kushtrimh / zana Goto Github PK

Is your feature request related to a problem? Please describe.
CICD pipeline takes too long when a new build or deployment is being executed.
Rust crates in services take most of the time, 2 minutes on average for build or tests run.

Describe the solution you'd like
Caching can be setup for the CICD pipeline, so the same dependencies are not download multiple times.
More information about caching for GitHub actions: https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows
A good examples from Cargo Lambda shows caching setup for Rust crates on GitHub Actions: https://www.cargo-lambda.info/guide/automating-deployments.html#step-3-creating-the-release-workflow

Describe alternatives you've considered
No other alternatives considered. This is supported by GitHub Actions, and would be one of the better solutions.

Additional context
Examples from some of the latest build and deployment runs.

Create a service that will fetch data from Google Books API.
Service should be written in Rust, and written as a library so it can be used later by other services which will be more focused on the environment that they will be used or deployed in.

The library should accept an API Key and the Google Books API URL.
When provided an ISBN, Google Books API should be queried for the selected book and return the data.
If the book cannot be found with ISBN, it should be search using the book name and author name, and if multiple books are returned in this case, the first one should be picked.

If an error occurs, an Error entity should be returned that has detailed information.

• HTTP Response status code if the error is HTTP related
• Description about the reason why the error happened

Zana willl be deployed in AWS, and the resource provisioning and deployment should be handled using AWS CDK.
Java will be used to provision the resources using AWS CDK.

More info to be added soon...

Add OpenLibrary support https://openlibrary.org/developers/api.

In this issue, a refactor should be made so that Google Books and OpenLibrary calls will return the same struct once called.
2 requests should be made for OpenLibrary.

1. Search by ISBN
2. Once step 1 returns, get the ratings on a seperate API call

Add a new project in services called zana-lambda.

This project will be a AWS Lambda written in Rust, which will use zana library to fetch data from Google Books API.
The responses from this Lambda should be in JSON, and it should accept Google Books API client id and secret and Google Books API URL as environment variables.

In cases where data of a book could be fetched from Google Books API, the response should be as follows

Status: 200

{
    "data": {
        "page_count": 200,
        "average_rating": 4.2,
        "ratings_count": 143,
        "description": "Some description about the book"
    }
}

In cases where an error occurs, and data cannot be fetched, the response should be as follows

Status: 429

{
    "error": "limit_reached",
    "details": "Cannot retrieve data from external services due to API request limits"
}

The error details and types are only placeholders.

2 environment variables should be provided to the created Lambda.

• GOOGLE_BOOKS_API_URL - which should have the value https://books.googleapis.com/books/v1
• GOOGLE_BOOKS_API_KEY - Google Books API Key