Quincy

This Node.js service is designed to scrape RSS feeds and store episode data in a SQLite database, leveraging Prisma as an ORM. It automatically processes episodes from a predefined list of RSS feed URLs, capturing essential information such as title, description, publication date, and duration.

Setup Instructions

Prerequisites

Node.js (v14 or newer recommended)
npm (v6 or newer)

Installation

Clone the Repository

git clone git@github.com:Noreaster76/quincy.git
cd quincy

Install Dependencies

Navigate to the project directory and run:
```
npm install
```

Setting Up the Database

This project uses SQLite, managed through Prisma. To set up and migrate the database:

Generate Prisma Client
```
npx prisma generate
```

Create a local .env file

echo 'DATABASE_URL="file:./dev.db"' > .env

Run Migrations

Ensure the SQLite database is set up with the correct schema:
```
npx prisma migrate dev
```
This command creates a SQLite database file according to the schema defined in prisma/schema.prisma.

Running the Application

To start the service and begin processing RSS feeds:

npm run build
npm run start

This compiles the TypeScript code and runs the compiled JavaScript from the dist directory.

Exploring the Downloaded Episodes

Since the episodes are downloaded and persisted to a local SQLite database, you can explore it from your terminal as follows:

➜  quincy git:(main) ✗ sqlite3 prisma/dev.db
SQLite version 3.43.2 2023-10-10 13:08:14
Enter ".help" for usage hints.
sqlite> select count(id) from Episode;
2060
sqlite> .exit
➜  quincy git:(main) ✗

Running the Tests

I included a few unit tests. Execute npm test to run them.

Notes

There was a requirement that sort of sounded like this service should be invoked from a network call. If that is the case, I respectfully push back on that requirement, and instead argue that npm run build could simply be invoked from a cron job running locally. There's no real reason why this service should have to be invoked over a network; it makes more sense for it to simply be kicked off on a regular schedule.
It looks like a lot of the podcast attributes are actually optional, per this Apple documentation.

Known Issues

Duration Parsing in hh:mm:ss Format: Currently, parsing episode durations provided in hh:mm:ss format from RSS feeds does not work as expected. Fixing this is probably pretty straightforward, but I have to submit what I have so far and consider it done.

License

Please create an issue if you want me to add a license.

Noreaster76 / quincy