Project Breakdown

Implement backend services

1. Create cloud functions to manage dealers and reviews

These are implemented on IBM Cloud Functions. Versions of these are stored locally in functions/sample/<nodejs or python>. The serverless Functions connect to Cloudant via a service API KEY and URL, which are configured as Parameters for the Function: if no alternative value is supplied in the request, the Parameter will be used by default, allowing for secure connection to the service. The endpoints are exposed over the web as follows:

• <URL>/api/dealership (GET only)
• <URL>/api/review (GET or POST)

2. Create Django models and views to manage car model and car make
3. Create Django proxy services and views to integrate dealers, reviews, and cars together

Add dynamic pages with Django templates

1. Create a page that shows all the dealers
2. Create a page that show reviews for a selected dealer
3. Create a page that let's the end user add a review for a selected dealer

Containerize your application

1. Add deployment artifacts to your application
2. Deploy your application

Setup

This project was created in a local development environment rather than in the web-container "Theia" environment provided by IBM.
As such, there are some differences to work around.

Using Couchimport

Couchimport is a node.js tool that imports data into Cloudant. In this case, we have pre-defined test data that we want to load into our Cloudant instance. We install Couchimport by running npm install -g couchimport.

Set the following environment variables using the service credentials of your Cloudant instance.

export IAM_API_KEY="REPLACED IT WITH GENERATED <apikey>"

export COUCH_URL="REPLACED IT WITH GENERATED <url>"

From the project root, run cat cloudant/data/dealerships.json | node node_modules/couchimport/bin/couchimport.bin.js --type "json" --jsonpath "<THE-TABLE-WE-WANT-TO-IMPORT>.*" --database dealerships --url <REPLACE-ME-WITH-CLOUDANT-URL>.

Setting Up PostgreSQL

Instead of using the instance provided by IBM using their Theia environment, I chose to set up a local instance using Docker Compose.

In a terminal from the project root, navigate to storage-docker-configuration and run docker-compose up. On the first run, this will download and install Postgresql into the container and run it. The database will be exposed on the default port 5432 so that we can connect to it. The configuration is implemented in settings.py.