WCMC Labs

The homepage for the WCMC Informatics team. A basic rails app for showing off the Informatics team's products, storing Servers, and information about the installations of products across servers.

Pre-requisites

Ensure you have Ruby 2.3.8 installed (with bundler version specified in bottom of Gemfile.lock).

• Rbenv:

rbenv install 2.3.8
gem install bundler -v 1.17.3

• RVM:

rvm install 2.3.8
gem install bundler -v 1.17.3

Setup

• Clone repo
• Run bundle install
• Run cp config/secrets.yml.sample config/secrets.yml
• Populate config/secrets.yml according to secrets.yml.sample which gives guidance on how to access the secrets. The Teams wiki may also be helpful.
• Either run rails secret and update the development secret_key_base in config/secrets.yml
• Or run sed -i -E "s/^ secret_key_base:.*/ secret_key_base: $(rake secret)/" config/secrets.yml to update all environment entries.
• Run yarn install
• Run rails db:drop db:create db:migrate

Run

• Open a terminal for both: rails s and ./bin/webpack-dev-server

KPI Page

NB: Make sure you have the secrets populated before running this task, else it will fail as it expects certain secrets to be present!

The KPI page will be initially blank if you don't have an instance of the KPI model in your database. Run rake kpi:regenerate to create a KPI, which should then populate the various charts on the KPI page. The data can be manually updated by running the same command if the latest data is required.

Google Analytics

This requires a bit more nuance compared to the standard KPI fields. Please note that the code applies to, and was built around, v4 of the Google Analytics API, and so in future if the response from the API changes in terms of output, then it will be the maintainers' responsibility to fix this.

1. Get a copy of the Service Account Credentials from Lastpass > Development Environments. Store that in the root of your repository as a JSON. The GoogleAnalytics::Base class specifically looks for it in that location to authenticate you against Google.

2. You need to have the correct GA tracking codes in for each project that's tracked via GA and and is also present on Labs, prior to running the KPI regeneration rake task, which now also includes a way to generate GA user counts (in the last 90 days) for each project. *Remember to add the client_email found in the Service Account Credentials to the GA account of the site you wish to track via the Admin panel, and that at least 'Read and Analyse' privileges are provided for the service account, otherwise an error will be registered when you try to access the statistics.

   A script has been written and added to the Google Scripts dashboards for analytics.wcmc@gmail.com and informatics.wcmc@googlemail.com. The script is called add_analytics_service_account.js and running the listAccounts function will run through all the analytics accounts for the logged in Google account and add this access.

There is a field in the Edit view of a product for entering in the Google Analytics View ID code for that particular product. Please note that this is not the Account ID, rather this is the View ID, which only ever contains numbers. To identify this, within the Google Analytics Dashboard, from the main menu, look for the Views column for each project, which should always contain 'All Web Site Data' as the default. Underneath each view is the View ID. Also whenever possible, use the code which corresponds to views that exclude the WCMC office as that is most reliable. When you then run the Rake task, it will compute the user counts for each project which has a tracking code and then updates the list of projects and KPI page accordingly.

Reviews

If the review questions are not showing, you may need to uun rake review:load to populate the review questions from db/review_template.json

Deployment

• Staging: cap staging deploy
• Production: cap production deploy

Sign In

Sign in uses Omniauth with Github, a method is then ran to query the Github API to see if the user is a member of the wcmc-core-dev team. If they are, login succeeds. If not, the login is rejected and they are directed to the root of the application.

You will need to register an application with Github to use this and stub out the is_dev_team? method to return true.

For local development, homepage URL for the github application will need to be http://0.0.0.0:3000 and Authorization Callback URL will be http://0.0.0.0:3000/users/auth/github/callback

User Interface for displaying the Database structure of our products in the Entity-Relationship format.

How it works:

1. The product of which DB structure needs to be uploaded to Labs, must include the following gem into the Gemfile: https://github.com/unepwcmc/domain_uploader

2. The domain_uploader gem includes the rails-erd gem which inspects the database and generates dotfiles for each entity, representing relationships in the E-R format. The rails-erd gem needs graphviz installed in the system in order to work properly.

3. Run the following rake task to upload the current product on Labs: bundle exec rake du:uploader['product_name','host'], where product_name must match the name of the product on Labs and host is the host you wish to upload the domain to, so for example http://localhost:3000 if you are trying it locally. Please note that there cannot be any space between the comma in the rake task params.

4. On Labs you will now have a new folder in public which is public/domains/product_name with a list of .png files, each of which represents the relationships of an entity/model.

5. On the header of the Labs website, click on Domains to have a list of all the uploaded domains. Then choose one by clicking on the eye looking icon.

6. On the left of the new page you will have a list of all entities of that product and by clicking on them you will have the corresponding E-R diagrams showing as an image.