tests in parallel it will generate report per process so if you want to see one report only you should run test directly without parallel functionality.

Deployment versions trigger: 7